Downloaded from UvA-DARE, the Institutional Repository of the University of Amsterdam (UvA)http://dare.uva.nl/document/164740

Description ThesisFile ID 164740Filename thesis.pdf

SOURCE, OR PART OF THE FOLLOWING SOURCE:Type DissertationTitle Thrombophilia ad dies vitaeAuthor D.M. CohnFaculty Faculty of MedicineYear 2010Pages 184

FULL BIBLIOGRAPHIC DETAILS: http://dare.uva.nl/record/331702

Copyrights It is not permitted to download or to forward/distribute the text or part of it without the consent of the copyright holder(usually the author), other then for strictly personal, individual use. UvA-DARE is a service provided by the Library of the University of Amsterdam (http://dare.uva.nl)

Thrombophilia Ad Dies Vitae

Thrombophilia For Life

Author: Danny Michael Cohn

Photo Cover painting: Ido Menco

Printed by: Gildeprint Drukkerijen B.V.

ISBN/EAN: 978-90-9025005-2

The printing of this thesis was financially supported by: J.E. Jurriaanse Stichting, Federatie van Nederlandse Trombosediensten, Stichting tot Steun Promovendi Vasculaire Geneeskunde, Stichting AMSTOL, Pfizer, LEO Pharma, GlaxoSmithKline, AstraZenica, MSD, Bayer, Schering-Plough, Boehriner-Ingelheim, Sanquin, Sanofi-Aventis.

Copyright ©2010, D.M. Cohn, Amsterdam, the Netherlands No part of this thesis may be reproduced, stored in a retrieval system or transmitted in any form or by any means, without prior written permission of the author.

Thrombophilia Ad Dies Vitae

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam

op gezag van de Rector Magnificus

prof. dr. D.C. van den Boom

ten overstaan van een door het college voor promoties

ingestelde commissie,

in het openbaar te verdedigen in de Aula der Universiteit

op woensdag 17 maart 2010, te 14.00 uur

door

Danny Michael Cohn

geboren te ‘s-Gravenhage

PROMOTIECOMMISSIE

Promotor: Prof. dr. H.R. Büller

Copromotores: Dr. S. Middeldorp

Dr. P.W. Kamphuisen

Overige Leden: Dr. R.G. Farquharson, MD, FRCOG

Dr. V.E. Gerdes

Prof. dr. M.J. Heineman

Prof. dr. M.M. Levi

Prof. dr. J.C.M. Meijers

Prof. dr. F.R. Rosendaal

Faculteit der Geneeskunde

Financial support by the Netherlands Heart Foundation for the publication of this thesis is

gratefully acknowledged.

TABLE OF CONTENTS

Chapter 1 General introduction and outline of the thesis 7

Part I Clinical and psychological aspects of venous thromboembolism and thrombophilia Chapter 2 Thrombophilia and venous thromboembolism: 15

implications for testing Chapter 3 Thrombophilia testing for prevention of recurrent 27 venous thromboembolism Chapter 4 The psychological impact of testing for thrombophilia: 37 a systematic review Chapter 5 Quality of life after pulmonary embolism: 49 the development of the Pulmonary Embolism Quality of Life questionnaire (PEmb-QoL) Chapter 6 Quality of life after pulmonary embolism: 55 validation of the Pulmonary Embolism Quality of Life questionnaire (PEmb-QoL) Part II Identification of new thrombophilic factors

Chapter 7 Venous thrombosis is associated with hyperglycaemia 75 at diagnosis: a case-control study

Chapter 8 Stress-induced hyperglycaemia and venous thromboembolism 85 following total hip or total knee arthroplasty

Chapter 9 Common genetic variation at the Endothelial 97

Lipase (LIPG) locus and the risk of coronary artery disease and deep venous thrombosis

Part III Reproductive aspects of venous thromboembolism and thrombophilia Chapter 10 Risk of postpartum haemorrhage in women receiving 117 therapeutic doses of low-molecular-weight heparins: a cohort-study Chapter 11 Recurrent miscarriage in women with and without 127

antiphospholipid syndrome: prognosis for the next pregnancy outcome

Chapter 12 Increased sperm count maintains high population 139 frequency of factor V Leiden Appendix A call for a uniform reference format for submission 147 of manuscripts Summary 151 Samenvatting 159 Dankwoord 169 Authors’ affiliations 175 List of publications and curriculum vitae 181

Chapter 1

General introduction and outline of the thesis

General introduction

Venous thromboembolism (VTE) is a frequently occurring disease in Western societies,

with an annual incidence of 2-3 per 1000 inhabitants.1-3 Its clinical spectrum ranges from

deep venous thrombosis of the leg to potentially fatal pulmonary embolism.

Even though anticoagulant therapy is highly effective, treatment success is counterbalanced

by a 15% annual risk of bleeding, and a 3% annual risk of major bleeding (such as

intracranial or gastrointestinal bleeding).4 After discontinuation of anticoagulant therapy,

the risk of recurrence following the initial event is as high as 17-30% after 7-8 years.5, 6

In addition, approximately half of the patients with deep venous thrombosis develop post

thrombotic complaints in the leg.7

It has been known for many decades that several exogenous factors increase the risk of VTE.

These factors include major trauma, prolonged immobilization, surgery, use of oral

contraceptives or hormonal replacement therapy, pregnancy, puerperium, cancer and

chemotherapy. In addition to these exogenous risk factors, several endogenous conditions

are known to be associated with an increased risk of VTE which are denoted as

“thrombophilia”.

The term thrombophilia was introduced in 1965 by Olav Egeberg, when he reported a

Norwegian family with a remarkable thrombotic tendency due to antithrombin deficiency.8

In 1982 and 1984, protein C and protein S deficiency were identified as new hereditary risk

factors for VTE.9,10 Deficiencies in antithrombin, protein C or protein S are rare in the

general population and in patients with VTE.11 It wasn’t until the mid-1990s when more

common risk factors for VTE were identified, such as factor V Leiden and the prothrombin

G20210A mutation.12,13

Factor V Leiden is a gain-of-function mutation that leads to resistance of activated clotting

factor V through inactivation by activated protein C.14 Besides VTE, carriers are also at a

slightly increased risk of spontaneous or recurrent miscarriages.15 Despite these obvious

disadvantages, the point mutation -which occurred about 21,000 - 34,000 years ago-16 has a

high prevalence of approximately 4 to 7% in Caucasians.17 It has long been speculated that

the high population frequency of factor V Leiden reflects some sort of evolutionary benefit

for carriers.18-23

Thrombophilia also comprises acquired conditions, of which the antiphospholipid

syndrome is the most established. Antiphospholipid syndrome is a disorder, which causes

arterial and/or venous thrombosis, as well as pregnancy-related complications such as

Chapter 1

8

1

2

3

4

5

6

7

8

9

10

11

12

miscarriage or preterm delivery. The syndrome occurs due to the autoimmune production

of antibodies against plasma proteins that bind to negatively charged phospholipids.

In approximately half of all patients with VTE, a thrombophilic defect can be

demonstrated.24 Moreover, one or more exogenous risk factors can be demonstrated in

circa 50% of all patients with VTE (the remaining episodes being denoted as

“unprovoked”). This implies that in about 25% of patients with VTE no obvious cause can

be found. It is therefore likely that other, yet unidentified risk factors are to be discovered.

During the past decade, the associations between various thrombophilic defects and VTE

have been clearly established. Nevertheless, several issues remain unresolved. These

include the clinical relevance of testing for thrombophilia, the identification of new risk

factors for VTE and aspects of thrombophilia related to reproduction. These issues will be

addressed in this thesis.

Outline of the Thesis

Originally, the main focus of this thesis was planned to be assessment of the efficacy of

testing patients with a first VTE for thrombophilia. For this purpose, we initiated the

NOSTRADAMUS (Necessity Of Screening for ThRombophilia At Diagnosis of venous

thromboembolism to Assess Most Unresolved iSsues) trial. This randomized controlled,

multicentre trial was the ideal instrument to demonstrate the benefits, if any, of testing for

thrombophilia. However, due to a low inclusion rate which was primarily the result of

competition with industry initiated intervention studies, this trial was terminated

prematurely.25

Nevertheless, we were able to perform several other studies focussing on consequences of

testing for thrombophilia and, in addition, on clinical and psychological aspects of VTE.

The results of these studies are presented in the first part of the thesis. Chapter 2 reviews

the associations between thrombophilia and risk of VTE. In addition, the implications of

testing for thrombophilia are highlighted. Whether the risk of recurrent VTE could be

reduced by testing patients with a first VTE for thrombophilia (by adjusting therapy or

taking preventive measures), is assessed by a systematic review of the available evidence in

chapter 3. In Chapter 4, the published research on psychological effects of testing for

thrombophilia is systematically reviewed and critically appraised. Even though quality of

life is increasingly being conceptualized as the central outcome of health care, instruments

General introduction and outline of the thesis

9

to specifically measure quality of life following pulmonary embolism are lacking. We

therefore created the “Pulmonary Embolism Quality of Life” (PEmb-QoL) questionnaire.

The development and validation analyses of this questionnaire are presented in chapters 5

and 6.

The second part of this thesis addresses a search of new risk factors for VTE. Chapter 7

describes a case-control study in which the association between elevated blood glucose

levels and the occurrence of deep venous thrombosis is investigated. Chapter 8 elaborates

on the risk of VTE and elevated glucose levels. In this study, the pre- and postsurgical

glucose levels in 12,383 patients who underwent knee or hip surgery are related to the

incidence of (a)symptomatic VTE. Whether an association is present between single

nucleotide polymorphisms of the gene encoding for endothelial lipase and VTE is

discussed in Chapter 9.

The third part of the thesis describes reproductive aspects of venous thromboembolism

and thrombophilia. Chapter 10 addresses the issue of safety of thromboprophylaxis in

pregnant women. Even though low-molecular-weight heparins have shown to be safe for

the fetus throughout pregnancy, the associated bleeding risk for the mother with high

(therapeutic) doses remains unknown. In this chapter, the risk of peripartum blood loss in

83 women who used therapeutic doses of low-molecular-weight heparin during pregnancy

is presented and compared to the risk in pregnant women who did not receive

thromboprophylaxis. The rate of successful pregnancy outcomes in couples with recurrent

miscarriage due to antiphospholipid syndrome of the women is presented in chapter 11.

Whether increased male fertility can account for the high population frequency of factor V

Leiden is described in chapter 12.

References

1. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and mortality of venous thrombosis: a population-based study. J Thromb Haemost 2007;5:692-9.

2. Oger E. Incidence of venous thromboembolism: a community-based study in Western France. EPI-GETBP Study Group. Groupe d'Etude de la Thrombose de Bretagne Occidentale. Thromb Haemost 2000;83:657-60.

3. Hansson PO, Welin L, Tibblin G, Eriksson H. Deep vein thrombosis and pulmonary embolism in the general population. 'The Study of Men Born in 1913'. Arch Intern Med 1997;157:1665-70.

4. Van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briët E. Bleeding complications in oral anticoagulant therapy. An analysis of risk factors. Arch Intern Med 1993;153:1557-62.

5. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical factors, and recurrent venous thrombotic events. JAMA 2005;293:2352-61.

10

Chapter 1

1

2

3

4

5

6

7

8

9

10

11

12

6. Prandoni P, Lensing AW, Cogo A et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996;125:1-7.

7. Brandjes DP, Büller HR, Heijboer H et al. Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet 1997;349:759-62.

8. Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh 1965;13:516-30.

9. Mannucci PM, Vigano S. Deficiencies of protein C, an inhibitor of blood coagulation. Lancet 1982;2:463-7. 10. Schwarz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S deficiency in familial

thrombotic disease. Blood 1984;64:1297-300. 11. Heijboer H, Brandjes DP, Büller HR, Sturk A, Ten Cate JW. Deficiencies of coagulation-inhibiting and

fibrinolytic proteins in outpatients with deep-vein thrombosis. N Engl J Med 1990;323:1512-6. 12. Bertina RM, Koeleman BP, Koster T et al. Mutation in blood coagulation factor V associated with resistance

to activated protein C. Nature 1994;369:64-7. 13. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3'-untranslated region

of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698-703.

14. Dahlback B, Hildebrand B. Inherited resistance to activated protein C is corrected by anticoagulant cofactor activity found to be a property of factor V. Proc Natl Acad Sci U S A 1994;91:1396-400.

15. Middeldorp S. Thrombophilia and pregnancy complications: cause or association? J Thromb Haemost 2007;5 Suppl 1:276-82.

16. Zivelin A, Griffin JH, Xu X et al. A single genetic origin for a common Caucasian risk factor for venous thrombosis. Blood 1997;89:397-402.

17. Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet 1995;346:1133-4. 18. Lindqvist PG, Svensson PJ, Dahlback B, Marsal K. Factor V Q506 mutation (activated protein C resistance)

associated with reduced intrapartum blood loss--a possible evolutionary selection mechanism. Thromb Haemost 1998;79:69-73.

19. Lindqvist PG, Zoller B, Dahlback B. Improved hemoglobin status and reduced menstrual blood loss among female carriers of factor V Leiden--an evolutionary advantage? Thromb Haemost 2001;86:1122-3.

20. Lindqvist PG, Dahlback B. Carriership of Factor V Leiden and evolutionary selection advantage. Curr Med Chem 2008;15:1541-4.

21. Gopel W, Ludwig M, Junge AK, Kohlmann T, Diedrich K, Moller J. Selection pressure for the factor-V-Leiden mutation and embryo implantation. Lancet 2001;358:1238-9.

22. van Dunne FM, Doggen CJ, Heemskerk M, Rosendaal FR, Helmerhorst FM. Factor V Leiden mutation in relation to fecundity and miscarriage in women with venous thrombosis. Hum Reprod 2005;20:802-6.

23. van Dunne FM, De Craen AJ, Heijmans BT, Helmerhorst FM, Westendorp RG. Gender-specific association of the factor V Leiden mutation with fertility and fecundity in a historic cohort. The Leiden 85-Plus Study. Hum Reprod 2006;21:967-71.

24. Weitz JI, Middeldorp S, Geerts W, Heit JA. Thrombophilia and new anticoagulant drugs. Hematology (Am Soc Hematol Educ Program ) 2004;424-38.

25. Cohn DM, Middeldorp S. [Early termination of the multicentre randomised clinical trial to evaluate the benefit of testing for thrombophilia following a first venous thromboembolism: the NOSTRADAMUS study]. Ned Tijdschr Geneeskd 2008;152:2093-4.

11

General introduction and outline of the thesis

Part I

Clinical and psychological aspects of venous thromboembolism and thrombophilia

Chapter 2

Thrombophilia and venous thromboembolism: implications for testing

Based on: Cohn DM, Roshani S, Middeldorp S. Seminars in Thrombosis and Hemostasis 2007 Sep; 33(6): 573-81.

Background of thrombophilia

Venous thromboembolism (VTE) is a common disease with an annual incidence of

approximately 2-3 per 1,000 inhabitants in Western societies.1-3 Its manifestations are deep

venous thrombosis, pulmonary embolism or a combination of both. VTE is a multicausal

disease, in which both exogenous and endogenous risk factors have been identified.4

Established exogenous risk factors are cancer, pregnancy, puerperium, surgery,

immobilisation and oral contraceptive use. The endogenous risk factors for VTE contribute

to the term “thrombophilia”. The definition of thrombophilia (an inherited or acquired

predisposition to VTE) has expanded over the past century. It was introduced in 1965 by a

physician called Egeberg, when he described a Norwegian family with a high tendency to

venous thrombosis due to a deficiency in the natural anticoagulant antithrombin.5

Antithrombin deficiency increases the risk of VTE by a less pronounced inhibition of

thrombin formation (and factor Xa) (see Figure 1 for the effects of thrombophilia on fibrin

formation). Subsequently, in the 1980s, deficiencies of the other natural anticoagulants,

protein C and its cofactor protein S, were found to increase the risk of VTE.6,7 A deficiency

of one of the latter proteins leads to a reduced inactivation of clotting factors Va and VIIIa.

Many mutations have been described in the coding genes for antithrombin, protein C and

protein S and, in fact, heterozygous mutations already cause a deficiency of the affected

protein. Deficiencies of natural anticoagulants (antithrombin, protein C and protein S) are

rather uncommon, with a prevalence of <1% in the general population and a prevalence of

at most 5% among patients with thrombosis.8-11

During the past two decades, newer and more prevalent thrombophilic defects have been

discovered, such as the factor V Leiden mutation (1994) and the prothrombin mutation

(1996).12,13 Factor V Leiden -the most common genetic risk factor for VTE- is a single point

mutation on coagulation factor V (G1691A) which leads to an amino acid substitution

(R506Q) on the major cleavage site for activated protein C (APC), thereby causing a

phenomenon called “APC resistance”. The protein C anticoagulant system is twofold

affected by factor V Leiden: 1) by impaired degradation of mutant factor Va by APC because

of the elimination of its most important cleavage site and 2) by impaired degradation of

factor VIIIa, since mutant factor V is a poor cofactor to activated protein C in the

degradation of factor VIIIa.14 A single point mutation (G20210A) in the 3`untranslated

region of the prothrombin (factor II) gene is the second most common genetic risk factor

for VTE. The function of the prothrombin protein is unaffected, however the plasma levels

of prothrombin are slightly increased as a result of the mutation.14

Chapter 2

16

1

2

3

4

5

6

7

8

9

10

11

12

Figure 1. Coagulation system and points of action of thrombophilic disorders

Also, persistently elevated levels of factor VIII have shown to be a risk factor for

thrombosis.15,16 It has been found that (mild) hyperhomocysteinaemia is a risk factor for

VTE, but its clinical relevance seems small, especially since lowering the homocysteine level

did not show a reduction in the recurrence of VTE.17,18

The only established acquired thrombophilic disorder is the primary antiphospholipid

syndrome (APS). This syndrome is characterized by a heterogenic group of auto-antibodies

directed against plasma proteins that bind to negatively charged phospholipids. The well

established, clinically most relevant antibodies can be divided into three categories: lupus

anticoagulant, anti-cardiolipin antibodies and anti-�2-glycoprotein-1 antibodies. APS is a

clinical diagnosis, based on at least one episode of arterial or venous thrombosis or

recurrent miscarriage, combined with the persistent presence of anti-phospholipid

antibodies.19

In approximately half of patients presenting with VTE, one or more thrombophilic defect

can be identified.20,21 This has led to widespread testing for thrombophilia, also in patients

17

Thrombophilia and VTE: implications for testing

with a first episode of VTE despite the fact that -at present- it is unclear whether this

strategy is beneficial.

Clinical implications of thrombophilia testing

The (dis)advantages and implications of thrombophilia testing are discussed in the

following section. Reasons for testing for thrombophilia might be clarification of the cause,

the opportunity to adjust therapeutic regimes of VTE in thrombophilic patients for the

optimal prevention of recurrence, and the opportunity to track asymptomatic family

members (and subsequently take preventive measures). Conversely, testing for

thrombophilia might lead to needless expenses, anxiety, and social problems.

Reasons to test for thrombophilia

It is often argued that patients and their doctors would like to have an explanation for the

episode of VTE, although this has never been explicitly studied. It should be realized

however, that the existence of a thrombophilic defect does not exclude other reasons for a

prothrombotic state. For example, a 60-year-old male presenting with an unprovoked deep

VTE of the leg might have an occult cancer as well as a thrombophilic defect. An important

argument in favour of testing for thrombophilia would be the opportunity to adjust

therapeutic measures for treatment of a VTE (by means of intensity or duration of

treatment). The optimal therapy for VTE depends on the risk of recurrence, the

(dis)comfort of the therapy and the risk of side effects, such as (major) bleeding. The

estimated risk of recurrence for VTE in general is ~5% per year 22,23 (although unprovoked

episodes tend to recur more frequently: ~20% in the first 2 years compared with provoked

episodes).24 Standard therapy for patients with a first VTE includes anticoagulant treatment

with vitamin K antagonists for 3 to 6 months, with international normalized ratios between

2.0 and 3.0.25 This therapy ensues an annual bleeding risk of 0.25% for fatal bleeding and

1.0% for life-threatening bleeding.26,27 A different approach to thrombophilic patients,

compared with non-thrombophilic patients is only justified if the former have a different

risk of recurrence. Even though thrombophilia has shown to increase the risk of a first VTE,

it is to date still controversial whether thrombophilia also increases the risk of recurrent

VTE. The estimated relative risk of recurrence in patients with thrombophilia is small, as

compared with patients without thrombophilia (Table 1). The estimated odds ratios (ORs)

for the natural anticoagulant deficiencies, as described mainly in retrospective analyses,

were 2.5.28-30 In one prospective study, the follow-up of the Leiden Thrombophilia Study

Chapter 2

18

1

2

3

4

5

6

7

8

9

10

11

12

(LETS), the risk of recurrence appeared even more moderate, with an OR of 1.8.11 Two meta-

analyses studied the risks of recurrence in patients with the common thrombophilias:

factor V Leiden and the prothrombin G20210A mutation. The risk of recurrence was found

consistently to be 1.2 to 1.4-fold higher in patients with factor V Leiden and 0.7 to 1.7-fold

Table 1. Estimated Relative Risk of VTE recurrence in patients with thrombophilia

type of thrombophilia relative risk

natural anticoagulant deficiencies 11,28-30 1.8-2.5

factor V Leiden 11,31-33 1.2-1.4

prothrombin G20210A 11,31-33 0.7-1.7

elevated levels of factor VIII:c 11,34,35 1.3-6

elevated levels of factor IX 11 1.2

elevated levels factor XI 11 0.6

mild hyperhomocysteinaemia 18,36-38 1.8-2.7

antiphospholipid antibodies 39-42 2-6

higher in patients with the prothrombin mutation.11,31-33 For the other thrombophilic defects,

less data are available. Three studies assessed the risk of recurrence in patients with high

levels of FVIII coagulant activity (FVIII:c) compared with patients with normal levels. One

case-control study showed that elevated levels of FVIII:c above 150% were associated with

an approximate 2-fold increased risk of recurrent VTE, compared with patients with a

single episode. In two cohort studies, the estimated relative risk of recurrent VTE was 6-

fold increased in those with FVIII:c levels above the 90th percentile, corresponding to 234%

and 294%, respectively.34,35 These results could not be reproduced in the LETS follow-up

study, in which an OR of only 1.3 was found.11 The data on estimated risk of recurrence for

elevated levels of FIX and FXI are scarce, but their impact on recurrence seems negligible.11

The attribution of mild hyperhomocysteinaemia in terms of risk of recurrence appears low

(1.8 to 2.7), and treating hyperhomocysteinaemia did not show a decrease in the number of

recurrences.18,36-38 The risk of recurrence in antiphospholipid or anticardiolipin antibodies

was investigated in four studies.39-42 The outcomes regarding relative risk for recurrence

ranged between 2- and 6-fold. These results are difficult to interpret, given that in these

studies the antiphospholipid and anticardiolipin antibodies or lupus anticoagulant were

not tested repetitively (as suggested by international guidelines).19 Moreover, duration of

anticoagulant treatment differed substantially. Adjustment of anticoagulant treatment in

19

Thrombophilia and VTE: implications for testing

thrombophilic patients after a first VTE has only been addressed for a difference in

intensity. This has not shown to be beneficial in patients with VTE, regardless of

thrombophilia. Reducing the intensity of VKAs below 2.0 led to an increase of recurrence

risk (1.9 versus 0.6%)43,44, whereas major bleeding complications did not differ between a

low-intensity and regular-intensity treatment (0.96 versus 0.93%).44 A higher intensity of

VKA in patients with antiphospholipid antibodies showed no reduction in the risk of

recurrence, but led to an increase in the bleeding risk.45,46 Whether clinical outcome of

patients with VTE and thrombophilia improves with prolongation of anticoagulant has

never been investigated. Current trials focus on whether such an intervention outweighs the

bleeding risk, as oral anticoagulant medication is known to prevent VTE by more than 90%,

as long as it is used.47 Finally, a potential advantage of testing patients with VTE for

thrombophilia may be the identification of asymptomatic family members. These

individuals have a 2- to 10-fold increased risk for VTE as compared with non-carriers.48-51

Regardless of this increased relative risk, the overall absolute risk remains low (Table 2). It

is often argued that asymptomatic family members with thrombophilia may benefit from

targeted prevention in high-risk situations (e.g., pregnancy, puerperium, surgery,

immobilization, and trauma), and the avoidance of acquired risk factors, most notably oral

contraceptives.

Table 2. Absolute risk of VTE in asymptomatic carriers of thrombophilia

type of thrombophilia overall (%/year) surgery, trauma, or immobilization (%/episode)

pregnancy (%/pregnancy)

oral contraceptive use (%/year of use)

natural anticoagulant deficiencies 48,61-65 0.4-4.0 8.1 4.1 4.3

factor V Leiden 48,50,51,61,64,66,67 0.1-0.7 1.8-2.4 1.9-2.1 0.5-2.0

prothrombin 20210A 67-69 0.1-0.4 2.0 2.8 0.2

elevated FVIII:c 70 0.3 1.2 1.3 0.6

mild hyperhomocysteinaemia 71,72 0-0.2 0.9 0.5 0.1

It is clear from Table 2 that bleeding risk associated with continuous anticoagulant

treatment outweighs the risk of VTE. It is notable that the figures considering surgery,

trauma and immobilization, as shown in Table 2, have been collected for the larger part in

times before standard prophylaxis was routine patient care. For pregnancy, ~80% of the

episodes occur in the postpartum period. Whether this should lead to administration of

prophylaxis in the postpartum period is a matter of physician and patient preference, given

Chapter 2

20

1

2

3

4

5

6

7

8

9

10

11

12

that the number needed to treat is 25 in case of a deficiency in the natural anticoagulants

and approximately twice as high in patients with the common thrombophilias. Finally, it is

clear from data in Table 2 that the use of oral contraceptives should be weighed against the

disadvantages of other contraceptive methods.

Reasons not to test for thrombophilia

Disadvantages of testing patients with a VTE for thrombophilia might be the cost of testing,

which is approximately €500.52 Several sophisticated studies focused on the cost-

effectiveness of testing for thrombophilia.53,54 These studies focused on selected patient

groups, because universal testing was considered less cost effective. It is of note that the

external validity of the results may be distorted by the fact that the findings were based on a

range of various assumptions. One study, by Marchetti et al, assessed the cost effectiveness

of testing for double heterozygosity of factor V Leiden and the prothrombin mutation, and

subsequently prolonging anticoagulant therapy in those tested positive for both common

thrombophilias.53 This strategy was considered cost effective, as testing all patients with

VTE provided one additional day of life at the cost of $13.624/quality-adjusted life-years.

The Thrombosis: Risk and Economic Assessment of Thrombophilia Screening study

assumed that testing for thrombophilia might be efficacious in high-risk situations. The

cost effectiveness of four different testing scenarios was calculated: (1) testing all women

prior to prescription of oral contraceptives and restricting prescription only to those tested

negative for thrombophilia; (2) testing all women prior to prescribing hormone

replacement therapy and restricting prescription only to those tested negative for

thrombophilia; (3) testing women at the onset of pregnancy and prescribing prophylaxis to

those tested positive for thrombophilia; and (4) testing all patients prior to major elective

orthopaedic surgery and prescribing extended thromboprophylaxis to those tested positive

for thrombophilia. It was concluded in this study that the second scenario would be most

cost effective, compared with the other scenarios. Nevertheless, selective screening based

on the presence of previous personal or family history of VTE was considered to be more

cost effective than universal testing in the four different scenarios. Furthermore, the

psychological impact and consequences of knowing that one is a carrier of a (genetic)

thrombophilic defect could be regarded as a drawback of testing. Most studies that focused

on impact of testing for thrombophilia showed that patients had experienced low

psychological distress following thrombophilia testing.55,56 Nevertheless, qualitative studies

described several negative effects. Bank et al conclude from their study that parents were

Thrombophilia and VTE: implications for testing

21

worried that their children ‘‘would be negatively influenced by factor V Leiden’’ and that

some carriers ‘‘felt stigmatized.’’57 Finally, a disadvantage of testing for thrombophilia

could be its potential social consequences (for instance problems with health insurance or

life insurance), although little data are available on this issue.

Thrombophilia testing scrutinized

As mentioned above, thrombophilia tests are performed widely, even though their clinical

benefits are yet to be established. We attempted to assess whether testing for

thrombophilia after a first VTE and prolonging anticoagulant treatment in patients with

thrombophilia is justified.58 The NOSTRADAMUS (Necessity Of Screening for

ThRombophilia At Diagnosis of venous thromboembolism to Assess Most Unresolved

iSsues) trial, a multicentre randomised controlled trial, was designed to assess whether this

strategy is beneficial in terms of clinical outcomes (the composite endpoint of recurrent

VTE and bleeding risk), quality of life and costs. We aimed to include 1336 patients with a

first VTE, whom we would randomly assign to one of two groups. The intervention group

would be tested for thrombophilia and subsequently receive the test results. Additional

anticoagulant treatment for a predefined period would be initiated in patients found to have

thrombophilia, while others would receive a standard predefined duration of treatment.

Patients in the control group they would receive the standard predefined duration of

treatment. Primary outcomes were the risk of recurrent VTE, clinically important bleeding

and the composite outcome of both. Other outcomes included overall quality of life and

costs associated with outcome measures 18 months after the initial episode of VTE.

However, the NOSTRADAMUS trial was prematurely terminated due to a low inclusion

rate.59 Several reasons account for this: delay of approval by medical ethics committees in

participating centres, competition with industry-initiated intervention studies with a higher

financial compensation in potentially eligible patients, and the lack of study staff and

allowance of expenses. To date, no randomized controlled trials have been completed to

assess whether testing for thrombophilia and subsequent therapeutic consequences

reduces the risk of recurrent VTE. However, in a case-control study the rate of recurrence

was shown not to be influenced by testing for thrombophilia.60

Chapter 2

22

1

2

3

4

5

6

7

8

9

10

11

12

References 1. Hansson PO, Welin L, Tibblin G, Eriksson H. Deep vein thrombosis and pulmonary embolism in the general

population. 'The Study of Men Born in 1913'. Arch Intern Med 1997;157:1665-70. 2. Oger E. Incidence of venous thromboembolism: a community-based study in Western France. EPI-GETBP

Study Group. Groupe d'Etude de la Thrombose de Bretagne Occidentale. Thromb Haemost 2000;83:657-60. 3. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and

mortality of venous thrombosis: a population-based study. J Thromb Haemost 2007;5:692-9. 4. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999;353:1167-73. 5. Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh

1965;13:516-30. 6. Mannucci PM, Vigano S. Deficiencies of protein C, an inhibitor of blood coagulation. Lancet 1982;2:463-7. 7. Schwarz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S deficiency in familial

thrombotic disease. Blood 1984;64:1297-300. 8. Miletich J, Sherman L, Broze G, Jr. Absence of thrombosis in subjects with heterozygous protein C deficiency.

N Engl J Med 1987;317:991-6. 9. Tait RC, Walker ID, Perry DJ et al. Prevalence of antithrombin deficiency in the healthy population. Br J

Haematol 1994;87:106-12. 10. Tait RC, Walker ID, Reitsma PH et al. Prevalence of protein C deficiency in the healthy population. Thromb

Haemost 1995;73:87-93. 11. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical

factors, and recurrent venous thrombotic events. JAMA 2005;293:2352-61. 12. Bertina RM, Koeleman BP, Koster T et al. Mutation in blood coagulation factor V associated with resistance

to activated protein C. Nature 1994;369:64-7. 13. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3'-untranslated region

of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698-703.

14. Dahlback B. Blood coagulation and its regulation by anticoagulant pathways: genetic pathogenesis of bleeding and thrombotic diseases. J Intern Med 2005;257:209-23.

15. Koster T, Blann AD, Briët E, Vandenbroucke JP, Rosendaal FR. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 1995;345:152-5.

16. Kraaijenhagen RA, In 't Anker PS, Koopman MM et al. High plasma concentration of factor VIIIc is a major risk factor for venous thromboembolism. Thromb Haemost 2000;83:5-9.

17. Den Heijer M, Koster T, Blom HJ et al. Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. N Engl J Med 1996;334:759-62.

18. Den Heijer M, Willems HP, Blom HJ et al. Homocysteine lowering by B vitamins and the secondary prevention of deep-vein thrombosis and pulmonary embolism. A randomized, placebo-controlled, double blind trial. Blood 2007;109:139-44.

19. Miyakis S, Lockshin MD, Atsumi T et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295-306.

20. Lane DA, Mannucci PM, Bauer KA et al. Inherited thrombophilia: Part 2. Thromb Haemost 1996;76:824-34. 21. Bauer KA. The thrombophilias: well-defined risk factors with uncertain therapeutic implications. Ann Intern

Med 2001;135:367-73. 22. Prandoni P, Bernardi E, Marchiori A et al. The long term clinical course of acute deep vein thrombosis of the

arm: prospective cohort study. BMJ 2004;329:484-5. 23. Van Dongen CJ, Vink R, Hutten BA, Büller HR, Prins MH. The incidence of recurrent venous

thromboembolism after treatment with vitamin K antagonists in relation to time since first event: a meta-analysis. Arch Intern Med 2003;163:1285-93.

24. Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet 2003;362:523-6.

25. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133:454S-545S.

26. Van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briët E. Bleeding complications in oral anticoagulant therapy. An analysis of risk factors. Arch Intern Med 1993;153:1557-62.

Thrombophilia and VTE: implications for testing

23

27. Palareti G, Leali N, Coccheri S et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy. Lancet 1996;348:423-8.

28. Margaglione M, D'Andrea G, Colaizzo D et al. Coexistence of factor V Leiden and Factor II A20210 mutations and recurrent venous thromboembolism. Thromb Haemost 1999;82:1583-7.

29. Van den Belt AG, Sanson BJ, Simioni P et al. Recurrence of venous thromboembolism in patients with familial thrombophilia. Arch Intern Med 1997;157:2227-32.

30. De Stefano V, Leone G, Mastrangelo S et al. Clinical manifestations and management of inherited thrombophilia: retrospective analysis and follow-up after diagnosis of 238 patients with congenital deficiency of antithrombin III, protein C, protein S. Thromb Haemost 1994;72:352-8.

31. Vink R, Kraaijenhagen RA, Levi M, Büller HR. Individualized duration of oral anticoagulant therapy for deep vein thrombosis based on a decision model. J Thromb Haemost 2003;1:2523-30.

32. Ho WK, Hankey GJ, Quinlan DJ, Eikelboom JW. Risk of recurrent venous thromboembolism in patients with common thrombophilia: a systematic review. Arch Intern Med 2006;166:729-36.

33. Van Hylckama Vlieg A, Baglin CA, Bare LA, Rosendaal FR, Baglin TP. Proof of principle of potential clinical utility of multiple SNP analysis for prediction of recurrent venous thrombosis. J Thromb Haemost 2008;6:751-4.

34. Legnani C, Benilde C, Michela C, Mirella F, Giuliana G, Gualtiero P. High plasma levels of factor VIII and risk of recurrence of venous thromboembolism. Br J Haematol 2004;124:504-10.

35. Kyrle PA, Minar E, Hirschl M et al. High plasma levels of factor VIII and the risk of recurrent venous thromboembolism. N Engl J Med 2000;343:457-62.

36. Eichinger S, Stumpflen A, Hirschl M et al. Hyperhomocysteinemia is a risk factor of recurrent venous thromboembolism. Thromb Haemost 1998;80:566-9.

37. Keijzer MB, Blom HJ, Bos GM, Willems HP, Gerrits WB, Rosendaal FR. Interaction between hyperhomocysteinemia, mutated methylenetetrahydrofolatereductase (MTHFR) and inherited thrombophilic factors in recurrent venous thrombosis. Thromb Haemost 2002;88:723-8.

38. Den Heijer M, Blom HJ, Gerrits WB et al. Is hyperhomocysteinaemia a risk factor for recurrent venous thrombosis? Lancet 1995;345:882-5.

39. Prandoni P, Simioni P, Girolami A. Antiphospholipid antibodies, recurrent thromboembolism, and intensity of warfarin anticoagulation. Thromb Haemost 1996;75:859.

40. Rance A, Emmerich J, Fiessinger JN. Anticardiolipin antibodies and recurrent thromboembolism. Thromb Haemost 1997;77:221-2.

41. Schulman S, Svenungsson E, Granqvist S. Anticardiolipin antibodies predict early recurrence of thromboembolism and death among patients with venous thromboembolism following anticoagulant therapy. Duration of Anticoagulation Study Group. Am J Med 1998;104:332-8.

42. De Godoy JM, De Godoy MF, Braile DM. Recurrent thrombosis in patients with deep vein thrombosis and/or venous thromboembolism associated with anticardiolipin antibodies. Angiology 2006;57:79-83.

43. Hull R, Hirsh J, Jay R et al. Different intensities of oral anticoagulant therapy in the treatment of proximal-vein thrombosis. N Engl J Med 1982;307:1676-81.

44. Kearon C, Ginsberg JS, Kovacs MJ et al. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin therapy for long-term prevention of recurrent venous thromboembolism. N Engl J Med 2003;349:631-9.

45. Crowther MA, Ginsberg JS, Julian J et al. A comparison of two intensities of warfarin for the prevention of recurrent thrombosis in patients with the antiphospholipid antibody syndrome. N Engl J Med 2003;349:1133-8.

46. Finazzi G, Marchioli R, Brancaccio V et al. A randomized clinical trial of high-intensity warfarin vs. conventional antithrombotic therapy for the prevention of recurrent thrombosis in patients with the antiphospholipid syndrome (WAPS)1. Journal of Thrombosis and Haemostasis 2005;3:848-53.

47. Hutten BA, Prins MH. Duration of treatment with vitamin K antagonists in symptomatic venous thromboembolism. Cochrane Database Syst Rev 2006;CD001367.

48. Simioni P, Sanson BJ, Prandoni P et al. Incidence of venous thromboembolism in families with inherited thrombophilia. Thromb Haemost 1999;81:198-202.

49. De Stefano V, Rossi E, Paciaroni K, Leone G. Screening for inherited thrombophilia: indications and therapeutic implications. Haematologica 2002;87:1095-108.

Chapter 2

24

1

2

3

4

5

6

7

8

9

10

11

12

50. Middeldorp S, Henkens CM, Koopman MM et al. The incidence of venous thromboembolism in family members of patients with factor V Leiden mutation and venous thrombosis. Ann Intern Med 1998;128:15-20.

51. Middeldorp S, Meinardi JR, Koopman MM et al. A prospective study of asymptomatic carriers of the factor V Leiden mutation to determine the incidence of venous thromboembolism. Ann Intern Med 2001;135:322-7.

52. Machin SJ. Pros and cons of thrombophilia testing: cons. J Thromb Haemost 2003;1:412-3. 53. Marchetti M, Quaglini S, Barosi G. Cost-effectiveness of screening and extended anticoagulation for carriers

of both factor V Leiden and prothrombin G20210A. QJM 2001;94:365-72. 54. Wu O, Robertson L, Twaddle S et al. Screening for thrombophilia in high-risk situations: systematic review

and cost-effectiveness analysis. The Thrombosis: Risk and Economic Assessment of Thrombophilia Screening (TREATS) study. Health Technol Assess 2006;10:1-110.

55. Korlaar IM, Vossen CY, Rosendaal FR et al. Attitudes toward genetic testing for thrombophilia in asymptomatic members of a large family with heritable protein C deficiency. Journal of Thrombosis and Haemostasis 2005;3:2437-44.

56. Legnani C, Razzaboni E, Gremigni P, Ricci Bitti PE, Favaretto E, Palareti G. Psychological impact of testing for thrombophilic alterations. Thromb Haemost 2006;96:348-55.

57. Bank I, Scavenius MP, Büller HR, Middeldorp S. Social aspects of genetic testing for factor V Leiden mutation in healthy individuals and their importance for daily practice. Thromb Res 2004;113:7-12.

58. Cohn DM, Middeldorp S. [A multicentre randomised clinical trial to evaluate the benefit of testing for thrombophilia following a first venous thromboembolism: the NOSTRADAMUS study]. Ned Tijdschr Geneeskd 2007;151:371-3.

59. Cohn DM, Middeldorp S. [Early termination of the multicentre randomised clinical trial to evaluate the benefit of testing for thrombophilia following a first venous thromboembolism: the NOSTRADAMUS study]. Ned Tijdschr Geneeskd 2008;152:2093-4.

60. Coppens M, Reijnders JH, Middeldorp S, Doggen CJ, Rosendaal FR. Testing for inherited thrombophilia does not reduce recurrence of venous thrombosis. J Thromb Haemost 2008.

61. Vossen CY, Conard J, Fontcuberta J et al. Risk of a first venous thrombotic event in carriers of a familial thrombophilic defect. The European Prospective Cohort on Thrombophilia (EPCOT). J Thromb Haemost 2005;3:459-64.

62. Bucciarelli P, Rosendaal FR, Tripodi A et al. Risk of venous thromboembolism and clinical manifestations in carriers of antithrombin, protein C, protein S deficiency, or activated protein C resistance: a multicenter collaborative family study. Arterioscler Thromb Vasc Biol 1999;19:1026-33.

63. Sanson BJ, Simioni P, Tormene D et al. The incidence of venous thromboembolism in asymptomatic carriers of a deficiency of antithrombin, protein C, or protein S: a prospective cohort study. Blood 1999;94:3702-6.

64. Faioni EM, Franchi F, Bucciarelli P et al. Coinheritance of the HR2 haplotype in the factor V gene confers an increased risk of venous thromboembolism to carriers of factor V R506Q (factor V Leiden). Blood 1999;94:3062-6.

65. Tormene D, Fortuna S, Tognin G et al. The incidence of venous thromboembolism in carriers of antithrombin, protein C or protein S deficiency associated with the HR2 haplotype of factor V: a family cohort study. Journal of Thrombosis and Haemostasis 2005;3:1414-20.

66. Heit JA, Sobell JL, Li H, Sommer SS. The incidence of venous thromboembolism among Factor V Leiden carriers: a community-based cohort study. Journal of Thrombosis and Haemostasis 2005;3:305-11.

67. Martinelli I, Bucciarelli P, Margaglione M, De Stefano V, Castaman G, Mannucci PM. The risk of venous thromboembolism in family members with mutations in the genes of factor V or prothrombin or both. Br J Haematol 2000;111:1223-9.

68. Bank I, Libourel EJ, Middeldorp S et al. Prothrombin 20210A Mutation: A Mild Risk Factor for Venous Thromboembolism but Not for Arterial Thrombotic Disease and Pregnancy-Related Complications in a Family Study. Arch Intern Med 2004;164:1932-7.

69. Coppens M, Van de Poel MH, Bank I et al. A prospective cohort study on the absolute incidence Of venous thromboembolism and arterial cardiovascular disease in asymptomatic carriers of the prothrombin 20210A mutation. Blood 2006;108:2604-7.

70. Bank I, Coppens M, Van de Poel MH et al. A prospective cohort study of asymptomatic individuals with elevated factor VIII:c to determine the absolute incidence of venous and arterial thromboembolism. J Thromb Haemost 2005;3 (Suppl.1):P1056.

Thrombophilia and VTE: implications for testing

25

71. Lijfering WM, Coppens M, Van de Poel MH et al. The risk of venous and arterial thrombosis in hyperhomocysteinaemia is low and mainly depends on concomitant thrombophilic defects. Thromb Haemost 2007;98:457-63.

72. Makelburg AB, Lijfering WM, Middeldorp S et al. Low absolute risk of venous and arterial thrombosis in hyperhomocysteinaemia - A prospective family cohort study in asymptomatic subjects. Thromb Haemost 2009;101:209-12.

Chapter 2

26

Chapter 3

Thrombophilia testing for prevention of recurrent venous thromboembolism

Cohn DM, Vansenne F, De Borgie CAJ, Middeldorp S. Thrombophilia testing for prevention of recurrent venous thromboembolism. Cochrane Database Syst Rev 2009;CD007069.

Abstract

Background

Tests for thrombophilia are being performed on a large scale in patients with venous

thromboembolism (VTE), even though the benefits of testing are still subject of debate. The

most important benefit would be reduction of the risk of recurrent VTE due to additional

prophylactic measures.

Objective

To systematically review the available evidence for testing for thrombophilia after VTE, in

terms of risk reduction of recurrent VTE.

Methods

We searched the Cochrane Peripheral Vascular Diseases (PVD) Group trials register (last

searched 22 April 2008), MEDLINE, EMBASE, and the CENTRAL database (last searched

2008, Issue 2), and reference lists. Randomized controlled trials (RCTs) and Controlled

clinical trials (CCTs) comparing the rate of recurrent VTE in patients with VTE who were

tested for thrombophilia and patients with VTE who were not tested for thrombophilia

were eligible. Data from identified studies was to be extracted and recorded on data

extraction forms to allow pooling for meta-analysis.

Results

No studies were included because no RCTs or CCTs could be identified.

Conclusion

There are no trials that assessed the benefit(s) of testing for thrombophilia on the risk of

recurrent VTE. RCTs are needed as tests for thrombophilia are being performed widely,

even though the benefits have not been demonstrated yet.

28

Chapter 3

1

2

3

4

5

6

7

8

9

10

11

12

Background

Thrombophilia is the term used to describe a hereditary or acquired pre-disposition to

venous thromboembolism (VTE). VTE may manifest as either pulmonary embolism or deep

vein thrombosis, or a combination of both. The term thrombophilia was first mentioned by

Egeberg in 1965, when he described a Norwegian family with a high tendency to

thrombosis due to a deficiency in the natural anticoagulant antithrombin.1 Subsequently, in

the 1980s, deficiencies of the other natural anticoagulants, protein C and protein S, were

found to increase the risk of VTE.2, 3

These deficiencies are rather uncommon, with a prevalence of <1% in the general

population4-6 and a prevalence of at most 5% among patients with thrombosis.7 During the

last two decades, newer and more prevalent thrombophilic defects have been discovered,

such as the factor V Leiden mutation which causes activated protein C resistance and the

prothrombin G20210A mutation.8, 9 Also, elevated levels of factor VIII have shown to be a

risk factor for thrombosis.10, 11 It has been found that (mild) hyperhomocysteinaemia is a

risk factor for VTE12, but its clinical relevance seems small, especially since lowering the

homocysteine level did not show a reduction in recurrence of VTE.13 As nowadays a

thrombophilic defect can be demonstrated in at least 50% of patients with a VTE14, 15, testing

patients with a first VTE for thrombophilia, has gained in interest tremendously. Potential

advances of testing patients might be the opportunity to elucidate the cause of the

thrombosis in patients or to track unaffected family members. On the other hand, testing

for thrombophilia has several potential disadvantages. The psychological consequences of

testing patients for thrombophilia might be considered as a drawback of testing. It is not

inconceivable that a patient's knowledge of being a carrier of a genetic risk factor might

influence his/her well being. In addition, a positive test result for thrombophilia testing

might cause problems with health or life insurance. The assessment whether a patient

should be tested should mainly depend on the feasibility to reduce the risk of recurrent VTE

(for example, by prolonging the duration of anticoagulant treatment). To assess whether

testing for thrombophilia reduces the risk of recurrent VTE, a systematic literature search

was undertaken. If possible, we aimed to perform a meta-analysis to assess the overall

reduction in recurrent VTE after testing for thrombophilia.

29

Thrombophilia testing for prevention of recurrent VTE

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials (RCTs) and controlled clinical trials (CCTs) comparing the

rate of recurrent VTE in patients with existing VTE tested for thrombophilia and patients

who were not tested for thrombophilia were considered eligible.

Types of participants

Patients with VTE (either deep venous thrombosis or pulmonary embolism, or both).

Types of interventions

The investigated intervention was testing for thrombophilia. Thrombophilia was defined as:

� antithrombin deficiency;

� protein C deficiency;

� protein S deficiency;

� factor V Leiden mutation;

� factor II mutation (prothrombin mutation);

� (mild) hyperhomocysteinaemia;

� persistently elevated levels of clotting factor VIII:c;

� presence of antiphospholipid or anticardiolipin antibodies, or lupus anticoagulant.

Types of outcome measures

The primary outcome measure was recurrent VTE. Secondary outcome measures included:

� major bleeding;

� clinically relevant non-major bleeding

� quality of life

30

Chapter 3

1

2

3

4

5

6

7

8

9

10

11

12

Search methods for identification of studies

The Cochrane Peripheral Vascular Diseases (PVD) Group searched their Trials Register

(last searched 22 April 2008) and the Cochrane Central Register of Controlled Trials

(CENTRAL) in The Cochrane Library (last searched 2008, Issue 2) for publications

describing randomized controlled trials (RCTs) or controlled clinical trials (CCTs). The

PVD Group's Trials Register was compiled from electronic searches of MEDLINE (1966 to

date), EMBASE (1980 to date), and CINAHL (1982 to date), and through hand searching

relevant journals. The full list of journals that have been hand searched, as well as the

search strategies used are described in the 'Search strategies for the identification of

studies' section within the editorial information about the Cochrane PVD Group in The

Cochrane Library.

We searched for potentially eligible articles in the MEDLINE, EMBASE and CENTRAL

databases. Highly sensitive search filters were used for the identification of RCTs or CCTs.

Table 1. Highly sensitive search filter for MEDLINE

MEDLINE highly sensitive search filter #1 randomized controlled trial [pt] #2 controlled clinical trial [pt] #3 randomized [tiab] #4 placebo [tiab] #5 drug therapy [sh] #6 randomly [tiab] #7 trial [tiab] #8 groups [tiab] #9 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 #10 humans [mh] #11 #9 and #10

We used the "Cochrane Highly Sensitive Search Strategy: sensitivity-maximizing version

(2008 revision)” filter, published at: http://www.cochrane-handbook.org (Chapter 6.4.11

(search filter 6.4a). for identifying randomized trials in MEDLINE (Table 1). For EMBASE,

we used the highly sensitive search filter described in the Peripheral Vascular Diseases

Group's module in The Cochrane Library (Table 2).

31

Thrombophilia testing for prevention of VTE

Table 2. Highly sensitive search filter for EMBASE

EMBASE highly sensitive search filter 1. random$.ti,ab. 2. factorial$.ti,ab. 3. (crossover$ or cross over$ or cross-over$).ti,ab. 4. placebo$.ti,ab. 5. (doubl$ adj blind$).ti,ab. 6. (singl$ adj blind$).ti,ab. 7. assign$.ti,ab. 8. allocat$.ti,ab. 9. volunteer$.ti,ab. 10. CROSSOVER PROCEDURE/ 11. DOUBLE-BLIND PROCEDURE/ 12. RANDOMIZED CONTROLLED TRIAL/ 13. SINGLE-BLIND PROCEDURE/ 14. or/1-13 15. exp ANIMAL/ or NONHUMAN/ or exp ANIMAL EXPERIMENT/ 16. exp HUMAN/ 17. 16 and 15 18. 15 not 17 19. 14 not 18

The following search terms, structured as PI(C)O, were used as MeSH and EMTREE terms

and free text:

Patients: thromboembolism; venous thrombosis; pulmonary embolism. Intervention:

thrombophilia; prothrombin; protein C deficiency; protein S deficiency; antithrombin III

deficiency; activated protein C resistance; factor VIII; lupus coagulation inhibitor;

antibodies, antiphospholipid; antibodies, anticardiolipin; thrombophil*;

hypercoagulabil*; at III; antithrombin; protein C; protein S; apc resistance; factor V;

antiphospholipid antibody; anticardiolipin antibody; hyperhomocysteinaemia; homocyst*;

hyperhomocyst*. The term "factor V/genetics" is only searched as a MeSH term.

Comparison: no specific search terms. Outcome: recurrent VTE. The following words were

searched in free text: relapse; recrudescence; recurr*.

The restrictions "pregnancy loss" and "hemophilia" were applied to filter out articles

focussing on the association between thrombophilia and recurrent pregnancy loss and

articles on hemophilia.

Chapter 3

32

1

2

3

4

5

6

7

8

9

10

11

12

Data collection and analysis

Selection of trials

Two authors (DC and FV) independently screened the titles, abstracts and subsequently

review the text of articles that appeared to be eligible. We searched the reference sections of

relevant papers for additional articles. The criteria for selection of trials were as specified in

the above section 'Criteria for considering studies for this review'.

Quality of trials

Two authors (DC and FV) independently assessed the quality of the methods used in the

trials based on methods of randomization. Disagreements were resolved by discussion.

Only trials that explicitly affirmed a randomization process were considered eligible. We

coded allocation of concealment as adequate (A), unclear (B), inadequate (C) or not used

(D), as described in the Cochrane Handbook for Systematic Reviews of Interventions

(Cochrane Handbook). Only trials with adequate allocation of concealment were

considered eligible. Quality of blinding in this case was not applicable.

Data extraction

Two authors (DC and FV) independently extracted data. Disagreements were resolved by

discussion. We recorded the collected data on data extraction forms. If necessary, we

contacted the trialists for additional data.

Statistical analysis

If possible, we aimed to perform a meta-analysis to assess the overall effect of testing for

thrombophilia on recurrent VTE. Variance would be calculated using the Peto odds ratio,

since this method was considered the most appropriate for dichotomous, sparse data

without a large effect. Heterogeneity would be calculated by the I-square test and trials were

considered heterogenic if I-square > 50%.

Results

The search trough the Trial Register revealed no eligible trials. The above mentioned search

strategies in MEDLINE and EMBASE yielded 556 and 123 hits, respectively. After screening

for title we excluded 523 of the MEDLINE articles. Reasons for exclusion were: review

article (95), cohort study (18); case report or case series (141), other outcome measure/other

33

Thrombophilia testing for prevention of recurrent VTE

focus (256) and editorial/rebuttal (13). Of the 33 remaining articles we screened the

abstracts. All these studies were excluded, as 11 were reviews, 6 were cohort studies, 1 was a

case report, 13 focused on other outcome measures and 1 was an editorial. We identified

one outline of an intervention trial that fulfilled the inclusion criteria.16, 17 However, this trial

(our trial) has recently been stopped early due to a low inclusion rate.18 No conclusions can

be drawn from the 23 patients that were included in this intervention trial before it was

stopped. From EMBASE, 123 hits were identified, of which at least 47 overlapped with the

search in the MEDLINE database. Of the remaining 76 hits, 22 were excluded because they

were review articles, 1 was a case report, and 49 had a different outcome measure/other

focus. The abstracts of the remaining 3 articles were screened but had to be excluded as 2

articles were reviews and 1 had another focus. From a search through CENTRAL we

identified 20 potential trials, of which 19 were not relevant. The remaining trial was the

above mentioned NOSTRADAMUS trial.16, 17

Discussion

We did not identify completed trials in which testing for thrombophilia was the

intervention and recurrent VTE was the outcome measure. Most of the published studies on

thrombophilia focused on the prevalence of various thrombophilic defects and lacked the

appropriate design. The only trial that assessed the potential benefits and disadvantages of

testing for thrombophilia was terminated early due to slow inclusion rates.18

Implications for practice

There is no information available from RCTs or CCTs on the benefits of thrombophilia

testing to reduce the risk of recurrent VTE.

Implications for research

As tests for thrombophilia are being performed widely, even though the benefits have not

been demonstrated yet, RCTs are needed. A useful design would be randomization between

disclosure and no disclosure of the thrombophilia test results (for both treating physicians

and participants). Those patients allocated to the "disclosure" group with a thrombophilic

defect should receive a more intense treatment (such as prolongation of the anticoagulant

therapy). The primary endpoint would be the composite of recurrent VTE and/or bleeding.

34

Chapter 3

1

2

3

4

5

6

7

8

9

10

11

12

References 1. Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh

1965;13:516-30. 2. Mannucci PM, Vigano S. Deficiencies of protein C, an inhibitor of blood coagulation. Lancet 1982;2:463-7. 3. Schwarz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S deficiency in familial

thrombotic disease. Blood 1984;64:1297-300. 4. Miletich J, Sherman L, Broze G, Jr. Absence of thrombosis in subjects with heterozygous protein C deficiency.

N Engl J Med 1987;317:991-6. 5. Tait RC, Walker ID, Perry DJ et al. Prevalence of antithrombin deficiency in the healthy population. Br J

Haematol 1994;87:106-12. 6. Tait RC, Walker ID, Reitsma PH et al. Prevalence of protein C deficiency in the healthy population. Thromb

Haemost 1995;73:87-93. 7. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical

factors, and recurrent venous thrombotic events. JAMA 2005;293:2352-61. 8. Bertina RM, Koeleman BP, Koster T et al. Mutation in blood coagulation factor V associated with resistance

to activated protein C. Nature 1994;369:64-7. 9. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3'-untranslated region

of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698-703.

10. Koster T, Blann AD, Briët E, Vandenbroucke JP, Rosendaal FR. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 1995;345:152-5.

11. Kraaijenhagen RA, In 't Anker PS, Koopman MM et al. High plasma concentration of factor VIIIc is a major risk factor for venous thromboembolism. Thromb Haemost 2000;83:5-9.

12. Den Heijer M, Koster T, Blom HJ et al. Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. N Engl J Med 1996;334:759-62.

13. Den Heijer M, Willems HP, Blom HJ et al. Homocysteine lowering by B vitamins and the secondary prevention of deep-vein thrombosis and pulmonary embolism. A randomized, placebo-controlled, double blind trial. Blood 2007;109:139-44.

14. Bauer KA. The thrombophilias: well-defined risk factors with uncertain therapeutic implications. Ann Intern Med 2001;135:367-73.

15. Lane DA, Mannucci PM, Bauer KA et al. Inherited thrombophilia: Part 2. Thromb Haemost 1996;76:824-34. 16. Cohn DM, Middeldorp S. [A multicentre randomised clinical trial to evaluate the benefit of testing for

thrombophilia following a first venous thromboembolism: the NOSTRADAMUS study]. Ned Tijdschr Geneeskd 2007;151:371-3.

17. Cohn DM, Middeldorp S. Necessity of screening for thrombophilia at diagnosis of venous thromboembolism: outline of the NOSTRADAMUS intervention trial. Journal of Thrombosis and Haemostasis 2007;2007:P-M-470.

18. Cohn DM, Middeldorp S. [Early termination of the multicentre randomised clinical trial to evaluate the benefit of testing for thrombophilia following a first venous thromboembolism: the NOSTRADAMUS study]. Ned Tijdschr Geneeskd 2008;152:2093-4.

Thrombophilia testing for prevention of recurrent VTE

35

Chapter 4

The psychological impact of testing for thrombophilia: a systematic review

Cohn DM, Vansenne F, Kaptein AA, De Borgie CA, Middeldorp S. The psychological impact of testing for thrombophilia: a systematic review. J Thromb Haemost 2008; 6:1099-104.

Abstract

Background

Large numbers of patients are tested for thrombophilia nowadays, even though the benefits

are unclear. A potential disadvantage of this (mainly) genetic testing might be its

psychological impact, such as fear, depression and worry.

Objectives

To systematically review studies that determined the nature and extent of psychological

impact of testing for thrombophilia.

Methods

We searched the MEDLINE database (1966 to February 2008), the EMBASE database (1985

to 2008, week 5) and the PsychInfo database (1806 to February 2008) for relevant trials,

without language restrictions. Bibliographies of relevant articles were scanned for

additional articles. We reviewed all relevant studies that focused on the psychological

impact of testing for thrombophilia. Only full papers of studies that included 15 patients or

more were considered eligible for this review. Two reviewers independently extracted data

and assessed quality.

Results

Six studies fulfilled the eligibility criteria. As these studies varied tremendously in

methodology; pooling of the data was not possible. Studies of psychological impact of

genetic testing for thrombophilia report few negative results, but most assessments are

limited to the short term or lack methodological accuracy.

Conclusion

No valid conclusions can be drawn about the psychological impact of genetic testing in

patients based on the current available literature. Given the large number of patients that

are being exposed to thrombophilia testing, and the uncertain benefits, there is an urgent

need for more uniformity in the measurement of psychological impact of thrombophilia

testing.

Chapter 4

38

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Venous thromboembolism (VTE) commonly manifests as deep vein thrombosis or

pulmonary embolism, or a combination of both. VTE affects approximately 2-3/1000

persons each year.1-3

It is a multifactorial disease, in which both exogenous and endogenous conditions are

known to increase the risk.4 Established exogenous risk factors are cancer, pregnancy,

puerperium, surgery, immobilisation and oral contraceptive use. The endogenous risk

factors for VTE contribute to the term “thrombophilia”. Thrombophilia has gained interest

since a growing number of ‘common’ abnormalities were discovered in the 1990’s, such as

factor V Leiden and the prothrombin mutation.5,6 Due to the discovery of these more

prevalent thrombophilias, in approximately 50-60% of patients with VTE at least one

thrombophilic defect can be demonstrated nowadays.7 Thrombophilic abnormalities can

be acquired or inherited. The most common acquired thrombophilic abnormality is the

antiphospholipid antibody syndrome. Inherited forms of thrombophilia are deficiencies of

the natural anticoagulants (antithrombin, protein C and protein S) and the more common

mutations: factor V Leiden (FVL) and prothrombin 20210A. Moreover, persistently elevated

levels of FVIII:c increase the risk of VTE; this condition is at least partially hereditary.8

Whether (mild) hyperhomocysteinaemia should be considered a risk factor for venous

thrombosis remains controversial.9,10

Thrombophilia tests are also being performed within the scope of other disorders with

which it is associated, such as recurrent miscarriage and intra-uterine fetal death, but also

for conditions for which no association has been established, such as arterial diseases

including stroke. Finally, healthy individuals are often tested for thrombophilia, usually

within the context of family testing after diagnosing thrombophilia in an affected

proband.11

The question whether testing for thrombophilia is justified is still a matter of debate and

can not be answered unequivocally.12-14 Potential advantages of testing may be the possibility

to take preventive measures or to avoid exposure to evitable risk factors. However, in

accordance to other genetic tests, a potential drawback of testing for thrombophilia could

be social consequences, such as problems with acquiring life or disability insurances.15

Moreover, results of genetic tests may have substantial psychological consequences, such

as depression, anxiety, and persistent fear.16,17 Not all psychological consequences following

genetic testing are negative; testing may also lead to relief and fear reduction.18 As the

benefits and disadvantages of testing for thrombophilia are not fully clear, the

The psychological impact of testing for thrombophilia: a systematic review

39

psychological consequences of testing for thrombophilia should also be considered in the

decision whether testing for thrombophilia is indicated or can be justified.

Here we systematically review the studies on psychological consequences of testing for

thrombophilia. Furthermore, we critically appraise the methods of assessment of

psychological impact in the reviewed studies.

Methods

All full papers that reported the psychological impact of testing for thrombophilia were

eligible. We considered all study designs and all indications for testing eligible, as long as

at least 15 individuals were included. The primary outcome was the degree of psychological

impact of testing for thrombophilia. Secondary outcome measures were descriptions of the

degree of satisfaction after testing for thrombophilia and uptake of behavioural changes

after disclosure. Potential eligible articles were systematically searched in the MEDLINE

(1966 to February 2008), EMBASE (1985 to 2008, week 5) and PsychInfo databases (1806 to

February 2008). The used search strategy was highly sensitive, without language

restrictions. One restriction was made: “NOT hemophilia”, since there was considerable

overlap between “Factor VIII” AND “hemophilia”. The used search terms (structured as

PICO) are stated in table 1.

Table 1. Literature search terms

search terms Patients no specific search terms Intervention thrombophilia; thrombophil*; prothrombin; hyperhomocysteinemia; homocystein; hyperhomocyst*;

homocystein*; protein c deficiency; protein s deficiency; antithrombin deficiency; activated protein c resistance; “factor V/genetics” [MeSH]; hypercoagulabil*; “apc resistance”; “factor V”; “protein c”; “protein s”; antithrombin; “at III”; F VIII; factor VIII; thromboplastinogen

Comparison no specific search terms Outcomes “psychology”[Subheading]; anxiety; anxieties[tw]; nervousness[tw]; stresses[tw]; distress[tw];

psycholog*[tw]; psychosocial[tw]; worry[tw]; worrying[tw]; “stress, psychological”[MeSH]; anguish*[tw]; “mental suffering”; distress*; emotion*[tw]; “Stress”[MeSH]; depressive[tw]; “Depression”[MeSH]; worries[tw], “risk perception”, “illness cognition”, “cognition”, “health beliefs”, “lay beliefs”, “treatment beliefs”, “emotion regulation” and “common sense model”

They were used both as MESH terms and free text (unless noted otherwise). Bibliographies

of relevant articles were scanned for additional articles. In case more publications

concerning the same participants suited the search terms, only the most relevant article was

included. Two reviewers (DC, FV) independently screened the results of searches to identify

potentially relevant papers, and independently extracted the data from each paper using the

40

Chapter 4

1

2

3

4

5

6

7

8

9

10

11

12

same form. Data extraction was checked and discrepancies were resolved through

discussion. If necessary, we contacted authors of selected papers to provide additional data.

We aimed to pool the results in a meta-analysis in case of sufficient comparability of the

used methods, or otherwise to assess the psychological consequences of testing for

thrombophilia qualitatively.

Results

A search in Medline yielded 1823 potentially relevant studies. Given the substantial

heterogeneity of interventions and methods across studies, it was not possible to pool the

results for meta-analysis.

After screening for title and abstract, 6 relevant studies were identified.19-24 Searches in the

EMBASE and PsychInfo databases yielded 1301 and 387 hits, respectively, but after

screening for title and abstract no additional relevant articles were found.

Table 2 summarises the included studies. Most participants reported to be satisfied with

the knowledge of being carrier of (a) thrombophilia (respectively 97/110 (88%) and 202/215

(94%) and testing for thrombophilia was perceived as less harmful than a genetic test for

breast cancer or a non-genetic test for diabetes.24 However, a substantial number of

participants reported to be more worried (respectively 47/110 (43%) and 58/215 (27%).19,20

Furthermore, thrombophilia testing might be associated with negative effects, as reported

in a small, qualitative study: asymptomatic carriership of FVL can influence daily life by

having concerns, stigmatisation and problems with insurance eligibility.21 It is of note that

43/110 (39%) of the participants in one study did not recall to have consented for their

genetic test and that 14/110 (13%) of the participants were unaware of their abnormal test

result.19

The number of participants ranged from 17 to 168 per study. Four-hundred of 692

participants were tested for research purposes in three studies.20-22 The selected participants

of two of these studies were a subset of a large group of individuals that were tested

previously to assess the incidence and thrombotic risk of carriership of FVL.25,26 The

participants of the third study had been tested for protein C deficiency to clarify the high

frequency of VTE in their kindred.27

The greater part (266/692; 38%) of all participants was apparently healthy and never

experienced VTE. One third (227/692; 33%) were asymptomatic relatives of probands with

VTE. Only 160/692 (23%) of the participants experienced a VTE. The VTE or thrombophilia

history in the remaining 6% was unknown.

The psychological impact of testing for thrombophilia: a systematic review

41

Tabl

e 2. O

verv

iew

of t

he in

clude

d st

udie

s

pa

rtic

ipan

ts

setti

ng

thro

mbo

phili

c def

ects

in

stru

men

ts

poin

t in

time

outc

ome

Hel

lman

n 20

0319

11

0 con

secu

tive i

ndivi

dual

s 83

per

sona

l hist

ory o

f VTE

27

reas

on fo

r tes

ting

unkn

own

clini

cal p

urpo

ses

fact

or V

Leid

en

1 not

valid

ated

que

stio

nnai

re,

base

d on

pre

vious

pu

blica

tions

conc

erni

ng o

ther

ge

netic

test

s

mos

tly se

vera

l ye

ars a

fter

disc

losu

re o

f te

st re

sults

know

ledge

of g

enet

ic st

atus

incr

ease

d aw

aren

ess

of th

rom

botic

risk

, but

the m

agni

tude

of t

he ri

sk

is of

ten

over

estim

ated

kn

owled

ge of

fact

or V

Leid

en st

atus

incr

ease

d wo

rry i

n 43

% o

f the

par

ticip

ants

, alth

ough

88%

of

all p

artic

ipan

ts w

as g

lad

to kn

ow th

e out

com

e

Lind

qvis

t 20

0320

4 p

erso

nal h

istor

y of V

TE*

211 h

ealth

y con

trols

* repo

rted

in

corr

espo

nden

ce b

y th

e au

thor

s

rese

arch

pur

pose

s:

to as

sess

the i

ncid

ence

of

APC

resis

tanc

e am

ongs

t pre

gnan

t wo

men

fact

or V

Leid

en*

in ca

se o

f alte

red

test

resu

lt of

AP

C re

sista

nce

* repo

rted

in co

rres

pond

ence

by

the

auth

ors

2 not

valid

ated

que

stio

nnai

res

rega

rdin

g sa

tisfa

ctio

n, th

e aw

aren

ess a

nd b

ehav

iour

af

ter r

eceiv

ing

a pos

itive

test

re

sult

6-12

mon

ths

afte

r disc

losu

re

of te

st re

sults

94%

was

satis

fied

with

the a

ware

ness

of b

eing

APC-

resis

tant

27

% d

ecla

red

to b

e mor

e wor

ried

Bank

20

0421

17

asym

ptom

atic

rela

tives

of

indi

vidua

ls wi

th fa

ctor

V

Leid

en

rese

arch

pur

pose

s: to

asse

ss th

e inc

iden

ce

of V

TE in

indi

vidua

ls wi

th th

rom

boph

ilia

fact

or V

Leid

en

qual

itativ

e,

sem

i-stru

ctur

ed in

terv

iews

4 to

7 yea

rs

afte

r disc

losu

re

of te

st re

sults

asym

ptom

atic

carr

iersh

ip o

f fac

tor V

Leid

en

mig

ht in

fluen

ce d

aily

life b

y con

cern

s, st

igm

atisa

tion

and

prob

lems w

ith in

sura

nce

eligi

bilit

y

Van

Korla

ar

2005

22

168 f

amily

mem

bers

of o

ne

kind

red

with

a hi

gh

incid

ence

of p

rote

in C

de

ficien

cy

rese

arch

pur

pose

s: to

asse

ss th

e he

ritab

ility

of a

rare

pr

otein

C d

efici

ency

prot

ein C

def

icien

cy

valid

ated

risk

per

cept

ion

and

worr

y sca

les

and

valid

ated

trai

t anx

iety

(STA

I) qu

estio

nnai

re

attit

udes

abou

t tes

ting

mos

tly te

n ye

ars

afte

r disc

losu

re

of te

st re

sults

risk p

erce

ptio

n an

d wo

rry i

ncre

ased

in

indi

vidua

ls wi

th p

rote

in C

def

icien

cy

no si

gnifi

cant

diff

eren

ces i

n at

titud

es ab

out

gene

tic te

stin

g

Sauk

ko

2006

23

42 p

artic

ipan

ts

10 p

erso

nal h

istor

y of V

TE

20 fa

mily

hist

ory o

f VTE

or

thro

mbo

phili

a 12

oth

er re

ason

or

unkn

own

clini

cal p

urpo

ses

fact

or V

Leid

en

prot

hrom

bin

mut

atio

n*

prot

ein S

defic

iency

* pr

otein

C d

efici

ency

* an

tithr

ombi

n de

ficien

cy*

*s sel

f rep

orte

d by

the

part

icip

ants

qual

itativ

e,

sem

i-stru

ctur

ed in

terv

iews

at m

ost t

wo

year

s afte

r te

stin

g fo

r th

rom

boph

ilia

test

ing

for t

hrom

boph

ilia w

as g

ener

ally

cons

ider

ed to

be l

ess s

erio

us th

an a

gene

tic te

st

for b

reas

t can

cer o

r a n

on-g

enet

ic te

st fo

r di

abet

es

Legn

ani

2006

24

140 p

artic

ipan

ts

63 p

erso

nal h

istor

y of V

TE

22 fa

mily

hist

ory o

f VTE

or

thro

mbo

phili

a 55

appa

rent

ly he

alth

y in

divid

uals

clini

cal p

urpo

ses

fact

or V

Leid

en

prot

hrom

bin

mut

atio

n pr

otein

S de

ficien

cy

prot

ein C

def

icien

cy

antit

hrom

bin

defic

iency

hy

perh

omoc

yste

inae

mia

lu

pus a

ntico

agul

ant

Perc

eived

Hea

lth Sc

ore

and

valid

ated

CBA

scal

e A&B

qu

estio

nnai

re

prio

r to t

estin

g

and

20

day

s afte

r di

sclo

sure

of

test

resu

lts

no (s

igni

fican

t) ha

rmfu

l effe

cts o

f gen

etic

test

ing

in in

divid

uals

with

thro

mbo

phili

a a n

on-s

igni

fican

t dec

reas

e of P

erce

ived

Heal

th

Scor

e in

the s

ubjec

t s w

ithou

t a p

erso

nal h

istor

y of

VTE

42

Chapter 4

1

2

3

4

5

6

7

8

9

10

11

12

characteristics of individuals tested for clinical purposes (N=292)

14%

19%

13%

54%

personal history of VTErelative with VTE / thrombophiliahealthy controlunknown

characteristics of individuals tested for research purposes (N=400)

1%

46%

53%

Figure 1. Characteristics of individuals tested for clinical and research purposes

The characteristics of individuals tested for clinical purposes differed from those tested for

research purposes (Figure 1).

Almost all studies assessed the impact between 6 months and 10 years or more after

testing.19-23 One study (140 individuals) had a prospective design with a baseline

measurement.24

Two studies used qualitative, semi-structured interviews21,23, two studies used validated

questionnaires22,24 and two studies used non-validated questionnaires.19,20 The

thrombophilic tests that participants had undergone varied: FVL19-21, protein C deficiency

(without mutation analysis, but with awareness of its inheritable nature)22 and a set of

thrombophilic defects.23,24

Discussion

Following a highly sensitive search strategy through the main literature databases, only 6

studies reporting the psychological impact of thrombophilia testing could be identified. As

the applied methods varied considerably amongst these studies, it was impossible to pool

the results for meta-analysis. Therefore, we described the results of the individual studies.

The large majority of participants reported to be satisfied by the knowledge of being a

carrier of thrombophilia (88% and 94%), however several participants reported to be more

worried (27% and 43%). Qualitative studies showed that participants experienced testing

for thrombophilia not as serious as a genetic test for heritable cancer or a non-genetic test

for diabetes, although other participants reported that positive testing for FVL could have

led to stigmatisation and problems related to insurance. However, given the large

heterogeneity between the studies, no valid conclusion can be drawn. This heterogeneity

43

The psychological impact of testing for thrombophilia: a systematic review

refers to large differences in selection of individuals, instruments and points in time, which

will be discussed in detail below.

The results are difficult to extrapolate to daily clinical practice for two reasons. Firstly, in

58% of the participants thrombophilia tests were performed for research purposes. It is

likely that the way testing is performed and how tested individuals are informed about

indication and results differs considerably between settings. Secondly, the majority of the

participants did not have a personal history of VTE (71%). It is probable that results of

thrombophilia testing have different impact on symptomatic carriers compared to healthy

individuals. This was observed in the study by Legnani et al.: a (non-significant decrease) of

perceived health score was found in relatives and controls tested positively for

thrombophilia compared to baseline, but not in symptomatic carriers.24

The method of counselling or information provision on the test results affects the way

these results are perceived by the patient.28 Remarkably, none of the studies takes this into

consideration. Three of the included studies did not mention the way counselling had

transpired at all19,22,23, two studies informed the participants only in writing20,24 and one study

reported that “[participants] have been informed the same way and seen in the same place

as would be individuals for purposes of clinical practice”, without providing details.21

The applied instruments were in depth interviews, validated questionnaires and non-

validated questionnaires. The major advantage of application of standardised evaluation

approaches, such as validated questionnaires, is the possibility of pooling of the results to

perform meta-analyses and the possibility of comparison of different genetic tests. Genetic

tests for varying conditions may have a different impact, which might depend on e.g. the

severity of the underlying disease or the possible therapeutic options. At present, there are

no papers that compare different validated questionnaires in the context of the

psychological impact of genetic testing, but there are a few validated questionnaires that

have been used several times in genetic evaluation studies.29,30

The points in time of assessment differed substantially amongst all studies (between 20

days and 10 years or more after disclosure of the test results). Except one study, all studies

had a retrospective design. Hence, no baseline measurements could be performed in the

latter studies, because participants were already informed on their test results. The

interpretation of these results is insofar confined, that it is impossible to discriminate

between the impact of being ill and the impact of receiving an altered test result. A design

in which a baseline measurement is included allows this discrimination. Other conditions

in which the impact of genetic testing was assessed applied a prospective design (such as

Chapter 4

44

1

2

3

4

5

6

7

8

9

10

11

12

Huntington’s disease, Familial Hemochromatosis, Hereditary Breast and Ovarian Cancer

and Familial Adenomatous Polyposis).31-34

Besides the application of a baseline measurement, a serial assessment of impact of genetic

testing provides more insight in the impact of testing, especially since psychological

outcomes might change over time. Shortly after receiving the (positive) genetic test result

participants may feel anxious and shocked, leading to changes on psychological scales.35,36

This initial reaction may change after a while as participants think about their result and

incorporate this into their lives. After several years people appear to have adapted their lives

and thus may not even remember being shocked or anxious after disclosure of the test

result.37 This phenomenon is also known as response shift.22 Repeated measurements and a

long term follow-up after the genetic test result might improve the insight into this

phenomenon.

A potential drawback of our systematic review (and systematic reviews in general) might be

the possible occurrence of publication bias: those studies with significant, positive, results

are easier to find than those with non-significant or 'negative' results.38 Nevertheless, as it

is concluded that testing for thrombophilia appears to have no major adverse effects, and,

taken into consideration that this result might well be an overestimation, it seems unlikely

that publication bias affected our results gravely.

Conclusions and recommendations for future research

Overall, we conclude that the external validity of the studies is too much confined to

extrapolate the results on psychological impact of testing for thrombophilia to current

practice. The published data lack uniformity in various areas, such as types of participants

and counselling methods, points in time of measurement(s) and applied instruments. For

this reason, the external validity is seriously limited.

Given the large number of individuals tested yearly for thrombophilia, together with

uncertain benefits of testing, there is a need for more research into the psychological

impact of genetic testing for thrombophilia.

For future research, we would like to make some recommendations. Ideally, a randomised

clinical trial between testing and no testing could be performed to assess psychological

impact of testing for thrombophilia on top of receiving the diagnosis of VTE. Moreover,

this study design limits the potential effects of bias and confounding. However, such a

study is difficult to perform and recently, we had to stop our study due to slow

recruitment.39

The psychological impact of testing for thrombophilia: a systematic review

45

In any type of research, either interventional or observational, we would recommend the

application of validated questionnaires as they facilitate comparability within patients with

a specific condition and between diseases. Furthermore, completion of these

questionnaires both before and after disclosure of the test results, at short term and long

term, allows clarification of the course of potential harmful psychological effects.

References

1. Oger E. Incidence of venous thromboembolism: a community-based study in Western France. EPI-GETBP Study Group. Groupe d'Etude de la Thrombose de Bretagne Occidentale. Thromb Haemost 2000;83:657-60.

2. Hansson PO, Welin L, Tibblin G, Eriksson H. Deep vein thrombosis and pulmonary embolism in the general population. 'The Study of Men Born in 1913'. Arch Intern Med 1997;157:1665-70.

3. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and mortality of venous thrombosis: a population-based study. J Thromb Haemost 2007;5:692-9.

4. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999;353:1167-73. 5. Bertina RM, Koeleman BP, Koster T et al. Mutation in blood coagulation factor V associated with resistance

to activated protein C. Nature 1994;369:64-7. 6. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3'-untranslated region

of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698-703.

7. Lensing AW, Prandoni P, Prins MH, Büller HR. Deep-vein thrombosis. Lancet 1999;353:479-85. 8. Kamphuisen PW, Houwing-Duistermaat JJ, Van Houwelingen HC, Eikenboom JC, Bertina RM, Rosendaal

FR. Familial clustering of factor VIII and von Willebrand factor levels. Thromb Haemost 1998;79:323-7. 9. den Heijer M, Lewington S, Clarke R. Homocysteine, MTHFR and risk of venous thrombosis: a meta-

analysis of published epidemiological studies. J Thromb Haemost 2005;3:292-9. 10. Bezemer ID, Doggen CJ, Vos HL, Rosendaal FR. No association between the common MTHFR 677C->T

polymorphism and venous thrombosis: results from the MEGA study. Arch Intern Med 2007;167:497-501. 11. Coppens M, Van Mourik JA, Eckmann CM, Buller HR, Middeldorp S. Current practise of testing for inherited

thrombophilia. J Thromb Haemost 2007;5:1979-81. 12. Martinelli I. Pros and cons of thrombophilia testing: pros. J Thromb Haemost 2003;1:410-1. 13. Cohn DM, Roshani S, Middeldorp S. Thrombophilia and venous thromboembolism: implications for

testing. Semin Thromb Hemost 2007;33:573-81. 14. Machin SJ. Pros and cons of thrombophilia testing: cons. J Thromb Haemost 2003;1:412-3. 15. Homsma SJ, Huijgen R, Middeldorp S, Sijbrands EJ, Kastelein JJ. Molecular screening for familial

hypercholesterolaemia: consequences for life and disability insurance. Eur J Hum Genet 2007. 16. Green MJ, Botkin JR. "Genetic exceptionalism" in medicine: clarifying the differences between genetic and

nongenetic tests. Ann Intern Med 2003;138:571-5. 17. Lerman C, Croyle RT, Tercyak KP, Hamann H. Genetic testing: psychological aspects and implications. J

Consult Clin Psychol 2002;70:784-97. 18. Lim J, Macluran M, Price M, Bennett B, Butow P. Short- and long-term impact of receiving genetic mutation

results in women at increased risk for hereditary breast cancer. J Genet Couns 2004;13:115-33. 19. Hellmann EA, Leslie ND, Moll S. Knowledge and educational needs of individuals with the factor V Leiden

mutation. J Thromb Haemost 2003;1:2335-9. 20. Lindqvist PG, Dahlback B. Reactions to awareness of activated protein C resistance carriership: a descriptive

study of 270 women. Acta Obstet Gynecol Scand 2003;82:467-70. 21. Bank I, Scavenius MP, Büller HR, Middeldorp S. Social aspects of genetic testing for factor V Leiden

mutation in healthy individuals and their importance for daily practice. Thromb Res 2004;113:7-12. 22. Korlaar IM, Vossen CY, Rosendaal FR et al. Attitudes toward genetic testing for thrombophilia in

asymptomatic members of a large family with heritable protein C deficiency. Journal of Thrombosis and Haemostasis 2005;3:2437-44.

Chapter 4

46

1

2

3

4

5

6

7

8

9

10

11

12

23. Saukko PM, Richards SH, Shepherd MH, Campbell JL. Are genetic tests exceptional? Lessons from a qualitative study on thrombophilia. Soc Sci Med 2006;63:1947-59.

24. Legnani C, Razzaboni E, Gremigni P, Ricci Bitti PE, Favaretto E, Palareti G. Psychological impact of testing for thrombophilic alterations. Thromb Haemost 2006;96:348-55.

25. Middeldorp S, Henkens CM, Koopman MM et al. The incidence of venous thromboembolism in family members of patients with factor V Leiden mutation and venous thrombosis. Ann Intern Med 1998;128:15-20.

26. Lindqvist PG, Svensson PJ, Marsaal K, Grennert L, Luterkort M, Dahlback B. Activated protein C resistance (FV:Q506) and pregnancy. Thromb Haemost 1999;81:532-7.

27. Bovill EG, Bauer KA, Dickerman JD, Callas P, West B. The clinical spectrum of heterozygous protein C deficiency in a large New England kindred. Blood 1989;73:712-7.

28. Marteau TM, Croyle RT. The new genetics. Psychological responses to genetic testing. BMJ 1998;316:693-6. 29. Vadaparampil ST, Ropka M, Stefanek ME. Measurement of psychological factors associated with genetic

testing for hereditary breast, ovarian and colon cancers. Fam Cancer 2005;4:195-206. 30. Marteau TM, Weinman J. Self-regulation and the behavioural response to DNA risk information: a

theoretical analysis and framework for future research. Soc Sci Med 2006;62:1360-8. 31. Wiggins S, Whyte P, Huggins M et al. The psychological consequences of predictive testing for Huntington's

disease. Canadian Collaborative Study of Predictive Testing. N Engl J Med 1992;327:1401-5. 32. Watson M, Foster C, Eeles R et al. Psychosocial impact of breast/ovarian (BRCA1/2) cancer-predictive genetic

testing in a UK multi-centre clinical cohort. Br J Cancer 2004;91:1787-94. 33. Power TE, Adams PC. Psychosocial impact of C282Y mutation testing for hemochromatosis. Genet Test

2001;5:107-10. 34. DudokdeWit AC, Tibben A, Duivenvoorden HJ, Niermeijer MF, Passchier J. Predicting adaptation to

presymptomatic DNA testing for late onset disorders: who will experience distress? Rotterdam Leiden Genetics Workgroup. J Med Genet 1998;35:745-54.

35. Van Roosmalen MS, Stalmeier PF, Verhoef LC et al. Impact of BRCA1/2 testing and disclosure of a positive test result on women affected and unaffected with breast or ovarian cancer. Am J Med Genet A 2004;124:346-55.

36. Bonadona V, Saltel P, Desseigne F et al. Cancer patients who experienced diagnostic genetic testing for cancer susceptibility: reactions and behavior after the disclosure of a positive test result. Cancer Epidemiol Biomarkers Prev 2002;11:97-104.

37. Meiser B, Dunn S. Psychological impact of genetic testing for Huntington's disease: an update of the literature. J Neurol Neurosurg Psychiatry 2000;69:574-8.

38. Phil Alderson SG. Cochrane Collaboration open learning material for reviewers, version 1.1. 2002. 39. Cohn DM, Middeldorp S. Necessity of screening for thrombophilia at diagnosis of venous thromboembolism: outline of the NOSTRADAMUS intervention trial. Journal of Thrombosis and Haemostasis 2007;2007:P-M-470.

The psychological impact of testing for thrombophilia: a systematic review

47

Chapter 5

Quality of life after pulmonary embolism: the development of the Pulmonary Embolism Quality of Life questionnaire (PEmb-QoL)

Cohn DM, Nelis EA, Busweiler LA, Kaptein AA, Middeldorp S. Quality of life after pulmonary embolism: the development of the PEmb-QoL questionnaire. J Thromb Haemost 2009 Jun;7(6):1044-6.

1

2

3

4

5

6

7

8

9

10

11

12

Pulmonary Embolism (PE) is a rather common disease, with an annual incidence of

approximately 0.5 in 1000 persons in Western countries.1 As PE shares many features with

deep vein thrombosis (DVT), both diseases are considered to be different manifestations of

the same entity: venous thromboembolism (VTE).2 PE and DVT share the same risk factors

such as thrombophilia, pregnancy, cancer, surgery, immobilisation, and oral contraceptive

use. Furthermore, both manifestations occur in venous blood and coincide frequently.3-5

Accordingly, treatment recommendations are similar.2

PE is a leading cause of mortality and morbidity: death occurs in about 15% of the cases

within 6 months of its presentation.6 In addition, VTE often can be considered a chronic

disorder: recurrence is common with an incidence of approximately 30% within 10 years.7,8

Moreover, residual complaints (known as post thrombotic syndrome) are reported in 30%

of patients with DVT within 2 years after the initial event, despite the use of compression

stockings.9 The most important long-term complication of PE is chronic pulmonary

hypertension (which may manifest as fatigue, limited exercise tolerance or shortness of

breath) which was shown to affect 3.8% of PE patients within 2 years following the initial

event in one study.10

Quality of life is conceptualized increasingly as the central outcome of health care.

“Perceived health, health-related quality of life, and health-state utilities bring health

assessment progressively closer to the patient’s perspective”, is a conclusion in a paper by

Sullivan on taking the patient’s point of view on health care and health into account.11 To

illustrate the gained interest in quality of life, we performed a broad-brush search in

Medline with the MeSH term “quality of life”. This yielded 7143 articles published in 2007,

compared to 4923 in 2002. Surprisingly, a more sophisticated search without any

restriction yielded not a single publication on quality of life after PE (neither by a disease

generic questionnaire, nor by a disease specific questionnaire). In contrast, quality of life

following DVT has been subject of investigation and several DVT specific questionnaires

have been developed over the past decade.12-16

We aimed to develop a disease specific questionnaire to assess quality of life after PE using

the principles of grounded theory. We performed qualitative, semi-structured interviews in

10 outpatients (4 males / 6 females) whom we selected for the gravity of their complaints

following PE. These patients did not have other cardiopulmonary diseases that might have

resembled PE related complaints. Two investigators (LB and EN) visited the subjects at

their homes and structured the interviews into social functioning, physical complaints and

emotional disturbances. The interviews were tape-recorded with consent and transcribed

Quality of life after pulmonary embolism: development of the PEmb-QoL questionnaire

51

later. Characteristics of the interviewed patients are listed in Table 1. The most remarkable

complaints were shortness of breath/difficulty in breathing, fatigue, fear of recurrence after

discontinuing anticoagulant treatment, more readily emotionally disturbed (which

bothered a subgroup of the patients) and more social isolation than prior to the PE. The

authors (of whom 2 are experienced clinicians with a specific interest in patients with VTE)

remodelled the outcomes of the interviews into the draft questionnaire.

The original version was developed in Dutch. For the creation of the English version, the

Dutch version was independently translated by two native English speakers and

subsequently back-translated by a third native English speaker. The structure of the

questionnaire, which we named PEmb-QoL (Pulmonary Embolism Quality of Life), was

modelled in line with the existing generic SF-36 (short form 36) questionnaire and the

disease specific VEINES-QOL/Sym questionnaire, which has been developed for DVT. The

PEmb-QoL currently contains 10 questions (40 items) covering 6 dimensions: Frequency of

complaints (8 items), Activities of Daily Living (ADL) limitations (13 items), Work related

problems (4 items), Social limitations (1 item), Intensity of complaints (2 items) and

Emotional complaints (10 items). Two questions provide descriptive information. The

PEmb-QoL is a self administered questionnaire, in line with the SF-36 and Veines-

QOL/Sym questionnaires.

Our future aims are to further validate this questionnaire. We are currently distributing the

PEmb-QoL questionnaire amongst patients with a recent PE to assess construct validity.

The PEmb-QoL questionnaire will be distributed together with the disease generic SF-36

questionnaire to measure criterion validity. A subgroup of patients with PE will receive the

PEmb-QoL a second time for analysis of the test-retest reliability.

Chapter 5

52

1

2

3

4

5

6

7

8

9

10

11

12

Table 1. Characteristics of interviewed patients

ggender, age (years), marital status

PE event main functional complaints

main psychological complaints

main social limitations

1 female, 37, married

7 months prior to interview. Massive PE, resuscitated

fatigue, muscle weakness

anxiety: more readily emotional; fear of recurrent PE; worried to stop anticoagulant treatment

afraid to be a burden of relatives and friends, afraid of being alone

2 male, 31, unknown

6 years prior to interview first PE; 2 recurrences (22 months and 7 months respectively)

pain behind the shoulder blades, pain in the chest, tiredness, difficulty in breathing

fear of recurrent PE, more readily emotional (experienced as annoying), depressed

not able to work, limited in social contacts

3 female, 84, widow

1 year ago fatigue, not able to exert herself

depressed (at times) becomes easily weary after having a visit from friends/relatives

4 female, 43, married

13 months and 5 months prior to the interview

pain in the back and fatigue

fear of recurrence, depressed feeling, more readily emotional

avoids visits to friends and shopping in the centre of the city, which is too exhausting

5 female, 73, widow

4 years prior to the interview

breathlessness, sensation of pressure, fatigue

fear of recurrence more socially isolated, limited in taking a trip

6 female, 32, single

14 months prior to PE difficulty in breathing, pain at the back and between shoulder blades

anxious of recurrence, worried to stop anticoagulant treatment

no complaints

7 male, 34, married

4 months prior to interview

difficulty in breathing, fatigue, shortness of breath, pain behind shoulder blades, chest pain

fear of recurrence, more readily emotional (which bothers the patient)

feeling of lack of attention to his children, not able to work, not able to have visitors too often (too exhausting)

8 male, 63, married

13 months prior to interview

shortness of breath, tired, difficulty in breathing

much more emotional disturbed

not able to perform as much as he intends

9 female, 79, widow

first PE 7 years prior to interview, recurrent PE 6 years ago

difficulty in climbing stairs and fatigue, shortness of breath

no typical complaints no typical complaints

10 male, 55, married

18 months prior to interview

more easily tired, difficulty in breathing following exercise, chest pain

afraid of a recurrent PE, worried of stopping anticoagulant treatment

prefers staying at home

PE=pulmonary embolism

Quality of life after pulmonary embolism: development of the PEmb-QoL questionnaire

53

References 1. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and

mortality of venous thrombosis: a population-based study. J Thromb Haemost 2007;5:692-9. 2. Büller HR, Agnelli G, Hull RD, Hyers TM, Prins MH, Raskob GE. Antithrombotic therapy for venous

thromboembolic disease: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126:401S-28S.

3. Hull RD, Raskob GE, Ginsberg JS et al. A noninvasive strategy for the treatment of patients with suspected pulmonary embolism. Arch Intern Med 1994;154:289-97.

4. Moser KM, Fedullo PF, LitteJohn JK, Crawford R. Frequent asymptomatic pulmonary embolism in patients with deep venous thrombosis. JAMA 1994;271:223-5.

5. Nielsen HK, Husted SE, Krusell LR, Fasting H, Charles P, Hansen HH. Silent pulmonary embolism in patients with deep venous thrombosis. Incidence and fate in a randomized, controlled trial of anticoagulation versus no anticoagulation. J Intern Med 1994;235:457-61.

6. Murin S, Romano PS, White RH. Comparison of outcomes after hospitalization for deep venous thrombosis or pulmonary embolism. Thromb Haemost 2002;88:407-14.

7. Prandoni P, Lensing AW, Cogo A et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996;125:1-7.

8. Heit JA, Mohr DN, Silverstein MD, Petterson TM, O'Fallon WM, Melton LJ, III. Predictors of recurrence after deep vein thrombosis and pulmonary embolism: a population-based cohort study. Arch Intern Med 2000;160:761-8.

9. Brandjes DP, Büller HR, Heijboer H et al. Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet 1997;349:759-62.

10. Pengo V, Lensing AW, Prins MH et al. Incidence of chronic thromboembolic pulmonary hypertension after pulmonary embolism. N Engl J Med 2004;350:2257-64.

11. Sullivan M. The new subjective medicine: taking the patient's point of view on health care and health. Soc Sci Med 2003;56:1595-604.

12. Lamping DL, Schroter S, Kurz X, Kahn SR, Abenhaim L. Evaluation of outcomes in chronic venous disorders of the leg: development of a scientifically rigorous, patient-reported measure of symptoms and quality of life. J Vasc Surg 2003;37:410-9.

13. Kahn SR, M'Lan CE, Lamping DL, Kurz X, Berard A, Abenhaim L. The influence of venous thromboembolism on quality of life and severity of chronic venous disease. J Thromb Haemost 2004;2:2146-51.

14. Korlaar van I, Vossen CY, Rosendaal FR et al. The impact of venous thrombosis on quality of life. Thromb Res 2004;114:11-8.

15. Kahn SR, Lamping DL, Ducruet T et al. VEINES-QOL/Sym questionnaire was a reliable and valid disease-specific quality of life measure for deep venous thrombosis. J Clin Epidemiol 2006;59:1049-56.

16. Hedner E, Carlsson J, Kulich KR, Stigendal L, Ingelgard A, Wiklund I. An instrument for measuring health-related quality of life in patients with Deep Venous Thrombosis (DVT): development and validation of Deep Venous Thrombosis Quality of Life (DVTQOL) questionnaire. Health Qual Life Outcomes 2004;2:30.

Chapter 5

54

Chapter 6

Quality of life after pulmonary embolism: validation of the Pulmonary Embolism Quality of Life questionnaire (PEmb-QoL)

Based on: Cohn DM, Klok FA, Middeldorp S, Scharloo M, Büller HR, van Kralingen KW, Kaptein AA, Huisman MV. J Thromb Haemost 2010; in press.

Abstract

Background

Even though quality of life (QoL) has become a key component of medical care, there is no

instrument available that specifically measures QoL after pulmonary embolism (PE).

Recently, the Pulmonary Embolism Quality of Life (PEmb-QoL) questionnaire has been

developed to address this gap.

Objective

To evaluate the validity of the PEmb-QoL questionnaire.

Methods

We distributed the PEmb-QoL questionnaire and the Short Form-36 (SF-36) questionnaire

twice among consecutive subjects with a history of objectively confirmed acute PE. Internal

consistency reliability, test-retest reliability, convergent validity and criterion validity of the

PEmb-QoL were assessed using standard-scale construction techniques.

Results

90 participants completed the questionnaires twice. Internal consistency was adequate

(Cronbach’s � 0.62-0.94), as well as test-retest reliability (intra-class correlation

coefficients: 0.78-0.94). Furthermore, correlation between the PEmb-QoL questionnaire

and the SF-36 questionnaire supported convergent validity.

Conclusion

The PEmb-QoL questionnaire is a reliable instrument to specifically assess QoL following

PE, which is helpful in the identification of patients with decreased QoL following acute PE.

56

Chapter 6

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Pulmonary Embolism (PE) is a common disorder characterized by the obstruction of the

pulmonary arterial tree by floating thrombi predominantly originating from the leg or

pelvic veins.1 Although PE has traditionally been considered to be an acute disease, the long

term natural course in patients surviving the acute thromboembolic event can be

complicated by recurrent episodes of PE or deep vein thrombosis, bleeding complications

caused by anticoagulant treatment, arterial cardiovascular events and in rare cases by

chronic thromboembolic pulmonary hypertension (CTEPH).1-4 CTEPH may present as

fatigue, limited exercise tolerance or shortness of breath and affects approximately 4% of

PE patients within 2 years following the initial event, as reported in one study.3 Moreover,

patients often have residual dyspnoea complaints years after the acute thromboembolic

event.5

Quality of life (QoL) has become an important outcome aspect of medical care. QoL can be

assessed by generic or disease-specific questionnaires. The latter are more sensitive than

generic questionnaires to detect and quantify small changes that are relevant to patients.

Several disease-specific QoL instruments have been developed for deep venous thrombosis

(DVT), a condition closely related to PE and considered a manifestation of the same disease

entity.6-10 Furthermore, several specific questionnaires for symptoms of the respiratory tract

have been created, such as the Cambridge Pulmonary Hypertension Outcome Review

(CAMPHOR)11or the Chronic Respiratory Disease Questionnaire (CRQ).12 However, since

respiratory or other symptoms that affect QoL after PE have never been purposely studied,

we have developed a new measure -the Pulmonary Embolism Quality of Life (PEmb-QoL)

questionnaire-, based on symptoms as reported by 10 interviewed participants with severe

complaints following PE. Details on the development of the PEmb-QoL questionnaire have

been described previously.13 The complete questionnaire is presented in the Appendix.

The PEmb-QoL was modelled on the quality of life after DVT (VEINES-QOL/Sym)

questionnaire6,7,9 Both questionnaires assess the frequency of symptoms, the time of day at

which the complaints are at their worst, and ADL as well as work related problems.

However, the PEmb-QoL questionnaire is distinct from the VEINES-QOL/Sym in the

inclusion of pulmonary-specific symptoms, adding questions on limitations in daily

physical activities, and extending the number of questions on emotional functioning.

Moreover, in order not to lose valuable information, we decided to assess the different

areas of limitations as separate dimensions, instead of combining items into two subscales

(symptoms and QoL), as is the case in the VEINES-QOL questionnaire. In the present

57

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

paper, we report results from the validation study that was performed to assess the

psychometric and clinical characteristics of the questionnaire.

Methods

Participants

The Dutch version of the PEmb-QoL was distributed among consecutive participants of a

large follow-up study in patients with a history of acute PE referred to the Leiden University

Medical Centre. Inclusion criteria were objectively confirmed PE diagnosed between

January first 2001 and July first 2007. All surviving patients were invited for a control visit in

our outpatient clinic. We asked a random, consecutive subsample of 93 participants to

complete the PEmb-QoL and Short-Form 36 (SF-36) questionnaires shortly before this

visit. After first review, incomplete questionnaires were completed at the study visit. For

assessment of test-retest reliability, participants were instructed to complete both

questionnaires for a second time (within a two week period) at home shortly after the visit

and return these by mail. Incomplete returned questionnaires were completed by the

patients following contact by telephone. We excluded participants with language barriers

who could not complete the questionnaires in Dutch. The study protocol was approved by

the Medical Review Ethics Committee of the Leiden University Medical Centre and all

patients provided written informed consent.

Measures

PEmb-QoL questionnaire

We applied the disease specific PEmb-QoL questionnaire which we developed to assess

QoL in patients with PE.13 The original version of this questionnaire was developed in

Dutch. For the creation of the English version (see Appendix), the Dutch version was

independently translated by two native English speakers and subsequently back-translated

by a third native English speaker. The PEmb-QoL questionnaire contains 6 dimensions:

frequency of complaints, ADL (activities of daily living) limitations, work related problems,

social limitations, intensity of complaints and emotional complaints. Higher scores

indicate worse outcome.

SF-36 questionnaire

The SF-36 is a generic QoL measure containing 8 scales (physical functioning, social

functioning, physical role functioning, emotional role functioning, mental health, vitality,

58

Chapter 6

1

2

3

4

5

6

7

8

9

10

11

12

bodily pain, and general health), scoring 0 to 100, with higher values indicating better

health.14 Two summary scores are created by combining scales into a physical health

summary score and mental health summary score.

Outcome measures

We expected that the PEmb-QoL dimensions frequency of complaints, ADL limitations,

work related problems, social limitations and intensity of complaints would have higher

correlations with the physical health summary score of the SF-36, whereas emotional

complaints would have a higher correlation with the mental health summary score. The

outcome measures of this analysis were internal consistency reliability (which assesses

whether several items that propose to measure the same general construct produce similar

scores), test- retest reliability, convergent validity, criterion validity (as assessed by

comparing the PEmb-QoL dimensions with the dimensions of the SF-36 disease generic

questionnaire), and the association of patient demographics, comorbid conditions and PE

characteristics with higher or lower QoL in our patient population.

Statistical Analyses

The responses were entered and analysed in SPSS version 16.0.2. Means and standard

deviations were calculated for normally distributed variables. Non-normally distributed

variables were expressed in medians with ranges. We performed a factor analysis on the

items of the PEmb-QoL with varimax rotation to examine the underlying constructs.

Internal consistency reliability was calculated with Cronbach’s �.16 Following the

recommendations of DeVellis, internal consistency reliability was considered adequate if

Cronbach’s � was higher than 0.7.17 Demographic associations of the PEmb-QoL

dimensions were tested using independent sample T-tests in case of two categories and

analysis of variance with post-hoc Scheffé tests18 in case of more than two categories. Test

re-test reliability was expressed as intra-class correlation coefficients. We calculated inter-

dimension correlations and criterion validity with bivariate Spearman correlation

coefficients.

Results

Patients

The questionnaires were distributed amongst 93 participants, of whom 90 completed the

questionnaire after a median period of 38 months (range 10 – 91 months) following the PE.

59

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

Three participants (3.2%) were excluded due to inability to complete this questionnaire in

Dutch because of language barriers. The number of missing items was very low; however

the exact number could not be calculated as all missing items were completed by the

respondents following contact by phone with the researchers. Demographic and clinical

characteristics are presented in Table 1.

Table 1. Demographics of 93 included patients

demographics of the study population Male sex (n, %) 44 (47) Age (years ±SD) 56±14 Recurrent PE (n, %) 19 (20) Concurrent illness

malignancy (n, %) 12 (13%) pulmonary disease (e.g. COPD) (n, %) 14 (15%) cardiac disease (n, %) 6 (7%)

Time to registration event* (range) 3 years and 2 months (10 months – 7 years and 7 months)

PE=pulmonary embolism, n=number.* time span between registration acute PE and study inclusion.

Factor analysis (with varimax rotation) supported the underlying dimensions producing 6

factors which accounted for 72% of the total variance. The rotated component matrix is

presented in Table 2. Table 3 lists the internal reliabilities of the dimensions, as expressed

by Cronbach’s �, as well as inter-dimension correlations between the PEmb-QoL

dimensions. Internal reliability was high (�0.87) for the dimensions frequency of

complaints, ADL limitations, work related problems, and emotional complaints but lower

for the dimension intensity of complaints (� = 0.62). We assessed whether deletion of any

of the items in the various dimensions could increase the internal reliability of any of the

dimensions (and hence whether the PEmb-QoL questionnaire could be abridged).

However, deletion of any of the items from the various dimensions did not lead to

substantial improvements of the dimensions’ internal consistency reliability. The highest

correlations between dimensions were found between intensity of complaints and all other

dimensions (0.60 � r � 0.79). Except for work related problems and frequency of

complaints (r=0.42), all dimensions were moderately correlated (0.56� r �0.82).

60

Chapter 6

1

2

3

4

5

6

7

8

9

10

11

12

Table 2. Rotated component matrix of the PEmb-QoL questionnaire

Pemb-QoL items

Frequency of complaints

ADL limitations

Work related problems

Social limitations

Intensity of complaints

Emotional complaints

1a 0.771 0.230 0.202 -0.095 0.223 0.023 1b 0.806 0.172 0.029 0.062 0.115 0.047 1c 0.608 0.232 -0.014 -0.044 0.040 0.298 1d 0.757 0.154 -0.078 0.062 0.151 0.255 1e 0.628 0.195 -0.063 -0.359 0.321 0.083 1f 0.766 0.075 0.163 -0.061 -0.042 0.242 1g 0.772 0.144 0.101 -0.044 -0.019 0.298 1h 0.267 0.424 0.045 0.000 00.617 0.190 2 N.A. 3 N.A. 4a 0.200 0.158 -0.049 --0.779 -0.029 -0.020 4b 0.239 00.674 0.322 0.275 0.166 0.129 4c 0.522 00.568 0.273 0.182 0.051 0.129 4d 0.118 00.601 0.321 0.098 0.433 0.098 4e 0.186 00.785 0.251 0.099 0.142 0.085 4f 0.109 00.860 0.121 0.011 0.073 0.188 4g 0.138 00.757 0.215 0.073 0.391 0.059 4h 0.122 00.832 0.084 0.086 -0.009 0.156 4i 0.104 00.778 0.231 0.166 0.083 0.162 4j 0.250 00.833 0.184 -0.116 0.164 0.101 4k 0.243 00.845 0.054 -0.033 -0.006 0.197 4l 0.224 00.845 0.020 -0.002 -0.052 0.228 4m 0.291 00.696 -0.057 0.070 -0.155 0.157 5a 0.182 0.463 00.603 0.133 0.099 0.250 5b 0.105 0.514 00.651 0.056 0.132 0.282 5c 0.113 0.559 00.672 -0.045 0.147 0.194 5d 0.171 0.565 00.633 -0.058 0.171 0.194 6 0.349 00.600 0.192 -0.005 0.038 0.311 7 0.792 0.314 0.061 -0.167 0.232 0.109 8 0.314 00.554 0.178 -0.032 00.507 0.291 9a 0.458 0.144 0.108 0.324 -0.044 00.503 9b 0.170 0.105 0.164 0.007 0.161 00.811 9c 0.042 0.147 -0.003 0.062 0.089 00.266 9d 0.284 0.210 0.111 0.066 0.184 00.834 9e 0.339 0.231 0.125 0.061 0.141 00.791 9f 0.097 0.218 0.138 0.113 0.156 00.811 9g 0.433 0.333 0.368 0.009 0.093 00.439 9h 0.059 0.374 0.245 0.250 0.043 00.460 9i 0.500 0.389 0.347 0.252 -0.076 0.283 9j 0.352 0.353 -0.055 0.158 -0.072 00.592

Highest factor loadings are stated in bold.

61

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

Table 3. Internal Consistency Reliability and Correlations between PEmb-QoL dimensions

PEmb-QoL dimensions

PEmb-QoL questions

Number of items

Cronbach’s �

Mean score (SD)

Frequency of complaints

ADL limitations

Work related problems

Social limitations

Intensity of complaints

Frequency of complaints Question 1* 8 0.90 1.9 (0.9) - - - -

ADL limitations

Question 4* 13 0.94 1.4 (0.5) 0.61*** - - -

Work related problems

Question 5* 4 0.87 1.3 (0.4) 0.42*** 0.68*** - -

Social limitations

Question 6 1 N.A. 1.5 (0.9) 0.66*** 0.72*** 0.59*** -

Intensity of complaints

Questions 7/8 2 0.62 2.2 (1.1) 0.79*** 0.74*** 0.60*** 0.74***

Emotional complaints

Question 9* 10 0.91 2.0 (1.1) 0.69*** 0.66*** 0.59*** 0.70*** 0.71***

ADL = Activities of Daily Living; *items reversely scored (higher scores indicate more complaints); N.A.= not applicable; ***p < 0.001

The results of the test re-test analysis are presented in Table 4. Intra-class correlation

coefficients for test-retest analysis varied between 0.78 for work related problems and 0.94

for frequency of complaints.

Table 4. Test-retest reliability

PEmb-QoL dimensions intra-class correlation coefficients Frequency of complaints 0.94*** ADL limitations 0.87*** Work related problems 0.78*** Social limitations 0.83*** Intensity of complaints 0.85*** Emotional complaints 0.81***

***p<0.001

The results of the criterion validity are reported in Table 5. As expected, the PEmb-QoL

dimensions frequency of complaints, ADL limitations, work related problems, social

limitations and intensity of complaints had higher associations with the physical health

summary score of the SF-36 questionnaire, whereas emotional complaints were most

strongly associated with the mental health summary score. Frequency of complaints was

most strongly correlated with vitality (r=-0.56), social functioning (r=-0.55) and physical

functioning (r=-0.46). The strongest correlations for ADL limitations were physical

functioning (r=-0.78), social functioning (r=-0.61) and vitality (r=-0.66). Work-related

problems most strongly correlated with physical role functioning (r=-0.58) and physical

62

Chapter 6

1

2

3

4

5

6

7

8

9

10

11

12

functioning (r=-0.53). The dimension social limitations was correlated with several SF-36

dimensions. We observed strong correlations with physical functioning (r=-0.54), social

functioning (r=-0.55) and vitality (r=-0.53). Intensity of complaints was strongly correlated

with the same SF-36 dimensions as frequency of complaints (but with higher correlations

coefficients). The strongest correlations for emotional complaints were mental health (r=-

0.57), vitality (r=-0.69) and social functioning (r=-0.60).

Table 5. Criterion validity: Pearson correlations between PEmb-QoL dimensions and SF-36 subscales

SF-36

PEmb-QoL Physical functioning

Social functioning

Physical role functioning

Emotional role functioning

Mental health

Vitality Bodily pain General health

Physical health summary

Mental health summary

Frequency of complaints -0.46** -0.55** -0.33** -0.28** -0.45** -0.56** -0.42** -0.55** -0.44** -0.42**

ADL limitations

-0.78** -0.61** -0.53** -0.39** -0.45** -0.66** -0.49** -0.63** -0.66** -0.39**

Work related problems -0.53** -0.50** -0.58** -0.43** -0.40** -0.54** -0.29** -0.53** -0.51** -0.39**

Social limitations -0.54** -0.55** -0.45** -0.31** -0.38** -0.53** -0.35** -0.51** -0.50** -0.35**

Intensity of complaints

-0.62** -0.66** -0.44** -0.32** -0.55** -0.67** -0.49** -0.60** -0.55** -0.48**

Emotional complaints -0.45** -0.60** -0.38** -0.43** -0.57** -0.69** -0.41** -0.59** -0.40** -0.57**

*p<0.05; **p<0.01

Analysis of the associations between PEmb-QoL scores and various demographic

parameters did not reveal substantial differences in QoL. Men reported slightly lower

frequency of complaints (mean score 1.6 vs. 2.1; mean difference -0.48 [95CI -0.86; -0.11]).

In addition, patients younger than 50 years (at the time of a first pulmonary embolism)

reported higher scores for frequency of complaints (mean score 2.2 vs. 1.7; mean

difference 0.59 [95%CI 0.14; 1.03]), as compared to respondents over 50 years of age.

Furthermore patients with a concurring cardiac condition (such as heart failure or a prior

myocardial infarction) reported more ADL limitations (mean score 2.0 vs. 1.3; mean

difference 0.62 [95%CI 0.03-1.21]) and slightly more work related problems (mean score

1.7 vs. 1.2; mean difference 0.52 [95%CI 0.09-0.94]) as compared to respondents without

cardiac or pulmonary comorbidity, or a concurring malignancy.

63

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

Discussion

The results from this validation study indicate that this newly developed disease-specific

health-related QoL instrument PEmb-QoL is a valid and reliable instrument to assess QoL

following PE. Internal reliability for all dimensions (except intensity of complaints) was

adequate and comparable to the reliability of the VEINES-QOL/Sym scales.6 Test-retest

reliability was also adequate. The inter-correlations between the PEmb-QoL dimensions

demonstrated logical relationships. Intensity of complaints correlated with a worse

outcome in all other dimensions. This was expected as this dimension might actually affect

a person’s well-being in general. Also, its association with frequency of complaints is high,

suggesting these dimensions could be taken together to form one summary score for

symptom severity, comparable to the VEINES-QoL/Sym summary score.

We observed a tendency towards small floor and ceiling effects in some of the PEmb-QoL

dimensions. This was assumed to be attributable to the time between the events and

completion of the questionnaires. Therefore, we expect that other patient samples

including those with a more recent event will show less floor or ceiling effects.

Furthermore, we observed that work related problems most strongly correlated with

physical role functioning and physical functioning. We hypothesized that this observation

is explained by the fact that both dimensions focus on the extent of limitations performing

work or physical exercise. Emotional complaints were more strongly associated with

mental health and vitality as compared to emotional role functioning. This is also

conceivable, as the wording of the items of this PEmb-QoL dimension closely matches the

items of the SF-36 dimensions mental health and vitality. Also, the correlation between

social limitations and social functioning was (almost) as high as the correlation with

physical functioning and vitality. This is a plausible observation as well, since social

activities are also influenced by the capability to perform exercises such as climbing stairs

or walking a certain distance.

Limitations of our study comprise the exclusion of 3 participants due to language barriers

and the lack of detailed comparison to healthy subjects. This comparison is difficult since

the PEmb-QoL was designed for patients with acute PE and is by definition not applicable

to subjects without this disease. Furthermore, the PEmb-QoL dimensions were created

based on the contents of the items and not on the results of factor analysis. Post hoc

varimax rotation analysis demonstrated that although most items fitted well in the

proposed dimensions, few items had more than one high loading (for instance item 8 and

9i) or had higher loadings in other than those they were designated to. On the other hand,

64

Chapter 6

1

2

3

4

5

6

7

8

9

10

11

12

item 9c had low loadings on all dimensions. As this is the first report of factor analysis of

the PEmb-QoL questionnaire, future studies should assess whether some items might fit

better in other dimensions, possibly allowing removal or regrouping of some items.

In summary, the PEmb-QoL is a valuable instrument for determining the disease-specific

QoL in patients with previous acute PE. The clinical applicability of the PEmb-QoL remains

to be studied in clinical outcome studies.

References 1. Tapson VF. Acute pulmonary embolism. N Engl J Med 2008; 358:1037-52. 2. Spencer FA, Gore JM, Lessard D, Douketis JD, Emery C, Goldberg RJ. Patient outcomes after deep vein

thrombosis and pulmonary embolism: the Worcester Venous Thromboembolism Study. Arch Intern Med 2008; 168: 425-30.

3. Pengo V, Lensing AW, Prins MH, Marchiori A, Davidson BL, Tiozzo F, Albanese P, Biasiolo A, Pegoraro C, Iliceto S, Prandoni P; Thromboembolic Pulmonary Hypertension Study Group. Incidence of chronic thromboembolic pulmonary hypertension after pulmonary embolism. N Engl J Med 2004; 350: 2257-64.

4. Klok FA, Mos IC, Broek L, Tamsma JT, Rosendaal FR, de Roos A, Huisman MV. Risk of arterial cardiovascular events in patients after pulmonary embolism. Blood 2009; doi:10.1182/blood-2009-05-220491 [Epub ahead of print].

5. Klok FA, Tijmensen JE, Haeck ML, van Kralingen KW, Huisman MV. Persistent dyspnea complaints at long-term follow-up after an episode of acute pulmonary embolism: results of a questionnaire. Eur J Intern Med 2008; 19: 625-9.

6. Lamping DL, Schroter S, Kurz X, Kahn SR, Abenhaim L. Evaluation of outcomes in chronic venous disorders of the leg: development of a scientifically rigorous, patient-reported measure of symptoms and quality of life. J Vasc Surg 2003; 37: 410-9.

7. Kahn SR, M'Lan CE, Lamping DL, Kurz X, Berard A, Abenhaim L. The influence of venous thromboembolism on quality of life and severity of chronic venous disease. J Thromb Haemost 2004; 2: 2146-51.

8. Korlaar van I, Vossen CY, Rosendaal FR, Bovill EG, Cushman M, Naud S, Kaptein AA. The impact of venous thrombosis on quality of life. Thromb Res 2004; 114: 11-8.

9. Kahn SR, Lamping DL, Ducruet T, Arsenault L, Miron MJ, Roussin A, Desmarais S, Joyal F, Kassis J, Solymoss S, Desjardins L, Johri M, Shrier I. VEINES-QOL/Sym questionnaire was a reliable and valid disease-specific quality of life measure for deep venous thrombosis. J Clin Epidemiol 2006; 59: 1049-56.

10. Hedner E, Carlsson J, Kulich KR, Stigendal L, Ingelgard A, Wiklund I. An instrument for measuring health-related quality of life in patients with Deep Venous Thrombosis (DVT): development and validation of Deep Venous Thrombosis Quality of Life (DVTQOL) questionnaire. Health Qual Life Outcomes 2004; 2: 30.

11. McKenna SP, Doughty N, Meads DM, Doward LC, Pepke-Zaba J. The Cambridge Pulmonary Hypertension Outcome Review (CAMPHOR): a measure of health-related quality of life and quality of life for patients with pulmonary hypertension. Qual Life Res 2006; 15: 103-15.

12. Guyatt GH, Berman LB, Townsend M, Pugsley SO, Chambers LW. A measure of quality of life for clinical trials in chronic lung disease. Thorax 1987; 42: 773-8.

13. Cohn DM, Nelis EA, Busweiler LA, Kaptein AA, Middeldorp S. Quality of life after pulmonary embolism: the development of the PEmb-QoL questionnaire. J Thromb Haemost 2009; 7: 1044-6.

14. Ware JE, Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992; 30: 473-83.

15. Kahn SR, Shbaklo H, Lamping DL, Holcroft CA, Shrier I, Miron MJ, Roussin A, Desmarais S, Joyal F, Kassis J, Solymoss S, Desjardins L, Johri M, Ginsberg JS. Determinants of health-related quality of life during the 2 years following deep vein thrombosis. J Thromb Haemost 2008; 6:1105-12.

16. Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika 1951; 16: 297-334. 17. DeVellis R.F. Scale Development: Theory and Applications (Applied Social Research Methods) 2nd. Edition.

Sage; 2003. 18. Scheffé H. A method for judging all contrasts in the analysis of variance. Biometrika 1953; 40: 87-104.

65

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

Appendix: the PEmb-QoL Questionnaire

Questionnaire

After having a

PULMONARY EMBOLISM

INSTRUCTIONS HOW TO ANSWER: Answer every question by marking the answer as indicated. If you are unsure about how to answer a question, please give the best answer you can. These questions are about your llungs. The information you give should describe how you feel. You can also indicate how capable you are of carrying out your normal activities.

Chapter 6 — Appendix

66

1

2

3

4

5

6

7

8

9

10

11

12

1. During the past 4 weeks, how often have you had any of the following symptoms from your lungs?

(Circle 1 answer on each line) Every day Several

times a week About once a

week Less than

once a week Never

Pain behind or between the shoulder blades?

1 2 3 4 5

Pain on or in the chest? 1 2 3 4 5

Pain in the back? 1 2 3 4 5

Sensation of pressure? 1 2 3 4 5

Feeling that there is “still something there”?

1 2 3 4 5

“Burning sensation” in the lungs? 1 2 3 4 5

“Nagging feeling” in the lungs? 1 2 3 4 5

Difficulty in breathing or breathlessness?

1 2 3 4 5

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

67

2. At what time of day are your llung symptoms most intense? (circle one answer)

1. On waking

2. At mid-day

3. At the end of the day

4. During the night

5. At any time of the day

6. Never

3. Compared to one year ago, how would you rate the ccondition of your llungs in general now? (circle one answer)

1. Much better now than one year ago

2. Somewhat better now than one year ago

3. About the same now as one year ago

4. Somewhat worse now than one year ago

5. Much worse now than one year ago

6. I did not have any problems with my lungs

Chapter 6

68

Chapter 6 — Appendix

1

2

3

4

5

6

7

8

9

10

11

12

4. TThe following items are about activities that you might do in a typical day. Do your llung symptoms now limit you in these activities? If so, how much?

(Circle one answer on each line) I do not work

YES, Limited A Lot

YES, Limited A Little

NO, Not Limited At All

a. DDaily aactivities at work 0 1 2 3

b. DDaily aactivities at home (e.g. housework, ironing, doing odd jobs/repairs around the house, gardening, etc…)

1 2 3

c. SSocial or activities (such as travelling, going to the cinema, parties, shopping)

1 2 3

d. VVigorous activities, such as running, lifting heavy objects, participating in strenuous sports

1 2 3

e. MModerate activities, such as moving a table, hovering, swimming or cycling

1 2 3

f. Lifting or carrying groceries 1 2 3

g. Climbing sseveral flights of stairs 1 2 3

h. Climbing oone flight of stairs 1 2 3

i. Bending, kneeling, or squatting 1 2 3

j. Walking mmore than half a mile 1 2 3

k. Walking a couple of hundred yards 1 2 3

l. Walking about one hundred yards 1 2 3

m. Washing or dressing yourself 1 2 3

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

69

5. During the past four weeks, have you had any of the following problems with your work or other regular daily activities as a result of your llung symptoms?

(Circle one answer on each line) YES NO

a. Cut down the amount of time you spent on work or other activities 1 2

b. AAccomplished less than you would like 1 2 c. Were limited in the kkind of work or other

activities 1 2

d. Had ddifficulty performing the work or other activities (e.g. it took extra effort) 1 2

6. During the past four weeks, to what extent have your llung symptoms interfered with your normal social activities with family, friends, neighbours or groups? (Circle one answer)

1. Not at all

2. Slightly

3. Moderately

4. Quite a bit

5. Extremely

7. How much pain around your shoulder blades / pain in your chest have you experienced during the past four weeks? (Circle one answer)

1. None

2. Very slight

3. Slight

4. Quite a bit

5. Serious

6. Very serious

8. How much breathlessness have you experienced in the past four weeks? (Circle one answer)

1. None

2. Very slight

3. Slight

4. Quite a bit

5. Serious

6. Very serious

Chapter 6

70

Chapter 6 — Appendix

1

2

3

4

5

6

7

8

9

10

11

12

9. These questions are about how you feel and how things have been with you during the past four weeks as a result of your llung symptoms. For each question, please give the one answer that comes closest to the way you have been feeling. How much of the time during the past four weeks -

(Circle one answer on each line)

All of the Time

Most of the Time

A good Bit of the Time

Some of the Time

A Little of the Time

None of the Time

were you worried about having another pulmonary embolism?

1 2 3 4 5 6

did you feel irritable? 1 2 3 4 5 6 would you have been worried if you had to stop taking anticoagulant medication?

1 2 3 4 5 6

did you become emotional more readily? 1 2 3 4 5 6

did it bother you that you became emotional more quickly?

1 2 3 4 5 6

were you depressed or in low spirits? 1 2 3 4 5 6

did you feel that you were a burden to your family and friends?

1 2 3 4 5 6

were you afraid to exert yourself? 1 2 3 4 5 6

did you feel limited in taking a trip? 1 2 3 4 5 6

were you afraid of being alone? 1 2 3 4 5 6

Legend to the Appendix: Higher scores indicate worse outcome. Scores for all dimensions are calculated by the sum of the scores for each item of the dimension divided by the number of the items. Questions 1, 4, 5 and 9 are reversed scored. Questions 2 and 3 provide descriptive information.

Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire

71

Part II

Identification of new thrombophilic factors

Chapter 7

Venous thrombosis is associated with hyperglycaemia at diagnosis: a case-control study

Cohn DM, Hermanides J, DeVries JH, Kamphuisen PW, Huijgen R, Meijers JC, Hoekstra JB, Buller HR. J Thromb Haemost 2009 Jun;7(6):945-9.

Abstract

Background

Patients with (undiagnosed) diabetes mellitus, impaired glucose tolerance or stress-

induced hyperglycaemia may be at greater risk for venous thrombosis and present with

relative hyperglycaemia during the thrombotic event.

Objectives

To assess whether venous thrombosis is associated with hyperglycaemia at diagnosis.

Methods

We performed a case-control study, derived from a cohort of consecutive patients referred

for suspected deep venous thrombosis. Cases were patients with confirmed symptomatic

venous thrombosis of the lower extremity. Controls were randomly selected in a 1:2 ratio

from individuals in whom this diagnosis was excluded. We measured plasma glucose levels

upon presentation to the hospital.

Results

A total of 188 patients with thrombosis and 370 controls were studied. The glucose cut-off

level for the 1st quartile was < 5.3 mmol/l, the 2nd 5.3-5.7 mmol/l, the 3rd 5.7-6.6 mmol/l and

the 4th quartile � 6.6 mmol/l. When adjusted for body mass index, a known history of

diabetes mellitus, age, sex, ethnicity and whether known risk factors for deep venous

thrombosis were present, the OR for deep venous thrombosis in the 2nd, 3rd and 4th quartile

of glucose levels compared to the 1st quartile was 1.59 (95% CI 0.89-2.85), 2.04 (95 % CI

1.15-3.62) and 2.21 (95% CI 1.20-4.05), respectively, P for trend =0.001.

Conclusions

Increased glucose levels measured at presentation were associated with venous

thrombosis. Experimental evidence supports a potential causal role for hyperglycaemia in

this process. As this is the first report about the association between (stress)

hyperglycaemia and venous thrombosis, confirmation in other studies is required.

Chapter 7

76

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Venous thromboembolism (VTE) is a common disease with an annual incidence of 2-3 per

1,000 inhabitants.1 Both acquired and inherited risk factors are known to play a role in the

development of thrombosis.2 Nevertheless, in approximately 25% of patients with VTE

neither an acquired nor an inherited risk factor can be demonstrated.3,4 Evidence is growing

that classic risk factors for arterial disease are also involved in the development of VTE.5,6

Hyperglycaemia is associated with arterial thrombosis7-9 and, indeed, patients with diabetes

mellitus or the metabolic syndrome10) also have an increased risk of VTE.11,12 This increased

risk of VTE can in part be explained by the platelet activation and hypercoagulability present

in diabetes mellitus.13 Activation of the coagulation system has also been observed in acute

experimentally induced hyperglycaemia in healthy male volunteers.14 During acute illness

such as myocardial infarction, a phenomenon called stress hyperglycaemia may occur

independently of the presence of known diabetes.

Since hyperglycaemia stimulates coagulation, we hypothesized that higher glucose levels -

independent of known diabetes mellitus- would be more common in patients presenting

with acute VTE.

Methods

We performed a case-control study. Cases and controls were selected from the Amsterdam

Case-Control Study on Thrombosis (ACT) which was initiated in 1999 to identify new risk

factors for DVT.

All consecutive outpatients older than 18 years referred to the Academic Medical Centre in

Amsterdam between September 1999 and August 2006 with clinically suspected deep

venous thrombosis (DVT) of the lower extremity were eligible for this study. The study

protocol was approved by the Medical Ethics Review Committee and all participants

provided written informed consent. At presentation, the patient’s medical history was

obtained through a standardized questionnaire including specific questions about

symptom duration, presence of known risk factors (concomitant malignancy, pregnancy,

use of Hormonal Replacement Therapy, Oral Contraceptives or selective oestrogen receptor

modulators, recent trauma (within last 60 days), bedridden > 3 days, uncommon travel (>

6hrs) within the last 3 months, paralysis of the symptomatic leg, or surgery within the last

4 weeks), concomitant diseases and medication use. In addition, body mass index (BMI)

was calculated. Cases were patients with thrombosis of the lower extremity confirmed by

compression ultrasonography, including proximal DVT (i.e. proximal thrombosis of the

77

Venous thrombosis is associated with hyperglycaemia at diagnosis: a case-control study

iliac or superficial femoral vein, calf vein thrombosis, involving at least the upper third part

of the deep calf veins), symptomatic calf vein thrombosis, and superficial

thrombophlebitis. The diagnosis was confirmed following a diagnostic management

strategy, based on the Wells’ criteria15 and a Tinaquant D-dimer assay (Roche diagnostics,

Basel, Switzerland), followed by compression ultrasound if indicated as validated and

described before.16 Controls were selected in a 2:1 ratio from those individuals in whom

thrombosis was ruled out using the above mentioned strategy. Selection was performed

randomly, only taking into account the male/female ratio of the cases.

Sample storage and laboratory analysis

On admission and prior to diagnostic testing, blood samples were drawn and collected in

tubes containing 0.109 M trisodium citrate. Within one hour after collection, platelet poor

plasma was obtained by centrifugation twice for 20 min at 1600 g and 4°C. The plasma was

stored in 2 ml cryovials containing 0.5 ml plasma at -80° C.

Glucose was measured using the HK/G-6PD method (Roche/Hitachi, Basel, Switzerland)

and corrected for the 10% dilution with sodium citrate. To assess whether elevated glucose

levels were related to an acute phase response induced by the thrombotic event itself, we

analyzed C-reactive protein (CRP) levels from 20 cases and 20 controls, randomly selected

from each quartile, hence in a total of 160 patients.

Statistical analysis

Results are presented as mean ± standard deviation (SD) or median with interquartile range

(IQR), depending on the observed distribution. The primary objective of this study was to

assess the relationship between glucose levels at presentation and VTE, which was

expressed in Odds Ratios (OR), with 95% confidence intervals. Glucose levels of the

controls were divided into quartiles, as distribution of glucose levels is typically non-

normally distributed. Subsequently, the cases were assigned to these quartiles according to

their admission glucose values. We used binary logistic regression. The regression model

was created based on clinical relevant potential confounders (BMI, concomitant known

diabetes mellitus, sex, ethnicity, age at diagnosis, and whether known risk factors for VTE

were present). Furthermore, ORs were calculated for cases in which the diagnosis of

thrombosis was restricted to DVT only, excluding calf vein thrombosis and superficial

thrombophlebitis. To assess the correlation between glucose levels and CRP levels we

78

Chapter 7

1

2

3

4

5

6

7

8

9

10

11

12

performed a scatter plot and calculated the correlation coefficient (expressed in r). All

statistical analyses were performed in SPSS version 15.0.

Results

In total 188 patients with confirmed thrombosis and 370 controls were included in this

study, as blood samples were not available for 2 cases and 10 controls. The baseline

characteristics of the two study groups are displayed in Table 1. Mean age and gender

distribution were comparable. The control group consisted of a smaller proportion of

white patients and a greater proportion of black patients as compared to the cases. The

mean BMI was higher in the control group, as tended to be the number of patients with

known diabetes mellitus. Median glucose levels were 5.9 mmol/l (IQR 5.3-6.6) in patients

with thrombosis, and 5.6 mmol/l (IQR 5.2-6.6) in the controls.

Table 1. Baseline characteristics of the two study groups

CCases (n=188) Controls (n=370)

Age, years (mean ±SD) 57±17 56±16 Female (%) 57.4 58.1 Ethnicity (%)

White Black Asian/Pacific islander Other

78.7 10.1 2.7 8.5

69.2 18.1 4.2 8.5

BMI kg/m² (median, IQR) 26.6 (23.9-29.1) 27.2 (24.2-31.3) Diabetes type 1 or 2 (%) 3.7 10.0 Glucose mmol/l (median, IQR) 5.6 (5.2-6.6) 5.9 (5.3-6.6)

Known risk factors: concomitant malignancy, pregnancy, use of Hormonal Replacement Therapy, Oral Contraceptives or selective oestrogen receptor modulators, recent trauma (within last 60 days), bedridden > 3 days, uncommon travel (> 6hrs) within the last 3 months, paralysis of the symptomatic leg, or surgery within the last 4 weeks

In thrombosis patients, 38% had one or more acquired risk factors and the distribution of

thrombosis was: 82% DVT, 6% calf vein thrombosis, and the remaining 12% had

superficial thrombophlebitis.

The cut-off glucose levels for the 1st quartile was < 5.3 mmol/l (n=141), the 2nd 5.3-5.7

mmol/l (n=134), the 3rd 5.7-6.6 mmol/l (n=139) and the 4th quartile � 6.6 mmol/l (n=144)

(Table 2a).

After adjustment for BMI, concomitant known diabetes mellitus, sex, ethnicity, age at

diagnosis, and whether known risk factors for VTE were present, the OR for thrombosis in

the 2nd, 3rd and 4th quartile of glucose levels compared to the 1st quartile were 1.40 (95%

Venous thrombosis is associated with hyperglycaemia at diagnosis: a case-control study

79

confidence interval (CI) 0.82-2.38), 1.69 (95 % CI 1.00-2.87) and 1.94 (95% CI 1.12-3.39),

respectively, P for trend = 0.01. The same trend of an increasing OR could be observed

when adjusting only for BMI, age, sex and known diabetes mellitus.

Table 2a. Odds Ratio (OR) for all thrombotic events (including calf vein thrombosis and superficial thrombophlebitis)

Quartile (glucose mmol/l)

Cases Controls Crude OR (95% CI)

Adjusted OR* (95% CI)

1st : < 5.3 39 102 reference reference 2nd : 5.3-5.7 46 88 1.37 (0.82-2.28) 1.40 (0.82-2.38) 3rd : 5.7-6.6 52 87 1.56 (0.94-2.59) 1.69 (1.00-2.87) 4th : � 6.6 51 93 1.43 (0.87-2.37) 1.94 (1.12-3.39) P for trend = 0.14 P for trend = 0.01

CI=confidence interval, * adjusted for BMI, known diabetes mellitus, sex, age, ethnicity and known risk factors As most risk factors predominantly relate to DVT a separate analysis was planned for DVT

only, thus excluding patients with superficial thrombophlebitis or calf vein thrombosis,

leaving 154 cases and 370 controls (Table 2b). Also here, the OR increases with glucose

levels: 1.59 (0.89-2.85), 2.04 (1.15-3.62) and 2.21 (1.20-4.05) for the 2nd, 3rd and 4th quartile

of glucose levels respectively (see Table 2b, P for trend = 0.001). Finally, the Spearman’s

rank correlation coefficient for CRP and plasma glucose was 0.09, with a P-value of 0.27.

Table 2b. Odds Ratio for deep venous thrombosis only

Quartile (glucose mmol/l)

Cases Controls Crude OR (95% CI)

Adjusted OR* (95% CI)

1st : < 5.3 28 102 reference reference 2nd : 5.3-5.7 37 88 1.53 (0.87-2.70) 1.59 (0.89-2.85) 3rd : 5.7-6.6 46 87 1.93 (1.11-3.34) 2.04 (1.15-3.62) 4th : � 6.6 43 93 1.68 (0.97-2.93) 2.21 (1.20-4.05) P for trend = 0.05 P for trend = 0.001

CI=confidence interval, * adjusted for BMI, known diabetes mellitus, sex, age, ethnicity and known risk factors

Discussion

In this case-control study, increased glucose levels measured at the time of presentation

were associated with venous thrombosis. This could be a relevant clinical concept as the

general population is becoming increasingly glucose intolerant. The relation between

glucose and DVT was not readily explained by an acute phase reaction due to the

thrombotic event itself.

Chapter 7

80

1

2

3

4

5

6

7

8

9

10

11

12

Whereas our results indicate that increased glucose levels and VTE coincide, it is

impossible with the current design to demonstrate a causal relationship. However, a causal

relationship seems plausible. To assess this one can apply the diagnostic criteria for

causation. 17

A causal relationship is supported by the available biological evidence from experiments in

humans. Stegenga and co-workers showed that experimentally induced acute

hyperglycaemia activates the coagulation system in healthy volunteers.14 From a

pathophysiological point of view, hyperglycaemia is known to induce coagulation

activation through glycocalyx damage18, up regulation of tissue factor19,20, non-enzymatic

glycation and the development of increased oxidative stress.21 Long term exposure to

hyperglycaemia such as in diabetes mellitus, is a known risk factor for VTE.22 In addition,

the effect of hyperglycaemia on coagulation seems modifiable in diabetes patients, as

treating hyperglycaemia among these patients lead to down regulation of coagulation

activation in several randomized controlled trials.23,24 Furthermore, our results are in line

with the findings by Mraovic and colleagues who demonstrated that hyperglycaemia

increases the risk of pulmonary embolism after major orthopaedic surgery.25 Thus, direct

and indirect evidence supports a possible association for acute and chronic hyperglycaemia

in the development of VTE. Furthermore, the association is consistent from this study to

other studies, which is in line with the criterion for repetitive demonstration of causality.

In this study, an adjusted OR of 2.21 (95 % CI 1.20-4.05) for DVT was observed in the

highest quartile which suggests a strong relationship. In comparison, the OR for the well

established risk factor for VTE, the prothrombin 20210A mutation is 2.8 (95% CI 1.4-5.6).26

The OR for venous thrombosis is increasing with increasing glucose levels, from 1.40

(0.82-2.38) in the 2nd quartile to 1.69 (1.00-2.87) in the 3rd quartile and 1.94 (95 % CI 1.12-

3.39) in the 4th quartile. We tested for differences in ORs amongst the quartiles of glucose

levels and found a significant linear trend (P=0.01). This is in concordance with a dose-

response gradient, another criterion for causality.

The question arises whether elevated glucose levels during a VTE result from the

inflammatory and counter regulatory hormone action initiated by the VTE event itself, or

whether hyperglycaemia preceded the VTE event. Although a significant proportion of the

patients with hyperglycaemia during an episode of VTE will have an undisturbed glucose

tolerance at follow-up,27 undiagnosed impaired glucose tolerance is likely to have been

present in a proportion of patients before the VTE event itself, which may therefore have

contributed to the development of thrombosis. We therefore suspect a temporal

81

Venous thrombosis is associated with hyperglycaemia at diagnosis: a case-control study

relationship. In addition, no correlation was found between the acute phase reaction,

measured by CRP, and glucose levels. The presence of stress hyperglycaemia during a

thrombotic event, independent of its cause, could be relevant: it has been shown to have

evident clinical consequences in patients with myocardial infarction and patients admitted

to the Intensive Care Unit (especially without known diabetes mellitus)28,29, although results

from recent intervention trials have been disappointing.30,31

Interestingly, we found a greater proportion of patients with diabetes in the control group

compared to the case group. In fact, this higher rate can be caused by referral bias. Patients

with diabetes are usually under chronic medical care and are prone to leg- and foot

problems which can resemble DVT, such as erysipelas. Thus, they are more easily referred

for suspicion of DVT. We had no information on the use of anti-diabetic drugs. Differences

in distribution of diabetes treatment in both cases and controls could be a source of bias.

However, we believe this effect to be limited. The model was adjusted for diabetes mellitus

as a potential confounder.

Our study has several limitations. Firstly, the control group consisted of patients with

complaints of their legs instead of healthy controls. Consequently, it may be possible that

glucose levels were increased in the controls due to an underlying disease such as infection.

However, this would have led to an underestimation of the association between glucose

and DVT. Secondly, the glucose levels were measured on admission, and it was unknown

whether these were fasting or non-fasting samples. However, due to the study design there

were no differences between cases and controls with respect to the time of presentation and

therefore larger dispersion of glucose levels would have affected cases as much as controls.

Thirdly, citrated plasma is not the plasma of choice for determining glucose and CRP

levels, because of the dilution with sodium citrate. Also, sodium citrate does not inhibit ex-

vivo glycolysis. However, we corrected glucose levels for the dilution and the obtained

blood samples were centrifuged and stored within one hour thereby minimizing glycolysis.

Again, as blood samples of both cases and controls were obtained and processed in the

same manner, the results of glucose levels were affected equally. Since this is the first

report about the association between stress-hyperglycaemia and VTE, confirmation in

other studies is required.

82

Chapter 7

1

2

3

4

5

6

7

8

9

10

11

12

In conclusion, our findings suggest that higher glucose levels are a risk factor for the

development of venous thrombosis. It will therefore be of importance to analyse whether

disturbed glucose homeostasis persists after the acute phase of venous thrombosis.

Acknowledgements

We would like to acknowledge Professor M.M. Levi for assessing the manuscript and for

his significant intellectual contribution. In addition we would like to acknowledge the

“vasculists”, a group of medical students responsible for execution of this study, ranging

from designing of the study to recruitment and data collection. References 1. Cushman M, Tsai AW, White RH, Heckbert SR, Rosamond WD, Enright P, Folsom AR. Deep vein

thrombosis and pulmonary embolism in two cohorts: the longitudinal investigation of thromboembolism etiology. The American Journal of Medicine 2004; 117:19-25.

2. Rosendaal FR. Venous thrombosis: a multicausal disease. The Lancet 1999; 353:1167-73. 3. Lensing AW, Prandoni P, Prins MH, Buller HR. Deep-vein thrombosis. The Lancet 1999; 353:479-85. 4. Spencer FA, Emery C, Lessard D, Anderson F, Emani S, Aragam J, Becker RC, Goldberg RJ. The Worcester

Venous Thromboembolism study: a population-based study of the clinical epidemiology of venous thromboembolism. J Gen Intern Med 2006; 21:722-7.

5. Eliasson A, Bergqvist D, Bjorck M, Acosta S, Sternby NH, Ogren M. Incidence and risk of venous thromboembolism in patients with verified arterial thrombosis: a population study based on 23 796 consecutive autopsies. Journal of Thrombosis and Haemostasis 2006; 4:1897-902.

6. Sorensen HT, Horvath-Puho E, Pedersen L, Baron JA, Prandoni P. Venous thromboembolism and subsequent hospitalisation due to acute arterial cardiovascular events: a 20-year cohort study. Lancet 2007; 370:1773-9.

7. Bruno A, Levine SR, Frankel MR, Brott TG, Lin Y, Tilley BC, Lyden PD, Broderick JP, Kwiatkowski TG, Fineberg SE. Admission glucose level and clinical outcomes in the NINDS rt-PA Stroke Trial. Neurology 2002; 59:669-74.

8. Malmberg K, Ryden L, Hamsten A, Herlitz J, Waldenstrom A, Wedel H. Effects of insulin treatment on cause-specific one-year mortality and morbidity in diabetic patients with acute myocardial infarction. DIGAMI Study Group. Diabetes Insulin-Glucose in Acute Myocardial Infarction. Eur Heart J 1996; 17:1337-44.

9. Laakso M. Hyperglycemia and cardiovascular disease in type 2 diabetes. Diabetes 1999; 48:937-42. 10. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ,

Smith SC, Jr., Spertus JA, Costa F. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation 2005; 112:2735-52.

11. Ay C, Tengler T, Vormittag R, Simanek R, Dorda W, Vukovich T, Pabinger I. Venous thromboembolism a manifestation of the metabolic syndrome. Haematologica 2007; 92:374-80.

12. Petrauskiene V, Falk M, Waernbaum I, Norberg M, Eriksson JW. The risk of venous thromboembolism is markedly elevated in patients with diabetes. Diabetologia 2005; 48:1017-21.

13. Grant PJ. Diabetes mellitus as a prothrombotic condition. J Intern Med 2007; 262:157-72. 14. Stegenga ME, van der Crabben SN, Levi M, de Vos AF, Tanck MW, Sauerwein HP, van der Poll T.

Hyperglycemia Stimulates Coagulation, Whereas Hyperinsulinemia Impairs Fibrinolysis in Healthy Humans. Diabetes 2006; 55:1807-12.

15. Wells PS, Anderson DR, Bormanis J, Guy F, Mitchell M, Gray L, Clement C, Robinson KS, Lewandowski B. Value of assessment of pretest probability of deep-vein thrombosis in clinical management. The Lancet 350:1795-8.

83

Venous thrombosis is associated with hyperglycaemia at diagnosis: a case-control study

16. Kraaijenhagen RA, Piovella F, Bernardi E, Verlato F, Beckers EA, Koopman MM, Barone M, Camporese G, Potter Van Loon BJ, Prins MH, Prandoni P, Buller HR. Simplification of the diagnostic management of suspected deep vein thrombosis. Arch Intern Med 2002; 162:907-11.

17. How to read clinical journals: IV. To determine etiology or causation. Can Med Assoc J 1981; 124:985-90. 18. Nieuwdorp M, van Haeften TW, Gouverneur MC, Mooij HL, van Lieshout MH, Levi M, Meijers JC, Holleman

F, Hoekstra JB, Vink H, Kastelein JJ, Stroes ES. Loss of endothelial glycocalyx during acute hyperglycemia coincides with endothelial dysfunction and coagulation activation in vivo. Diabetes 2006; 55:480-6.

19. Khechai F, Ollivier V, Bridey F, Amar M, Hakim J, de Prost D. Effect of advanced glycation end product-modified albumin on tissue factor expression by monocytes. Role of oxidant stress and protein tyrosine kinase activation. Arterioscler Thromb Vasc Biol 1997; 17:2885-90.

20. Min C, Kang E, Yu SH, Shinn SH, Kim YS. Advanced glycation end products induce apoptosis and procoagulant activity in cultured human umbilical vein endothelial cells. Diabetes Res Clin Pract 1999; 46:197-202.

21. Ceriello A. Coagulation activation in diabetes mellitus: the role of hyperglycaemia and therapeutic prospects. Diabetologia 1993; 36:1119-25.

22. Ageno W, Becattini C, Brighton T, Selby R, Kamphuisen PW. Cardiovascular Risk Factors and Venous Thromboembolism: A Meta-Analysis. Circulation 2008; 117:93-102.

23. Caballero AE, Delgado A, Guilar-Salinas CA, Herrera AN, Castillo JL, Cabrera T, Gomez-Perez FJ, Rull JA. The Differential Effects of Metformin on Markers of Endothelial Activation and Inflammation in Subjects with Impaired Glucose Tolerance: A Placebo-Controlled, Randomized Clinical Trial. J Clin Endocrinol Metab 2004; 89:3943-8.

24. De Jager J, Kooy A, Lehert P, Bets D, Wulffele MG, Teerlink T, Scheffer PG, Schalkwijk CG, Donker AJ, Stehouwer CD. Effects of short-term treatment with metformin on markers of endothelial function and inflammatory activity in type 2 diabetes mellitus: a randomized, placebo-controlled trial. J Intern Med 2005; 257:100-9.

25. Mraovic B, Hipszer BR, Epstein RH, Pequignot EC, Parvizi J, Joseph JI. Preadmission Hyperglycemia is an Independent Risk Factor for In-Hospital Symptomatic Pulmonary Embolism After Major Orthopedic Surgery. J Arthroplasty 2008.

26. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3'-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996; 88:3698-703.

27. Ishihara M, Inoue I, Kawagoe T, Shimatani Y, Kurisu S, Hata T, Nakama Y, Kijima Y, Kagawa E. Is admission hyperglycaemia in non-diabetic patients with acute myocardial infarction a surrogate for previously undiagnosed abnormal glucose tolerance? Eur Heart J 2006; 27:2413-9.

28. Ishihara M, Kojima S, Sakamoto T, Asada Y, Tei C, Kimura K, Miyazaki S, Sonoda M, Tsuchihashi K, Yamagishi M, Ikeda Y, Shirai M, Hiraoka H, Inoue T, Saito F, Ogawa H. Acute hyperglycemia is associated with adverse outcome after acute myocardial infarction in the coronary intervention era. Am Heart J 2005; 150:814-20.

29. Whitcomb BW, Pradhan EK, Pittas AG, Roghmann MC, Perencevich EN. Impact of admission hyperglycemia on hospital mortality in various intensive care unit populations. Crit Care Med 2005; 33:2772-7.

30. Mehta SR, Yusuf S, Diaz R, Zhu J, Pais P, Xavier D, Paolasso E, Ahmed R, Xie C, Kazmi K, Tai J, Orlandini A, Pogue J, Liu L. Effect of glucose-insulin-potassium infusion on mortality in patients with acute ST-segment elevation myocardial infarction: the CREATE-ECLA randomized controlled trial. JAMA 2005; 293:437-46.

31. NICE-SUGAR Study Investigators, Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, Bellomo R, Cook D, Dodek P, Henderson WR, Hebert PC, Heritier S, Heyland DK, McArthur C, McDonald E, Mitchell I, Myburgh JA, Norton R, Potter J, Robinson BG et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009; 360:1283-97

84

Chapter 7

Chapter 8

Stress-induced hyperglycaemia and venous thromboembolism following total hip or total knee arthroplasty

Cohn DM, Hermanides J, De Vries JH, Kamphuisen PW, Kuhls S, Homering M, Hoekstra JBL, Lensing AWA, Büller HR. To be submitted.

Abstract

Background

Stress-induced hyperglycaemia is common during orthopaedic surgery. In addition,

hyperglycaemia has recently been shown to activate coagulation. Whether stress-induced

hyperglycaemia is associated with symptomatic or asymptomatic venous

thromboembolism (VTE), is unknown.

Objective

To investigate whether hyperglycaemia measured pre-surgery (day 0) and post-surgery

(day1) is associated with an increased risk of VTE.

Methods

We performed post-hoc analyses in the four RECORD studies (REgulation of Coagulation

in major Orthopaedic surgery reducing the Risk of Deep venous thrombosis and pulmonary

embolism). Separate analyses were performed for patients undergoing total hip or knee

replacement. Outcome measures were symptomatic VTE and “total VTE” (defined as the

composite of symptomatic VTE, asymptomatic DVT and all cause mortality) during the

predefined treatment periods. Glucose levels were measured at day 0 and day 1 and

categorized into quartiles, based on the distribution in the respective cohorts. The

influence of glucose and other relevant covariates on VTE was assessed by stepwise

selection (entry/stay level P=0.20) in multivariate analyses. Variables for treatment group

and study were forced into the model. Adjusted odds ratios (ORs) and 95% confidence

intervals were calculated.

Results

A total of 12,383 patients were eligible for assessment of symptomatic VTE and 8,512

patients were eligible for assessment of total VTE. Glucose levels measured at day 1 were

associated with both symptomatic and total VTE in patients undergoing hip surgery,

adjusted OR highest versus lowest quartile 2.6 (95%CI 1.0 to 6.6) and 2.0 (95%CI 1.3 to 3.1),

respectively. No association between hyperglycaemia and knee replacement was observed.

Conclusion

Hyperglycaemia following total hip replacement was associated with symptomatic and

asymptomatic VTE. This was, however, not observed in patients total undergoing knee

replacement, which is likely due to the surgical procedure.

Chapter 8

86

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Venous thromboembolism (VTE) may manifest as either deep venous thrombosis (DVT) or

pulmonary embolism (PE), or a combination of both. VTE affects 2-3 persons per 1,000

inhabitants annually in Western Societies.1 Acquired risk factors can be identified in

approximately 50% of VTE patients, and, in addition, one or more inherited risk factors can

be demonstrated in an equal proportion of patients.2,3 This implies, that in approximately

25% of patients with VTE no risk factors can be identified.

Hyperglycaemia has recently been shown to be associated with VTE in an outpatient

population.4 In several reports, acute hyperglycaemia was shown to cause coagulation

activation in an experimental setting as demonstrated by increased levels of thrombin-

antithrombin complexes and soluble tissue factor.5,6 During acute illness or surgery, a

phenomenon called stress hyperglycaemia may occur independently of the presence of

known diabetes.7 Indeed, hip surgery has been shown to induce hyperglycaemia peaking

the days after the procedure, followed by post-operative procoagulant activity peaking on

the third and fourth postoperative days.8 Whether stress-hyperglycaemia is associated with

VTE in hospitalized patients, such as orthopaedic patients, has not been previously

addressed. The associated risk of hyperglycaemia could be of particular interest in surgical

patients, since the coagulation system is already activated due to the surgical procedure.9

Furthermore, VTE is a frequently occurring complication following orthopaedic surgery.

Despite anticoagulant prophylaxis, symptomatic VTE occurs in approximately 2% of all

patients undergoing hip or knee replacements.10 In this study we aimed to assess whether

pre- and postoperative hyperglycaemia is associated with (a)symptomatic VTE.

Methods

Patients

For the present analysis we made use of the large phase III trial programme: “REgulation of

Coagulation in major Orthopaedic surgery reducing the Risk of Deep venous thrombosis

and pulmonary embolism studies” (RECORD 1-4).11-14 These trials compared the efficacy

and safety of Rivaroxaban (a direct factor Xa inhibitor) 10 mg o.d. relative to standard

treatment with Enoxaparin 40 mg o.d.11-13 or 30 mg b.i.d.14 As efficacy and safety outcomes,

as well as independent, blinded adjudication committees were the same in all RECORD

studies, the studies allow pooling of the data. Patients were eligible if they were aged 18

years or older and were scheduled for elective total hip or total knee arthroplasty. Major

exclusion criteria included active bleeding or high risk of bleeding, significant liver disease,

Stress-induced hyperglycaemia and VTE following total hip or total knee arthroplasty

87

anticoagulant therapy that could not be interrupted, use of HIV-protease-inhibitors and a

contraindication for prophylaxis with enoxaparin or a condition that would require a dose

adjustment for enoxaparin. Detailed in- and exclusion criteria were reported in the original

publications of the RECORD studies.11-14

Procedures

Deep-vein thrombosis was assessed per protocol, by ascending, bilateral venography with a

standardized technique.15 This was performed at day 36 (range 32 to 40) in patients with

total hip replacements (RECORD 1 and 2)11,12 and at day 13 (range 11 to 15) in patients with

total knee replacement (RECORD 3 and 4).13,14 Symptomatic VTE was objectively confirmed

by standard imaging techniques. Treatment duration was 5 weeks for RECORD study 1 and

2 (including the 3 weeks of placebo treatment in the comparator group). Treatment

duration in RECORD 3 and 4 was 2 weeks. Patients were followed up for 30–35 days after

the last dose. Glucose levels were sampled upon admission (day 0) and 6 hours after

surgery (day 1). All samples were analyzed in the central laboratory.

Outcomes

The aim of this study was to assess whether hyperglycaemia (measured pre- and post-

surgery) was associated with an increased risk of VTE. We assessed the effect of

hyperglycaemia on symptomatic VTE in the participants included in the safety population

of the four RECORD studies. The safety population was defined as patients who had

undergone surgery and had received at least one dose of study medication. Furthermore, we

investigated the association between hyperglycaemia and total VTE, which was defined as

the composite outcome of symptomatic VTE and asymptomatic DVT as detected by per

protocol venography and all cause mortality. The analyses for total VTE were performed in

the modified intention to treat (mITT) population, compromising all patients from the

safety analyses with adequate assessment of both proximal and distal veins on venography.

Only events occurring in the treatment phase of the studies were considered.

Statistical Analyses

Descriptive results are presented as mean ± standard deviation (SD) or median with

interquartile range (IQR).

To investigate the association of hyperglycaemia and VTE multiple logistic regression

analyses were performed separately for patients undergoing total hip and total knee

Chapter 8

88

1

2

3

4

5

6

7

8

9

10

11

12

arthroplasty. Separate analyses were carried out to assess the influence of glucose on Day 0

(pre-surgery) and Day 1 (post-surgery), and of the difference in glucose levels (Day 1-Day 0)

on symptomatic and total VTE. In all analyses glucose levels were categorized into quartiles,

based on the distribution in the respective cohorts. Glucose quartiles and other relevant

covariates (BMI (<25, 25-<35, �35), age (<65, 65-75, �75), gender and concomitant known

diabetes mellitus) were screened via stepwise selection (entry/stay level P=0.20) for the

creation of the logistic regression models. Variables for treatment group and study were

not screened but forced into the model. Odds ratios (OR) and p-values of the likelihood

ratio test for testing whether the selected variables have a significant influence on the

outcome were calculated. All statistical analyses were performed in SAS version 9.1 (SAS

institute, Cary, NC). All outcomes are reported as adjusted ORs.

Table 1. Baseline characteristics of both study cohorts

TTotal hip arthroplasty (RECORD 1&2)11,12 (n=6890)

Total knee arthroplasty (RECORD 3&4)13,14 (n=5493)

Age – years (mean ± SD) 63 (12) 66 (10) Sex, female n (%) 3780 (55%) 3652 (66%) Body-mass index – kg/m² (mean ± SD) 28 (5) 30 (6) History of VTE, n (%) 132 (2%) 156 (3%) Previous orthopaedic surgery, n (%) 1447 (21%) 1690 (31%) Type of surgery, n (%)

Primary Revision Missing/no surgery

6562 (95%) 254 (4%) 74 (1%)

5329 (97%) 119 (2%) 45 (1%)

Type of anaesthesia General only General and regional Regional only Missing None

1983 (29%) 619 (9%) 4215 (61%) 73 (1%) 0 (0%)

1034 (19%) 1126 (21%) 3292 (60%) 0 (0%) 41 (1%)

Duration of surgery – minutes (mean ± SD) 98 (48) 99 (40) Time to mobilization – days (mean ± SD) 2.2 (4) 1.6 (2) Ethnic origin, n (%)

White Asian Hispanic Black Other/missing

5687 (83%) 498 (7%) 329 (5%) 103 (1%) 273 (4%)

4037 (73%) 736 (14%) 353 (6%) 181 (3%) 186 (3%)

Randomization allocation, n (%) Rivaroxaban Comparator (enoxaparin)

3437 (50%) 3453 (50%)

2746 (50%) 2747 (50%)

On glucose lowering therapy, n (%) 407 (6%) 654 (12%) Diabetes Mellitus n (%) 532 (8%) 890 (16%)

VTE=venous thromboembolism, SD=standard deviation, N.A.=not available.

Stress-induced hyperglycaemia and VTE following total hip or total knee arthroplasty

89

Results

Baseline characteristics

The safety population of the four RECORD studies consisted of 12,383 patients, of whom

6,890 underwent total hip arthroplasty and 5,493 total knee arthroplasty, respectively. The

baseline characteristics of both cohorts are shown in Table 1. High BMI and female gender

were more common in the knee arthroplasty cohort (mean BMI 30.3 kg/m2 vs. 27.5 kg/m2,

and 66% vs. 55%, respectively). Mean age was slightly higher in the knee surgery cohort: 66

years vs. 63 years. The mITT population comprised 8,512 patients (hip replacement 4,886

patients, knee replacement 3,626 patients).

Total hip arthroplasty

Median pre-operative glucose level (day 0) was 97 mg/dL (IQR 88-114) (Table 2). Glucose

levels on Day 0 were not associated with either symptomatic VTE or total VTE, as the

variable “glucose” was not maintained in the regression model following stepwise

selection regression analysis (Table 2).

At day 1 (post surgery), the median glucose level was 117 mg/dL (IQR 99 to 142). Glucose

levels in the highest quartile (>142 mg/dL) were associated with both symptomatic VTE and

total VTE, when compared to the lowest glucose quartile (�99 mg/dL): adjusted OR 2.6

(95%CI 1.0 to 6.6) and OR 2.0 (95%CI 1.3 to 3.1), respectively. The p-value of the

Likelihood Ratio test for glucose on Day 1 was significant for both outcomes (p=0.0142 for

symptomatic VTE and p<0.0001 for total VTE). In addition, the amount of increase of the

glucose level between day 1 and day 0 was associated with total VTE (p=0.0082). Median

difference of glucose levels between these two time points was 16 mg/dL (IQR -5 to 40).

The highest quartile of this difference was associated with a nearly twofold increased risk

for total VTE compared to the lowest quartile (adjusted OR 1.8, 95%CI 1.2 to 2.8). The

adjusted OR for symptomatic VTE was not maintained in the regression model (see Table

2).

Total knee arthroplasty

The distribution in glucose levels at day 0 and day 1 in patients undergoing knee

arthroplasty was comparable to that in the hip surgery cohort. Median glucose was 99

mg/dL (IQR 88 to 115) at day 0 and 123 mg/dL (IQR 103 to 150) at day 1 (Table 3).

For glucose levels measured at day 0, there was no association between glucose levels and

symptomatic VTE.

Chapter 8

90

1

2

3

4

5

6

7

8

9

10

11

12

Table 2. Odds Ratio for glucose levels measured at day 0, day 1 and difference in glucose levels between day 1 and day 0 in relation to symptomatic VTE and total VTE in hip surgery patients (RECORD 1&2)11,12

quartiles glucose levels measured at day 0 (mg/dL) �88 >88 to 97 >97 to 114 >114 all patients (N) 1487 1713 2014 1630 symptomatic VTE (n) 5 13 11 6

crude OR (95%CI) 1 (ref) 1.2 (0.6 to 2.5) 0.8 (0.4 to 1.8) 1.3 (0.6 to 2.8) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

patients valid for the modified intention to treat analyses (N) 1051 1220 1449 1141 total VTE (n) 37 41 60 35

crude OR (95%CI) 1 (ref) 1.2 (0.9 to 1.7) 1.1 (0.8 to 1.6) 1.5 (1.1 to 2.0) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

quartiles glucose levels measured at day 1 (mg/dL) �99 >99 to 117 >117 to 142 >142 all patients (N) 1590 1722 1671 1697 symptomatic VTE (n) 6 4 6 17

crude OR (95%CI) 1 (ref) 0.6 (0.2 to 2.2) 0.9 (0.3 to 2.9) 2.6 (1.0 to 6.6) adjusted OR (95%CI) 1 (ref) 0.6 (0.2 to 2.2) 0.9 (0.3 to 2.9) 2.6 (1.0 to 6.6) LR test p=0.0142

patients valid for the modified intention to treat analyses (N) 1148 1241 1219 1210 total VTE (n) 32 31 33 75

crude OR (95%CI) 1 (ref) 0.9 (0.5 to 1.4) 0.9 (0.6 to 1.5) 2.2 (1.4 to 3.4) adjusted OR (95%CI) 1 (ref) 0.9 (0.5 to 1.4) 0.9 (0.5 to 1.5) 2.0 (1.3 to 3.1) LR test p<0.0001

quartiles difference in glucose levels day 1 - day 0 (mg/dL) �-5 >-5 to 16 >16 to 40 >40 all patients (N) 1633 1655 1699 1661 symptomatic VTE (n) 5 6 8 14

crude OR (95%CI) 1 (ref) 1.2 (0.4 to 3.9) 1.5 (0.5 to 4.6) 2.7 (1.0 to 7.5) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

patients valid for the modified intention to treat analyses (N) 1169 1192 1241 1197 total VTE (n) 33 30 43 64

crude OR (95%CI) 1 (ref) 0.9 (0.5 to 1.5) 1.2 (0.7 to 1.8) 1.8 (1.1 to 2.8) adjusted OR (95%CI) 1 (ref) 0.9 (0.6 to 1.6) 1.2 (0.7 to 1.9) 1.8 (1.2 to 2.8) LR test p=0.0082

VTE=venous thromboembolism; OR=Odds Ratio, LR test=Likelihood Ratio test for testing whether increase in glucose levels (in quartiles) has a significant influence on the occurrence of VTE. Total VTE is the composite of symptomatic VTE, asymptomatic DVT and all cause mortality.

Again, the variable “glucose” was not maintained in the regression model (Table 3). A non-

significant trend was observed for total VTE at day 0: adjusted OR 1.4 (95%CI 1.0 to 1.9,

Likelihood Ratio test: p=0.16). Furthermore, glucose levels measured at day 1 were not

significantly associated with symptomatic VTE: adjusted OR 1.6 (95%CI 0.7 to 3.6), nor

with total VTE (see Table 3). In addition, no association was found between the increase in

glucose levels (median difference 19 mg/dL, IQR -3 to 46) and VTE: adjusted OR for

symptomatic VTE: 0.8 (95%CI 0.3 to 1.8), the variable “glucose” was not maintained in the

regression model for total VTE (Table 3).

Stress-induced hyperglycaemia and VTE following total hip or total knee arthroplasty

91

Table 3. Odds Ratio for glucose levels measured at day 0, day 1 and difference in glucose levels between day 1 and day 0 in relation to symptomatic VTE and total VTE in knee surgery patients (RECORD 3&4)13,14

quartiles glucose levels measured at day 0 (mg/dL) �88 >88 to 99 >99 to 115 >115 all patients (N) 1216 1480 1320 1417 symptomatic VTE (n) 12 18 11 19

crude OR (95%CI) 1 (ref) 1.2 (0.6 to 2.5) 0.8 (0.4 to 1.8) 1.3 (0.6 to 2.8) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

patients valid for the modified intention to treat analyses (N) 813 996 878 903 total VTE (n) 72 115 93 124

crude OR (95%CI) 1 (ref) 1.2 (0.9 to 1.7) 1.1 (0.8 to 1.6) 1.5 (1.1 to 2.0) adjusted OR (95%CI) 1 (ref) 1.2 (0.9 to 1.7) 1.1 (0.8 to 1.5) 1.4 (1.0 to 1.9) LR test p=0.16

quartiles glucose levels measured at day 1 (mg/dL) �103 >103 to 123 >123 to 150 >150 all patients (N) 1476 1300 1266 1281 symptomatic VTE (n) 10 22 12 14

crude OR (95%CI) 1 (ref) 2.5 (1.2 to 5.3) 1.4 (0.6 to 3.2) 1.6 (0.7 to 3.6) adjusted OR (95%CI) 1 (ref) 2.5 (1.2 to 5.3) 1.4 (0.6 to 3.2) 1.6 (0.7 to 3.6) LR test p=0.08

patients valid for the modified intention to treat analyses (N) 1005 876 846 831 total VTE (n) 103 93 99 107

crude OR (95%CI) 1 (ref) 1.0 (0.8 to 1.4) 1.1 (0.8 to 1.5) 1.3 (0.9 to 1.7) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

quartiles difference in glucose levels day 1 - day 0 (mg/dL) �-3 >-3 to 19 >19 to 46 >46 all patients (N) 1331 1304 1347 1290 symptomatic VTE (n) 13 21 13 10

crude OR (95%CI) 1 (ref) 1.6 (0.8 to 3.3) 1.0 (0.4 to 2.1) 0.8 (0.3 to 1.8) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

patients valid for the modified intention to treat analyses (N) 887 899 902 840 total VTE (n) 100 100 102 95

crude OR (95%CI) 1 (ref) 1.0 (0.8 to 1.4) 1.0 (0.8 to 1.4) 1.0 (0.8 to 1.4) adjusted OR (95%CI) glucose not maintained in the regression model LR test p>0.20

VTE=venous thromboembolism; OR=Odds Ratio, LR test=Likelihood Ratio test for testing whether increase in glucose levels (in quartiles) has a significant influence on the occurrence of VTE. Total VTE is the composite of symptomatic VTE, asymptomatic DVT and all cause mortality.

Discussion

This study shows an association between post–operative glucose levels and both

symptomatic VTE and total VTE in patients undergoing total hip arthroplasty. In addition,

the increase of glucose levels from pre to post operative was also significantly associated

with total VTE in these patients. In contrast, after adjustment for potential confounders,

these associations were not found in patients undergoing knee arthroplasty. Only a non-

significant trend was observed for total VTE and glucose levels, measured either at day 0 or

day 1.

Chapter 8

92

1

2

3

4

5

6

7

8

9

10

11

12

The association between preoperatively elevated glucose levels and VTE in patients

undergoing orthopaedic surgery has previously been reported.16 Mraovic and colleagues

found that preadmission glucose levels exceeding 200 mg/dL independently increased the

risk of pulmonary embolism (OR=3.19, 95%CI 1.25 to 8.10) when compared to glucose

levels of less than 110mg/dL in patients who underwent hip or knee arthroplasty. We did

not find a relation between preoperative hyperglycaemia and VTE. However, 200 mg/dL is

well above the cut-off value of the highest quartile in our cohort, 145 mg/dL, and is likely to

reflect patients with undiagnosed diabetes mellitus before surgery, a known risk factor for

postsurgical complications17 We also assessed the influence of post-operative stress-

induced glucose increase in relation to the development of VTE to investigate whether

stress-hyperglycaemia is an independent risk factor for VTE following orthopaedic surgery.

Available pathophysiological evidence supports the relation between (acute)

hyperglycaemia and hypercoagulability. In patients with diabetes mellitus, the

concentration of several procoagulant factors are increased (fibrinogen, von Willebrand

antigen, factor VII antigen, factor VIII) and antifibrinolytic factors are decreased, such as

plasminogen activator inhibitor-1 (PAI-1).18 Furthermore, hip surgery has been shown to

induce hyperglycaemia, which preceded a rise of factor VIII clotting activity, von

Willebrand ristocetin cofactor activity, von Willebrand factor antigen and prothrombin

fragment 1+2.8 In healthy volunteers, acute hyperglycaemia activates the coagulation

system in an experimental setting.5

In this study, only those hip surgery patients in the highest quartile of post-operative stress-

induced hyperglycaemia were at increased risk for VTE, instead of a linear increase in risk.

This implies that glucose levels apparently need to exceed a certain threshold to reach a

significant association with VTE. Interestingly, the cut-off between the third and fourth

quartiles, 142 mg/dL, is close to the classic cut-off for impaired glucose tolerance following

the glucose tolerance test, 140 mg/dL.19

Hyperglycaemia was not clearly associated with VTE in patients undergoing TKR, in

contrast with patients undergoing THR. This discrepancy cannot be explained by major

differences in patients’ characteristics, as these were included as covariates in the

regression model. The most important factor may concern the surgical procedure. TKR is

performed with application of a tourniquet, occluding arterial and venous flow.20

Tourniquet use is related to the formation of thrombi.21-24 It is thus possible that the effect

of hyperglycaemia is completely outweighed by other risk factors for VTE in patients

undergoing TKR.

Stress-induced hyperglycaemia and VTE following total hip or total knee arthroplasty

93

Although preoperative glucose samples were drawn at predefined time points, patients

were not mandatory in a fasting state. We have, however, no reason to assume that there

was a difference in distribution in fasting/non-fasting samples in patients with- or without

VTE. Furthermore, this substudy involved a non-prespecified analysis. Nevertheless, we

have investigated a prospective database, with a very large sample size. In addition, the

most important assessments, VTE and glucose, were collected prospectively and

comprehensively. The outcome “total VTE” not only included symptomatic VTE and

asymptomatic DVT, but also “all cause mortality”. As the number of subjects who died

during follow up is very low in the four RECORD studies (n=23/8,512, 0.3%), it is not likely

that the inclusion of the latter outcome measure affected the results.

In conclusion, stress-induced postoperative hyperglycaemia is associated with

asymptomatic and symptomatic VTE following hip replacement. Future studies should

assess whether the risk of VTE can be decreased by glucose lowering therapy in patients

with postoperative hyperglycaemia. References 1. Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrom J. Incidence and

mortality of venous thrombosis: a population-based study. J Thromb Haemost 2007;5:692-9. 2. Lensing AW, Prandoni P, Prins MH, Büller HR. Deep-vein thrombosis. Lancet 1999;353:479-85. 3. Spencer FA, Emery C, Lessard D et al. The Worcester Venous Thromboembolism study: a population-based

study of the clinical epidemiology of venous thromboembolism. J Gen Intern Med 2006;21:722-7. 4. Hermanides J, Cohn DM, Devries JH et al. Venous thrombosis is associated with hyperglycaemia at diagnosis:

a case-control study. J Thromb Haemost 2009;7:945-9. 5. Stegenga ME, van der Crabben SN, Levi M et al. Hyperglycaemia stimulates coagulation, whereas

hyperinsulinemia impairs fibrinolysis in healthy humans. Diabetes 2006;55:1807-12. 6. Stegenga ME, van der Crabben SN, Blumer RM et al. Hyperglycaemia enhances coagulation and reduces

neutrophil degranulation, whereas hyperinsulinemia inhibits fibrinolysis during human endotoxemia. Blood 2008;112:82-9.

7. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet 2009;373:1798-807. 8. Hermanides J, Huijgen R, Henny CP et al. Hip surgery sequentially induces stress hyperglycaemia and

activates coagulation. Neth J Med 2009;67:226-9. 9. Galster H, Kolb G, Kohsytorz A, Seidlmayer C, Paal V. The pre-, peri-, and postsurgical activation of

coagulation and the thromboembolic risk for different risk groups. Thromb Res 2000;100:381-8. 10. Warwick D, Friedman RJ, Agnelli G et al. Insufficient duration of venous thromboembolism prophylaxis

after total hip or knee replacement when compared with the time course of thromboembolic events: findings from the Global Orthopaedic Registry. J Bone Joint Surg Br 2007;89:799-807.

11. Eriksson BI, Borris LC, Friedman RJ et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty. N Engl J Med 2008;358:2765-75.

12. Kakkar AK, Brenner B, Dahl OE et al. Extended duration rivaroxaban versus short-term enoxaparin for the prevention of venous thromboembolism after total hip arthroplasty: a double-blind, randomised controlled trial. Lancet 2008;372:31-9.

13. Lassen MR, Ageno W, Borris LC et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty. N Engl J Med 2008;358:2776-86.

14. Turpie AG, Lassen MR, Davidson BL et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty (RECORD4): a randomised trial. Lancet 2009;373:1673-80.

15. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch Surg 1972;104:134-44.

Chapter 8

94

1

2

3

4

5

6

7

8

9

10

11

12

16. Mraovic B, Hipszer BR, Epstein RH, Pequignot EC, Parvizi J, Joseph JI. Preadmission Hyperglycaemia is an Independent Risk Factor for In-Hospital Symptomatic Pulmonary Embolism After Major Orthopedic Surgery. J Arthroplasty 2008.

17. Marchant MH, Jr., Viens NA, Cook C, Vail TP, Bolognesi MP. The impact of glycemic control and diabetes mellitus on perioperative outcomes after total joint arthroplasty. J Bone Joint Surg Am 2009;91:1621-9.

18. Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15:44-54. 19. Diagnosis and classification of diabetes mellitus. Diabetes Care 2008;31 Suppl 1:S55-S60. 20. Sharrock NE, Go G, Sculco TP, Ranawat CS, Maynard MJ, Harpel PC. Changes in circulatory indices of

thrombosis and fibrinolysis during total knee arthroplasty performed under tourniquet. J Arthroplasty 1995;10:523-8.

21. Parmet JL, Horrow JC, Pharo G, Collins L, Berman AT, Rosenberg H. The incidence of venous emboli during extramedullary guided total knee arthroplasty. Anesth Analg 1995;81:757-62.

22. Parmet JL, Horrow JC, Berman AT, Miller F, Pharo G, Collins L. The incidence of large venous emboli during total knee arthroplasty without pneumatic tourniquet use. Anesth Analg 1998;87:439-44.

23. Westman B, Weidenhielm L, Rooyackers O, Fredriksson K, Wernerman J, Hammarqvist F. Knee replacement surgery as a human clinical model of the effects of ischaemia/reperfusion upon skeletal muscle. Clin Sci (Lond) 2007;113:313-8.

24. Carles M, Dellamonica J, Roux J et al. Sevoflurane but not propofol increases interstitial glycolysis metabolites availability during tourniquet-induced ischaemia-reperfusion. Br J Anaesth 2008;100:29-35.

Stress-induced hyperglycaemia and VTE following total hip or total knee arthroplasty

95

Chapter 9

Common genetic variation at the Endothelial Lipase (LIPG) locus and the risk of coronary artery disease and deep venous thrombosis

Vergeer M, Cohn DM, Boekholdt SM, Sandhu MS, Prins H, Ricketts SL, Wareham, NJ, Kastelein JJP, Khaw KT, Kamphuisen PW, Dallinga-Thie GM Submitted for publication.

Abstract

Background

Low levels of high-density lipoprotein cholesterol (HDL-C) are a risk factor for coronary

artery disease (CAD) and possibly for deep venous thrombosis (DVT). Endothelial lipase is

involved in HDL-C metabolism. Common variants in the endothelial lipase gene (LIPG)

have been reported to be associated with HDL-C levels and atherothrombosis, but these

findings were not consistent.

Objectives

We determined whether 5 tagging single nucleotide polymorphisms (SNP) in LIPG were

associated with lipid parameters, the risk of CAD and the risk of DVT.

Methods

We used the prospective case-control study nested in the EPIC-Norfolk cohort (1138 CAD

cases, 2237 matched controls) for initial association studies and, subsequently, the ACT

study (185 patients with documented DVT, 586 patients in which DVT was ruled out) to

replicate our findings regarding DVT risk.

Results

In EPIC-Norfolk, we found that the minor allele of one SNP, rs2000813 (p.T111I), was

associated with moderately higher HDL-C and apolipoprotein A-I levels, higher HDL

particle number and larger HDL size. No variants were associated with CAD risk, but 3

variants were associated with DVT risk (odds ratios 0.60 (95%CI 0.43-0.84), 2.04 (95%CI

1.40-2.98) and 1.67 (95%CI 1.18-2.38) per minor allele for rs2000813, rs6507931 and

rs2097055 respectively, p<0.005 for each). However, the association between LIPG SNPs

and DVT risk could not be replicated in the ACT study.

Conclusion

Our data support a modest association between the LIPG rs2000813 variant and parameters

of HDL metabolism, but no association between common genetic variants in LIPG and

CAD or DVT risk.

98

Chapter 9

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Plasma high-density lipoprotein cholesterol (HDL-C) levels correlate inversely with the risk

of coronary artery disease (CAD)1 and are also suggested to be associated with venous

thromboembolism.2-4 Family studies underline that 40-60% of the variation in HDL-C levels

is explained by genetic factors5 and many candidate genes involved in human HDL

metabolism have been identified.6 Endothelial lipase constitutes such a candidate and is a

member of the lipase family. These lipolytic enzymes are involved in lipid absorption,

transport, and metabolism. Endothelial lipase is synthesized in endothelial cells and

possesses phospholipase activity preferentially directed at HDL phospholipids.7

Overexpression of endothelial lipase reduces HDL-C, whereas deficiency of endothelial

lipase leads to elevation of HDL-C levels in mice.7-9 In humans, high levels of endothelial

lipase are significantly associated with features of the metabolic syndrome and with

coronary artery calcification.10

Genetic association studies have yielded conflicting results with respect to associations

between common genetic variants in LIPG and HDL-C levels.9,11-18 However, in several

genome-wide association studies single nucleotide polymorphisms (SNPs) near LIPG were

identified as being associated with HDL-C.19-24 Furthermore, rare loss-of-function variants

in LIPG were recently shown to be a cause of elevated HDL-C in humans.25

The data regarding associations between LIPG gene variants and CAD risk are mixed,

however. The minor allele of a common variant resulting in a Thr to Ile substitution at

codon 111 (rs2000813), was found to occur less frequently in patients who had a history of

myocardial infarction compared with those without.13 Similar findings were reported from

two small case-control studies.15,16 Furthermore, the same SNP has been reported to be

associated with atherothrombotic cerebral infarction in Japanese women.26 However, in a

recent large investigation in three independent prospective cohorts by Jensen et al, no

association between this SNP and the risk of CAD was found.18

The association between genetic variation in LIPG and the occurrence of deep venous

thrombosis (DVT) has, to our knowledge, never been specifically addressed. Since the

potential interrelation between endothelial lipase, high-density lipoprotein metabolism

and (athero)thrombosis is currently unclear, we aimed to evaluate whether common

genetic variants in LIPG associate with plasma lipid parameters, CAD or DVT risk.

99

Common genetic variation at the LIPG locus and the risk of CAD and DVT

Methods

Studies were approved by Institutional Review Boards and conducted according to the

Declaration of Helsinki. Written informed consent was obtained from all participants.

Data sources

The EPIC (European Prospective Investigation into Cancer and Nutrition)-Norfolk cohort

consists of a prospective population of 25,663 men and women between ages 45 and 79.

EPIC-Norfolk is part of the 10-country collaborative EPIC study designed to investigate

determinants of cancer. From the outset, additional data were obtained in EPIC-Norfolk to

enable the assessment of determinants of other diseases, including self-reported history of

DVT at baseline and prospective data on the occurrence of CAD. We performed analyses in

a subset of this cohort, a case-control cohort consisting of men and women who developed

fatal or nonfatal CAD during 7 years of follow-up (n=1138), and controls, matched for age,

sex, and enrolment time (n=2237). CAD was defined as codes 410 to 414 according to the

International Classification of Diseases, Ninth Revision. Participants were identified as

having CAD during follow-up if they had had a hospital admission or had died with CAD as

the underlying cause. Previous validation studies in our cohort indicate high specificity

of such case ascertainment.27 To replicate the findings from this study, we genotyped all

tagSNPs in the Amsterdam Case-control Thrombophilia (ACT) study, consisting of 771

outpatients who were referred to our hospital for evaluation of a suspected deep venous

thrombosis (DVT).28 In this study, 185 patients with objectively confirmed first DVT were

compared to 586 patients in whom this diagnosis was ruled out.

Laboratory analyses

Genotyping

We used the HAPMAP database and the TAGGER algorithm to capture most of the

common variation in the LIPG locus (NM_006033). For an in-depth discussion of this

method, we refer to reference 29. Briefly, we selected from HAPMAP all common genetic

variants (minor allele frequency > 0.1) in the LIPG locus in a population of European

ancestry. However, because of high correlations (high r2) among some pairs of SNPs,

genotyping both SNPs would offer only limited extra information. A tagging SNP approach

uses the knowledge of associations between genetic variants (linkage disequilibrium [LD]

structure) to limit the number of SNPs that needs to be genotyped. TagSNPs are those SNPs

which most effectively represent (or ‘tag’) all the SNPs in a particular locus. We selected

100

Chapter 9

1

2

3

4

5

6

7

8

9

10

11

12

tagSNPs using an r2 cut-off level > 0.8. For the EPIC-Norfolk study genotyping was

conducted by KBioscience (http://www.kbioscience.co.uk) using KASPar technology.

Genotyping of SNPs in the other cohort was carried out on an ABI 7900 system, using

Assay by DesignTM assays (Applied Biosystems, Foster City, CA, USA).

Biochemical analyses

Total cholesterol, HDL cholesterol and triglycerides were determined using standard

laboratory procedures within one hour after (non-fasting) blood sampling. Low-density

lipoprotein LDL cholesterol levels were calculated with the Friedewald formula. Serum

levels of apolipoprotein A-I (apoA-I) and B (apoB) were measured by rate

immunonephelometry (Behring Nephelometer BNII, Marburg, Germany) with calibration

traceable to the International Federation of Clinical Chemistry primary standards. The

interassay coefficient of variation (CV) of the apoA-I and apoB measurements were 5% and

3%, respectively. Serum concentrations of apolipoprotein A-II (apoA-II) were measured

with a commercially available immunoturbidimetric assay (Wako Pure Chemicals

Industries, Ltd, Osaka, Japan) on a Cobas-Mira autoanalyzer (Roche, Basel, Switzerland).

The intra-assay and interassay CVs for this assay were 2.5% and 3.1%, respectively. HDL

particle number and HDL size were measured with an automated nuclear magnetic

resonance (NMR) spectroscopic assay as described previously.30 In brief, particle

concentrations of lipoprotein subclasses of different size were obtained directly from the

measured amplitudes of their spectroscopically distinct lipid methyl group NMR signals.

Summation of the HDL subclass levels provides total HDL particle concentration. For the

present study, we grouped HDL subclasses as follows: small HDL (7.3 nm to 8.2 nm),

medium HDL (8.2 to 8.8 nm), and large HDL (8.8 to 13 nm). NMR spectroscopy–measured

HDL size was calculated as the mass-weighted average diameter of the HDL particles in a

particular plasma sample. Plasma concentrations of C-reactive protein (CRP) were

measured with a sandwich-type enzyme-linked immunosorbent assay as previously

described.31 Samples were analyzed in random order to avoid systematic bias. Researchers

and laboratory personnel had no access to identifiable information and could identify

samples by number only.

Statistical analyses

All statistical analyses were performed in cases with complete data, using SPSS version

16.0.2. Two-sided probability values of less than 0.05 were considered statistically

101

Common genetic variation at the LIPG locus and the risk of CAD and DVT

significant. Effects of SNPs on continuous variables were examined by ANOVA. To estimate

the relative risk of CAD in EPIC-Norfolk, conditional logistic regression was used to

calculate odds ratios (ORs) and 95% confidence intervals. Conditional logistic regression

took into account the matching for sex, age and enrolment time, and was adjusted for

Framingham risk score. We furthermore performed a meta-analysis on the relationship

between rs2000813 and cardiovascular risk using our own data as well as the published

data by Jensen et al.18 To estimate the relative risk of DVT in EPIC-Norfolk and in the ACT

study, differences between cases and controls were analysed by standard contingency table

analysis using two-tailed chi-square test probabilities; linearity of this relationship was

assessed using logistic regression analysis. LD plots were created with Haploview, version

4.1.32

Results

Using HAPMAP, we selected five LIPG tagSNPs. These tagSNPs were rs2097055 (c.460-

320T>CT>C), rs8093249 (c.571+1480A>C), rs2000813 (c.332C>T; p.T111I), rs3819166

(c.793+142A>G) and rs6507931 (c.1377-108C>T). LD plots for the five tagSNPs in EPIC-

Norfolk are displayed in Figure 1. The SNPs were not in LD as all r2 values were below 0.8,

confirming their non-redundant status as tagSNPs. All SNPs were in Hardy-Weinberg

equilibrium (data not shown).

Figure 1. LD plot of LIPG tagSNPs

The relationships between the five selected LIPG tagSNPs are shown in an LD plot. The values in the plots represent r2 between the appropriate SNPs (·10-2) and were calculated using Haploview. All are below 0.8, confirming their non-redundancy.

Chapter 9

102

1

2

3

4

5

6

7

8

9

10

11

12

Common genetic variants in LIPG and lipid levels

Of the five tagSNPs that were genotyped, only rs2000813 (p.T111I) showed a consistent

relationship with HDL parameters. Characteristics of patients stratified according to T111I

genotype are given in Table 1; characteristics of patients according to the genotype of the

other four SNPs are presented in Supplementary Tables 1-4. Carriers of the I allele had

higher HDL-C (0.034 mmol/L (95%CI 0.012-0.055) per allele, p=0.002), higher apoA-I

levels (3.16 mg/dL (95%CI 1.48-4.85) per allele, p<0.0005), a larger HDL size as measured

by NMR spectroscopy (40 pm (95%CI 15-65) per allele, p=0.002) as well as by gradient gel

electrophoresis (33 pm (95%CI 10-56) per allele, p=0.005) and a higher concentration of

HDL particles (0.34 nmol/L (95%CI 0.04-0.64) per allele, p=0.026). Body mass index, waist

circumference, history of diabetes mellitus, blood pressure and CRP levels did not differ

among carriers of different alleles. Total cholesterol, LDL cholesterol or apolipoprotein B

levels were also similar across genotypes.

Table 1. Characteristics of individuals according to LIPG T111I genotype in EPIC-Norfolk

rs2000813 (p.T111I) CC CT TT P N 1605 1269 304 BMI, kg/m2 26.7±3.6 26.6±3.7 26.5±3.8 0.43 Waist, cm 93 ±11 92±12 91±12 0.08 Blood pressure, mmHg systolic diastolic

140±18 84±11

141±18 84±11

142±18 85±11

0.62 0.61

Diabetes, n (%) 58 (3.6%) 42 (3.3%) 7 (2.3%) 0.50 CRP, mg/L 1.8 (0.8-3.9) 1.6(0.8-3.6) 1.7(0.8-3.9) 0.30 Cholesterol, mmol/L total LDL HDL

6.3±1.1 4.1±1.0 1.31±0.39

6.4±1.2 4.1±1.0 1.35±0.39

6.4±1.4 4.1±1.1 1.38±0.41

0.25 0.97 0.008

TG, mmol/L 1.8 (1.2-2.4) 1.7(1.2-2.5) 1.6(1.2-2.4) 0.28 ApoA-I, mg/dL 158±29 161±30 164±30 0.001 ApoB, mg/dL 133±32 132±32 131±33 0.627 HDL particles, nmol/L total large medium small

33.6±5.6 5.7±3.5 3.3±3.0 24.6±4.9

34.1±5.8 5.8±3.6 3.5±3.1 24.8±5.1

34.1±5.7 6.1±3.6 3.4±3.2 24.5±5.2

0.039 0.12 0.26 0.44

HDL size, nm 8.87±0.47 8.90±0.48 8.96±0.49 0.005 HDL size (GGE), nm 8.82±0.42 8.84±0.44 8.89±0.44 0.016

Data are presented as mean (±SD) or number (percentage). Data for CRP and TG are presented as median (interquartile range). P-values for diabetes are calculated by Pearson chi-square. P-values for CRP and TG are calculated using Kruskal-Wallis tests. Other p-values are calculated by one-way ANOVA. BMI = body mass index, CRP = C-reactive protein, TG = triglycerides, ApoA-I = apolipoprotein A-I, ApoB = apolipoprotein B, GGE = gradient gel electrophoresis.

Common genetic variation at the LIPG locus and the risk of CAD and DVT

103

Common genetic variants in LIPG and risk of CAD

None of the SNPs we genotyped showed a significant relationship with CAD risk. For

rs8093249, the OR for CAD was 0.97 per allele (95%CI 0.82-1.14, p=0.68; Fig. 2A). For

rs6507931, the OR for CAD was 1.09 per allele (95%CI 0.97-1.22, p=0.16; Fig. 2B). For

rs2097055, the OR was 1.06 per allele (95%CI 0.95-1.19, p=0.30; Fig. 2C), for rs2000813, it

was 0.94 per allele (95%CI 0.83-1.07, p=0.33; Fig. 2D), and, finally, for rs3819166, the OR

for CAD was 0.95 per allele (95%CI 0.81-1.11, p=0.50; Fig. 2E). Adjusting for individual risk

factors instead of the Framingham risk score did not materially change the results (data not

shown). To obtain a better estimate of the cardiovascular risk associated with rs2000813,

we performed a meta-analysis of our own data combined with the published data by Jensen

et al.18 We found that, in this combined data set of 2639 CAD cases and 4843 controls, the

OR for CAD was 0.96 per allele (95% CI 0.89-1.03, p=0.27), with heterozygous carriers of

the minor allele having an OR for CAD of 0.96 (95% CI 0.87-1.06) and homozygous carriers

having an OR for CAD of 0.91 (95%CI 0.76-1.10).

Figure 2. Odds ratios for CAD according to genotype in EPIC-Norfolk

AA GA GG0.125

0.25

0.5

1

2

4

744 248 26CAD1452 508 49No CAD

rs8093249

OR

for

CA

D

TT TC CC0.125

0.25

0.5

1

2

4

340 511 208CAD715 998 396No CAD

rs6507931

OR

for

CA

D

CC CT TT0.125

0.25

0.5

1

2

4

331 510 228CAD690 1021 430No CAD

rs2097055

OR

for

CA

D

CC TC TT0.125

0.25

0.5

1

2

4

97 413 97CAD207 857 207No CAD

rs2000813

OR

for

CA

D

GG GA AA0.125

0.25

0.5

1

2

4

774 259 31CAD1515 562 53No CAD

rs3819166

OR

for

CA

D

A CB

D E

Odds ratios are adjusted for the Framingham risk score. Numbers below the charts represent the number of people per genotype, with and without subsequent CAD.

Chapter 9

104

1

2

3

4

5

6

7

8

9

10

11

12

Common genetic variants in LIPG and risk of DVT

In the EPIC-Norfolk prospective CAD case-control study, 72 subjects had been diagnosed

with DVT. Of the five LIPG tagSNPs we genotyped, three SNPs showed a significant

relationship with DVT (Figure 3). For rs6507931 the per-allele OR for DVT was 2.04 (95%CI

1.40-2.98, p<0.0005; Fig. 3B), for rs2097055 the per-allele OR for DVT was 1.67 (95%CI

1.18-2.38, p=0.004; Fig. 3C), and for rs2000813, the per-allele OR for DVT was 0.60 (95%CI

0.43-0.84, p=0.003; Fig. 3D).

Figure 3. Odds ratios for DVT according to genotype in EPIC-Norfolk

AA GA GG0.0625

0.125

0.25

0.5

1

2

4

8

48 18 3DVT2147 738 72No DVT

rs8093249

OR

for

DVT

TT TC CC0.0625

0.125

0.25

0.5

1

2

4

8

p=0.001

37 27 5DVT1018 1418 599No DVT

rs6507931

OR

for

DVT

CC CT TT0.0625

0.125

0.25

0.5

1

2

4

8

p=0.014

31 33 6DVT990 1497 652No DVT

rs2097055

OR

for

DVT

CC TC TT0.0625

0.125

0.25

0.5

1

2

4

8

p=0.002

27 27 15DVT1578 1242 289No DVT

rs2000813

OR

for

DVT

GG GA AA0.0625

0.125

0.25

0.5

1

2

4

8

52 17 1DVT2236 804 83No DVT

rs3819166

OR

for

DVT

A CB

D E

Numbers below the charts represent the number of people per genotype, with and without a history of DVT.

We set out to replicate these results in an independent study, the ACT study. Assuming true

effects, the statistical power to replicate findings of this magnitude with a confidence level

of 0.95 in the ACT study would be > 0.99 in each case. However, in the ACT study, no

associations between these SNPs and the risk of DVT were found (Figure 4). Here, the OR

for DVT was 0.96 per allele (95%CI 0.76-1.21, p=0.74; Fig 4B) for rs6507931; 1.04 per allele

(95%CI 0.83-1.32, p=0.72; Fig 4C) for rs2097055 and 1.16 (95%CI 0.92-1.48, p=0.22; Fig

4D) per allele for rs2000813. We repeated the analysis after excluding all cases of provoked

Common genetic variation at the LIPG locus and the risk of CAD and DVT

105

DVT, potentially enriching the sample with patients with a genetic predisposition, but this

did not substantially change the results (data not shown). To confirm the suitability of the

ACT study to assess genetic risk factors for DVT, we genotyped Factor V Leiden (FVL) and

found that this variant was associated with a per-allele OR for DVT of 3.72 (95%CI 2.26-

6.13, p<0.0001).

Figure 4. Odds ratios for DVT according to genotype in the ACT study

AA GA GG0.0625

0.125

0.25

0.5

1

2

4

8

133 51 8DVT358 94 22No DVT

rs8093249

OR

for

VTE

TT TC CC0.0625

0.125

0.25

0.5

1

2

4

8

40 97 57DVT112 222 144No DVT

rs6507931

OR

for

VTE

CC CT TT0.0625

0.125

0.25

0.5

1

2

4

8

37 104 54DVT100 226 152No DVT

rs2097055O

R fo

r VT

E

CC TC TT0.0625

0.125

0.25

0.5

1

2

4

8

69 93 27DVT210 191 70No DVT

rs2000813

OR

for

VTE

GG GA AA0.0625

0.125

0.25

0.5

1

2

4

8

124 62 6DVT355 108 12No DVT

rs3819166

OR

for

VTE

A CB

D E

Numbers below the charts represent the number of people per genotype, with and without DVT.

Discussion

In the present study we could determine that, out of 5 tagSNPs in the LIPG locus, only the

minor allele of rs2000813 (p.T111I) was modestly, but consistently associated with higher

HDL-C levels, higher apoA-I levels, a higher concentration of HDL particles and larger HDL

size. We found no association between any of the tagSNPs and the risk of future CAD.

Three SNPs, including rs2000813, were associated with DVT risk in the prospective CAD

case-control study nested in the EPIC-Norfolk cohort, but these results could not be

replicated in the ACT study.

Chapter 9

106

1

2

3

4

5

6

7

8

9

10

11

12

Common genetic variants in LIPG and HDL metabolism

The association between the T111I variant and HDL-C levels has been evaluated in several

studies, but the results have been inconsistent. Whereas some studies found higher HDL-C

levels in carriers of the minor allele9,12,13,16, others could not confirm such a

relationship.11,14,15,18 It has been proposed that a small sample size and established

cardiovascular disease among study participants has limited the interpretation of some of

these studies.18 In the current study, we show in more than 3000 apparently healthy

Caucasian individuals that the minor allele of T111I associates with higher HDL-C levels.

Generally, a higher HDL-C is accompanied by increased apolipoprotein A-I levels and HDL

particle number, as well as a larger HDL size, as is the case here. Recently, it has been

shown in vitro that the T111I endothelial lipase variant possesses a lipolytic activity similar

to that of wildtype endothelial lipase. Accordingly, in silico analysis in Polyphen suggested

that the mutation is benign.25 Therefore, the possibility should be considered that T111I is

not itself a functional variant, but may be in linkage disequilibrium with another, yet

unknown, variant that is functional.

Common genetic variants in LIPG and CAD risk

We found no associations between any of the LIPG tagSNPs and CAD risk, which is in line

with the most recent findings regarding T111I from a prospective analysis in Caucasians by

Jensen et al.18 Moreover, in a meta-analysis of our data combined with those of the Jensen

study, no significant association between T111I and CAD risk was observed. It has been

suggested that large studies are needed to demonstrate the modest impact that single

genetic variants have on complex outcomes such as CAD.33 With the 2639 CAD cases and

4843 controls represented by the combined studies, we estimated there was 80% statistical

power to exclude a relative risk of CAD of 0.93 or less. Therefore, to confidently exclude a

potential 4-5% risk reduction associated with this variant, further studies would be

necessary. The current null-finding is however in line with a recent string of data on genetic

variants that do affect HDL-C levels but do not affect cardiovascular risk.34,35 In fact, a direct

head-to-head comparison between genetic variants influencing either plasma LDL-C or

HDL-C in a study of over two thousand CAD cases and over twelve thousand controls

indicated that only the former predict CAD.36 Findings such as these have fuelled a

controversy over the atheroprotective potential of HDL.

Common genetic variation at the LIPG locus and the risk of CAD and DVT

107

Common genetic variants in LIPG and DVT risk

Recent evidence suggests a direct link between HDL and thrombosis. HDL has been shown

to inhibit several coagulation factors, such as tissue factor, factor Va, VIIIa and Xa,37 and

purified HDL has been shown to enhance the inactivation of factor Va by activated protein C

(APC) and protein S, whereas purified LDL could not.38 HDL can also scavenge anionic

phospholipids, thereby abolishing their pro-coagulant properties.39

We found large differences in DVT risk across LIPG genotypes for 3 different tagSNPs in

EPIC-Norfolk, suggesting a role for endothelial lipase in thrombosis susceptibility.

However, we considered these results suspect for several reasons. First, the T111I variant is

associated with higher HDL-C levels and earlier studies suggested a protective effect of

T111I regarding CAD risk. Thus, under the assumption that DVT and CAD share a part of

their aetiology, our finding that T111I was associated with increased DVT risk was

counterintuitive. Second, such a high risk magnitude (a per-allele OR for DVT of ~1.7) in

combination with such a high allele frequency (~0.3) would render this genetic risk factor

more important than FVL and potentially explain a substantial part of DVT occurrence in

the general population. Such a finding would be in stark contrast with the results from

recent genome-wide association studies, in which no SNPs near LIPG were identified as

risk markers for venous thromboembolism.40,41 Finally, the EPIC-Norfolk data on DVT was

self-reported, increasing the possibility of bias. For these reasons, and because we could

not replicate the results in the ACT study, which was specifically designed to evaluate risk

factors for DVT, we interpret the original finding regarding LIPG SNPs and DVT risk in

EPIC-Norfolk as the result of a type I error.

In conclusion, our data support a modest association between the LIPG rs2000813 variant

and parameters of HDL metabolism, but no association between common genetic variants

in LIPG and CAD or DVT risk. In addition, our results reinvigorate the notion that findings

of genetic association need to be replicated in independent studies.

Acknowledgements

We gratefully acknowledge J.D. Otvos (Liposcience, Raleigh, NC, USA) for performing

NMR spectroscopy measurements in EPIC-Norfolk samples. We are indebted to M.W.

Tanck for his expert counsel in genetic statistics. SLR and MSS are supported by the British

Heart Foundation and the Medical Research Council.

Chapter 9

108

1

2

3

4

5

6

7

8

9

10

11

12

Supplementary Table 1. Characteristics of individuals according to LIPG genotype rs8093249 in EPIC-Norfolk

rs8093249 AAA AG GG P N 2195 756 75 BMI, kg/m2 26.6±3.6 26.8±3.6 26.8±3.7 0.26 Waist, cm 92±12 94±11 93±12 0.006 Blood pressure, mmHg systolic diastolic

140±18 84±11

140±18 85±12

145±21 85±13

0.12 0.80

Diabetes, n (%) 76 (3.5) 23 (3.0) 3 (4.0) 0.82 CRP, mg/L 1.7 (0.8-3.7) 1.7 (0.8-4.1) 1.8 (0.9-4.1) 0.68 Cholesterol, mmol/L total LDL HDL

6.4±1.2 4.1±1.0 1.35±0.39

6.3±1.1 4.1±1.0 1.31±0.39

6.3±1.1 4.1±0.9 1.31±0.36

0.17 0.92 0.03

TG, mmol/L 1.7 (1.2-2.5) 1.7(1.2-1.4) 1.7(1.2-2.5) 0.88 ApoA-I, mg/dL 161±30 157±29 160±26 0.009 ApoB, mg/dL 133±33 133±31 133±29 0.99 HDL particles, nmol/L total large medium small

34.0±5.7 5.8±3.6 3.5±3.1 24.7±5.0

33.6±5.8 5.6±3.5 3.3±3.0 24.8±4.9

33.2±5.2 5.6±3.6 3.7±3.4 24.0±4.7

0.13 0.18 0.22 0.38

HDL size, nm 8.91±0.48 8.85±0.46 8.83±0.45 0.15 HDL size (GGE), nm 8.85±0.44 8.80±0.43 8.83±0.40 0.14

Data are presented as mean (±SD) or number (percentage). Data for CRP and TG are presented as median (interquartile range). P-values for diabetes are calculated by Pearson chi-square. P-values for CRP and TG are calculated using Kruskal-Wallis tests. Other p-values are calculated by one-way ANOVA. BMI = body mass index, CRP = C-reactive protein, TG = triglycerides, ApoA-I = apolipoprotein A-I, ApoB = apolipoprotein B, GGE = gradient gel electrophoresis.

Common genetic variation at the LIPG locus and the risk of CAD and DVT

109

Supplementary Table 2. Characteristics of individuals according to LIPG genotype rs6507931 in EPIC-Norfolk

rs6507931 TT TC CC P N 1055 1508 604 BMI, kg/m2 26.7±3.8 26.5±3.6 26.7±3.6 0.38 Waist, cm 93 ±12 92 ±12 93 ±12 0.14 Blood pressure, mmHg systolic diastolic

141±18 85±12

140±18 84±11

140±18 84±11

0.28 0.12

Diabetes, n (%) 36 (3.4%) 46 (3.0%) 27 (4.5%) 0.27 CRP, mg/L 1.7 (0.8-3.9) 1.7(0.8-3.7) 1.7(0.8-3.8) 0.95 Cholesterol, mmol/L total LDL HDL

6.4±1.3 4.2±1.1 1.33±0.39

6.3±1.2 4.1±1.0 1.34±0.39

6.4±1.16 4.1±1.04 1.33±0.41

0.60 0.68 0.79

TG, mmol/L 1.7 (1.3-2.5) 1.7 (1.2-2.5) 1.7 (1.2-2.5) 0.78 ApoA-I, mg/dL 161±30 160±29 158±29 0.41 ApoB, mg/dL 133±33 132±32 133±33 0.52 HDL particles, nmol/L total large medium small

33.8±5.7 5.7±3.6 3.4±3.1 24.8±5.0

34.0±5.8 5.8±3.6 3.4±3.1 24.8±5.0

33.5±5.7 5.8±3.5 3.3±3.1 24.5±4.7

0.17 0.73 0.57 0.41

HDL size, nm 8.88±0.47 8.89±0.48 8.88±0.48 0.87 HDL size (GGE), nm 8.84±0.43 8.84±044 8.83±0.43 0.98

Data are presented as mean (±SD) or number (percentage). Data for CRP and TG are presented as median (interquartile range). P-values for diabetes are calculated by Pearson chi-square. P-values for CRP and TG are calculated using Kruskal-Wallis tests. Other p-values are calculated by one-way ANOVA. BMI = body mass index, CRP = C-reactive protein, TG = triglycerides, ApoA-I = apolipoprotein A-I, ApoB = apolipoprotein B, GGE = gradient gel electrophoresis.

Chapter 9

110

1

2

3

4

5

6

7

8

9

10

11

12

Supplementary Table 3. Characteristics of individuals according to LIPG genotype rs2097055 in EPIC-Norfolk

rs2097055 CC CT TT P N 1021 1530 658 BMI, kg/m2 26.6±3.7 26.7±3.6 26.6±3.7 0.98 Waist, cm 92±12 93±11 92±12 0.81 Blood pressure, mmHg systolic diastolic

141 ±19 84 ±12

141 ±18 85 ±11

140 ±18 84 ±11

0.22 0.99

Diabetes, n (%) 37 (3.6%) 46 (3.0%) 24 (3.6%) 0.61 CRP, mg/L 1.7 (0.8-3.9) 1.8 (0.8-3.9) 1.6 (0.8-3.5) 0.09 Cholesterol, mmol/L total LDL HDL

6.4±1.3 4.1±1.0 1.33±0.39

6.4±1.2 4.1±1.0 1.34±0.39

6.3±1.1 4.2±1.0 1.32±0.39

0.96 0.93 0.32

TG, mmol/L 1.7 (1.2-2.5) 1.7 (1.2-2.5) 1.8 (1.2-2.5) 0.97 ApoA-I, mg/dL 160±29 160±30 158±29 0.14 ApoB, mg/dL 132±33 133±33 132±31 0.78 HDL particles, nmol/L total large medium small

33.8±5.7 5.7±3.6 3.4±3.1 24.7±4.9

34.0±5.7 5.8±3.6 3.4±3.1 24.8±5.1

33.4±5.5 5.7±3.4 3.2±2.9 24.5±4.9

0.61 0.90 0.14 0.50

HDL size, nm 8.89±0.48 8.89±0.47 8.88±0.47 0.84 HDL size (GGE), nm 8.85±0.44 8.84±0.43 8.83±0.43 0.66

Data are presented as mean (±SD) or number (percentage). Data for CRP and TG are presented as median (interquartile range). P-values for diabetes are calculated by Pearson chi-square. P-values for CRP and TG are calculated using Kruskal-Wallis tests. Other p-values are calculated by one-way ANOVA. BMI = body mass index, CRP = C-reactive protein, TG = triglycerides, ApoA-I = apolipoprotein A-I, ApoB = apolipoprotein B, GGE = gradient gel electrophoresis.

Common genetic variation at the LIPG locus and the risk of CAD and DVT

111

Supplementary Table 4. Characteristics of individuals according to LIPG genotype rs3819166 in EPIC-Norfolk

rs3819166 GG GA AA P N 2289 820 84 BMI, kg/m2 26.6±3.7 26.7±3.5 26.9±6.5 0.73 Waist, cm 92±12 92.4±11 93±12 0.92 Blood pressure, mmHg systolic diastolic

140±18 84±12

141±18 85±11

142±17 84±11

0.72 0.76

Diabetes, n (%) 82 (3.6%) 25 (3.0%) 1 (1.2%) 0.41 CRP, mg/L 1.7 (0.8-3.7) 1.7 (0.8-3.9) 1.25 (0.7-3.9) 0.66 Cholesterol, mmol/L total LDL HDL

6.4±1.2 4.1±1.0 1.35±0.4

6.3±1.2 4.1±1.0 1.32±0.37

6.3±1.1 4.2±1.0 1.27±0.45

0.70 0.84 0.08

TG, mmol/L 1.7 (1.2-2.4) 1.7 (1.2-2.5) 1.8 (1.3-2.5) 0.88 ApoA-I, mg/dL 160±30 158±27 156±30 0.12 ApoB, mg/dL 133±33 132±31 133±30 0.74 HDL particles, nmol/L total large medium small

34.0±5.7 5.8±3.6 3.4±3.1 24.8±4.9

33.6±5.7 5.6±3.4 3.4±3.1 24.6±5. 3

32.7±5.3 5.7±3.7 2.7±2.4 24.3±4.1

0.046 0.34 0.13 0.55

HDL size, nm 8.89±0.48 8.88±0.47 8.86±0.51 0.69 HDL size (GGE), nm 8.84±4.34 8.82±0.43 8.84±0.42 0.54

Data are presented as mean (±SD) or number (percentage). Data for CRP and TG are presented as median (interquartile range). P-values for diabetes are calculated by Pearson chi-square. P-values for CRP and TG are calculated using Kruskal-Wallis tests. Other p-values are calculated by one-way ANOVA. BMI = body mass index, CRP = C-reactive protein, TG = triglycerides, ApoA-I = apolipoprotein A-I, ApoB = apolipoprotein B, GGE = gradient gel electrophoresis.

Chapter 9

112

1

2

3

4

5

6

7

8

9

10

11

12

References 1. Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High density lipoprotein as a protective

factor against coronary heart disease : The Framingham study. Am J Med 1977;62(5):707-714. 2. Ageno W, Becattini C, Brighton T, Selby R, Kamphuisen PW. Cardiovascular risk factors and venous

thromboembolism: a meta-analysis. Circulation 2008;117(1):93-102. 3. Doggen CJ, Smith NL, Lemaitre RN, Heckbert SR, Rosendaal FR, Psaty BM. Serum lipid levels and the risk of

venous thrombosis. Arterioscler Thromb Vasc Biol 2004;24(10):1970-1975. 4. Eichinger S, Pecheniuk NM, Hron G et al. High-density lipoprotein and the risk of recurrent venous

thromboembolism. Circulation 2007;115(12):1609-1614. 5. Heller DA, de Faire U, Pedersen NL, Dahlen G, McClearn GE. Genetic and Environmental Influences on

Serum Lipid Levels in Twins. N Engl J Med 1993;328(16):1150-1156. 6. Holleboom AG, Vergeer M, Hovingh GK, Kastelein JJ, Kuivenhoven JA. The value of HDL genetics. Curr

Opin Lipidol 2008;19(4):385-394. 7. Jaye M, Lynch KJ, Krawiec J et al. A novel endothelial-derived lipase that modulates HDL metabolism. Nat

Genet 1999;21(4):424-428. 8. Ishida T, Zheng Z, Dichek HL et al. Molecular cloning of nonsecreted endothelial cell-derived lipase

isoforms. Genomics 2004;83(1):24-33. 9. Ma K, Cilingiroglu M, Otvos JD, Ballantyne CM, Marian AJ, Chan L. Endothelial lipase is a major genetic

determinant for high-density lipoprotein concentration, structure, and metabolism. Proc Natl Acad Sci U S A 2003;100(5):2748-2753.

10. Badellino KO, Wolfe ML, Reilly MP, Rader DJ. Endothelial Lipase Concentrations Are Increased in Metabolic Syndrome and Associated with Coronary Atherosclerosis. PLoS Med 2006;3(2):e22.

11. Halverstadt A, Phares DA, Ferrell RE, Wilund KR, Goldberg AP, Hagberg JM. High-density lipoprotein-cholesterol, its subfractions, and responses to exercise training are dependent on endothelial lipase genotype. Metabolism 2003;52(11):1505-1511.

12. Hutter CM, Austin MA, Farin FM et al. Association of endothelial lipase gene (LIPG) haplotypes with high-density lipoprotein cholesterol subfractions and apolipoprotein AI plasma levels in Japanese Americans. Atherosclerosis 2006;185(1):78-86.

13. Mank-Seymour AR, Durham KL, Thompson JF, Seymour AB, Milos PM. Association between single-nucleotide polymorphisms in the endothelial lipase (LIPG) gene and high-density lipoprotein cholesterol levels. Biochim Biophys Acta 2004;1636(1):40-46.

14. Paradis ME, Couture P, Bosse Y et al. The T111I mutation in the EL gene modulates the impact of dietary fat on the HDL profile in women. J Lipid Res 2003;44(10):1902-1908.

15. Shimizu M, Kanazawa K, Hirata K et al. Endothelial lipase gene polymorphism is associated with acute myocardial infarction, independently of high-density lipoprotein-cholesterol levels. Circ J 2007;71(6):842-846.

16. Tang NP, Wang LS, Yang L et al. Protective effect of an endothelial lipase gene variant on coronary artery disease in a Chinese population. J Lipid Res 2008;49(2):369-375.

17. Yamakawa-Kobayashi K, Yanagi H, Endo K, Arinami T, Hamaguchi H. Relationship between serum HDL-C levels and common genetic variants of the endothelial lipase gene in Japanese school-aged children. Hum Genet 2003;113(4):311-315.

18. Jensen MK, Rimm EB, Mukamal KJ et al. The T111I variant in the endothelial lipase gene and risk of coronary heart disease in three independent populations. Eur Heart J 2009;ehp145.

19. Kathiresan S, Willer CJ, Peloso GM et al. Common variants at 30 loci contribute to polygenic dyslipidemia. Nat Genet 2009;41(1):56-65.

20. Loci influencing lipid levels and coronary heart disease risk in 16 European population cohorts. Nat Genet 2009;41(1):47-55.

21. Sabatti C, Service SK, Hartikainen AL et al. Genome-wide association analysis of metabolic traits in a birth cohort from a founder population. Nat Genet 2009;41(1):35-46.

22. Kathiresan S, Melander O, Guiducci C et al. Six new loci associated with blood low-density lipoprotein cholesterol, high-density lipoprotein cholesterol or triglycerides in humans. Nat Genet 2008;40(2):189-197.

23. Willer CJ, Sanna S, Jackson AU et al. Newly identified loci that influence lipid concentrations and risk of coronary artery disease. Nat Genet 2008;40(2):161-169.

113

Common genetic variation at the LIPG locus and the risk of CAD and DVT

24. Heid IM, Boes E, Muller M et al. Genome-Wide Association Analysis of High-Density Lipoprotein Cholesterol in the Population-Based KORA Study Sheds New Light on Intergenic Regions. Circ Cardiovasc Genet 2008;1(1):10-20.

25. Edmondson AC, Brown RJ, Kathiresan S et al. Loss-of-function variants in endothelial lipase are a cause of elevated HDL cholesterol in humans. J Clin Invest 2009;119(4):1042-1050.

26. Yamaguchi S, Yamada Y, Metoki N et al. Genetic risk for atherothrombotic cerebral infarction in individuals stratified by sex or conventional risk factors for atherosclerosis. Int J Mol Med 2006;18(5):871-883.

27. Matthijs Boekholdt S, Peters RJG, Day NE et al. Macrophage migration inhibitory factor and the risk of myocardial infarction or death due to coronary artery disease in adults without prior myocardial infarction or stroke: The EPIC-Norfolk Prospective Population study. The American Journal of Medicine 2004;117(6):390-397.

28. Hermanides J, Cohn DM, Hans DJ et al. Venous thrombosis is associated with hyperglycemia at diagnosis: a case-control study. J Thromb Haemost 2009.

29. De Bakker PIW, Yelensky R, Pe'er I, Gabriel SB, Daly MJ, Altshuler D. Efficiency and power in genetic association studies. Nat Genet 2005;37(11):1217-1223.

30. El-Harchaoui K, Arsenault BJ, Franssen R et al. High-density lipoprotein particle size and concentration and coronary risk. Ann Intern Med 2009;150(2):84-93.

31. Bruins P, Velthuis Ht, Yazdanbakhsh AP et al. Activation of the Complement System During and After Cardiopulmonary Bypass Surgery : Postsurgery Activation Involves C-Reactive Protein and Is Associated With Postoperative Arrhythmia. Circulation 1997;96(10):3542-3548.

32. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21(2):263-265.

33. Thompson A, Di AE, Sarwar N et al. Association of cholesteryl ester transfer protein genotypes with CETP mass and activity, lipid levels, and coronary risk. JAMA 2008;299(23):2777-2788.

34. Frikke-Schmidt R, Nordestgaard BG, Stene MC et al. Association of loss-of-function mutations in the ABCA1 gene with high-density lipoprotein cholesterol levels and risk of ischemic heart disease. JAMA 2008;299(21):2524-2532.

35. Johannsen TH, Kamstrup PR, Andersen RV et al. Hepatic Lipase, Genetically Elevated High-Density Lipoprotein, and Risk of Ischemic Cardiovascular Disease. J Clin Endocrinol Metab 2009;94(4):1264-1273.

36. Willer CJ, Sanna S, Jackson AU et al. Newly identified loci that influence lipid concentrations and risk of coronary artery disease. Nat Genet 2008;40(2):161-169.

37. Nofer JR, Kehrel B, Fobker M, Levkau B, Assmann G, Eckardstein Av. HDL and arteriosclerosis: beyond reverse cholesterol transport. Atherosclerosis 2002;161(1):1-16.

38. Griffin JH, Kojima K, Banka CL, Curtiss LK, Fernandez JA. High-density lipoprotein enhancement of anticoagulant activities of plasma protein S and activated protein C. J Clin Invest 1999;103(2):219-227.

39. Oslakovic C, Krisinger MJ, Andersson A, Jauhiainen M, Ehnholm C, Dahlback B. Anionic Phospholipids Lose their Procoagulant Properties when Incorporated into High-Density Lipoproteins. J Biol Chem 2009;M807286200.

40. Tregouet DA, Heath S, Saut N et al. Common susceptibility alleles are unlikely to contribute as strongly as the FV and ABO loci to VTE risk: results from a GWAS approach. Blood 2009;113(21):5298-5303.

41. Bezemer ID, Bare LA, Doggen CJ et al. Gene variants associated with deep vein thrombosis. JAMA 2008;299(11):1306-1314.

114

Chapter 9

Part III

Reproductive aspects of venous thromboembolism and thrombophilia

Chapter 10

Risk of postpartum haemorrhage in women receiving therapeutic doses of low-molecular-weight heparin

Cohn DM, Roshani S, Stehouwer AC, Hamers S, Wolf H, van der Post JAM, Büller HR, Kamphuisen PW, Middeldorp S. Submitted for publication.

Abstract

Background

Low-molecular-weight heparin (LMWH) is the drug of choice to prevent venous

thrombosis in pregnancy, but the optimal dose to prevent thrombosis while avoiding

bleeding complications is unclear. We investigated whether use of therapeutic doses of

LMWH increases the risk of postpartum haemorrhage (PPH).

Methods

We identified all pregnant women who received therapeutic doses of LMWH between 1995

and 2008 in the Academic Medical Centre, Amsterdam, The Netherlands. We compared the

risk of PPH and estimated amount of blood loss in 83 women treated with LMWH to 523

pregnant women who did not use LMWH.

Results

PPH risk after vaginal delivery was 12% (7/83) in LMWH users and 21% (97/523) in non-

users (RR 0.6; 95%CI 0.3 to 1.2). After caesarean section, PPH risk was 9% (2/22) in LMWH

users and 4% (2/51) in non-users (RR 2.3; 0.3 to 17). Median amount of blood loss in

vaginal deliveries was 200 mL in LMWH users and 300 mL in non-users, (difference 100 mL;

41 to 159). In caesarean sections the median blood loss did not differ between LMWH users

and non-users (425 and 400 mL respectively, difference 25 mL; -133 to 183). In emergency

caesarean sections this was higher in LMWH users than in non-users (450 and 200 mL

respectively, difference 250 mL; 9 to 491).

Conclusion

Therapeutic doses of LMWH are relative safe to use for pregnant women who deliver in the

hospital setting of optimal obstetric care. An increased risk of bleeding appears to occur in

the setting of emergency caesarean section.

Chapter 10

118

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Low-molecular-weight heparin (LMWH) is the drug of choice in pregnant women requiring

prophylaxis or treatment for venous thrombosis. However, the optimal dose with respect to

efficacy and safety is uncertain.1 LMWH has one crucial disadvantage, as its anticoagulant

effect can only be partially antagonized. This is of particular importance with respect to its

use in high doses and raises concerns about an increased risk of postpartum haemorrhage

(PPH) when used in pregnant women.

PPH is defined by the World Health Organization (WHO) as postpartum blood loss in

excess of 500 mL.2 However, other definitions have been suggested, such as the addition of

blood loss exceeding 1000 mL following caesarean section.3 PPH has an incidence of 19% in

nulliparous deliveries in the Netherlands.4 The diagnosis encompasses excessive blood loss

from uterus, cervix, vagina and perineum. The commonest cause of primary PPH (PPH < 24

hours following delivery) is lack of efficient uterine contraction (uterine atony).5

In order to limit the risk of PPH, current guidelines recommend discontinuation of LMWH

12 to 24 hours prior to delivery.1;6 However, as labour can commence spontaneously, timely

discontinuation cannot be guaranteed. The risk of PPH associated with use of therapeutic

doses of LMWH has been assessed in few studies.3;7-12 These studies either included a small

or an unknown number of women treated with therapeutic doses of LMWH3;7-10 or they

lacked a control group of women who did not use LMWH.7;8;10 Three studies assessed

bleeding risk in pregnant women (including those who were treated with therapeutic doses

of LMWH).3;11;12 However, only one study reports the bleeding risk associated with

antepartum therapeutic doses of LMWH.11 In this prospective multicentre survey in the UK

and Ireland, blood loss > 500 mL occurred in 6/126 (4.8%) women who were treated with

therapeutic doses of LMWH. The two other studies also assessed PPH associated with

LMWH, however the bleeding rate associated with therapeutic doses only can not be

deduced from these papers.3;12

In our hospital all pregnant women whom we judge to require anticoagulant prophylaxis

are treated with therapeutic doses of LMWH. This protocol was based on a systematic

review that we performed in 1998.13 In this review of several cohorts of women, recurrent

venous thromboembolism (VTE) occurred in 2.0% (3/149) of pregnant women, all of whom

were treated with prophylactic or intermediate doses of LMWH. Similar findings were

reported in another large cohort study in which 7 of 8 recurrent episodes of VTE occurred

in women on prophylactic or intermediate doses of enoxaparin.14

Postpartum haemorrhage in women receiving therapeutic doses of LMWH: a cohort study

119

We performed a controlled cohort study in our hospital to assess the risk of PPH associated

with therapeutic doses of LMWH in pregnant women.

Methods

Identification of study cohorts

Women who had used therapeutic doses of LMWH during pregnancy were identified by

collection of all hospital ID numbers in whom anti-Xa measurements were performed

between mid-August 1995 and mid-February 2008. We reviewed charts to assess whether

the anti-Xa measurements were performed during pregnancy. Inclusion criteria were:

therapeutic doses of LMWH, pregnancy duration of at least 25 weeks gestation, and

delivery in the Academic Medical Centre (AMC).

The control cohort consisted of women who had been registered for antenatal care in the

AMC before 24 weeks gestational age, delivered in the AMC and did not use LMWH during

their pregnancy. Cases and controls were matched by random electronic selection for age

(±2 years), parity (nulliparous or multiparous) and date of delivery (±1 year) in a 1:6 ratio.

Intervention

LMWH was dosed on body weight prior to pregnancy, according to the hospital’s protocol.

Measurements of anti-Xa levels were performed in all women who used LMWH during

their regular visits to the outpatient clinic of the Department of Vascular Medicine. Dose-

adjustments were only performed if peak anti-Xa activity was lower than 0.4 or higher than

1.2 anti-Xa units on repeated occasions. A multidisciplinary team of obstetricians and

vascular medicine experts discussed patients with regular intervals. Women were advised to

discontinue LMWH as soon as either contractions started, membranes ruptured or the

evening before the induction of labour or a caesarean section was planned.

Outcomes

The primary outcome was PPH, defined as blood loss > 500 mL following vaginal delivery

or >1000 mL following caesarean section within 24 hours of delivery. Secondary outcomes

were the estimated amount of blood loss in mL, blood transfusions in the first week

postpartum, and recurrent VTE.

Chapter 10

120

1

2

3

4

5

6

7

8

9

10

11

12

Statistical analysis

We calculated the absolute risk of PPH and its 95% confidence interval (95%CI) for both

groups of LMWH users and non-users. Subsequently relative risk (RR) of PPH and 95%CI

in pregnant women treated with therapeutic doses of LMWH compared to non-users was

calculated. Non-normally distributed data are presented as medians. We calculated the

median blood loss difference and its 95%CI between the two groups of women. The

difference of median blood loss between groups was evaluated by Mann-Whitney test.

Furthermore, we compared the median blood loss of both groups in strata of a priori

defined other risk factors (i.e. type of vaginal delivery [normal versus assisted] or caesarean

section [emergency versus primary] and ethnicity) to investigate their interaction with

LMWH on PPH risk. Blood transfusion in the first 24 hours of delivery was compared

between two groups of the study using the �2 test.

Results

We identified 83 women who used therapeutic doses of LMWH during pregnancy for

various indications (see Flowchart for case selection).

Baseline characteristics of the study groups are shown in Table 1. Mean gestational age

(±SD) was 39±3 weeks in LMWH users and 38±4 in non-users. In both cohorts, 93% of

vaginal deliveries proceeded spontaneously (normal vaginal delivery) and 7% needed

assistance. Almost one-quarter (27%) of the women treated with LMWH delivered by

caesarean sections; half of these were planned before onset of labour (primary caesarean

section). In the control cohort only 10% of the women underwent caesarean sections, most

were emergency caesarean sections.

Inclusion flowchart of women treated with LMWH

Postpartum haemorrhage in women receiving therapeutic doses of LMWH: a cohort study

121

PPH occurred in 11% of the women who used therapeutic doses of LMWH and in 19% of the

controls (RR for PPH: 0.6; 95%CI 0.3 to 1.1) (Table 2).The absolute risks of PPH in women

using LMWH were within the same range after vaginal delivery (12%) and after caesarean

section (9%). However, in the control cohort these risks were 21% and 4% respectively (RR

for PPH after vaginal delivery 0.6; 95%CI 0.3 to 1.2; RR after caesarean section 2.3; 95%CI

0.3 to 17).

Table 1. Baseline characteristics of two study groups

Women who used

therapeutic dose of LMWH (N=83)

Women who did not use LMWH (N=523)

Age, years Mean± SD 32±5 31±5 Ethnicity N (% )

Caucasian 59 (71) 263 (50) African 11 (13) 167 (32) Others/unknown 13 (16) 93 (18)

Gestational age, weeks Mean± SD 39±3 38±4 Delivery route

Vaginal N (% of all women) 61 (73) 472 (90) Normal delivery, (% of vaginal deliveries) 57 (93) 437 (93) Assisted delivery, (% of vaginal deliveries) 4 (7) 35 (7)

Caesarean section N (% of all women) 22 (27) 51 (10) Primary caesarean section, (% of caesarean sections) 11 (50) 5 (10) Emergency caesarean section, (% of caesarean sections) 11 (50) 46 (90)

Perineal laceration degree N (% of vaginal deliveries) 1st degree 4 (7) 43 (9) 2nd degree, Episiotomy 10 (16) 59 (12) 2nd degree, Spontaneous rupture 22 (36) 100 (22) 3rd degree 0 (0) 7 (1) No laceration 24 (39) 263 (56) Unknown 1 (2) -

Birth weight, grams Mean± SD 3044±736 3053±885 Indication for LMWH administration N (% of all women)

History of VTE 14 (17) History of VTE and thrombophilia 46 (55) Current VTE* 11 (13) Current VTE* and thrombophilia 2 (3) Antiphospholipid syndrome 2 (3) Pre-eclampsia 1 (1) Prosthetic heart valve 5 (6) Prostatic heart valve+ current heart thrombosis 1 (1) Current CVA 1 (1)

LMWH=low-molecular -weight heparin, VTE= venous thromboembolism. * VTE during current pregnancy

Chapter 10

122

1

2

3

4

5

6

7

8

9

10

11

12

Median blood loss after vaginal delivery was 200 (range, 50 to 4000) and 300 (20 to 3600)

mL in LMWH users and non-users respectively (median difference 100; 95%CI: 41 to 159).

Blood loss did not differ between the groups after stratification for delivery subtypes (i.e.

normal or assisted vaginal delivery, and primary or emergency caesarean section), ethnicity

and perineal laceration degree (data not shown), except for the stratum of no-laceration

(intact perineum) where the median of estimated blood loss (range) was 175 (50 to 1600) in

cases and 300 (50 to 3500) in controls with median difference of 125; 95% CI 87 to 213.

Blood transfusion was given as judged by the attending obstetrician in 5% of LMWH users

and 3% of non-users after delivery (OR 1.4; 95%CI: 0.5 to 4.3).

In terms of efficacy, recurrent VTE was suspected in one woman (1.2%, 95%CI 0.6-5.8)

despite the use of therapeutic doses of LMWH. However, a recurrent episode was not

established as ventilation/perfusion scintigraphy revealed a perfusion defect on the same

localization of the previous PE.

Table 2. Risk of PPH, median and range of blood loss, stratified for subtypes of deliveries in women with and without therapeutic doses of LMWH

Women who used therapeutic doses of LMWH (N=83)

Women who did not use LMWH (N=523)

RR / Median difference 95%CI

PPH events N (%) 9 (11) 99 (19) 0.6* 0.3 to 1.1 Vaginal delivery 7 (12) 97 (21) 0.6* 0.3 to 1.2 Caesarean section 2 (9) 2 (4) 2.3* 0.3 to 17

Blood loss Median (range) Vaginal delivery 200 (50 to 4000) 300 (20 to 3600) -100+ 41 to 159

Normal vaginal delivery 200 (50 to 4000) 300 (20 to 3600) -100+ 39 to 161 Assisted vaginal delivery 400 (250 to 550) 400 (100 to 2500) 0+ -772 to 772

Caesarean section 425 (200 to 2000) 400 (100 to 2000) 25+ -133 to 183 Primary caesarean section 400 (200 to 2000) 400 (100 to 2000) 0+ -225 to 225 Emergency caesarean section 450 (200 to 1200) 200 (100 to 400) 250+ 9 to 491

Blood transfusion N (%) 4 (5) 18 (3) 1.4* 0.5 to 4.3

PPH= postpartum haemorrhage (>500 mL blood loss in vaginal delivery and >1000 mL blood loss in caesarean section), LMWH= low-molecular-weight heparin. Blood loss is reported in mL. * depicts relative risk (RR) and + depicts median difference. Discussion We observed that in vaginal deliveries the risk of PPH was not increased in women who

used therapeutic doses of LMWH compared with those who did not. The upper limit of the

95% CI is compatible with only a 10% increase and corresponds with 160 mL of blood loss

in vaginal deliveries. This finding is in line with a previous study that reported similar risks

(5.7%) of PPH in vaginal deliveries in women who used LMWH (doses not specified) and

Postpartum haemorrhage in women receiving therapeutic doses of LMWH: a cohort study

123

those who did not use LMWH (OR 1.0; 95%CI: 0.2 to 4.7).3 However, the absolute risk of

PPH in both our study groups (12% in LMWH users and 21% in non-LMWH users) was

higher. Although the PPH rate in our control group appears high as compared to other

studies that assessed PPH in general populations,15-17 a previously performed population-

based cohort in the Netherlands showed a comparable PPH incidence of 19% (PPH defined

as blood loss >500 mL).4 An explanation for this high incidence could be the difference in

blood loss estimation and in treatment regimens. In the Netherlands, an active

management during the third stage of delivery (such as prophylactic administration of

oxytocics, immediate cord clamping or controlled cord traction) is not routinely performed.

Oxytocics administered in the third stage of delivery have previously been shown to reduce

the amount of blood loss.18 Therefore we assume that withholding oxytocics might have led

to a higher incidence of PPH in our control cohort, whereas this was not observed in the

cases since LMWH use warranted an active management of the third stage of delivery.

Furthermore, as our hospital is a tertiary referral centre, the observed high incidence of

blood loss >500 mL in the control cohort may be explained by comorbidities that increase

the risk of a complicated delivery.

In caesarean sections, PPH risk was 2.3 times higher (95%CI: 0.3 to 17) in women who

used LMWH as compared to those who did not, although the certainty of this estimate is

limited by the small number of patients. In another study, the risk of PPH for LMWH users

(5%) in caesarean sections was half of the controls (12.5%) (OR 0.4; 95%CI: 0.04 to 3.4).3

We were not able to calculate the risk of PPH in two subtypes of caesarean sections because

of few PPH events; 2 in cases (one underwent emergency and one primary caesarean

section) and 2 in controls (both had emergency caesarean section). In primary caesarean

sections, the amount of blood loss was not different between LMWH users and non-users.

The main limitation of this cohort study is its retrospective design. We were unable to

retrieve the anti-Xa level shortly prior to delivery and the time interval between LMWH

cessation and delivery. However, evidence about the association between this duration and

the risk of PPH is conflicting.8;9;19

In conclusion, therapeutic doses of LMWH are relatively safe to use in women who deliver

vaginally in the hospital setting of optimal obstetric care. However an increased risk of

bleeding in the setting of emergency caesarean section can not be excluded.

Chapter 10

124

1

2

3

4

5

6

7

8

9

10

11

12

References 1. Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J. Venous thromboembolism, thrombophilia, antithrombotic

therapy, and pregnancy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6 Suppl):844S-886S.

2. World Health Organization. Managing Complications in Pregnancy and Childbirth: A guide for midwives and doctors. Integrated Management Of Pregnancy And Childbirth. 2000;S25.

3. Kominiarek MA, Angelopoulos SM, Shapiro NL et al. Low-molecular-weight heparin in pregnancy: peripartum bleeding complications. J Perinatol. 2007;27(6):329-334.

4. Bais JM, Eskes M, Pel M, Bonsel GJ, Bleker OP. Postpartum haemorrhage in nulliparous women: incidence and risk factors in low and high risk women. A Dutch population-based cohort study on standard (> or = 500 ml) and severe (> or = 1000 ml) postpartum haemorrhage. Eur J Obstet Gynecol Reprod Biol. 2004;115(2):166-172.

5. Dildy GA, III. Postpartum hemorrhage: new management options. Clin Obstet Gynecol. 2002;45(2):330-344. 6. Royal College of Obstetricians and Gynaecologists. Thromboembolic disease in pregnancy and the

puerperium: acute management. Guideline No 28. 2007;10. 7. Dulitzki M, Pauzner R, Langevitz P et al. Low-molecular-weight heparin during pregnancy and delivery:

preliminary experience with 41 pregnancies. Obstet Gynecol. 1996;87(3):380-383. 8. Maslovitz S, Many A, Landsberg JA et al. The safety of low molecular weight heparin therapy during labor. J

Matern Fetal Neonatal Med. 2005;17(1):39-43. 9. Rowan JA, McLintock C, Taylor RS, North RA. Prophylactic and therapeutic enoxaparin during pregnancy:

indications, outcomes and monitoring. Aust N Z J Obstet Gynaecol. 2003;43(2):123-128. 10. Nelson-Piercy C, Letsky EA, De Swiet M. Low-molecular-weight heparin for obstetric thromboprophylaxis:

experience of sixty-nine pregnancies in sixty-one women at high risk. Am J Obstet Gynecol. 1997;176(5):1062-1068.

11. Voke J, Keidan J, Pavord S, Spencer NH, Hunt BJ. The management of antenatal venous thromboembolism in the UK and Ireland: a prospective multicentre observational survey. Br J Haematol. 2007;139(4):545-558.

12. Bauersachs RM, Dudenhausen J, Faridi A et al. Risk stratification and heparin prophylaxis to prevent venous thromboembolism in pregnant women. Thromb Haemost. 2007;98(6):1237-1245.

13. Sanson BJ, Lensing AW, Prins MH et al. Safety of low-molecular-weight heparin in pregnancy: a systematic review. Thromb Haemost. 1999;81(5):668-672.

14. Lepercq J, Conard J, Borel-Derlon A et al. Venous thromboembolism during pregnancy: a retrospective study of enoxaparin safety in 624 pregnancies. BJOG. 2001;108(11):1134-1140.

15. Begley CM. A comparison of 'active' and 'physiological' management of the third stage of labour. Midwifery. 1990;6(1):3-17.

16. Khan GQ, John IS, Wani S, Doherty T, Sibai BM. Controlled cord traction versus minimal intervention techniques in delivery of the placenta: a randomized controlled trial. Am J Obstet Gynecol. 1997;177(4):770-774.

17. Rogers J, Wood J, McCandlish R et al. Active versus expectant management of third stage of labour: the Hinchingbrooke randomised controlled trial. Lancet. 1998;351(9104):693-699.

18. Nordstrom L, Fogelstam K, Fridman G, Larsson A, Rydhstroem H. Routine oxytocin in the third stage of labour: a placebo controlled randomised trial. Br J Obstet Gynaecol. 1997;104(7):781-786.

19. van Wijk FH, Wolf H, Piek JM, Buller HR. Administration of low molecular weight heparin within two hours before caesarean section increases the risk of wound haematoma. BJOG. 2002;109(8):955-957.

Postpartum haemorrhage in women receiving therapeutic doses of LMWH: a cohort study

125

Chapter 11

Recurrent miscarriage in women with and without antiphospholipid syndrome: prognosis for the next pregnancy outcome

Cohn DM, Middeldorp S, Korevaar JC, Dawood F, Goddijn M, Farquharson RG. Submitted for publication.

Abstract

Background

Antiphospholipid syndrome (APS) is frequently diagnosed in women with recurrent

miscarriage. Nevertheless, the outcome of a subsequent pregnancy in these women is not

clearly established.

Objective

To assess the outcome of a subsequent pregnancy in women with APS and recurrent

miscarriage.

Methods

We performed a cohort study among all women who attended the Miscarriage Clinic at

Liverpool Women’s Hospital between 1987 and 2006 referred with recurrent miscarriage

(�2 consecutive pregnancy losses). We compared the pregnancy outcome (live birth rate,

miscarriage rate, birth weight, gestational age and intra-uterine growth restriction) of the

subsequent index pregnancy in women with APS with the outcome of women with

unexplained recurrent miscarriage. Furthermore, the influence of treatment on live birth

rate was assessed.

Results

A total of 693 women fulfilled the selection criteria, of whom 176 (25%) had APS. 122 (69%)

women with APS had a subsequent live birth compared to 324 (63%) women with

unexplained recurrent miscarriage (OR 1.3, 95%CI 0.9 to 1.9). No differences were found

for birth weight, gestational age, and intra-uterine growth restriction between these two

groups. When treatment choice was analyzed, 53/67 (79%) of women with APS who had

received aspirin and heparin during their pregnancy had a live birth, compared to 64/104

(62%) of women with APS who received aspirin only (adjusted OR 2.7, 95%CI 1.3 to 5.8)

and compared to 204/305 (67%) of women with unexplained miscarriage who received no

treatment (adjusted OR 2.2, 95%CI 1.1 to 4.2).

Conclusion

Live birth rate between women with recurrent miscarriage and APS and women with

unexplained recurrent miscarriage was comparable. In women with APS, combined use of

aspirin and heparin was associated with a higher live birth rate as compared to women with

APS who were treated with aspirin only and as compared to women with unexplained

recurrent miscarriage without treatment.

Chapter 11

128

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Recurrent miscarriage is fairly common with a prevalence of 0.4-2% amongst couples who

try to conceive (depending on the definition of 2 or 3 consecutive miscarriages).1,2 Major

determinants of the prognosis following recurrent miscarriage are maternal age, the

number of preceding miscarriages, and whether or not an underlying cause is found.

Therefore, diagnosing an underlying cause is an essential part of investigation and for

appropriate counselling of couples with recurrent miscarriage. Known risk factors for

recurrent miscarriage include anatomical, hormonal or chromosomal abnormalities and

the antiphospholipid syndrome (APS).3 However, the cause of recurrent miscarriage

remains unexplained in more than 50% of couples with recurrent miscarriage.4,5

Antiphospholipid syndrome (APS) is an acquired condition, defined as the presence of

thrombosis or pregnancy loss or maternal morbidity and persistent circulating

antiphospholipid antibodies in plasma.6,7 The prognosis of a subsequent pregnancy in

women with APS and recurrent miscarriage is not clearly established. Most estimates are

based on a few small randomised trials that have assessed the efficacy of aspirin, with or

without heparin, to improve the live birth in women with APS after recurrent miscarriage,

mostly without a no treatment arm.8 Since participants of trials do not necessarily reflect

the general APS population, these results are not easily translated to daily practice.

Treatment guidelines vary with regard to the administration of heparin for APS and

recurrent miscarriage. The American College of Chest Physicians (ACCP) guidelines

recommend the combination of low-dose aspirin and a low dose of either unfractionated or

low molecular weight heparin (LMWH) (level of evidence: 1B),9 whereas the European

Society for Human Reproduction and Embryology (ESHRE) guidelines recommend the

prescription of aspirin with or without LMWH (level of evidence: 2B).10

We aimed to assess the prognosis for a subsequent pregnancy in a large cohort of women

with recurrent miscarriage and APS and compared this with women with unexplained

recurrent miscarriage.

Material and Methods

We performed a cohort study including all women with recurrent miscarriages (defined as

miscarriage <24 weeks of gestation) who attended the Miscarriage Clinic at Liverpool

Women’s Hospital, Liverpool, UK between 1986 and 2006. We compared women with

established APS to women in whom APS was excluded (for diagnostic criteria see Table 1).

Recurrent miscarriage in women with and without antiphospholipid syndrome

129

Table 1. Selection criteria of the study participants

Inclusion criteria

�2 consecutive miscarriages < 24 weeks gestational age delivery at Liverpool Women’s Hospital APS definition according to Sapporo criteria6 IgM anticardiolipin antibodies � 6 U/mL IgG anticardiolipin antibodies � 11 U/mL diluted Russell venom viper test (DRVVT) � 1.10

Exclusion criteria

other causes for recurrent miscarriage chromosomal abnormalities in the participant or in the male partner major (congenital) uterine abnormalities endocrine disorders at the time of previous miscarriages (including diabetes mellitus, thyroid dysfunction) pregnancy losses due to documented fetal formation or auto-immune disorders (e.g. SLE and thrombophilia) incomplete data sets (including outcome of tests for other causes for recurrent miscarriages)

This study was approved by the Local Research Ethics Committee (LREC reference number:

08/H1017/72).

All women underwent a standardised investigation sequence, as previously reported.11,12

This included testing for thrombophilia (APS and acquired activated protein C resistance

(APCR)), chromosome abnormalities (in both partners), thyroid dysfunction, diabetes

mellitus, bacterial vaginosis and uterine abnormalities. By hospital protocol, all women

with APS received aspirin. Heparin was prescribed according to the pregnancy loss type and

patient choice. From 1988 onward, heparin was administered as low-molecular-weight

heparin and was offered as a choice to women with a history of late pregnancy loss.

The primary outcome measure was the live birth rate in the first index pregnancy

subsequent to the referral and investigation visit to the clinic. Secondary outcome measures

were miscarriage rate within 13 weeks of gestation, rate of miscarriage between 13 and 24

weeks of gestation, stillbirth (loss > 24 weeks of gestation), and APS related obstetric

complications: intra uterine growth restriction (IUGR) (birth weight < 10th percentile) and

premature delivery (prior to the 36th week of gestation).

To calculate differences in distribution of the data, independent sample T-tests (two tailed)

were used for continuous variables in case of two groups and one-way ANOVA-tests were

performed for comparison between more than two groups. �2-tests were used in case of

categorical variables. To adjust for potential confounders, we performed binary logistic

regression analysis. Covariates which showed a linear relationship were entered as a

continuous variable. Mean differences and 95% confidence intervals were calculated for

continuous data, and odds ratios (OR) and 95% confidence intervals were calculated for

tests positive on two different occasions with an interval of 6 or more weeks

130

Chapter 11

1

2

3

4

5

6

7

8

9

10

11

12

categorical data. Baseline characteristics were stratified for therapy, to identify differences

in prognostic variables between treatment groups. For the primary outcome measure, we

calculated crude odds ratios (OR), as well as ORs adjusted for maternal age and number of

previous miscarriages. We also stratified the primary outcome measure for therapy (aspirin,

combined treatment of aspirin and heparin or none), in which women who had

unexplained recurrent miscarriage and had received no treatment were the reference group.

For all secondary outcome measures, we calculated crude ORs (only for miscarriage we also

calculated ORs adjusted for age and number of previous miscarriages).

All data entries were double checked by a second independent investigator. In addition,

random validation checks were performed. Missing or inconsistent data were assessed for

random distribution by comparison of baseline characteristics, primary and secondary

outcomes with those in women in whom the respective data were not missing. Random

distribution was assumed, if this comparison did not demonstrate a difference.

Subsequently, we excluded participants with missing or inconsistent data from the

particular analysis. All data were analysed with SPSS software (version 16.0.2).

Figure. Flow chart of patient selection

APS = antiphospholipid syndrome

131

Recurrent miscarriage in women with and without antiphospholipid syndrome

Results

A total of 693 women met the selection criteria, of whom 176 (25%) were diagnosed with

APS (for exclusions: see Figure). Baseline characteristics are listed in Table 2 and were

similar in women with APS and women with unexplained recurrent miscarriage. Live birth

was observed in 122 (69%) women with treated APS and in 324 (63%) women with

unexplained recurrent miscarriage (OR adjusted for number of previous miscarriage and

maternal age 1.4, 95%CI 0.9 to 2.0), as shown in Table 3.

Miscarriage (<13 weeks) occurred in 49/176 women with APS (28%) and in 176/517 women

with unexplained recurrent miscarriage (34%), OR 0.7 (95%CI 0.5 to 1.1) (Table 4).

Late pregnancy loss (13-24 weeks) occurred in 3/176 women with APS (2%) and in 15/517

women with unexplained recurrent miscarriage (3%), OR 0.6 (95%CI 0.2 to 2.0) and

stillbirth (>24 weeks) occurred in 2/176 women with APS (1.1%) and in 2/517 women with

unexplained recurrent miscarriage (0.4%), OR 3.0 (95%CI 0.4 to 21) (Table 4).

No differences were found for birth weight or gestational age between both study groups

(Table 4). IUGR was observed in 7 (7%) women with APS as compared to 29 (10%) women

with unexplained recurrent miscarriage OR 0.6 (95%CI 0.3 to 1.5). Premature delivery

occurred in 8 (7%) women with APS and 25 (8%) women with unexplained recurrent

miscarriage, OR 0.9 (95%CI 0.4 to 2.0).

Determinants for heparin therapy were a history of late miscarriage as well as patient

preference. Women with a history of two or more late pregnancy losses were approximately

four times more likely to receive heparin treatment as compared to women with a history of

no or one late pregnancy loss (p=0.014 for women with APS and p=0.003 for women with

unexplained recurrent miscarriage). Women with unexplained recurrent miscarriage who

received no treatment were significantly younger than women with unexplained recurrent

miscarriage who received treatment with aspirin, either or not combined with heparin

(mean age 31 vs. 33 years, respectively) (Table 2). Stratification for therapy showed that the

combination of low dose aspirin and heparin was associated with a higher chance of live

birth in women with APS (53/67; 79%) as compared to women with APS who were treated

with aspirin only (live birth 64/104; 62%), adjusted OR 2.7 (95% CI 1.3 to 5.8), and as

compared to women with unexplained recurrent miscarriage who had had no treatment

(live birth 204/305; 67%), adjusted OR 2.2 (95%CI 1.1 to 4.2) (Table 3).

132

Chapter 11

1

2

3

4

5

6

7

8

9

10

11

12

Table 2. Baseline characteristics of wom

en with recurrent m

iscarriage.

All wom

en N=693

APS N=165

*

Unexplained recurrent m

iscarriage N=482

*

APS aspirin only N=98

**

APS aspirin and heparin N

=62**

Unexplained recurrent m

iscarriage aspirin only N

=146**

Unexplained recurrent m

iscarriage aspirin and heparinN=36

**

Unexplained recurrent m

iscarriage no treatm

ent N=294

**

Mean age, years (SD)

32 (5.6) 32 (5.3)

32 (5.7) p=0.66

32 (5.9) 32 (4.2)

p=0.55 33 (5.6)

33 (5.6) 31 (5.5)

p=0.008Num

ber of previous m

iscarriages 1

p=0.62

p=0.013

p=0.54

2 miscarriages

196 (30%)

46 (28%)

150 (31%)

26 (27%

) 20 (32%

)

42 (29%)

9 (25%)

98 (33%)

3 miscarriages

263 (41%)

75 (46%)

188 (39%)

46 (47%

) 25 (40%

)

60 (40%)

13 (36%)

113 (39%)

4 miscarriages

99 (15%)

24 (14%)

75 (15%)

17 (17%

) 7 (11%

)

22 (15%)

8 (22%)

45 (15%)

5 miscarriages

46 (7%)

9 (5%)

37 (8%)

7 (7%

) 1 (2%

)

8 (6%)

4 (11%)

23 (8%)

� 5 miscarriages

43 (7%)

11 (7%)

32 (7%)

2 (2%

) 9 (15%

)

14 (10%)

2 (6%)

15 (5%)

APS = antiphospholipid syndrome, SD= standard deviation; 1 m

iscarriage within 24 weeks of gestation. *Reproductive history was not recorded in 11 women with APS and 35 wom

en unexplained recurrent m

iscarriage **Treatment was not recorded in 5 wom

en with APS and 6 women unexplained recurrent m

iscarriage. Table 3. Pregnancy outcom

e in wom

en with recurrent m

iscarriage

All wom

en N=693

APS N=176

Unexplained recurrent m

iscarriage N=517

APS aspirin only N=104

*

APS aspirin and heparin N=67

*

Unexplained recurrent m

iscarriage aspirin only N=163

*

Unexplained recurrent m

iscarriage aspirin and heparin N

=43*

Unexplained recurrent m

iscarriage no treatm

ent N=305

*

Live birth, N (%)

446 (64%)

122 (69%)

324 (63%)

64 (62%)

53 (79%)

93 (57%)

25 (58%)

204 (67%)

OR live birth (95%CI)

- 1.3 (0.9 to 1.9)

1(ref) 0.8 (0.5 to 1.3)

1.9 (1.0 to 3.5) 0.7 (0.4 to 1.0)

0.7 (0.4 to 1.3) 1 (ref)

OR live birth adjusted (95%CI) **

- 1.4 (0.9 to 2.0)

1 (ref) 0.8 (0.5 to 1.3)

2.2 (1.1 to 4.2) 0.8 (0.5 to 1.1)

0.7 (0.4 to 1.4) 1 (ref)

APS = antiphospholipid syndrome, OR = odds ratio, CI = confidence interval. *Treatm

ent was not recorded in 5 women with APS and 6 wom

en unexplained recurrent miscarriage, **Adjusted for

age and number of previous m

iscarriages.

133

Recurrent miscarriage in women with and without antiphospholipid syndrome

In women with unexplained recurrent miscarriage, treatment was not associated with

higher live birth rates: 163 women received aspirin, of whom 93 (57%) had a live birth,

adjusted OR 0.8 (95%CI 0.5 to 1.1) and 43 women received aspirin and heparin treatment,

of whom 25 (58%) had a live birth, adjusted OR 0.7 (95% CI 0.4 to 1.4).

Table 4. Secondary outcomes in women with recurrent miscarriage

All women N=693

APS N=176

Unexplained recurrent miscarriage N=517

First trimester miscarriage (loss <13 weeks) N (%) 225 (33%) 49 (28%) 176 (34%)

OR miscarriage (95%CI) - 0.7 (0.5 to 1.1) 1 (ref)

OR miscarriage adjusted (95%CI)* - 0.7 (0.5 to 1.1) 1 (ref)

Late miscarriage (loss between 13 and 24 weeks) N (%) 18 (3%) 3 (2%) 15 (3%)

OR late pregnancy loss (95%CI) - 0.6 (0.2 to 2.0) 1 (ref)

Stillbirth (loss >24 weeks) N (%) 4 (0.6%) 2 (1.1%) 2 (0.4%)

OR IUFD - 3.0 (0.4 to 21) 1 (ref)

Mean birth weight, grams (SD) 3211 (692) 3168 (603) 3265 (665)

Mean difference, grams (95%CI) - -96 (-240 to 48)

Mean gestational age, weeks (SD) 39 (2.6) 39 (2.4) 39 (2.8)

Mean difference, weeks (95%CI) - -0.3 (-0.9 to 0.3)

IUGR, N (%) 36 (9%) 7 (7%) 29 (10%)

OR IUGR - 0.6 (0.3 to 1.5) 1 (ref)

Premature delivery, N (%) 33 (8%) 8 (7%) 25 (8%)

OR premature delivery - 0.9 (0.4 to 2.0) 1 (ref)

APS = antiphospholipid syndrome, IUGR = intra-uterine growth restriction, OR = odds ratio, CI = confidence interval. * Adjusted for age and number of previous miscarriages

Discussion

In this large cohort study, performed in a tertiary referral centre, we observed that the

prognosis of a subsequent pregnancy following recurrent miscarriage was similar in

women with APS and women with unexplained recurrent miscarriage. Live births were

observed in 69% of 176 women with APS and in 63% of 517 women with unexplained

recurrent miscarriage. Previously reported live birth rates in women with APS varied

between 42% and 100%.11,13-22 However, comparison between studies is difficult for various

reasons.

This study was an observational cohort study, whereas all other studies were randomized

controlled trials, allocating patients with APS to either one or more of the following

treatments: aspirin11,13-22, prednisone17, heparin11,14,18-20, IVIG22, placebo15-17 or usual care13.

134

Chapter 11

1

2

3

4

5

6

7

8

9

10

11

12

These trials applied various diagnostic criteria for APS (single or repetitive positivity for

IgM/IgG ACA only and/or LA) and for recurrent miscarriage (2 or 3 miscarriages,

consecutive or non consecutive miscarriages), and had relatively small sample sizes

ranging from only 16 to 202 participants. It is interesting to note that studies with the

smallest sample sizes reported the highest successful pregnancy rates. This may be the

result of publication, referral or selection bias, or lack of concealed allocation bias of small

trials with positive results. When the studies are confined to those with more than 50

participants, the successful pregnancy rates vary between 42% and 80%, which is more in

line with our findings.11,17-19

Successful pregnancy outcome in women with unexplained recurrent miscarriage occurred

less frequently as compared to a previous report from the same clinic (63% vs. 75%).23

However, the definition of “successful pregnancy outcome” differed between the two

studies: live birth in our study versus “survival beyond 24 weeks” in the latter study.

Furthermore, in contrast to the latter study, we also included women with a history of late

miscarriage which may have accounted for the observed difference in live birth rates. The

observed difference is not easily interpreted otherwise, as the most important predictors of

a future live birth (i.e. maternal age and obstetric history) were similar in both reports. The

observed live birth rates in this study are similar to the observed live birth rates as reported

in other studies.24,25

We observed that combined treatment of aspirin and heparin was associated with a higher

live birth incidence in women with APS as compared to women with APS who received

aspirin only. This supports findings from previous trials that showed a beneficial effect of

combined treatment with aspirin and unfractionated heparin over aspirin alone in women

with APS and recurrent miscarriage.18,19 However, these results contrast with level 1B

evidence from the largest reported prospective RCT from the same centre as the present

study, in which treatment with aspirin was compared to treatment with aspirin combined

with low-molecular-weight heparin.11 Only a small, non significant difference in live birth

was observed in the group treated with aspirin and heparin as compared to the group who

received aspirin alone (78% vs. 72%, OR live birth 1.39, 95%CI 0.55 to 3.47). Age and

obstetric history were equally comparable between the study participants of this trial and

our study.

Our study has several strengths. First, the large number of women with recurrent

miscarriage and APS in our study enabled us to perform clinically relevant subgroup

analyses, such as stratification for therapy. Furthermore, the observational design provides

Recurrent miscarriage in women with and without antiphospholipid syndrome

135

a good reflection of the course of a subsequent pregnancy in APS patients with recurrent

miscarriage. Third, the diagnosis of APS in all participants was performed in the same

laboratory, thereby ensuring homogeneity of our patient population.

Some limitations warrant comment. First, although the combined treatment of aspirin and

heparin was associated with a higher rate of live births in women with APS and recurrent

miscarriage, our observational study design (grade 2 level of evidence) does not rule out

that this observation is confounded by indication. Indeed, heparin treatment was

prescribed four times more often in women with a history of two or more late pregnancy

losses. Nevertheless, the presence of more known and unknown unfavourable prognostic

variables in women who have received heparin is likely to underestimate, not overestimate,

the association between treatment and live birth. Second, the retrospective design of the

study posed us with a number of missing data. However, comparison of baseline

characteristics of women with and without missing data showed random distribution, thus

making us confident that this has not affected our results. Third, we used the 1999 Sapporo

criteria to diagnose APS.6 These criteria have been adapted since the initiation of this study.7

Thus, our findings may not apply to women testing only positive for anti-�2-glycoprotein-1

antibodies. The extension of the time between initial and repeated testing for APS related

antibodies from 6 weeks to 12 weeks possibly has attenuated our findings, as we may have

included women with assumed APS who had only transient antiphospholipid antibodies.

In conclusion, our large cohort study showed that the overall prognosis of a successful

pregnancy in women with recurrent miscarriage and APS treated with heparin and aspirin

is good.

Acknowledgements:

We would like to greatly acknowledge Linda Roberts who was pivotal in the data collection.

Furthermore we would like to acknowledge the IT department of Liverpool Women’s

Hospital for their assistance.

Dr S. Middeldorp is Clinical Established Investigator of the Netherlands Heart Foundation

(2008T056).

References

1. Salat-Baroux J. [Recurrent spontaneous abortions]. Reprod Nutr Dev 1988;28:1555-68. 2. Alberman E. Antenatal and perinatal causes of handicap: epidemiology and causative factors. Baillieres Clin

Obstet Gynaecol 1988;2:9-19. 3. Rai R, Regan L. Recurrent miscarriage. Lancet 2006;368:601-11.

Chapter 11

136

1

2

3

4

5

6

7

8

9

10

11

12

4. Stephenson MD. Frequency of factors associated with habitual abortion in 197 couples. Fertil Steril 1996;66:24-9.

5. Porter TF, Scott JR. Evidence-based care of recurrent miscarriage. Best Pract Res Clin Obstet Gynaecol 2005;19:85-101.

6. Wilson WA, Gharavi AE, Koike T et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999;42:1309-11.

7. Miyakis S, Lockshin MD, Atsumi T et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295-306.

8. Empson M, Lassere M, Craig J, Scott J. Prevention of recurrent miscarriage for women with antiphospholipid antibody or lupus anticoagulant. Cochrane Database Syst Rev 2005;CD002859.

9. Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J. Venous thromboembolism, thrombophilia, antithrombotic therapy, and pregnancy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133:844S-86S.

10. Jauniaux E, Farquharson RG, Christiansen OB, Exalto N. Evidence-based guidelines for the investigation and medical treatment of recurrent miscarriage. Hum Reprod 2006;21:2216-22.

11. Farquharson RG, Quenby S, Greaves M. Antiphospholipid syndrome in pregnancy: a randomized, controlled trial of treatment. Obstet Gynecol 2002;100:408-13.

12. Drakeley AJ, Quenby S, Farquharson RG. Mid-trimester loss--appraisal of a screening protocol. Hum Reprod 1998;13:1975-80.

13. Cowchock S, Reece EA. Do low-risk pregnant women with antiphospholipid antibodies need to be treated? Organizing Group of the Antiphospholipid Antibody Treatment Trial. Am J Obstet Gynecol 1997;176:1099-100.

14. Noble LS, Kutteh WH, Lashey N, Franklin RD, Herrada J. Antiphospholipid antibodies associated with recurrent pregnancy loss: prospective, multicenter, controlled pilot study comparing treatment with low-molecular-weight heparin versus unfractionated heparin. Fertil Steril 2005;83:684-90.

15. Tulppala M, Marttunen M, Soderstrom-Anttila V et al. Low-dose aspirin in prevention of miscarriage in women with unexplained or autoimmune related recurrent miscarriage: effect on prostacyclin and thromboxane A2 production. Hum Reprod 1997;12:1567-72.

16. Pattison NS, Chamley LW, Birdsall M, Zanderigo AM, Liddell HS, McDougall J. Does aspirin have a role in improving pregnancy outcome for women with the antiphospholipid syndrome? A randomized controlled trial. Am J Obstet Gynecol 2000;183:1008-12.

17. Laskin CA, Bombardier C, Hannah ME et al. Prednisone and aspirin in women with autoantibodies and unexplained recurrent fetal loss. N Engl J Med 1997;337:148-53.

18. Kutteh WH. Antiphospholipid antibody-associated recurrent pregnancy loss: treatment with heparin and low-dose aspirin is superior to low-dose aspirin alone. Am J Obstet Gynecol 1996;174:1584-9.

19. Kutteh WH, Ermel LD. A clinical trial for the treatment of antiphospholipid antibody-associated recurrent pregnancy loss with lower dose heparin and aspirin. Am J Reprod Immunol 1996;35:402-7.

20. Rai R, Cohen H, Dave M, Regan L. Randomised controlled trial of aspirin and aspirin plus heparin in pregnant women with recurrent miscarriage associated with phospholipid antibodies (or antiphospholipid antibodies). BMJ 1997;314:253-7.

21. Silver RK, MacGregor SN, Sholl JS, Hobart JM, Neerhof MG, Ragin A. Comparative trial of prednisone plus aspirin versus aspirin alone in the treatment of anticardiolipin antibody-positive obstetric patients. Am J Obstet Gynecol 1993;169:1411-7.

22. Branch DW, Peaceman AM, Druzin M et al. A multicenter, placebo-controlled pilot study of intravenous immune globulin treatment of antiphospholipid syndrome during pregnancy. The Pregnancy Loss Study Group. Am J Obstet Gynecol 2000;182:122-7.

23. Brigham SA, Conlon C, Farquharson RG. A longitudinal study of pregnancy outcome following idiopathic recurrent miscarriage. Hum Reprod 1999;14:2868-71.

24. Clifford K, Rai R, Regan L. Future pregnancy outcome in unexplained recurrent first trimester miscarriage. Hum Reprod 1997;12:387-9.

25. Carp H, Feldman B, Oelsner G, Schiff E. Parental karyotype and subsequent live births in recurrent miscarriage. Fertil Steril 2004;81:1296-301.

Recurrent miscarriage in women with and without antiphospholipid syndrome

137

Chapter 12

Increased sperm count maintains high population frequency of Factor V Leiden

Cohn DM, Repping S, Büller HR, Meijers JC, Middeldorp S. J Thromb Haemost 2010; in press.

Abstract

Background

Factor V Leiden (FVL) increases the risk of venous thrombosis and pregnancy loss in

carriers. Nevertheless, this relatively old mutation is prevalent in Caucasian populations,

which could be explained by positive selection pressure. Men with FVL have previously been

found to have higher fecundity (the time between marriage and first pregnancy). Whether

this is caused by increased sperm counts in men with FVL is unknown.

Objectives

To assess whether men with Factor V Leiden have increased sperm counts.

Methods

We performed a prospective cohort study among 1139 consecutively included male partners

of subfertile couples presenting at our university hospital for fertility workup between

January 2000 and July 2007. All potential candidates who gave informed consent were

included, irrespective of their fertility workup. In this retrospective analysis, we excluded

participants with known causes of spermatogenic function or azoospermia. Subsequently,

we genotyped all participants and compared sperm counts between FVL carriers and non-

carriers.

Results

We identified 37 FVL carriers and 921 non-carriers. FVL carriers had higher total sperm

counts and total motile sperm counts than non-carriers: 236 x106 (95%CI 158-292 x106)

versus 163 x106 (95%CI 147-178 x106) and 81 x106 (95%CI 54-105 x106) versus 52 x106 (95%CI

48-57 x106), respectively.

Conclusion

Our results provide a possible explanation for the high prevalence of FVL among

Caucasians. To our knowledge, this is the first study that indicates that an increased

prevalence of a genotype is controlled by increased sperm counts.

140

Chapter 12

1

2

3

4

5

6

7

8

9

10

11

12

Introduction

Factor V Leiden (rs6025, FVL) is the most common inherited thrombophilic defect in

Western countries.1 This gain-of-function mutation leads to resistance of activated clotting

factor V through inactivation by activated protein C.2,3 Carriers are at increased risk of

venous thrombosis and spontaneous or recurrent miscarriage.3,4

Despite these obvious disadvantages, the mutation -which occurred about 21,000 - 34,000

years ago5- has a high prevalence of approximately 4-7% in Caucasians.6 It has long been

thought that the high population frequency of FVL reflects some sort of evolutionary

benefit for carriers.7-11 Interestingly, male, but not female FVL carriers display a higher

fecundity rate (the time between marriage and first pregnancy) than non-carriers.12 We

hypothesized that men with FVL have high sperm counts thereby increasing their chance to

establish a pregnancy and spreading their genotype.

Methods

Pilot study

We performed a pilot study between August and November 2008 among nineteen male FVL

carriers, who had been tested for the mutation in a previous study that investigated the

associated risk of venous thrombosis13 or who had been tested for clinical purposes because

of a history of venous thrombosis. These participants donated two semen samples with a

two-week time interval, each after three or more days of sexual abstinence, at the fertility

laboratory of the Academic Medical Center, Amsterdam, the Netherlands. The technicians

who performed the semen analyses were unaware of the participants’ FVL status. Donated

semen was analyzed for volume, sperm concentration, sperm motility and sperm

morphology according to the WHO guidelines.14 Total sperm count and total motile sperm

count were calculated. The outcomes were compared to previously published cohorts of the

general population.15-18

Cohort study

Participant identification

Subsequently, we performed a formal cohort analysis. We genotyped a cohort of

consecutive male partners of subfertile couples presenting at the Center for Reproductive

Medicine of the Academic Medical Center for a fertility workup, from January 2000 until

July 2007. All men were consecutively included prior to semen analyses. None of the

participants of the pilot study had also been included in this cohort study. The study was

141

Increased sperm count maintains high population frequency of factor V Leiden

approved by the Institutional Review Board of the Academic Medical Center. Written

informed consent to store and use their DNA for research purposes was obtained from all

men.

Exclusion criteria

We excluded men with known causes of spermatogenic failure: i.e. hyperprolactinaemia,

hypogonadotrophic hypogonadism, previous chemo- or radiotherapy, bilateral

cryptorchidism, congenital absence or surgery of the vas deferens, history of orchitis, and

bilateral orchidectomy. Participants were also excluded from this analysis if the fertility

workup identified retrograde ejaculation, azoospermia, an AFZa, P5/proximal-P1, P5/distal-

P1, AZFc or gr/gr deletion, or numerical or structural chromosome abnormalities.

PCR analysis

The FVL mutation (G1691A) was identified by PCR on a Biorad CFX96 or Roche Lightcycler

480 Real Time System. The laboratory technicians who performed the PCR analyses were

unaware of the results of the semen analyses.

Statistics

The average of the participant’s two or more semen analyses was included in the statistical

analysis. In case of normally distributed data, we calculated means and 95% confidence

intervals using SPSS version 16.0. For non-normally distributed data, we calculated median

values and 95% confidence intervals by performing bootstrap procedures in “The R Project

for Statistical Computing”, available at: http://www.r-project.org/.

Results

The pilot study suggested that FVL carriers had increased sperm concentration (median 75

x106/mL, 95%CI 58 - 95 x106/mL) and increased total sperm count (median 192 x106, 95%CI

132 - 247 x106) as compared to previously published cohorts of the general population (see

Figure 1).15-18 We also assessed total motile sperm count: median 77 x106 (95%CI 55 - 105

x106), however, this parameter could not be compared to the general population, as it had

not been reported.15-18

Of 1139 screened participants of the cohort study, 181 met at least one exclusion criterion

(for reasons of exclusion: see Figure 2). At least two semen samples were available for each

participant. We identified 37 men heterozygous for FVL and 921 men without the mutation.

142

Chapter 12

1

2

3

4

5

6

7

8

9

10

11

12

Figure 1. Semen analyses of 19 factor V Leiden carriers (pilot study) as compared to the general population15-18

Volume

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

mean

mL

Concentration

0

25

50

75

median

x10^

6/m

L

Motility

0102030405060708090

median

% m

otile

cel

ls

Morphology

0

10

20

30

40

mean

% n

orm

al

Total sperm count

0

50

100

150

200

median

x10^

6

Pilot StudyAndersen15

Jensen16

Jensen17

Paasch ("Hamburg" cohort)18

Paasch ("Leipzig" cohort)18

As depicted in the Table, median total sperm count was increased in FVL carriers as

compared to non-carriers: 236 x106 (95%CI 158-292 x106) and 163 x106 (95%CI 147-178 x106),

respectively. In addition, we found increased total motile sperm counts in FVL carriers as

compared to our controls: 81 x106 (95%CI 54-105 x106) versus 52 x106 (95%CI 48-57 x106).

Increased sperm count maintains high population frequency of factor V Leiden

143

Figure 2. Flow chart of the cohort study

1139 potentially eligible participants

PCR analysis in 961 participants

37 FVL+ 921 FVL-

178 participants excluded:karyotype abnormalities (n=12)bilateral cryptorchidism (n=39)no semen analysis available (n=33)no DNA available (n=31)azoospermia (n=52)retrograde ejaculation (n=2)AZFc deletions (n=7)AZFa deletion (n=1)P5/distal-P1 deletion (n=1)

PCR failed (n=3)

FVL=factor V Leiden Table. Semen analyses of 37 carriers of the factor V Leiden mutation as compared to 921 non-carriers

FFVL + FVL -

participants (n) 37 921

mean age (y) 36 (34 - 38) 37 (36 - 37)

mean days of sexual abstinence (n) 4.0 (3.5 - 4.5) 4.2 (4.1 - 4.4)

semen quality

mean volume (mL) 3.7 (3.3 - 4.2) 3.4 (3.3 - 3.5)

median concentration (106/mL) 58 (47 - 80) 56 (52 - 60)

median motility (% progressive) 36 (28 - 41) 34 (32 - 35)

mean morphology (% normal) 38 (34 - 43) 35 (34 - 36)

median total count (106) 236 (158 - 292) 163 (147 - 178)

median total motile count (106) 81 (54 - 105) 52 (48 - 57)

FVL= factor V Leiden. Means and medians are presented with 95% confidence intervals

Chapter 12

144

1

2

3

4

5

6

7

8

9

10

11

12

Discussion

Our data show that men with FVL have increased total sperm counts and total motile sperm

counts. The observed high frequency of FVL among the population could thus be the

evolutionary result of a balance between allele loss through thrombosis and pregnancy loss,

and allele gain through increased male fertility. Higher sperm counts are likely to increase

the chance of a successfully established pregnancy, as lower sperm counts are associated

with decreased fertility. However, no studies have yet focused on determinants of increased

male fertility.

Our results provide an explanation for the increased fecundity found in men with FVL as

described in a previous study.12 In this large case-control study among 1176 subjects, the

relative risk of conception within 3 months after marriage was 3.5 (95%CI 2.1-5.7) for men

with FVL as compared to men without FVL.

Also, a biological explanation may be possible for the association between FVL and higher

sperm counts, even though a direct function of the factor V protein in spermatogenesis has

not been investigated. However, two studies have found an excess of breakpoints in

chromosome 1, including the FVL locus on 1q23, among infertile males.19,20 Moreover, two

infertile brothers with a pericentric inversion of chromosome 1 including the FVL locus

(p34q23) were shown to suffer from severe oligozoospermia.21 Thus, it could be that the

1q23 locus harbours an important gene for spermatogenesis.

Our data suggest two directions for future research. First, an investigation into reasons for

increased fertility and fecundity, in particular higher sperm counts, is needed. Second, the

biological explanation of the relationship between FVL and high sperm counts may offer an

opportunity to gain more fundamental insight into male fertility regulation.

Acknowledgements

We would like to thank Professor F.R. Rosendaal for his useful methodological

suggestions, J. Peter for technical assistance and M. Vergeer for assistance with the

statistical analyses. We thank Professor F. van der Veen for his intellectual contribution.

Dr S. Middeldorp is Clinical Established Investigator of the Netherlands Heart Foundation

(2008T056).

References 1. Rees DC. The population genetics of factor V Leiden (Arg506Gln). Br J Haematol 1996;95:579-86. 2. Dahlback B, Hildebrand B. Inherited resistance to activated protein C is corrected by anticoagulant cofactor

activity found to be a property of factor V. Proc Natl Acad Sci U S A 1994;91:1396-400.

145

Increased sperm count maintains high population frequency of factor V Leiden

3. Bertina RM, Koeleman BP, Koster T et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994;369:64-7.

4. Middeldorp S. Thrombophilia and pregnancy complications: cause or association? J Thromb Haemost 2007;5 Suppl 1:276-82.

5. Zivelin A, Griffin JH, Xu X et al. A single genetic origin for a common Caucasian risk factor for venous thrombosis. Blood 1997;89:397-402.

6. Cohn DM, Roshani S, Middeldorp S. Thrombophilia and venous thromboembolism: implications for testing.Semin Thromb Hemost 2007;33:573-81.

7. Lindqvist PG, Svensson PJ, Dahlback B, Marsal K. Factor V Q506 mutation (activated protein C resistance) associated with reduced intrapartum blood loss--a possible evolutionary selection mechanism. Thromb Haemost 1998;79:69-73.

8. Lindqvist PG, Zoller B, Dahlback B. Improved hemoglobin status and reduced menstrual blood loss among female carriers of factor V Leiden--an evolutionary advantage? Thromb Haemost 2001;86:1122-3.

9. Lindqvist PG, Dahlback B. Carriership of Factor V Leiden and evolutionary selection advantage. Curr Med Chem 2008;15:1541-4.

10. Gopel W, Ludwig M, Junge AK, Kohlmann T, Diedrich K, Moller J. Selection pressure for the factor-V-Leiden mutation and embryo implantation. Lancet 2001;358:1238-9.

11. van Dunne FM, Doggen CJ, Heemskerk M, Rosendaal FR, Helmerhorst FM. Factor V Leiden mutation in relation to fecundity and miscarriage in women with venous thrombosis. Hum Reprod 2005;20:802-6.

12. van Dunne FM, De Craen AJ, Heijmans BT, Helmerhorst FM, Westendorp RG. Gender-specific association of the factor V Leiden mutation with fertility and fecundity in a historic cohort. The Leiden 85-Plus Study. Hum Reprod 2006;21:967-71.

13. Middeldorp S, Meinardi JR, Koopman MM et al. A prospective study of asymptomatic carriers of the factor V Leiden mutation to determine the incidence of venous thromboembolism. Ann Intern Med 2001;135:322-7.

14. World Health Organization. WHO Laboratory Manual for the Examination of Human Semen En Sperm-Cervical Mucus Interaction. Cambridge University Press., 1999.

15. Andersen AG, Jensen TK, Carlsen E et al. High frequency of sub-optimal semen quality in an unselected population of young men. Hum Reprod 2000;15:366-72.

16. Jensen TK, Andersson AM, Jorgensen N et al. Body mass index in relation to semen quality and reproductive hormones among 1,558 Danish men. Fertil Steril 2004;82:863-70.

17. Jensen TK, Jorgensen N, Asklund C et al. Self-rated health and semen quality among 3,457 young Danish men. Fertil Steril 2007;88:1366-73.

18. Paasch U, Salzbrunn A, Glander HJ et al. Semen quality in sub-fertile range for a significant proportion of young men from the general German population: a co-ordinated, controlled study of 791 men from Hamburg and Leipzig. Int J Androl 2008;31:93-102.

19. Bache I, Assche EV, Cingoz S et al. An excess of chromosome 1 breakpoints in male infertility. Eur J Hum Genet 2004;12:993-1000.

20. Manvelyan M, Schreyer I, Hols-Herpertz I et al. Forty-eight new cases with infertility due to balanced chromosomal rearrangements: detailed molecular cytogenetic analysis of the 90 involved breakpoints. Int J Mol Med 2007;19:855-64.

21. Meschede D, Froster UG, Bergmann M, Nieschlag E. Familial pericentric inversion of chromosome 1 (p34q23) and male infertility with stage specific spermatogenic arrest. J Med Genet 1994;31:573-5.

Chapter 12

146

Appendix

A call for a uniform reference output format for submission of manuscripts

Even though uniform requirements of output styles exist1,2, each biomedical journal has its

own editorial requirements for formatting -such as composition of the reference list-. I do

not question the relevance of individual output styles, as journals may consider this an

important characteristic its identity. However I would like to advocate a uniform output

style for submission of manuscripts (such as the Vancouver Style2). Converting the

manuscript to the various required output styles can become time consuming cumulatively,

especially if a manuscript is submitted multiple times due to prior rejections. Moreover, in

case a manuscript is not considered for peer review, the conversion according to the

journal’s stipulations had been redundant. Of course, the authors could be requested to

adapt the output style according to the journal’s requirements, but not before the journal’s

editor has decided to accept the manuscript for publication. In fact, saving the effort which

is nowadays needed for conversion of the format would allow us, researchers, to spend

more time on our actual experiments. In this manner both authors and editors would

benefit, as this could enhance the number of relevant scientific results. 1Patrias, Karen. Citing medicine: the NLM style guide for authors, editors, and publishers [Internet]. 2nd ed. Wendling, Daniel L., technical editor. Bethesda (MD): National Library of Medicine (US); 2007 [2009/02/17]. Available from: http://www.nlm.nih.gov/citingmedicine 2Monash University Library. Vancouver style (uniform requirements for manuscripts submitted to biomedical journals); 2006 [2009/02/17]. Available from: http://www.lib.monash.edu.au/tutorials/citing/vancouver.html

A call for a uniform reference output format for submission of manuscripts

149

Summary

Thrombophilia, which is the common term for a tendency towards excessive venous clot

formation, is the focus of this thesis. Chapter 1 constitutes a short introduction on venous

thromboembolism (VTE) and thrombophilia. The remaining chapters are classified into

three parts. The first part (chapters 2-6) discusses clinical and psychological aspects of VTE

and thrombophilia, the second part (chapters 7-9) addresses the search for new

thrombophilic factors and the third part (chapters 10-12) focuses on reproductive aspects of

VTE and thrombophilia.

Part I: clinical and psychological aspects of VTE and thrombophilia

Chapter 2 reviews the various thrombophilic factors (antithrombin deficiency, protein C

deficiency, protein S deficiency, factor V Leiden, prothrombin G20210A mutation, high

levels of clotting factor VIII and antiphospholipid syndrome) and their association with

VTE. Furthermore, the potential advantages of testing for thrombophilia are discussed, as

well as the disadvantages of testing. In addition, an outline of the NOSTRADAMUS

(Necessity Of Screening for ThRombophilia At Diagnosis of venous thromboembolism to

Assess Most Unresolved iSsues) study is presented. This study was designed to assess

whether testing for thrombophilia and prolonging treatment duration is beneficial in terms

of clinical outcomes (the composite endpoint of recurrent VTE and bleeding risk), quality

of life and costs. Unfortunately, this study was terminated early. This chapter concludes

with the various reasons for the early termination of the NOSTRADAMUS study.

In chapter 3, the available evidence for testing for thrombophilia after VTE is systematically

reviewed. We investigated whether testing for thrombophilia leads to reduction of the risk

of recurrent VTE by adjusting therapy or taking preventive measures. The Cochrane

Peripheral Vascular Diseases Group trials register (last searched 22 April 2008), MEDLINE,

EMBASE, and the CENTRAL database (last searched 2008, Issue 2), and reference lists were

searched for relevant papers. However, no trials were identified and, therefore, the

conclusion of this chapter is that randomized controlled trials are urgently needed to

address this issue.

Chapter 4 investigates the extent of psychological impact of testing for thrombophilia, such

as fear, depression and worry, which is a potential disadvantage of testing for

thrombophilia. Studies that determined the nature and extent of psychological impact

following testing for thrombophilia were systematically reviewed. We searched the

Summary

153

MEDLINE database (1966 to February 2008), the EMBASE database (1985 to 2008, week 5)

and the PsychInfo database (1806 to February 2008) for relevant trials that focused on the

psychological impact of testing for thrombophilia, without language restrictions. In

addition, bibliographies of relevant articles were scanned for additional articles. Six studies

fulfilled the eligibility criteria, but pooling of the data was not possible since these studies

varied tremendously in methodology. These six studies reported few negative results, but

most assessments were limited to the short term or lacked methodological accuracy.

Therefore, no valid conclusions can be drawn about the psychological impact of genetic

testing in patients based on the current available literature.

Chapters 5 and 6 address the development and validation of the Pulmonary Embolism

Quality of Life (PEmb-QoL) questionnaire. Quality of life has become a key component of

medical care and an important outcome measure of clinical trials. Indeed, several

questionnaires have been developed and validated to specifically assess quality of life

following deep venous thrombosis. However, no questionnaires exist to measure quality of

life following pulmonary embolism. Therefore, we created the PEmb-QoL questionnaire,

by determining relevant social, physical and emotional complaints by interviewing 10

patients with severe lingering symptoms following pulmonary embolism. Subsequent to

the creation, the PEmb-QoL questionnaire was distributed together with the Short Form-36

(SF-36) questionnaire twice among 90 consecutive subjects with a history of objectively

confirmed acute PE. Internal consistency was adequate (Cronbach’s � 0.62 to 0.94), as well

as test-retest reliability (intra-class correlation coefficients: 0.78 to 0.94). Furthermore,

correlation between the PEmb-QoL questionnaire and the SF-36 questionnaire supported

convergent validity. These data indicate that the PEmb-QoL questionnaire is a reliable

instrument to specifically assess QoL following PE, which is helpful in the identification of

patients with decreased QoL following acute PE.

Part II: identification of new thrombophilic factors

Chapter 7 reports a case-control study among 188 patients with confirmed venous

thrombosis (including calf vein thrombosis and superficial thrombophlebitis) and 370

controls with suspected venous thrombosis but in whom the diagnosis was excluded.

Glucose levels were measured upon presentation and divided into quartiles based on the

distribution in the controls and binary regression analyses were performed to assess

whether high glucose levels were associated with venous thrombosis. When adjusted for

Summary

154

body mass index, a known history of diabetes mellitus, age, sex, ethnicity and whether

known risk factors for deep venous thrombosis were present, the OR for deep venous

thrombosis in the 2nd, 3rd and 4th quartile of glucose levels compared to the 1st quartile was

1.59 (95% CI 0.89 to 2.85), 2.04 (95 % CI 1.15 to 3.62) and 2.21 (95% CI 1.20 to 4.05),

respectively, P for trend=0.001. These data support the assumption that increased glucose

levels measured at presentation are associated with venous thrombosis, but a causal role

needs to be investigated in future studies.

Chapter 8 elaborates on the association between hyperglycaemia and VTE. A post-hoc

analysis of four phase III clinical trials in 6890 patients undergoing elective total hip and

5493 patients undergoing total knee replacement was performed. Glucose levels were

measured at day 0 (admission) and day 1 (after surgery) and categorized into quartiles,

based on the distribution in the respective cohorts. Glucose levels measured at day 1 were

associated with both symptomatic VTE and total VTE (the composite of symptomatic VTE,

asymptomatic deep venous thrombosis and all cause mortality) in patients undergoing hip

surgery, adjusted OR highest versus lowest quartile 2.6 (95%CI 1.0 to 6.6) and 2.0 (95%CI

1.3 to 3.1), respectively. However, no association between hyperglycaemia and VTE

following knee replacement was observed, which is likely due to the surgical procedure.

Chapter 9 describes the results of two case-control studies (the EPIC-Norfolk study and the

ACT study) in which an association between 5 tag single nucleotide polymorphisms in the

endothelial lipase gene (LIPG) and lipid parameters, coronary artery disease and VTE was

investigated. In EPIC-Norfolk, we found that the minor allele of one SNP, rs2000813

(p.T111I), was associated with moderately higher HDL-C and apolipoprotein A-I levels,

higher HDL particle number and larger HDL size. No variants were associated with risk of

coronary artery disease, but 3 variants were associated with DVT risk OR 0.60 (95%CI 0.43

to 0.84); OR 2.04 (95%CI 1.4 to 2.98) and OR 1.67 (95%CI 1.18 to 2.38) per minor allele for

rs2000813, rs6507931 and rs2097055 respectively, p<0.005 for each single nucleotide

polymorphism. However, the association between LIPG single nucleotide polymorphisms

and DVT risk could not be replicated in the ACT study. In conclusion, these data support a

modest association between the LIPG rs2000813 variant and parameters of HDL

metabolism, but no association between common genetic variants in LIPG and coronary

artery disease or DVT.

Summary

155

Part III: reproductive aspects of VTE and thrombophilia

Chapter 10 reports the results of a cohort study that investigated whether the use of

therapeutic doses of low-molecular-weight heparin (LMWH) during pregnancy is

associated with an increased risk of post-partum haemorrhage (defined as blood loss

exceeding 500mL following vaginal delivery or over 1000mL following caesarean section).

We compared the bleeding risk in 83 pregnant women who received therapeutic doses of

low-molecular-weight heparin to the risk in 523 pregnant women who did not receive

anticoagulant treatment. The risk of post partum haemorrhage after vaginal delivery was

12% in LMWH users and 21% in non-users (RR 0.6; 95%CI 0.3 to 1.2). After caesarean

section, PPH risk was 9% (2/22) in LMWH users and 4% (2/51) in non-users (RR 2.3; 0.3 to

17). Median amount of blood loss in vaginal deliveries was 200mL in LMWH users and

300mL in non-users, (difference 100mL; 41 to 159). In caesarean sections the median blood

loss did not differ between LMWH users and non-users (425 and 400mL respectively,

difference 25mL; -133 to 183). In emergency caesarean sections this was higher in LMWH

users than in non-users (450 and 200mL respectively, difference 250mL; 9 to 491). These

data suggest that therapeutic doses of LMWH are relative safe for pregnant women who

deliver in the hospital setting of optimal obstetric care. An increased risk of bleeding

appears to occur in the setting of emergency caesarean section, but more research is

needed.

Chapter 11 details the prognosis of a subsequent pregnancy in 176 women with

antiphospholipid syndrome (APS) and recurrent miscarriage and 517 women with

unexplained recurrent miscarriage. A total of 122/176 women with APS had a subsequent

live birth (69%) compared to 324/517 women with unexplained recurrent miscarriage

(63%); OR 1.3 (95%CI 0.9 to 1.9). No differences were found for birth weight, gestational

age and intra-uterine growth restriction between these two groups. Following stratification

for therapy, 53/67 of women with APS who had received aspirin and heparin during their

pregnancy had a live birth (79%), compared to 64/104 women with APS who received

aspirin only (62%); adjusted OR 2.7 (95%CI 1.3 to 5.8) and compared to 204/305 of women

with unexplained miscarriage who received no treatment (67%); adjusted OR 2.2 (95%CI

1.1 to 4.2). In conclusion, the live birth rate between women with recurrent miscarriage and

APS and women with unexplained recurrent miscarriage is comparable. In women with

APS, combined use of aspirin and heparin appears associated with a higher live birth rate as

Summary

156

compared to women with APS who are treated with aspirin only and as compared to

women with unexplained recurrent miscarriage without treatment.

Chapter 12 reports high sperm counts as a possible explanation for the high population

frequency of factor V Leiden. A pilot study in 19 men with factor V Leiden showed increased

sperm concentration and total sperm counts in the 19 factor V Leiden carriers as compared

to men from the general population. For confirmation, we performed a cohort study

among 37 factor V Leiden carriers and 921 non-carriers. Factor V Leiden carriers had higher

total sperm counts and total motile sperm counts than non-carriers: 236 x106 (95% CI 158

to 292 x106) versus 163 x106 (95% CI 147 to 178 x106) and 81 x106 (95% CI 54 to 105 x106)

versus 52 x106 (95% CI 48 to 57 x106), respectively. These results provide a possible

explanation for the high prevalence of factor V Leiden among Caucasians. To our

knowledge, this is the first study that indicates that an increased prevalence of a genotype is

controlled by increased sperm counts.

Summary

157

Samenvatting

Inleiding op dit proefschrift

Veneuze trombo-embolie (VTE) is een veelvoorkomende ziekte in Westerse landen, met een

jaarlijkse incidentie van 2-3 per 1000 inwoners. De uiting van het ziektebeeld varieert van

diep veneuze trombose van het been tot in potentie fatale longembolieën.

Alhoewel antistollingstherapie zeer effectief is, wordt het succes van de behandeling voor

een groot gedeelte beperkt door een bloedingsrisico van 15% per jaar (het risico op een

ernstige bloeding, zoals hersenbloeding of bloeding vanuit het spijsverteringskanaal, is

ongeveer 3%). Na het staken van antistollingstherapie is het risico op een nieuwe episode

van VTE 17-30% binnen 7-8 jaar. Bovendien ontwikkelt ongeveer de helft van de patiënten

met diep veneuze trombose klachten van het post-trombotisch syndroom in het been.

Het is al vele jaren bekend dat er exogene risicofactoren zijn die het risico op VTE verhogen,

zoals ernstig trauma, langdurige immobilisatie, chirurgie, gebruik van de anticonceptiepil

of hormonale substitutietherapie, zwangerschap, kraambed, kanker en chemotherapie.

Naast deze exogene risicofactoren zijn er ook endogene factoren die het risico op VTE

verhogen, deze worden ook wel trombofilie genoemd.

De naam trombofilie werd in 1965 geïntroduceerd door een Noorse arts, Olav Egeberg,

toen hij een familie beschreef met een sterk verhoogde neiging tot trombose als gevolg van

een tekort van het antistollingseiwit antitrombine. In 1982 en 1984 werd ontdekt dat een

tekort aan proteïne C en proteïne S ook een erfelijke aanleg voor VTE kunnen veroorzaken.

Deficiënties van de antistollingseiwitten antritrombine, proteïne C en proteïne S zijn

zeldzaam, zowel in de algemene populatie als in patiënten met VTE. Pas halverwege de

jaren ’90 van de vorige eeuw werden er frequent voorkomende risicofactoren voor VTE

ontdekt, zoals factor V Leiden en de protrombine G20210A mutatie.

Factor V Leiden is een puntmutatie die leidt tot weerstand van geactiveerd stollingsfactor V

via inactivatie door geactiveerd proteïne C. Dragers van factor V Leiden hebben behalve op

VTE ook een licht verhoogd risico op spontane of recidiverende miskramen. Ondanks deze

duidelijke nadelen komt de puntmutatie (die ongeveer 21.000-34.000 jaar geleden

ontstond) met een prevalentie van 4-7% nog vaak voor. Er is lang gespeculeerd over de

hoge prevalentie van factor V Leiden, die een uiting zou kunnen zijn van een zeker

evolutionair voordeel.

Trombofilie omvat ook verworven condities, waarvan het antifosfolipidensyndroom het

meest bekend is. Antifosfolipidensyndroom is een afwijking die leidt tot arteriële en/of

veneuze trombose, of tot zwangerschapscomplicaties zoals miskraam of vroeggeboorte.

Samenvatting

161

Het syndroom ontstaat als gevolg van productie van auto-antistoffen tegen eiwitten in het

plasma die gebonden zijn aan het oppervlak van negatief geladen fosfolipiden.

In ongeveer de helft van alle patiënten met een VTE kan een trombofiliedefect worden

aangetoond. Daarnaast is de episode in ongeveer 50% van de patiënten uitgelokt door één

of meer exogene risicofactoren. Dit impliceert dat in ongeveer een kwart van de patiënten

met VTE geen duidelijke oorzaak kan worden achterhaald. Het is aannemelijk dat hiervoor

tot nu toe onbekende risicofactoren verantwoordelijk zijn.

Het oorspronkelijke onderwerp van dit proefschrift omvatte de vraagstelling of het zinvol is

om patiënten met een eerste VTE te testen op trombofilie. Hiervoor hadden wij het

NOSTRADAMUS (Necessity Of Screening for ThRombophilia at Diagnosis of venous

thromboembolism to Assess Most Unresolved iSsues) onderzoek opgezet. Dit multi-

centrum gerandomiseerde onderzoek was de ideale manier om de eventuele voordelen van

testen op trombofilie in kaart te kunnen brengen. Dit onderzoek moest echter vroegtijdig

worden beëindigd door een te laag aantal deelnemende patiënten. Desalniettemin waren

wij in staat om een aantal onderzoeken uit te voeren die gericht waren op de consequenties

van testen op trombofilie en op de klinische en psychologische gevolgen van VTE.

Samenvatting

Hoofdstuk 1 bestaat uit een korte introductie over VTE en trombofilie. De overige

hoofdstukken zijn onderverdeeld in drie delen. Het eerste gedeelte (hoofdstuk 2 t/m 6)

behandelt klinische en psychologische aspecten van VTE en trombofilie, het tweede

gedeelte (hoofdstuk 7 t/m 9) bespreekt de zoektocht naar nieuwe trombofiliefactoren en

het derde gedeelte richt zich op voortplantingsgerelateerde aspecten van VTE en

trombofilie.

Deel I: klinische en psychologische aspecten van VTE en trombofilie

Hoofdstuk 2 geeft een overzicht van de verscheidene trombofiliefactoren (antitrombine

deficiëntie, proteïne C deficiëntie, proteïne S deficiëntie, factor V Leiden, de protrombine

G20210A mutatie, verhoogde concentraties van stollingsfactor VIII en het

antifosfolipidensyndroom) en hun associatie met VTE. Daarnaast worden de mogelijke

voor- en nadelen van testen op trombofilie besproken. Bovendien wordt er een overzicht

gepresenteerd van het NOSTRADAMUS onderzoek. Dit onderzoek, dat was opgezet om te

Samenvatting

162

achterhalen of testen op trombofilie en verlengde behandelduur zinvol is ten aanzien van

klinische eindpunten (de combinatie van recidief VTE en bloedingsrisico, kwaliteit van

leven en kosten), werd helaas vroegtijdig gestaakt. Dit hoofdstuk besluit met de

verscheidene oorzaken voor de vroegtijdige beëindiging van het NOSTRADAMUS

onderzoek.

In hoofdstuk 3 wordt het beschikbare bewijs om patiënten na een VTE te testen op

trombofilie bediscussieerd. We onderzochten of testen op trombofilie leidt tot een

verlaging van het risico om een recidief VTE te krijgen door de therapie aan te passen of

aanvullende maatregelen te nemen. Het onderzoeksregister van de “Cochrane Peripheral

Vascular Diseases Group” (meest recent geraadpleegd op 22 april 2008), MEDLINE,

EMBASE en de CENTRAL database (meest recent geraadpleegd in 2008, uitgave 2) en de

relevante bronvermeldingen werden geraadpleegd voor relevante artikelen. Er bleken

echter geen relevante artikelen over dit onderwerp te zijn gepubliceerd. De conclusie van dit

hoofdstuk luidt dan ook dat er dringend behoefte is aan gerandomiseerd controleerde

klinische onderzoeken om meer duidelijkheid over dit onderwerp te creëren.

Hoofdstuk 4 onderzoekt de psychologische consequenties van testen op trombofilie, zoals

angst, depressie en ongerustheid, wat een mogelijk nadeel van testen op trombofilie zou

kunnen zijn. Er wordt een systematisch overzicht gegeven van onderzoeken die de aard en

de mate van psychologische gevolgen van testen op trombofilie in kaart brachten. We

raadpleegden de MEDLINE database (vanaf 1966 tot februari 2008), de EMBASE database

(vanaf 1985 tot 2008, week 5) en de PsychInfo database (vanaf 1806 tot februari 2008) om te

zoeken naar relevante artikelen, zonder ons te beperken tot de taal waarin het artikel

verschenen was. Bovendien onderzochten we de referentielijsten van relevante artikelen om

additionele artikelen op het spoor te komen. Zes onderzoeken voldeden aan de

selectiecriteria, maar het bleek niet mogelijk om de uitkomsten van deze onderzoeken

samen te voegen aangezien ze onderling aanzienlijk verschilden op het gebied van de

toegepaste methoden. In deze zes onderzoeken werden weinig nadelige gevolgen

gerapporteerd, maar de meeste metingen vonden alleen op korte termijn plaats of misten

methodologische precisie. Hierdoor kan er, op basis van de beschikbare literatuur, geen

duidelijk oordeel worden gegeven over de psychologische gevolgen van testen op

trombofilie.

Samenvatting

163

De hoofdstukken 5 en 6 behandelen de ontwikkeling en de validatie van de Pulmonary

Embolism Quality of Life (PEmb-QoL) vragenlijst. Kwaliteit van leven speelt tegenwoordig

een centrale rol in de gezondheidszorg en is een belangrijke uitkomstmaat in klinisch

onderzoek. Er zijn in het verleden meerdere vragenlijsten ontwikkeld en gevalideerd die

specifiek kwaliteit van leven na een trombosebeen meten. Echter, er bestaat nog geen

vragenlijst die specifiek de kwaliteit van leven meet na een longembolie. We hebben

daarom de PEmb-QoL ontwikkeld door de relevante sociale, lichamelijke en emotionele

klachten na een longembolie in kaart te brengen bij een tiental longemboliepatiënten met

ernstige restverschijnselen. Na de ontwikkeling van de PEmb-QoL vragenlijst, werd hij

tweemaal, samen met de Short Form-36 (SF-36) vragenlijst, ter validatie rondgestuurd naar

90 patiënten bij wie in het verleden objectief een longembolie was vastgesteld. De interne

consistentie was adequaat (Cronbach’s � 0,62 tot 0,94), net als de test-hertest

betrouwbaarheid (intra-class correlatiecoëficiënt: 0,78 tot 0,94). Bovendien werd

convergente validiteit ondersteund door de correlatie tussen de PEmb-QoL vragenlijst en de

SF-36 vragenlijst. Deze data tonen aan dat de PEmb-QoL vragenlijst een betrouwbaar

instrument is om specifiek de kwaliteit van leven na een longembolie te meten en van pas

kan komen bij het opsporen van patiënten met een verminderde kwaliteit van leven na een

acute longembolie.

Deel II: de zoektocht naar nieuwe trombofilie factoren

Hoofdstuk 7 beschrijft de resultaten van een patiënt-controle onderzoek onder 188

patiënten met een bewezen veneuze trombose (inclusief kuitvene trombose en

oppervlakkige tromboflebitis) en 370 controles met verdenking op veneuze trombose,

maar bij wie deze diagnose was verworpen. Op het moment van presentatie werd bij alle

deelnemers het glucosegehalte gemeten en verdeeld in kwartielen (gebaseerd op de

verdeling in de controlegroep). Om te onderzoeken of een hoge glucoseconcentratie

geassocieerd was met veneuze trombose werd een binaire regressieanalyse uitgevoerd. Na

correctie voor body mass index, bestaande diabetes, leeftijd, geslacht, etniciteit en of de

veneuze trombose was uitgelokt was de odds ratio voor diep veneuze trombose bij een

glucosegehalte in het 2e, 3e en 4e kwartiel ten opzichte van het eerste kwartiel: 1,59 (95%

betrouwbaarheidsinterval (BI) 0,89 tot 2,85), 2,04 (95%BI 1,15 tot 3,62) en 2,21 (95%BI

1,20 tot 4,05), respectievelijk (p voor trend=0,001). Deze data ondersteunen de aanname dat

een toegenomen glucosegehalte ten tijde van presentatie geassocieerd is met veneuze

trombose. Of dit verband causaal is, dient nader te worden onderzocht.

Samenvatting

164

Hoofdstuk 8 gaat dieper in op de associatie tussen hyperglycemie en VTE. Wij voerden een

een post-hoc analyse uit van vier fase III klinische onderzoeken in 6890 patiënten die een

geplande heupvervanging ondergingen en 5493 patiënten die een geplande knievervanging

ondergingen. Het glucosegehalte werd bepaald op dag 0 (opname) en dag 1 (na de

operatie) en werd verdeeld in kwartielen, gebaseerd op de distributie in het betreffende

cohort. Glucoseconcentratie op dag 1 was geassocieerd met zowel symptomatische VTE als

“totale” VTE (de combinatie tussen symptomatische VTE, asymptomatische diep veneuze

trombose en sterfte) bij patiënten die een totale heup vervanging ondergingen,

gecorrigeerde odds ratio van het hoogste versus het laagste kwartiel: 2,6 (95%BI 1,0 tot 6,6)

en 2,0 (95%BI 1,3 tot 3,1), respectievelijk. Er werd echter geen associatie waargenomen

tussen hyperglycemie en VTE na totale knievervanging, wat waarschijnlijk een gevolg is van

de toegepaste operatietechniek.

Hoofdstuk 9 beschrijft de resultaten van twee patiënt-controle onderzoeken (het EPIC-

Norfolk onderzoek en het ACT onderzoek) waarin een associatie werd onderzocht tussen

vijf tag single nucleotide polymorphisms (SNP=enkel-nucleotide polymorfismen) in het

gen dat codeert voor endotheel lipase en lipiden parameters, kransslagaderlijden en diep

veneuze trombose. In het EPIC-Norfolk cohort vonden we dat het minder vaak

voorkomende allel van één van de SNPs, rs2000813 (p.T111I), geassocieerd was met matig

verhoogde concentraties van het HDL-cholesterol en apolipolipoproteïne A-1 en een groter

aantal en grotere omvang van de HDL-partikels. Geen van de SNPs was geassocieerd met

het risico op kransslagader lijden, maar drie varianten waren wel geassocieerd met het

risico op diep veneuze trombose, odds ratio 0,60 (95%BI 0,43 tot 0,84); odds ratio 2,04

(95%BI 1,40 tot 2,98) en odds ratio 1,67 (95%BI 1,18 tot 2,38) per minder vaak voorkomend

allel van respectievelijk rs2000813, rs6507931 en rs2097055, p<0,005 voor elke SNP.

Echter, de associatie tussen deze SNPs en VTE werd niet teruggevonden in het ACT

onderzoek. Deze bevindingen ondersteunen een matige associatie tussen de rs2000813

variant van het endotheliaal lipase gen en parameters van het HDL-metabolisme, maar niet

een associatie tussen veel voorkomende genetische varianten in het endotheel lipase gen en

kransslagader lijden of diep veneuze trombose.

Deel III: voortplantingsgerelateerde aspecten van VTE en trombofilie

In hoofdstuk 10 worden de resultaten weergegeven van een cohort onderzoek waarin werd

onderzocht of gebruik van therapeutische doseringen laagmoleculair gewichtsheparine

Samenvatting

165

(LMWH) tijdens de zwangerschap gepaard ging met een verhoogd risico op bloeding

rondom de bevalling (gedefinieerd als bloedverlies van meer dan 500ml na een vaginale

bevalling of meer dan 1000ml na een keizersnede). We vergeleken het bloedingsrisico van

83 zwangere vrouwen die therapeutische doseringen LMWH kregen met het

bloedingsrisico van 523 vrouwen die geen antistollingstherapie ontvingen. Het risico op

een bloeding na vaginale bevalling was 12% (7/61) bij vrouwen die LMWH gebruikten en

21% (97/472) bij vrouwen die geen LMWH gebruikten (relatief risico 0,6; 95%BI 0,3 tot

1,2). Na een keizersnede was het risico op een bloeding 9% (2/22) bij vrouwen die LMWH

gebruikten en 4% (2/51) bij vrouwen die geen LMWH ontvingen (relatief risico 2,3; 95%BI

0,3 tot 17). De mediane hoeveelheid bloedverlies na vaginale bevallingen was 200ml bij

LMWH gebruiksters en 300ml bij de vrouwen die geen LMWH kregen (verschil 100ml;

95%BI 41ml tot 159ml). Het bloedverlies na een keizersnede verschilde niet tussen beide

groepen (425ml en 400ml respectievelijk, verschil 25ml; 95%BI -133ml tot 183ml). Het

bloedverlies na een spoedkeizersnede was hoger bij vrouwen die LMWH gebruikten ten

opzichte van vrouwen die dat niet deden (450ml en 200ml respectievelijk, verschil 250ml;

95%BI 9ml tot 491ml). Deze resultaten suggereren dat therapeutische doseringen LMWH

relatief veilig zijn toe te dienen aan zwangere vrouwen die bevallen in een ziekenhuis met

optimale verloskundige zorg. Er lijkt een toegenomen bloedingsrisico te zijn na een

spoedkeizersnede, maar er is meer onderzoek nodig om dat te bevestigen.

Hoofdstuk 11 richt zich op de prognose van een volgende zwangerschap bij 176 vrouwen

met recidiverende miskramen en antifosfolipidensyndroom en bij 517 vrouwen met

onverklaarbare recidiverende miskramen. In totaal hadden 122/176 vrouwen met

antifosfolipidensyndroom een levend geboren kind (69%) ten opzichte van 324/517

vrouwen met onverklaarbare recidiverende miskramen (63%); odds ratio 1,3 (95%BI 0,9 tot

1,9). Er was geen verschil in geboortegewicht, zwangerschapsduur en intra-uteriene groei

vertraging tussen beide groepen. Na stratificatie voor therapie bleek dat 79% van de

vrouwen met antifosfolipidensyndroom die tijdens hun zwangerschap een combinatie van

heparine en aspirine ontvingen, een levend geboren kind hadden, ten opzichte van 62% van

de vrouwen met antifosfolipidensyndroom die alleen aspirine ontvingen, gecorrigeerde

odds ratio 2,7 (95%BI 1,3 tot 5,8) en ten opzichte van 67% van de vrouwen met

onverklaarde recidiverende miskramen, gecorrigeerde odds ratio 2,2 (95%BI 1,1 tot 4,2).

Kortom, een succesvolle uitkomst van de zwangerschap is vergelijkbaar tussen vrouwen

met antifosfolipiden syndroom en recidiverende miskramen en vrouwen met

Samenvatting

166

onverklaarbare recidiverende miskramen. Combinatietherapie van heparine en aspirine

lijkt geassocieerd te zijn met een hoger aantal levend geboorten bij vrouwen met

antifosfolipidensyndroom ten opzichte van vrouwen met antifosfolipidensyndroom die

uitsluitend heparine ontvingen en ten opzichte van vrouwen met onverklaarbare

recidiverende miskramen zonder therapie.

In hoofdstuk 12 wordt een mogelijke verklaring gegeven voor de hoge prevalentie onder de

Kaukasische bevolking van factor V Leiden. Een pilot onderzoek onder 19 mannen met

factor V Leiden toonde een verhoogde spermaconcentratie en een verhoogd aantal

spermacellen aan ten opzichte van mannen uit de algemene bevolking. Ter bevestiging van

deze bevinding voerden we een cohort onderzoek uit onder 37 mannen met factor V Leiden

en 921 mannen zonder factor V Leiden. Dragers van factor V Leiden hadden een hoger

aantal spermacellen en een hoger aantal beweeglijke spermacellen ten opzichte van niet-

dragers: 236 x106 (95%BI 158 tot 292 x106) versus 163 x106 (95%BI 147 tot 178 x 106) en

respectievelijk 81 x106 (95%BI 54 tot 105 x106) versus 52 x106 (95%BI 48 tot 57 x106). Deze

resultaten leveren een mogelijke verklaring voor de hoge prevalentie van factor V Leiden

onder Kaukasiërs. Voor zover wij weten is dit het eerste onderzoek dat er op duidt dat een

verhoogde prevalentie in stand wordt gehouden door een verhoogd aantal spermacellen.

Samenvatting

167

Dankwoord

Promotieonderzoek doe je niet alleen. Deze stelling mag misschien clichématig

overkomen, maar zij wordt alleen al ondersteund door het aantal mede-auteurs (n=39) van

de hoofdstukken in dit proefschrift. Graag wil ik dan ook iedereen die direct of indirect

betrokken is geweest bij de totstandkoming van dit proefschrift heel hartelijk bedanken:

De leden van de promotiecommissie

Dr. S. Middeldorp, co-promotor en dagelijks begeleider “op afstand”. Beste Saskia, al vanaf

het moment dat ik in 2006 bij jou op sollicitatiegesprek kwam, was ik ervan overtuigd dat

promoveren onder jouw supervisie een kans was die ik niet kon laten liggen. Het kostte me

daarom ook enige moeite om mijn teleurstelling te verhullen, toen jij een week voor mijn

eerste werkdag aangekondigde dat je niet langer in het AMC, maar voortaan in het LUMC

werkzaam zou zijn. Mijn reactie: “o, wat leuk voor je, nou gefeliciteerd hoor” zal dan ook

niet erg overtuigend zijn overgekomen. Echter, dankzij jouw ongekende

organisatievermogen (emails werden bijna altijd dezelfde dag nog beantwoord,

manuscripten kreeg ik meestal binnen een week uitvoerig gecorrigeerd terug), heeft jouw

vertrek geen moment tot vertraging van mijn promotie geleid. Bedankt voor de geweldige

samenwerking en ik hoop oprecht dat, met de totstandkoming van dit proefschrift, onze

samenwerking niet is beëindigd!

Dr. P.W. Kamphuisen, co-promotor en dagelijks begeleider van dichtbij. Beste PW, je

promovendi komen bij jou altijd op de eerste plaats. Je bent altijd beschikbaar voor overleg.

Daarnaast verbreek jij alle records met het beoordelen of corrigeren van manuscripten

(soms wel binnen 1 uur). Ten slotte, je ongekende relativeringsvermogen zorgt er voor, dat

je na een tegenslag de onderzoeksdraad toch weer wilt oppakken.

Professor dr. H.R. Büller, promotor. Waarde Harry, alhoewel je vaker in het buitenland was

dan in het AMC, was je in alle fasen van mijn promotie betrokken. Als ik je tijdens onze

voortgangsgesprekken of in de wandelgangen wilde bijpraten over een van mijn lopende

onderzoeksprojecten, had jij altijd aan een half woord voldoende om niet alleen het hele

onderzoek te overzien, maar om bovendien met nieuwe suggesties te komen die we nog

niet eerder hadden overwogen. Je bent een geweldige leermeester op wetenschappelijk

gebied, maar ook van de manier waarop jij anderen weet te inspireren heb ik bijzonder veel

geleerd.

Dankwoord

171

Professor dr. M.M. Levi, opponent. Beste Marcel, aangezien jij degene bent geweest die mij

met mijn (co-)promotores in contact heeft gebracht, heb ik deze promotie zeker ook aan

jou te danken. Onze samenwerking op onderzoeksgebied was weliswaar niet heel groot,

maar des te meer heb ik van je geleerd als clinicus. Ik kan wel begrijpen waarom jij

bovenaan de verkiezing eindigde naar welke internist je je eigen moeder zou doorverwijzen!

Ik kijk er dan ook naar uit om als internist in opleiding veel van je te mogen leren!

Professor dr. F.R. Rosendaal, opponent. Beste Frits, graag wil ik je heel hartelijk bedanken

voor de gelegenheid om als gastmedewerker in de op de afdeling Klinische Epidemiologie

in Leiden te mogen werken. Ik heb tijdens de verscheidene cursussen bijzonder veel van je

geleerd. Het is een eer dat je op mijn promotie wilt opponeren.

Professor dr. J.C.M. Meijers, opponent. Beste Joost, het was een groot plezier om met jou

samen te mogen werken (en tegen bovengenoemde opponent te klaverjassen). Je kritische

commentaar tijdens de journal club en je bereidwilligheid om altijd in te springen wanneer

er geen docent was voor het vasculaire onderwijs op donderdagmiddag, hebben grote

indruk op me gemaakt.

Dr. V.E. Gerdes, opponent. Beste Victor, ook van jouw visie tijdens menig

patiëntbespreking of congres heb ik veel geleerd. Bedankt dat je zitting wilt nemen in mijn

promotiecommissie.

R.G. Farquharson, MD FRCOG, opponent. Dear Roy, thank you very much for the

opportunity to cooperate in the LIVERPOOL project. Learning how early pregnancy is

managed in a tertiary referral centre in the UK has been a great experience, as well as

attending an Everton’s home game.

Prof. dr. M.J. Heineman, opponent. Geachte professor Heineman, na mijn voordracht op

het ochtendrapport van de afdeling gynaecologie enige tijd geleden waren uw vragen een

ware voorproef van de oppositie die nu op handen is. Graag wil ik u heel hartelijk danken

dat u bereid bent om in deze promotiecommissie plaats te nemen.

Dankwoord

172

Mijn collega’s van de afdeling Vasculaire Geneeskunde

Het verrichten van een promotieonderzoek was voor mij een buitengewoon leerzame

ervaring. Alhoewel ik de gedwongen vroegtijdige beëindiging van het NOSTRADAMUS

onderzoek als een dieptepunt heb beschouwd, kijk ik toch vooral terug op de vele

hoogtepunten tijdens mijn promotietijd. Wat het werken de afdeling Vasculaire

Geneeskunde bij uitstek zo plezierig heeft gemaakt, is volgens mij de uitstekende werksfeer

die er heerst. Deze sfeer is hoogstwaarschijnlijk voortgekomen uit het door Harry Büller zo

vaak gelauwerde anarchisme dat een “conditio sine qua non” zou zijn voor het verrichten

van vruchtbaar wetenschappelijk onderzoek. Ik wil dan ook iedereen van onze afdeling

bedanken voor de plezierige sfeer en fijne samenwerking! In de eerste plaats: mijn (ex-)F4-

139 collega’s Olav de Peuter, Hans Avis, Raphaël Duivenvoorden, Michiel Coppens en

Meeike Kusters. En verder (in willekeurige volgorde) Joyce Jansen, Henriëtte Bekkers-den

Haan, Remco Franssen, Menno Vergeer, Karim El-Harchaoui, Corien Eckhardt, Reneé

Douma, Bregje van Zaane, Bas van den Bogaard, Nanne Frenkel, Suthesh Sivapalaratnam,

Lysette Broekhuizen, Onno Holleboom, Katrijn Rensing, Sara Rafi, Aart Strang, Carlijne

Hassing, Anne Reuwer, Steffie Maiwald, Fatima Akdim, Geerte van Sluis, Brigitte

Sondermeijer, Diederik van Wijk, Hans Mooij, Ankie Kleinjan, Nadine Reyn-Gibson, Ester

Löwenberg, Andrea Bochem, Roeland Huijgen, Frederiek van Doormaal, Ward van Beers,

Federico Lussana, Annemieke Bakker, Elise Eerenberg, Dirk-Jan Enkelaar, Lily Jakulj,

Maartje Visser, Erik Stroes, John Kastelein, Bert-Jan van den Born, Iris Wichers, Marjin

Meuwese, Sander van Leuven, Wim van der Steeg, Anouk van der Graaf, Maud Vissers,

Geesje Dallinga-Thie, Marcel Twickler, Lizzy Brewster, Gert van Montfrans, Mieke Trip,

Houshang Monajemi, Marije ten Wolde, Sanne van Wissen, Kees Hovingh, Max

Nieuwdorp, Debbie Visser-Bus, Agnes Vree en Margreet Ruiter, Paul van der Valk en de

medewerkers van het trial bureau, in het bijzonder Belia Rekké.

De studenten die betrokken waren bij één of meerdere projecten

Alexander Stehouwer, Linde Busweiler, Elise Nelis, Hester Prins, Sofie Hamers en Juliëtte

van Baal, bedankt voor jullie geweldige ondersteuning. Het was een plezier om met jullie

samen te werken.

De overige collega’s met wie ik direct heb samengewerkt

Jeroen Hermanides, Hans de Vries, Erik Klok, Menno Huisman, Sjoerd Repping, Sara

Roshani, Joris van der Post, Hans Wolf, Mariëtte Goddijn, Fleur Vansenne, Corianne de

Dankwoord

173

Borgie, Ad Kaptein, Margreet Scharloo, Joke Korevaar en Ton Lensing, bedankt voor de

samenwerking en de tijd die jullie in ons onderzoek hebben geïnvesteerd.

Mijn paranimfen

Beste Daniël en Jonathan, wie had tien jaar geleden gedacht dat, toen onze paden rond de

collegezalen c.q. de practica elkaar kruisten, wij zo een hechte vriendschap zouden

opbouwen. Ik ben vereerd dat jullie mijn paranimfen willen zijn en ik vind het een

geruststellend idee dat ik alle vragen van de opponenten naar jullie door kan spelen, als ik

zelf het antwoord niet weet!

Mijn familieleden

Lieve mamma en pappa, oma en opa, te allen tijde waren jullie geïnteresseerd en betrokken

bij mijn opleiding, al vóór mijn allereerste schooldag. Dat ik nu in de gelegenheid ben om

(hopelijk) tot doctor in de geneeskunde te promoveren, is vooral ook te danken aan jullie

onvolprezen steun. Het is een zegen dat jullie allen van mijn promotie getuigen mogen zijn!

Verder gaan mijn gedachten natuurlijk ook uit naar oma Betty ( �''� ), opa Paul ( �''� ) en in het

bijzonder ook oma Gerda ( �''� ), die nog wel getuige is geweest van het begin van mijn

promotietijd. Jeffrey en Lisa, wat een enorm hechte band hebben wij samen. Jullie beseffen

niet hoe fijn ik het vind, dat jullie bij mijn promotieplechtigheid aanwezig zijn.

Tenslotte, last but not least, mijn lieve Petra, een lievere vrouw had ik me niet kunnen

wensen. Elke dag met jou samen zijn is voor mij een onophoudelijk feest!

Dankwoord

174

Authors’ Affiliations

S. Matthijs Boekholdt Department of Cardiology, Academic Medical Centre, Amsterdam, the Netherlands Corianne A.J.M. de Borgie Department of Clinical Epidemiology and Biostatistics and Bioinformatics, Academic Medical Centre, Amsterdam, the Netherlands Harry R. Büller Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Linde A. Busweiler Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Danny M. Cohn Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Geesje M. Dallinga-Thie Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Feroza Dawood Department of Obstetrics and Gynaecology, Liverpool Women’s Hospital, Liverpool, United Kingdom Roy G. Farquharson Department of Obstetrics and Gynaecology, Liverpool Women’s Hospital, Liverpool, United Kingdom Mariëtte Goddijn Department of Obstetrics and Gynecology, Academic Medical Centre, Amsterdam, the Netherlands Sofie Hamers Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Jeroen Hermanides Department of Internal Medicine, Academic Medical Centre, Amsterdam, the Netherlands Joost B.L. Hoekstra Department of Internal Medicine, Academic Medical Centre, Amsterdam, the Netherlands Martin Homering BayerHealthCare AG, Wuppertal, Germany

177

Authors' Affiliations

Roeland Huijgen Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Menno V. Huisman Section of Vascular Medicine, Department of General Internal Medicine-Endocrinology, Leiden University Medical Centre, Leiden the Netherlands Pieter-Willem Kamphuisen Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Ad A. Kaptein Unit of Psychology, Leiden University Medical Centre, Leiden, the Netherlands John J.P. Kastelein Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Kay-Tee Khaw Department of Public Health and Primary Care, Institute of Public Health, University of Cambridge, Cambridge, United Kingdom Frederikus A. Klok Section of Vascular Medicine, Department of General Internal Medicine-Endocrinology, Leiden University Medical Centre, Leiden the Netherlands Joke C. Korevaar Department of Clinical Epidemiology and Biostatistics and Bioinformatics, Academic Medical Centre, Amsterdam, the Netherlands Klaas W. van Kralingen Department of Pulmonary Medicine, Leiden University Medical Centre, Leiden, the Netherlands Sylvia Kuhls BayerHealthCare AG, Wuppertal, Germany Anthonie W.A. Lensing Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands BayerHealthCare AG, Wuppertal, Germany Joost C.M. Meijers Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands

178

Authors' Affiliations

Saskia MMiddeldorp Department of Clinical Epidemiology and Department of General Internal Medicine, Leiden University Medical Centre, Leiden, the Netherlands Elise A. Nelis Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Joris A.M. van der Post Department of Obstetrics and Gynecology, Academic Medical Centre, Amsterdam, the Netherlands Hester Prins Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Sjoerd Repping Centre for Reproductive Medicine, Department of Obstetrics and Gynaecology, Academic Medical Centre, Amsterdam, the Netherlands Sally L. Ricketts Department of Public Health and Primary Care, Institute of Public Health, University of Cambridge, Cambridge, United Kingdom Sara RRoshani Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Department of Clinical Epidemiology, Leiden University Medical Centre, Leiden, the Netherlands Majinder S. Sandhu Department of Public Health and Primary Care, Institute of Public Health, University of Cambridge, Cambridge, United Kingdom Margreet Scharloo Unit of Psychology, Leiden University Medical Centre, Leiden, the Netherlands Alexander C. Stehouwer Department of Vascular Medicine, Academic Medical Centre, Amsterdam, the Netherlands Fleur VVansenne Department of Clinical Epidemiology and Biostatistics and Bioinformatics, Academic Medical Centre, Amsterdam, the Netherlands A. Menno H. Vergeer Department of Internal Medicine, Academic Medical Centre, Amsterdam, the Netherlands

179

Authors' Affiliations

J. Hans de Vries Department of Internal Medicine, Academic Medical Centre, Amsterdam, the Netherlands Nicholas J. Wareham Medical Research Council, Epidemiology Unit, Cambridge, United Kingdom Hans WWolf Department of Obstetrics and Gynecology, Academic Medical Centre, Amsterdam, the Netherlands

180

Authors' Affiliations

List of publications and

curriculum vitae

1. Klok FA*, Cohn DM*, Middeldorp S, Scharloo M, Büller HR, Van Kralingen KW, Kaptein AA, Huisman MV. Quality of life after pulmonary embolism: validation of the PEmb-QoL questionnaire. JJournal of Thrombosis and Haemostasis 2010 in press

2. Cohn DM, Repping S, Büller HR, Meijers JCM, Middeldorp S. Increased sperm count may account for high population frequency of factor V Leiden. JJournal of Thrombosis and Haemostasis 2010 in press

3. Cohn DM, Vansenne F, De Borgie CAJM, Middeldorp S. Thrombophilia testing for prevention of recurrent venous thromboembolism. CCochrane Database of Systematic Reviews 2009 CD007069.

4. Hermanides J, Cohn DM*, Devries JH, Kamphuisen PW, Huijgen R, Meijers JCM, Hoekstra JBL, Büller HR. Venous thrombosis is associated with hyperglycemia at diagnosis: a case-control study. JJournal of Thrombosis and Haemostasis 2009;7:945-949.

5. Cohn DM, Nelis EA, Busweiler LA, Kaptein AA, Middeldorp S. Quality of life after pulmonary embolism: the development of the PEmb-QoL questionnaire. JJournal of Thrombosis and Haemostasis 2009;7:1044-1046.

6. Cohn DM, Vansenne F, Kaptein AA, De Borgie CA, Middeldorp S. The psychological impact of testing for thrombophilia: a systematic review. JJournal of Thrombosis and Haemostasis 2008;6:1099-1104.

7. Cohn DM, Middeldorp S. Vroegtijdige beëindiging van het onderzoek naar het nut van trombofilietests bij een eerste veneuze trombo-embolie: het NOSTRADAMUS-onderzoek. NNederlands Tijdschrift voor Geneeskunde 2008;152:2093-2094.

8. Cohn DM*, Roshani S*, Middeldorp S. Thrombophilia and venous thromboembolism: implications for testing. SSeminars in Thrombosis and Hemostasis 2007;33:573-581.

9. Cohn DM, Middeldorp S. Trombofiliescreening bij VTE: het NOSTRADAMUS onderzoek. NNederlands Tijdschrift voor Hematologie 2007;4:144-145.

10. Cohn DM, Middeldorp S. Een multicentrisch gerandomiseerd klinisch onderzoek naar het nut van trombofilietests bij een eerste veneuze trombo-embolie: het NOSTRADAMUS-onderzoek. NNederlands Tijdschrift voor Geneeskunde 2007;151:371-373.

11. Zelinkova Z, Derijks LJ, Stokkers PC, Vogels EW, van Kampen AH, Curvers WL, Cohn DM, van Deventer SJ, Hommes DW. Inosine triphosphate pyrophosphatase and thiopurine s-methyltransferase genotypes relationship to azathioprine-induced myelosuppression. CClinical Gastroenterology and Hepatology 2006;4:44-49.

12. Braat H, Stokkers PC, Hommes T, Cohn DM, Vogels EW, Pronk I, Spek AL, van Kampen AH, van Deventer SJ, Pepeppenlenbosch MP, Hommes DW. Consequence of functional Nod2 and Tlr4 mutations on gene transcription in Crohn's disease patients. Journal of Molecular Medicine 2005;83:601-609.

* Equal contribution

List of publications

183

Danny Cohn, zoon van Frank en Yvonne Cohn en broer van Jeffrey Cohn, werd op 28 april

1980 geboren in ’s-Gravenhage. Na het doorlopen van de lagere school (Wolters, ’s-

Gravenhage) en het gymnasium (Vrijzinnig Christelijk Lyceum, ’s-Gravenhage) slaagde hij

“cum laude” voor zijn eindexamen in acht vakken. Tijdens deze periode speelde hij in

diverse theater- en televisieproducties, waaronder Macbeth en Wachten op Godot met Het

Nationaal Toneel. In 1999 begon hij aan de studie geneeskunde aan de Universiteit van

Amsterdam. In 2003 behaalde hij zijn doctoraalexamen en in 2006 “cum laude” zijn

artsexamen. Tijdens zijn studie begeleidde hij practica fysiologie als student-assistent en

was hij bestuurschauffeur bij AFC Ajax.

Aansluitend aan het behalen van zijn artsexamen startte hij met zijn promotieonderzoek

aan de afdeling Vasculaire Geneeskunde van het Academisch Medisch Centrum, waarvan

dit proefschrift het directe resultaat is. Gedurende zijn promotieonderzoek was hij actief in

het bestuur van APROVE, de promovendivereniging van het AMC en gaf hij onderwijs over

bloedstolling op de Hogeschool van Amsterdam. Per 1 april 2010 zal hij in het AMC

beginnen aan zijn opleiding tot internist. Danny is in 2009 getrouwd met Petra Cohn-

Hokke.

Curriculum Vitae

184