Orthostatic blood pressure response, carotid intima–media thickness, and plasma fibrinogen in...

Original Article

Orthostatic blood pressure response, carotid intima^media thickness, and plasma fibrinogen in oldernondiabetic adults

Artur Fedorowskia,b, Gerd Ostlinga,b, Margaretha Perssona,b, Joachim Struckc, Gunnar Engstromb,Peter M. Nilssona,b, Bo Hedblada,b, and Olle Melandera,b

Journal of Hypertension 2012, 30:522–529aCenter for Emergency Medicine, bLund University, Department of Clinical Sciences,Clinical Research Center, Skane University Hospital, Malmo, Sweden and cResearchDepartment, BRAHMS AG, Hennigsdorf, Germany

Correspondence to Artur Fedorowski, MD, PhD, Center for Emergency Medicine,Entrance 35, Floor 2, Skane University Hospital, 205 02 Malmo, Sweden. Tel: +46 4033 10 00; fax: +46 40 33 62 08; e-mail: [email protected]

Received 14 July 2011 Revised 6 September 2011 Accepted 24 November 2011

J Hypertens 30:522–529 � 2012 Wolters Kluwer Health | Lippincott Williams &Wilkins.

52

Objective: Although recent studies have indicated thatboth orthostatic hypotension and orthostatic hypertension(OHTN) independently predict cardiovascular events, theunderlying mechanisms are still debatable.

Methods: A total of 700 nondiabetic adults (43% men,age 64 years) were examined by orthostatic blood pressure(BP) test, carotid artery ultrasonography, and biochemicaltests including plasma fibrinogen and lipid profile.Multivariate-adjusted logistic regression was applied toassess association of intima–media thickness (IMT) andP-fibrinogen with orthostatic hypotension and OHTN. Inaddition, distribution of IMT and P-fibrinogen acrossquintiles of orthostatic systolic BP (SBP) response wasanalyzed.

Results: Orthostatic hypotension and OHTN were found in40 (5.7%) and 45 (6.4%) study participants, respectively.Both IMT [odds ratio (OR), 95% confidence interval (CI)per one-SD increment: 1.27, 1.01–1.60; P¼0.04] andP-fibrinogen (OR 1.44, 1.07–1.93; P¼0.02) wereassociated with orthostatic hypotension in a crude model.After adjustment relationship between orthostatichypotension and IMT was slightly attenuated (OR 1.26,0.96–1.65; P¼ 0.09) but was substantially unchanged inregard to P-fibrinogen (OR 1.45, 1.06–1.99; P¼ 0.02). Incontrast, OHTN showed no association with either IMT orP-fibrinogen (adjusted OR 1.09, 0.78–1.52; P¼0.61, and0.97, 0.70–1.34; P¼0.84, respectively). Distribution ofIMT across quintiles of orthostatic SBP response was U-shaped, whereas that of fibrinogen was more linear butnone of borderline quintiles (with pronounced hypertensiveor hypotensive response) significantly differed from themiddle quintiles in a fully adjusted model.

Conclusion: In older nondiabetic adults only orthostatichypotension seems to independently correlate withincreased carotid atherosclerosis and systemicinflammation.

Keywords: atherosclerosis, biomarkers, fibrinogen,hypertension, intima–media thickness, orthostatichypotension

Abbreviations: ACEIs, angiotensin-converting enzymeinhibitors; ARBs, angiotensin receptor blockers; CCBs,calcium channel blockers; CVD, cardiovascular disease;

Copyright © Lippincott Williams & Wilkins. Unaut2 www.jhypertension.com

hs-CRP, high-sensitive C-reactive protein; IMT, intima–media thickness; Lp-PLA2, lipoprotein-associatedphospholipase 2; MDCS, Malmo Diet and Cancer Study;MDCS-CC, Malmo Diet and Cancer Study – CardiovascularCohort; MR-proADM, midregional proadrenomedullin;MR-proANP, midregional proatrial natriuretic peptide;NT-proBNP, N-terminal pro-B-type natriuretic peptide;OHTN, orthostatic hypertension; SBP, systolic bloodpressure

INTRODUCTION

Although recent studies have indicated that ortho-static hypotension and hypertension (OHTN) con-stitute independent risk factors for mortality and

cardiovascular morbidity [1–6], the underlying mechanismsare still the subject of debate. In particular, whether or notthe directionality of orthostatic blood pressure (BP)response is associated with progression of arteriosclerosisand vascular remodeling remains an open question. Aspreviously reported, both carotid intima–media thickness(IMT) and arterial stiffness may be linked to orthostatichypotension [7,8] in general population. In contrast, verylittle is known about relationship between OHTN and thesesurrogate markers of vascular damage. In a small study(n¼ 136) of community dwelling octogenarians, anincreased IMT was associated with a pronounced posturalsystolicBP (SBP) response (decreaseor increase�20mmHg)in women [9]. IMT is generally seen as a useful screeningtool for subclinical atherosclerosis and an independentpredictor of cardiovascular events [10–12]. In parallel, it

horized reproduction of this article is prohibited.

DOI:10.1097/HJH.0b013e32834fa860

Volume 30 � Number 3 � March 2012

mailto:[email protected]

Malmö diet and cancer study (MDCS, 1991-96)n = 28 449

men and women born 1923-50

MDCS- cardiovascular cohortn = 6 103

IMT, CRP, cystatin C, LpPLa2, MR-proADM,MR-proANP, and NT - proBNP

MDCS- cardiometabolicsubcohort (1999-2000)

n = 901no history of diabetes

IMT, lipids, and fibrinogen

Orthostatic blood pressurestudy

n = 700supine and standing BP

FIGURE 1 Overall study design. CRP, C-reactive protein; IMT, intima–media thick-ness; Lp-PLA2, lipoprotein-associated phospholipase 2; MDCS, Malmo Diet andCancer Study; MR-proADM, midregional proadrenomedullin; MR-proANP, midre-gional proatrial natriuretic peptide; NT-proBNP, N-terminal pro-B-type natriureticpeptide.

Orthostatic BP response, IMT and fibrinogen

has been demonstrated that carotid (and coronary) athero-sclerosis correlates with soluble markers of inflammationsuch as plasma fibrinogen [13,14]. If disorders of orthostatichemodynamics could be related to changes in carotidarteries and inflammatory biomarkers, the longitudinalassociation of manifest cardiovascular disease (CVD) withorthostatic hypotension and OHTN would be partiallyexplained.

The aim of this cross-sectional study was to investigateIMT and plasma fibrinogen among older nondiabetic adultsin relation to early postural BP response defined as ortho-static hypotension or OHTN, and as a continuous variableranked by quintiles of SBP change on standing.

METHODS

Study populationWithin the prospective population-based ‘Malmo Diet andCancer’ Study (MDCS, 1991–1996), a random sample of6103 out of 28 449 initially included persons formed acardiovascular cohort (MDCS-CC) in order to more pro-foundly study risk factors for CVD (Fig. 1) [15]. Participantsof MDCS-CC were additionally investigated by carotidartery ultrasonography, lipid profile and fasting bloodsampling [16]. At the next stage, after a mean of 6.6 years,a subset of 909 individuals without known diabetes wasselected for assessment of cardiometabolic status. Thereinvestigation procedure was described previously [17].Although not included in the baseline MDCS examination,orthostatic BP and plasma fibrinogen measurements wereadded to the extended re-examination protocol. Completedata on orthostatic BP response were available among 700re-examined individuals. In addition, authors had access todata on baseline IMT and baseline cardiovascular bio-markers, high-sensitive C-reactive protein (hs-CRP), cysta-tin C, lipoprotein-associated phospholipase 2 (Lp-PLA2),midregional proadrenomedullin (MR-proADM), midre-gional proatrial natriuretic peptide (MR-proANP), andN-terminal pro-B-type natriuretic peptide (NT-proBNP) inthe studied sample.

Basic examinationThe invitedparticipantswere instructed to abstain from food,alcohol and tobacco for 12h prior to the examination butwere allowed to drink water ad libitum. The examinationwas performed between 0800 and 1100h. The body massindex (BMI)was calculatedasweight inkilogramsdividedbyheight squared in meters. BP was measured twice after 5minrest in semi-recumbentpositionby trainednurseswhousedasphygmomanometer with a three-cuff and a mercury man-ometer. Themean figurewas recorded.Next, theparticipantsrested for 5min in supine position, and BP was measuredwith the arm placed paralleled to the body and cuff sup-ported at the heart level. Then, the participantswere asked tostand up and the measurement was repeated after 120 s instandingpositionwith the armhanging along thebody.Basicfacts regardinghistoryofCVD(definedasprevalent coronaryevent or stroke), smoking, medication against hypertensionand hyperlipidemia were collected through a questionnaire.Those who confirmed regular or occasional current smokingwere classified as smokers.

Copyright © Lippincott Williams & Wilkins. UnauthJournal of Hypertension

Definition of orthostatic hypotension andhypertensionExtreme orthostatic BP response was defined as eitherorthostatic hypotension [decrease in SBP �20mmHg ordecrease in diastolic BP (DBP) �10mmHg] [18] or ortho-static hypertension (OHTN, increase in SBP �20mmHg)[19] according to widely accepted criteria.

Carotid artery ultrasonographyThe methodology of IMT measurement was previouslydescribed in detail [17,20]. Briefly, an Acuson 128 XP4 Ultra-sound System (Acuson, Mountain View, California, USA)with a 7-MHz transducer was used. The examination wasperformed by trained and certified ultrasonographers. Theright carotid bifurcation was scanned within a predefinedwindow comprising 3 cm of the distal common carotidartery, the bifurcation and 1 cm of the internal and externalcarotid arteries, respectively. The common carotid artery IMTin the far wall was measured off-line along a 1-cm section justproximal to the bulb. Images were taken in the longitudinalprojection and all measurements were done according to theleading edge principle using a computer-assisted imageanalyzing system based on automated detection of the echostructures. The investigators had the option to manuallymodify the automatically outlined echoes. Presence ofplaques, defined as a focal thickening of the intima–mediacomplex greater than 1.2mm, was assessed.

orized reproduction of this article is prohibited.www.jhypertension.com 523

Fedorowski et al.

Laboratory analysisPlasma fibrinogen concentration was measured photo-optically according to the modified Clauss method usingMultifibren U reagent (Dade Behring Marburg GmbH,Marburg, Germany) in Department of Clinical Chemistry,Malmo University Hospital. Mean interassay coefficient ofvariation was 1.5% at level 3.0 g/l and 3.8% at level 1.0 g/l.The methods of measuring low-density lipoprotein (LDL)and high-density lipoprotein (HDL) cholesterol, and trigly-ceride (all in mmol/l) were described previously [16] aswere analytical assays for determination of CRP, cystatin C,Lp-PLA2, MR-proADM, MR-proANP, and NT-proBNP atbaseline [21].

Statistical analysisGroup-wise differences in continuous variables betweenorthostatic hypotension or OHTN-positive individuals andboth orthostatic hypotension and OHTN-negative controlswere compared using one-way ANOVA test, whereas categ-orical variables were compared using Pearson’s chi-squared test. Biomarker concentrations with right-skeweddistribution (NT-proBNP and hs-CRP) were reported asmedian and interquartile range. Cross-sectional associationbetween IMT, plasma fibrinogen and orthostatic hypoten-sion/OHTN was assessed by applying three logisticregression models: univariate, multivariate-adjusted forage and sex, and a final model additionally adjusted forBMI, SBP and DBP, antihypertensive treatment, lipid-low-ering treatment, current smoking, prevalent CVD, LDL andHDL-cholesterol, and triglycerides. In a subsidiary analysis,the predictive value of baseline IMT and cardiovascularbiomarkers in relation to orthostatic hypotension or OHTNwas assessed using corresponding logistic regressionmodels as above. Odds ratios (ORs) were calculated perone SD increment in the analyzed parameter unless specifi-cally indicated. Biomarker concentrations with right-skewed distribution (NT-proBNP and CRP) were log-trans-formed before analysis. In addition, orthostatic SBPresponse (standing – supine SBP) was ranked by quintiles.Then, authors applied one-way ANOVA to assess IMT andplasma fibrinogen distribution across quintiles of SBPresponse. Quintiles with pronounced hypertensive andhypotensive response (Q1 and Q5, respectively) werecompared as dependent variables to merged middle quin-tiles (Q2 to Q4) using three analogical logistic regressionmodels by entering IMT or P-fibrinogen as an independentvariable, and the same covariates as in the orthostatichypotension and OHTN analysis.

All analyses were performed using IBM SPSS Statisticsversion 19.0 for Windows (SPSS Inc., Chicago, Illinois,USA). All tests were two-sided whereby P less than 0.05was considered statistically significant.

RESULTS

General characteristics of orthostatichypertension and hypotensionOrthostatic hypotension and hypertension were found in40 (5.7%) and 45 (6.4%) out of 700 studied individuals,respectively. Distribution of SBP and DBP response toorthostasis was normal [mean SBP and DBP increment


on standing (SD): þ2.6 (12.3) mmHg, and þ5.5 (7.5)mmHg, respectively]. As can be seen in Table 1, bothIMT and P-fibrinogen were increased in individuals withorthostatic hypotension, whereas those with OHTN hadhigher age and BMI and were more frequently treated withantihypertensive agents as compared to those withoutorthostatic hypotension or OHTN, in particular with diu-retics (17.8 vs. 6.8%). Moreover, in a retrospective analysis,baseline levels of biomarkers (except for hs-CRP andLp-PLA2 mass) were increased among those who werefound to have OHTN but not orthostatic hypotension atreinvestigation. In contrast, proportions of those takingantihypertensive agents at baseline did not significantlydiffer among the groups, although there were more treatedindividuals among those with hypertensive orthostaticresponse (þ 6.1%) as compared to the reference group.

Orthostatic hypertension and hypotension vs.intima–media thickness, fibrinogen andbiomarkersAs shown in Fig. 2, both IMT (OR, 95% confidence interval(CI) per one-SD increment: 1.27, 1.01–1.60; P¼ 0.04] andP-fibrinogen (OR 1.44, 1.07–1.93; P¼ 0.02) were associatedwith orthostatic hypotension in a univariate logisticregression model. After adjustment for traditional risk fac-tors the relationship between orthostatic hypotension andIMT was slightly attenuated (OR 1.26, 0.96–1.65; P¼ 0.09)but was substantially unchanged in regard to P-fibrinogen(OR 1.45, 1.06–1.99; P¼ 0.02). In the fully adjusted model,OHTN was associated with age (OR per 5 years: 1.55, 1.21–1.92; P¼ 0.001), higher BMI (OR per 5 kg/m2: 1.42, 1.01–1.85; P¼ 0.04), and lower supine SBP (OR per 10 mmHgdecrement: 1.27, 1.01–1.53; P¼ 0.04) but not with IMT or P-fibrinogen (OR 1.09, 0.78–1.52; P¼ 0.61, and 0.97, 0.70–1.34; P¼ 0.84, respectively). In contrast, none of the cova-riates included in model 3 was significantly associated withorthostatic hypotension (data not shown). Furthermore, outof baseline MDCS biomarkers, higher NT-proBNP concen-tration was predictive of OHTN in a multivariate-adjustedlogistic regression model (OR 1.79, 1.25–2.57; P¼ 0.002),whereas baseline LpPLA2 mass level was inversely corre-lated with OHTN (OR 0.58, 0.39–0.85; P¼ 0.006). Otherbiomarkers were not independently predictive of ortho-static hypotension or OHTN. Finally, increased baselineIMT was associated with presence of orthostatic hypoten-sion at reinvestigation even after multivariate adjustmentaccording to model 3 (OR 1.44, 1.07–1.95; P¼ 0.02).

Orthostatic systolic blood pressure response vs.intima–media thickness and fibrinogenAs shown in Fig. 3, the distribution of IMT across quintilesof orthostatic SBP response was U-shaped, whereas that ofP-fibrinogen was more linear with slightly higher concen-tration in the first quintile characterized by the most hyper-tensive response. The fifth quintile (i.e. those withpronounced hypotensive response) had significantly largerIMT as compared to the middle quintiles in the crudelogistic regression model (OR, 95% CI per one-SD incre-ment of IMT: 1.23, 1.04–1.45; P¼ 0.02). Adjustment for ageand sex moderately attenuated this relationship (OR 1.19,

horized reproduction of this article is prohibited.Volume 30 � Number 3 � March 2012

TABLE 1. Characteristics of study participants (n¼700) according to status of orthostatic response at reinvestigation

Characteristics OH (n¼40) No OH or OHTN (n¼615) OHTN (n¼45)

Reinvestigation (1999–2000)Age (years) 65.2 (6.4); P¼0.13 63.7 (5.8) 66.7 (5.6); P¼0.001

IMT cca (mm) 0.89 (0.29); P¼0.04 0.83 (0.18) 0.86 (0.15); P¼0.32

P-fibrinogen (g/l) 4.1 (0.7); P¼0.015 3.8 (0.8) 3.8 (0.9); P¼0.77

Sex (male, %) 45.0; P¼0.78 42.8 40.0; P¼0.72

BMI (kg/m2) 27.0 (3.3); P¼0.97 27.0 (4.4) 28.7 (4.7); P¼0.008

Smoking (%) 17.5; P¼0.25 25.7 13.3; P¼0.06

Systolic blood pressure, supine (mmHg) 138 (20); P¼0.16 134 (18) 132 (17); P¼0.45

Diastolic blood pressure, supine (mmHg) 84 (9); P¼0.13 82 (10) 81 (10); P¼0.68

Antihypertensive treatment (%) 20.5; P¼0.80 18.9 31.1; P¼0.05

Beta-blockers (%) 15.4; P¼0.47 11.6 17.8; P¼0.22

CCBs (%) 2.6; P¼0.61 4.2 8.9; P¼0.15

ACEIs/ARBs (%) 2.6; P¼0.85 2.1 2.2; P¼0.96

Diuretics (%) 5.1; P¼0.68 6.8 17.8; P¼0.02

LDL-cholesterol (mmol/l) 3.7 (1.1); P¼0.61 3.6 (0.8) 3.7 (0.9); P¼0.56

HDL-cholesterol (mmol/l) 1.5 (0.4); P¼0.72 1.5 (0.4) 1.5 (0.3); P¼0.92

Triglycerides (mmol/l) 1.4 (0.8); P¼0.25 1.3 (0.6) 1.4 (0.6); P¼0.47

Lipid-lowering treatment (%) 2.6; P¼0.81 3.3 4.4; P¼0.67

History of CVD (%) 7.5; P¼0.24 3.7 4.4; P¼0.81

Baseline examination (1991–1996)Age (years) 59.0 (6.3); P¼0.15 57.7 (5.8) 60.7 (5.5); P¼0.001

IMT cca (mm) 0.80 (0.20); P¼0.003 0.73 (0.15) 0.75 (0.14); P¼0.28

Antihypertensive treatment (%) 17.5; P¼0.83 16.1 22.2; P¼0.29

Hs-CRP� (mg/l) 0.13 [0.08–0.23]; P¼0.45 0.15 [0.07–0.32] 0.11 [0.06–0.36]; P¼0.61

Cystatin C (mg/l) 0.83 (0.16); P¼0.09 0.79 (0.13) 0.84 (0.16); P¼0.01

Lp-PLA2 mass (ng/ml) 261 (67); P¼0.20 279 (86) 247 (62); P¼0.01

MR-proANP (pmol/l) 80 (42); P¼0.13 71 (33) 89 (62); P¼0.002

MR-proADM (nmol/l) 0.50 (0.12); P¼0.11 0.47 (0.12) 0.53 (0.14); P¼0.003

NT-proBNP� (pg/ml) 65 [40–125]; P¼0.15 57 [31–102] 88 [52–197]; P<0.001

Data presented as mean (SD), percentage or median with inter-quartile range�.ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin receptor blockers; BMI, body mass index; CCBs, calcium channel blockers; HDL, high-density lipoprotein; hs-CRP,high-sensitive C-reactive protein; IMT cca, intima–media thickness in common carotid artery; LDL, low-density lipoprotein; Lp-PLA2, lipoprotein-associated phospholipase 2; MR-proADM,midregional proadrenomedullin; MR-proANP, midregional proatrial natriuretic peptide; NT-proBNP, N-terminal pro-B-type natriuretic peptide; OH, orthostatic hypotension; OHTN,orthostatic hypertension.


1.00–1.41; P¼ 0.05), which in the fully adjusted model 3 wasno longer significant (OR 1.15, 0.95–1.37; P¼ 0.15). Inparallel, the first quintile had also larger IMT but only ofborderline significance (OR 1.19, 0.98–1.44; P¼ 0.09) in acrude model. After adjustment for age and sex this relation-ship was remarkably attenuated (OR 1.14, 0.92–1.40;P¼ 0.22). Moreover, in the crude model, the first quintileshowed a distinct trend toward higher P-fibrinogen as com-pared to the middle quintiles (OR, 95% CI per one-SDincrementofP-fibrinogen: 1.19, 0.98–1.44;P¼ 0.08),moder-ately attenuated after adjustment for age and sex (OR 1.16,0.96–1.41; P¼ 0.14), but there was no difference betweenQ5 and the middle quintiles (OR 0.96, 0.79–1.17; P¼ 0.67).

DISCUSSIONIn recent years, first hypotensive and then hypertensiveresponse of autonomic nervous system during early phase(1–3min) of orthostatic challenge have been studied indifferent settings and prospectively related to cardiovascu-lar events [4,6]. However, mechanisms linking pronouncedorthostatic BP variability with increased mortality and car-diovascular morbidity are largely unknown. The presentstudy provides a new insight into this association.

Orthostatic hypotensionBoth carotid artery IMT and plasma fibrinogen wereincreased in orthostatic hypotension, the latter reported


for the first time to the best of authors’ knowledge. Enlarge-ment of IMT, suggesting accelerated progression of athero-sclerosis in orthostatic hypotension, was in concordancewith previous studies on orthostatic hypotension-relatedIMT increment observed in middle-aged individuals withorthostatic hypotension [The Atherosclerosis Risk in Com-munities (ARIC) study] [7], and nearly 70% higher preva-lence of carotid stenosis (>50%) among community-dwelling older adults with orthostatic hypotension (theCardiovascular Health Study) [22]. Interestingly, an absolutedifference of IMT value between those who had and thosewho had not orthostatic hypotension was identical (0.80 vs.0.73 mm) in both studied subcohort of MDCS-CC at baseline(mean age 58 years) and ARIC (mean age 54 years) [7],indicating that results of the present study are generalizable.The enlargement of IMT was paralleled by increase inplasma fibrinogen. Based on a previous meta-analysis[23], increment of 1 g/l in plasma fibrinogen confers analmost doubled risk of all-cause mortality, coronary heartdisease and stroke. Taken together, these two well estab-lished predictors of mortality and cardiovascular morbidity(i.e. IMT and fibrinogen) offer a plausible explanation forthe increased risk of various fatal and nonfatal eventsobserved in orthostatic hypotension. At this stage, wecannot discern whether autonomic hyporesponsivness iscausally related with carotid atherosclerosis and hyperfi-brinogenemia. It has been demonstrated that prolongedstanding leads to increase in plasma fibrinogen and state of


Odds ratio (95 % confidence interval)

3.02.52.01.51.00.5

1.05 (0.77-1.42)

1.27 (1.01-1.60)OH IMT Model 1

1.24 (0.97-1.60)Model 2

1.26 (0.96-1.65)Model 3

1.44 (1.07-1.93)OH Fib Model 1

1.43 (1.06-1.93)Model 2

1.45 (1.06-1.99)Model 3

1.15 (0.88-1.51)OHTN IMT Model 1

1.02 (0.75-1.38)Model 2

1.09 (0.78-1.52)Model 3

OHTN Fib Model 1

1.00 (0.72-1.37)Model 2

0.97 (0.70-1.34)Model 3

FIGURE 2 Carotid intima–media thickness (IMT) and plasma fibrinogen (P-Fib) in relation to orthostatic hypotension and hypertension (OH and OHTN) according todifferent logistic regression models. Odds ratios are presented per one-standard deviation increment of IMT and P-Fib. Model 1, crude; model 2, adjusted for age and sex;model 3, adjusted for age, sex, body mass index, systolic and diastolic BP, antihypertensive treatment, lipid-lowering treatment, current smoking, history of cardiovasculardisease, low-density lipoprotein and high-density lipoprotein cholesterol, and triglycerides.

Fedorowski et al.

‘orthostatic hypercoagulation’ [24]. We propose that pro-nounced and recurrent BP fall during orthostatic challengemay augment this effect by additional activation ofendogenous vasoconstrictors such as noradrenalin, angio-tensin, vasopressin, and endothelins. In parallel, posturalBP swings may cause an increased arterial wall stressleading to accelerated progression of atherosclerosisthrough both minor mechanical injuries and release ofvasoactive substances such as nitric oxide, analogically tothe presence of high pulse pressure [25], a correlate oforthostatic hypotension [26]. Moreover, nocturnal hyper-tension, often associated with orthostatic hypotension [27],can promote carotid (and systemic) atherosclerosis byincreasing the tonic pressure load during sleep [28]. Inter-estingly, association of orthostatic BP fall with IMT seems tobe quasi dose-dependent, as shown by quintiles of SBPresponse, whereas that with P-fibrinogen probably has adistinct threshold in the vicinity of a standard orthostatichypotension definition (SBP decrement �20mmHg) asP-fibrinogen level did not differ between Q5 and the middlequintiles.

Orthostatic hypertensionOrthostatic hypertension, as a categorical variable, did notdemonstrate an independent association with IMT and P-fibrinogen. On the contrary, authors noted a nonsignificanttrend for increment of IMT and P-fibrinogen within thequintile characterized by hypertensive BP response (Q1).Here, the hypothetical mechanisms might include primaryhyperactivation of vasoconstrictors during orthostasis.Increase of BP on standing can involve similar vascular


mechanisms as those found in essential hypertension(increased tonic pressure and pulsatile stress), additionallyaugmented by oscillatory shear stress related to increased BPvariability [12,28]. Moreover, it has been shown that P-fibri-nogen increases parallelly to SBP [23]. Thus, the orthostaticstress in combination with temporary hypertension may leadto the state of ‘orthostatic hypercoagulability’, as describedbefore [24]. However, increment of IMT and plasma fibri-nogen was more expressed in orthostatic hypotension,which suggests that other orthostatic hypotension-specificmechanisms could be involved. Indeed, in contrast withorthostatic hypotension, hypertensive response to orthosta-sis was consistently shown to correlate with cerebrovasculardisease only [29] but not with coronary disease [2].

Age, higher BMI, and lower supine SBP were all inde-pendent determinants of OHTN. Two latter are logicallyassociated with OHTN as both lower BMI and higher supineSBP correlate with orthostatic hypotension [2], whereas agemay contribute to OHTN through both vascular stiffeningand enhanced sympathetic activity [30]. Strikingly, in theanalysis of DSBP quintiles, adjustment for age and sex dis-tinctly attenuated the observed trends (i.e. Q1 vs. Q2–4),which suggests that advancing age is the crucial confound-ing factor related to hypertensive response. Among base-line biomarkers, only NT-proBNP independently predictedorthostatic hypertension, although MR-proADM, MR-proANP, and cystatin C were also increased. This findingis very intriguing in the light of previous studies indicatingthat hypertensive orthostatic response is inversely relatedwith incident heart failure [31], and that NT-proBNP predictscardiovascular events [21]. A possible explanation of this


Car

oti

d in

tim

a–m

edia

thic

knes

s (m

m)

0.90

0.88

0.86

0.84

0.82

0.80

0.87

0.82

0.820. 83

0.85

over 13 mmHg 6 to 12 mmHg 1 to 5 mmHg -7 to 0 mmHg under -8 mmHg

Q5Q4Q3Q2Q1

P -

Fib

rin

og

en (

g/l

)

4.0

3.9

3.8

3.7

3.6

3.5

3.733.75

3.84

3.74

3.90

over 13 mmHg 6 to 12 mmHg 1 to 5 mmHg -7 to 0 mmHg under -8 mmHg

Q5Q4Q3Q2Q1

(b)

(a)

FIGURE 3 (a and b) Mean carotid intima–media thickness (a) and plasma fibrinogen (b) by quintiles of orthostatic systolic blood pressure response. Quintiles [number,mean (SD), range]: Q1, n¼131, þ19.3 (6.9) mmHg, 13.0 to 50.0 mmHg; Q2, n¼166, þ8.8 (2.0) mmHg, 6.0 to 12.0 mmHg; Q3, n¼87, þ3.1 (1.2) mmHg, 1.0 to5.0 mmHg; Q4, n¼172, �2.1 (2.3) mmHg, �7.0 to 0.0 mmHg; Q5, n¼144, �14.4 (8.0) mmHg, �50.0 to �8.0 mmHg.


paradox might be a more frequent use of diuretics in OHTNgroup as diuretics have been linked to increased levels ofnatriuretic peptides [32]. An apparent dissociation ofcoupling between OHTN, NT-proBNP, and incident heartfailure warrants more studies in settings in which OHTNand biomarker levels are determined and prospectivelyrelated to incident cardiovascular disease.


Orthostatic response vs. blood pressureinstability

Resting BP is usually assessed in supine or sitting positionand orthostatic measurements are recommended in symp-tomatic patients only [33]. In recent years several authorshave raised the importance of BP variability and instability


Fedorowski et al.

as independent predictors of increased cardiovascular risk,especially in regard to cerebrovascular disease [34]. Bothorthostatic hypotension and OHTN are logically implicatedin postural instability and diurnal variability of BP, and bothincrease the risk of stroke. The presence of U-shaped IMTcurve in population stratified by quintiles of orthostatic SBPresponse provides a plausible explanation for this relation-ship. However, a clearly stronger association between theclassically defined orthostatic hypotension and both IMTand fibrinogen might explain why orthostatic hypotensionis in general a more potent risk factor for cardiovasculardisease than is OHTN, even if only stroke incidence is takeninto account [29].

Study strengths and limitationsThis study is based on a diabetes-free representativepopulation sample, in which prospective associationbetween baseline IMT and cardiovascular events waspreviously demonstrated [12]. Combined surrogatemarkers of atherosclerosis and systemic inflammationwere assessed and reported for the first time in a subsetof individuals stratified by orthostatic BP response. How-ever, there are several limitations that should beaddressed. Firstly, cardiovascular biomarkers were ana-lyzed at baseline examination (approximately 7 yearsbefore this cross-sectional study), which mostly reflectstheir predictive role towards subsequent orthostatic BPinstability. It should be emphasized that this part of thestudy was subsidiary only and did not influence the mainresults but a more comprehensive biomarker study inorthostatic hypotension and OHTN would be of a greatvalue. Secondly, orthostatic BP response was assessed ina single measurement, which may have led to eitherunderestimation or overestimation of orthostatic hypo-tension and hypertension. On the contrary, prevalence oforthostatic hypotension and OHTN in the studied popu-lation was in agreement with previous studies, in which5–10% of similarly aged individuals were found to haveorthostatic hypotension [1,2]. Thirdly, although analyseswere adjusted for multiple potential confounders, someother factors influencing hemodynamic homeostasis,such as sodium and alcohol consumption, physical exer-cise, use of antidepressants, may have partially explainedthe studied associations. The distribution of these factorsdid not, however, significantly differ across the groupsand they were consequently omitted in the adjustedmodels with the exception of sodium consumption inwhich case data were not available.

In conclusion, in a diabetes-free population of olderadults (�64 years) both increased carotid IMT and plasmafibrinogen are associated with orthostatic hypotension butnot hypertension. Distribution of carotid IMT across quin-tiles of orthostatic SBP response is U-shaped, whereas thatof plasma fibrinogen shows a trend toward a slightly higherlevel in a quintile with pronounced hypertensive response.However, increment of IMT and plasma fibrinogen amongthose with hypertensive SBP response could be to someextent explained by age difference. Taken together, onlyhypotensive orthostatic response seems to independentlycorrelate with increased atherosclerosis and systemicinflammation.


ACKNOWLEDGEMENTThe work was supported by grants from the SwedishMedical Research Council, the Swedish Heart and LungFoundation, the Medical Faculty of Lund University, SkaneUniversity Hospital, the Albert Pahlsson Research Founda-tion, the Crafoord Foundation, the Ernhold LundstromsResearch Foundation, the Region Skane, the Hulda andConrad Mossfelt Foundation, the King Gustaf V and QueenVictoria Foundation, the Wallenberg Foundation, and theLennart Hanssons Memorial Fund.

Conflicts of interestDr Engstrom is employed as senior epidemiologist byAstraZeneca R&D, Sweden. J.S. is employed at ResearchDepartment, BRAHMS AG, Hennigsdorf, Germany.

REFERENCES1. Rose KM, Eigenbrodt ML, Biga RL, Couper DJ, Light KC, Sharrett AR, et

al. Orthostatic hypotension predicts mortality in middle-aged adults:the Atherosclerosis Risk In Communities (ARIC) Study. Circulation2006; 114:630–636.

2. Fedorowski A, Stavenow L, Hedblad B, Berglund G, Nilsson PM,Melander O. Orthostatic hypotension predicts all-cause mortalityand coronary events in middle-aged individuals (The Malmo Preven-tive Project). Eur Heart J 2010; 31:85–91.

3. Fedorowski A, Hedblad B, Engstrom G, Gustav Smith J, Melander O.Orthostatic hypotension and long-term incidence of atrial fibrillation:the Malmo Preventive Project. J Intern Med 2010; 268:383–389.

4. Benvenuto LJ, Krakoff LR. Morbidity and mortality of orthostatichypotension: implications for management of cardiovascular disease.Am J Hypertens 2011; 24:135–144.

5. Fan XH, Wang Y, Sun K, Zhang W, Wang H, Wu H, et al. Disorders oforthostatic blood pressure response are associated with cardiovasculardisease and target organ damage in hypertensive patients. Am J Hyper-tens 2010; 23:829–837.

6. Robertson D. Orthostatic hypertension: the last hemodynamic frontier.Hypertension 2011; 57:158–159.

7. Rose KM, Tyroler HA, Nardo CJ, Arnett DK, Light KC, Rosamond W,et al. Orthostatic hypotension and the incidence of coronary heartdisease: the Atherosclerosis Risk in Communities study. Am J Hypertens2000; 13:571–578.

8. Mattace-Raso FU, van der Cammen TJ, Knetsch AM, van den MeirackerAH, Schalekamp MA, Hofman A, et al. Arterial stiffness as the candidateunderlying mechanism for postural blood pressure changes and ortho-static hypotension in older adults: the Rotterdam Study. J Hypertens2006; 24:339–344.

9. Fujisawa M, Okumiya K, Matsubayashi K, Hamada T, Endo H, Doi Y.Factors associated with carotid atherosclerosis in community-dwellingoldest elderly aged over 80 years. Geriatr Gerontol Int 2008; 8:12–18.

10. Nguyen-Thanh HT, Benzaquen BS. Screening for subclinical coronaryartery disease measuring carotid intima media thickness. Am J Cardiol2009; 104:1383–1388.

11. Lorenz MW, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction ofclinical cardiovascular events with carotid intima-media thickness: asystematic review and meta-analysis. Circulation 2007; 115:459–467.

12. Engstrom G, Melander O, Hedblad B. Carotid intima-media thickness,systemic inflammation, and incidence of heart failure hospitalizations.Arterioscler Thromb Vasc Biol 2009; 29:1691–1695.

13. Corrado E, Rizzo M, Coppola G, Fattouch K, Novo G, Marturana I, et al.An update on the role of markers of inflammation in atherosclerosis.J Atheroscler Thromb 2010; 17:1–11.

14. De Luca G, Verdoia M, Cassetti E, Schaffer A, Cavallino C, Bolzani V,et al. High fibrinogen level is an independent predictor of presence andextent of coronary artery disease among Italian population. J ThrombThrombolysis 2011; 31:458–463.

15. Berglund G, Elmstahl S, Janzon L, Larsson SA. The Malmo Diet andCancer Study. design and feasibility. J Intern Med 1993; 233:45–51.

16. Hedblad B, Nilsson P, Janzon L, Berglund G. Relation between insulinresistance and carotid intima-media thickness and stenosis in



nondiabetic subjects. Results from a cross-sectional study in Malmo,Sweden. Diabet Med 2000; 17:299–307.

17. Nilsson PM, Engstrom G, Hedblad B, Frystyk J, Persson MM, BerglundG, et al. Plasma adiponectin levels in relation to carotid intima mediathickness and markers of insulin resistance. Arterioscler Thromb VascBiol 2006; 26:2758–2762.

18. Consensus statement on the definition of orthostatic hypotension, pureautonomic failure, and multiple system atrophy. The Consensus Com-mittee of the American Autonomic Society and the American Academyof Neurology. Neurology 1996; 46:1470.

19. Fessel J, Robertson D. Orthostatic hypertension: when pressor reflexesovercompensate. Nat Clin Pract Nephrol 2006; 2:424–431.

20. Hedblad B, Wikstrand J, Janzon L, Wedel H, Berglund G. Low-dosemetoprolol CR/XL and fluvastatin slow progression of carotid intima-media thickness: main results from the Beta-Blocker Cholesterol-Lowering Asymptomatic Plaque Study (BCAPS). Circulation 2001;103:1721–1726.

21. Melander O, Newton-Cheh C, Almgren P, Hedblad B, Berglund G,Engstrom G, et al. Novel and conventional biomarkers for prediction ofincident cardiovascular events in the community. JAMA 2009; 302:49–57.

22. Rutan GH, Hermanson B, Bild DE, Kittner SJ, LaBaw F, Tell GS.Orthostatic hypotension in older adults. The Cardiovascular HealthStudy. CHS Collaborative Research Group. Hypertension 1992; 19:508–519.

23. Danesh J, Lewington S, Thompson SG, Lowe GD, Collins R, Kostis JB, etal. Plasma fibrinogen level and the risk of major cardiovascular dis-eases and nonvascular mortality: an individual participant meta-analysis. JAMA 2005; 294:1799–1809.

24. Masoud M, Sarig G, Brenner B, Jacob G. Orthostatic hypercoagulabil-ity: a novel physiological mechanism to activate the coagulationsystem. Hypertension 2008; 51:1545–1551.

25. Montalcini T, Gorgone G, Pujia A. Association between pulsepressure and subclinical carotid atherosclerosis in normotensive


and hypertensive postmenopausal women. Clin Exp Hypertens2009; 31:64–70.

26. Fedorowski A, Stavenow L, Hedblad B, Berglund G, Nilsson PM,Melander O. Consequences of orthostatic blood pressure variabilityin middle-aged men (The Malmo Preventive Project). J Hypertens 2010;28:551–559.

27. Okamoto LE, Gamboa A, Shibao C, Black BK, Diedrich A, Raj SR, et al.Nocturnal blood pressure dipping in the hypertension of autonomicfailure. Hypertension 2009; 53:363–369.

28. Shintani Y, Kikuya M, Hara A, Ohkubo T, Metoki H, Asayama K, et al.Ambulatory blood pressure, blood pressure variability and the preva-lence of carotid artery alteration: the Ohasama study. J Hypertens 2007;25:1704–1710.

29. Yatsuya H, Folsom AR, Alonso A, Gottesman RF, Rose KM. Posturalchanges in blood pressure and incidence of ischemic stroke subtypes:the ARIC study. Hypertension 2011; 57:167–173.

30. Lakatta EG. Age-associated cardiovascular changes in health: impacton cardiovascular disease in older persons. Heart Fail Rev 2002; 7:29–49.

31. Fedorowski A, Hedblad B, Engstrom G, Melander O. Directionality ofblood pressure response to standing may determine development ofheart failure: prospective cohort study. Eur J Heart Fail 2011; 13:496–503.

32. Park HJ, Baek SH, Jang SW, Kim DB, Shin DI, Shin WS, et al. Directcomparison of B-type natriuretic peptide and N-terminal pro-BNP forassessment of cardiac function in a large population of symptomaticpatients. Int J Cardiol 140:336–343.

33. O’Brien E, Asmar R, Beilin L, Imai Y, Mancia G, Mengden T, et al.Practice guidelines of the European Society of Hypertension for clinic,ambulatory and self blood pressure measurement. J Hypertens 2005;23:697–701.

34. Rothwell PM. Limitations of the usual blood-pressure hypothesis andimportance of variability, instability, and episodic hypertension. Lancet2010; 375:938–948.


Orthostatic blood pressure response, carotid intima–media thickness, and plasma fibrinogen in...

Documents

Transcript of Orthostatic blood pressure response, carotid intima–media thickness, and plasma fibrinogen in...