Progressive Dilatation of the Main Pulmonary Artery Is a Characteristic of Pulmonary Arterial...

Insights into the progression of right ventricular failure in pulmonary arterial hypertension

Gert-Jan Mauritz

insights into the progression of right ventricular failure in pulmonary arterial hypertension

VRIJE UNIVERSITEIT

Insights into the progression of right ventricular failure

in pulmonary arterial hypertension

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aande Vrije Universiteit Amsterdam,op gezag van de rector magnificus

prof.dr. L.M. Bouter,in het openbaar te verdedigen

ten overstaan van de promotiecommissievan de faculteit der Geneeskunde

op maandag 19 december 2011 om 15.45 uurin de aula van de universiteit,

De Boelelaan 1105

door

Gerrit-Jan Mauritz

geboren te Rijssen

promotor prof.dr. A. Vonk Noordegraaf

copromotor dr. J.T. Marcus

Aan mijn ouders

Promotiecommissie prof.dr. A.C. van Rossum (VU medisch centrum, Amsterdam) prof.dr. W.J. Paulus (VU medisch centrum, Amsterdam) prof.dr. A. de Roos (Leids Universitair Medisch Centrum, Leiden) prof.dr. R.M.F. Berger (Universitair Medisch Centrum Groningen, Groningen) prof.dr. A. Torbicki (National Tuberculosis and Lung Disease Research Institute, Warsaw) dr. A. Boonstra (VU medisch centrum, Amsterdam)

Paranimfen Taco Kind Ard ten Bolscher

Financial support for printing of this thesis was kindly provided by

ActelionBayer HealthcareGlaxoSmithKlineKlein Kromhof HoutvezelsPluimers IsolatiePfizerTherabel

Insights into the progression of right ventricular failure in pulmonary arterial hypertension

Lay-out, illustrations and cover design: J.W. LankhaarPrinted by Drukwerkconsultancy, Utrecht

Copyright © 2011 by G.J. Mauritz, Amsterdam, the Netherlands.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise, without the written permission of the author. The rights of published chapters have been transferred to the publishers of the respective journals.

isbn 978-90-9026454-7

vii

Contents

1 General introduction: Pulmonary arterial hypertension and right ventricular failure 11

1.1 General introduction 111.2 Aim and outline of the thesis 15

2 Right ventricular ejection fraction is better reflected by transverse rather than longitudinal wall motion in pulmonary hypertension 19

2.1 Introduction 202.2 Materials and methods 202.3 Results 232.4 Discussion 292.5 Conclusions 32

3 Progressive changes in right ventricular geometric shortening and long-term survival in pulmonary arterial hypertension 35

3.1 Introduction 363.2 Methods 363.3 Results 383.4 Discussion 453.5 Conclusions 47

Contentsviii Contents

4 Prolonged right ventricular post-systolic isovolumic period in pulmonary arterial hypertension is not a reflection of diastolic dysfunction 51

4.1 Introduction 524.2 Methods 524.3 Results 544.4 Discussion 584.5 Conclusions 60

5 Pulmonary endarterectomy normalizes interventricular asynchrony and right ventricular systolic wall stress 65

5.1 Introduction 665.2 Methods 665.3 Results 685.4 Discussion 745.5 Conclusion 75

6 Usefulness of serial N-terminal pro-B-type natriuretic peptide measurements for determining prognosis in patients with pulmonary arterial hypertension 79

6.1 Introduction 796.2 Methods 806.3 Results 826.4 Discussion 86

7 Non-invasive stroke volume assessment in patients with pulmonary arterial hypertension: left-sided data mandatory 91


8 Early onset of retrograde flow in the main pulmonary artery is a characteristic of pulmonary arterial hypertension 105

8.1 Introduction 1058.2 Materials and methods 1068.3 Results 1108.4 Discussion 112

Contents ixContents

9 Progressive dilatation of the main pulmonary artery is a characteristic of pulmonary arterial hypertension and is not related to changes in pressure 119


10 Insights into the progression of right ventricular failure in pulmonary arterial hypertension: summary and future perspectives 131

10.1 Summary 13110.2 Future perspectives 134

Nederlandse samenvatting 139

Dankwoord 145

List of publications 151

About the author 155

C H A P T E R 1

11

Department of Pulmonary Diseases, Institute for Cardiovascular ResearchVU University Medical Center, Amsterdam

G.J. Mauritz

General introduction:Pulmonary arterial hypertension and right ventricular failure

1.1 GENERAL INTRODUCTION

1.1.1 The right heart and the circulation

‘THUS THE rIGHT VENTrICLE may be said to be made for the sake of trans-mitting blood through the lungs, not for nourishing them.’ In 1616, Sir Wil-liam Harvey, an English physician, was the first to describe with these words

the importance of right ventricular function in his seminal treatise, De Motu Cordis.1,2 In the normal heart, the right ventricle (rv) is complex and has been likened to a trun-

cated ellipsoid pyramid, draped around the cylindrical left ventricle (Figure 1.1A).3 The rv under normal circumstances is a thin-walled, low-pressure volume pump that moves desaturated blood across the pulmonary valve into the pulmonary circulation. Desaturated blood returns from the systemic circulation via the vena cava and the right atrium, finally crossing the tricuspid valve and entering the inlet portion of the rv cavity. The rv is con-nected in series with the lv and is, therefore, obligated to pump on average the same effec-tive stroke volume.

Since resting pulmonary vascular resistance (pvr) is, however, only one-seventh to one-tenth of normal systemic vascular resistance, the rv pressures are ten times lower and it operates at a far lower energy cost than the lv. The low-pressure environment within the pulmonary circulation is advantageous in that it prevents fluid migrating into the inter-stitial space and so provides optimal conditions for gas exchange. These low pressures do however make the pulmonary circulation in general, and the thin-walled rv in particular, poorly able to cope with changes in pulmonary vascular resistance and rising pulmonary artery pressures.4

12 Chapter 1

1.1.2 Pulmonary arterial hypertensionPulmonary hypertension (ph) is a pathophysiologic hemodynamic condition, defined by an increased mean pulmonary artery pressure of ≥ 25 mmHg at rest, measured (invasively) during right heart catheterisation.5,6 Pulmonary hypertension is divided into five subgroups (Table 1.1). The most common forms of pulmonary hypertension are ‘passive’, relating to left ventricular dysfunction with increased left ventricular end-diastolic pressure (Group 2) or copd (Group 3). The first groups consist of pulmonary arterial hypertension (pah). pah is a rare disease and has a prevalence of 15 per million in the general population.7 The patients are predominantly female (male-female ratio of 1 : 2) and it may develop at all ages. pah can be idiopathic, familial, or associated with a number of conditions or diseases, such as connective tissue disease, congenital heart disease, portal hypertension, hiv infection and exposure to toxins and drugs, including appetite suppressants.

Pathologically, pah is characterized by lesions of the small pulmonary arteries progress-ing from early medial hypertrophy to end-stage plexiform fibrosis.8,9 The consequent ob-struction and obliteration of the small arteries leads to increased pulmonary vascular resis-tance, increased right-sided pressures and finally right ventricular failure and a 15% annual mortality rate.10,11

At present, no curative treatment exists for this disease. In the past decades, however, new therapeutic options have become available that selectively lower pulmonary vascular resistance,12 thereby reducing the right ventricular afterload. Nevertheless, despite the improved survival due to this advanced medical therapy,13-15 pah still remains a progressive, fatal disease.

LV

RV

LV

RV

LV

RV

RV RV

RA

LA

LV

LV

LV

RV

LVRV

Normal heart Heart in PAHA B

Figure 1.1 Right ventricular (RV) configuration in the normal heart (A) and in pulmonary arterial hypertension (PAH, B). In PAH, the RV is characterized by increased end-diastolic volume, a change of the normal configura-tion to a more spherical one with a greater cross-sectional area than the left ventricle and RV hypertrophy.

13General introduction: PAH and RV failure

1

1.1.3 Right ventricular function and prognosis in pulmonary arterial hypertensionAlthough pulmonary artery pressure rise is the diagnostic criterion of pulmonary hyper-tension, the level of pressure itself has only modest prognostic significance in patients with pah. It is rather the ability of the right ventricle to cope with the progressive increase in pulmonary arterial pressure that mainly determines the patient's symptoms (i.e. breathless-ness, fatigue, peripheral oedema) and survival.

In studies addressing hemodynamic variables and survival with pah, a low cardiac index and high mean right atrial pressure are consistently associated with poorer survival.16-18 The first ph-registry already identifies parameters of rv function as independent prognostic variables, in contrast to the limited prognostic importance of parameters that reflect pul-monary arterial function.18 recently, data from the French registry showed that a relatively preserved right ventricular function may in turn lead to better outcomes in pah.14 This powerful prognostic influence reflects the fact that progressive rv failure is the usual and, unfortunately inevitable, cause of death in the vast majority of patients who develop pah. Therefore, rv structural and functional assessment should play a central role in both diag-nosis and follow-up of patients with pah.

1.1.4 Pathophysiology of right ventricular failure in PAHAs a consequence of the increased pressure overload, significant morphological and func-tional adaptive changes of the rv develop in pah (Figure 1.1B). It follows from Laplace's law that an increase in intraventricular pressure results in an increase in wall stress, unless the thickness of the chamber wall is augmented or the internal radius of the chamber is re-duced.19 Thus, due to the gradual increase in pulmonary vascular resistance, the rv adapts by thickening of the myocardial wall.20 The hypertrophy due to rv pressure overload causes an increase in the cell size of the cardiomyocytes, which in turn increases the diffusion distance of oxygen into the cell. In spite of the adaptive rv hypertrophy, the stroke volume (sv) often cannot be maintained and decreases. This loss of stroke volume in pah causes left ventricular underfilling, thus a further sv decrease and the systemic hypotension in these patients. The hypotension causes renal hemodynamic changes and the retention of sodium and water mediated by the systemic baroreceptors.21

This fluid retention causes rv volume overload and increased rv volume (dilatation) which in turn contributes to the increase in rv wall tension. It is the balance between rv mass (hypertrophy) and ventricular volume (dilatation) that determines the change in rv diastolic wall (muscle cell) stress. This balance between rv mass and rv volume is disrupted by progressive rv dilatation in pah. Both the increased cell size and wall stress cause im-pairment of cardiac muscle cell function and thus of rv systolic and diastolic function. The mechanisms involved in the pathophysiology of rv failure are illustrated in Figure 1.2.

Table 1.1 Clinical classification of pulmonary hypertension

Group 1 Pulmonary arterial hypertension (PAH)

Group 2 Pulmonary hypertension owing to left heart disease

Group 3 Pulmonary hypertension owing to lung diseases and/or hypoxia

Group 4 Chronic thromboembolic pulmonary hypertension (CTEPH)

Group 5 Pulmonary hypertension with unclear multifactorial mechanisms

14 Chapter 1

1.1.5 Increased right ventricular wall stress ventricular dyssynchronyAs shown in previous studies, the increased rv wall stress causes lengthening of the rv myocardial shortening period.22 This increased duration of rv free wall shortening leads to a reversed trans-septal pressure gradient and subsequent leftward septal motion at the start of lv diastole. The rv thereby diverts its energy into a leftward septal displacement. Not only does this impair the systolic efficiency of the rv ejection: it also impairs early diastolic efficiency of lv filling.23 This leads to a decrease in stroke volume which in turn contributes to the vicious circle of rv failure. The consequences of this septal bowing on rv function have been explored earlier,22 however need further investigation.

1.1.6 NT-proBNP in chronic right ventricular pressure overloadThe chronic increase in rv wall stress leads to myocardial release of atrial natriuretic pep-tide and brain natriuretic peptides (bnp).24 The final step of bnp synthesis consists of a high molecular weight precursor, probnp cleaved into biologically inactive N-terminal segment (nt-probnp) and the proper low molecular weight bnp. nt-probnp has a longer half-life and a better stability both in circulating blood and after sampling. Several studies showed that a single determination of nt-probnp plasma level reflects the severity of rv stretch and thus dysfunction and determines prognosis in pah.25,26 Therefore, bnp/nt-probnp plasma levels are used for initial risk stratification and may be considered for monitoring the effects

PAHRV pressure overload

Pulmonary vascular damageRV volume overload

Fluid retention+

RV adaptation RV dilatationRV hypertrophy

RV failureRV dilatation

RV hypertrophy

RV systolic function RV wall stress

L-R dyssynchrony

LV �lling

LV preload

Stroke volume

RV diastolic function

Figure 1.2 Pathophysiological mechanisms involved in right ventricular failure. RV = right ventricle, LV = left ven-tricle, L-R = left-to-right.


1

of treatment, in view of their prognostic implications. However, the value of serial determi-nations of nt-probnp in everyday clinical practice is still not established.

1.2 AIM AND OUTLINE OF THE THESIS

The premise on which this thesis is based is that rv function dominates prognosis in patients with pah. Therefore a better means of detecting, measuring and understanding the mechan-ics of rv failure might result in an improved outcome for pah patients. In this thesis, insights are to be obtained into the progression of right ventricular failure in pah patients. We used magnetic resonance imaging (mri) to characterize the rv geometry, the rv myocardial strain and the flow pattern in the pulmonary artery.

In clinical routine, the rv status is generally judged on the basis of geometric rv measures that are easily obtainable. rv geometric systolic shortening can be measured in the longitu-dinal and transverse dimension. In Chapter 2, we studied rv geometric shortening in pah and compared longitudinal versus transverse shortening as estimators for rvef. As yet, there is no consensus about the most practical measure to monitor rv function during follow-up. Therefore, in Chapter 3, we analyzed the one-year changes in rv geometric shortening and the association with subsequent mortality.

We observed that in severe pah, leftward ventricular septum bowing is a prominent fea-ture. As shown in earlier studies, this septum bowing is caused by interventricular asynchro-ny. This was the rationale to explore, in Chapter 4, the effect of interventricular mechanical asynchrony on rv function. The underlying mechanism of prolonged rv myocardial shorten-ing may be the increased rv wall stress. In Chapter 5, the role of rv end-systolic wall stress on interventricular mechanical asynchrony was tested by an intervention study: in cteph patients, both rv wall stress and interventricular asynchrony were measured before and after pulmonary endarterectomy. The increased rv wall stress triggers the release of the cardiac hormone nt-probnp. Previous studies have shown the prognostic value of nt-probnp mea-surements in pah at the time of diagnosis. In Chapter 6, we studied the value of serial nt-probnp measurements in predicting prognosis in pah.

A key parameter for rv function is the effective rv forward stroke volume. This is usually measured by quantifying the pulmonary artery (pa) flow. Previous studies showed that the velocity profile in pah is non-laminar. We hypothesized that this might influence the accu-racy of the sv in pah patients. For that reason, In Chapter 7, we assessed the accuracy of the pa flow for measuring sv in pah patients, by comparing sv from this pa flow with the sv as-sessed by the invasive reference standard. retrograde flow in the main pa is observed in pah patients. This retrograde flow might be indicative for pah. In Chapter 8 we studied the flow patterns in the pa in pah patients. These flow patterns may well be affected by the increased pa size, as observed in pah. It is however unclear whether changes in pa size reflect changes in pulmonary vascular hemodynamics. Therefore, in Chapter 9, we evaluated the pa size dur-ing follow-up in treated patients with pah and tested whether it reflects pulmonary vascular hemodynamics. Finally, Chapter 10 summarizes the thesis results and provides additional recommendations for future research. ■

16 Chapter 1

References

1. Harvey W. Exercitatio Anatomica de Motu Cordis et Santguinis in Animalibus. 1628.2. Goldstein J. The right ventricle: what’s right and what’s wrong. Coron Artery Dis 2005;16:1-3.3. Ho SY, Nihoyannopoulos P. Anatomy, echocardiography, and normal right ventricular dimensions. Heart

2006;92 Suppl 1:i2-13.4. Sheehan F, Redington A. The right ventricle: anatomy, physiology and clinical imaging. Heart

2008;94:1510-5.5. Galie N, Hoeper MM, Humbert M, et al. Guidelines for the diagnosis and treatment of pulmonary

hypertension: The Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J 2009;30:2493-537.

6. McLaughlin VV, Archer SL, Badesch DB, et al. ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association developed in collaboration with the American College of Chest Physicians; American Thoracic Society, Inc.; and the Pulmonary Hypertension Association. J Am Coll Cardiol 2009;53:1573-619.

7. Humbert M, Sitbon O, Chaouat A, et al. Pulmonary arterial hypertension in France: results from a national registry. Am J Respir Crit Care Med 2006;173:1023-30.

8. Farber HW, Loscalzo J. Pulmonary arterial hypertension. N Engl J Med 2004;351:1655-65.9. Cool CD, Kennedy D, Voelkel NF, Tuder RM. Pathogenesis and evolution of plexiform lesions in pulmonary

hypertension associated with scleroderma and human immunodeficiency virus infection. Hum Pathol 1997;28:434-42.

10. Runo JR, Loyd JE. Primary pulmonary hypertension. Lancet 2003;361:1533-44.11. Rubin LJ. Primary pulmonary hypertension. N Engl J Med 1997;336:111-7.12. Humbert M, Sitbon O, Simonneau G. Treatment of pulmonary arterial hypertension. N Engl J Med

2004;351:1425-36.13. Benza RL, Miller DP, Gomberg-Maitland M, et al. Predicting survival in pulmonary arterial hypertension:

insights from the Registry to Evaluate Early and Long-Term \ Pulmonary Arterial Hypertension Disease Management (REVEAL). Circulation 2010; 122:164-72.

14. Humbert M, Sitbon O, Chaouat A, et al. Survival in patients with idiopathic, familial, and anorexigen-associated pulmonary arterial hypertension in the modern management era. Circulation 2010;122:156-63.

15. McLaughlin VV, Suissa S. Prognosis of pulmonary arterial hypertension: the power of clinical registries of rare diseases. Circulation 2010;122:106-8.

16. Sandoval J, Bauerle O, Palomar A, et al. Survival in primary pulmonary hypertension. Validation of a prognostic equation. Circulation 1994;89:1733-44.

17. Rubin LJ. Primary pulmonary hypertension. Chest 1993;104:236-50.18. D’Alonzo GE, Barst RJ, Ayres SM, et al. Survival in patients with primary pulmonary hypertension. Results

from a national prospective registry. Ann Intern Med 1991;115:343-9.19. Arts T, Bovendeerd PH, Prinzen FW, Reneman RS. Relation between left ventricular cavity pressure and

volume and systolic fiber stress and strain in the wall. Biophys J 1991;59:93-102.20. Dias CA, Assad RS, Caneo LF, et al. Reversible pulmonary trunk banding. II. An experimental model for

rapid pulmonary ventricular hypertrophy. J Thorac Cardiovasc Surg 2002;124:999-1006.21. Schrier RW, Bansal S. Pulmonary hypertension, right ventricular failure, and kidney: different from left

ventricular failure? Clin J Am Soc Nephrol 2008;3:1232-7.22. Marcus JT, Gan CT, Zwanenburg JJ, et al. Interventricular mechanical asynchrony in pulmonary arterial

hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol 2008;51:750-7.


1

23. Gan CT, Lankhaar JW, Marcus JT, et al. Impaired left ventricular filling due to right-to-left ventricular interaction in patients with pulmonary arterial hypertension. Am J Physiol Heart Circ Physiol 2006;290:H1528-33.

24. Weber M, Hamm C. Role of B-type natriuretic peptide (BNP) and NT-proBNP in clinical routine. Heart 2006;92:843-9.

25. Blyth KG, Groenning BA, Mark PB, et al. NT-proBNP can be used to detect right ventricular systolic dysfunction in pulmonary hypertension. Eur Respir J 2007;29:737-44.

26. Fijalkowska A, Kurzyna M, Torbicki A, et al. Serum N-terminal brain natriuretic peptide as a prognostic parameter in patients with pulmonary hypertension. Chest 2006;129:1313-21.

C H A P T E R 2

19

ABSTRACT

J Cardiovasc Magn Reson. 2010 Jun 4;12(1):35

Background Longitudinal wall motion of the right ventricle (rv), generally quantified as tricuspid annular systolic excursion (tapse), has been well studied in pulmonary hypertension (ph). In contrast, transverse wall motion has been examined less. Therefore, the aim of this study was to evaluate regional rv transverse wall motion in ph, and its relation to global rv pump function, quantified as rv ejection fraction (rvef).Methods In 101 ph patients and 29 control subjects cardiovascular magnetic resonance was performed. From four-chamber cine imag-ing, rv transverse motion was quantified as the change of the septum-free-wall (sf) distance between end-diastole and end-systole at seven levels along an apex-to-base axis. For each level, regional absolute and fractional transverse distance change (sfd and fractional-sfd) were computed and related to rvef. Longitudinal measures, including tapse and fractional tricuspid-annulus-apex distance change (frac-tional-taad) were evaluated for comparison.Results Transverse wall motion was significantly reduced at all levels compared to control subjects (p < 0.001). For all levels, fractional-sfd and sfd were related to rvef, with the strongest relation at mid rv (R2 = 0.70, p < 0.001 and R2 = 0.62, p < 0.001). For tapse and fractional-taad, weaker relations with rvef were found (R2 = 0.21, p < 0.001 and R2 = 0.27, p < 0.001).Conclusion regional transverse wall movements provide important information of rv function in ph. Compared to longitudinal motion, transverse motion at mid rv reveals a significantly stronger relation-ship with rvef and thereby might be a better predictor for rv func-tion. ◦

Departments of aPulmonary Diseases, bPhysics and Medical Technology and cPhysiology,Institute for Cardiovascular Research, VU University Medical Center, Amsterdam

T. Kind,a G.J. Mauritz,a J.T. Marcus,b M. van de Veerdonk,a N. Westerhof,a,c

A. Vonk-Noordegraafa

Right ventricular ejection fraction is better reflected by transverse rather than longitudinal wall motion in pulmonary hypertension

20 Chapter 2

2.1 INTRODUCTION

PULMONAry HyPErTENSION (ph) is characterized by increased pulmonary pres-sure, leading to right ventricular (rv) overload. The course of the illness varies from patient to patient, with the worst prognosis seen in patients with the greatest degree

of rv dysfunction.1,2 This highlights the importance of knowledge of rv function for the determination of prognosis and therapy strategies.

In general, rv ejection fraction (rvef) is assumed to be a major determinant of systolic rv function, and has been shown to be of prognostic value in ph.3,4 However, determining rvef is time consuming and depends on geometric assumptions, and this has limited the application in clinical practice.5,6 A simpler approach is to approximate rvef by measuring the tricuspid annular plane systolic excursion (tapse). This measure quantifies the longitu-dinal shortening of the rv and its clinical value has been well established by echocardiogra-phy.7-10 It has also recently been applied using cardiovascular magnetic resonance (cmr).11 Less attention has been paid to rv transverse wall motion in the literature,12-14 despite the fact that transverse movements of the rv free wall towards the septum are important in rv ejection (bellows action).12,15,16

Therefore, the aim of the present study was to measure rv transverse motion in ph and to assess its relationship with rvef employing cmr. Longitudinal rv motion measurements were included for comparison purposes. results in ph patients were compared to control subjects.

2.2 MATERIALS AND METHODS

2.2.1 Patients and control subjectsThe local Ethics Committee of the vu University Medical Center approved the study pro-tocol and all participants gave written informed consent. Between September 2004 and September 2008, 658 patients were referred to our hospital for evaluation of ph. ph was diagnosed according to a standard protocol,17 including right heart catheterization (rhc), and was confirmed when the mean pulmonary artery pressure at rest was >25 mmHg and the pulmonary capillary wedge pressure was < 15 mmHg. Inclusion criteria were: patients with ph with aetiologies from who group 1 (pulmonary arterial hypertension) or group 4 (chronic thrombotic or embolic disease), and who had undergone cmr within 14 days after rhc had been performed. In total 123 patients met these criteria. Of this group, 15 patients were excluded because some cmr cines were lacking and 7 patients due to technically in-adequate images. Thus, in total 101 patients were included in the study. In addition, a total of 29 healthy non-smoking, age and gender matched control subjects were included as a reference group. These healthy subjects had had no rhc.

2.2.2 Cardiopulmonary exercise testing In a subset of patients (n = 66), cardiopulmonary exercise testing was performed on an electromagnetically braked cycle ergometer (rehcor, Lode Groningen, The Netherlands).

21RVEF and transverse and longitudinal wall motion

2

A progressive increase in pedalling workload (5-20 W/min) was applied until maximum tolerance was reached. Peak oxygen consumption (vo2,peak) was considered to express the patient’s exercise capacity, which was measured using a metabolic cart (Vmax 229; Viasys, yorba Linda, ca). To ensure accuracy, the measurements were time-averaged over a mini-mum of 20 seconds. The equipment was calibrated according the manufacturer’s specifica-tions.

2.2.3 Cardiovascular magnetic resonancecmr was performed by means of a 1.5T Siemens Avanto mri system (Siemens Medical So-lutions, Germany), equipped with a 6-element phased-array coil. ecg-gated cine imaging was performed using a balanced steady state free precession pulse sequence, during repeat-ed breath-holds. Long-axis slices were acquired in the four, three and two-chamber views. The four-chamber image plane was localized by the following steps: a basal short-axis im-age at end-diastole was used as the first localizer. Orthogonal to this short-axis image, the four-chamber view was obtained by rotating the planning line to such an orientation that it passes through the middle of the mitral and tricuspid valvular rings. The lv vertical long-axis view was used as the second localizer, to assure that the planned four-chamber cine passes through the most apical point of the lv cavity. Additionally, short-axis slices were obtained with a typical slice thickness of 5 mm and an interslice gap of 5 mm, fully cover-ing both ventricles from base to apex. The mr parameters used were: temporal resolution between 35 to 45 ms, voxel size 1.5 × 1.8 × 5.0 mm3, flip angle 60°, receiver bandwidth 930 Hz/pixel, Tr/TE 3.2/1.6 ms, matrix 256 × 156.

2.2.4 Image analysisAn apical four-chamber view was used to analyze longitudinal and transverse movements of the rv myocardium (Figure 2.1). The software for the analysis was implemented in mat-lab release r2008a (The MathWorks, Inc., Natick, United States).

Transverse measuresTransverse rv movements were analyzed using a method that quantifies regional changes from septum to free-wall (sf) between end-diastole and end-systole. First, lines were drawn between the right-lateral point of the tricuspid annulus and the apex, and between the left-lateral point of the mitral annulus and the apex. Then, the line intersecting the centers of the left and right lateral annulus- apex lines was drawn (Figure 2.1), both in end-diastole and in end-systole. sf dimensions were considered parallel to the intersecting line. The advantage of this approach is that clear markers are used to determine the geometric orien-tation of the sf dimensions. Next, contours were carefully drawn within the compact layer but outside the trabeculated layer of the rv myocardium. If at end-systole the interstices between hypertrophied trabeculae were no longer visible the corresponding boundary line was estimated after careful observation in cine mode. Subsequently, sf dimensions were computed at seven different levels covering the whole rv cavity, indicated as apex-1 through base-7, with level mid-4 exactly halfway through the rv. For each level, the absolute sf dis-tance change (sfd) was determined, and the fractional sf distance change (fractional-sfd) was then computed as sfd divided by the sf dimension at end-diastole. End-diastole was

22 Chapter 2

defined as the onset of the r-wave in the ecg. End-systole was determined visually as the moment of end-shortening of the rv free wall. This definition of end-systole was applied because a time delay often exists between pulmonary valve closure and end-shortening in ph. In this situation leftward septal bowing is observed, where rv end-shortening coincides with maximum bowing.18

Longitudinal measuresLongitudinal rv movements were quantified as modifiedtapse, adapted from echocardiog-raphy.11 cmr tapse was computed as the absolute distance change between end-diastole and end-systole of the tricuspid annulus-apex dimension. Fractional tricuspid annulus-apex distance change (fractional-taad) was calculated as tapse divided by the tricuspid annulus-apex dimension at end-diastole.

Area changerv area change in 4-chamber view was quantified from the endocardial contours at end-diastole and end-systole. Fractional area change (fac) was calculated as the absolute area change divided by the end-diastolic area.

Longitudinal Heart axis

RA

LA

LV

apex-123mid-456

base-7

A

ST

Septalbowing

Severe PHModerate PHControl

End-

dias

tole

End-

syst

ole RV

Figure 2.1 Four-chamber and short axis views of a healthy control subject, a patient with moderate PH and a patient with severe PH. The figure illustrates how the longitudinal dimension (from tricuspid annulus to apex (TA)) and transverse dimensions (from septum to free-wall (SF)) are determined. Firstly, both in end-diastole and end-systole, the left and right lateral annulus-apex lines were drawn. Secondly, the intersection through the centres of these lines was drawn. SF dimensions were considered parallel to this intersecting line. Thirdly, RV endocardial contours were drawn to determine the SF dimension at seven different levels covering the whole RV (indicated as apex-1 through base-7, with level mid-4 exactly halfway through the RV). The white lines in the short axis views indicate the intersections of the four-chamber views. RV = right ventricle, RA = right atrium, LV = left ventricle, LA = left atrium, A = Apex, T = lateral tricuspid annulus, S = RV endocardial septum.


2

RVEFEndocardial surfaces were carefully manually traced from the stack of short-axis cine im-ages, using Mass Analysis software (medis Medical Imaging Systems, Leiden, The Nether-lands) to obtain rv end-diastolic and end-systolic volumes. Trabeculae and papillary mus-cles were excluded from these measurements. The rv end-systolic image was identified by selecting the smallest ventricular surface. Based on these volumes, rvef and stroke volume were calculated.

2.2.5 StatisticsNormal distribution of the data was verified using a normal probability plot and log trans-formed if necessary. All data are presented as mean ± sd, unless stated otherwise. A p-value < 0.05 was considered statistically significant. Data between controls and patients were compared using the 2-tailed Students t-test for unpaired data or using one-way anova when more than two groups were tested. The Fisher exact test was used for categorical data. Linear regression was performed to test the relationships between the transverse mea-sures (sfd and fractional-sfd) and rvef, and between the longitudinal measures (tapse and fractional-taad) and rvef. In the regression analysis only patients were included in order to avoid regression bias by the control group. receiver operating characteristic (roc) analysis was used to test sensitivity and specificity of all measures to detect a rvef less than the median rvef value in patients.

To correct for multiple testing, the threshold for significance was adjusted using Bonfer-roni correction for families of tests with 0.05 divided by the amount of tests giving the ad-justed threshold for significance. Intra-observer and inter-observer variability of the endo-cardial wall measurements were assessed using the analysis of agreement method described by Bland and Altman.19 To this end, the same observer repeated cmr measurements of 10 patients and 10 control subjects within a period of one month to determine the intra-ob-server variability. A second observer repeated the same measurements to obtain the inter-observer variability. All statistical analyses were performed with spss Statistics 15.0 (spss Inc., Chicago, United States).

2.3 RESULTS

2.3.1 Characteristics of the study populationThere was no difference between the 101 ph patients and 29 control subjects with respect to age (ph = 50 ± 15 years vs. control = 46 ± 19 years, p = 0.082) and gender (ph = 68% female vs. control = 67% female, p = 0.35). Table 2.1 summarizes the clinical and hemodynamic characteristics of the ph patients. Most patients were from who group 1 (80%), with the remainder being from group 4 (chronic thromboembolic pulmonary hypertension). Ap-proximately 60% of all patients were medically treated for ph at enrollment, and many of them went through one or more regimens.

24 Chapter 2

Table 2.1 Baseline characteristics of the patients

Functional status

NYHA functional class II/III/IV (n) 30/51/20

6-minute walking distance (m), (predicted) 443 ± 142 (589 ± 102)

Dyspnoea score (Borg index) 4 ± 2

Diagnosis (n)

Idiopathic/familial PAH 41/9

PAH associated with

Systemic sclerosis 27

Portal hypertension 1

HIV 2

Chronic thromboembolic pulmonary hypertension 21

Treatment† (n)

Bosentan 17

Bosentan + Sildenafil 15

Epoprostenol 1

Sildenafil 7

Epoprostenol + Sildenafil 7

Sitaxentan 5

Sildenafil + Sitaxentan 4

Treprostinil + Bosentan + Sildenafil 3

Epoprostenol + Sildenafil + Bosentan 3

Treprostinil + Sitaxentan 2

Calcium Antagonist 2

Bosentan + Epoprostenol 2

Treprostinil + Sildenafil 1

Hemodynamics

Heart rate (bpm) 79 ± 13

Mean pulmonary artery pressure (mmHg) 46 ± 16

Mean right atrial pressure (mmHg) 6 ± 4

Pulmonary capillary wedge pressure (mmHg) 8 ± 4

Pulmonary vascular resistance (dyn · s / cm5) 632 ± 370

Cardiac output (l/min) 5.5 ± 1.9

Cardiac index (l/min/m2 ) 3.0 ± 0.9

Mixed venous O2 saturation (%) 67 ± 9

Cardiopulmonary exercise testing*

VO2,peak (ml/min) 1039 ± 544

VO2,peak-predicted 46.0 ± 20.7

Values expressed as mean ± SD. *Data obtained in a subset of patients (n = 66). †A considerable number of pa-tients went through different regimens before study enrollment. NYHA = New York Heart Association functional class.


2

Baseline cmr measurements of ph patients and control subjects are presented in Table 2.2. Values of rvef were significantly different between the groups, but no significant differ-ences were found for lvef. However, there was a significant difference for lvedv between the two groups. Figure 2.1 illustrates four-chamber and short-axis views of a healthy subject and two patients with moderate or severe ph.

2.3.2 Comparison of RV regional wall motion between patients and controls Figure 2.2 shows results of fractional-sfd, evaluated for seven levels ranging from apex to base, and fractional-taad in patients and controls. In ph patients, virtually no region-al variation in fractional-sfd was found. In contrast, control subjects showed the high-est fractional-sfd at the apical segment, and the lowest at the basal segment. At all levels, there were significant differences in transverse movements between patients and control subjects (Table 2.3). From Table 2.3 it can be seen that for all subjects, sfd values are consistently smaller than tapse, but that this does not hold for fractional-sfd compared to fractional-taad due to the shorter end-diastolic transverse dimension (sf) compared to the longitudinal dimension (taa). No significant differences were found in longitudinal and transverse motion between the patients from who group 1 and 4 (data not shown).

2.3.3 RV wall motion with reference to RV ejection fraction and exercise capacityFractional-sfd was correlated positively with rvef at each level of the rv free wall (Table 2.4), but the strongest correlations were found at level 4 (mid-level; R2 = 0.70, p < 0.001; Figure 2.3) and level 5 (R2 = 0.66, p < 0.001). Weaker relationships were found for sfd, fractional-taad and tapse (Figure 2.3 and Table 2.4).

Figure 2.4 and Table 2.5 show the results of roc-analysis, indicating that fractional-sfd at mid-rv is a sensitive and specific indicator of depressed rvef below the median value in patients (< 35%, range 10-61%, range control 41-77%). Compared to fractional-sfd, there were statistically significant differences among the areas under the curves for sfd (p = 0.005), fractional-taad (p = 0.003) and tapse (p < 0.001) after Bonferroni correction.

Table 2.2 CMR Measurements of control subjects and PH patients

Control PH p-value

Cardiac index (l/min/m2) 3.3 ± 0.9 2.6 ± 0.9 0.267

Heart rate (bpm) 73 ± 12 82 ± 16 0.107

Stroke volume (ml) 72 ± 18 54 ± 21 0.001

Stroke volume index (ml/m2) 35 ± 19 30 ± 11 0.001

RVEF (%) 56 ± 8 36 ± 12 0.001

LVEF (%) 65 ± 8 66 ± 9 0.689

RVEDV (ml) 126 ± 36 155 ± 61 0.001

LVEDV (ml) 119 ± 34 90 ± 29 0.001

Values expressed as mean ± SD. RV = right ventricle, LV = left ventricle, EF = ejection fraction, EDV = end-diastolic volume.

26 Chapter 2

mid-4 5 62 3 base-7apex-10

10

20

30

40

50

60

70

Chan

ge (%

)

Fraction-SFD

RV cavityTricuspid

valve

T

A

F

S

Control

PH

Fractional-TAAD

apex-1

5mid-4

32

6

base-7

Figure 2.2 Fractional-SFD in control subjects and PH patients, for seven longitudinal levels in the RV, from apex (level 1) to base (level 7). Fractional-TAAD is shown on the right. In patients, the RV exhibits for each ventricu-lar level approximately the same fractional-SFD. In contrast, control subjects show the highest fractional-SFD around the apex, with less fractional-SFD around the base. For every longitudinal level, differences in fractional change between controls and patients were significant with p < 0.0001. A = apex, T = lateral annulus of tricus-pid valve, S = RV endocardial septum, F = RV endocardial free wall.

Table 2.3� SF and TA distance at end-diastole and end-systole in controls and PH patientsControl PH

Transverse movements

Level SFed (mm) SFes (mm) SFD (mm) Fractional-SFD (%) SFed (mm) SFes (mm) SFD (mm) Fractional-

SFD (%)

apex-1 17 ± 5 7 ± 3 10 ± 4 56 ± 14 27 ± 11 25 ± 13 2 ± 7* 11 ± 28*

2 24 ± 5 11 ± 4 13 ± 4 54 ± 13 34 ± 10 31 ± 13 3 ± 7* 11 ± 22*

3 30 ± 5 15 ± 4 17 ± 6 52 ± 10 40 ± 9 35 ± 12 5 ± 6* 13 ± 18*

mid-4 36 ± 5 19 ± 5 19 ± 6 48 ± 9 46 ± 9 40 ± 12 7 ± 6* 16 ± 14*

5 41 ± 5 24 ± 5 18 ± 5 42 ± 9 52 ± 9 44 ± 12 8 ± 6* 16 ± 12*

6 44 ± 6 28 ± 6 16 ± 4 37 ± 10 55 ± 9 46 ± 11 8 ± 5* 16 ± 10*

base-7 45 ± 7 30 ± 7 15 ± 4 33 ± 9 55 ± 9 46 ± 10 9 ± 4* 17 ± 9*

Longitudinal movements

TAAed (mm) TAAes (mm) TAPSE (mm) Fractional-TAAD (%) TAAed (mm) TAAes (mm) TAPSE (mm) Fractional-

TAAD (%)

92 ± 11 65 ± 12 27 ± 7 29 ± 8 99 ± 12 82 ± 14 16 ± 6* 17 ± 7

RV area change

Areaed (mm2) Areaes (mm2) AC (mm2) Fractional-AC (%) Areaed (mm2) Areaes (mm2) AC (mm2) Fractional-AC

(%)

2840 ± 720 1397 ± 407 1443 ± 374 51 ± 6 3937 ± 1209 2798 ± 1271 1139 ± 421* 31 ± 13*

Values expressed as mean ± SD. SF is determined at different levels along an apex to base axis. Values of fractional-SFD are also shown in Figure 2.2. RV area change (AC) and fractional AC are determined in the 4-chamber view. *Significantly lower compared to control subjects (p < 0.001).


2

Transverse motion

0 20 40 60

fractional SFD, mid-level (%)

0

20

40

60

RVEF

(%)

C

R2 = 0.70, p < 0.001

0 20 40 60

fractional-TAAD (%)

D

R2 = 0.27, p < 0.001

0 10 20 30 400

20

40

60

SFD, mid-level (mm)

RVEF

(%)

A

R2 = 0.62, p < 0.001

Longitudinal motion

0 10 20 30 40

TAPSE (mm)

B

R2 = 0.21, p < 0.001

PAHControl

Figure 2.3� Regression between RVEF and SFD at mid-RV level (A), TAPSE (B), fractional-SFD at mid-RV level (C) and fractional-TAAD (D).

Table 2.4� R2-values resulting from linear regression of transverse and longitudinal parameters, and RVEF in PH patients

Transverse movements

Level Fractional-SFD SFD

apex-1 0.30* 0.30*

2 0.39* 0.35*

3 0.55* 0.48*

mid-4 0.70* 0.62

5 0.66* 0.55*

6 0.51* 0.37*

base-7 0.32* 0.17*

Longitudinal movements

Fractional-TAAD 0.27*

TAPSE 0.21*

RV Area change (AC)

AC 0.25*

Fractional-AC 0.76*

Values are mean values. Data determined in PH-patients only. RV area change (AC) and Fractional-AC are deter-mined in the 4-chamber view. *Significantly lower compared to control subjects.

28 Chapter 2

Table 2.5� Diagnostic performance of transverse and longitudinal parameters for detection of depressed RVEF below the median value (< 35%)

Transverse measures

Area under the curve* Standard error

Cut-off value

Sensitivity† (%)

Specificity† (%) p-value†

Fractional SF at mid-4 level 0.938 [0.892, 0.985] 0.0236 17.7% 84.2 88.1 < 0.001

ΔSF at mid-4 level 0.906 [0.849, 0.963] 0.0291 7.4 mm 80.7 83.1 < 0.001

Longitudinal measures

Area under the curve* Standard error Cut-off value Sensitivity†

(%)Specificity†

(%) p-value†

Fractional-TA 0.819 [0.741, 0.896] 0.0397 17.2% 75.4 78.0 < 0.001

TAPSE 0.767 [0.680, 0.854] 0.0442 16.3 mm 70.2 76.3 < 0.001

*Data shown with the 95% confidence interval between brackets. †At receiver operating characteristic curve analysis.

0 20 40 60 80 1000

20

40

60

80

100RVEF < 35%

100 – Speci�city (%)

Sens

itivi

ty (%

)

Fractional SFD

Fractional TAAD

SFD

TAPSE

Transverse mid-RV level:

Longitudinal:

Figure 2.4� Receiver operating characteristic curves showing the ability of fractional-SFD at mid-RV, SFD at mid-RV, fractional-TAAD and TAPSE to detect a RVEF below the median value (< 35%).

p = 0.002

Transverse motion Longitudinal motionA B

0

5

10

15

20

25

30

> 44 < 44

*

Frac

tiona

l−SF

D (%

)

VO2, peak-predicted (%)

0

5

10

15

20

25

30

> 44 < 44

VO2, peak-predicted (%)

Figure 2.5� Bar graphs to compare values of VO2,peak-predicted above and below the median value (44%) in com-parison with fractional-SFD (A) and Fractional-TAAD (B). VO2,peak-predicted values were obtained in a subset of patients (n = 66).


2In a subset of patients (n = 66), exercise capacity was measured using vo2,peak and vo2,peak-predicted (Table 2.1). Figure 2.5 illustrates vo2,peak-predicted below and above the median value (44%, range: 10-110%) in comparison with fractional-sfd and fractional-taad. A significant difference was found for fractional-sfd (p = 0.002) but not for fractional-taad.

2.3.4 VariabilityFigure 2.6 shows the intra- and inter-observer variability using Bland-Altman plots. The intra- and inter-observer variability for all longitudinal and transverse measures was not statistically significant.

2.4 DISCUSSION

2.4.1 General discussionWe have evaluated regional transverse motion of the rv myocardium, defined as move-ments of the rv free-wall to the septum, in a group of ph patients and control subjects. The most important finding of this study was that rvef was more closely related to fractional transverse movements than to longitudinal movements. The former might consequently be used as an alternative measure of rv pump function. In addition, we observed that frac-tional transverse movements in control subjects were largest near the apex, but smaller near the base of the heart, while ph patients showed consistently smaller movements and little regional variation.

2.4.2 RV ejection fractionA significant relationship between fractional-sfd and rvef was found for all levels, with the strongest relation at mid-rv level (R2 = 0.70, p < 0.001). The relationship between tapse and rvef, however, was much lower (R2 = 0.21, p < 0.001). Although this value corresponds to

−4

0

4Δ

(mm

)

Δ (%

)Δ

(%)

Δ (m

m)

0 5 10 15 20 25

−4

0

4

SFD average (mm)0 20 40 60

−8

0

8

Fractional-SFDavarage (%)

0 10 20 30−8

−4

0

4

8

TAPSE average (mm)0 10 20 30 40

−8

0

8

Fractional-TAADaverage (%)

Intr

a-ob

s.In

ter-

obs.

A B

F G H

C D

E

p = 0.176 p = 0.132 p = 0.096 p = 0.074

p = 0.433 p = 0.936 p = 0.978 p = 0.956

−8

0

8

−8

−4

0

4

8

−8

0

8

Figure 2.6 Analysis of agreement with Bland-Altman plots to illustrate the intra- and inter-observer variability of SF (A and E), fractional SFD (B and F), TAPSE (C and G), and fractional-TAAD (D and H). Mean difference and 95% limits of agreement are shown. Blue circles: PH patients, yellow circles: control subjects.

30 Chapter 2

what was found by Kjaergaard et al.20 (R2 = 0.23), other studies have reported higher cor-relations (R2 > 0.38) inclusion of control subjects in their regression analysis.

From a practical perspective, the measurement of fractional-sfd appears clinical useful. A cutoff value of 17.7% was used to differentiate between patients with a low (< 35%) or preserved (>35%) rvef (sensitivity of 84.2%; specificity of 88.1%), which had a significantly higher discriminatory power compared to sfd, fractional-taad or tapse (Figure 2.4; Table 2.5). Because rvef is of prognostic importance3,4 and our results showed that fractional-sfd has a high predictive value for rvef, fractional-sfd might be useful in clinical practice.

2.4.3 Functional considerationsThe importance of transverse wall movements in ejection was first acknowledged by rush-mer et al.15 They described that although longitudinal movements are easily identified, they are probably much less important for ejection than compression of the rv chamber by movement of the free wall toward the septum (bellows action). However, only a few studies have been performed on (regional) rv transverse wall motion and its contribution to rv function.12-14,21,22 and none of these were performed in ph. Moreover, while some reports ex-amined transverse movements just below the tricuspid valve,21,22 our results, in ph, showed that loss of transverse motion at this level was lower compared to the other regions (Figure 2.2 and Table 2.3). Evidence for a hypokinetic apex in ph was shown earlier by strain analy-sis employed by tissue Doppler echocardiography.23-25 However, this has not been observed using mr myocardial tagging.26

Considering the transverse measures at mid-rv a remarkably large difference exists be-tween ph patients and control subjects (Figure 2.2 and Table 2.3). Moreover, this difference is much larger than is found for the longitudinal measures. To be able to explain these re-sults accurately, the definitions of the measures need to be considered in more detail. sfd is defined as the difference between end-diastolic and end-systolic sf dimension, and thereby quantifies disturbances of rv free-wall movements. Additionally, paradoxical leftward sep-tal bowing increases the end-systolic sf dimensions and, as a consequence, also affects sfd. Thus, especially in severe patients the existence of a septal bowing may contribute to the large differences seen between patients and controls.

Fractional-sfd is defined as sfd divided by the end-diastolic sf dimension. The unusu-ally high end-systolic pressure in the lumen of the rv in patients with ph affects the cross-sectional shape of the cavity. Whereas it normally tends to be crescent shaped in cross section, it adopts a more circular cross section when contracting against pressure that ap-proaches or exceeds that of the left ventricle (Figure 2.1). This alteration of geometry main-ly affects transverse rather than longitudinal dimensions, and therefore contributes more to reduction of fractional-sfd rather than fractional-taad. In addition to any impairment of rv myocardial contractility, this ‘transverse’ dilation may contribute to the large difference between ph patients and controls.

2.4.4 Anatomical considerationsDespite several anatomical studies, the relationship between fiber orientation and right-heart mechanics is still not completely clear.27,28 Presently, there is increasing consensus that both the septum and rv free wall play an important role in rv contraction. The septum


2

consists of oblique longitudinal fibers with spiral architecture,29 resulting in the twisting motion required for efficient ejection against increased vascular resistance. In contrast, the predominant transverse fiber orientation of the rv free wall leads to circumferential compression or bellows action, which maintains rvef with normal pulmonary artery pres-sure.28,30,31

In the setting of pulmonary vascular disease, it has been shown that the fiber orientation of the septum becomes more transverse.30 An altered fiber orientation in the rv free wall is also likely to occur as the rv dilates. This dilation is profound toward the apical segments and results in an increased apical angle (Figure 2.1).32 Changes in fiber direction in the free wall are supported by a study of Pettersen et al.33 in which mr strain analysis was applied to patients with a systemic rv. There results indicated a predominant circumferential over longitudinal free wall shortening at mid-rv, while the reverse has been observed in healthy control subjects.33,34 Therefore, we hypothesize that disturbed rv wall motion in ph can be explained by impaired rv myocardial contractility due to altered fiber orientation. How-ever, further studies are needed to explain how changes of fiber geometry contribute to both longitudinal and transverse wall motion.

2.4.5 Practical implicationsrv function is the primary determinant of prognosis in ph. Therefore, clinicians need sim-ple and reproducible tests to assess rv function in order to improve their management of ph. In general, rvef is assumed to be a major determinant of systolic rv function. How-ever, its determination is time consuming and is limited by large inter and intra-observer variability. tapse has been shown to correlate with rvef and has been considered a simple method for semi-quantitative assessment of rv function. In this study we have showed that fractional-sfd around mid-rv level is more strongly correlated with rvef. Since sf dimen-sion around mid-level is as simple to measure as tapse (only two points are needed), it is an easy and accurate way to assess rv function with cmr. The clinical value of the transverse measures would be even stronger if these could be measured using echocardiography. This should be investigated in future research.

2.4.6 LimitationsSome limitations of the current study should be noted. Firstly, geometry and heavily tra-beculated myocardium of the rv make it sensitive to errors in the determination of en-docardial definition. Secondly, four-chamber views were acquired with equal geometric orientation at enddiastole and end-systole, without accounting for through plane motion. It is unknown whether rv torsion may affect the end-systolic dimension. A previous study reported regional differences in rotation in the systemic rv. However, when values were averaged over different segments, this resulted in almost absent rotation at the basal and the apical level with no absolute global ventricular torsion.35

32 Chapter 2

2.5 CONCLUSIONS

In ph, transverse myocardial motion is signifi cantly declined. Moreover, measures of trans-myocardial motion is signifi cantly declined. Moreover, measures of trans-yocardial motion is significantly declined. Moreover, measures of trans-verse movement at mid-rv reveal a significant relationship with rvef, which is much stron-ger than for measures of longitudinal movement. Since fractional-sfd at mid-level is as feasible and reliable to measure as tapse, it might be a better predictor of rv function in ph. Further studies are needed to evaluate the usefulness in clinical practice. ■

References

1. McLaughlin VV, Presberg KW, Doyle RL, Abman SH, McCrory DC, Fortin T, Ahearn G: Prognosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest 2004, 126:78S-92S.

2. D’Alonzo GE, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, Fishman AP, Goldring RM, Groves BM, Kernis JT, et al.: Survival in patients with primary pulmonaryhypertension. Results from a national prospective registry. Ann Intern Med 1991, 115:343-49.

3. Kawut SM, Horn EM, Berekashvili KK, Garofano RP, Goldsmith RL, Widlitz AC, Rosenzweig EB, Kerstein D, Barst RJ: New predictors of outcome in Idiopathic pulmonary arterial hypertension. Am J Cardiol 2005, 95:199-203.

4. Bleasdale RA, Frenneaux MP: Prognostic importance of right ventricular dysfunction. Heart 2002, 88:323-24.

5. Coghlan JG, Davar J: How should we assess right ventricular function in 2008? Eur Heart J Suppl 2007, 9(suppl_H):H22-H28.

6. Chin KM, Kim NH, Rubin LJ: The right ventricle in pulmonary hypertension. Coron Artery Dis 2005, 16:13-18.

7. Park JH, Park YS, Park SJ, Lee JH, Choi SW, Jeong JO, Seong IW: Midventricular peak systolic strain and Tei index of the right ventricle correlated with decreased right ventricular systolic function in patients with acute pulmonary thromboembolism. Int J Cardiol 2008, 125:319-24.

8. Ghio S, Klersy C, Magrini G, D’Armini AM, Scelsi L, Raineri C, Pasotti M, Serio A, Campana C, Vigano M: Prognostic relevance of the echocardiographic assessment of right ventricular function in patients with idiopathic pulmonary arterial hypertension. Int J Cardiol 2010, 140:272-8.

9. Forfia PR, Fisher MR, Mathai SC, Housten-Harris T, Hemnes AR, Borlaug BA, Chamera E, Corretti MC, Champion HC, Abraham TP, Girgis RE, Hassoun PM: Tricuspid annular displacement predicts survival in pulmonary hypertension. Am J Respir Crit Care Med 2006, 174:1034-41.

10. Kaul S, Tei C, Hopkins JM, Shah PM: Assessment of right ventricular function using two-dimensional echocardiography. Am Heart J 1984, 107:526-31.

11. Nijveldt R, Germans T, McCann GP, Beek AM, van Rossum AC: Semiquantitative assessment of right ventricular function in comparison to a 3D volumetric approach: A cardiovascular magnetic resonance study. Eur Radiol 2008, 18:2399-405.

12. Sakuma M, Ishigaki H, Komaki K, Oikawa Y, Katoh A, Nakagawa M, Hozawa H, Yamamoto Y, Takahashi T, Shirato K: Right ventricular ejection function assessed by cineangiography--Importance of bellows action. Circ J 2002, 66:605-9.

13. McConnell MV, Solomon SD, Rayan ME, Come PC, Goldhaber SZ, Lee RT: Regional right ventricular dysfunction detected by echocardiography in acute pulmonary embolism. Am J Cardiol 1996, 78:469-73.

14. Sheehan FH, Bolson EL, Dodge HT, Mathey DG, Schofer J, Woo HW: Advantages and applications of the centerline method for characterizing regional ventricular function. Circulation 1986,74:293-305.

15. Rushmer RF: Cardiovascular dynamics. 4th edition. Philadelphia: W.B. Saunders Company; 1976.16. Rushmer RF, Thal N: The mechanics of ventricular contraction; a cinefluorographic study. Circulation 1951,

4:219-28.


2

17. Barst RJ, McGoon M, Torbicki A, Sitbon O, Krowka MJ, Olschewski H, Gaine S: Diagnosis and differential assessment of pulmonary arterial hypertension. J Am Coll Cardiol 2004, 43:40S-7S.

18. Marcus JT, Gan CT, Zwanenburg JJ, Boonstra A, Allaart CP, Gotte MJ, Vonk- Noordegraaf A: Interventricular mechanical asynchrony in pulmonary arterial hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol 2008, 51:750-57.

19. Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986, 1:307-10.

20. Kjaergaard J, Petersen CL, Kjaer A, Schaadt BK, Oh JK, Hassager C: Evaluation of right ventricular volume and function by 2D and 3D echocardiography compared to MRI. Eur J Echocardiogr 2006, 7:430-38.

21. Anavekar NS, Gerson D, Skali H, Kwong RY, Yucel EK, Solomon SD: Twodimensional assessment of right ventricular function: an echocardiographic-MRI correlative study. Echocardiography 2007, 24:452-56.

22. Molinari G, Sardanelli F, Gaita F, Ottonello C, Richiardi E, Parodi RC, Masperone MA, Caponnetto S: Right ventricular dysplasia as a generalized cardiomyopathy? Findings on magnetic resonance imaging. Eur Heart J 1995, 16:1619-24.

23. Huez S, Vachiery JL, Unger P, Brimioulle S, Naeije R: Tissue Doppler imaging evaluation of cardiac adaptation to severe pulmonary hypertension. Am J Cardiol 2007, 100:1473-78.

24. Dambrauskaite V, Delcroix M, Claus P, Herbots L, D’hooge J, Bijnens B, Rademakers F, Sutherland GR: Regional right ventricular dysfunction in chronic pulmonary hypertension. J Am Soc Echocardiogr 2007,20:1172-80.

25. Borges AC, Knebel F, Eddicks S, Panda A, Schattke S, Witt C, Baumann G:Right ventricular function assessed by two-dimensional strain and tissue Dopplerechocardiography in patients with pulmonary arterial hypertension and effect of vasodilator therapy. Am J Cardiol 2006, 98:530-34.

26. Fayad ZA, Ferrari VA, Kraitchman DL, Young AA, Palevsky HI, Bloomgarden DC, Axel L: Right ventricular regional function using MR tagging:normals versus chronic pulmonary hypertension. Magn Reson Med 1998, 39:116-23.

27. Ho SY, Nihoyannopoulos P: Anatomy, echocardiography, and normal right ventricular dimensions. Heart 2006, 92(Suppl 1):i2-13.

28. Hristov N, Liakopoulos OJ, Buckberg GD, Trummer G: Septal structure and function relationships parallel the left ventricular free wall ascending and descending segments of the helical heart. Eur J Cardiothorac Surg 2006, 29(Suppl 1):S115-25.

29. Buckberg G, Mahajan A, Saleh S, Hoffman JI, Coghlan C: Structure and function relationships of the helical ventricular myocardial band. J Thorac Cardiovasc Surg 2008, 136:578-89. 589.e1-11

30. Buckberg GD: The ventricular septum: the lion of right ventricular function, and its impact on right ventricular restoration. Eur J Cardiothorac Surg 2006, 29 (Suppl 1):S272-8.

31. Armour JA, Randall WC: Structural basis for cardiac function. Am J Physiol 1970, 218:1517-23.32. Lopez-Candales A, Dohi K, Iliescu A, Peterson RC, Edelman K, Bazaz R: An abnormal right ventricular apical

angle is indicative of global right ventricular impairment. Echocardiography 2006, 23:361-68.33. Pettersen E, Helle-Valle T, Edvardsen T, Lindberg H, Smith HJ, Smevik B, Smiseth OA, Andersen K:

Contraction pattern of the systemic right ventricle shift from longitudinal to circumferential shortening and absent global ventricular torsion. J Am Coll Cardiol 2007, 49:2450-56.

34. Haber I, Metaxas DN, Geva T, Axel L: Three-dimensional systolic kinematics of the right ventricle. Am J Physiol Heart Circ Physiol 2005,289:H1826-33.

35. Fogel MA, Weinberg PM, Fellows KE, Hoffman EA: A study in ventricularventricular interaction. Single right ventricles compared with systemic right ventricles in a dual-chamber circulation. Circulation 1995,92:219-30.

C H A P T E R 3

35

ABSTRACT Background Until now, many investigators have focused on describ-Until now, many investigators have focused on describ-ing right ventricular (rv) dysfunction in groups of patients with pul-monary arterial hypertension (pah) but very few have addressed the deterioration of rv function over time. The aim of this study was to investigate time courses of rv geometric changes during the progres-sion of rv failure.Methods 42 pah patients were selected who underwent right heart catheterization and cmr at baseline and after 1-year follow-up. Based on the survival after this one-year run-in period, patients were clas-sified into two groups: survivors (26 patients): subsequent survival of more than 4 years, and non-survivors (16 patients): subsequent survival of less than 4 years. Four-chamber cine imaging was used to quantify rv longitudinal shortening (apex-base distance change), rv transverse shortening (septum-free wall distance change), and rv fractional area change (rvfac) between end-diastole and end-systole. Results Longitudinal shortening, transverse shortening, and rvfac measured at the beginning of the run-in period and 1 year later, were significantly higher in subsequent survivors than in non-survivors (p < 0.05). Longitudinal shortening did not change during the run-in period in either patients group. Transverse shortening and rvfac did not change during the run-in period in subsequent survivors, but decreased in subsequent non-survivors (p < 0.05). This decrease was caused by increased leftward septal bowing. Conclusions Progressive rv failure in pah is associated with a parallel decline in longitudinal and transverse shortening until a floor effect is reached for longitudinal shortening. A further reduction of rv func-tion is due to progressive leftward septal displacement. Because trans-verse shortening incorporates both free wall and septum movements, this parameter can be used to monitor the decline in rv function in end-stage pah. ◦

Chest (e-pub ahead of print)

*Both authors contributed equally

G.J. Mauritz,a* T. Kind,a* J.T. Marcus,b H.J. Bogaard,c M. van de Veerdonk,a P.E. Postmus,aA. Boonstra,a N. Westerhof,a,d A. Vonk-Noordegraafa

Departments of aPulmonary Diseases, bPhysics and Medical Technology and dPhysiology, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam; cDepartment of Medicine, Division of Pulmonary and Critical Care, Virginia Commonwealth University, Richmond, VA

Progressive changes in right ventricular geometric shortening and long-term survival in pulmonary arterial hypertension

36 Chapter 3

3.1 INTRODUCTION

PULMONAry ArTErIAL HyPErTENSION (pah) is a progressive and devastating disease characterized by pathological lesions of the pulmonary arteries leading to increased pulmonary artery pressure.1 Although the pulmonary pressure rise is the

distinctive characteristic of this disease, it is right ventricular (rv) function that dominates prognosis and long-term survival.2,3 Several geometrical measures exist that quantify rv function. rv longitudinal shortening (tapse) has been shown to be a good reflection of rv systolic function.4-6 However, tapse only accounts for the longitudinal rv shortening and ignores the septal contribution to rv ejection.7 recently, our group showed that rv trans-verse shortening reflects global rv function at least as much as longitudinal shortening.8 Furthermore, rv fractional area change (rvfac) has been shown to be a robust measure of rv systolic function and has a good correlation with invasive hemodynamic data in pah patients.9, 10 This rvfac has the advantage of combining transverse and longitudinal rv shortening in one measure. However, it is largely unknown how these different rv param-eters change during the follow up of patients with progressive rv failure. Therefore, the aim of this study was to evaluate the pathophysiological changes of rv geometric shortening in patients with progressive pah, who died between 1 and 5 years after the initial diagnosis and to compare these results with stable pah patients, surviving more than 5 years. To reach this objective, rv longitudinal and transverse shortening, as well as rvfac, were measured in a group of pah patients at baseline and after one year of follow-up.

3.2 METHODS

3.2.1 Patients populationThe present study was performed in an observational cohort of pah patients, and was part of a prospective ongoing research project aimed to evaluate the rv in pah by means of mri. Between May 2003 and May 2005, 287 patients were referred to our hospital for the evaluation of pulmonary hypertension (ph). A diagnosis of pah was established following a standard protocol which included right heart catheterization (rhc).11 We selected treat-ment naive pah patients who underwent cmr at the time of diagnosis and again after one year of follow-up. Patients with congenital systemic-to-pulmonary shunts were excluded from this study.

In total, 42 patients were selected and grouped according to subsequent survival after the one-year run-in period: survivors (n = 26) were those patients with a transplant-free survival period of at least five years after the baseline cmr. The non-survivors (n = 16) were defined as the patients who died between one and five years after baseline cmr due to progression of the disease. Although the choice of five years is arbitrary, this period was chosen to make a clear distinction between long and short-term survivors. In addition, previous reports and clinical experience show that patients surviving five years can be clas-sified as in a stable clinical condition, since annual mortality rate after five years is low for this group.12, 13

37RV geometric shortening changes and survival

3

Twenty-five patients met the diagnostic criteria for the study and survived for more than one year after diagnosis but could not be selected because they did not undergo follow-up cmr measurements (e.g. dropout, withdrawal or protocol deviators). These patients were analyzed for potential selection bias.

Institutional research Board approval of this protocol was obtained and all patients con-sented to participate in the study.

3.2.3 Cardiac magnetic resonance imagingcmr examination was performed using a 1.5T Siemens Sonata mri system (Siemens Medi-cal Solutions, Germany), equipped with a 6-element phased-array coil. ecg-gated cine im-aging was performed using a balanced steady-state free precession pulse sequence, during breath-holds. Long-axis slices were acquired in the four-, three- and two-chamber views. Additionally, short-axis slices were obtained with a typical slice thickness of 5 mm and a slice gap of 5 mm, fully covering both ventricles from base to apex. mr parameters used were: temporal resolution between 35 to 45 ms, voxel size 1.5 × 1.8 × 5.0 mm3, flip angle 60°, receiver bandwidth 930 Hz/pixel, tr/te 3.2/1.6 ms, matrix 256 × 156.

3.2.4 Image analysis

Global LV and RV function analysisFrom the stack of short-axis cine images endocardial surfaces were carefully manually traced, using Mass Analysis software (medis Medical Imaging Systems, Leiden, The Neth-erlands) to obtain rv and lv end-diastolic, end-systolic volumes, stroke volumes and ejec-tion fractions. All volumetric cmr measures were corrected for body surface area (bsa).

RV Geometric shortening analysisThe four-chamber cine images were analyzed according to a method we described previ-ously.8 End-diastole was defined as the onset of the r-wave of the ecg. End-systole was determined as the moment of end-shortening of the rv free wall (using mr cine imaging).

RV Longitudinal shortening ■ was calculated as the distance change between end-diastole and end-systole of the tricuspid annulus to apex distance. RV Transverse shortening ■ was calculated as the change between end-diastole and end-systole of the rv free wall to septum distance (Figure 3.1). This transverse shortening was measured at seven different levels covering the whole rv cavity from base (level 1) through apex (level 7), with mid-level (level 4) exactly halfway through the rv. RV free wall displacement ■ was defined as the displacement of the rv free wall with respect to the tricuspid annulus to apex line Septal displacement ■ is the displacement of the interventricular septum to the tricuspid annulus to apex line (Figure 3.1). Positive displacement means towards the tricuspid an-nulus to apex line, and negative displacement means away from this line. RV fractional area change ■ is calculated from the rv end-diastolic and end-systolic areas measured from the four-chamber view as: rvfac = 100 × (rv end-diastolic area – rv end-systolic area) / rv end-diastolic area.

The procedural duration of post-acquisition geometric assessments depends on the num-

38 Chapter 3

ber of contours, which have to be drawn. To calculate rvef, endocardial contours are drawn at end-diastole and end-systole on all short-axis slices (generally 8 to 12). rvfac is obtained from a single four-chamber view at end-diastole and end-systole; this process is up to 8 to 12 times faster than the determination of rvef. For the longitudinal and transverse mea-sures, the procedure is even simpler since it requires only single dimensions at end-diastole and end-systole.

3.2.5 Statistical analysisNormal distribution of the data was verified using a normal probability plot and log trans-formed if necessary. All data are presented as mean ± sd, unless stated otherwise. Compari-sons between and within groups were calculated using unpaired and paired Student t-tests. A one-way anova was used when more than two groups were tested. Linear regression analysis was used to estimate the correlation between rvef and geometric shortening mea-sures. The Fisher exact test was used for categorical data. Statistical tests were performed with spss software 18 (spss Inc, Chicago, United States). Statistical significance was consid-ered when p < 0.05.

3.3 RESULTS

3.3.1 Patients characteristicsThe baseline demographic and hemodynamic data of both the survivors and non-survivors are summarized in Table 3.1. In 4 survivors and 4 non-survivors no rhc was performed at one-year follow-up within 2 weeks of cmr assessment. The non-survivors had a mean survival of 2.5 ± 1 year. There was no difference between the survivors and non-survivors

RV End-diastoleRV End-systole

RA

RV

Tricuspid annulus

Apex

Free wall

Septum

b

dc

Transverse shortening = (a + b) – (c + d)Free wall displacement = a – cSeptum displacement = b – d

a

Figure 3�.1 Schematic depiction of RV transverse shortening and RV free wall and septal displacement. RV trans-verse shortening is defined as a change in the distance change from RV free wall to septum between RV end-diastole and RV end-systole. Displacement of the free wall and septum is defined as the displacement relative to the tricuspid annulus to apex line between RV end-diastole and RV end-systole.


3

Table 3�.1 Baseline characteristics of the study population

Entire population(n = 42)

Survivors(n = 26)

Non-survivors(n = 16) p-value

General characteristics

Age (y) 44 ± 14 43 ± 11 46 ± 18 0.46

Sex (male/female) 11/31 8/18 3/13 0.61

Diagnosis, n (%)

Idiopathic PAH 28 (67) 20 (77) 8 (50) 0.19

Heritable PAH 3 (7) 2 (8) 1 (6) 1.00

Drugs and toxins induced PAH 1 (2) 1 (4) 0 (0) 1.00

PAH associated with

Connective tissue disease 8 (19) 1 (4) 7 (44) 0.02

HIV infection 1 (3) 1 (4) 0 (0) 1.00

Portal hypertension 1 (3) 1 (4) 0 (0) 1.00

Functional status

NYHA functional class, n (%)

II 13 (31) 10 (38) 3 (18) 0.31

III 21 (50) 15 (58) 6 (38) 0.22

IV 8 (19) 1 (4) 7 (44) 0.001

6-minute walking distance (m) 434 ±136 465 ±142 380 ±106 0.04

Hemodynamics

Heart rate (bpm) 80 ± 17 78 ± 14 83 ± 13 0.30

mSAP (mmHg) 90 ± 17 91 ± 9 89 ± 13 0.38

mPAP (mmHg) 52 ± 15 54 ± 14 50 ± 16 0.46

Cardiac output (L/min) 4.6 ± 1.4 5.0 ± 1.5 3.9 ± 0.9 0.017

PVR (dyn·s·cm‒5) 878 ± 455 821 ± 413 970 ± 516 0.31

RAP (mmHg) 8 ± 6 7 ± 4 10 ± 8 0.08

PCWP (mmHg) 8 ± 5 8 ± 6 7 ± 4 0.34

Sa02 (%) 95 ± 3 95 ± 2 94 ± 4 0.37

Sv02 (%) 64 ± 9 66 ± 9 60 ± 11 0.08

Medication use, n (%)

Intravenous prostacyclin 12 (29) 6 (16) 6 (33) 0.25

Endothelin receptor antagonist 24 (58) 17 (45) 7 (39) 0.15

Sildenafil 15 (36) 9 (24) 6 (33) 0.70

Calcium antagonist 5 (12) 4 (11) 1 (6) 0.37

Data expressed as mean ± SD. mSAP = mean systemic arterial pressure, mPAP = mean pulmonary arterial pressure; PVR = pulmonary vascular resistance, RAP = mean right atrial pressure, PCWP = pulmonary capillary wedge pressure, SaO2 = arterial oxygen pressure, SvO2 = mixed venous oxygen saturation.

40 Chapter 3

with respect to age and gender. The non-survivors group contained more patients with connective tissue disease, a greater number of patients nyha Class iv and poorer average distance 6MWT than survivors. Furthermore, cardiac output (co) was significantly lower in non-survivors, but pulmonary artery pressure and pvr at baseline were similar in both groups. All patients were optimally treated and there was no significant difference in the use of the ph medication between survivors and non-survivors. Figure 3.2 shows the hemo-dynamics in the subset of patients (22 survivors and 12 non-survivors) with rhc at baseline and follow-up. Although statistically not significant, both the survivors and non-survivors showed a reduction in pvr after one-year follow-up. Heart rate increased significantly dur-ing follow-up in the non-survivors.

Selection bias was tested on patients without follow-up measurements (n = 25: see meth-ods). Baseline characteristics of these patients were not significantly different compared to the total study population, except for age: included patients were significantly younger than the patients without follow-up measurements (Table 3.2).

3.3.2 Image analysis

Global RV and LV functionFigure 3.3 illustrates a long-axis view of a survivor and a non-survivor patient at baseline in end-diastole. Compared to the survivor, the rv configuration in the non-survivor shows a markedly dilated atrium and ventricle. Volumetric and geometric cmr measurements at baseline and follow-up for the subsequent survivors and non-survivors are presented

Survivors Non-survivors0

250

500

750

1000

1250(d

yn . s

. cm

–5)


25

50

75

100

(bpm

)Pulmonary vascular resistanceA

Heart rateC


1

2

3

4

5

6

7

(L /

min

)

Cardiac outputBp = 0.03

p = 0.03

BaselineFollow-up


10

20

30

40

50

(mL

/ mm

2)

Stroke volume indexD

Figure 3�.2 Pulmonary vascular resistance (A), cardiac ouput (B), heart rate (C) and stroke volume index (D) in 34 PAH patients (22 survivors and 12 non-survivors) with a right heart catheterization at baseline and one-year follow-up. All data are presented as mean ± SEM.


3

Table 3�.2 Characteristics of patients with and without follow-up

With follow-up(n = 42)

Without follow-up(n = 25) p-value


Age (y) 44 ± 14 56 ± 17 0.002

Sex (male/female) 11/31 8/17 0.78

Diagnosis, n (%)

Idiopathic PAH 28 (67) 16 (64 ) 1.00

Heritable PAH 3 (7) 2 (8) 1.00

Drugs and toxins induced PAH 1 (2) 0 (0) 1.00

PAH associated with

Connective tissue disease 8 (19) 5 (20) 1.00

HIV infection 1 (3) 1 (4) 1.00

Portal hypertension 1 (3) 1 (4) 1.00

Functional status


II 13 (31) 10 (40) 0.43

III 21 (50) 13 (52) 0.80

IV 8 (19) 2 (8) 0.31

6-minute walking distance (m) 434 ± 136 407 ±173 0.56

Hemodynamics

Heart rate (bpm) 80 ± 17 81± 17 0.94

mPAP (mmHg) 52 ± 15 52 ± 18 0.99

Cardiac output (L/min) 4.6 ± 1.4 5.4 ± 3 0.10

PVR (dyn·s·cm‒5) 862 ± 461 897 ± 788 0.83

RAP (mmHg) 8 ± 6 8 ± 7 0.98

PCWP (mmHg) 8 ± 5 8 ± 6 0.75

Sa02 (%) 95 ± 3 92 ± 9 0.29

Sv02 (%) 64 ± 10 64 ± 9 0.96

CMR

RVEDVI 77 ± 25 72 ± 20 0.47

RVESVI 53 ± 22 47 ± 19 0.29

RVSVI 24 ± 9.2 25 ± 9.1 0.56

RVEF 32 ± 11 37 ± 13 0.15

LVEF 66 ± 11 69 ± 10 0.15

Data expressed as mean ± SD. mSAP = mean systemic arterial pressure, mPAP = mean pulmonary arterial pressure; PVR = pulmonary vascular resistance, RAP = mean right atrial pressure, PCWP = pulmonary capillary wedge pressure, SaO2 = arterial oxygen pressure, SvO2 = mixed venous oxygen saturation, EDVI = end-diastolic volume index, ESVI = end-systolic volume index, SVI = stroke volume index, EF = ejection fraction.

42 Chapter 3

in Table 3.3. rvef at the beginning and at the end of the run-in period was significantly lower in non-survivors than in survivors. lv end-diastolic volume index (edvi) and rv and lv stroke volume index (svi) were significantly lower at the end of the run-in period in non-survivors than in survivors. Although baseline rvedvi was not significantly different between the survivors and non-survivors, the rv dilated progressively during follow-up only in non-survivors. rvef remained stable in the survivors while it further decreased in the non-survivors.

RV Geometric shorteningLongitudinal shortening ■ There was a significant difference between subsequent survi-vors and non-survivors as shown in Figure 3.4A; the non-survivors had a reduced rv longitudinal shortening as compared to the survivors (14 ± 7 vs. 20 ± 5 mm, p < 0.05). At follow-up, the change in rv longitudinal shortening was not significantly different

Survivor Non-survivor

RV

RA LV

RVRA

LV

pericardiale�usion

Figure 3�.3� CMR images of a PAH-survivor (left) and a non-survivor (right). They encountered equally raised mean pulmonary artery pressure (47 vs. 48 mmHg, respectively) but differed in their response to afterload. Note the enlarged right atrium and RV in the non-survivor as signs of a failing RV. All data are presented as mean ± SEM.

Table 3�.3� Global baseline and follow-up RV and LV function

Survivors Non-survivors

Baseline Follow-up pintra Baseline Follow-up pintra pinter, baseline pinter, follow-up

Left ventricle

EDVI (ml/m2) 43 ± 12 44 ± 15 0.84 39 ± 13 32 ± 11 0.03 0.14 0.02

ESVI (ml/m2) 15 ± 5 16 ± 9 0.31 14 ± 9 13 ± 7 0.27 0.33 0.28

SVI (ml/m2) 28 ± 9 27 ± 10 0.39 24 ± 7 19 ± 7 0.01 0.15 0.008

EF (%) 66 ± 9 64 ± 12 0.41 65 ± 14 62 ± 16 0.46 0.76 0.62

Right ventricle

EDVI (ml/m2) 74 ± 21 71 ± 21 0.34 80 ± 31 94 ± 39 0.04 0.56 0.02

ESVI (ml/m2) 49 ± 18 48 ± 21 0.96 57 ± 24 75 ± 34 0.01 0.15 0.003

SVI (ml/m2) 25 ± 5 22 ± 8 0.07 22 ± 10 19 ± 7 0.21 0.24 0.03

EF (%) 35 ± 11 33 ± 12 0.21 28 ± 9 19 ± 7 0.002 0.03 0.001

Data expressed as mean ± SD. EDVI = end-diastolic volume index, ESVI = end-systolic volume index, SVI = stroke volume index, EF = ejection fraction, pintra = p-values between baseline and follow-up within the group of survivors and non-survivors, pinter= p-values between the survivors and non-survivors groups.


3

A

B

0

3

6

9

12

15

18

21

0

3

6

9

12

15

18

21

Long

itudi

nal s

hort

enin

g (m

m)

Non-survivors

*

*

*

*

*

*

Survivors

BaselineFollow-up

–4–20246810 –4–20246810

†

†

†

†

†

Transversal shortening (mm)

Base-1

2

3

Mid-4

5

6

Apex-7

Figure 3�.4� The averaged RV longitudinal (A) and transverse shortening (B) of survivors and non-survivors for baseline (dark bars) and follow-up (light bars). The longitudinal shortening in the non-survivors is significantly decreased at baseline compared to survivors, whereas in both groups during follow-up no significant change is seen. The transverse shortening in the non-survivors is significantly decreased for level 2 to 7 at baseline compared to survivors (*: p < 0.05). During follow-up the RV transverse shortening, for levels 2 to 6, further de-creases significantly (†: p < 0.05) from baseline and even shows lengthening at mid-ventricular to apical level, whereas in the survivors it did not decrease. Data are presented as mean ± SEM.

44 Chapter 3

between baseline and follow-up values, both for the survivors (p = 0.25) as well as for the non-survivors (p = 0.13). In the non-survivors there was no significant correlation between the change in longitudinal shortening and change in rvef (ρ = 0.4, p = 0.13).Transverse shortening ■ The results of the transverse shortening, measured both at base-line and follow-up for the subsequent survivors and non-survivors, are given in Figure 3.4B. For the survivors, the rv transverse shortening did not change significantly at all levels between baseline and follow-up (e.g. at mid-rv level: 9.7 ± 3.6 vs. 9.5 ± 4.7 mm, p = 0.82). For the non-survivors, rv shortening at baseline was significantly reduced for levels two to seven as compared to the survivors (e.g. at mid-rv level: 5.1 ± 5.5 vs. –0.5 ± 5.2 mm, p < 0.001). During follow-up, the rv transverse shortening further decreased significantly from baseline at levels two to six and even showed lengthening at levels from mid to apex. There was a significant correlation between the change in rvef and the change in rv transverse shortening obtained at mid-level (ρ = 0.66, p = 0.005) in the non-survivors. RV Wall and septal displacement ■ As mentioned above, the decrease in transverse short-ening at follow-up is a feature of non-survivors. This shortening is determined by the displacements of rv free wall and septum. Note that shortening is a change of the dis-tance between rv free wall and the septum, while displacement is an isolated property of the rv free wall and septum separately. In Figure 3.5 we illustrated the transverse dis-placements of rv free wall and septum in non-survivors separately. For the rv free wall displacement, there was no significant change at follow-up for any level. In contrast, for the septum there was a significant increase of leftward displacement for levels 2 through 7, demonstrating that the decrease in transversal shortening in progressive pah is mainly due to worsening of leftward septum bowing.

RV

Baseline Follow-upED ESEDES

Base-1

Apex-7

Mid-4

2

3

5

6

00 5 0 5–5

ED ESEDES

0 –5Displacement (mm) Displacement (mm)

Free wallSeptum

Figure 3�.5� RV free wall and septal transverse displacement in non-survivors at baseline and follow-up. For the RV free wall all the (blue) bars are to the right, which means displacement toward the RV cavity. For the septum (red bars), a bar to the left means displacements towards the RV cavity, and a bar to the right means displace-ment outward from the RV cavity. For the free-wall movements, there was no significant change at follow-up compared to baseline at any level, but there was a significant increase of leftward septum displacement for levels 2 through 7 at follow-up compared to baseline (*: p < 0.05).


3

RV fractional area change ■ The change in rvfac is illustrated in Figure 3.6A. Similar to rv transverse shortening, the rvfac in the survivors remained stable during one-year of follow-up. The non-survivors showed a decreased rvfac at baseline already compared to the survivors (24 ± 10 vs. 31 ± 9%, respectively, p = 0.03). At follow-up the non-survivors showed a further significant (p < 0.001) decrease to 17 ± 10%. Also, there was a moderate correlation between the change in rvfac and the change in rvef (ρ = 0.67, p = 0.005, Figure 3.6B).

3.4 DISCUSSION

3.4.1 General discussionWe investigated the pathophysiological changes in rv geometry in pah patients in the first year after diagnosis and related the geometric changes occurring during this initial year to subsequent survival. The major finding of this study is the following characterization of rv properties in subsequent non-survivors:

■ Longitudinal shortening and transverse shortening are already reduced at baseline. ■ Both longitudinal shortening and rv free wall motion stay the same over time in non-survivors, while transverse shortening shows a further decline over time. ■ The end-stage decline in rv function is due to a progressive leftward septal displacement, rather than due to a further decrease in rv free wall transverse or longitudinal displace-ment.

3.4.2 RV Geometric shorteningAt baseline, longitudinal and transverse shortening was significantly lower in subsequent non-survivors. The longitudinal shortening is comparable with the tricuspid annular plane systolic excursion (tapse) as used in echocardiography.14 Forfia et al.15 showed that patients with an echocardiographically derived tapse of less than 18 mm had a significantly re-duced survival and an especially poor outcome was shown in those patients with a tapse of less than 15 mm. The reduced tapse of the non-survivors in the current study (mean 15 ±

RV Fractional area Non-survivors


10

20

30Ch

ange

(%)

40

ΔRV fractional area change (%)

ΔRV

EF (%

)

ρ = 0.67, p =0.005

*

†

A B

–30–30

–20

–20

–10

–10

0

0

10

10BaselineFollow-up

Figure 3�.6 A: RV Fractional area change at baseline and follow-up. B: For the non-survivors the change in RVFAC correlated with the change in RVEF. (*: p <0.05 compared to baseline survivors; †: p < 0.001 compared to base-line non-survivors).

46 Chapter 3

6 mm) corresponds to the observations by Forfia et al. 15 During a one-year run-in period, we observed no further changes in longitudinal shortening in subsequent non-survivors, whereas stroke volume and rvef showed a progressive decline in these patients.

The absence of a further loss in tapse in the group of non-survivors may be explained by a ‘floor effect’ beyond a mean tapse of 15 mm in the group of non-survivors. We argue that rv dysfunction may start with a loss of tapse and that tapse continues to decrease until a lower limit, or floor, is reached. During further progression of the disease, rv function may continue to deteriorate via a loss of transverse shortening. This floor effect may also explain the absence of a significant correlation between the change in rvef and change in tapse. However, it has been shown that clinical improvement is associated with improved tapse.16

During the initial one-year of follow-up there was a significant reduction in transverse shortening in non-survivors, which was related to a reduction in rvef (ρ = 0.66, p = 0.005). The transverse shortening is composed of both displacement of the rv free wall and the septum. As for longitudinal shortening, a lower limit or floor was also reached for the rv free wall displacement in the group of non-survivors (Figure 3.5). Thus, increased leftward septal bowing is the main explanation for a further decline in rv transverse shortening in progressive pah.

recently, Marcus et al.17 have demonstrated that maximal leftward septal bowing coin-cides with peak rv myocardial strain. This supports the explanation that just at the mo-ment of peak displacement of the rv free wall, the septum bows maximally to the left and thus the effective transverse shortening is low or even negative (Figure 3.4). Since leftward septal bowing is a consequence of l-r dyssynchrony,18-20 our finding is in line with Lopez-Candales et al.21 who demonstrated in pah patients that the presence of l-r dyssynchrony correlates well with markers of disease severity.

rvfac combines the effect of both longitudinal and transverse shortening in one mea-sure, and therefore also includes septal motion. rvfac obtained both by cmr and echocar-diography has been shown to correlate well with cmr derived rvef.8, 9, 22 This study also showed that there was a good correlation between the change in rvfac and the change in rvef over a one-year period. These results suggest that a decrease in rvfac is an accurate reflection of rv deterioration in patients with severe pah.

Although the change in transverse shortening correlated well with the change in rv func-tion, it has to be noted that other factors may also contribute to rv dysfunction. Among these factors are tricuspid regurgitation and the effect of the lv contraction on rv ejec-tion.23 recently, the importance of the rv outflow tract in the determination of rv function has been also demonstrated.24

3.4.3 Clinical implications An important current challenge in pah is to improve the monitoring of treatment outcomes in clinical practice and clinical trials.25 Since prognosis in pah is mainly determined by rv function, there is a need for simple and reproducible measures of rv function, which reflect rv deterioration over time.

This study shows that both longitudinal measures and rv free wall movements reach a lower limit in non-survivors. Clinically, the lack of improvement in longitudinal and trans-


3

verse shortening of the rv free wall during treatment can be regarded as a sign of end stage rv failure. On the other hand, it is still possible that an improvement of these parameters is associated with improvement in rv function and clinical signs.16, 26

rvfac has the advantage that it quantifies both shortening of the rv free wall and of the septum and was shown to correlate well with global rv function. However, the tracing of the rv endocardial area by 2D echo is difficult, due to the limited range of the echo-window and the trabeculated endocardial surface. This limits the use of this measurement in clinical practice. Another parameter that quantifies leftward septal bowing is the eccentricity index, which might also be very useful.27

3.4.4 LimitationsSome general methodological considerations apply when discussing the findings. In this study, both the rv free wall and septal movements are determined in a 4-chamber view. It should be noted that septum bowing, if present, is less distinct in this view compared to a short-axis view, because the septum is intersected near its posterior attachment to the rv and lv. In addition, the number of patients who met the diagnostic criteria, but did not receive a follow up mri was high. There were a variety of reasons for patients being lost to follow-up and baseline characteristics of these patients were not different from the other patients in this study. We therefore think it is unlikely that the outcome of our study was affected by selection bias.

3.5 CONCLUSIONS

Progressive rv failure in pah is associated with a parallel decline in rv longitudinal and transverse free wall displacement until a floor effect is reached for both. A further reduction of rv function is due to progressive leftward septal displacement. Because transverse short-ening incorporates both free wall and septum displacement, this parameter can be used to monitor the decline of rv function in end-stage pah. ■

References

1. Humbert M, Sitbon O, Simonneau G. Treatment of pulmonary arterial hypertension. N Engl J Med. 2004;351:1425-1436.

2. McLaughlin VV, Presberg KW, Doyle RL, Abman SH, McCrory DC, Fortin T, Ahearn G. Prognosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest. 2004;126:78S-92S.

3. D’Alonzo GE, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, Fishman AP, Goldring RM, Groves BM, Kernis JT, et a. Survival in patients with primary pulmonary hypertension. Results from a national prospective registry. Ann Intern Med. 1991;115:343-349.

4. Ueti OM, Camargo EE, Ueti Ade A, de Lima-Filho EC, Nogueira EA. Assessment of right ventricular function with Doppler echocardiographic indices derived from tricuspid annular motion: comparison with radionuclide angiography. Heart. 2002;88:244-248.

5. Arce OX, Knudson OA, Ellison MC, Baselga P, Ivy DD, Degroff C, Valdes-Cruz L. Longitudinal motion of the atrioventricular annuli in children: reference values, growth related changes, and effects of right ventricular volume and pressure overload. J Am Soc Echocardiogr. 2002;15:906-916.

48 Chapter 3

6. Kaul S, Tei C, Hopkins JM, Shah PM. Assessment of right ventricular function using two-dimensional echocardiography. Am Heart J. 1984;107:526-531.

7. Lindqvist P, Calcutteea A, Henein M. Echocardiography in the assessment of right heart function. Eur J Echocardiogr. 2008;9:225-234.

8. Kind T, Mauritz GJ, Marcus JT, van de Veerdonk M, Westerhof N, Vonk-Noordegraaf A. Right ventricular ejection fraction is better reflected by transverse rather than longitudinal wall motion in pulmonary hypertension. J Cardiovasc Magn Reson. 2010;12:35.

9. Anavekar NS, Gerson D, Skali H, Kwong RY, Yucel EK, Solomon SD. Two-dimensional assessment of right ventricular function: an echocardiographic-MRI correlative study. Echocardiography. 2007;24:452-456.

10. Hinderliter AL, Willis PWt, Barst RJ, Rich S, Rubin LJ, Badesch DB, Groves BM, McGoon MD, Tapson VF, Bourge RC, Brundage BH, Koerner SK, Langleben D, Keller CA, Murali S, Uretsky BF, Koch G, Li S, Clayton LM, Jobsis MM, Blackburn SDJ, Crow JW, Long WA. Effects of long-term infusion of prostacyclin (epoprostenol) on echocardiographic measures of right ventricular structure and function in primary pulmonary hypertension. Primary Pulmonary Hypertension Study Group. Circulation. 1997;95:1479-1486.

11. Barst RJ, McGoon M, Torbicki A, Sitbon O, Krowka MJ, Olschewski H, Gaine S. Diagnosis and differential assessment of pulmonary arterial hypertension. J Am Coll Cardiol. 2004;43:40S-47S.

12. McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hypertension: the impact of epoprostenol therapy. Circulation. 2002;106:1477-1482.

13. Sitbon O, Humbert M, Nunes H, Parent F, Garcia G, Herve P, Rainisio M, Simonneau G. Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: prognostic factors and survival. J Am Coll Cardiol. 2002;40:780-788.

14. Nijveldt R, Germans T, McCann GP, Beek AM, van Rossum AC. Semi-quantitative assessment of right ventricular function in comparison to a 3D volumetric approach: A cardiovascular magnetic resonance study. Eur Radiol. 2008;18:2399-2405.

15. Forfia PR, Fisher MR, Mathai SC, Housten-Harris T, Hemnes AR, Borlaug BA, Chamera E, Corretti MC, Champion HC, Abraham TP, Girgis RE, Hassoun PM. Tricuspid annular displacement predicts survival in pulmonary hypertension. Am J Respir Crit Care Med. 2006;174:1034-1041.

16. Brown SB, Raina A, Katz D, Szerlip M, Wiegers SE, Forfia PR. Longitudinal Shortening Accounts for the Majority of Right Ventricular Contraction in Normal Subjects and in Pulmonary Arterial Hypertension and Improves After Pulmonary Vasodilator Therapy. Chest. 2010

17. Marcus JT, Gan CT, Zwanenburg JJ, Boonstra A, Allaart CP, Gotte MJ, Vonk-Noordegraaf A. Interventricular mechanical asynchrony in pulmonary arterial hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol. 2008;51:750-757.

18. Kalogeropoulos AP, Georgiopoulou VV, Howell S, Pernetz MA, Fisher MR, Lerakis S, Martin RP. Evaluation of right intraventricular dyssynchrony by two-dimensional strain echocardiography in patients with pulmonary arterial hypertension. J Am Soc Echocardiogr. 2008;21:1028-1034.

19. Lopez-Candales A, Rajagopalan N, Dohi K, Edelman K, Gulyasy B. Normal range of mechanical variables in pulmonary hypertension: a tissue Doppler imaging study. Echocardiography. 2008;25:864-872.

20. Lopez-Candales A, Dohi K, Bazaz R, Edelman K. Relation of right ventricular free wall mechanical delay to right ventricular dysfunction as determined by tissue Doppler imaging. Am J Cardiol. 2005;96:602-606.

21. Lopez-Candales A, Dohi K, Rajagopalan N, Suffoletto M, Murali S, Gorcsan J, Edelman K. Right ventricular dyssynchrony in patients with pulmonary hypertension is associated with disease severity and functional class. Cardiovasc Ultrasound. 2005;3:23.

22. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise J, Solomon S, Spencer KT, St John Sutton M, Stewart W. Recommendations for chamber quantification. Eur J Echocardiogr. 2006;7:79-108.

23. Rushmer RF. Cardiovascular dynamics. Philadelphia: W.B. Saunders Company; 1976.24. Calcutteea A, Chung R, Lindqvist P, Hodson M, Henein MY. Differential right ventricular regional function

and the effect of pulmonary hypertension: three-dimensional echo study. Heart. 201125. McLaughlin VV, Suissa S. Prognosis of pulmonary arterial hypertension: the power of clinical registries of

rare diseases. Circulation. 2010;122:106-108.


3

26. Basil A, Liu T, Arkles J, Opotowsky A, Taichman D, Palevsky HI, Forfia PR. Serial Changes in TAPSE Correlate with Functional Assessment and Clinical Events in Patients with PAH. Am J Respir Crit Care Med. 2009;179:A4882.

27. Raymond RJ, Hinderliter AL, Willis PW, Ralph D, Caldwell EJ, Williams W, Ettinger NA, Hill NS, Summer WR, de Boisblanc B, Schwartz T, Koch G, Clayton LM, Jobsis MM, Crow JW, Long W. Echocardiographic predictors of adverse outcomes in primary pulmonary hypertension. J Am Coll Cardiol. 2002;39:1214-1219.

C H A P T E R 4

51

ABSTRACT Background In pulmonary arterial hypertension (pah) a prolonged time interval between pulmonary valve closure and tricuspid valve opening is found. This period is interpreted as prolonged right ven-tricular (rv) relaxation, and thus a reflection of diastolic dysfunction. This concept recently has been questioned, since it was shown that rv contraction continues after pulmonary valve closure causing a post-systolic contraction period.Objectives To investigate in pah whether the increased rv post-sys-tolic isovolumic period is caused by either an additional post-systolic contraction period, or an increased relaxation period (diastolic dys-function).Methods Twenty three patients with pah (mean pulmonary arterial pressure 54 ± 12 mmHg), and 18 healthy subjects were studied us-ing cardiac mri. In a rv two-chamber view, times of pulmonary valve closure (tpvc) and tricuspid valve opening (ttvo) were measured, defining the total post-systolic isovolumic period. Time to peak of rv free wall contraction (TpeakrV) was determined with myocardial tag-ging. Post-systolic contraction and relaxation periods were defined as the time intervals between tpvc and TpeakrV and between TpeakrV and ttvo, respectively. These periods were normalised to an rr-interval.Results The total post-systolic isovolumic period was longer in pa-tients than in healthy subjects (0.15 ± 0.04 vs. 0.04 ± 0.02, p < 0.001), but the relaxation period was not different (0.06 ± 0.02 vs 0.05 ± 0.02, p = 0.09). The post-systolic contraction period in patients was strongly related to the total post-systolic isovolumic period (y = 0.98 x – 0.05; ρ = 0.89, p < 0.001), and was associated with disease severity.Conclusion In pah, the prolonged post-systolic isovolumic period is caused by an additional post-systolic contraction period, rather than by an increased relaxation period. ◦

Heart 2011 Mar; 97(6):473-8

Departments of aPulmonary Diseases, bPhysics and Medical Technology and cPhysiology, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam

G.J. Mauritz,a J.T. Marcus,b N. Westerhof,a,c P.E. Postmus,a A. Vonk-Noordegraafa

Prolonged right ventricular post-systolic isovolumic period in pulmonary arterial hypertension is not a reflection of diastolic dysfunction

52 Chapter 4

4.1 INTRODUCTION

IN PULMONAry ArTErIAL HyPErTENSION (pah), a prolonged post-systolic iso-volumic time interval, here called post-systolic isovolumic period, between pulmonary valve closure and tricuspid valve opening is seen.1-4 Several reports have shown that this

post-systolic isovolumic period is related to disease severity,5,6 and is assumed to mirror prolonged right ventricular (rv) isovolumic relaxation3 and therefore interpreted as a re-flection of rv diastolic dysfunction.2,7-10 However, Marcus et al.11 demonstrated that in pah the shortening (contraction) of the rv free wall continues after pulmonary valve closure, as an indication of a post-systolic (or post-ejection) contraction (see Figure 4.1).

Consequently, this raises the question whether the current interpretation of the increased post-systolic isovolumic period as prolonged rv relaxation (diastolic dysfunction) is valid in pah. Alternatively, a post-systolic contraction period followed by a normal isovolumic relaxation period might explain this prolonged time interval. Therefore, the aim of this study was to explore in pah whether the increased post-systolic isovolumic period is caused by either a post-systolic contraction, or an increased relaxation period (diastolic dysfunc-tion), or both. To answer this question a group of patients with pah of different disease severity were studied using cardiac mri.

4.2 METHODS

4.2.1 Study population This study was conducted within the Pulmonary Hypertension Program at the vu Univer-sity Medical Center. The institutional review board approved the conduct of this study and all patients gave informed consent before enrolment. We recruited 23 consecutive patients referred for right heart catheterisation in the follow-up of their pah. All patients also un-derwent cardiovascular magnetic resonance (cmr) imaging as part of their disease evalua-tion. A mean pulmonary artery pressure (pap) of > 25 mm Hg with a pulmonary capillary wedge pressure of < 15 mm Hg was considered to be pah.12, 13 Patients with any form of car-diopulmonary disease and those who were unable to undergo cmr were excluded. results were compared with 18 age- and gender-matched, non-smoking controls subjects, without a history of cardiopulmonary diseases.

Pulmonary valve closure Tricuspid valve opening

Ejection

Peak RV contraction

Post-systolic isovolumic periodFilling

Contraction Relaxation

Figure 4�.1 Post-systolic isovolumic period in PAH, consisting of a contraction and a relaxation period.

53Prolonged RV post-systolic isovolumic period

4

4.2.2 CMR protocol cmr imaging was performed on a 1.5T scanner (Magnetom Sonata, Siemens, Erlangen, Germany). Complementary tagged (cspamm) myocardial images with high temporal reso-lution (14 ms) were acquired in all patients using steady-state free precession (ssfp) imag-ing and a multiple brief expiration breath-hold scheme as described by Zwanenburg et al.14 Parameters were: three phase encoding lines/beat, repetition time 4.7 ms, echo time 2.3 ms, no view sharing, flip angle 20º, voxel size 1.2 × 3.8 × 6.0 mm3. Images for two-dimensional strain analysis were acquired in the mid-ventricular short-axis planes. With these tagging cine images, contraction and relaxation within the myocardial wall can be assessed. In pa-tients and control subjects ssfp cine imaging (without tagging) was performed with full coverage of the left ventricle and right ventricle (stack of short-axis slices) to assess ven-tricular volumes and ejection fraction (ef). In addition, to determine the moments of the tricuspid and pulmonary valve closures, a long-axis cine image in the rv two-chamber view was acquired. For this purpose, ssfp imaging with view sharing was used to obtain a temporal resolution of 15 ms in a single breath-hold.

4.2.3 CMR data analysis The tagged images were analysed with the harmonic phase procedure.15 Circumferential shortening, assumed to reflect contraction, was calculated as a function of time over the cardiac cycle. For the left ventricular (lv) and rv free wall the strains and strain timing pa-rameters were derived as described previously.11 For the whole rv and lv free wall at mid-level, the time to peak of rv circumferential shortening (TpeakrV) was calculated related to the ecg r-wave by automated routines.14 The times to pulmonary valve closure (tpvc) and tricuspid valve opening (ttvo) were assessed from rv two-chamber cine images. In three patients pulmonary valve timing was derived from the most basal short-axis cine that showed the valves during the last part of systole. The stack of short-axis cine images was used for the calculation of the lv and rv end-diastolic volumes and the rv end-systolic volume. Since stroke volume (sv) from rv volumes has limited accuracy,16 lv volumes were used to calculate sv. rvef was calculated using rv end-diastolic volumes and lv svs. The time to maximal leftward septal bowing was measured at the most basal short-axis cine slice that still showed the lv and rv myocardium through the cardiac cycle.

4.2.4 Definitions of cardiac phases In patients and healthy controls the post-systolic isovolumic period was defined as the time interval between tpvc and ttvo. In the healthy subjects this time interval was designated as isovolumic relaxation period.17 In the patients this post-systolic isovolumic period was found to consist of a contraction period and a relaxation period (Figure 4.1). The isovolumic contraction period was defined as the time interval between tpvc (end systole) and TpeakrV. The rv isovolumic relaxation period was defined as the time interval between TpeakrV and ttvo. To minimise the influence of the heart rate, all periods were normalised to the heart period, the r-to-r interval.

54 Chapter 4

4.2.5 Statistical analysis results are expressed as mean ± sd. Statistical significance was set at a value of p < 0.05. Comparisons between patients and control subjects were made with unpaired t-tests, with-out correction for multiple comparisons. The relations between the post-systolic contrac-tion period versus the post-systolic isovolumic period and the post-systolic relaxation period versus the post-systolic isovolumic period were tested by linear regression. Linear correlations between the post-systolic contraction period and pulmonary vascular resis-tance (pvr), systolic pap, stroke volume index and rvef, were also calculated. The inter-observer variation was determined for the time to pulmonary valve closure and time to tricuspid valve opening by BlandeAltman analysis, for a subset of 10 patients. All statistical analyses were performed with spss Statistics 15.0 (spss Inc.)

4.3 RESULTS

4.3.1 Patients’ characteristicsThere were no differences between the 23 patients with pah and 18 controls with respect to age (pah 43 ± 14 vs. controls 38 ± 9 years) and proportion of male to female subjects (pah 6 : 17 vs. controls 6 : 12, Fisher’s exact test, p = 0.8). Patients’ characteristics and hemodynamic variables are presented in Table 4.1. On the basis of the ecg morphology, three patients had an incomplete, and two patients a complete, right bundle branch block (rbbb). Seventeen patients were diagnosed as having idiopathic pah, whereas six had chronic thromboembo-lic pah. The majority of patients were in New york Heart Association functional class iii. Hemodynamics yielded characteristics of rv pressure overload. mri data of healthy con-trols and patients with pah are shown in Table 4.2.

4.3.2 Images, strains and valves timingFigure 4.2 shows in a patient with pah, rv two-chamber images at the time of pulmonary valve closure and at the time of tricuspid valve opening, a short-axis cine image at TpeakrV and the circumferential strain curves during the cardiac cycle for the lv and rv free walls. The contraction of the left and right ventricles starts simultaneously, but the right ventricle reaches its peak later than the left ventricle. As can be seen, rv contraction continues after tpvc (post-systolic contraction) and reaches its peak before ttvo, dividing the post-sys-tolic isovolumic period into a contraction period and a relaxation period. The post-systolic isovolumic period divided by rr interval in the patients was longer than in the control sub-jects (0.15 ± 0.04 vs 0.04 ± 0.017, p < 0.001). However, the time to pulmonary valve closure, divided by rr interval, was not earlier in the patients (Table 4.2).

4.3.3 Relation of contraction and relaxation with the post-systolic isovolumic periodAs shown in Figure 4.3, in the patients with pah a significant relation exists between the rr-normalised post-systolic contraction period and the post-systolic isovolumic period (ρ = 0.89, y = 0.98x – 0.05, p = 0.0001), but no significant relation is found for the relaxation pe-riod and the post-systolic isovolumic period (ρ = 0.05, p = 0.82). Additionally, in Table 4.2


4

and Figure 4.4, a comparison of the relaxation period between patients and healthy controls (0.057 ± 0.018 and, 0.047 ± 0.017, respectively) shows no significant difference (p = 0.09).

4.3.4 Regression analysis of the post-systolic contraction period and disease severityAs shown in Figure 4.5, there was an association between the post-systolic contraction pe-riod and pvr (p < 0.001, ρ = 0.75) and with systolic pap (p = 0.002, ρ = 0.61). Furthermore, the post-systolic contraction period was negatively related to stroke volume index (ρ = –0.46, p = 0.03) and rvef (ρ = –0.58, p = 0.003). None of these associations was found for the post-systolic relaxation period.

Table 4�.1 Characteristics and hemodynamic variables of the PAH-patients


Age (y) 43 ± 14

Gender (m/f) 6/17

Functional status

NYHA Class

II 3

III 19

IV 1

6-minute walking distance (m) 411 ± 107

Medication use

Intravenous prostacyclin 10

Endothelin receptor antagonist 14

Sildenafil 8

Calcium antagonist 2

Hemodynamics

Mean RAP (mmHg) 8 ± 4

Systolic PAP (mmHg) 88 ± 19

Diastolic PAP (mmHg) 33 ± 9

Mean PAP (mmHg) 54 ± 12

PCWP (mmHg) 5 ± 3

PVR (dyne · s · cm-5) 957 ± 354

HR (bpm) 84 ± 16

CI (L/min/m2) 2.5 ± 0.8

SVI (ml/m2) 31 ± 12

SaO2 (%) 94 ± 2

SvO2 (%) 64 ± 7

Values expressed as mean ± SD. NYHA = New York Heart Association functional class, RAP = right atrial pres-sure, PAP = pulmonary artery pressure, SvO2 = mixed venous oxygen saturation, SaO2 = arterial oxygen satura-tion, CI = cardiac index, PVR = pulmonary vascular resistance, HR = heart rate.

56 Chapter 4

Table 4�.2 RV volumentric data and timing parameters in controls

Control(n = 18)

PAH(n = 23) p-value

RVEDVI (ml/m2) 76 ± 22 110 ± 37 < 0.001

RVMI (g/m2) 22 ± 4.2 49 ± 27 < 0.001

RVEF (%) 54 ± 9 29 ± 12 < 0.001

SVI (ml/m2) 43 ± 14 30 ± 9 < 0.001

RR (ms) 914 ± 127 773 ± 140 0.003

Time to pulmonary valve closure (ms) 356 ± 40 319 ± 40 0.006

Time to pulmonary valve closure / RR 0.39 ± 0.05 0.42 ± 0.05 0.13

Time to tricuspid valve opening (ms) 388 ± 42 431 ± 47 0.025

Time to tricuspid valve opening / RR 0.44 ± 0.04 0.57 ± 0.06 < 0.001

Post-systolic isovolumic period (ms) 36 ± 15 112 ± 25 < 0.001

Post-systolic isovolumic period / RR 0.047 ± 0.02 0.15 ± 0.04 < 0.001

Post-systolic contraction period (ms) n.m. 69 ± 29

Post-systolic contraction period /RR n.m. 0.09 ± 0.04

Isovolumic relaxation period (ms) 36 ± 15 43 ± 14 0.38

Isovolumic relaxation period / RR 0.047 ± 0.02 0.057 ± 0.02 0.09

Values expressed as mean ± SD. RVEDVI = RV end-diastolic volume index, RVMI = RV mass index, SVI = stroke volume index, RR = heart period, n.m. = not measured at the intensive care unit.

Figure 4�.2 MR images of the RV at pulmonary valve closure (left), tricuspid valve opening (right) and the short-axis tagged image (middle) at TpeakRV and RV and LV strain curves. The RV strain curve shows that the RV continues active contraction after pulmonary valve closure during 58 ms, while isovolumic relaxation following RV peak contraction continues for 32 ms. The R-R interval is 767 ms.

Time (ms)

Circ

umfe

rent

ial

stra

in (

%)

100 200 300 500 600 700 8000

–2

2

–4

–6

–8

–10

–12

LVRV

Post-systolic isovolumic period

Pulmonary valve closure Tricuspid valve openingPeak RV contraction

RV

t = 341 ms t = 431 msTpeak RV = 399 ms

RV

LV

RA


4

0.10 0.15 0.20 0.250.00

0.05

0.10

0.15

0.20

Post-systolic isovolumic period / RR

Cont

ract

ion

peri

od /

RR

Post-systolic isovolumic period / RR

A B

0.00

0.05

0.10

0.15

0.20

Rela

xatio

n pe

riod

/ RR

0.10 0.15 0.20 0.25

0.00

0.05

0.10

0.15

0.20

0.25

Isov

olum

ic p

erio

d / R

R

Control PAH

Contraction period

Relaxationperiod

Contraction period

Relaxationperiod

n.s.

0 500 1000 1500 20000.00

0.05

0.10

0.15

ρ = 0.75, p = 0.0001 ρ = 0.61, p = 0.002Post

-sys

tolic

con

trac

tion

pe

riod

/ RR

(%

)

0 35 70 105 140

Systolic pulmonary artery pressure (mmHg)Pulmonary vascular resistance (dyn . s . cm-5)

A B

Figure 4�.3� Correlation between the post-systolic contraction period and the post-systolic isovolumic period (A), p < 0.001, ρ = 0.89. No correlation exist between the relaxation period and the post-systolic isovolumic period (B), p = 0.81, ρ = 0.05. All periods are normalised for the R-R interval (RR). Results in two patients with pulmonary arterial hypertension with a complete right bundle branch block are highlighted.

Figure 4�.4� Post-systolic isovolumic periods in PAH patients and control subjects. The only difference between patients and control subjects is the post-systolic isovolumic contraction period, which is present in the patients but absent in the controls. No significant (n.s.) difference was found in post-systolic isovolumic relaxation period between PAH patients and control subjects (p = 0.09).

Figure 4�.5� Post-systolic contraction time normalized for the R-R interval versus pulmonary vascular resistance (A) and systolic pulmonary artery pressure (B).

58 Chapter 4

4.3.5 Reader agreementThe interobserver variation in the time to pulmonary valve closure was given by a corre-lation coefficient of 0.87 with p < 0.001, and a bias of 9 ms with 95% confidence limits of agreement of –25 and 45 ms, respectively. For the time to tricuspid valve opening, interob-server variation was given by a correlation coefficient of 0.89 with p < 0.001, and a bias of 4 ms with 95% confidence limits of agreement of –37 and 44 ms, respectively.

4.4 DISCUSSION

4.4.1 General discussionIn normal physiology, the period between pulmonary valve closure and tricuspid valve opening represents the rv isovolumic relaxation period only.17,18 Until now, prolongation of this period has been interpreted as rv diastolic dysfunction.4, 10, 19-21 In this study the time of tricuspid valve opening was measured, which made it possible to analyse two separate time intervals in the rv post-systolic isovolumic period: (a) the postsystolic rv contraction time, between pulmonary valve closure and time to peak strain, and (b) the true isovolu-mic relaxation time, between time to peak strain and tricuspid valve opening. Thus, in pah the rv post-systolic isovolumic period cannot be interpreted as a measure of rv diastolic function only, and thus should not be labelled as an isovolumic relaxation time. Second, the post-systolic rv contraction is associated with pap and pvr. And finally, the true rv iso-volumic relaxation time is not associated with pap and pvr, and not prolonged. Therefore in patients with pah the increased post-systolic isovolumic period results from prolonged rv contraction duration rather than slower relaxation.

4.4.2 Integral conceptFigure 4.6 shows a pathophysiological concept, integrating our observations with previous findings in pulmonary hypertension. rv contraction starts simultaneously with lv con-traction. Then follows the ejection period until pulmonary valve closure. After pulmonary valve closure (defined as end systole), rv contraction continues with a post-systolic con-traction. This prolonged rv contraction, still continuing after the peak of lv contraction11,

22 (see also Figure 4.2) leads to a reversal of the trans-septal pressure gradient, which causes the post-systolic leftward bowing of the septum into the left ventricle and thereby impairs lv filling.20, 23, 24 Thus the post-systolic isovolumic period starts with an abnormal contrac-tion period, during which the rv contraction energy is wasted in the non-functional sep-tum bowing, which causes inefficiency in both rv systole and lv diastole.25 Only after peak rv contraction, which coincides with maximum leftward septum bowing, does a normal rv relaxation period begin.

4.4.3 Post-systolic contraction period and disease severity Previous findings have shown that in pulmonary hypertension prolongation of the post-systolic isovolumic period, is related to pvr, pap systolic and rvef.2, 4, 5 These findings are in line with our results. However, we show here that the interpretation needs revision. Based


4

on our results, it is the post-systolic contraction period that depends on disease severity (Figure 4.5), in contrast to the isovolumic relaxation period, which appears to be disease in-dependent. The mechanism of the prolonged contraction period is probably the increased afterload that is, higher systolic pap (Figure 4.5). Increased contraction duration with in-creased load has been shown in whole heart26, 27 and cardiac muscle28, 29 studies. At the level of the sarcomeres, an increased load leads to prolonged time of shortening, as found in iso-lated cardiac trabeculae.30 At the organ level of the right ventricle, the prolonged shortening with constant rv volume is made possible by the early pressure decline in the left ventricle, and the subsequent leftward septal bowing.11

4.4.4 Clinical implicationsAs mentioned above, the observed increase in the post-systolic isovolumic period is a re-flection of increased rv contraction period. A prolongation of this period is a manifestation of the inefficient post-systolic rv contraction which is an important sign of rv pressure overload in pah. This is in contrast with the current concept so far that this period is com-pletely a relaxation period. This study therefore shows that in the right ventricle of patients with pah, this prolonged isovolumic period is not a sign of diastolic dysfunction. Because of the twofold disadvantageous effect of prolonged rv shortening both on rv and lv function, treatment could focus on reduction of the prolonged rv post-systolic contraction. Dual pacing could be attempted to reduce the l-r delay in peak contraction, as shown earlier in an animal model of PAH31 and by a computer simulation analysis.32 Another option is to

0

60

30

90

120

Pres

sure

(m

mH

g)

Time (s)0.4 0.80 0.2 0.6

Ejection Filling

Pulm valveclosure

Tric valveopening

Peakcontraction

Isovol periodContr Relax

RV

Septum

Aortic valveclosure

Ejection FillingRelaxLV

Figure 4�.6 Integrated concept of the post-systolic isovolumic period in pulmonary arterial hypertension. For the RV, the post-systolic isovolumic period starts at pulmonary valve closure and ends at tricuspid valve opening. As shown in Figure 4.2, this period starts with an RV contraction period, from pulmonary valve closure until peak RV contraction. This continued RV contraction is possible, without pressure increase, due to the leftward septal bowing. At the peak of RV contraction, the right-left pressure difference is maximal and at the same time the septum bows maximally to the left. Subsequently, the isovolumic relaxation period begins.

60 Chapter 4

reduce the imbalance in contraction duration between the chambers with pharmacologi-cal methods that limit the variability and load dependence of contraction. Such treatments may improve both rv systolic efficiency and lv diastolic filling.25, 33 How can we measure the rv post-systolic isovolumic period in daily practice, and thereby the effect of any treat-ment? It is the time interval between pulmonary valve closure and tricuspid valve opening, which can easily be measured by echocardiography.1, 3-6, 17 Therefore, measurement of this period provides an easy tool for monitoring individual patients during treatment.

4.4.5 Study limitationsWe acknowledge some limitations of our study. The temporal resolution of 15 ms limits the accuracy of the timing of valves and peak strain. Furthermore, the tagged myocardial cine images and the rv two-chamber cine images could not be acquired simultaneously. How-ever, the acquisitions of both cine images were obtained within the same mri session, with no discernible difference in cardiac conditions, such as heart rate. The lack of longitudinal and radial strain is a limitation. However, it is unlikely that these other strain components would have a different time to peak value. Finally, patients with and without targeted medi-cation were included, nevertheless, a prolonged contraction was found for all patients.

4.5 CONCLUSIONS

In pah, the time interval between pulmonary valve closure and tricuspid valve opening (called post-systolic isovolumic period) is increased and consists of a contraction period and a relaxation period. The increase in post-systolic isovolumic period is only caused by increased rv contraction duration, and is not a reflection of diastolic dysfunction. ■

References

1. Brechot N, Gambotti L, Lafitte S, Roudaut R. Usefulness of right ventricular isovolumic relaxation time in predicting systolic pulmonary artery pressure. Eur J Echocardiogr 2008 July;9(4):547-54.

2. Gan CT, Holverda S, Marcus JT et al. Right ventricular diastolic dysfunction and the acute effects of sildenafil in pulmonary hypertension patients. Chest 2007 July;132(1):11-7.

3. Lindqvist P, Waldenstrom A, Wikstrom G, Kazzam E. Right ventricular myocardial isovolumic relaxation time and pulmonary pressure. Clin Physiol Funct Imaging 2006 January;26 (1):1-8

4. Dambrauskaite V, Delcroix M, Claus et al. The evaluation of pulmonary hypertension using right ventricular myocardial isovolumic relaxation time. J Am Soc Echocardiogr 2005 November;18(11):1113-20

5. Fahmy EM, Abdelraouf DA. Right ventricular myocardial isovolumic relaxation time as novel method for evaluation of pulmonary hypertension: correlation with endothelin-1 levels. J Am Soc Echocardiogr 2007 May;20(5):462-9

6. Yeo TC, Dujardin KS, Tei C, Mahoney DW, McGoon MD, Seward JB. Value of a Doppler-derived index combining systolic and diastolic time intervals in predicting outcome in primary pulmonary hypertension.Am J Cardiol 1998 May 1;81(9):1157-61

7. Emilsson K, Loiske K. Isovolumetric relaxation time of the right ventricle assessed by tissue Doppler imaging. Scand Cardiovasc J 2004 October;38(5):278-82


4

8. Boissiere J, Gautier M, Machet MC, Hanton G, Bonnet P, Eder V. Doppler tissue imaging in assessment of pulmonary hypertension-induced right ventricle dysfunction. Am J Physiol Heart Circ Physiol 2005 December;289(6):H2450-H2455

9. Lakoumentas JA, Panou FK, Kotseroglou VK, Aggeli KI, Harbis PK. The Tei index of myocardial performance: applications in cardiology. Hellenic J Cardiol 2005 January;46(1):52-8.

10. Lindqvist P, Caidahl K, Neuman-Andersen G et al. Disturbed right ventricular diastolic function in patients with systemic sclerosis: a Doppler tissue imaging study. Chest 2005 August;128(2):755-63.

11. Yu CM, Sanderson JE, Chan S et al. Right ventricular diastolic dysfunction in heart failure. Circulation 1996 Apr 15; 93(8):1509-14.

12. Marcus JT, Gan CT, Zwanenburg JJ et al. Interventricular mechanical asynchrony in pulmonary arterial hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol 2008 February 19;51(7):750-7.

13. Galiè N, Torbicki A, Barst R et al. Task Force. Guidelines on diagnosis and treatment of pulmonary arterial hypertension. The Task Force on Diagnosis and Treatment of Pulmonary Arterial Hypertension of the European Society of Cardiology. Eur Heart J 2004 December;25(24):2243-78.

14. Barst RJ, McGoon M, Torbicki A, Sitbon O, Krowka MJ, Olschewski H, Gaine S.. Diagnosis and differential assessment of pulmonary arterial hypertension. J Am Coll Cardiol 2004 June 16;43(12 Suppl S):40S-7S.

15. Zwanenburg JJ, Gotte MJ, Kuijer JP, Heethaar RM, van Rossum AC, Marcus JT. Timing of cardiac contraction in humans mapped by high-temporal-resolution MRI tagging: early onset and late peak of shortening in lateral wall. Am J Physiol Heart Circ Physiol 2004 May;286(5):H1872-H1880.

16. Osman NF, Kerwin WS, McVeigh ER, Prince JL. Cardiac motion tracking using CINE harmonic phase (HARP) magnetic resonance imaging. Magn Reson Med 1999 December;42(6):1048-60.

17. Zwanenburg JJ, Gotte MJ, Kuijer JP, Heethaar RM, van Rossum AC, Marcus JT. Timing of cardiac contraction in humans mapped by high-temporal-resolution MRI tagging: early onset and late peak of shortening in lateral wall. Am J Physiol Heart Circ Physiol 2004 May;286(5):H1872-H1880.

18. Mauritz GJ, Marcus JT, Boonstra A, Postmus PE, Westerhof N, Vonk-Noordegraaf A. Non-invasive stroke volume assessment in patients with pulmonary arterial hypertension: left-sided data mandatory. J Cardiovasc Magn Reson 2008;10(1):51.

19. Burstin L. Determination of pressure in the pulmonary artery by external graphic recordings. Br Heart J 1967 May;29(3):396-404.

20. Tei C, Dujardin KS, Hodge DO et al. Doppler echocardiographic index for assessment of global right ventricular function. J Am Soc Echocardiogr 1996 November;9(6):838-47.

21. Stojnic BB, Brecker SJ, Xiao HB et al. Left ventricular filling characteristics in pulmonary hypertension: a new mode of ventricular interaction. Br Heart J 1992 July;68(1):16-20.

22. Marangoni S, Scalvini S, Schena M, Vitacca M, Quadri A, Levi G. Right ventricular diastolic function in chronic obstructive lung disease. Eur Respir J 1992 April;5(4):438-43.

23. López-Candales A, Dohi K, Rajagopalan N et al. Right ventricular dyssynchrony in patients with pulmonary hypertension is associated with disease severity and functional class. Cardiovasc Ultrasound 2005;3:23.

24. Gan CT, Lankhaar JW, Marcus JT et al. Impaired left ventricular filling due to right-to-left ventricular interaction in patients with pulmonary arterial hypertension. Am J Physiol Heart Circ Physiol 2006 April;290(4):H1528-H1533.

25. Vonk-Noordegraaf A, Marcus JT, Gan CT, Boonstra A, Postmus PE. Interventricular mechanical asynchrony due to right ventricular pressure overload in pulmonary hypertension plays an important role in impaired left ventricular filling. Chest 2005 December;128(6 Suppl):628S-30S.

26. Beyar R. Heart inefficiency in pulmonary hypertension: a double jeopardy. J Am Coll Cardiol 2008 February 19;51(7):758-9.

27. Fahmy EM, Abdelraouf DA. Right ventricular myocardial isovolumic relaxation time as novel method for evaluation of pulmonary hypertension: correlation with endothelin-1 levels. J Am Soc Echocardiogr 2007 May;20(5):462-9.

28. Kind T, Westerhof N, Faes TJ, Lankhaar JW, Steendijk P, Vonk-Noordegraaf A. Cardiac phase-dependent time normalization reduces load dependence of time-varying elastance. Am J Physiol Heart Circ Physiol 2009 February;296(2):H342-H349.

62 Chapter 4

29. Hunter WC. End-systolic pressure as a balance between opposing effects of ejection. Circ Res 1989 February;64(2):265-75.

30. McCrossan ZA, Billeter R, White E. Transmural changes in size, contractile and electrical properties of SHR left ventricular myocytes during compensated hypertrophy. Cardiovasc Res 2004 August 1;63(2):283-92.

31. Brooksby P, Levi AJ, Jones JV. Investigation of the mechanisms underlying the increased contraction of hypertrophied ventricular myocytes isolated from the spontaneously hypertensive rat. Cardiovasc Res 1993 July;27(7):1268-77.

32. Handoko ML, Lamberts RR, Redout EM et al.. Right ventricular pacing improves right heart function in experimental pulmonary arterial hypertension: a study in the isolated heart Am J Physiol Heart Circ Physiol 2009 September;296(2):H1522-H1539.

33. Lurz P, Puranik R, Nordmeyer J et al. Improvement in left ventricular filling properties after relief of right ventricle to pulmonary artery conduit obstruction: contribution of septal motion and interventricular mechanical delay. Eur Heart J 2009 June 26.


4

C H A P T E R 5

65

ABSTRACT Background Interventricular mechanical asynchrony is a characteris-tic of pulmonary hypertension. We studied the role of right ventricular (rv) wall stress in the recovery of interventricular asynchrony, after pulmonary endarterectomy (pea) in chronic thromboembolic pulmo-nary hypertension (cteph).Methods In 13 consecutive patients with cteph, before and 6 months after pulmonary endarterectomy, magnetic resonance imaging myo-cardial tagging was applied. For the left ventricular (lv) and rv free walls, the time to peak (Tpeak) of circumferential shortening (strain) was calculated. Pulmonary artery pressure (pap) was measured by right heart catheterization within 48 hours of pea. Then the rv free wall systolic wall stress was calculated by the Laplace law. Results After pea, the left to right free wall delay (l-r delay) in Tpeak strain decreased from 97 ± 49 ms to –4 ± 51 ms (p < 0.001), which was not different from normal reference values of –35 ± 10 ms (p = 0.18). The rv wall stress decreased significantly from 15.2 ± 6.4 kPa to 5.7 ± 3.4 kPa (p < 0.001), which was not different from normal reference val-ues of 5.3 ± 1.39 kPa (p = 0.78). The reduction of l-r delay in Tpeak was more strongly associated with the reduction in rv wall stress (ρ = 0.69, p = 0.007) than with the reduction in systolic pap (ρ = 0.53, p = 0.07). The reduction of l-r delay in Tpeak was not associated with estimates of the reduction in rv radius (ρ = 0.37, p = 0.21) or increase in rv systolic wall thickness (ρ = 0.19, p = 0.53). Conclusion After pea for cteph, the rv and lv peak strains are resyn-chronized. The reduction in systolic rv wall stress plays a key role in this resynchronization. ◦

J Cardiovasc Magn Reson (under review)

Departments of aPulmonary Diseases, bPhysiology and cPhysics and Medical Technology, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam; departments of dPulmonary Diseases and eCardiothoracic Surgery, Academic Medical Center, Amsterdam; department of fRespiratory Medicine, Onze Lieve Vrouwe Gasthuis, Amsterdam

G.J. Mauritz,a A. Vonk-Noordegraaf,a T. Kind,a S. Surie,d J.J. Kloek,e P. Bresser,f N. Saouti,a N. Westerhof,a,b J.T. Marcusc

Pulmonary endarterectomy normalizes interventricular asynchrony and right ventricular systolic wall stress

66 Chapter 5

5.1 INTRODUCTION

INTErVENTrICULAr MECHANICAL ASyNCHrONy is a characteristic of right ventricular (rv) pressure overload.1-3 It occurs at the end of rv myocardial shortening, when the rv free wall continues shortening while the left ventricular (lv) wall is already

in its early diastolic phase.4-7 Consequently, the ventricular septum bows to the left, and the rv shortens without ejection thereby making the rv very inefficient,8 and in addition impairing early lv filling.9, 10 The underlying mechanism of this prolonged rv contraction duration is unknown. In an isolated Langendorf-perfused heart, Handoko et al.11 created a l-r asynchrony in peak pressure by increasing rv pressure using inflation of a balloon. Two earlier studies have found a relation between l-r asynchrony and wall stress.5,12 Since wall stress is the combined effect of pressure, volume and wall thickness, the questions remains whether asynchrony is best explained by rv pressure or the combination of the variables as expressed by wall stress. This insight is relevant for a better understanding of the adaptation mechanisms of rv structure in the presence of right ventricular overload.

The aim of the present study is twofold. First to assess the effect of rv unloading on asynchrony in chronic thrombo-embolic pulmonary hypertension (cteph). Secondly, to separate the effects on asynchrony induced by pressure, volume, wall thickness and wall stress. We assessed these effects on l-r asynchrony in cteph patients, before and after pul-monary endarterectomy.

5.2 METHODS

5.2.1 Patients populationThirteen of 17 consecutive patients with surgically accessible cteph, referred to the Aca-demic Medical Center of the University of Amsterdam, were prospectively studied before and after pulmonary endarterectomy (pea). One patient refused to participate because of claustrophobia, one patient died postoperatively and two patients refused to undergo a second mri after surgery. Diagnosis of cteph and eligibility for pea were established on the basis of previously reported procedures and criteria.13 Diagnosis and cardiopulmonary he-modynamics were determined by pulmonary angiography and right heart catheterization. Coronary angiography was routinely performed in all patients older than 50 years of age, and in patients older than 40 years of age if they had a history of smoking.

In addition, eight healthy subjects (called ‘control’) were included (age 55 ± 6 years, 3 women), with normal electrocardiogram (ecg) and qrs width of 80 ± 12 ms, where rv and lv wall strains were obtained and compared with the cteph group. For the estimation of normal rv wall stress, we included also 8 patients (called ‘non PH’) suspected of having ph (age 59 ± 11 years, 5 women) but with normal right-sided pressures confirmed by right heart catheterisation. All patients and controls gave informed consent to the study proto-col, which was approved by the the institutional review board of the vu University Medical Center.

67PEA normalizes asynchrony and RV wall stress

5

5.2.2 Right heart catheterization All patients underwent right heart catheterization within 48 h of their pre-operative mri. right heart catheterisation gave right atrial pressure, pulmonary artery pressure (pap), pul-monary capillary wedge pressure (pcwp), and cardiac output (thermodilution). Pulmonary vascular resistance (pvr) was calculated as: pvr = 80 · (mean pap − mean pcwp) / cardiac output. Postoperative hemodynamic measurements were repeated on the first or second day following pea, before removal of the Swan-Ganz catheter (Edwards LifeSciences, Ir-vine, ca, usa).

5.2.3 CMR Imaging acquisitionAll patients underwent mri myocardial tagging at baseline before, and at least 6 months after endarterectomy. A 1.5 Siemens Avanto whole body mri system, equipped with a 6-el-ement phased-array coil was used (Siemens Medical Solutions, Erlangen, Germany). mri myocardial tagging with high temporal resolution (29 ms) was applied with Complemen-tary Spatial Modulation of Magnetization (7 mm tag distance) and steady state free preces-sion imaging. Parameters: eight phase-encoding lines per heart beat, tr 3.6 ms, te 1.8 ms, flip angle 20°, voxel size 1.2 × 3.8 × 6.0 mm3. In all patients and control subjects, this tagging cine was acquired in the mid-ventricular short-axis plane. After the tagging acquisitions, the lv and rv were covered by a stack of short-axis cine mri images for volumetric assess-ment, using steady state free precession imaging with a temporal resolution between 25 and 35 ms.

5.2.4 CMR Image analysisEnd-diastolic volume (edv), end-systolic volume (esv), ejection fraction and myocardial mass were calculated using mr Analytical Software System (Medis, Leiden, The Nether-lands). In order to assess lv peak filling rate (pfr), lv volumes throughout the cardiac cycle were calculated. The tagged images were analyzed with the Harmonic Phase procedure.14 Circumferential shortening was calculated over time during the cardiac cycle. For the lv free wall, septum, and rv free wall, the peak time (Tpeak) of circumferential shortening was calculated related to the ecg r-wave by automated routines.15

5.2.5 LV free wall, RV free wall, and septum definitionsThe lv free wall was subdivided in five equal segments. The two segments of the lv wall that were in direct continuity with the septum were not included as part of the lv free wall. The rv free wall was delineated in the same way. The complete septum was taken for the calculation of the septal strain, from the anterior until the posterior connections with the ventricular wall. For the lv free wall, rv free wall, and septum, the strains and strain timing parameters were derived.

5.2.6 RV End-systolic wall stressOur estimation of rv end-systolic (es) wall stress for both the patients and control subjects starts from the law of Laplace:16

68 Chapter 5

RV end-systolic wall stress = 0.5 × RV systolic pressure × RV end-systolic radiusRV end-systolic wall thickness

The systolic rv pressure is estimated by the systolic pap. The rves radius of curvature is dif-ficult to measure directly due to the rv’s irregular shape. Therefore, we estimate this radius from the rvesv by assuming that this volume can be described by a sphere in both the patients and controls. Then the rves radius is:

RV end-systolic radius = 0.620 × (RV end-systolic volume)1/3

The rv es wall thickness is estimated by dividing the rv free wall es volume by the rv free wall es surface area. The rv free wall es volume is obtained by contouring the rv free wall on every short-axis es slice, and then applying Simpson’s rule. The total rv es surface area was calculated as 4πr2, with r the radius from the above equation. The rv free wall fraction of total rv surface is estimated as 2 / 3 part. Thus the rv free wall es surface area is calculated as 2 / 3 times the total rv es surface area.

5.2.7 StatisticsGraphPad Prism version 4.0 (GraphPad Software, San Diego, California) was used for sta-tistical calculations. All data was tested for normal distribution. We performed a 2-tailed paired Student’s t-test to compare pre- and postoperative mri measurements and hemody-namic measurements. The relations between the l-r delay in Tpeak versus lv stroke volume, lv pfr, six minute walking distance and rv wall stress were tested by linear regression. All data are described as mean ± SD. A p value of < 0.05 was considered statistically signifi-cant

5.3 RESULTS

5.3.1 Patients’ characteristicsPatient characteristics are shown in Table 1. The median age of the patient population was 63 years (range 45-83 y) and 46% were female. The ecg-qrs width before pea was 96 ± 9 ms. On the basis of ecg morphology, right bundle branch block was present in one patient. Coronary artery disease was not present in any of the patients studied. pea was successful in all patients, resulting in a significant reduction in mean pap (45 ± 12 vs. 25 ± 6 mmHg; p < 0.001) and tpr (870 ± 391 vs. 406 ± 171 dyn·s/cm5; p < 0.001). Invasive hemodynamic measurements are summarized in Table 5.2. The 6-minute walking distance increased from 409 ± 109 to 510 ± 91 m (p < 0.001).

5.3.2 MRI-derived ventricular volumes and function As shown in Table 5.3, after pea, there was a significant reduction in rvedv and rvesv at 6


5

months after pea. rv stroke volume tended to increase and rvef increased. The preopera-tive hypertrophy of the rv decreased after pea. Left ventricular edv increased significantly with a stable lvesv. Therefore lv stroke volume improved significantly. In addition, lv peak filling rate (pfr) as corrected for lv-edv increased significantly.

Table 5�.1 Patient characteristics before PEA

Patient Sex Age (y) Heart rate (bpm)

BP s/d (mmHg)

PAP s/d/m(mmHg)

PCWP(mmHg)

CO(l/min)

QRS width(ms) NYHA Medication

1 m 78 74 120/70 80/30/47 5 3.7 104 III ERA

2 f 50 74 120/80 110/30/56 13 2.7 108 III None

3 f 58 75 115/75 83/37/54 14 3.6 108 III ERA

4 f 43 72 120/80 78/24/45 15 4.2 88 III ERA

5 m 66 49 150/85 55/14/29 11 5.8 92 II ERA

6 m 69 75 130/80 42/14/26 9 5.9 94 II ERA

7 f 68 88 110/70 109/33/58 5 2.3 88 IV ERA

8 f 61 87 95/65 78/33/50 6 5.5 92 III ERA

9 m 51 78 175/85 72/29/45 8 4.8 108 III ERA

10 f 60 69 131/90 60/47/53 12 4.0 82 III ERA

11 f 51 76 120/75 48/17/26 10 3.7 86 III ERA

12 m 74 80 160/90 70/49/59 13 3.7 102 III ERA

13 m 55 79 130/80 55/18/32 12 5.2 96 II ERA

BP = blood pressure, CO = cardiac output, ERA = endothelin receptor antagonist, NYHA = New York Heart As-sociation functional class, PAP = pulmonary arterial pressure, PCWP = pulmonary capillary wedge pressure, s/d(/m) = systolic/diastolic(/mean).

Table 5�.2 Invasive hemodynamic data before and after PEA

Pre PEA Post PEA ICU p-value

Heart rate (bpm) 75 ± 10 74 ± 8 n.s.

Systolic PAP (mmHg) 72 ± 21 39 ± 14 < 0.001

Diastolic PAP (mmHg) 28 ± 17 11 ± 5 0.003

Mean PAP (mm Hg) 45 ± 12 25 ± 6 < 0.001

PVR (dyne · s · cm-5) 661 ± 338 n.m.

TPR (dyne · s · cm-5) 870 ± 391 406 ± 171 0.001

Cardiac output (l/min) 4.2 ± 1.1 4.8 ± 0.8 n.s.

Systolic BP (mmHg) 120 ± 39 n.m.

Diastolic BP (mmHg) 78 ± 8 n.m.

PCWP (mmHg) 7 ± 5 n.m.

RAP (mmHg) 10 ± 3 n.m.

Values expressed as mean ± SD. PAP = pulmonary artery pressure, PVR = pulmonary vascular resistance, TPR = total pulmonary resistance, BP = blood pressure, PCWP = pulmonary capillary wedge pressure, RAP = right atrial pressure, n.s. = not significant, n.m. = not measured.

70 Chapter 5

5.3.3 Images and strainsFigure 5.1 shows short-axis cine images and short-axis tagged images at the time of rv peak strain before and after pea. Figure 5.2 shows an example of the circumferential shortening curves during the cardiac cycle for the lv and rv free walls and the septum before and after pea in one patient. Pre pea, the lv and rv start simultaneously, but the rv reaches its peak later than the lv. Post operatively, the rv peak is not later than the lv peak. In the patients before pea, rv free wall peak circumferential shortening was decreased (pre pea –13 ± 3% vs. control –18 ± 2%, p < 0.001), while peak circumferential shortening of lv free wall did not differ as compared with the healthy controls. After pea, rv peak circumferential short-ening increased to normal values. (Figure 5.2).

5.3.4 Timing parameters The results of the timing parameters pre-and post pea are shown in Figure 5.3 and Table 5.4. Before pea, the time to peak rv strain (Tpeak,rV) was significantly longer compared with Tpeak,LV, resulting in a l-r delay in peak strain. Postoperatively, Tpeak,rV was significantly re-duced; whereas Tpeak,LV did not change. Consequently, the l-r delay decreased from 97 ± 49 ms to values not different from the controls i.e., they reached normal values: –4 ± 51 ms (p < 0.001). In addition the septal to rv delay also normalized post pea.

5.3.5 Correlation and linear regression analysis All variables satisfied the condition of normal distribution. The results of correlation and linear regression analysis are shown in Figure 5.4. Several recovery parameters were signifi-cantly associated with the reduction in l-r delay: increase in lv stroke volume (ρ = 0.75, p < 0.001), increase in normalized lv peak filling rate (ρ = 0.64, p < 0.001), and increase in 6 minute walking distance (ρ = 0.67, p < 0.001). In contrast, there was no significant cor-

Table 5�.3� MRI volumetric parameters before and after PEA

Pre PEA Post PEA Change p-value

Left ventricle

End-diastolic volume (ml) 98 ± 15 111 ± 19 13 ± 11 < 0.001

End-systolic volume (ml) 28 ±47 31 ± 46 1 ± 15 0.89

Stroke volume (ml) 59 ± 13 72 ± 10 13 ± 11 < 0.001

Ejection fraction (%) 62 ± 13 67 ± 5 5 ± 15 0.24

PFR (ml/s) 309 ± 89 474 ± 172 165 ± 150 0.002

PFR / end-diastolic volume (s-1) 2.9 ± 0.8 4.2 ± 1.3 1.3 ± 1.1 0.003

Right ventricle

End-diastolic volume (ml) 173 ± 38 125 ± 18 –47 ± 41 0.001

End-systolic volume (ml) 107 ± 34 46 ± 16 –61 ± 31 < 0.001

Stroke volume (ml) 65 ± 19 78 ± 14 13 ± 22 0.07

Ejection fraction (%) 39 ± 12 63 ± 10 24 ± 14 < 0.001

Mass (g) 75 ± 19 51 ±14 –24 ± 13 < 0.001

Values expressed as mean ± SD. PFR = peak filling rate.


5

relation between the decrease in systolic pulmonary artery pressure and the decrease in l-r delay (ρ = 0.53, p = 0.07). Also there was no relation between the decrease in rv-radius and decrease in l-r delay (ρ = 0.37, p = 0.21), and no relation between the increase in end-systolic wall thickness and decrease in l-r delay (ρ = 0.19, p = 0.53).

5.3.6 RV end-systolic wall stressAfter pea, the estimated rv end-systolic wall stress decreased significantly from 15.2 ± 6.4 kPa to 5.7 ± 3.4 kPa (p < 0.001), which was not different from the normal reference values of 5.3 ± 1.39 kPa (p = 0.78, Figure 5.5A). In addition the rv end-systolic free wall thickness increased significantly after pea (0.98 ± 0.17 vs. 1.21. ± 0.38 cm, p = 0.01). Furthermore, as shown in Figure 5.5B, the change in l-r delay correlated significantly with the reduction in rv end-systolic wall stress (ρ = 0.69, p = 0.007).

Pre PEA Post PEA

RV LV

10 cm 10 cm

pericardiale�usion

sternumwires

Figure 5�.1 Short-axis CMR cine (top panels) and tagged images (bottom panels), at the time of peak RV shorten-CMR cine (top panels) and tagged images (bottom panels), at the time of peak RV shorten-at the time of peak RV shorten-ing in a patient with chronic thromboembolic pulmonary hypertension before and after PEA. Leftward ventricular septal bowing, as present before PEA, has recovered 6 months after PEA. LV = left ventricle.

20 40 60 80 100 20 40 60 80 100

–20

–15

–10

–5

0

5Pre PEA Post PEA

Relative time over the cardiac cycle (%) Relative time over the cardiac cycle (%)

Circ

umfe

rent

ial s

trai

n (%

)

LVSeptum

RV

Figure 5�.2 Circumferential strain over time, before and after PEA after the ECG R-wave for the left ventricular (LV) and RV free walls and the septum for a patient before(left) and after (right) PEA. Before PEA, the LV, RV, and septum start simultaneously with shortening (negative strain), but the RV reaches its peak later than the LV and RV peak strain is lower. After PEA, the L-R synchrony and RV peak strain have recovered.

72 Chapter 5

–100

–50

0

50

100

150

200

Pre PEA Post PEA Control

L-R Delay

Tim

e (m

s)

p < 0.001

C S-R Delay

Pre PEA Post PEA Control

D

–28

–24

–20

–16

–12

–8

–4

0Pre PEA Post PEA Control

Cir

cum

fere

ntia

l str

ain

(%)

Left ventricleA Right ventricleBPre PEA Post PEA Control

n.s.

n.s.

n.s.

n.s.

p < 0.001

p < 0.001

n.s.

Figure 5�.3� Peak circumferential strain and L-R delay for the left ventricle (A) and the RV (B), in the 13 CTEPH patients and the healthy subjects before and after PEA. Panels C and D show the L-R delay (C) and the S-R delay (D), before and after PEA. Bars and error bars indicate mean ± SE.

Table 5�.4� MRI strain and timing parameters before and after PEA

Pre PEA Post PEA Change p-value

LV peak strain (%) –19 ± 3 –20 ± 4 –0.5 ± 5 0.70

RV peak strain (%) –13 ± 3 –17 ± 3 –4 ± 3 0.001

SP peak strain (%) –14 ± 3 –16 ± 4 –2 ± 3 0.09

RR (ms) 823 ± 70 840 ± 90 17 ± 134 0.64

Tmax LVSB (ms) 397 ± 77 Not observed

Tpeak RV (ms) 405 ± 61 352 ± 67 –53 ±80 0.02

Tpeak LV (ms) 310 ± 46 356 ± 45 46 ±80 0.09

Tpeak SP (ms) 296 ± 43 320 ± 58 –23 ±82 0.33

LV to RV delay in Tpeak (ms) 97 ± 49 –4 ± 51 –101 ± 49 < 0.001

SP to RV delay in Tpeak (ms) 110 ± 78 25 ± 51 –85 ± 84 0.004

Values expressed as mean ± SD. LV = left ventricular, LVSB = leftward septal bowing, RV = right ventricular, RR = heart period, SP = septum, Tpeak = time to peak, Tmax = time to maximum LVSB.


50 50 100 150 200

–10

0

10

20

30

40

ρ = 0.75

Stro

ke v

olum

e in

crea

se (m

l)

0

50

100

150

200

250

Decrease in L-R delay (ms)

I

ncre

ase

in 6

MW

T (m

)

–1

0

1

2

3

4

Incr

ease

in L

V PF

R /E

D (s

-1)

ρ = 0.64

ρ = 0.67

A

B

C

RV e

nd-s

ysto

lic w

all s

tres

s(k

PA)

Pre PEA Post PEA non PH

A

ρ = 0.69, p = 0.007

Decrease in RV end-systolic wall stress (kPA)

Dec

reas

e in

L-R

del

ay (m

s)

0

5

10

15

20

25

30 B

0 5 10 15 200

50

100

150

200p < 0.001 n.s.

Figure 5�.4� Decrease in L-R delay versus increase in stroke volume (A), increase in left ventricular peak filling rate (B) and increase in six minute walking test (C). LV PFR = left ventricular peak filling rate, 6MWT = 6-minute walk-ing test.

Figure 5�.5� A: RV end-systolic wall stress before and after PEA in the 13 CTEPH patients and the 8 non PH pa-tients. B: Linear regression between the decrease in left-to-right (L-R) delay in time to peak of circumferential shortening as dependent variable and the decrease in right ventricular (RV) end-systolic wall stress. Bars and error bars indicate mean ± SEM.

74 Chapter 5

5.4 DISCUSSION

5.4.1 General discussionThe major findings of the present study are: (1) after pea, both the l-r delay in peak strains and estimates of end-systolic wall stress reverse to normal values; and (2) the change in l-r delay is predominantly associated with the change in rv end-systolic wall stress.

Our findings of a decrease in l-r asynchrony after unloading the rv, agree with the previous observations by Lurz et al. 12 in patients where an rv-pulmonary artery conduit obstruction was relieved. Similar to our study, they reported that the reduction in rv wall stress correlated with the reduction in l-r delay. We showed that this also applies for pa-tients with pulmonary hypertension and in addition we proved that the synchrony-recov-ery is mainly explained by the combined effect of reduction in pap pressure, end-systolic radius and increased wall thickness. Since we also included healthy subjects, we were able to prove that rv peak strain, wall stress, as well as l-r synchrony reversed to normal values (Figures 5.3 and 5.5). This implies that the failing rv is capable of functional recovery after an intervention that reduces rv wall stress.

In this study, we performed the mri measurements at least six months post-operatively, since previous reports showed that unloading the ventricle results in morphological and functional improvement over a period of at least 3 months.17, 18 In addition, reesink et al.19 showed, by performing mri after at least 4 months, that an almost complete reverse rv remodelling had taken place. However for the invasive pressures, we used the direct post-operative pressure measurements for the calculation of wall stress. The combination of post-operative mri after 6 months and the postoperative pressure values is a limitation in this study, since the mri derived values reflect a reverse-remodeled state whereas the post-pea hemodynamics do not. Nevertheless, previous reports indicate that the pulmonary artery pressure remains stable during follow-up after pea.20, 21

5.4.2 Experimental data Our results agree with data on cardiac muscle. Brutsaert et al.22 and others23 investigated the relationship between load and contraction duration by showing that an acute higher af-terload imposed on isolated muscles increased the duration of contraction. recently, Han-doko et al.11 used an isolated langendorff-perfused heart from a chronic ph rat model and showed that ventricular asynchrony in peak pressure was induced by increasing rv volume and pressure, using an inflatable ventricular balloon. Thus the increase of myocardial wall stress leads to prolonged contraction duration, in experimental conditions as well.

5.4.3 End-systolic wall stress in pulmonary hypertension The finding that end–systolic wall stress is increased in cteph, is in accordance with earlier reports on pulmonary hypertension.24, 25 The increased wall stress does not only affect rv contraction duration, but also negatively affects myocardial perfusion26 and subsequent glu-cose metabolism,25 and increases oxygen demand.27 Furthermore, Grossman et al.28 showed that increased wall stress induced dilatation, which in turn leads to further increase in wall stress and thereby a vicious circle of positive feedback is maintained.


5

Lowering end-systolic wall stress can be achieved by both a reduction of rv pressure and by an improvement in rv adaptation through concentric hypertrophy.29 The relevance of this adaptation is manifest in patients with ph due to congenital heart disease: In these patients, the rv is more capable to cope with the increased afterload, probably because the rv has had more time to adapt by developing compensatory hypertrophy. This underscores the relevance of hypertrophy for lowering wall stress, and thereby for reducing l-r asynchrony in rv pressure overload.

5.4.4 LimitationsThe number of patients in this study was small. This is a consequence of the very invasive surgical pea procedure, which can only be applied in a small subset of cteph patients. However, all statistical analyses were performed within patient, comparing the data before and after surgery by paired samples t-testing. This made it possible to obtain significant results in a small number.

Furthermore, a simplifying assumption is the description of the rv end-systolic volume by a spherical configuration, which is used in the calculation of rv wall stress for both the patients and the control subjects.

5.5 CONCLUSION

In cteph patients, the l-r asynchrony in peak strain recovers to normal values after pea. The rv end-systolic wall stress plays a key role in this recovery, reflecting a complex inter-play of pulmonary artery pressure, rv radius and wall thickness. ■

References

1. Roeleveld RJ, Marcus JT, Faes TJ, Gan TJ, Boonstra A, Postmus PE,Vonk-Noordegraaf A. Interventricular septal configuration at MR imaging and pulmonary arterial pressure in pulmonary hypertension. Radiology 2005; 234(3):710-717.

2. Kingma I, Tyberg JV, Smith ER. Effects of diastolic transseptal pressure gradient on ventricular septal position and motion. Circulation 1983; 68(6):1304-1314.

3. Tanaka H, Tei C, Nakao S, Tahara M, Sakurai S, Kashima T, Kanehisa T. Diastolic bulging of the interventricular septum toward the left ventricle. An echocardiographic manifestation of negative interventricular pressure gradient between left and right ventricles during diastole. Circulation 1980; 62(3):558-563.

4. Dohi K, Onishi K, Gorcsan J, López-Candales A, Takamura T, Ota S, Yamada N, Ito M. Role of radial strain and displacement imaging to quantify wall motion dyssynchrony in patients with left ventricular mechanical dyssynchrony and chronic right ventricular pressure overload. Am J Cardiol 2008; 101(8):1206-1212

5. Marcus JT, Gan CT, Zwanenburg JJ, Boonstra A, Allaart CP, Götte MJ, Vonk-Noordegraaf A. Interventricular mechanical asynchrony in pulmonary arterial hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol 2008; 51(7):750-757.

6. Lopez-Candales A, Dohi K, Rajagopalan N, Suffoletto M, Murali S, Gorcsan J, Edelman K. Right ventricular dyssynchrony in patients with pulmonary hypertension is associated with disease severity and functional class. Cardiovasc Ultrasound 2005; 3:23.

76 Chapter 5

7. Kalogeropoulos AP, Georgiopoulou VV, Howell S, Pernetz MA, Fisher MR, Lerakis S, Martin RP. Evaluation of right intraventricular dyssynchrony by two-dimensional strain echocardiography in patients with pulmonary arterial hypertension. J Am Soc Echocardiogr 2008; 21(9):1028-1034.

8. Beyar R. Heart inefficiency in pulmonary hypertension: a double jeopardy. J Am Coll Cardiol 2008; 51(7):758-759.

9. Gan CT, Lankhaar JW, Marcus JT, Westerhof N, Marques KM, Bronzwaer JG, Boonstra A, Postmus PE, Vonk-Noordegraaf A. Impaired left ventricular filling due to right-to-left ventricular interaction in patients with pulmonary arterial hypertension. Am J Physiol Heart Circ Physiol 2006; 290(4):H1528-1533.

10. Stojnic BB, Brecker SJ, Xiao HB, Helmy SM, Mbaissouroum M, Gibson DG. Left ventricular filling characteristics in pulmonary hypertension: a new mode of ventricular interaction. Br Heart J 1992; 68(1):16-20.

11. Handoko ML, Lamberts RR, Redout EM, de Man FS, Boer C, Simonides WS, Paulus WJ, Westerhof N, Allaart CP, Vonk-Noordegraaf A. Right ventricular pacing improves right heart function in experimental pulmonary arterial hypertension: a study in the isolated heart. Am J Physiol Heart Circ Physiol 2009; 297(5):H1752-1759.

12. Lurz P, Puranik R, Nordmeyer J, Muthurangu V, Hansen MS, Schievano S, Marek J, Bonhoeffer P, Taylor AM. Improvement in left ventricular filling properties after relief of right ventricle to pulmonary artery conduit obstruction: contribution of septal motion and interventricular mechanical delay. Eur Heart J 2009; 30(18):2266-2274.

13. Auger WR, Fedullo PF, Moser KM, Buchbinder M, Peterson KL. Chronic major-vessel thromboembolic pulmonary artery obstruction: appearance at angiography. Radiology 1992; 182(2):393-398.

14. Osman NF, Kerwin WS, McVeigh ER, Prince JL. Cardiac motion tracking using CINE harmonic phase (HARP) magnetic resonance imaging. Magn Reson Med 1999; 42(6):1048-1060.

15. Zwanenburg JJ, Gotte MJ, Kuijer JP, Heethaar RM, van Rossum AC, Marcus JT. Timing of cardiac contraction in humans mapped by high-temporal-resolution MRI tagging: early onset and late peak of shortening in lateral wall. Am J Physiol Heart Circ Physiol 2004; 286(5):H1872-1880.

16. Norton JM. Toward consistent definitions for preload and afterload. Adv Physiol Educ 2001; 25(1-4):53-61.17. Cihak R, Kolar F, Pelouch V, Prochazka J, Ostadal B, Widimsky J. Functional changes in the right and

left ventricle during development of cardiac hypertrophy and after its regression. Cardiovasc Res 1992; 26(9):845-850.

18. D’Armini AM, Zanotti G, Ghio S, Magrini G, Pozzi M, Scelsi L, Meloni G, Klersy C, Viganò M. Reverse right ventricular remodeling after pulmonary endarterectomy. J Thorac Cardiovasc Surg 2007; 133(1):162-168.

19. Reesink HJ, Marcus JT, Tulevski, Jamieson S, Kloek JJ, Vonk Noordegraaf A, Bresser P. Reverse right ventricular remodeling after pulmonary endarterectomy in patients with chronic thromboembolic pulmonary hypertension: utility of magnetic resonance imaging to demonstrate restoration of the right ventricle. J Thorac Cardiovasc Surg 2007; 133(1):58-64.

20. Giusca S, Dambrauskaite V, Scheurwegs C, D’hooge J, Claus P, Herbots L, Magro M, Rademakers F, Meyns B, Delcroix M, Voigt JU. Deformation imaging describes right ventricular function better than longitudinal displacement of the tricuspid ring. Heart 2010; 96(4):281-288.

21. Freed DH, Thomson BM, Tsui SS, Dunning JJ, Sheares KK, Pepke-Zaba J, Jenkins DP. Functional and haemodynamic outcome 1 year after pulmonary thromboendarterectomy. Eur J Cardiothorac Surg 2008; 34(3):525-529; discussion 529-530.

22. Brutsaert DL, Rademakers FE, Sys SU. Triple control of relaxation: implications in cardiac disease. Circulation 1984; 69(1):190-196.

23. van Heuningen R, Rijnsburger WH, ter Keurs HE. Sarcomere length control in striated muscle. Am J Physiol 1982; 242(3):H411-420.

24. Quaife RA, Chen MY, Lynch D, Badesch DB, Groves BM, Wolfel E, Robertson AD, Bristow MR, Voelkel NF. Importance of right ventricular end-systolic regional wall stress in idiopathic pulmonary arterial hypertension: a new method for estimation of right ventricular wall stress. Eur J Med Res 2006; 11(5):214-220.

25. Oikawa M, Kagaya Y, Otani H, Sakuma M, Demachi J, Suzuki J, Takahashi T, Nawata J, Ido T, Watanabe J, Shirato K. Increased [18F]fluorodeoxyglucose accumulation in right ventricular free wall in patients with pulmonary hypertension and the effect of epoprostenol. J Am Coll Cardiol 2005; 45(11):1849-1855.


5

26. Gomez A, Bialostozky D, Zajarias A, Santos E, Palomar A, Martínez ML, Sandoval J.. Right ventricular ischemia in patients with primary pulmonary hypertension. J Am Coll Cardiol 2001; 38(4):1137-1142.

27. Westerhof N, Boer C, Lamberts RR, Sipkema P. Cross-talk between cardiac muscle and coronary vasculature. Physiol Rev 2006; 86(4):1263-1308.

28. Grossman W, Jones D, McLaurin LP. Wall stress and patterns of hypertrophy in the human left ventricle. J Clin Invest 1975; 56(1):56-64.

29. Vonk Noordegraaf A, Westerhof N. Right ventricular ejection fraction and NT-proBNP are both indicators of wall stress in pulmonary hypertension. Eur Respir J 2007; 29(4):622-623.

C H A P T E R 6

79

ABSTRACT

Am J Cardiol (epub ahead of print)

Previous studies have shown the prognostic benefit of nt-probnp in pulmonary arterial hypertension (pah) at time of diagnosis. However, there are only limited data on the clinical utility of serial measure-ments of the inactive peptide nt-probnp in pah. This study examines the value of serial nt-probnp measurements in predicting prognosis pah. We retrospectively analyzed all available nt-probnp plasma sam-ples in 198 patients who were diagnosed with who group i pah from January 2002 to January 2009. At the time of diagnosis, median nt-probnp levels were significantly different between survivors (610 pg/ml; range, 6 to 8714) and nonsurvivors (2609 pg/ml, range 28-9828) (p < 0.001). In addition nt-probnp was significantly associated (p<0.001) with other parameters of disease severity (6 minutes walking distance; functional class). receiver operating curve analysis identified ≥ 1256 pg/ml as the optimal nt-probnp cut-off for predicting mortality at the time of diagnosis. Serial measurements allowed calculation of baseline nt-probnp (i.e., the intercept obtained by back-extrapolation of the concentration-time graph) providing a better discrimination between survivors and non-survivors than nt-probnp at time of diagnosis alone (p = 0.010). Furthermore, a decrease of nt-probnp of more than 15 percent per year is associated with survival. In conclusion, a serum nt-probnp level of ≥ 1256 pg/ml at time of diagnosis identifies pah patients with poor outcome. In addition a decrease in nt-probnp of more then 15% per year is associated with survival in pah. ◦

Departments of aPulmonary Diseases, bPhysiology, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam; departments of cBiostatistics and dEpidemiology, Erasmus University Medical Center, Rotterdam; 5University of Giessen Lung Centre, Germany

G.J. Mauritz,a D. Rizopoulos,c H. Groepenhoff,a H. Tiede,d,e J. Felix,d P. Eilers,c J. Bosboom,a P. E. Postmus,a N. Westerhof,a,b A. Vonk-Noordegraafa

Usefulness of serial N-terminal pro-B-type natriuretic peptide measurements for determining prognosis in patients with pulmonary arterial hypertension

6.1 INTRODUCTION

N-TErMINAL PrO-BrAIN NATrIUrETIC PEPTIDE (nt-probnp) concentra-tions have been shown to correlate with invasive pulmonary hemodynamics and outcome.1 Andreassen and colleagues2 showed that a nt-probnp < 553 pg/mL was

related to a better 6-month and 1-year survival in 68 patients with pulmonary arterial hy-pertension (pah) associated with scleroderma. Fijalkowska et al.3 showed that nt-probnp

80 Chapter 6

levels > 1400 pg/mL were predictive of 3-year outcome in 55 patients with severe pah. Fur-thermore, nt-probnp correlates with cardiac magnetic resonance imaging-derived mea-sures of right ventricular (rv) function in pulmonary hypertension.4,5 These studies, how-ever, have only evaluated the prognostic value of nt-probnp at a single point in time. From left–sided heart failure it is known that serial measurements provide additional prognostic value in comparison to single values alone.6-9 Therefore, this study aims to investigate in pah patients: 1) the optimal cut-off value of nt-probnp at time of original diagnosis in pre-dicting mortality in a large patient population and 2) whether serial nt-probnp measure-ments provide better information in addition to that of the value at diagnosis in predicting prognosis.

6.2 METHODS

Serial cardiac biomarker measurement had been introduced at the department of Pul-monology of vu Medical Center of Amsterdam, the Netherlands since 2002 as part of routine clinical assessment. We retrospectively identified all patients who were diagnosed with who group i pulmonary hypertension (pah) between November 2002 and September 2009. The study was approved by our institutional review board and ethics committee. All who group i subgroups with at least one nt-probnp measurement were included in this analysis (n = 198 patients). Thirty-seven of these only had a single measurement at baseline and these were not included in the longitudinal analysis. Survival data were available for all patients until the date of death or 1st January 2010. Follow-up was performed by regular visits to the outpatient clinic at 3-6 month-intervals, and by telephone contact.

Blood samples for nt-probnp measurements were drawn from a peripheral vein in all patients while the patient was in a stable hemodynamic state before the six minute walking test was performed. Figure 6.1 shows the available longitudinal data of nt-probnp measure-ments for the 161 patients (determination at time of diagnosis). Due to the observational nature of this study the visiting patterns of the patients are rather variable. The nt-probnp plasma levels were analyzed with the Elecsys 1010 electrochemiluminescence immunoas-say (roche Diagnostics Netherlands)), as described previously.5

Baseline NT-proBNPavailable (n = 198)

Follow-up available (n = 161)

No follow-up available (n = 37)

2-3 samplesavailable (n = 51)



≥ 10 samplesavailable (n = 25)

Figure 6.1 The PAH patients with blood samples available for N-terminal pro-B-type natriuretic peptide testing at baseline and follow-up visits.

81Usefulness of NT-proBNP for prognosis in PAH

6

Differences in the baseline covariates between survivors and non-survivors were tested with the Mann-Whitney u-test. Multivariable Cox regression was used to assess the association between the baseline log10(nt-probnp) level and the risk of death, adjusting for sex, age, nyha class, mean right atrial pressure and baseline estimated glomerular filtration rate.

To investigate the added value of the serial nt-probnp measurements a two-stage pro-cedure was employed. In the first stage, the longitudinal information for each patient was summarized, while accounting for the correlation in the nt-probnp serial measurements, by means of a linear mixed effects model10 with a random intercepts and random slopes structure. The aim of this model is to estimate the longitudinal evolution of nt-probnp for each patient while accounting for the biological variation. Typically, biomarker mea-surements, such as nt-probnp, contain intra-patient ‘error’ which is also known ‘biologi-cal variation’. More specifically, by biological variation we imply the fluctuations that are observed in the biomarker levels human physiology. As shown in Figure 6.2, patient b has a relatively stable profile of nt-probnp levels in time; however, not all nt-probnp measure-ments fall exactly on the dashed line denoting this profile. The fluctuations of the observed nt-probnp levels around the profile we call biological variation. From a statistical point of view, this variability should be taken into account in order to obtain optimal results, since failure to account for this variation will lead to an underestimation of the effect size of nt-probnp.

At the second stage, the patient-specific intercepts (baseline value) and slopes were in-cluded as covariates in a similar multivariable Cox regression model as in the previous analysis, which also corrected for the effects of sex, age, nyha class, mean right atrial pres-sure and baseline estimated glomerular filtration rate.

Based on an roc analysis at 80 months, optimal cut points for the baseline nt-probnp and patient-specific intercept were determined as the points for which the product of sen-sitivity and specificity attains its maximum. All analyses have been performed in r version 2.11.1 (2010-05-31) using packages jm (version 0.7-0), nlme (version 3.1-97), survival (ver-sion 2.35-8) and survival roc (version 1.0.0).

10 20 30 40 50 60 702.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

Time (months)

log1

0(N

T-pr

oBN

P)

Observed NT-proBNP

Estimated evolutionPatient A

Patient B

0

Figure 6.2 Examples of observed NT-proBNP measurements and estimated underlying true evolutions from the linear mixed effects model for two patients (A and B). The slope quantifies the expected average change of the marker not the observed one, where data scatter.

82 Chapter 6

6.3 RESULTS

Baseline characteristics of all 198 patients are presented in Table 6.1. The population was predominantly female (71.2%) with a mean age of 54 ± 17 years. Fift y-six percent had id-4 ± 17 years. Fift y-six percent had id-years. Fifty-six percent had id-iopathic pah and over 58% were in modified nyha/who functional class iii and vi. pah baseline therapies included prostacyclin analogs in 26 (13.1%), endothelin receptor antago-nists in 118 (59.5%), and phosphodiesterase-5 inhibitors in 40 patients (20.2%). A total of 8 patients received combination pah therapies. Calcium channel blockers were used for pah treatment for 5 patients (2.5%, data not shown).

During a mean follow-up period of 38 ± 23 months, 54 patients (27.3%) died. Figure 6.3 shows the Kaplan-Meier estimate of the survival function for the whole group. The overall survival estimates for our pah population at 1, 3 and 4 years were 93% (95%-ci: 90-97%), 80% (95%-ci: 74-86%) and 71% (95%-ci: 65-79%), respectively. Both, pulmonary artery pressure and pvr at baseline were similar for survivors and nonsurvivors. In contrast, those who died had significantly worse functional class and walked a significantly shorter dis-tance in 6 min. Furthermore, their hemodynamic parameters (higher rap and lower car-diac output and mixed venous saturation) indicated more compromised rv function.

Baseline nt-probnp levels of survivors and nonsurvivors were significantly different (Table 6.1). Plasma nt-probnp increased significantly (Kruskal-Wallis test, p < 0.0001) with the severity of nyha functional class (Figure 6.4). Of our patients, 147 had diagnostic assessment and simultaneous nt-probnp measurements at baseline. nt-probnp was posi-tively correlated with mean right atrial pressure (Spearman’s correlation coefficient 0.26, p < 0.001), serum creatinine (Spearman’s correlation coefficient 0.48, p < 0.001), and pul-monary vascular resistance (Spearman’s correlation coefficient 0.54, p < 0.001), and heart rate (Spearman’s correlation coefficient 0.28, p < 0.001), whereas negative associations were found with cardiac index (Spearman’s correlation coefficient –0.43, p < 0.001), and mixed venous oxygen saturation (Spearman’s correlation coefficient –0.42, p < 0.001).

An initial serum nt-probnp cutoff value of 1256 pg/ml (95%-ci: 811-1700 pg/mL) for prediction of death was identified by roc analysis (area under the curve is 0.747). A serum nt-probnp concentration of 1256 pg/ml showed 63.5% sensitivity, 77.5% specificity, and a negative predictive value of 88.5% for a fatal outcome. Cox regression modeling confirmed that baseline nt-probnp was highly significantly associated with mortality in our pah pa-

Kaplan-Meier estimate

Time (months)

Surv

ival

(%)

95% CI

0 20 40 60 800

40

60

80

100

20

Figure 6.3� Kaplan-Meier estimates of survival from the time of enrolment. Dashed lines represent the 95% CI for the estimates.


6

tient population (p = 0.010; Table 6.2), adjusting for gender, age, nyha functional class, mean right atrial pressure and baseline estimated glomerular filtration rate. The hazard ratio for one unit increase in the baseline log10 (nt-probnp) was 1.92 (95%-ci: 1.01-3.64). Kaplan-Meier curves for survival in those with nt-probnp levels above 1256 pg/ml and

Table 6.1 Patients’ characteristics according to survival

Entire population Survivors Non-survivors p-value*


Number of patients 198 144 54

Age (y) 54 ± 17 51 ± 17 63 ± 15 < 0.001

Sex (m/f) 49/149 31/113 18/36 0.1261

Diagnosis, n (%)

Idiopathic PAH 110 (56) 92 (64) 18 (34) 0.002

PAH associated with

Connective tissue disease 58 (28) 35 (24) 23 (43) 0.021

Portal hypertension 11 (6) 6 (3) 5 (9) 0.290

HIV infection 6 (3) 5 (4) 1 (2) 0.889

Drugs and toxins induced PAH 5 (3) 2 (1) 3 (6) 0.247

Other 7 (4) 5 (4) 3 (6) 0.791

Functional status


II 82 (41) 71 (49) 11 (20) < 0.001

III 94 (47) 67 (47) 27 (50) 0.773

IV 22 (12) 6 (4) 16 (30) < 0.001

6-minute walking distance (m) 387 ± 137 406 ± 137 326 ± 121 0.002

Hemodynamics

Heart rate (bpm) 79 ± 14 79 ± 15 82 ±13 0.177

PAP (mmHg) 45 ± 13 45 ± 10 46 ± 14 0.980

RAP (mmHg) 8 ± 7 7 ± 6 10 ± 8 0.049

PCWP (mmHg) 9.3 ± 6 9 ± 5 10 ± 3 0.554

PVR (dyn·s·cm‒5) 607 ± 347 580 ± 344 676 ± 348 0.066

Cardiac output (l/min) 5.7 ± 2.3 5.9 ± 2.4 5.1 ± 2.1 0.039

SaO2 (%) 93 ± 10 90 ± 11 94 ± 9 0.036

SvO2 (%) 66 ± 11 67 ± 10 65 ± 11 0.046

NT-proBNP

NT-proBNP (log10 pg/L) 2.89 ± 0.69 2.75 ± 10 3.42 ± 0.63 < 0.001

NT-proBNP (pg/L)* 6818 (287-3,039) 610 (210-1,950) 2609 (872-5,335) < 0.001

Values expressed as mean ± SD. *Values expressed as median (25th-75th percentile). PAH = pulmonary arterial hypertension, PAP = pulmonary artery pressure, PVR = pulmonary vascular resistance, RAP = right atrial pres-sure, PCWP = pulmonary capillary wedge pressure, PVR = pulmonary vascular resistance, SaO2 = arterial oxygen saturation, SvO2 = mixed venous oxygen saturation.

84 Chapter 6

those with levels below this cut-off are shown in Figure 6.5. Differences between the 2 groups were significant (log-rank test, p < 0.0001).

More than the half of the patients provided up to three measurements. The overall aver-age was 5 ± 4 measurements per patient. Patients who died provided fewer measurements

II III IV

NYHA class

NT-

proB

NP

(pg/

ml)

0

2000

4000

6000

8000

10000

Kaplan-Meier Estimate

Time (months)

Surv

ival

(%)

0 20 40 60 800

40

60

80

100

20NT-proBNP ≤ 1256 pg/mLNT-proBNP > 1256 pg/mL

0

10,000

Change per year (%)

Base

line

NT-

proB

NP

(pg/

ml)

–30 30 60–60–90

100

1

1,000

–15

90

Survivors Survivors + Non-survivors

IncreaseDecrease

SurvivorsNon-survivors

Figure 6.4� Box plots of the relationship between NYHA status and NT-proBNP level. The plots display the median values (horizontal bars), 25th and 75th percentiles (lower and upper limits of boxes), and 1.5 times the inter-quartile range (error bars). Median NT-proBNP values for each group are: NYHA Class II: 431.5 pmol/mL; NYHA Class III: 1100 pmol/mL; NYHA Class IV: 2711 pmol/mL.

Figure 6.5� Kaplan-Meier estimates of survival for low and high baseline NT-proBNP values (low is defined as < 1256 pg/ml and high as ≥ 1256 pg/ml).

Figure 6.6 Scatterplot of the estimated baseline NT-proBNP (intercept) and the relative change in percentage per year (slope) derived from the linear mixed effects model. A decrease in NT-proBNP of 15% per year or more dur-ing follow-up is associated with survival.


6

Table 6.2 Cox regression models of single N-terminal pro-brain natriuretic peptide measurement

Value Hazard ratio SE p-value

Female 0.82 2.27 0.71 0.2473

Age 0.04 1.03 0.01 0.0004

NYHA III 0.33 1.39 0.39 0.3980

NYHA IV 1.09 2.97 0.46 0.0189

Mean right atrial pressure 0.02 1.02 0.02 0.4062

Estimated Glomerular iltration Rate –4.01 0.02 1.87 0.0324

Log10 (baseline NT-proBNP) 0.65 1.92 0.33 0.0469

Parameter estimates, hazard ratios, standard errors, and p-values from a Cox model using the baseline mea-surement of log10 (NT-proBNP).

Table 6.3� Cox regression models of estimated slopes and intercepts of the linear mixed effects model

Value Hazard ratio SE p-value

Female 0.82 2.27 0.71 0.2473

Age 0.04 1.03 0.01 0.0004

NYHA III 0.33 1.39 0.39 0.3980

NYHA IV 1.09 2.97 0.46 0.0189

Mean right atrial pressure 0.02 1.02 0.02 0.4062

Estimated glomerular filtration rate –4.01 0.02 1.87 0.0324

Log10 (baseline NT-proBNP) 0.65 1.92 0.33 0.0469

Parameter estimates, hazard ratios, standard errors, and p-values from a Cox model that includes the estimated intercepts and slopes of the linear mixed effects model.

compared to those who still alive on January 1st, 2010 (4 ± 4 vs. 6 ± 5 measurements, re-spectively). Table 6.3 presents the results of the Cox analysis for the association of longi-tudinal measurements of nt-probnp with survival, adjusted for gender, age, nyha class, mrap and baseline egfr. The estimated hazard ratios for one unit increase in the intercept and 0.01 units increase (i.e., one standard deviation increase) in the slope are 2.26 (95%-ci: 1.1-4.64), and 1.28 (95%-ci: 0.94-1.75), respectively. Furthermore, Figure 6.6 presents the scatter plot of the estimated nt-probnp intercepts and % change per year for each patient determined from the longitudinal data analysis using the log-linear mixed effects model. As shown from this figure, all patients with a decrease of nt-probnp at follow-up of more than 15 percent per year are surviving pah patients.

86 Chapter 6

6.4 DISCUSSION

This is the largest study to date to assess the prognostic value of using all available single and repeated measurements of nt-probnp in patients with pah. The major findings can be summarized as follows:

■ A serum nt-probnp level of ≥ 1256 pg/ml at time of diagnosis identified pah patients with poor outcome. ■ Longitudinal available nt-probnp measurements provide additional prognostic value to a calculated baseline nt-probnp value. In particular, after accounting for the biologi-cal variation in the serial nt-probnp measurements by means of a linear mixed effects model, it was noted that the variation-free baseline nt-probnp (i.e., intercept) provided much better discrimination than the nt-probnp value measured at time of diagnosis (Table 6.3). ■ As observed from Figure 6.6, a decrease in nt-probnp of 15% per year or more is associ-ated with survival.

Our findings are in line with preceding studies that describe a correlation between nt-probnp and baseline clinical and hemodynamic parameters in pah.1,11-15 As other investi-gators,1,2,5 we found baseline nt-probnp to be a risk factor for all-cause mortality. Similar to other reports in pah, baseline nt-probnp (median 812 pg/ml) largely exceeded normal values (< 50-334 pg/ml)16 in the present study. According to recent work in left heart dis-ease, left ventricular end-systolic wall stress appears to be the key mechanical stimulus influencing cardiac nt-probnp release.17 When applicable to right heart failure in pah, an increased nt-probnp level might thus reflect increased rv wall stress, as has been hypoth-esized earlier.18 Our finding that a baseline nt-probnp value above a cut-off of 1256 pg/ml is associated with increased mortality risk is comparable with the findings of Fijalkowska and colleagues.3 They found in a smaller group that a serum nt-probnp level above 1400 pg/ml is related to increased mortality risk.

So far, no data are available investigating the importance of evolution over time in nt-probnp as a prognostic marker in patients with pah. The present study shows that serial determination of nt-probnp in clinical practice add significant prognostic value to a base-line serum nt-probnp measurement. This is in accordance with data from left heart failure and serial nt-probnp measurements.7,8,19 Although a significant drop of nt-probnp of more than 15 % in 1 year is associated with survival (Figure 6.6), the predictive value of a change less than 15 % is absent. Interestingly, Sitbon and associates demonstrated that a reduction in pvr of more than 30% relative to baseline after three months of epoprostenol therapy, is a clear indicator of better survival.20 In accordance with there findings this implies that long term survival is only guaranteed if significant treatment effects are reached.

risk stratification in patients with pah is crucial in order to guide decisions about treat-ment continuation versus intensification. The high predictive value of baseline values con-firm a common finding in pah that poor baseline values predict a poor prognosis.21-24 Our data confirms that early diagnosis of the disease is required to improve outcome. Second, based on the prognostic value of baseline nt-probnp it has been assumed that getting nt-probnp below this value during treatment should be considered as a treatment goal.25 Our data showed that a only a significant drop in nt-probnp, of more than 15% per year, guar-


6

antees long term survival rather then lowering of nt-probnp below a certain threshold, as currently is recommended. Finally, our data provide a basis for future prospective stud-ies incorporating nt-probnp to comprehensively determine the prognostic utility of serial measures of nt-probnp. Investigators undertaking cohort studies or therapeutic trials in pah should be encouraged to incorporate serial nt-probnp measurements in study de-signs.

Our study had limitations. First, our study had a retrospective nature, resulting in het-erogeneous spread of the time intervals of serial nt-probnp measurements. Therefore we were not able to study the importance of changes with regard to a defined follow-up period. By using a linear mixed effects model we were however able to analyse the data taking into account these different time intervals. Furthermore, we cannot exclude that different medi-cal therapy may result in different changes of nt-probnp plasma levels, although our data did not support this. Another limitations is that we estimated renal function on a simple estimation of creatinine clearance and do not provide repetitive data. ■

References

1. Leuchte HH, El Nounou M, Tuerpe JC, Hartmann B, Baumgartner RA, Vogeser M, Muehling O, Behr J. N-terminal pro-brain natriuretic peptide and renal insufficiency as predictors of mortality in pulmonary hypertension. Chest 2007;131:402-9.

2. Andreassen AK, Wergeland R, Simonsen S, Geiran O, Guevara C, Ueland T. N-terminal pro-B-type natriuretic peptide as an indicator of disease severity in a heterogeneous group of patients with chronic precapillary pulmonary hypertension. Am J Cardiol 2006;98:525-9.

3. Fijalkowska A, Kurzyna M, Torbicki A, Szewczyk G, Florczyk M, Pruszczyk P, Szturmowicz M. Serum N-terminal brain natriuretic peptide as a prognostic parameter in patients with pulmonary hypertension. Chest 2006;129:1313-21.

4. Blyth KG, Groenning BA, Mark PB, Martin TN, Foster JE, Steedman T, Morton JJ, Dargie HJ, Peacock AJ. NT-proBNP can be used to detect right ventricular systolic dysfunction in pulmonary hypertension. Eur Respir J 2007;29:737-44.

5. Gan CT, McCann GP, Marcus JT, van Wolferen SA, Twisk JW, Boonstra A, Postmus PE, Vonk-Noordegraaf A. NT-proBNP reflects right ventricular structure and function in pulmonary hypertension. Eur Respir J 2006;28:1190-4.

6. Daly MG, Frampton CM, Troughton RW. Improving risk stratification for heart failure. A role for serial testing of B-type natriuretic peptides? J Am Coll Cardiol 2010 Feb 2;55:451-3.

7. Latini R, Masson S, Wong M, Barlera S, Carretta E, Staszewsky L, Vago T, Maggioni AP, Anand IS, Tan LB, Tognoni G, Cohn JN. Incremental prognostic value of changes in B-type natriuretic peptide in heart failure. Am J Med 2006;119:70 e23-30.

8. Masson S, Latini R, Anand IS, Barlera S, Angelici L, Vago T, Tognoni G, Cohn JN. Prognostic value of changes in N-terminal pro-brain natriuretic peptide in Val-HeFT (Valsartan Heart Failure Trial). J Am Coll Cardiol 2008;52:997-1003.

9. Wu AH. Serial testing of B-type natriuretic peptide and NTpro-BNP for monitoring therapy of heart failure: the role of biologic variation in the interpretation of results. Am Heart J 2006;152:828-34.

10. Verbeke G, Molenberghs G. Linear Mixed Models for Longitudinal Data. Springer-Verlag, New York, 2000.11. Souza R, Bogossian HB, Humbert M, Jardim C, Rabelo R, Amato MB, Carvalho CR. N-terminal-pro-brain

natriuretic peptide as a hemodynamic marker in idiopathic pulmonary arterial hypertension. Eur Respir J 2005;25:509-13.

12. Williams MH, Handler CE, Akram R, Smith CJ, Das C, Smee J, Nair D, Denton CP, Black CM, Coghlan JG. Role of N-terminal brain natriuretic peptide (N-TproBNP) in scleroderma-associated pulmonary arterial hypertension. Eur Heart J 2006;27:1485-94.

88 Chapter 6

13. Allanore Y, Borderie D, Meune C, Cabanes L, Weber S, Ekindjian OG, Kahan A. N-terminal pro-brain natriuretic peptide as a diagnostic marker of early pulmonary artery hypertension in patients with systemic sclerosis and effects of calcium-channel blockers. Arthritis Rheum 2003;48:3503-8.

14. Mukerjee D, Yap LB, Holmes AM, Nair D, Ayrton P, Black CM, Coghlan JG. Significance of plasma N-terminal pro-brain natriuretic peptide in patients with systemic sclerosis-related pulmonary arterial hypertension. Respir Med 2003;97:1230-6.

15. Hesselstrand R, Ekman R, Eskilsson J, Isaksson A, Scheja A, Ohlin AK, Akesson A. Screening for pulmonary hypertension in systemic sclerosis: the longitudinal development of tricuspid gradient in 227 consecutive patients, 1992-2001. Rheumatology (Oxford) 2005;44:366-71.

16. Galasko GI, Lahiri A, Barnes SC, Collinson P, Senior R. What is the normal range for N-terminal pro-brain natriuretic peptide? How well does this normal range screen for cardiovascular disease? Eur Heart J 2005;26:2269-76.

17. Maeder MT, Mariani JA, Kaye DM. Hemodynamic determinants of myocardial B-type natriuretic peptide release: relative contributions of systolic and diastolic wall stress. Hypertension 2010 Oct;56:682-9.

18. Vonk Noordegraaf A, Westerhof N. Right ventricular ejection fraction and NT-proBNP are both indicators of wall stress in pulmonary hypertension. Eur Respir J 2007;29:622-3.

19. Richards AM. Serial measurements of plasma B-type natriuretic peptides: what do they tell us? J Am Coll Cardiol 2008;52:1004-5.

20. Sitbon O, Humbert M, Nunes H, Parent F, Garcia G, Herve P, Rainisio M, Simonneau G. Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: prognostic factors and survival. J Am Coll Cardiol 2002;40:780-8.

21. Benza RL, Miller DP, Gomberg-Maitland M, Frantz RP, Foreman AJ, Coffey CS, Frost A, Barst RJ, Badesch DB, Elliott CG, Liou TG, McGoon MD. Predicting survival in pulmonary arterial hypertension: insights from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL). Circulation 2010;122:164-72.

22. Humbert M, Sitbon O, Chaouat A, Bertocchi M, Habib G, Gressin V, Yaici A, Weitzenblum E, Cordier JF, Chabot F, Dromer C, Pison C, Reynaud-Gaubert M, Haloun A, Laurent M, Hachulla E, Cottin V, Degano B, Jais X, Montani D, Souza R, Simonneau G. Survival in patients with idiopathic, familial, and anorexigen-associated pulmonary arterial hypertension in the modern management era. Circulation 2010;122:156-63.

23. McLaughlin VV, Presberg KW, Doyle RL, Abman SH, McCrory DC, Fortin T, Ahearn G. Prognosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest 2004;126:78S-92S.

24. McLaughlin VV, Suissa S. Prognosis of pulmonary arterial hypertension: the power of clinical registries of rare diseases. Circulation 2010;122:106-8.

25. McLaughlin VV, McGoon MD. Pulmonary arterial hypertension. Circulation 2006;114:1417-31.


6

C H A P T E R 7

91

ABSTRACT

J Cardiovasc Magn Reson. 2008 Nov 5;10(1):5

Background Cardiovascular magnetic resonance (cmr) is an emerg-ing modality in the diagnosis and follow-up of patients with pulmo-nary arterial hypertension (pah). Derivation of stroke volume (sv) from the pulmonary flow curves is considered as a standard in this respect. Our aim was to investigate the accuracy of pulmonary artery (pa) flow for measuring sv. Methods Thirty-four pah patients underwent both cmr and right-sided heart catheterisation. cmr-derived sv was measured by pa flow, left (lv) and right ventricular (rv) volumes, and, in a subset of nine patients also by aortic flow. These sv values were compared to the sv obtained by invasive Fick method.Results For sv by pa flow versus Fick, ρ = 0.71, mean difference was –4.2 ml with limits of agreement 26.8 and –18.3 ml. For sv by lv vol-umes versus Fick, ρ = 0.95, mean difference was –0.8 ml with limits of agreement of 8.7 and –10.4 ml. For sv by rv volumes versus Fick, ρ = 0.73, mean difference –0.75 ml with limits of agreement 21.8 and –23.3 ml. In the subset of nine patients, sv by aorta flow versus Fick yielded ρ = 0.95, while in this subset sv by pulmonary flow versus Fick yielded ρ = 0.76. For all regression analyses, p < 0.001.Conclusion In conclusion, sv from pa flow has limited accuracy in pah patients. lv volumes and aorta flow are to be preferred for the measurement of sv. ◦


G.J. Mauritz,a J.T. Marcus,b A. Boonstra,c P.E. Postmus,a N. Westerhof,a,c A. Vonk-Noordegraaf a

Non-invasive stroke volume assessment in patients with pulmonary arterial hypertension: left-sided data mandatory

7.1 INTRODUCTION

IN PULMONAry ArTErIAL HyPErTENSION (pah), cardiovascular magnetic reso-nance (cmr) has been proposed as a standard for the assessment of right ventricular function and characteristics of the pulmonary vascular bed.1,2 Accurate assessment of

stroke volume (sv) by cmr is critical in this respect, since earlier studies revealed that sv is closely related to prognosis and that a change in sv reflects treatment effects.3,4 Since most

92 Chapter 7

of the cmr protocols used in pah measured pulmonary artery flow,5,6 sv can be assessed by measuring flow in the main pulmonary artery (pa). Previous studies have shown that this method is accurate to measure sv from pa flow in healthy subjects.7-9 Whether this also holds true in pah is questionable, since the velocity profile in pah is non-laminar, in contrast to the profile in healthy subjects.10-13 For this test of accuracy, a clinical standard is required. This standard is provided by the measurement of sv by the direct Fick principle during right heart catheterisation (rhc).14 However, this is an invasive procedure and thus not well suited for either screening or frequently repeated follow-up measurements. There-fore the aim of the present study is to assess the accuracy of the pa flow by cmr for measur-ing sv in pah patients, by comparing sv from this pa flow with the sv assessed by the Fick method. In addition, other cmr-derived sv measures from aorta flow and cine imaging in the same patients will also be compared versus the Fick-derived sv.

7.2 METHODS

7.2.1 Patients This study was approved by the institutional review Board on research Involving Human Subjects of the vu University medical center, and all participants gave written informed consent. Between January 2004 and April 2007, a total of 34 patients who were given a final diagnosis of pah after a complete diagnostic workup including rhc, underwent cmr. rhc and cmr were performed within 12 hours. The study group consisted of 23 female (68%) and 11 male (32%) patients with a mean age of 45 ± 17 years (sd), and an age range of 20-84 years.

7.2.2 CMR imaging protocol

CMR flow measurementssv was measured using both phase contrast flow and volumetric methods as described be-low. cmr was performed with a Siemens 1.5 T Sonata whole body scanner (Siemens Medi-cal Solutions, Erlangen. Germany), equipped with a phased-array body coil. Phase-contrast cmr was acquired during continuous breathing, with a gradient echo mr sequence, with velocity encoding perpendicular to the imaging plane and a velocity sensitivity of 120 cm/sec. The flow sequence was run with the following parameters: orientation orthogonal to the main pa, slice thickness 6 mm, field of view 240 × 320 mm2, matrix size 140 × 256, echo time 4.8 ms, repetition time 11 ms, temporal resolution 22 ms, flip angle 25°. To explore whether there is inaccurate determination of flow-derived sv due to inherent technical lim-itations of phase-contrast cmr, aortic flow was also measured in a subset of nine patients, approximately 2-4 cm above the aortic valve and distal to the coronary arterial ostia, and the aortic flow-derived sv was also compared to the Fick-derived sv. After the flow images were acquired, a 7 liter bottle containing H2O, with per liter 1.25g NiSO4.6H2O + 2.6g NaCl (‘phantom’) was then imaged with identical imaging parameters, to serve as correction for the background phase error in the main pa and aorta.15

93Non-invasive stroke volume assessment

7

RV and LV volumetric measurementsThe short-axis slices needed to encompass the entire left and right ventricle volumes to measure right ventricular (rv) and left ventricular (lv) volumes-derived sv, were obtained by steady state free precession imaging: 11 phase encoding lines per heart beat (which means 11 lines per segment), flip angle 60°, slice thickness 6 mm, slice gap 4 mm, temporal resolution 36.3 ms and retrospectively ecg gated. Acquisition was in breathhold, acquisi-tion time was 14 heartbeats.

7.2.3 Imaging analysis

CMR flow measurements cmr post-processing was performed using the ‘flow’ software package (Department of ra-diology, Leiden University Medical Center, Leiden, The Netherlands). The heart rate during cardiac cmr was recorded from the images. No aliasing due to high peak systolic veloci-ties was encountered. The contours of the mean pulmonary artery mpa were automatically traced, with manual correction when necessary, simultaneously on magnitude and velocity-map images of all reconstructed phases. The software then calculated the velocity in each of the pixels included within the contours. The flow in each pixel (velocity times pixel area) is calculated and the pixel flow within the contour is summed. This is done for every temporal phase, resulting in blood flow as a function of time through the main pa. Following image acquisition, pulmonary blood flow was corrected using the offset values from the phantom (Figure 7.1). After background correction with the phantom (Figure 7.2), we calculated pa forward flow volume as the area under the curve until the zero-crossing of the downward limb. We checked whether there was more than 5% reverse flow that would indicate any pulmonary regurgitation (pr). However we did not observe any pr, thus we did not exclude any pr patient. The same was done for the aorta flow.

RV and LV volumetric measurements We measured rv and lv volumes as follows.16 The endocardial contours of the ventricles were manually traced on short-axis slices in end-diastolic (first cine phase of the r-wave triggered acquisition) and end-systolic (image phase with smallest cavity area) phases us-ing commercial software (mass software package, version 5.0, Figure 7.3). In the present study, papillary muscles were excluded from manual tracings of the endocardial contours of the right end left ventricle. The ventricular areas were then measured and ventricular volumes calculated by adding the ventricular areas and multiplying by the slice distance. end-diastolic volumes (edv) and end-systolic volumes (esv) were used to calculate lv-and rv volumes-derived sv. One investigator analyzed both phase contrast Flow and volumetric images unaware of stroke volume values measured by direct Fick principle.

7.2.4 Right heart catheterizationDiagnostic right heart catheterization (rhc) was performed with a balloon tipped, flow directed 7f Swan-Ganz catheter (131hf7; Baxter Healthcare Corp; Irvine, ca). The direct Fick principle (df) was used. The patient was in stable condition, lying supine and breath-ing room air. right atrial, right ventricular, pulmonary artery and pulmonary capillary

94 Chapter 7

wedge pressures were measured. Heart rate was monitored continuously. Average steady state oxygen consumption was obtained at the same time during rhc using a timed collec-tion of expired air to measure oxygen consumption by an on-line analyzer (Vmax 229, Sen-sormedics, yorba Linda, usa) connected to a mouth piece. To ensure accuracy, the system was calibrated before each study and the values were time-averaged over at least 5 minutes. Simultaneous arterial and mixed venous blood samples were then drawn for measurement

BA

C D

PA

Phantomfor PA

Average �ow 315 mL/s

Flow 21 mL/s

Figure 7.1 Magnitude (A) and phase image (B) of a double-oblique gradient-echo phase-contrast CMR (repetition time ms/echo time ms 11/4.8, flip angle 25°, section thickness 6 mm, matrix 140 x 256), perpendicular to pulmonary trunk. Region of interest (ROI) placement is shown for measuring pulmonary artery (PA) flow. The images correspond to the time frame when maximum flow is measured in the PA. ROI placement for the cor-responding phantom magnitude (C) and velocity (D) image is shown to correct velocity offset error in PA.

–50

50

150

250

350Ejection time

1500 300 450 600 750 900

Time (ms)

Flow

(ml/s

)

Uncorrected MPA �owPhantom corrected MPA �ow

Figure 7.2 Flow curve of the main pulmonary artery (MPA) during a cardiac cycle in a PAH patient. Uncorrected stroke volume (SV) is obtained by integration of the flow curve during ejection time and is 58 ml, ejection time is 0.323 s and phantom offset value is 21.13 ml/s. Phantom corrected SV is 58 – 0.323 × 21.13 = 51 ml.


7

of arterial oxygen saturation (SaO2) and haemoglobin (Hb) concentration. Arterial blood samples were obtained through either a radial or femoral arterial puncture and the mixed venous blood samples were obtained from the distal port of the pulmonary artery catheter. The cardiac output (co) was then calculated by dividing the average oxygen consumption (VO2) value with the difference between the concentration of arterial oxygen (CaO2) and concentration of mixed venous oxygen content (CvO2). Eventually, sv (ml) was determined by dividing co (l/min) by hr (beats/min).

7.2.5 Intra-observer and inter-observer variability Interobserver variability of the different cmr methods was assessed by a second investigator analyzing all of data sets. Intraobserver variability was analyzed by assessing svs obtained by the different cmr methods twice by one observer. The two assessments were separated by a one month period, and the observer was blinded for his previous results.

7.2.6 Data analysis The results are expressed as mean ± sd. For each measurement of sv, the values for sv ob-tained invasively direct Fick and by the cmr-based methods were compared by regression analysis. Bland-Altman analysis was used to compare the degree of agreement between the three cmr methods and the Fick method for sv measurement.17 Bias was defined as the mean value of the differences between cmr methods and the Fick principle. Precision was defined as 1 standard deviation (sd) of the differences and limits of agreement as the bias ± 2 sds reported as ml. The percentage error was calculated as the ratio of 2 times the sd to mean sv. Tendency toward over- or underestimation of sv by the different cmr-based methods was assessed with the two-tailed Student's t-test for paired data. The mean differ-ence (bias) and the coefficient of variability (cov = sd of repeated measures as % of their mean) were used to assess intra- and interobserver variability. P values less than 0.05 were considered to indicate significant differences. All statistical analyses were performed using GraphPad prism (version 4.0, GraphPad software, San Diego, ca).

A B

RV

LV

Figure 7.3� End-diastolic (A) and end-systolic (B) cardiac short axis double-oblique steady-state free precession CMR images in a PAH patient (repetition time ms/echo time ms 3.3/1.65, flip angle 60°, section thickness 6 mm, matrix 256 x 150). The endocardial boundaries of the RV and left ventricle (LV) are traced for calculation of end-diastolic volume (EDV) and end-systolic volume (ESV). EDV and ESV are calculated by summation of the product (area x slice distance) for all slices. SV is then given by SV = EDV – ESV for RV and LV.

96 Chapter 7

7.3 RESULTS

The demographic data, causes of pah and resting clinical hemodynamic measurements of the 34 pah patients are shown in Table 7.1. Twenty-nine patients (out of 34) had measurable Tricuspid regurgitation by echo, and 4 patients had right-to-left shunting based on the pas-sage of contrast bubbles through a patent foramen ovale, observed by echo-cardiography.

Figure 7.4 shows the results of the three different cmr methods to measure sv in com-parison with sv measurements assessed by the direct Fick principle. From this figure it is clear that the pa flow-derived sv values and the rv volumes-derived sv values show only a limited correlation with the Fick standard. lv volumes-derived sv showed the best correla-tion with Fick.

Figure 7.5 shows the Bland-Altman plots of the difference between the cmr based meth-ods and the direct Fick against the mean of both values. Mean (± sd), mean difference (bias), sd of the difference (precision), limits of agreement, and percentage error accord-ing to different cmr based sv measurements are presented in Table 7.2. As shown in this Table 7.2, the limits of agreement of the pa flow-derived sv with Fick present almost the same range of dispersion as was found when rv volumes-derived sv was compared with Fick. The limits of agreement between Fick and lv volumes-derived sv are narrow. Bland–Altman analysis demonstrated a small degree of underestimation of sv by pa flow with mean difference of 4.2 ml, the underestimation was statistically significant (p = 0.039). If those patients with echo-derived right-to-left shunting were excluded, than the correlation coefficient between pa Flow-derived sv and Fick-derived sv increased to R² = 0.56 and the bias decreased to a non-significant level of 1.9 ± 3.7 ml. The slight overestimation of sv by rv and lv volumes, 0.75 ml and 0.8 ml respectively, was not significant. The sv assessed from the aortic flow (Figure 7.6A) in the subset of 9 patients showed a tight relation with the Fick-derived sv. The comparison with Fick was also made for the pa flow-derived sv, in the same subset of patients (Figure 7.6B). Table 7.3 displays the intra-observer and inter-observer variability data. Intra-observer and interobserver bias for pa flow (0.8 - 0.9ml), as well as for aorta flow (0.7-1.3 ml), rv volumes (2.1-2.6 ml) and lv volumes (1.8-2.57 ml), were negligible. Intra-observer and inter-observer variability was sufficiently low for all dif-ferent cmr methods for sv assessment, as shown by cov values (2.7-13.3%).

7.4 DISCUSSION

7.4.1 General discussionTo our knowledge this study is the first to assess the accuracy of sv derived non-invasively from pa flow, rv volumes, aorta flow and lv volumes by comparing these values with stroke volume assessed by means of the invasive Fick method in a group of pah patients. Our results showed that sv derived non-invasively from pa flow and rv volumes were in poor agreement with the Fick-derived sv. By contrast sv measurement from lv volumes and aortic flow showed good agreement with Fick. Although sv is highly variable in healthy subjects, an earlier study18 revealed that sv is fixed in pah and does not even change during


7

exercise. Thus although the invasive and cmr measurements were not performed synchro-nously, we expect similar sv values for each patient. The direct Fick principle was chosen as the method of reference because it is considered as a standard for measuring sv in patients with pah.14

Table 7.1 Baseline characteristics of the patients (n = 34)


Age (y) 54 ± 17

Sex (m/f) 11/23

Diagnosis, n (%)

Idiopathic PAH 14

Familial PAH 6

PAH associated with

Connective tissue disease 5

HIV infection 6

Portal hypertension 3

Functional status


II 18

III 12

IV 4

6-minute walking distance (m) 441 ± 109

Hemodynamics

Heart rate (bpm) 81.5 ± 12

mPAP (mmHg) 45 ± 10

sPAP (mmHg) 74 ± 21

dPAP (mmHg) 29 ± 11

Stroke volume, Fick (ml) 58.8 ± 16

Cardiac output, Fick (L/min) 4.7 ± 1.3

Cardiac index (L/min/m2) 2.5 ± 0.6

PVR (dyn·s·cm‒5) 777 ± 402

RAP (mmHg) 7 ± 4.5

PCWP (mmHg) 6.9 ± 3.3

Sa02 (%) 95 ± 2

Sv02 (%) 63 ± 8

Values expressed as n or mean ± SD. NYHA = New York Heart Association, mPAP = mean pulmonary artery pres-sure, sPAP = systolic pulmonary artery pressure, dPAP = diastolic pulmonary artery pressure, PVR = pulmonary vascular resistance, RAP = right atrial pressure, PCWP = pulmonary capillary wedge pressure, SaO2 = arterial oxygen saturation, SvO2 = mixed venous oxygen saturation.

98 Chapter 7

7.4.2 SV assessment with PA flow sv determined from the pulmonary artery flow curve was poorly related to stroke vol-ume assessed by means of Fick. There are several explanations for this discrepancy: First, we observed non-laminar velocity profiles during systole (Figure 7.1B) in the main pa of pah patients, which can suggest turbulent or helical flow patterns. Phase-contrast measure-ments are optimized for laminar flow, whereas in turbulent flow patterns the precision of the cmr flow measurements declines.19, 20 In case of helical flow, motion in the non-velocity encoding directions may also lead to phase-shifts which are not related to through-plane flow. Second, differences may result from cardiac right-to-left shunting, which causes un-derestimation of sv determination by pa flow. This is supported by our observation that sv using pa flow showed statistically significant underestimation of 4.2 ml; this underestima-tion disappeared by excluding those patients with a right-to-left shunt.

Another potential explanation is an inherent inaccuracy of the flow measurement by means of phase-contrast cmr. However, this explanation is unlikely, since in the subset of nine patients where aorta flow was also measured, a much better correlation with the direct Fick is found (Figure 7.6). The accuracy of the aorta flow measurements for sv has already been validated by comparison with invasive sv measurements by Hundley et al21 in 23 subjects.

7.4.3 SV assessment with RV volumes Several factors may explain the errors in the stroke volume measurements derived from the volumetric measures of the right ventricle. First, in the short axis view, in the most basal rv slice, with the rv outflow tract and the inflow region with the tricuspid valve, usually it is more complicated to define the rv contour than in the case of the left ventricle. Because of the large area of this basal slice, errors in choosing the exact rv contour can render inac-

Table 7.2 Bland-Altman analysis of CMR based stroke volume measurements

Mean ± SD Bias Precision 95% Limits of agreement Error (%)

Pulmonary artery flow (ml/s) 55.2 ± 13.1 –4.2 11.48 –18.3 to 26.8 42

Aorta flow* (ml/s) 51.4 ± 13.3 –2.3 3.85 –9.8 to 5.2 14

RV volumes (ml) 59.5 ± 15.7 –0.75 11.49 –23.3 to 21.8 39

LV volumes (ml) 59.6 ± 16.2 –0.8 4.87 –10.4 to 8.8 16

*Aorta flow in 9 patients. RV = right ventricle, LV = left ventricle.

Table 7.3� Intra- and inter-observer variability of CMR based measurements of stroke volume

Intra-observer Inter-observer

Bias 95% Limits of agreement CoV (%) Bias 95% Limits of

agreement CoV (%)

Pulmonary artery flow (ml/s) –0.8 ± 1.7 –4.1 to 2.4 3.1 –0.9 ± 2.2 –5.3 to 3.5 4.0

Aorta flow* (ml/s) –0.7 ± 2.6 –5.4 to 5.0 2.7 –1.3 ± 2.1 –4.1 to 5.2 5.1

RV volumes (ml) –2.1 ± 5.2 –12.3 to 8.1 8.7 –2.6 ± 7.9 –18 to 13.1 13.3

LV volumes (ml) –1.8 ± 3.2 –8.0 to 4.4 5.4 –2.6 ± 3.8 –10.1 to 4.9 6.4

Bias and limits of agreement determined according to the method of Bland and Altman. CoV = coefficient of variability, RV = right ventricle, LV = left ventricle.


7

20

40

60

80

100 y = 0.64 x + 17.04R2 = 0.51

SV: P

A F

low

(ml)

20 40 60 80 10020

40

60

80

100 y = 0.96 x + 2.96R2 = 0.91

SV: Direct Fick (ml)

SV: L

V vo

lum

es (m

l)

C

A

20

40

60

80

100 y = 0.72 x + 17.12R2 = 0.54

SV: R

V vo

lum

es (m

l)

B

Figure 7.4� Linear regression analysis of the correlation between stroke volume from direct Fick and from PA flow with phase-contrast MRI (A), from RV volumetric assessment (B) and from LV volumetric assessment (C) in 34 PAH patients. Dashed lines = 95% CIs.

curate volume estimates and thus inaccurate sv measurements. Second, evaluation of right ventricular volumes in patients with pah is difficult because of the complex anatomy of the right ventricle. Third rv endocardial boundary delineation is more complicated because of the considerably more trabeculation of the rv in comparison with the lv.22 These fac-tors may be improved with further software development to automate boundary detection and also assist in the selection of the basal slice during analyses. Fourth, in patients with pulmonary hypertension tricuspid regurgitation (tr) is common. Thus, with considerable tr, the volumetric sv overestimates the actual sv,23 because it is impossible to differentiate between the volume that moves back through the tricuspid valves and forward though the pulmonary valves.

7.4.4 SV assessment with LV volumes The correlation between the sv derived from lv end-diastolic and end-systolic volumes and the sv from Fick was very tight (Figure 7.5C). This is presumably so because for the lv it is rather easy to delineate the endocardial border and to select the basal slice. Additionally

100 Chapter 7

B

A

–30

–15

0

15

30 +2SD

–2SD

Bias

Fick

– P

A F

low

(ml)

–30

–15

0

15

30+2SD

–2SD

Bias

Fick

– R

V vo

lum

e (m

l)

20 40 60 80 100

–30

–15

0

15

30

+2SD

–2SD

Bias

Mean SV (ml)

Fick

– L

V vo

lum

e (m

l)C

Figure 7.5� Bland-Altman plots showing the degree of agreement between direct Fick and PA Flow (A), RV volumes (B) and LV volumes (C) in 34 PAH patients. Blue dashed line indicates the bias and red dashed lines indicate the upper and lower limits of agreement.

BA

20 40 60 80 100

y = 1.0 x – 6.5R2 = 0.58

SV: Direct Fick (ml)

Pulmonary artery

20

40

60

80

100 y = 1.02 x + 3.76R2 = 0.91

SV: F

low

(ml)

SV: Direct Fick (ml)20 40 60 80 100

Aorta

Figure 7.6 In a subset of nine patients both aorta and pulmonary flow was measured showing excellent correla-tion between the direct Fick and the aorta flow (A) in contrast to the direct Fick and the pulmonary flow (B) for stroke volume (SV). Dashed lines = 95% CIs.


7

there was no sign of systematic errors for low or high values indicating that the sv measure from lv volumes is valid for a wide range of sv values. There were some high sv values in three patients with porto pulmonary syndrome which can be explained by the hyperdy-namic circulation feature in these patients.24

7.4.5 Intra-observer and inter-observer variability The results of the current study indicated that the intra- and interobserver variability of sv measurements using pa flow, aortic flow and lv volumes measurements were low, showing small differences and low coefficients of variability. A larger interobserver and intraobserv-er variability was shown for the rv volumes. This could be explained by factors mentioned above. Together with the poor agreement of rv-derived sv versus Fick, this advises against the use of rv volumes for sv determination.

7.4.6 Clinical implications The accurate non-invasive sv assessment has clinical importance because monitoring of sv in pah patients now becomes feasible as a routine method for monitoring the effects of medical treatment and the follow-up of patients non-invasively. The results of this study have important implications for pah patients. The sv measurement using pa flow is less ac-curate in comparison to the sv derived from lv volumes and aorta flow. Therefore, lv vol-umes and aorta flow should be the methodology of choice in this patients group to measure sv accurately. Consequently, it is advisable to include in the cmr basic protocol a stack of contiguous short axis slices covering the full extend of the left ventricle, and an aortic flow measurement for accurate sv assessment.

7.4.7 Limitations We acknowledge some limitations of our study including the fact that catheter studies and cmr examinations could not be performed simultaneously. However, the results of this study demonstrate that sv derived from lv volumes correlate well with measurements ob-tained using the direct Fick method, supporting earlier findings that stroke volume does not vary over a short period of time in pah, making this parameter of value to monitor pah patients. Furthermore we acknowledge the small sample size of the aorta flow measure-ments. Finally, the present study does not provide conclusive evidence of the cause of the discrepancy between the pa flow and the direct Fick.

7.5 CONCLUSION

In patients with pah, taking the direct Fick as a standard invasive method, non-invasive derivation of sv should preferably be carried out using aortic flow or lv volumes. sv assess-ment with pa flow or rv volumes shows limited accuracy. ■

102 Chapter 7

References

1. Torbicki A. Cardiac magnetic resonance in pulmonary arterial hypertension: a step in the right direction. Eur Heart J 2007 May;28(10):1187-9.

2. McLure LE, Peacock AJ. Imaging of the heart in pulmonary hypertension. Int J Clin Pract Suppl 2007 September;(156):15-26.

3. van Wolferen SA, Marcus JT, Boonstra A, Marques KM, Bronzwaer JG, Spreeuwenberg MD et al. Prognostic value of right ventricular mass, volume, and function in idiopathic pulmonary arterial hypertension. Eur Heart J 2007 May;28(10):1250-7.

4. Roeleveld RJ, Vonk-Noordegraaf A, Marcus JT, Bronzwaer JG, Marques KM, Postmus PE et al. - Effects of epoprostenol on right ventricular hypertrophy and dilatation in pulmonary hypertension. Chest 2004 February;125(2):572-9.

5. Kreitner KF, Kunz RP, Ley S, Oberholzer K, Neeb D, Gast KK et al. - Chronic thromboembolic pulmonary hypertension - assessment by magnetic resonance imaging. Eur Radiol 2007 January;17(1):11-21.

6. Marcus JT, Vonk NA, De Vries PM, Van Rossum AC, Roseboom B, Heethaar RM et al. - MRI evaluation of right ventricular pressure overload in chronic obstructive pulmonary disease. J Magn Reson Imaging 1998 September;8(5):999-1005.

7. Kondo C, Caputo GR, Semelka R, Foster E, Shimakawa A, Higgins CB. Right and left ventricular stroke volume measurements with velocity-encoded cine MR imaging: in vitro and in vivo validation. AJR Am J Roentgenol 1991 July;157(1):9-16.

8. Firmin DN, Nayler GL, Klipstein RH, Underwood SR, Rees RS, Longmore DB. - In vivo validation of MR velocity imaging. J Comput Assist Tomogr 1987 September;11(5):751-6.

9. Caputo GR, Kondo C, Masui T, Geraci SJ, Foster E, O’Sullivan MM et al. - Right and left lung perfusion: in vitro and in vivo validation with oblique-angle, velocity-encoded cine MR imaging. Radiology 1991 September;180(3):693-8.

10. Bogren HG, Buonocore MH. - Complex flow patterns in the great vessels: a review. Int J Card Imaging 1999 April;15(2):105-13.

11. Bogren HG, Klipstein RH, Mohiaddin RH, Firmin DN, Underwood SR, Rees RS et al. - Pulmonary artery distensibility and blood flow patterns: a magnetic resonance study of normal subjects and of patients with pulmonary arterial hypertension. Am Heart J 1989 November;118(5 Pt 1):990-9.

12. Okamoto M, Miyatake K, Kinoshita N, Sakakibara H, Nimura Y. - Analysis of blood flow in pulmonary hypertension with the pulsed Doppler flowmeter combined with cross sectional echocardiography. Br Heart J 1989 April;51(4):407-15.

13. Reuben SR, Swadling JP, de Lee GJ. - Velocity profiles in the main pulmonary artery of dogs and man, measured with a thin-film resistance anemometer. Circ Res 1970 December;27(6):955-1001.

14. Hoeper MM, Maier R, Tongers J, Niedermeyer J, Hohlfeld JM, Hamm M et al. Determination of cardiac output by the Fick method, thermodilution, and acetylene rebreathing in pulmonary hypertension. Am J Respir Crit Care Med 1999 August;160(2):535-41.

15. Chernobelsky A, Shubayev O, Comeau CR, Wolff SD. Baseline correction of phase contrast images improves quantification of blood flow in the great vessels. J Cardiovasc Magn Reson 2007;9(4):681-5.

16. Marcus JT, Gotte MJ, DeWaal LK, Stam MR, Van der Geest RJ, Heethaar RM et al. The influence of through-plane motion on left ventricular volumes measured by magnetic resonance imaging: implications for image acquisition and analysis. J Cardiovasc Magn Reson 1999;1(1):1-6.

17. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986 February 8;1(8476):307-10.

18. Holverda S, Gan CT, Marcus JT, Postmus PE, Boonstra A, Vonk-Noordegraaf A. Impaired stroke volume response to exercise in pulmonary arterial hypertension. J Am Coll Cardiol 2006 April 18;47(8):1732-3.

19. Bakker CJ, Hoogeveen RM, Viergever MA. Construction of a protocol for measuring blood flow by two-dimensional phase-contrast MRA. J Magn Reson Imaging 1999 January;9(1):119-27.

20. Evans AJ, Blinder RA, Herfkens RJ, Spritzer CE, Kuethe DO, Fram EK et al. Effects of turbulence on signal intensity in gradient echo images. Invest Radiol 1988 July;23(7):512-8.


7

21. Hundley WG, Li HF, Hillis LD, Meshack BM, Lange RA, Willard JE et al. Quantitation of cardiac output with velocity-encoded, phase-difference magnetic resonance imaging. Am J Cardiol 1995 June 15;75(17):1250-5.

22. Grothues F, Moon JC, Bellenger NG, Smith GS, Klein HU, Pennell DJ. Interstudy reproducibility of right ventricular volumes, function, and mass with cardiovascular magnetic resonance. Am Heart J 2004 February;147(2):218-23.

23. Hoeper MM, Tongers J, Leppert A, Baus S, Maier R, Lotz J. Evaluation of right ventricular performance with a right ventricular ejection fraction thermodilution catheter and MRI in patients with pulmonary hypertension. Chest 2001 August;120(2):502-7.

24. Hoeper MM, Krowka MJ, Strassburg CP. Portopulmonary hypertension and hepatopulmonary syndrome. Lancet 2004 May 1;363(9419):1461-8.

C H A P T E R 8

105

ABSTRACT

J Magn Reson Imaging. 2011 Jun;33(6):1362-8

Purpose To evaluate if early onset of retrograde flow in the main pul-monary artery is a characteristic of pulmonary arterial hypertension (pah).Materials and methods Fifty-five patients with suspected pulmonary hypertension (ph) underwent right-sided heart catheterization and retrospectively ecg-gated mr phase-contrast velocity quantification in the main pulmonary artery. Pulmonary hypertension was defined by a mean pulmonary artery pressure being larger than 25 mmHg. The onset time of the retrograde flow relative to the cardiac cycle dura-tion (relative onset time = rot) was compared with mean pulmonary artery pressure.Results By the catheterization, 38 patients were identified as having pah. The rot for these pah patients was significantly different from those found in the 17 non-ph subjects (0.14 ± 0.06 vs. 0.37 ± 0.06, p < 0.001). The mean pulmonary artery pressure was related to the rot (R2 = 0.62, p < 0.001) and could be estimated from the rot with a standard deviation of 11.7 mmHg. With a cut-off value of 0.25, the rot distin-guished pah patients from non-ph subjects.Conclusion Early onset of retrograde flow in the main pulmonary ar-tery is a characteristic of pulmonary arterial hypertension and is vis-ible by standard mr phase-contrast velocity quantification. ◦

Departments of aPulmonary Diseases and bPhysics and Medical Technology, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam

G.J. Mauritz,a* F. Helderman,b* K. Andringa,a A. Vonk-Noordegraaf,a J.T. Marcusb

Early onset of retrograde flow in the main pulmonary artery is a characteristic of pulmonary arterial hypertension


8.1 INTRODUCTION

PULMONAry HyPErTENSION (ph) is a progressive disease defined by a chroni-cally elevated mean pulmonary artery pressure that exceeds 25 mmHg at rest.1 Vari-ous studies have shown that in ph patients the flow pattern in the main pulmonary

artery differs from that seen in healthy people.2–6 One of the main findings is the appear-ance of retrograde flow at the right dorsal side of the main pulmonary artery. Okamoto et al. suggested that this indicates the appearance of a vortex.6

106 Chapter 8

recently, reiter et al. found that the vortex duration is related to the mean pulmonary arterial pressure (r2 = 0.88).2 Long vortex duration may be associated with early onset of retrograde flow. If so, then early onset of retrograde flow could indicate the presence of ph. Many patients which are suspected of ph, undergo mr phase-contrast velocity quantifica-tion in the main pulmonary artery for the assessment of stroke volume.7,8 The measured velocity images show retrograde flow if present. The question is whether early onset of this retrograde flow indicates the presence of ph and what cut-off value should be used. There-fore, we conducted a study in pulmonary arterial hypertension (pah) patients and subjects suspected of having ph, in whom both right heart catheterization and mri are performed.

8.2 MATERIALS AND METHODS

8.2.1 Study population Our sample comprised of patients diagnosed with pulmonary arterial hypertension (pah) and non-ph subjects. ph was diagnosed using right heart catheterization, whereby an mpap of > 25 mmHg indicated ph. Non-ph subjects were suspected of ph but turned out to have a mean pulmonary artery pressure (mpap) smaller than 25 mmHg. All patients in this study were referred to the vu University Medical Center, from January 2003 until April 2009.

The inclusion criterion was the existence of right heart catheterization and mri velocity quantification measurements within a time interval of three weeks. Exclusion criteria were receipt of pneumonectomy and poor mri image quality. Some patients with ph can have characteristics other than pressure that can affect pulmonary flow. Therefore, we decided for clarity to confine to only patients with pulmonary arterial hypertension.

8.2.2 Right heart catheterization right-sided heart catheterization (rhc) was performed with a balloon tipped, flow directed 7 French Swan-Ganz catheter (131HF7, Baxter Healthcare Corp, Irvine, ca). The patients were in stable condition, lying supine and breathing room air, while their heart rate was continuously monitored. Measurements were made of mean right atrial pressure, right ventricular pressure, pulmonary artery pressure and pulmonary capillary wedge pressure (pcwp). Cardiac output (co) was obtained by thermodilution or by direct Fick method, from arterial and mixed venous oxygen saturation and O2 consumption. Pulmonary vascu-lar resistance was calculated as (mpap-pcwp) / co.

8.2.3 MR Velocity quantificationmri was performed with either a Siemens 1.5T Sonata or 1.5 Tesla (t) Avanto whole body scanner (Siemens Healthcare, Erlangen, Germany) equipped with a phased-array body coil. One investigator performed all mr acquisitions. Phase-contrast velocity imaging was performed without averaging during continuous breathing. We performed a prospectively electrocardiograph (ecg)-gated, spoiled gradient-echo mri sequence, with through-plane velocity encoding and a velocity sensitivity of 120 cm/s. The orientation of the image plane was orthogonal to the main pulmonary artery (Figure 8.1). The flow sequence was run with

107Early onset of retrograde flow in the MPA

8the following parameters: slice thickness 6 mm, field of view 240 × 320 mm2, matrix size 140 × 256, echo time 4.8 ms, repetition time 11 ms, temporal resolution 22 ms, flip angle 25°. If breathing artefacts were obvious, the acquisition was repeated. Phase offset errors were corrected by subtracting the velocity images measured at a stationary water-phantom using the same acquisition settings and an artificial ecg mimicking the patients ecg.

8.2.4 Image analysisA semiautomatic program, based on studies published by Li et al., 9,10 using matlab r2008a (The Mathworks, Inc., Natick, ma) was used to find the lumen boundary of the main pul-monary artery (mpa). The software reduced operator bias by using a minimum cost algo-rithm. This algorithm is very robust to image noise because it can sense the whole lumen boundary at once. Therefore, even if the vessel wall was only partly visible, this algorithm could find the contour that marked the lumen boundary. This tested and well-validated technique was developed for ultrasound images. A summary of the wall detection is shown in Figure 8.2. Our software created thirty starting points randomly around the marks se-

A B

C D

Figure 8.1 A-C: Steady-state free precession MR images to localize the pulmonary trunk (repetition time 3.2 ms, echo time 1.6 ms, flip angle 65°, section thickness 6 mm, matrix 256 × 96). First, an image was prescribed perpendicular to a transversal view (A) that included the pulmonary artery. This resulted in an oblique-sagittal image as shown in B. A third localizer image shown in C was acquired orthogonal to B, and through the pulmo-nary artery and then used together with B to obtain the image plane for the through-plane velocity measurement in the main pulmonary artery (MPA). D: The cross-section of the MPA, acquired as the magnitude image in the flow measurement. This acquisition was with a gradient-echo sequence with through-plane phase-contrast velocity quantification.

108 Chapter 8

lected by the operator. This way, 30 different contours rescinded the influence of the op-erator. From these contours, an average contour was calculated. The middle of the average contour was an estimator for the middle of the vessel in the next image. Again, 30 starting point were generated and the minimal cost algorithm was applied to the second image. Af-ter a contour was found for every image, a filter smoothed all contours whereby preserving the vessel shape. The operator, unaware of rhc results, visually checked all contours and performed all analyses.

8.2.5 Variables in the studyThe contours that marked the lumen boundary during the cardiac cycle were essential to calculate the variables in this study. The cross-sectional area (csa) was the smallest cross-sectional area of the mpa during the cardiac cycle. This was easily found as the smallest area enclosed by any contour. The average distance between the contour and its centre was multiplied by two to get the diameter. The difference in length between the longest and shortest contour was divided by the length of the shortest contour to get mpa strain. The mr velocity images showed velocity perpendicular to the image plane on a pixel-by pixel basis. Therefore, it was possible to locate and quantify retrograde flow. Flow (mL/s) was calculated by summing the product of velocity and pixel size. By keeping track of the flow direction for each pixel, antegrade, retrograde, and net flow were calculated (Figure 8.3).

A B C

D E F

Figure 8.2 Detection of the arterial wall. A: The user only once marks the centre of the vessel and a reference point outside the vessel. B: Sixty radial lines scan the intensity of the magnitude image. C: The sixty radial lines are transformed to an image showing the bright lumen in the bottom half. D: The gradient of signal intensity (SI) in the top-down direction of the image in C. A negative gradient corresponds to low SI (‘dark’), and thus to low cost. E: Cumulative cost image showing the path (yellow line) of minimal cost in the left–right direction of the gradient image in D. F: Transformation of the path of minimal cost to the original magnitude image. The path indicates the inside of the arterial wall.


8

Net flow was identical to the antegrade flow minus the retrograde flow. For some variables in this study, we had to know when the pulmonary valve opened and closed or when flow reached a maximum. These data were automatically detected based on the net flow. Valve opening was identified by net flow reaching 5% of its maximum. Valve closure was identi-fied by zero net flow or a minimum in net flow. Time between valve opening and maximal net flow was the acceleration time. Time between valve opening and valve closure was the ejection time. Onset of retrograde flow was identified by retrograde flow reaching 5% of its maximum. relative retrograde flow (rrf) was the sum of systolic retrograde flow relative to the sum of systolic antegrade flow. Furthermore, we calculated the averaged and peak velocities over the csa during systole. The most important variable in our study was the relative onset time (rot). The rot indicated the onset time of retrograde flow relative to the duration of the cardiac cycle (Figure 8.3). The duration of the cardiac cycle was retrieved from the data that accompanied the mr images.

8.2.6 Location of retrograde flowretrograde flow in the mpa is a complex phenomenon because the amount of flow depends on time and location (Figure 8.4). Furthermore, during the cardiac cycle the mpa stretches and shows some motion. We decided to express location of retrograde flow as one simple variable independent of time. Hence, in the coordination system of the velocity images the location was expressed as an angle between 0 and 360° as commonly done in a circle. To determine this angle data processing was required. The phase-contrast data in the mpa was averaged over the whole cardiac cycle. Hereby, the motion of the mpa was corrected by shifting the centre of the lumen of each phase-image to the same location. Then we calcu-lated the weighted mean of retrograde time averaged flow. Finally, the angle of retrograde flow in relation to the centre of the lumen was calculated.

8.2.7 Statistical analysisDifferences between patient groups were evaluated with an independent sample t-test. We

0.50 0.750.250 1.0–200

–100

0

100

200

300

400

Time (s)

Flow

(ml/

s)

Onset time

Figure 8.3� Antegrade and retrograde flow in the main pulmonary artery during one cardiac cycle. The antegrade flow (blue line) rapidly increases from the beginning of the cardiac cycle and reaches a maximum during systole. The retrograde flow (red line) starts after an onset time and often shows two peaks. The first peak is during systole and was only seen in PAH patients. The second peak is at valve closure and was visible in all subjects. The inset shows the velocity profile perpendicular to the image plane for maximal flow. This illustrates the occur-rence of antegrade and retrograde flow at the same time.

110 Chapter 8

compared mPAP with mri-derived variables. Statistical test included descriptive statistics, independent t-tests, linear regression and roc-curves. All tests were two tailed and a p-value smaller than 0.05 was considered significant. Some data were presented with a 95% confidence interval (ci). In addition to presenting the mean and standard deviation of the location of retrograde flow, a figure visualized the distribution. For this purpose, subjects were classified in intervals of 10 degrees and the distribution was shown in a bar plot and a circular plot. Analyses were performed with spss for Windows (version 15.0.0, spss Inc, Chicago, il).

8.3 RESULTS

8.3.1 Patients’ characteristicsThis study included 38 pah patients and seventeen non-ph subjects. The average time in-terval between mri and rhc was 5 days. As expected, the mpap, systolic pulmonary artery pressure, mean right atrial pressure, pulmonary vascular resistance, mixed venous oxygen saturation, cardiac output, and heart rate at rhc were significantly different between pah patients and non-ph subjects. There were no significant differences in terms of age or sex distribution, body surface area, time interval between procedures, or pulmonary capillary wedge pressure (Table 8.1).

8.3.2 Variables in the study The rot, average velocity, smallest csa and rrf of pah patients were significant different from those found in non-ph subjects (Table 8.2).

PAH

Non-PH

mPAP = 66 mmHg

mPAP = 7 mmHg

t = 22 ms t = 73 ms t = 123 ms t = 174 ms t = 224 ms

Figure 8.4� Examples of phase-contrast velocity quantification in the main pulmonary artery for a PAH patient on the top row and a non-PH subject below. The time specified is the time after opening of the pulmonary valves. All images show an area of 5 × 5 cm. The grey scale indicates the through-plane velocity. White corresponds with antegrade velocities, black corresponds with retrograde velocities. The images show early retrograde flow in the PAH patient and absence of retrograde flow in the non-PH subject.


8

The location of retrograde flowretrograde flow was found at an angle of 261° (95%-ci: 254–268°) with a standard devia-tion of 27°. The angle found in pah patients (264 ± 27°) did not differ from those found in non-ph subjects (254 ± 26°). The location of retrograde flow corresponded to the dorsal side of the pulmonary artery and was slightly to the right side (Figure 8.5).

Table 8.1 Demographic and hemodynamic data

Entire population Without PH PAH p-value*



Age (y) 50 ± 15 54 ±14 49 ±16 0.25

Sex (m/f) 11/44 3/14 8/30 0.9

Time between MRI and RHC (days) 0 ± 5 0 ± 6 0 ± 5 0.42

Body surface area (m2) 1.8 ± 0.3 1.9 ± 0.3 1.8 ± 0.3 0.85

Hemodynamics

Heart rate (bpm) 81 ± 14 74 ± 12 84 ± 13 < 0.05

RAP (mmHg) 6 ± 4 3 ± 2 7 ± 4 < 0.001

sPAP (mmHg) 61 ± 30 26 ± 8 77 ± 21 < 0.001

dPAP (mmHg) 23 ± 13 8 ± 3 30 ± 11 < 0.001

mPAP (mmHg) 39 ± 19 16 ± 5 49 ± 13 < 0.001

PCWP (mmHg) 8 ± 4 7 ± 3 8 ± 4 0.19

Pulmonary vascular resistance (dyn·s·cm‒5) 543 ± 412 134 ± 99 724 ± 365 < 0.001

Cardiac index (l/min/m2) 3.2 ± 1.1 4.1 ± 1.3 2.8 ± 0.8 < 0.001

Mixed venous oxygen saturation (%) 69 ± 7 75 ± 6 66 ± 6 < 0.001

Values expressed as means ± SD. *Comparison between patients without PH and patients with PAH. PH = pul-monary hypertension, MRI = magnetic resonance imaging, RHC = right-heart catheterization, mPAP = mean pul-monary artery pressure, sPAP = systolic pulmonary artery pressure, dPAP = diastolic pulmonary artery pressure, PVR = pulmonary vascular resistance, RAP = right atrial pressure, PCWP = pulmonary capillary wedge pressure.

Table 8.2 Parameters derived from MRI velocity quantification in the pulmonary artery


Average velocity (m/s) 0.23 ± 0.12 0.38 ± 0.10 0.16 ± 0.05 < 0.001

Peak velocity (m/s) 0.69 ± 0.25 0.88 ± 0.22 0.60 ± 0.21 < 0.001

Main pulmonary artery strain (%) 9 ± 8 16 ± 10 6 ± 3 < 0.001

Diameter (mm) 32 ± 6 25 ± 3 36 ± 4 < 0.001

Cross sectional area (mm²) 854 ± 316 487 ± 123 1017 ± 223 < 0.001

Acceleration time (ms) 95 ± 28 107 ± 20 89 ± 29 < 0.05

Acceleration time / ejection time 0.31 ± 0.07 0.34 ± 0.07 0.29 ± 0.07 < 0.05

Relative onset time 0.21 ± 0.12 0.37 ± 0.06 0.14 ± 0.06 < 0.001

Relative retrograde flow 0.06 ± 0.06 0.01 ± 0.0.1 0.09 ± 0.05 < 0.001

Values expressed as mean ± SD. *Comparison between patients without PH and patients with PAH. PH = pulmo-nary hypertension.

112 Chapter 8

Relation of variables with pressure At pulmonary valve closure, the retrograde flow showed a peak in all subjects. In pah pa-tients, the retrograde flow often showed a second peak (Figure 8.3). The rot of retrograde flow was related to mpap (R2 = 0.62, mpap = –125 × rot + 66). Thus, large pressures corre-sponded with small rot (Figure 8.6). The standard deviation of pressure around the linear fit was 11.7 mmHg. There was also a relation between mpap and average velocity (R2= 0.60, mpap = –119 × velocity + 66), csa (R2 = 0.57, mpap = 0.046 × csa) and rrf (R2 = 0.54, mpap = 238 × rrf + 24).

8.3.4 Distinguishing PAH patients from non-PH Subjects by rot and csa cut-off values of 0.25 for rot and 626 mm2 for the csa, gave the best sensitivity and specificity, in our preselected population, to predict pulmonary hyper-tension to be present. By combining rot with csa data, the pah patients could be distin-guished more clearly from the non-ph subjects (Figure 8.7).

8.4 DISCUSSION

We studied the retrograde flow in the pulmonary artery and found a relation between onset time and pulmonary artery pressure. Furthermore, relative onset time of retrograde flow in pah patients was significant smaller than those found in non-ph subjects.

In our study, the cut off value for rot of 0.25 distinguished pah patients from all non-ph subjects (Figure 8.7A). Sanz et al. found that of many variables, average velocity (R2 = 0.53) and csa (R2 = 0.42) related best with pressure.11 The relation between pressure and rot is consistent with, but inferior to the relation between pressure and the vortex duration of reiter et al. (R2 = 0.62 vs. R2 = 0.88). The large variation of pressure around the linear fit between rot and mpap is a limiting factor for using rot as an mpap estimate. The rot is

0

2

4

6

8

10

12

180° 210° 240° 360°330°300°270°

Angle

Num

ber o

f sub

ject

sA B

270°

0°

90°

180° 4 6 8 10 122

Figure 8.5� Location of retrograde flow. A: The histogram shows the location of retrograde flow for every subject. The number of subjects was classified in intervals of ten degrees. B: A circular plot showing the histogram with the location of retrograde flow at the right angle. Related to anatomical directions 0° is the left side, 90° the ventral side, 180° the right side, and 270° the dorsal side.


8

visible by standard mr phase contrast velocity quantification (Figure 8.4). Mild retrograde flow is normal in late systole (2% of the total flow), but in ph patients, the retrograde flow is on average 26% of the total flow.12

Our study confirms that relative retrograde flow was significantly larger in pah patients than in non-ph subjects (9% vs. 1%, p < 0.001). retrograde flow in our study was less than found by Bogren et al. However, our study calculated this figure based on systole only while Bogren et al. considered the whole cardiac cycle. Various investigators have documented this reversal of flow during systole and diastole in patients with ph.2–4,6,12 retrograde flow can show one or two peaks during the cardiac cycle. One peak always appears at end-systole

0.0 0.2 0.40.30.1 0.5 0.6 0.0 0.2 0.40.30.1 0.5 0.60

20

40

60

80

R2 = 0.62

A

PA p

ress

ure

(mm

Hg)

ROT

0.00 0.05 0.10 0.15 0.20 0.25

R2 = 0.54

D

RRF

B

Velocity (m/s)

0

20

40

60

80C

PA p

ress

ure

(mm

Hg)

0 500 1000 1500 2000

CSA

R2 = 0.57

R2 = 0.60

Figure 8.6 Relation of variables with pressure A: Relative onset time of retrograde flow (ROT). B: The average velocity. C: The cross-sectional area (CSA). D: The relative retrograde flow (RRF).

0.0 0.20.1 0.40.3 0.5 0.60

20

40

60

80

PA p

ress

ure

(mm

Hg)

ROT

A B

0 500 1000 15000.0

0.2

0.4

0.3

0.1

0.6

0.5

ROT

CSA (mm2)

PHNon-PH

Figure 8.7 Distinguishing PAH patients from non-PH subjects by ROT and CSA. A: Mean pulmonary arterial pressure as function of ROT. Black dots indicate PAH patients and grey squares indicate non-PH subjects. The vertical line indicates the cut-off value of 0.25. B: By combining ROT with CSA data, the PAH patients can be distinguished more clearly from the non-PH subjects.

114 Chapter 8

and represents valvular regurgitation. In some subjects another peak of retrograde flow appeared at mid systole.3 This mid systolic peak was not reported later, only the peak rep-resenting valvular closing has been detected.2,4 Our study, however, clearly shows the mid systolic peak in patients with pah. The described retrograde flow is presumably a mani-festation of a vortex, in patients with ph.2 The vortex rotates at the same location as where retrograde flow was detected: at the dorsal side of the mpa3,4,6 Blood flows forward along the vortex at the ventral side of the mpa.

Flow or more precisely the velocity field is always the result of pressure gradients. Just after opening of the pulmonary valves, the pressure in the right ventricle is larger than pulmonary pressure. This pressure gradient causes blood to accelerate. Obviously, blood reaches maximal velocity during systole just before the pressure gradient switches. The highest forward veloc-ity is close to the middle of the vessel, while close to the wall velocity is small. After the mo-ment of maximal forward velocity, the adverse pressure gradient causes blood to decelerate. Because of the large diameter and velocities, flow in the mpa is unsteady (Womersley number > 10). This means that velocity and the pressure gradient are out of phase. While blood in the middle of the vessel still decelerates, blood closer to the wall already accelerates in the other direction toward the ventricle.

The question is why in pah patients retrograde flow starts earlier. Patients with ph usually have an enlarged mpa. This means flow in ph is more unsteady (larger Womersley number). Furthermore, if the csa is much larger than the valve opening then flow will separate into a jet and boundary layer.13 The boundary layer in an enlarged mpa is expected to be much larger than in a normal mpa. This favors the start of retrograde flow. Furthermore, the pressure gradient possibly switches earlier in ph. Two mechanisms for this early switch of the pressure gradient are proposed. First, because of the conservation of mass, velocity has to slow down as blood proceeds through the widened mpa. The velocity can only change due to pressure gradients. Therefore, the widened mpa adds an adverse pressure gradient. As a result, the total pressure gradient will switch earlier. Second, due to the elastic properties of the vessel wall, a pulse wave travels down the mpa and returns to the ventricle after reflecting in the periphery. The reflected pulse wave adds an adverse pressure gradient to the pressure in the mpa.14 Be-cause the wave speed is proportional to pressure15 and inverse proportional to compliance,16 the pulse wave returns faster in ph. Therefore, the pressure gradient will switch earlier.

Streamline

Boundary layer

Vessel wall

Velocity vector

Velocity pro�le antegrade �ow

Velocity pro�le retrograde �ow

Figure 8.8 A cross-section of the MPA in axial direction. Drawn are streamlines, the velocity vectors at the image plane, velocity profiles, and the large boundary layer at the inner curve. Retrograde flow in the boundary layer is part of the vortex.


8

The shape of the mpa influences the location of retrograde flow. In a curved vessel, the jet tends to flow straight to the outer wall. Therefore, the boundary layer at the inner curve becomes larger while at the outer wall the boundary layer flattens. The large boundary layer at the inner curve is susceptible to retrograde flow and the formation of a vortex (Figure 8.8).2 In conclusion, early onset of retrograde flow in the main pulmonary artery is a char-acteristic of pulmonary arterial hypertension. ■

AcknowledgmentsThe authors thank Dominique van Ginkel and Nico Westerhof for their assistance.

References

1. McLaughlin VV, Archer SL, Badesch DB, et al. ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association developed in collaboration with the American College of Chest Physicians; American Thoracic Society, Inc.; and the Pulmonary Hypertension Association. J AmColl Cardiol 2009;53:1573–1619.

2. Reiter G, Reiter U, Kovacs G, et al. Magnetic resonance-derived 3-dimensional blood flow patterns in the main pulmonary artery as a marker of pulmonary hypertension and a measure of elevated mean pulmonary arterial pressure. Circ Cardiovasc Imaging 2008;1:23–30.

3. Kondo C, Caputo GR, Masui T, et al. Pulmonary hypertension: pulmonary flow quantification and flow profile analysis with velocity-encoded cine MR imaging. Radiology 1992;183:751–758.

4. Mousseaux E, Tasu JP, Jolivet O, Simonneau G, Bittoun J, Gaux JC. Pulmonary arterial resistance: noninvasive measurement with indexes of pulmonary flow estimated at velocity-encoded MR imaging--preliminary experience. Radiology 1999;212:896–902.

5. Kitabatake A, Inoue M, Asao M, et al. Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique. Circulation 1983;68:302–309.

6. Okamoto M, Miyatake K, Kinoshita N, Sakakibara H, Nimura Y. Analysis of blood flow in pulmonary hypertension with the pulsed Doppler flowmeter combined with cross sectional echocardiography. Br Heart J 1984;51:407–415.

7. Mauritz GJ, Marcus JT, Boonstra A, Postmus PE, Westerhof N, Vonk-Noordegraaf A. Non-invasive stroke volume assessment in patients with pulmonary arterial hypertension: left-sided data mandatory. J Cardiovasc Magn Reson 2008;10:51.

8. Nogami M, Ohno Y, Koyama H, et al. Utility of phase contrast MR imaging for assessment of pulmonary flow and pressure estimation in patients with pulmonary hypertension: comparison with right heart catheterization and echocardiography. J Magn Reson Imaging 2009;30:973–980.

9. Li W, Bosch JG, Zhong Y, et al. Semiautomatic frame-to-frame tracking of the luminal border from intravascular ultrasound. Comput Cardiol 1992;353–356.

10. Li W, von Birgelen C, Di Mario C, et al. Semi-automatic contour detection for volumetric quantification of intracoronary ultrasound. Comput Cardiol 1994;277–280.

11. Sanz J, Kuschnir P, Rius T, et al. Pulmonary arterial hypertension: noninvasive detection with phase-contrast MR imaging. Radiology 2007;243:70–79.

12. Bogren HG, Klipstein RH, Mohiaddin RH, et al. Pulmonary artery distensibility and blood flow patterns: a magnetic resonance study of normal subjects and of patients with pulmonary arterial hypertension. Am Heart J 1989;118(Pt 1):990–999.

13. Talbot L. Fundamentals of fluid mechanics. In: Berger SA, Goldsmith W, Lewis ER, editors. Introduction to bioengineering. New York: Oxford University Press; 1996. p 101–132.

14. Turkevich D, Groves BM, Micco A, Trapp JA, Reeves JT. Early partial systolic closure of the pulmonic valve relates to severity of pulmonary hypertension. Am Heart J 1988;115:409–418.

116 Chapter 8

15. Milnor WR, Conti CR, Lewis KB, O’Rourke MF. Pulmonary arterial pulse wave velocity and impedance in man. Circ Res 1969;25:637–649.

16. Lankhaar JW, Westerhof N, Faes TJ, et al. Quantification of right ventricular afterload in patients with and without pulmonary hypertension. Am J Physiol Heart Circ Physiol 2006;291: H1731–H1737.


8

C H A P T E R 9

119

ABSTRACT Background Pulmonary artery (pa) dilatation is one of the conse-quences of pulmonary arterial hypertension (pah) and is used for noninvasive detection. However, it is unclear how the size of the pa behaves over time and whether it is related to pressure changes. The aim of this study was to evaluate pa size during follow-up in treated patients with pah and whether it reflects pulmonary vascular hemo-dynamics.Methods Fifty-one patients with pah who underwent at least two right-sided heart catheterizations (rhcs) together with cardiac mri (cmr) were included in this study. Another 18 patients who had nor-mal pressure at rhc were included for comparison at baseline. From rhc, we derived pa pressures and cardiac output. From the cmr im-ages we derived pa diameter (pad) and the ratio of the pad and as-cending aorta diameter.Results The pad was significantly larger in patients with pah than in patients without pah (p < 0.001). A ratio of the pad and ascending aor-ta diameter 1.1 had a positive predictive value of 92% for pah. Mean follow-up time was 942 days, and there was a significant dilatation during this period (p < 0.001). The change of the pad did not correlate with the changes in pressure or cardiac output. A moderate correlation with follow-up time was found (ρ = 0.56; p < 0.001).Conclusions A dilatated pa is useful for identifying patients with pah. However, during patient follow-up, progressive dilatation of the pa is independent of the change in pa pressure and cardiac output and might become independent from hemodynamics. ◦

Chest. 2010 Dec;138(6):1395-401


G.J. Mauritz,a* B. Boerrigter,a* J.T. Marcus,b F. Helderman,a P.E. Postmus,a N. Westerhof,a,c A. Vonk-Noordegraafa

Progressive dilatation of the main pulmonary artery is a characteristic of pulmonary arterial hypertension and is not related to changes in pressure


120 Chapter 9

9.1 INTRODUCTION

PULMONAry ArTErIAL HyPErTENSION (pah) is a clinical syndrome charac-terized by an increase in pulmonary vascular resistance (pvr) leading to right-side heart failure and, ultimately, death.1 Diagnosis early in the course of the disease is

difficult because of its nonspecific nature and symptoms, such as dyspnea, exercise intoler-ance, and fatigue. Several secondary effects of abnormally elevated pulmonary artery pres-sure (pap) on right-sided structures, such as pulmonary artery (pa) dilatation and right ventricular hypertrophy, may be helpful in the diagnosis of pah. pa diameter (pad) can be obtained noninvasively and is, therefore, one of the parameters in pah that has been studied throughout the years. Early investigators found through chest radiography reasonable cor-relations between right descending pad and pap.2 After the introduction of helical ct imag-ing, several studies have been performed to measure the pad and showed that an increased main pad is a reliable indicator of pah, especially when the ratio of pad and ascending aorta diameter (rpad /aad) is used.3-9 More recently, using cardiac mri (cmr), various authors showed that pad or rpad /aad is useful in the noninvasive detection of pah. On the basis of these studies, it can be concluded that pad or rpad /aad is useful in the non-invasive detection of pah, but correlations of diameter and pressure vary considerably among the studies possibly because of differences in study populations.10-14

Although pap is supposed to be the driving force to dilate the pa, the influence of other parameters, such as time and flow, is unknown. In addition, it is unknown whether the changes in pap are followed by a similar change in pad. Therefore, the aim of this study is to investigate whether change in pad over time reflects changes in pressure, cardiac output (co), or both in patients with pah.

9.2 METHODS

9.2.1 Study group This study is part of a large prospective study aimed to evaluate the role of mri in pah. This study has institutional research board approval, and all patients gave informed consent to the procedures. Fifty-six patients with pah were initially included in our study. All had a baseline evaluation before the start of therapy and a follow-up evaluation of therapy at least 8 months after. Each evaluation consisted of a right-sided heart catheterization (rhc) and cmr measurement separated on average 2 days from each other. In five patients, the qual-ity of one of the mri images did not permit measurement because of respiratory artefacts. Consequently, 51 patients with pah were included this study. All patients were classified in group 1 of the clinical classification of pah,15 with 41 having idiopathic pah; nine, collagen vascular disease-associated pah; and one, hiv-associated pah. During follow-up, patients were treated according to the standard in pah (Table 9.1). Additionally, for baseline analy-sis, we included 18 patients suspected of having pah who underwent rhc and mri once but had normal pap at rhc (i.e., normotensive patients).

121Progressive dilation of the main pulmonary artery

9

9.2.2 CMR ProtocolIn this study, we used cmr images that were part of the routine clinical evaluation of the pa-tients or part of former studies in this center. cmr was performed with either a Siemens 1.5 t Sonata or Avanto (Siemens Medical Solutions, Erlangen, Germany) whole-body scanner equipped with a phased-array body coil. pad was assessed from magnitude images used for phase contrast imaging because these images were assessed in all patients in this retrospec-tive study (Figure 9.1). The image plane for measuring the ascending aorta (aa) was deter-

A B

C D

Figure 9.1 MRI Images to localize the pulmonary trunk CMR cine. A, B and C are steady-state free precession MR images to localize the pulmonary trunk (repetition time ms/echo time ms 3.2/1.6, flip angle 65°, section thick-ness 6 mm, matrix 256 × 96). First, an image was prescribed perpendicular to a transversal view that included the pulmonary artery (A). The resulting oblique-sagittal image (B) was used to obtain a third image shown in C, acquired along the straight line in B, and used together with B to obtain the image plane for the through-plane velocity measurement in the main pulmonary artery. D shows the cross section of the main PA, acquired as the magnitude image in the flow measurement. This acquisition was with a gradient-echo sequence with through-plane phase-contrast velocity quantification. A also shows the plane orthogonal to the ascending aorta, from which the aortic diameter is measured as used for normalization of the PA size to the aorta.

Table 9.1Treatment of the patients with pulmonary arterial hypertension (n = 51)

Treatment Number of patients*

Ambrisentan 2

Bosentan 36

Sitaxentan 11

Calcium channel blocker 3

Sildenafil 41

Epoprostenol (IV) 20

Treprostenol (subcutaneous) 13

Iloprost (inhaled) 1

*The total number of treatments exceeds the number of patients in the study because of treatment combina-tions and treatment regimen changes.

122 Chapter 9

mined as follows: a set of coronal localizer images was acquired, and then the image that showed the aa was selected. On this coronal image, a set of axial images was acquired that intersected the aa. From these axial images, the image was selected that showed the aa in a circular cross section. This axial plane was at the level of the right pa. During the follow-up acquisition, position of the orthogonal slice of the main pa at baseline measurements was used for positioning the orthogonal slice at follow-up of the main pa to minimize differ-ences in measurement localization from baseline.

9.2.3 RHC Protocolrhc was performed with a balloon, flow-directed 7f Swan-Ganz catheter. The patients were in stable condition, lying supine and breathing room air, while heart rate was continuously monitored. Measurements were made of mean right atrial pressure, right ventricular pres-sure, systolic pap, diastolic pap (dpap), mean pap (mpap), and pulmonary capillary wedge pressure (pcwp). co obtained by either thermodilution or direct Fick method from arte-rial and mixed venous oxygen saturation and oxygen consumption. Afterward, pvr was calculated as (mpap – pcwp) / co. The median interval from cmr to rhc was 2 days (range 2-21 days).

9.2.4 Image analysisIn the magnitude images, the wall of the pa is automatically detected. A semiautomatic wall detection program, based on a study published by Li et al.16 using matlab r2008a (The MathWorks, Natick, ma) was used to acquire more-accurate measurements of the pa, reduce operator bias, and minimize detection difficulties due to low signal intensity dur-ing diastole. Cross-sectional areas (csas) were obtained through the entire cardiac cycle (25-60 images) and visually checked by the operator. The minimal csa was considered the diastolic csa. From the csa of this contour, the average diameter was calculated and used in this study. For scaling to aorta size, we measured the aad in the same way as the pad at end diastole at the level of the pa bifurcation. All analyses were performed by one investi-gator who was unaware of rhc results. Ten patients were randomly selected for repeated measurements by both the first investigator and a second blinded investigator in order to test intra- and interobserver variability.

9.2.5 Statistical analysesHemodynamic values are presented as mean 6 sds or median (interquartile ranges). Dif-ferences between patient groups were evaluated with an independent-sample t-test or a Mann-Whitney U-test when not normally distributed. The ability of the rpad /aad to pre-dict pah was tested using a receiver operating characteristic curve. Differences between baseline and follow-up where evaluated with a paired-sample t-test. For analysis of tertile differences, a one-way analysis of variance with post hoc Bonferroni correction was used. Intra- and interobserver agreement was assessed by Bland-Altman analysis. All tests were two-tailed, and a p < 0.05 was considered significant. Analyses were performed with statis-tical software spss for Windows, version 15.0.0 (spss Inc, Chicago, il).


99.3 RESULTS

9.3.1 RHC Resultsrhc measurements confirmed the diagnosis of pah in the 51 study patients and normal pressures in the 18 normotensive patients. The demographic and hemodynamic characteristics obtained at rhc of both patients groups are shown in Table 9.2. As expected, mpap, systolic pap, dpap, mean right atrial pressure, pvr and heart rate at rhc were significantly higher in patients with pah than in the normotensive patients. In addition, co, cardiac index, stroke volume, and mixed venous oxygen saturation were significantly lower in patients with pah. Between the study groups, there were no significant differences in terms of sex distribution, body surface area, arterial oxygen satu-ration, or pcwp. The normotensive patients were significantly older than the patients with pah.

Table 9.2 General and hemodynamic characteristics




Age (y) 44.8 ± 15 53.7 ± 16.9 41.6 ± 13 0.004

Sex (m/f) 16/53 3/15 13/38 n.s.

Body surface area (m2) 1.86 ± 0.21 1.85 ± 0.16 1.90 ± 0.2 n.s.

Hemodynamics

Heart rate (bpm) 83 ± 15 73 ± 10 87 ± 14 0.03

sPAP (mmHg) 66.6 ± 30.3 27.1 ± 6.0 80.5 ± 21.9 < 0.001

dPAP (mmHg) 25.9 ± 13.2 8.1 ± 3.7 32.2 ± 8.9 < 0.001

mPAP (mmHg) 42.2 ± 19.1 15.9 ± 4.1 51.5 ± 12.5 < 0.001

PCWP (mmHg) 8.2 ± 4.7 7.3 ± 3.1 8.5 ± 5.1 n.s.

PVR (dynes · sec · cm-5) 660 ± 465 130 ± 77 837 ± 401 < 0.001

RAP (mmHg) 665 ± 450 130 ± 79 830 ± 395 < 0.001

dRVP (mmHg) 7.7 ± 5.2 3.4 ± 2.1 8.9 ± 5.1 < 0.001

Cardiac output (L/min) 5.3 ± 1.8 6.7 ± 1.7 4.8 ± 1.6 < 0.001

Cardiac index (L/min/m2) 2.9 ± 1.1 3.7 ± 1.0 2.6 ± 1.0 < 0.001

Stroke volume (ml) 66.7 ± 27.4 92.0 ± 22.1 55.8 ± 24.1 < 0.001

SaO2 (%) 95.6 ± 3.0 95.7 ± 4.3 95.6 ± 2.6 n.s.

SvO2 (%) 67.1 ± 9.1 72.9 ± 6.5 65.2 ± 9.0 0.002

PAD (mm) 31.9 ± 5.8 25.0 ± 6.8. 33.7 ± 5.3 < 0.001

AAD (mm) 28.3 ± 5.8 31.8 ± 6.1 27.0 ± 5.3 0.02

PAD / AAD 1.16 ± 0.27 0.87 ± 0.17 1.26 ± 0.22 < 0.001

Values expressed as means ± SD. *Comparison between patients without PH and patients with PAH. PH = pulmonary hypertension, n.s. = not significant, mPAP/sPAP/dPAP = mean/systolic/diastolic pulmonary artery pressure, PVR = pulmonary vascular resistance, RAP = right atrial pressure, PCWP = pulmonary capillary wedge pressure, PVR = pulmonary vascular resistance, SvO2 = mixed venous oxygen saturation, PAD = pulmonary artery diameter, AAD = ascending aorta diameter.

124 Chapter 9

9.3.2 PAD and pressureThe diastolic pad was significantly larger in patients with pah than in normotensive patients (Table 9.2). In the entire group at baseline, the correlation coefficient between pad and mpap was 0.58 (p < 0.001, Figure 9.2A). In all patients with pah, there was only a weak, but signifi-cant relation (ρ = 0.29; p = 0.04) between pad and mpap. A significant difference was found between the rpad /aad in the pah group and normotensive group. In the entire group, the correlation coefficient of rpad /aad with mpap was 0.71 (p < 0.001, Figure 9.2B). Again, for patients with pah, the relation between rpad /aad and dpap was weak (ρ = 0.49, p < .001). The correlation coefficient of rpad /aad with dpap was 0.69 (p < 0.001). The area under the receiver operating characteristic curve for the rpad /aad in the detection of pah was 0.93 (95%-ci: 0.86-0.99, Figure 9.3). The optimal rpad /aad was 1.1, with a sensitivity of 80% and a specificity of 94%. For an rpad /aad of 1.0 we found a sensitivity of 92%, a specificity of 72%, and a positive predictive value of 92%.

9.3.3 Changes of PAD during follow-upMean follow-up time was 942 days (range 242-2,359 days). During follow-up, the pad in-creased significantly (Table 9.3). In 37 (73%) patients with pah, the diameter increased (mean, 3 ± 3 mm; range, 0.2-17 mm). In 11 patients (22%), the diameter decreased (mean –2 ± 2 mm, range –0.2-5.4 mm). In three patients, the diameter remained unchanged. pvr decreased, and co increased, both significantly, under treatment during follow-up (Table 9.3). Figure

0 20 40 60 800.5

1.0

1.5

2.0

mPAP (mmHg)0 20 40 60 80

mPAP (mmHg)

PAD

/ A

AD

20

30

40

50PA

D (m

m)

A B

ρ = 0.58 ρ = 0.71

PHNon-PH

Figure 9.2 Graphs showing the results of regression analysis in the study group for pulmonary artery diameter (PAD) vs. mPAP (A) and the ratio of PAD / ascending aorta diameter (AAD) vs. mPAP (B).

0 20 40 60 80 1000

20

40

60

80

100mPAP > 25 mmHg

100 – speci�city (%)

Sens

itivi

ty (%

)

AUC = 0.93CI = [0.86-0.99]p < 0.001

Figure 9.3� The receiver-operating characteristic curve of the ability of the ratio rPAD / AAD to detect mean PA pressure > 25 mmHg. AUC = area under the curve; CI = confidence interval.


9

9.4 shows a plot of the change of pad against the change in mpap. Of the 11 patients with a decreased diameter, only three had a decrease > 2 mm. The mpap in these three patients was normalized or almost normalized under treatment; the mpap at follow-up ranged from 15 to 28 mm Hg. In patients with an increased pad at follow-up, the pressure either increased or decreased, and no relation was found between pad change and pressure change. Figure 9.5 shows that there were no significant differences between tertiles regarding co, mpap at baseline, and pad at baseline. In contrast, diameter change across tertiles of follow-up time differed significantly for the outermost tertiles, whereas the changes in co and mpap were not different over the tertiles. Overall, a moderate correlation between follow-up time and diameter change was found (ρ = 0.56, p < 0.001). Neither a relation between changes in pad and pulse pressure at baseline (ρ = –0.21, p = 0.14) nor a relation between changes in pad and pulse pressure changes during follow-up (ρ = 0.16, p = 0.27) was found.

9.3.4 Reproducibility of PAD measurementsAlthough the detection of the pad is semiautomatic and the program used reduces operator bias, intra- and interobserver variability was checked. Bland-Altman analysis revealed an intraobserver bias of 0.1 ± 0.1 mm and an interobserver variability bias of 0.1 ± 0.2 mm, showing good reproducibility of pad measurements.

Table 9.3� Hemodynamic characteristics and PAD at baseline and follow-up (n = 51)

Without PH PAH p-value*

PAD (mm) 35.7 ± 6.5 33.7 ± 5.3 0.001

mPAP (mmHg) 53.7 ± 16.9 41.6 ± 13 0.15

Cardiac output (l/min) 5.2. ± 1.2 4.8 ± 1.65 0.005

Heart rate (bpm) 83 ± 12.5 87 ± 14 0.13

Stroke volume (ml) 60.0 ± 23.3 55.8 ± 24.2 0.6

PVR (dynes · sec · cm-5) 730 ± 365 837 ± 401 0.026

Values expressed as mean ± SD. Mean follow-up time: 942 days (range 224-2,359 days). *Paired-samples t-test. PAD = pulmonary artery diameter, mPAP = mean pulmonary artery pressure, PVR = pulmonary vascular resistance.

–40 –20 20 40–10

–5

5

0

0

10

15

20

n = 51

Δ mPAP (mmHg)

ΔD

iam

eter

(mm

)

Figure 9.4� Change in pulmonary artery diameter versus the change in mean pulmonary artery pressure (ΔmPAP) indicating the change in pressure and diameter between RHC + CMR evaluation in 59 patients. Four patients (highlighted) showed a decrease in diameter of more than 2 mm; in all four patients the pressure was (almost) normalized during follow-up.

126 Chapter 9

9.4 DISCUSSION

In this study, we show that pad and rpad /aad are useful in discriminating patients with pah from patients without pah. This finding is in line with previous reports. However, dur-ing follow-up, most pas showed progressive dilatation, which is not related to the changes in pressure. Furthermore, we found that the change in pad was not related to changes in flow.

9.4.1 PAD and pressureOur results showed that the diastolic pad has diagnostic value in pah. We chose the dia-stolic diameter because it is the customary choice in ct images. The diameters used were averaged diameters derived from the csa. There was no difference in the relation of rpad /aad with dpap and mpap, which can be explained by the strong proportional relation be-tween dpap and mpap.17,18 The mpap was used in this study because diagnosis customarily is based on it. As seen in prior studies, we found that an enlarged pad is related to the pres-ence of pah and, thus, can be used in the detection of pah. We found that an rpad /aad of 1.1 yields the highest diagnostic accuracy. The use of the rpad /aad also is recommended because its relationship with pap is independent of body surface area and sex.5 This cut-off point of 1.1 is different from earlier studies that found that a ratio of 1 is the best diagnostic cut-off point. The explanation of this discrepancy might be that these studies used 20 mm Hg of mpap as the diagnostic criterion for pah instead of the currently used criterion of 25 mm Hg.1 However, if we apply the cut-off point of 1.0, a reasonable sensitivity of 92% and specificity of 72% were found. The diagnostic accuracy of the rpad /aad might be overes-

< 31 31 - 36 > 360

1

2

3

4

5

6

Baseline PA diameter (mm)

Chan

ge P

AD

(mm

)

< 46 46 - 58 580

1

2

3

4

5

6

Baseline mPAP (mmHg)

< 0 0 - 1.1 > 1.10

1

2

3

4

5

6

Change in cardiac output (l/min)

Chan

ge P

AD

(mm

)

420 - 1176< 420 > 11760

1

2

3

4

5

6

Follow-up time (days)

A B

C D

p = 0.34

p < 0.01

p < 0.05p = 0.23

Figure 9.5� Diameter change against the different tertiles of PA diameter at baseline (A), mPAP at baseline (C) changes in cardiac output and duration of follow-up (D).


9

timated because there is a relatively large hiatus in the levels of mpap between the normo-tensive patients and this well-defined group of patients with pah. In this study group, there were no patients with aortic abnormalities or systemic hypertension, which could affect the size of the aa and clearly could affect the reliability of the rpad /aad.

9.4.2 Changes of PAD during follow-upDuring follow-up, the pa dilated in most of the patients. The changes of the pad between the two cmr measurements did not reflect changes in mpap between the corresponding rhcs. In all the patients in whom the pressure was higher than initially, the pad had in-creased. However, if the pressure was lower than initially, the majority of this group’s pads still increased. Only in three patients did the pad decrease > 2 mm; in all these patients, the pressure was normalized or almost normalized.

Thus, a progressive dilatation was found in the patients with pah, which was not ex-plained by pressure changes. Even in the majority of the patients in which the pressure decreased, there was an ongoing dilatation of the pa. Figure 9.5A suggests that a pa with a large diameter at baseline tends toward greater dilatation during follow-up. It seems logi-cal that if dilatation during follow-up becomes independent of hemodynamics, a larger diameter at baseline will lead to larger dilatation during the follow-up. However, this rela-tion was not found to be statistically significant. The finding that neither pressure nor flow are related to the progressive dilatation in most of the patients shows that other expla-nations than changes in pulmonary hemodynamics underlie this progressive dilatation, a phenomenon well-known from aneurysms of the aorta. Although systemic hypertension is an important underlying cause of this disease, a further dilatation of the aorta in aneu-rysms is independent of systemic bp.19 The main question was to investigate whether during follow-up changes in pa dilatation exist and whether they are related to changes in pressure. For this reason, we did not evaluate specific other hemodynamic parameters as possible explanations for changes in pad. Nevertheless, we found that the diameter increase did not correlate with age, use of epoprostanol, and pulsatility in pad.

Absence of a direct relation between changes in pressure or flow and changes in diameter does not exclude that increased pap or reduced flow is the cause of pa dilatation. Although there is an absence of radiologic studies investigating the structure of the pa wall, recent histologic studies of the proximal parts of the PAs provide evidence that significant remod-eling of the proximal pa wall occurs in pah.20,21 Structural changes in elastin and collagen under the influence of an increased pap might eventually become a cause of pa dilatation, irrespective of changes in pressure and flow. In addition, altered flow in pah affects wall shear stress and, subsequently, matrix properties of the vessel wall22 that might lead to pa dilatation.

9.4.3 Clinical implicationsOur findings have several clinical implications. First, the pad, although useful in the diag-nosis of pah, is not useful for follow-up of the disease or evaluation of treatment effects. Second, we show that the dilatation of the pa in pah is more related to follow-up time than to pressure change, indicating that the pa needs time to dilate. A severe dilated pa at the time of diagnosis thus indicates that the pah was already present for a long period. Third,

128 Chapter 9

our data indicate that although increased pap leads to dilation of the pa, further dilation is a process most likely due to a change of the intrinsic vessel properties, which is independent of the pulmonary hemodynamics.

9.4.4 LimitationsThere are some study limitations that might have influenced our results. First, pressure and cmr measurements could not be performed simultaneously for logistic and patient reasons and might have affected the strength of the associations we found. However, given the aver-age time of follow-up (942 days), it is unlikely that a median interval of 2 days between rhc and cmr measurements affected the conclusions. Second, given the cone-shaped main pa, differences in level of image acquisition on the different time points might have created a bias. We tried to overcome this limitation by using an experienced investigator to acquire all cmr images and by using reference points acquired at the baseline measurement during the follow-up measurement.

9.5 CONCLUSION

In conclusion, in patients with pah, the rpad /aad can be used for the detection of pah. During follow-up, dilatation of the pa does not reflect changes in pressure or flow; there-fore, changes in pad cannot be used in clinical practice for the evaluation of the course of the disease, therapeutic response, or estimation of pressure. ■

References

1. McLaughlin VV, Archer SL, Badesch DB, Barst RJ, Farber HW, Lindner JR, et al. ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association developed in collaboration with the American College of Chest Physicians; American Thoracic Society, Inc.; and the Pulmonary Hypertension Association. J Am Coll Cardiol 2009;53(17):1573-619.

2. Teichmann V, Jezek V, Herles F. Relevance of width of right descending branch of pulmonary artery as a radiological sign of pulmonary hypertension. Thorax 1970 Jan;25(1):91-6.

3. Haimovici JB, Trotman-Dickenson B, Halpern EF, Dec GW, Ginns LC, Shepard JA, et al. Relationship between pulmonary artery diameter at computed tomography and pulmonary artery pressures at right-sided heart catheterization. Massachusetts General Hospital Lung Transplantation Program. Acad Radiol 1997;4(5):327-34.

4. Kuriyama K, Gamsu G, Stern RG, Cann CE, Herfkens RJ, Brundage BH. CT-determined pulmonary artery diameters in predicting pulmonary hypertension. Invest Radiol 1984;19(1):16-22.

5. Ng CS, Wells AU, Padley SP. A CT sign of chronic pulmonary arterial hypertension: theratio of main pulmonary artery to aortic diameter. J Thorac Imaging 1999;14(4):270-8.

6. Schmidt HC, Kauczor HU, Schild HH, Renner C, Kirchhoff E, Lang P, et al. Pulmonary hypertension in patients with chronic pulmonary thromboembolism: chest radiograph and CT evaluation before and after surgery. Eur Radiol 1996;6(6):817-25.

7. Heinrich M, Uder M, Tscholl D, Grgic A, Kramann B, Schafers HJ. CT scan findings in chronic thromboembolic pulmonary hypertension: predictors of hemodynamic improvement after pulmonary thromboendarterectomy. Chest 2005 May;127(5):1606-13.


9

8. Moore NR, Scott JP, Flower CD, Higenbottam TW. The relationship between pulmonary artery pressure and pulmonary artery diameter in pulmonary hypertension. Clin Radiol 1988;39(5):486-9.

9. Grubstein A, Benjaminov O, Dayan DB, Shitrit D, Cohen M, Kramer MR. Computed tomography angiography in pulmonary hypertension. Isr Med Assoc J 2008;10(2):117-20.

10. Frank H, Globits S, Glogar D, Neuhold A, Kneussl M, Mlczoch J. Detection and quantification of pulmonary artery hypertension with MR imaging: results in 23 patients. AJR Am J Roentgenol 1993;161(1):27-31.

11. Murray TI, Boxt LM, Katz J, Reagan K, Barst RJ. Estimation of pulmonary artery pressure in patients with primary pulmonary hypertension by quantitative analysis of magnetic resonance images. J Thorac Imaging 1994;9(3):198-204.

12. Sanz J, Kariisa M, Dellegrottaglie S, Prat-Gonzalez S, Garcia MJ, Fuster V, et al. Evaluation of pulmonary artery stiffness in pulmonary hypertension with cardiac magnetic resonance. JACC Cardiovasc Imaging 2009;2(3):286-95.

13. Sanz J, Kuschnir P, Rius T, Salguero R, Sulica R, Einstein AJ, et al. Pulmonary arterial hypertension: noninvasive detection with phase-contrast MR imaging. Radiology 2007;243(1):70-9.

14. Bouchard A, Higgins CB, Byrd BF, 3rd, Amparo EG, Osaki L, Axelrod R. Magnetic resonance imaging in pulmonary arterial hypertension. Am J Cardiol 1985 Dec 1;56(15):938-42.

15. Galie N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera JA, et al. Guidelines for the diagnosis and treatment of pulmonary hypertension: The Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J 2009 Oct;30(20):2493-537.

16. Li W, von Birgelen C, Di Mario C, Boersma E, Gussenhoven EJ, van der Putten N and Bom N. Semiautomatic contour detection for volumetric quantification of intracoronary ultrasound. Comp Cardiol Los Alamitos, California: IEEE Computers Society Press; 1994. p. 277-80.

17. Chemla D, Castelain V, Humbert M, Hebert JL, Simonneau G, Lecarpentier Y, et al. New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure. Chest 2004 Oct;126(4):1313-7.

18. Syyed R, Reeves JT, Welsh D, Raeside D, Johnson MK, Peacock AJ. The Relationship Between the Components of Pulmonary Artery Pressure Remains Constant Under All Conditions in Both Health and Disease. Chest 2008 March 2008;133(3):633-9.

19 Brady AR, Thompson SG, Fowkes FG, Greenhalgh RM, Powell JT. Abdominal aortic aneurysm expansion: risk factors and time intervals for surveillance. Circulation 2004 Jul 6;110(1):16-21.

20. Kobs RW, Chesler NC. The mechanobiology of pulmonary vascular remodeling in the congenital absence of eNOS. Biomech Model Mechanobiol 2006 Nov;5(4):217-25.

21. Lammers SR, Kao PH, Qi HJ, Hunter K, Lanning C, Albietz J, et al. Changes in the structure-function relationship of elastin and its impact on the proximal pulmonary arterial mechanics of hypertensive calves. Am J Physiol Heart Circ Physiol 2008 Oct;295(4):H1451-9.

22. Botney Mitchell D. Role of Hemodynamics in Pulmonary Vascular Remodeling. Implications for Primary Pulmonary Hypertension. Am J Respir Crit Care Med 1999 February 1, 1999;159(2):361-4.

C H A P T E R 10

131

Department of Pulmonary Diseases, Institute for Cardiovascular Research, VU University Medical Center, Amsterdam

G.J. Mauritz

Insights into the progression of right ventricular failurein pulmonary arterial hypertension:summary and future perspectives

10.1 SUMMARy

10.1.1 Introduction

PULMONAry ArTErIAL HyPErTENSION (pah) is characterized by excessive pul-monary vascular remodelling, resulting in elevated pulmonary artery and right heart pressures. Although pulmonary pressure rise is the distinctive characteristic of this

disease, the patients’ functional capacity and survival is predominantly determined by right ventricular (rv) function. Knowledge about the role of the rv in health and disease histori-cally has lagged behind that of the left ventricle. The rv has generally been considered a mere bystander, a victim of pathological processes.0 Consequently, in pah comparatively little at-tention has been devoted to how rv dysfunction is best detected and measured. Little at-tention has also been paid to what specific molecular and cellular mechanisms contribute to adaptation of right ventricular function or failure, as well as how rv dysfunction evolves structurally and functionally in the course of the disease, or what interventions might best preserve rv function. The studies described in this thesis aimed to improve our understand-ing of the mechanical principles underlying the progression of rv failure in pah

10.1.2 Right ventricular geometric shortening in PAHIn Chapter 2, we evaluated regional transverse shortening of the rv myocardium, defined as movements of the rv free-wall to the septum, in a group of pah patients and control subjects. As yet, tricuspid annular plane systolic excursion (tapse) has been well studied in pulmonary hypertension (pah) as a surrogate for rv function. This measure quantifies the longitudinal shortening of the rv and its clinical and prognostic value has been well estab-lished by echocardiography. However, tapse ignores the septal contribution to rv ejection. We hypothesized that in pah, transverse movements of the rv free wall towards the septum are important in rv ejection. From 101 pah patients, mri four-chamber cine images were analyzed to quantify rv transverse shortening at seven levels along an apex-to-base axis.

132 Chapter 10

For each level, regional absolute and fractional transverse shortening were computed and related to rvef. regional transverse wall movements provide important information of rv function in pah. Our results showed a close relation between transverse motion at mid-rv reveals and global right ventricular function as measured by rvef.

Encouraged by the findings that rvef is better reflected by rv transverse shortening than longitudinal shortening in pah, we performed the study of Chapter 3 to assess which of these geometric shortening parameters provides the best reflection of rv functional decline over time in a cohort of pah patients with progressive right ventricular failure. We reviewed 42 consecutive pah patients who underwent right heart catheterization and cardiac mri at baseline and after 1-year follow-up. Based on the survival after this one-year run-in pe-riod, patients were classified into two groups: survivors (26 patients): subsequent survival of more than 4 years, and non-survivors (16 patients): subsequent survival of less than 4 years. Four-chamber cine imaging was used to quantify rv longitudinal shortening, rv transverse shortening, and rv fractional area change between end-diastole and end-systole. Our study revealed that a progressive decline in rv transverse shortening and rv fractional area change is associated with mortality in these pah patients. Interestingly, this decline in right ventricular function was predominantly the consequence of increased leftward septal bowing.

10.1.3 Interventricular asynchrony and RV wall stress in PAHrecent research demonstrated the presence of interventricular mechanical asynchrony in severe pah patients. This asynchrony is caused by prolonged duration of shortening and de-layed peak shortening of the rv free wall compared with that in the left ventricular free wall and interventricular septum. This rv free wall shortening even continued after pulmonary valve closure and appeared to be related to leftward septal bowing during lv early diastole. In Chapter 4 we investigated whether in pah the prolonged period between pulmonary valve closure and tricuspid valve opening is caused by an additional post-systolic contrac-tion period, or an increased relaxation period (representing rv diastolic dysfunction). Twenty-three pah patients underwent cardiac mri. From rv two-chamber imaging we de-termined the times of pulmonary valve closure and tricuspid valve opening. In addition, the time to peak rv shortening (contraction) was determined with myocardial tagging. Prolongation of the period between pulmonary valve closure and tricuspid valve opening in pah was strongly related to the increased post-systolic contraction period rather than increased relaxation period. Therefore, this prolonged isovolumic period in pah cannot be considered as a mere reflection of rv diastolic dysfunction, as had been the assumption until now, but is a parameter of prolonged contraction duration.

The underlying mechanism of this prolonged contraction duration of the rv free wall is unknown in pah. In Chapter 5 we hypothesized that rv end-systolic wall stress plays a key role in this interventricular mechanical asynchrony in pulmonary arterial hyperten-sion. Thirteen consecutive patients with chronic tromboembolic pulmonary hypertension underwent magnetic resonance imaging myocardial tagging before and after pulmonary endarterectomy. For the left ventricular free wall, septum and rv free wall, the time to peak of circumferential shortening (Tpeak) was calculated. Pulmonary artery pressure was mea-sured by right heart catheterization. Then, for the rv free wall, the rv systolic wall stress

133Summary and future perspectives

10

was calculated by the Laplace law. Post-operatively, the left to right free wall delay in Tpeak decreased to normal reference values. The rv wall stress decreased to values which were not different from the normal reference values. Moreover, the reduction of l-r delay in Tpeak was associated with the reduction in rv wall stress, but not with the reduction in systolic pap, reduction in rv radius or increase in rv wall thickness. These observations support the role of increased rv wall stress as a cause for the interventricular asynchrony in pah.

Since elevated myocardial wall stress induces an increased release of n-terminal pro-brain natriuretic peptide (nt-probnp), we subsequently investigated the prognostic value of serial nt-probnp measurements a large group of pah patients in Chapter 6. We retro-spectively analyzed all available nt-probnp plasma samples in 198 patients who were diag-nosed with who group ipah from January 2002 to January 2009. At the time of diagnosis, median nt-probnp levels were significantly different between survivors (610 pg/ml; range, 6 to 8714) and nonsurvivors (2609 pg/ml; range, 28 to 9828) (p < 0.001). In addition nt-probnp was significantly associated (p < 0.001) with other parameters of disease severity (6 minutes walking distance; functional class). receiver operating curve analysis identi-fied ≥ 1256 pg/ml as the optimal nt-probnp cut-off for predicting mortality at the time of diagnosis. Serial measurements allowed calculation of baseline nt-probnp (i.e., the inter-cept obtained by back-extrapolation of the concentration-time graph) providing a better discrimination between survivors and non-survivors than nt-probnp at time of diagnosis alone (p = 0.010). Furthermore, a decrease of nt-probnp of more than 15 percent per year is associated with long term survival.

10.1.4 Main pulmonary artery flow and sizeThe rv stroke volume directly reflects right ventricular (rv) function in response to its load, and contains prognostic information in pah as described in earlier studies. Since most of the mri protocols used in pah measure pulmonary artery flow, sv can be assessed by measuring flow in the main pulmonary artery (pa). Previous studies have shown that this method is accurate to measure sv from pa flow in healthy subjects. In Chapter 7 we investi-gated the accuracy of pulmonary artery (pa) flow for measuring sv in pah. Thirty-four pah patients underwent both cmr and right-sided heart catheterisation. cmr-derived stroke volume was measured by pa flow, left (lv) and right ventricular (rv) volumes, and, in a subset of nine patients also by aortic flow. These stroke volume values were compared to the stroke volume obtained by the invasive direct Fick method. The sv value, as derived from pa flow, appeared to have limited accuracy in pah patients. lv volumes and aorta flow are to be preferred for the measurement of sv.

As we observed that in pah patients the flow profile in the main pulmonary artery differs from that seen in healthy subjects, we subsequently studied the retrograde flow in the pul-monary artery in pah patients in Chapter 8. We hypothesised that early onset of retrograde flow in the main pulmonary artery is indicative for the presence of pulmonary hyperten-sion. Thirty-eight pah patients and 17 non-ph subjects underwent mri and right heart catheterisation. The onset time of the retrograde flow, as a fraction of the cardiac cycle dura-tion, was compared with the mean pulmonary artery pressure. A negative relation between onset time of retrograde flow and pulmonary artery pressure was found. Furthermore, the relative onset time of retrograde flow in pah patients was significantly smaller than those

134 Chapter 10

found in non-ph subjects. In addition, a cut off value for relative onset time (fraction of r-r interval) of 0.25 distinguished pah patients from all non-ph subjects hypertension (pah) and is used for non-invasive detection as described earlier. However, it is unclear how the size of the pa behaves over time and whether it is related to pressure changes. In Chapter 9 we evaluated the pa size during follow-up in treated patients with pah and tested whether it reflects pulmonary vascular hemodynamics. Fifty-one patients with pah who underwent at least two right-sided heart catheterizations together with cardiac mri were included in this study. Another 18 patients who had normal pressures at rhc were included for com-parison. From rhc, we derived pa pressures and cardiac output. From the mri images we derived pa diameter and the ratio of the pad and ascending aorta diameter. The pa diameter was significantly larger in patients with pah than in patients without pah. A ratio of the pa diameter and ascending aorta diameter more than 1 had a positive predictive value of 92% for pah. Mean follow-up time was 942 days, and there was a significant dilatation during this period. However, the change of the pad did not correlate with the changes in pressure or cardiac output. A moderate correlation between pad and follow-up time was found.

10.2 FUTURE PERSPECTIVES

10.2.1 Right ventricular geometric shortening in PAH: the importance of the septumThe rv longitudinal shortening, when taken as a measure of rv function, ignores the septal contribution to rv ejection (Chapter 2) which may become important to maintain overall rv function. A further reduction of rv function is due to progressive leftward septal dis-placement (Chapter 3). Because transverse shortening incorporates both free wall and sep- (Chapter 3). Because transverse shortening incorporates both free wall and sep-. Because transverse shortening incorporates both free wall and sep-tum displacement, this parameter can be used to monitor the decline of rv function. The clinical value of the transverse measures would be even stronger if these could be measured also by using echocardiography. Thus further development of right ventricular echocar-diography, already a wide-spread tool in the non-invasive work-up and follow-up of pa-tients with left ventricular dysfunction, is advised.1

10.2.2 Diastolic function in pulmonary arterial hypertension: how to measure?Diastolic function of the rv in pah is a concept that has not been well studied. Th ere is modest evidence of impaired diastolic rv function in pah. Further studies are needed to investigate the presence of relaxation impairment and the relative contribution to the pro-gression of rv dysfunction in pah patients. The gold standard for the assessment of dia-stolic dysfunction is to measure the load independent diastolic elastance as obtained by invasive pressure-volume relations.2

However, this measurement is rather invasive and not suitable for serial follow-up mea-surements. Although tissue Doppler imaging has been well validated as a modality to mea-sure left ventricular diastolic function, few studies have used it to examine rv diastolic function. The prolonged post-systolic isovolumic period, as previously interpreted as a re-flection of diastolic function, is not a sign of diastolic dysfunction (Chapter 4). This raises the question which measure should be used to accurately assess rv diastolic function non-


10

invasively. One option is diastolic myocardial strain rate derived from mri myocardial tag-ging.3 Alternatively, 3d velocity-encoded mr imaging4 might aid in the assessment of right ventricular diastolic function and the clinical implications in patients with pah.

10.2.3 Interventricular asynchrony: target for treatment?Interventricular asynchrony is often observed in progressive stages of pah-induced right ventricular failure.5-8 Th e underlying cause of this interventricular asynchrony is an in- e underlying cause of this interventricular asynchrony is an in-crease in pulmonary resistance and subsequently the increase in rv wall stress (Chapter 5). So, the principal target of therapy is the pulmonary microvasculature. However, as the rv adaptation capacity plays a central role, the rv itself may well be a target for treatment. This suggests that additional therapeutic methods, that improve the efficiency of rv contraction, may be beneficial in these patients. One option is resynchronization with biventricular pac-ing, such that the asynchrony in lv and rv peak shortening is reduced.9,10 An alternative treatment option would be to lower rv wall stress pharmacologically that will limit the variability and load dependence of relaxation

10.2.4 N-terminal pro B-type natriuretic peptide in PAH: serial measurements Measurement of the b-type natriuretic peptides is currently recommended by guidelines, despite a lack of appropriate validation in the pah population.11 With regard to nt-probnp, a value above 1400 pg⁄ml was associated with poorer survival in 55 patients with several types of pulmonary hypertension, all receiving pah targeted therapies.12 The reveal reg-istry has identified a similar cut-off (1500 pg ⁄ml) associated with worse outcomes.13 We observed in 198 pah patients a comparable cut-off value of 1256 pg/ml for predicting mor-tality at the time of diagnosis. Moreover, our result shows that decreasing nt-probnp levels during follow-up by more than 15% per year (Chapter 6) is an important treatment goal in pulmonary arterial hypertension. Prospective, goal oriented strategies are warranted to test whether this treatment goal is achievable in pah patients and whether this will alter the prognosis. Investigators undertaking cohort studies or therapeutic trials in pah should be encouraged to incorporate serial nt-probnp measurements in study designs.

10.2.5 Pulmonary artery dilatation: chance of PA dissection or rupture? Pulmonary artery dilatation is a characteristic in pah and further dilatation seems to be a process most likely due to a change of the intrinsic vessel properties independent of pul-monary hemodynamics (Chapter 9). Dissection and rupture of a dilated pulmonary artery is a rare, but deathly, event which is most difficult to predict in pah.14 These complications are possibly related to the size of the pulmonary arteries. Better identification of risk fac-tors for pulmonary artery rupture or dissection may help to select the high-risk patients as candidates for heart-lung transplantation, since the outcome is very poor of patients who develop these complications. ■

136 Chapter 10

References

0. Voelkel NF, Quaife RA, Leinwand LA et al. Right ventricular function and failure: report of a National Heart, Lung, and Blood Institute working group on cellular and molecular mechanisms of right heart failure. Circulation. 2006 Oct 24; 114(17):1883-91.

1. Badesch DB, Champion HC, Sanchez MA, Hoeper MM, Loyd JE, Manes A, McGoon M, Naeije R, Olschewski H, Oudiz RJ, Torbicki A. Diagnosis and assessment of pulmonary arterial hypertension. J Am Coll Cardiol. 2009;54(1 Suppl):S55-66.

2. Klotz S, Hay I, Dickstein ML, Yi GH, Wang J, Maurer MS, Kass DA, Burkhoff D. Single-beat estimation of end-diastolic pressure-volume relationship: a novel method with potential for noninvasive application. Am J Physiol Heart Circ Physiol. 2006;291(1):H403-412.

3. Germans T, Russel IK, Gotte MJ, Spreeuwenberg MD, Doevendans PA, Pinto YM, van der Geest RJ, van der Velden J, Wilde AA, van Rossum AC. How do hypertrophic cardiomyopathy mutations affect myocardial function in carriers with normal wall thickness? Assessment with cardiovascular magnetic resonance. J Cardiovasc Magn Reson.2010;12:13.

4. van der Hulst AE, Westenberg JJ, Kroft LJ, Bax JJ, Blom NA, de Roos A, Roest AA. Tetralogy of fallot: 3D velocity-encoded MR imaging for evaluation of right ventricular valve flow and diastolic function in patients after correction. Radiology. 2010; 256(3):724-734.

5. Marcus JT, Gan CT, Zwanenburg JJ, Boonstra A, Allaart CP, Gotte MJ, Vonk-Noordegraaf A. Interventricular mechanical asynchrony in pulmonary arterial hypertension: left-to-right delay in peak shortening is related to right ventricular overload and left ventricular underfilling. J Am Coll Cardiol. 2008;51(7):750-757.

6. Roeleveld RJ, Marcus JT, Faes TJ, Gan TJ, Boonstra A, Postmus PE, Vonk-Noordegraaf A. Interventricular septal configuration at mr imaging and pulmonary arterial pressure in pulmonary hypertension. Radiology. 2005;234(3):710-717.

7. Tanaka H, Tei C, Nakao S, Tahara M, Sakurai S, Kashima T, Kanehisa T. Diastolic bulging of the interventricular septum toward the left ventricle. An echocardiographic manifestation of negative interventricular pressure gradient between left and right ventricles during diastole. Circulation. 1980;62(3):558-563.

8. Vonk-Noordegraaf A, Marcus JT, Gan CT, Boonstra A, Postmus PE. Interventricular mechanical asynchrony due to right ventricular pressure overload in pulmonary hypertension plays an important role in impaired left ventricular filling. Chest. 2005;128(6 Suppl):628S-630S.

9. Handoko ML, de Man FS, Allaart CP, Paulus WJ, Westerhof N, Vonk-Noordegraaf A. Perspectives on novel therapeutic strategies for right heart failure in pulmonary arterial hypertension: lessons from the left heart. Eur Respir Rev.19(115):72-82.

10. Handoko ML, Lamberts RR, Redout EM, de Man FS, Boer C, Simonides WS, Paulus WJ, Westerhof N, Allaart CP, Vonk-Noordegraaf A. Right ventricular pacing improves right heart function in experimental pulmonary arterial hypertension: a study in the isolated heart. Am J Physiol Heart Circ Physiol. 2009;297(5):H1752-1759.

11. Galie N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera JA, Beghetti M, Corris P, Gaine S, Gibbs JS, Gomez-Sanchez MA, Jondeau G, Klepetko W, Opitz C, Peacock A, Rubin L, Zellweger M, Simonneau G. Guidelines for the diagnosis and treatment of pulmonary hypertension: the Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J. 2009;30(20):2493-2537.

12. Fijalkowska A, Kurzyna M, Torbicki A, Szewczyk G, Florczyk M, Pruszczyk P, Szturmowicz M. Serum N-terminal brain natriuretic peptide as a prognostic parameter in patients with pulmonary hypertension. Chest. 2006;129(5):1313-1321.

13. Benza RL, Miller DP, Gomberg-Maitland M, Frantz RP, Foreman AJ, Coffey CS, Frost A, Barst RJ, Badesch DB, Elliott CG, Liou TG, McGoon MD. Predicting survival in pulmonary arterial hypertension: insights from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL). Circulation. 2010;122(2):164-172.

14. Degano B, Prevot G, Tetu L, Sitbon O, Simonneau G, Humbert M. Fatal dissection of the pulmonary artery in pulmonary arterial hypertension. Eur Respir Rev. 2009;18(113):181-185.


10

139

Summary in Dutch

Afdeling longziekten, Institute for Cardiovascular Research, VU medisch centrum, Amsterdam

G.J. Mauritz

Nederlandse samenvatting

11.1 INTRODUCTIE

PULMONALE ArTErIëLE HyPErTENSIE (pah) is een zeldzame, progressieve aandoening gekenmerkt door vernauwing van de bloedvaten in de longen. Als ge-volg hiervan stijgt de bloeddruk in de longslagader, waardoor de rechterhartkamer

meer arbeid moet verrichten om het bloed door de vernauwde longvaten te pompen. De wandspanning in de rechterhartkamer neemt toe. De rechterhartkamer probeert zich hier-aan aan te passen door meer spiermassa te ontwikkelen en door uit te zetten. Als het de rechter hartkamer niet meer goed lukt om het bloed, tegen een verhoogde bloeddruk in, de longslagader in te pompen, spreekt men van rechter hartfalen. Dit uit zich bij patiënten vaak in lichamelijke beperkingen als kortademigheid en vermoeidheid.

Tot op heden is pah een ongeneeslijke ziekte. Ongeveer de helft van alle patiënten over-lijdt aan de gevolgen van deze ziekte binnen 5 jaar na diagnose, ondanks maximale therapie. Het falen van de rechterkamer is in de meeste gevallen van pah de doodsoorzaak. Tijdens het beloop van de ziekte is er een overgangsfase van een aangepaste rechterkamer, die in staat is zijn functie te behouden bij de hoge drukoverbelasting, naar een falende rechter-kamer. De huidige kennis van het proces van rechter hartfalen en de rol van de pulmonaal vaten is beperkt. Het onderzoek zoals beschreven in dit proefschrift, heeft als doel nieuw inzicht te verschaffen in de mechanische principes die de progressie van rechter hartfalen in pah karakteriseren. Deze nieuwe inzichten zijn van essentieel belang voor verbetering van zowel diagnostiek, monitoren en therapie van de progressie van rechter hartfalen bij patiënten met pah. Magnetische resonantie imaging (mri) is op dit moment een van de beste klinische beeldmodaliteiten gezien de hoge mate van precisie en reproduceerbaar-heid voor de evaluatie van de rechter hartkamer in pah. We hebben mri gebruikt voor het analyseren van de geometrie, structuur en verkorting van de rechter hartkamer, en voor de analyse van de bloedstroming in de longslagader. Hoofdstuk 1 geeft een algemene inleiding tot dit proefschrift en definieert het kader van het werk.

140 Summary in Dutch

11.2 GEOMETRISCHE VERkORTING EN PROGRESSIEF RECHTER HARTFALEN IN PAH

De verkorting van de lange as (de as van annulus tot apex, beter bekend als tapse) is een bekende afgeleide maat voor het functioneren van de rechter hartkamer bij patiënten met pah. Echter er is vrijwel niets bekend over de verkorting van de korte as (tussen de wand van de rechter hartkamer en het septum) bij pah-patiënten, terwijl die mogelijk juist be-langrijk wordt juist omdat ook het septum mee beweegt. In Hoofdstuk 2 bestuderen we de geometrische verkorting tijdens de hartcyclus van de rechter hartkamer bij pah-patiënten. Om dit te kunnen onderzoeken hebben we in 101 pah-patiënten, de lange-asverkorting, de korte-asverkorting en de oppervlakteverandering van de rechterhartkamer geanalyseerd uit mri-vierkameropnames en vergeleken met de gouden standaard voor globale rechter hartkamer functie (ejectiefractie). We konden aantonen dat bij pah-patiënten de korte-asverkorting een duidelijk betere maat is voor de functie van rechter hartkamer dan de lange-as verkorting.

Gebaseerd op de bevindingen van Hoofdstuk 2, bestudeerden we in Hoofdstuk 3 de veranderingen over 1 jaar van de korte- en lange as in twee groepen patiënten: pah met stabiele rechter hartfunctie en pah met progressieve rechter hartfalen. In de groep met progressief rechter hartfalen bleek dat op het moment van diagnose zowel de lange-as- als korte-asverkorting waren afgenomen tot een bepaald niveau. Tijdens follow-up werd ver-dere afname gezien van de korte-asverkorting, die veroorzaakt werd door de toename van buiging van het septum naar links. Klinisch zijn bij pah-patiënten het gebrek aan verbete-ring van zowel korte als lange-asbeweging tijdens follow-up en de toename van buiging van het septum naar links, signalen van eindstadium rechter hartfalen.

11.3 HARTINEFFICIëNTIE bIj PAH EN wANDSPANNING

Het doorbuigen van het septum naar links, dat karakteristiek is voor pah, wordt veroor-zaakt door een drukgradiënt over het septum waarbij de druk in de rechter hartkamer tijdelijk even groter is dan in de linker hartkamer. Uit eerder onderzoek is gebleken dat dit fenomeen wordt veroorzaakt doordat bij pah de rechter hartkamer langer doorgaat met samentrekken ten opzichte van de linkerhartkamer. In Hoofdstuk 4 konden we aantonen dat de rechterhartkamer bij pah inefficiënt gaat pompen in het laatste gedeelte van de sys-tolische fase van de hartcyclus. Op het moment dat de longslagaderklep al gesloten is, gaat de rechterkamer door met samentrekken, zonder volume uit het hart te pompen, waarbij alle energie wordt verbruikt voor het doorbuigen van het septum naar links. In Hoofdstuk 5 onderzochten we het effect van acuut ontladen van de rechterkamer drukoverbelasting op de asynchronie tussen de linker en rechter hartkamer bij pah. Om dit te kunnen onder-zoeken hebben we gekeken naar patiënten met chronische trombo-embolische pulmonale hypertensie (cteph). Deze selecte patiëntengroep kan effectief behandeld worden met pul-monale trombo-endarteriëctomie (pte). Tijdens deze chirurgische ingreep worden trombi en vernauwende lesies in het pulmonale vaatbed waar mogelijk verwijderd. pte resulteert in de meeste gevallen in een acute afname van de pulmonale vaatweerstand (en dus ook van de rechterkamerdruk). Na de daling van de rechterkamerdruk als gevolg van pte, zagen we

141Nederlandse samenvatting

NL

dat de asynchronie tussen beide hartkamers vrijwel volledig herstelde naar normaal waar-den. Ook het septum verkreeg weer zijn normale vorm, met buiging naar rechts. Tevens konden we aantonen dat de afgenomen wandspanning een centrale rol speelt in het herstel van asynchronie bij deze patiënten.

De toegenomen wandspanning van de rechterkamer bij pah-patiënten leidt tot stijging van b-type natriuretisch peptide (bnp) en n-terminal pro-b-type natriuretisch peptide (nt-probnp) in het bloed. Het bnp en nt-probnp zijn eiwitten die geproduceerd worden in de wand van de boezems en de kamers onder invloed van verhoogde wandspanning. Bij rech-ter hartfalen stijgen de bloedspiegels bij de pah-patiënten. Beide eiwitten kunnen worden gebruikt als prognostische markers in patiënten met pah. Echter de prognostische meer-waarde van herhaalde metingen van dit eiwit in het bloed tijdens verschillende tijdstippen in het follow-up traject is onbekend. In Hoofdstuk 6 bestuderen we de seriële nt-probnp metingen die zijn verricht in 198 pah-patiënten. We konden aantonen dat meerdere nt-probnp-bepalingen tijdens het ziekte proces een betere prognostische voorspelling geven dan een enkele bepaling. Verder zagen we dat een daling van minstens 15% tijdens follow-up geassocieerd is met een betere overleving.

11.4 PULMONAALARTERIE FLOw EN GROOTTE

Afname van het rechterkamer slagvolume is een maat voor rechter hartfalen bij pah, zoals in eerdere studies reeds is aangetoond. Omdat de meeste mri-protocollen die worden gebruikt bij pah ook de longslagader bloeddoorstroming (flow) meten, kan het slagvolume uit deze long-slagader flow worden gehaald. Eerdere studie studies hebben aangetoond dat dit accuraat is bij gezonde individuen. Echter op dit moment is onbekend welke methode het meest accuraat is voor het non-invasief meten van slagvolume bij pah-patiënten. In Hoofdstuk 7 meten we met mri op vier verschillende manieren het slagvolume en vergelijken dit met invasieve metingen verkregen middels rechter hartkatheterisatie. Deze studie laat zien dat bij pah het slagvolume het meest accuraat bepaald kan worden uit de aortaflow en de volumes van de linker hartkamer. We zagen dat de longslagaderflow bij pah-patiënten een minder accurate maat levert voor het slagvolume. Dit kan veroorzaakt worden door mogelijk turbulente flow in de longslagader die juist aanwezig is bij pulmonale hypertensie.

Verder zagen we dat bij pah-patiënten het flowpatroon in de longslagader verschilt van ge-zonde individuen. Dit was de aanleiding om in Hoofdstuk 8 de flowpatronen in de longslagader te bestuderen. Bij pah-patiënten werd al vroeg in de hartcyclus terugstroming gezien in een deel van de longslagader. We konden aantonen dat deze vroege terugstroming kenmerkend is voor een verhoogde longslagaderdruk. Tevens is een vergrote longslagader kenmerkend voor pulmo-nale hypertensie. Echter, hoe deze verandert over de tijd en wat de relatie is met de toegenomen druk is vrijwel onbekend. In Hoofdstuk 9 hebben we ten slotte gekeken naar de longslagader-diameter en de relatie met druk in 51 pah-patiënten. We konden aantonen dat een vergrote longslagader kenmerkend is voor patiënten met pah. Daarnaast zagen we dat tijdens follow-up er geen relatie was tussen de verandering van longslagader diameter en de verandering van de druk. Daarom kunnen de veranderingen in longslagader diameter niet gebruikt worden voor het evalueren van het verloop van de ziekte, de therapie respons of het schatten van de druk.

142 Summary in Dutch

11.5 CONCLUSIES

Samenvattend hebben we in dit proefschrift inzicht proberen te krijgen in de progressie van rechter hartfalen bij pah-patiënten. In de studies die verricht zijn hebben we dan ook aangetoond dat progressief rechter hartfalen wordt gekenmerkt door:

■ Een afgenomen rechterkamer lange-asverkorting en vrije wandverplaatsing tot een be-paald niveau en verdere afname van de korte-asverkorting tijdens follow-up door pro-gressief linkwaarts buigen van het septum. ■ Inefficiënte pompfunctie aan het einde van de systolische fase van de hartcyclus door mechanische asynchronie als gevolg van verlengde contractie van de vrije wand van de rechterkamer van die van de linkerkamer en het septum. ■ Verhoogde wandspanning van de rechterkamer die een centrale rol speelt bij de mecha-nische asynchronie tussen beide hartkamers. ■ Hoge nt-probnp waarden in het bloed op baseline en tijdens follow-up. ■

143Nederlandse samenvatting

NL

145

Dankwoord

Acknowledgements

EINDELIJK IS HET DAN ZOVEr. De laatste en waarschijnlijk ook de meest gelezen woorden van dit proefschrift. Nog meer wellicht dan in de wetenschappelijke tekst dient hier gewikt en gewogen te worden om de juiste woorden te vinden en niemand

te vergeten. Anderzijds moeten dankwoorden bij voorkeur uit je hart komen. Bij het tot stand komen van dit proefschrift zijn veel mensen betrokken geweest. Een

promotie kun je niet alleen en het is daarom van niet te onderschatten belang waardering uit te spreken richting de mensen die hun bijdrage hebben geleverd aan het tot stand ko-men van dit proefschrift. Hierbij wil ik graag de volgende personen noemen:

Afdelingshoofd van de afdeling longziekten van het vu medisch centrum, prof.dr. P.E. Postmus. Professor Postmus, u was het, samen met dr. A. Vonk Noordegraaf die mij heeft aangenomen om onderzoek te mogen doen op de afdeling longziekten. Ik wil u hartelijk bedanken voor de mogelijkheden die u mij heeft gegeven.

Mijn promotor prof.dr. A. Vonk Noordegraaf. Beste Anton, je bent een bijzonder man. Jouw passie voor de pulmonale hypertensie heeft mij aangestoken. Het overdraagbaar enthousi-asme voor het vak en jouw kritische kijk op de zaken (‘Wat zegt mij dat nou?’) hebben mij geïnspireerd vanaf het eerste moment dat ik bij je in de kamer zat. Ondanks de enorme vrijheid die je mij gaf, heb je me goed gestuurd en op de juiste weg gezet, me afgeremd als ik te hard van stapel liep en me aangemoedigd als ik dacht dat er geen oplossing voor de obstakels op het pad te vinden was. Anton, jouw deur stond altijd open, niet zelden kwam ik voor een korte vraag die soms twee uur tijd kostte vanwege het uitwisselen van kennis en ervaringen ten aanzien van hobby, geschiedenis, cultuur en religie. Voor de kans die je me hebt geboden om te promoveren en eens echte wetenschap te bedrijven ben ik je dan ook zeer dankbaar. Ik hoop dat wij in de toekomst nog samen met elkaar zullen werken.

146 Acknowledgements

Mijn copromotor dr. J.T. Marcus. Beste Tim, als onze mri-professional ben je cruciaal ge-weest in elk hoofdstuk van dit proefschrift. Maar niet alleen voor dit proefschrift, nee het paradepaardje, mri bij pulmonale hypertensie, was niet denkbaar geweest zonder jouw zorgvuldige en cruciale inbreng. Nog nooit heb ik je nee horen zeggen als het ging om plannen van patiënten of proefpersonen voor mri, al was het ook in de late avond (23.30 uur). Tim, ik heb ontzettend veel aan het overleg gehad waarmee ik te pas en te onpas mee aan mocht komen. Je maakte altijd tijd voor me. Tevens, heb ik al die jaren bewonderd hoe positief en onzettende ‘relaxed’ jij met wetenschappelijke tegenvallers omgaat. Een betere copromotor had ik mij niet kunnen wensen.

Emeritus hoogleraar fysiologie prof.dr. N. Westerhof. Beste Nico, nog altijd bent u volop bezig in de wetenschap. Het enthousiasme dat u uitstraalde toen ik u leerde kennen tijdens het Honours Program in het kader van de geneeskunde doctoraalopleiding, zijn me tot de dag van vandaag bij gebleven. De kamer waar de ‘pulmonale hypertensie’-wetenschap werd bedreven was niet compleet zonder uw aanwezigheid, kennis, wetenschappelijke inbreng en heldere visie. De tijd nemen, de openheid en de bijzonder aangename toegankelijkheid zijn karakteristiek voor onze emeritus hoogleraar fysiologie. Nico, ook u wil ik hartelijk danken voor de hulp en tijd die u hebt willen besteden aan de essentiële delen van mijn proefschrift.

De leden van de promotiecommissie wil ik bedanken voor deelname in de promotiecom-missie en het kritisch doorlezen en beoordelen van mijn proefschrift: prof.dr. R.M.F Berger, prof.dr. W.J. Paulus, prof.dr. A. de Roos en prof.dr. A.C. van Rossum. In addition, prof.dr. A. Torbicki, professor of Medicine and Head of the Department of Chest Medicine at the Insti-tute of Tuberculosis and Lung Diseases, Warsaw, Poland, a reference centre for pulmonary hypertension and embolism. Professor Torbicki, I would like to thank you for your willing-ness to participate in my PhD-committee and for the time you have spent on judging the thesis. I hope this thesis is ‘a further step in the right direction’.

De grondlegger van de ph-zorg in het vu medisch centrum, dr. Anco Boonstra. Beste Anco, je bent een geweldige arts met een hart voor je patiënt. Ik wil je bedanken voor al het en-thousiasme waarmee je mijn promotieonderzoek heb gevolgd. Zeker ook wil ik dr. Harm Jan Boogaard noemen. Beste Harm Jan jij ging weg toen ik kwam en jij kwam terug toen ik ging. Al was de periode kort toch heb je een essentiële bijdrage geleverd aan Hoofdstuk 3 van dit proefschrift. Hartelijk dank ook voor jouw snelle en correcte manier van becom-mentariëren. De overige stafleden van de afdeling longziekten tijdens mijn promotieonder-zoek wil ik ook hartelijk bedanken voor hun bijdrage en wetenschappelijke interesse, van wie ik met name prof.dr. Egbert Smit en dr. Thomas Sutedja noem.

Zeker ook wil ik de ph-verpleegkundigen, Iris van der Mark-Gobielje en Martha Wagenaar bedanken. Jullie onmisbare bijdrage aan de zorg en logistieke planning van onze patiënten is essentieel voor alle ph-promovendi.

Daarnaast wil ik ook physician-assistant, Frank Oosterveer bedanken. Beste Frank, als

147Dankwoord

Ack

al onze gespreksuren zouden moeten worden uitbetaald, dan hadden we een aardige 13e maand. Vaak heb ik gezien je sociale kwaliteiten gedacht: deze jongen moet huisarts worden. Frank bedankt voor je enthousiasme en uitleg tijdens de catheterisaties.

Hierbij wil ik ook de cardiologen en medewerkers van de catheterisatie kamer bedanken voor de prettige samenwerking en het uitvoeren van de betreffende catheterisaties: dr. K.M. Marques en dr. J.E.A. Appelman.

Furthermore, I would like to thank our Greek statistical expert dr. Dimitris Rizopoulos. Dear Dimitris, thank you very much for performing and discussing the statistical analysis of Chapter 6. Next time we plan our meeting for further discussing the most beautiful is-lands of the Greek seas. ‘Σας ευχαριστούμε’

Mijn collega-onderzoekers, Bart Boerrigter en Mariëlle van de Veerdonk. Veel hebben we samengewerkt in de afgelopen jaren. Bart, vanaf je sollicitatiegesprek was er al een goede klik tussen ons. Hoe kan het ook anders? Het hart gaat wel uit Twente, maar Twente niet uit het hart! De hilarische valpartij tijdens de ers in Wenen staat nog scherp op mijn netvlies bij elke ijsco die ik nuttig.

Mariëlle, 3 jaar geleden kwam je bij mij als student-assistent. Ik zie het nog voor me: een enorm enthousiaste studente, met de potentie om een artikel in het jacc te publiceren. Drie jaar later is het dan zover! Veel succes met je promotieonderzoek nu je bijna dokter bent.

Tevens wil ook mijn collega Frank Helderman, bedanken. Frank, de aanhouder wint. Ons project heeft vele vormen aangenomen, maar het uiteindelijke resultaat is toch gewel-dig. Ik wil jou bedanken voor de ontzettend fijne samenwerking. Je hebt de fantastische eigenschap dat je goed kan uitleggen en je weet waar je mee bezig bent. Nu is het dan zover dat we beiden promoveren op het onderzoek dat we samen hebben uitgevoerd. Hierbij hoort natuurlijk ook student-onderzoeker Kirsten Andringa. Beste Kirsten, je hebt je gewel-dig ingezet voor ons onderzoek. Je kwam terug in je coschappenperiode om het een en ander af te maken. Dit heeft Frank en mij regelmatig doen bewonderen. En mijn andere student-onderzoeker Joachim Bosboom. Beste Joachim, ook jij hebt je ingezet voor het onderzoek dat wij samen hebben opgezet. Het heeft wel veel tijd en discussie gekost maar dan heb je ook wat!

Collega onderzoeker Nabil Saouti. Beste Nabil, vanaf het eerste uur hebben wij samen de MrI’s gemaakt tot laat in de avond. Zeker in het begin was 23.00 uur geen uitzonderlijke tijd. Bedankt voor onze samenwerking een veel succes met je opleiding tot cardiothoracaal chirurg en verdere promotietraject.

Oud-collega Jan-Willem Lankhaar. Jan-Willem, ik vind het ontzettend leuk dat jij mijn proefschrift wilde opmaken. Jouw precisie en correctheid weerspiegelen in de opmaak van dit proefschrift. Mijn grote waardering voor het feit dat je, ondanks de drukke werkzaam-heden in het bedrijf waar je nu werkt, tijd heb willen vrijmaken om dit proefschrift een prachtige opmaak te geven.

De andere collega-onderzoekers: Frances en Louis, Gerrina, Pia, Tji-Joong, Nabil, Marieke, Heleen, Herman, Serge, Yeung Ying, Sophie en Jasmijn, bedankt voor de gezellige en leer-zame tijd met elkaar.

148 Acknowledgements

Mijn ene paranimf Taco Kind. Beste Taco, ik kan wel zeggen dat we naast goede colle-ga’s ook goede vrienden zijn geworden. We hebben de afgelopen jaren veel met elkaar samengewerkt en alle ups & downs samen beleefd. Ook jij hebt je promotieonderzoek bijna afgerond. Onvergetelijk die eerste reis naar San Diego voor de ats. Vooral de daaropvol-gende ‘avontuurlijke’ reis door West-Amerika. Het was geweldig dat onze promotor op het vliegveld in San Diego ons daar ‘zo maar’ een paar weken vrij gaf, terwijl hij zelf weer aan het werk ging. Vaak denk ik terug aan de vroege ochtend dat ‘die beer’ voor jullie tent stond in het Palomar Mountain State Park, terwijl ik vanuit de auto vaag de volgende tekst op het gele bord bestudeerde, dat niet ver van mij verwijderd was: Be aware: Bears and mountain lions are active in this area. If you encounter a bear, please do the following A) make yourself look as large as possible and wave with your hands. B) Convince the lion that you are not prey and that you may be dangerous yourself and C) don’t run. Hoewel er geen wetenschappelijke evidence voor was, leek het te helpen. Kortom, een reis om niet gauw te vergeten. Clau-dia, jij ook veel succes met je promotieonderzoek binnen de klinische genetica, ‘een pittig onderzoek’. Marian en ik hopen gauw nog eens langs te komen om ons volgende bordspel uit te spelen.

Daar aangekomen waar dankbaarheid niet meer in woorden uit te drukken is, toch maar een enkele aanduiding. Het is niet voor niets dat ik dit boekje heb opgedragen aan mijn ouders. Pa en ma, jullie hebben me altijd aangemoedigd, gesteund en gestimuleerd in school en studie en creëerden de financiële randvoorwaarden om mijn mogelijkheden te ontplooien. Papa jouw ‘denken in mogelijkheden (dim-denken)’ is van essentieel belang geweest tijdens mijn promotieonderzoek. Je zou een goede copromotor zijn geweest. Lieve ma, veel heb je met papa betekend in mijn leven. Alle (les)uren die je gebruikt hebt in het verleden om ons te overhoren en te helpen zijn ook een proefschrift waard. En zeker niet te vergeten mijn twee lieve zussen, Christina en Marieke, met hun fijne echtgenoten Bertin en Ard. Al zijn jullie twee sterke politieagentes, het is jullie nog niet gelukt om mij op de grond te krijgen. Beiden hebben jullie ook net als ik straks, een sociaal vak waar jullie beiden op je plek zijn. Ik wil jullie bedanken voor je steun, je meeleven en je interesse van je broer(tje). Ik zal jullie missen straks! Ard, ik vind het ontzettend leuk dat jij als zwager en chemicus mijn andere paranimf wil zijn.

Mijn schoonouders/-familie (Christiaan, rianne, Peter en Josephine) allemaal hartelijk dank voor jullie sterke interesse en intense meeleven. Ome Lauw en tante Mar, de band die is gegroeid in de afgelopen 4 jaar is meer dan woorden alleen. Ome Lauw, we waren vaak ’s morgens al heel vroeg op (1.00 uur). Tante Mar, wanneer begint u uw eigen vijfsterrenres-taurant? Voor alles wat jullie voor mij hebben gedaan mijn welgemeende dank. En voor de toekomst een goede buur is beter dan een verre vriend!

Als laatste wil ik mijn schat van een verloofde bedanken. Lieve Marian, je bent mij tot grote steun geweest, met name in de laatste maanden waarin ik je soms niet de aandacht kon geven die je verdient. Het schrijven van dit proefschrift was niet alleen voor mij een beproeving, mijn lange dagen en stress hebben ook veel van jouw energie en nachtrust gevraagd. Bij ‘leven en welzijn’ hoop ik over een half jaar jouw man te worden en jij mijn vrouw. Ik kan mij geen ‘lievere’ en ‘betere’ vrouw wensen.

149Dankwoord

Ack

I’ll close by thanking the Mighty Creator, who not only designed and created the whole universe, but created intelligent human beings able to discover all the beauty He created. I therefore end with a part of Psalm 104 ‘Man goeth forth unto his work and to his labor until the evening. O Lord, how manifold are thy works! In wisdom hast thou made them all: the earth is full of thy riches’. ■

151

List of publications

FULL PAPERS

1. Mauritz GJ, Kind T, Marcus JT , Bogaard HJ, van de Veerdonk M, Postmus PE, Boonstra A, Westerhof N, Vonk-Noordegraaf A. Progressive changes in right ventricular geometric shortening and long-term survival in pulmonary arterial hypertension. Chest. (published online ahead of print, September 2011)

2. Mauritz GJ, Rizopoulos D, Groepenhoff H, Tiede H Felix J, Eilers P, Bosboom J, Postmus PE, Boonstra A, Westerhof N, Vonk-Noordegraaf A. Usefulness of serial N-terminal pro-B-type natriuretic peptide measurements for determining prognosis in patients with pulmonary arterial hypertension. Am J Cardiol. (published online ahead of print, September 2011)

3. van de Veerdonk M , Kind T , Marcus JT, Mauritz GJ , Bogaard HJ , Boonstra A , Marques K, Westerhof N , Vonk-Noordegraaf A. Progressive right ventricular dysfunction in pulmonary arterial hypertension patients responding to therapy. J Am Coll Cardiol. (accepted May 2011)

4. Helderman F, Mauritz GJ, Andringa KE, Vonk-Noordegraaf A, Marcus JT. Early onset of retrograde flow in the main pulmonary artery is a characteristic of pulmonary arterial hypertension. J Magn Reson Imaging. 2011 Jun;33(6):1362-8.

5. Mauritz GJ, Marcus JT, Westerhof N, Postmus PE, Vonk-Noordegraaf A. Prolonged right ventricular post-systolic isovolumic period in pulmonary arterial hypertension is not a reflection of diastolic dysfunction. Heart. 2011 Mar;97(6):473-8.

6. van Wolferen SA, van de Veerdonk MC, Mauritz GJ, Jacobs W, Marcus JT, Marques KM, Bronzwaer JG, Heymans MW, Boonstra A, Postmus PE, Westerhof N, Vonk Noordegraaf A. Clinically significant change in stroke volume in pulmonary hypertension. Chest. 2011 May;139(5):1003-9.

7. Kind T, Mauritz GJ, Marcus JT, van de Veerdonk M, Westerhof N, Vonk-Noordegraaf A. Right ventricular ejection fraction is better reflected by transverse rather than longitudinal wall motion in pulmonary hypertension. J Cardiovasc Magn Reson. 2010 Jun 4;12:35

8. Boerrigter B, Mauritz GJ, Marcus JT, Helderman F, Postmus PE, Westerhof N, Vonk-Noordegraaf A. Progressive dilatation of the main pulmonary artery is a characteristic of pulmonary arterial hypertension and is not related to changes in pressure. Chest. 2010 Dec;138(6):1395-401.

9. Marcus JT, Mauritz GJ, Kind T, Vonk-Noordegraaf A. Interventricular mechanical dyssynchrony in pulmonary arterial hypertension: early or delayed strain in the right ventricular free wall? Am J Cardiol. 2009 Mar 15;103(6):894-5.

10. Mauritz GJ, Marcus JT, Boonstra A, Postmus PE, Westerhof N, Vonk-Noordegraaf A. Non-invasive stroke volume assessment in patients with pulmonary arterial hypertension: left-sided data mandatory. J Cardiovasc Magn Reson. 2008 Nov 5;10:51.

152

AbSTRACTS

1. Mauritz GJ, Bosboom J, Postmus PE, Westerhof N, Vonk Noordegraaf A. The Prognostic Value Of Repeated Measurements Of N-terminal Pro-B-type Natriuretic Peptide In Patients With Pulmonary Hypertension. Am J Resp Crit Care Med 2010.

2. Boerrigter B, Mauritz GJ, Marcus JT, Helderman F, Postmus PE, Westerhof N, Vonk-Noordegraaf A. Progressive dilatation of the main pulmonary artery is a characteristic of pulmonary arterial hypertension and is not related to changes in pressure. Am J Resp Crit Care Med 2010.

3. Kind T, Mauritz GJ, Marcus JT, van de Veerdonk M, Westerhof N, Vonk-Noordegraaf A Reduced right ventricular transverse wall motion is indicative of right ventricular dysfunction in pulmonary arterial hypertension. European Respiratory Society, Vienna 2009.

4. Mauritz GJ, van de Veerdonk M, Kind T, Marcus JT, Westerhof N, Vonk-Noordegraaf A. Prognostic value of right ventricular transverse wall motion in pulmonary arterial hypertension. European Respiratory Society, Vienna 2009.

5. Mauritz GJ, Marcus JT, Boonstra A, Postmus PE, Westerhof N, Vonk-Noordegraaf A. Non-invasive stroke volume assessment in patients with pulmonary arterial hypertension: left-sided data mandatory. Am J Resp Crit Care Med 2009.

ORAL PRESENTATIONS

1. Mauritz GJ. Changes in right ventricular configuration between survivors and nonsurvivors in pulmonary arterial hypertension. ERS Annual Congress, Barcelona 2010.

2. Mauritz GJ. The role of CMR in pulmonary arterial hypertension (Invited speaker). The annual scientific Meeting of EuroCMR, Florence 2010.

3. Mauritz GJ. Right ventricular post-systolic isovolumic phase reflects prolonged myocardial shortening in pulmonary arterial hypertension. ERS Annual Congress, Vienna 2009.

153Publications

Publ

155

Over de auteur

About the author

Gerrit-Jan Mauritz was born in rijssen, the Netherlands on November 22, 1983. He grew up in rijssen and completed secondary school (vwo) at the Jacobus Fruytier Scholengemeen-schap in Apeldoorn in 2001. Subsequently, he started studying biomedical engineering at the University of Twente. After one year, passing his first year examination (propaedeutics), he started medical school at the vu University Medical Center Amsterdam. During his study period, he was a board member of the medical faculty student organization (mfvu). Furthermore, he was student-assistant at the department of Physiology, and selected for the vumc Honors Program, which triggered his interest for medical research. Subsequently, before finishing medical school, he started the research for his thesis, at the department of Pulmonary Diseases, under the supervision of prof. A. Vonk Noordegraaf, PhD and J.T. Marcus, PhD. During his PhD-research period, he was member of the vumc/Institute for Cardiovascular research (icar) committee. In November 2010, he started an internship at the Leiden University Medical Center and he will obtain his medical license in September 2012. Gert-Jan is engaged to Marian raatgever, whom he will marry in May 2012.

Gerrit-Jan Mauritz werd geboren in rijssen op 22 november 1983. Hij groeide op in rijs-sen en haalde zijn vwo-diploma op de Jacobus Frutier Scholengemeenschap in Apeldoorn in 2001. Vervolgens haalde hij zijn propedeuse biomedische technologie aan de Univer-siteit Twente. Daarna studeerde hij geneeskunde aan het vu medisch centrum. Tijdens zijn studie was hij bestuurslid van de Medische Faculteitsvereniging (mfvu). Daarnaast was hij student-assistent bij de afdeling fysiologie en werd hij geselecteerd voor het vumc Honors Program, waar zijn interesse voor onderzoek werd gewekt. Alvorens zijn artsopleiding af te ronden, begon hij met het onderzoek voor zijn proefschrift op de afdeling longziekten van het vu medisch centrum onder begeleiding van prof.dr. A. Vonk Noordegraaf en dr. J.T. Marcus. Tijdens zijn promotieonderzoek was hij lid van de vumc/Institute for Cardiovas-cular research-commissie (icar). In november 2010 begon hij met zijn coschappen in het Leids Universitair Medisch Centrum. In september 2012 zal hij zijn artsbevoegdheid halen. Gert-Jan is verloofd met Marian raatgever met wie hij in mei 2012 gaat trouwen.

Dan wordt de mens door ’t rijzend morgenlicht Gewekt, gewenkt tot arbeid, tot zijn plicht; Hij plant, hij bouwt; men ziet hem zwoegen, draven; Tot ’s avonds toe laat hij niet af van slaven. Hoe schoon, hoe groot, o Oppermajesteit, Is al Uw werk, gevormd met wijs beleid! Uw wijsheid streelt oplettende gemoederen; Al ’t aardrijk is vervuld met Uwe goederen

Psalm 104 : 12 (psalmberijming 1773)

ISBN 978-90-9026454--7

Progressive Dilatation of the Main Pulmonary Artery Is a Characteristic of Pulmonary Arterial...

Documents

Transcript of Progressive Dilatation of the Main Pulmonary Artery Is a Characteristic of Pulmonary Arterial...