- Email: [email protected]
Arterial stiffness and orthostatic blood pressure changes in untreated and treated hypertensive subjects
Journal of the American Society of Hypertension 2(5) (2008) 372–377
Research Article
Arterial stiffness and orthostatic blood pressure changes in untreated and treated hypertensive subjects Athanase D. Protogerou, MD*, George S. Stergiou, MD, Panagiota Lourida, MD, and Apostolos Achimastos, MD Hypertension Center, 3rd University Department of Medicine, Sotiria Hospital, University of Athens, Athens, Greece Manuscript received December 20, 2007 and accepted March 26, 2008
Abstract Carotid-femoral pulse wave velocity (PWV), an integrated marker of segmental aortic stiffness, was recently proposed as one of the underlying mechanisms inducing orthostatic hypotension in the elderly with marked arterial rigidity. We examined the relationship between PWV (Complior; Colson, Paris, France) and orthostatic blood pressure (BP) changes, measured repeatedly, over a wide range of age and arterial stiffness. Sixty-nine hypertensive subjects (age, 37 to 76 years; 39 untreated and 30 treated) were studied. BP, in both sitting and erect position, was measured at two occasions a few weeks apart, and in between PWV was assessed by means of pulse wave analysis. In untreated hypertensive subjects, the orthostatic alterations in systolic, but not in diastolic blood pressure (DBP), were inversely related to PWV, independently from age, gender, mean BP, and diabetes mellitus. The greater the aortic stiffness the larger was the systolic blood pressure (SBP) decrease during upraises. On the contrary, no such association was found between PWV and orthostatic changes of BP in treated hypertensive subjects. These results suggest the presence of a pathophysiological association between arterial stiffening and BP postural changes. Antihypertensive drug treatment, as well as other factors that have not been evaluated in the present study, might have modulated this association. However, it might be argued that a causal association between arterial stiffness – disturbed baroreflex sensitivity – postural BP changes, even in subjects without pronounced vascular aging or orthostatic hypotension, is implied. J Am Soc Hypertens 2008;2(5): 372–377. © 2008 American Society of Hypertension. All rights reserved. Keywords: Aortic stiffness; pulse wave velocity; orthostatic hypotension; mean blood pressure.
Introduction Large artery (aortic) stiffness may be easily and reproducibly assessed by simple automatic devices that measure pulse wave velocity (PWV).1 Carotid-femoral PWV, an index of segmental aortic stiffness, is an independent predictor of overall and cardiovascular mortality in hypertensive subjects and also in the general population.2,3 The increase of PWV reflects not only the aging process but is also further accelerated by various cardiovascular risk factors including hypertension, diabetes mellitus, smoking, and hypercholesterolemia.4 Therefore it may be considered as an integrated marker of the arterial wall damage. Conflict of interest: none. *Corresponding author: Athanase D. Protogerou, MD, Hypertension Center, 3rd University Department of Medicine, 152 Mesogeion Av., Athens 115 27, Greece. Tel: 0030 210 7763117; fax: 0030 210 7719981. E-mail: [email protected]
It was recently suggested that aortic stiffness (carotidfemoral PWV) might be one of the underlying mechanisms inducing orthostatic hypotension in the elderly.5 Arterial wall stiffness in barosensitive regions (e.g., aorta) may reduce baroreceptor sensitivity by restricting their ability to stretch and relax in response to blood pressure (BP) changes.5–9 Data from repeated recordings of orthostatic changes of BP and their association with arterial stiffness in younger subjects are lacking. Because the association of arterial stiffness and baroreflex sensitivity has been described even in healthy subjects,7 we sought to determine the relation between aortic stiffness and postural BP changes in a wide range of age and PWV. Because antihypertensive drug treatment not only lowers peripheral resistance, and thereby brachial BP, but also affects large arteries’ function and structure,10 we analyzed untreated and treated hypertensive subjects separately, in order to investigate potential differences.
1933-1711/08/$ – see front matter © 2008 American Society of Hypertension. All rights reserved. doi:10.1016/j.jash.2008.03.011
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377 Table 1 Demographics of untreated and treated hypertensive subjects
Age (y) Gender (male) Weight (kg) Smokers (current) Diabetes mellitus Cardiovascular disease
Untreated Subjects (n ⫽ 39)
Treated Subjects (n ⫽ 30)
52.6 ⫾ 8.7 28 87.4 ⫾ 20.1 22 2 1
54.2 ⫾8.9 17 85.9 ⫾ 14.1 7 4 4
Data are n or mean ⫾ SD.
Methods Study Population and Study Protocol This is a retrospective study involving untreated and treated hypertensive subjects who were examined as outpatients in the hypertension clinic between January 2004 and December 2004. Inclusion criteria were: 1) at least two clinic visits (visit 1 and 2) for the assessment of office BP within a period of two months, without medication for untreated or with steady medication for at least one month for treated subjects, and 2) measurement of the arterial stiffness between visit 1 and 2. Subjects with cardiac arrhythmia were excluded from the study. Brachial BP measurements were performed by a mercury sphygmomanometer by three meticulously trained physicians who have been successfully tested for accuracy and agreement according to the international protocol.11 The setting of the BP evaluation was the day clinic of the Hypertension Center which is used for clinical research purposes.
Hemodynamic Parameters Triplicate brachial BP measurements were performed by mercury sphygmomanometer after a five-minute sitting rest,
373
using the first and the fifth Korotkoff sounds for the assessment of systolic blood pressure (SBP) and diastolic blood pressure (DBP), respectively, according to the international guidelines.12 The average of the second and third measurements was used for the analysis. Immediately after one minute in the erect position, SBP and DBP were measured. According to the consensus committee of the American Autonomic Society and the American Academy of Neurology, a reduction in SBP ⱖ 20 mm Hg and/or in DBP ⱖ 10 mm Hg, in any of the two visits, was used in order to define orthostatic hypotension.13 Carotid-femoral (aortic) PWV was measured in the supine position automatically using the Complior apparatus (Colson, Paris, France).1 Its determination is based on the simultaneous recording of the pulse wave in the common carotid and femoral arteries by two transducers. It is calculated as the distance separating the two transducers divided by the time delay between the onset (foot) of the two recorded waves. PWV was measured twice and the mean value was used for data analysis. The reproducibility of these measurements has been previously published.1
Statistical Analysis Bivariate linear correlation (Pearson’s correlation coefficient) was used in order to investigate the association between PWV and BP. Linear regression analysis was applied in order to study the independent effect of PWV on postural BP changes in the presence of confounding factors (age, gender, mean blood pressure [MBP], and pulse pressure [PP]). Bland-Altman plots were used in order to assess the reproducibility of SBP postural changes between visit 1 and 2. Statistical analysis was performed using SPSS 11.5 version software (SPSS Inc, Chicago, IL).
Table 2 Blood pressure parameters in seated and erect positions, orthostatic changes, and pulse wave velocity Untreated Subjects
sSBP (mm Hg) sDBP (mm Hg) sPP (mm Hg) eSBP (mm Hg) eDBP (mm Hg) ePP (mm Hg) orthSBP (mm Hg) orthDBP (mm Hg) orthPP (mm Hg) PWV (m/sec)*
Treated Subjects
Visit 1 (n ⫽ 39)
Visit 2 (n ⫽ 31)
Visit 1 (n ⫽ 30)
Visit 2 (n ⫽ 24)
154.3 ⫾ 7.5 99.5 ⫾ 4.7 54.8 ⫾ 7.3 151.7 ⫾ 17.6 99.6 ⫾ 9.8 52.1 ⫾ 17.2 ⫺2.6 ⫾ 9.1 0.3 ⫾ 5.5 ⫺2.7 ⫾ 10.2
147.5 ⫾ 16.4 98.3 ⫾ 13.8 51.1 ⫾ 15.4 146.4 ⫾ 18.1 97.0 ⫾ 8.8 49.3 ⫾ 16.0 ⫺0.7 ⫾ 7.7 ⫺1.8 ⫾ 10.2 ⫺0.7 ⫾ 7.5
158.2 ⫾ 18.1 100.1 ⫾ 10.4 58.1 ⫾ 15.7 154.2 ⫾ 19.4 98.1 ⫾ 12.2 56.1 ⫾ 18.0 ⫺4.1 ⫾ 6.7 ⫺2.1 ⫾ 6.7 ⫺1.9 ⫾ 6.6
159.8 ⫾ 18.9 99.7 ⫾ 11.1 60.1 ⫾ 16.4 155.8 ⫾ 20.8 98.1 ⫾ 12.8 57.7 ⫾ 18.,2 ⫺4.0 ⫾ 7.1 ⫺1.7 ⫾ 6.3 ⫺2.3 ⫾ 6.5
9.9 ⫾ 1.9
10.7 ⫾ 2.2
BP, blood pressure; DBP, diastolic blood pressure; orth, orthostatic changes, PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure; SD, standard deviation. Arterial blood pressure in the seated (s) and the erect (e) position, as well as orth of BP and PWV, in untreated and treated hypertensive subjects (mean values ⫾ SD). * Values of PWV correspond to n ⫽ 39 and n ⫽ 30, for untreated and treated hypertensive subjects, respectively.
374
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377 a
25
orthSBP difference between visits 1 & 2 (mmHg)
20 15 10 5 0 -5 -10 -15 -20 -25 -20
-15
-10
-5
0
5
10
15
20
Mean orthSBP of visit 1 & 2 (mmHg)
orthSBP difference between visits 1 & 2 (mmHg)
b
25 20 15 10 5 0 -5 -10 -15 -20 -25 -20
-15
-10
-5
0
5
10
15
20
Mean orthSBP of visit 1 & 2 (mmHg)
Figure 1. Bland-Altman plots of orthSBP (mm Hg) in (A) untreated and (B) treated hypertensive subjects between visit 1 and visit 2. The dotted lines in panel A represent the mean difference between orthSBP at visit 1 and visit 2 (mean, ⫺1.1 and 0.48 mm Hg, for untreated and treated subjects, respectively) and in panel B, ⫾ 2 SD (SD, 8.1 and 7.6 mm Hg, for untreated and treated subjects, respectively). orthSBP, orthostatic systolic blood pressure; SD, standard deviation.
Results Sixty-nine hypertensive subjects (age, 37 to 76 years; 39 untreated and 30 treated) fulfilled the inclusion criteria. Their characteristics are presented in Table 1. Eleven subjects were treated with beta-blockers, 14 with calcium channel blockers, 10 with thiazide diuretics, seven with angiotensin-converting enzyme (ACE) inhibitors, and seven with angiotensin receptor blockers (ARB). All quantitative parameters had a normal distribution. PWV (range, 5.2 to 17.8 m/sec) and BP parameters for both untreated and treated subjects, for visit 1 and 2, are presented in Table 2. Five (13%) untreated subjects and four (13%) treated subjects had orthostatic hypotension. In both untreated and treated subjects, the reproducibility of orthostatic SBP change between visits 1 and 2 was good as shown by Bland-Altman plots (Figure 1). The linear correlation between PWV and seated blood pressure (sBP), erect blood pressure (eBP), BP and orthostatic change (orth) of BP (orthBP; ⫽ sBP ⫺ eBP) in untreated hypertensive subjects is presented (for visit 1 and 2) in Table 3. PWV correlated with seated SBP (sSBP) and seated pulse pressure (sPP) but not with DBP. PWV did not correlate with eBP. On the other hand, PWV had an inverse association with orthSBP (Figure 2A and Table 3), and
Table 3 Linear correlation between pulse wave velocity and blood pressure, and orthostatic changes
sSBP sDBP sPP eSBP eDBP ePP orthSBP orthDBP orthPP
Untreated Subjects (Visit 1 and 2; n ⫽ 70)
Treated Subjects (Visit 1 and 2; n ⫽ 54)
0.300* ⫺0.105 0.381† 0.126 ⫺0.147 0.220 ⫺0.320† ⫺0.024 ⫺0.270*
0.204 ⫺0.238* 0.493† 0.245 ⫺0.273* 0.520† 0.154 ⫺0.068 0.235
BP, blood pressure; DBP, diastolic blood pressure; orth, orthostatic changes; PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure. Linear correlation (expressed by Pearson’s r correlation coefficient) between PWV and BP in seated (s) and erect (e) position, as well as orth, in treated and untreated hypertensive subjects. * P ⬍ .05. † P ⬍ .001.
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377
375
Figure 2. Linear correlation between PWV (m/sec) and (A) orthoSBP (mm Hg) (r ⫽ ⫺0.320; P ⬍ .001), and (B) orthPP (r ⫽ ⫺0.270; P ⬍ .005), in untreated hypertensive subjects at visits 1 and 2. orthPP, orthostatic pulse pressure; orthoSBP, orthostatic systolic blood pressure; PWV, pulse wave velocity.
orthPP (Figure 2B and Table 3), but not with orthDBP. This association was consistent even when visit 1 and visit 2 were analyzed separately (data not shown). In treated hypertensive subjects (Table 3), PWV had no association with sSBP, but it was correlated with sDBP and sPP. eBP and orthBP were not correlated with PWV. PWV was inversely correlated with orthSBP independently from age, gender, and MBP (Table 4). When PP was entered in the multivariate model, PWV was still the only independent predictor of orthSBP (data not shown). On the contrary, when PP was used instead of PWV in the multivariate model, PP was not an independent predictor of orthSBP. After exclusion of the subjects with diabetes mellitus, similar results were found.
Discussion In this study, we investigated the relationship between large artery stiffness and orth in BP in both untreated and
treated hypertensive subjects. We showed that in the untreated hypertensive subjects, the orthostatic changes found in SBP, but not in DBP, were inversely related to carotid-femoral PWV, independently from age, gender, MBP, and diabetes mellitus. The higher the aortic stiffness, the more SBP decreased during the uprise. On the contrary, we found no such association between PWV and the postural changes in BP in treated hypertensive subjects. These findings imply the presence of a pathophysiological association between arterial stiffening and BP postural changes, but it seems that this association is blurred in the presence of antihypertensive drug treatment. We measured aortic stiffness by means of carotidfemoral PWV, which is considered to be the “gold standard” for the assessment of segmental arterial stiffness.14,15 The distance traveled by the pulse wave was assessed directly by the line connecting the carotid artery to the femoral artery. This type of calculation is suggested to overestimate PWV by at least 2 meters per
376
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377
Table 4 Pulse wave velocity predicts postural changes in systolic blood pressure independently from age, gender, and mean blood pressure in untreated subjects Variables in the Model
P
Unstandardized Coefficient 
95% Confidence Intervals for 
PWV (m/sec) Age (y) Gender (female) MBP (mm Hg)
.037 .904 .962 .627
⫺1.228 0.029 ⫺0.118 0.092
⫺2.658 ⫺0.286 ⫺5.083 ⫺0.273
⫺0.088 0.265 4.845 0.166
MBP, mean blood pressure; ortho, orthostatic changes; PWV, pulse wave velocity; SBP, systolic blood pressure. PWV predicts postural changes in SBP independently from age, gender, and MBP, in untreated subjects (multivariate linear regression analysis; dependent variable, ortho in SBP).
second,5,15 introducing a systemic bias in our results. This type of systemic “error” is also present in the study of Mattace-Raso et al,5 which is the first large epidemiological study in 3,362 subjects (the Rotterdam Study) demonstrating that high arterial stiffness (PWV) is related with the presence of orthostatic hypotension, independently from age, gender, and MBP relation. Therefore the results of the present study may be directly compared with the results of the Rotterdam Study. In the present study, relatively young subjects have been included (mean age, 52 to 54 years vs. 71 to 73 years — all subjects ⬎55 years old — in the Rotterdam Study), and therefore, our sample had lower mean PWV (9.9 - 11.0 m/sec vs. 13.3 - 14 m/sec, respectively) than in the Rotterdam Study. Therefore, the studied population had lower prevalence of orthostatic hypotension (13% vs. almost 20% in the Rotterdam Study). Nevertheless, even in these younger untreated hypertensive subjects, increased aortic stiffness was consistently associated with a higher decrease of SBP during postural change, independently from age, gender, and MBP relation. Postural changes of BP were assessed on two occasions (a few weeks apart) with good repeatability as assessed by Bland-Altman plots (Figure 1), and PWV was assessed in a different setting in the interval between the two clinic visits. The presence of a consistent relationship of PWV with postural changes of SBP implies that this relationship might not be simply based on mathematical associations, and further reinforces the proposed theory suggesting a role of impaired baroreflex sensitivity due to, or associated with, increased vascular wall stiffness.5–9 Finally, in this study there was no relation between PWV and postural changes of SBP (and thus PP) in treated hypertensive subjects. Lowering of BP by drug treatment is known to act through the reduction of peripheral resistance and/or cardiac output, and may have direct or indirect effects on large artery stiffness and pressure wave reflections,10 but also on baroreceptor sensitivity.16,17 These effects appear to modulate the relation of PWV with SBP and DBP as depicted in Tables 3 and
4. In both untreated and treated hypertensive subjects, PWV is associated with PP. However, in untreated subjects this association appears to be mainly driven by a close relation of PWV with SBP, whereas in treated subjects the negative relation with DBP appears to be the dominant trait. Because potentially irreversible structural alterations in the aorta are more likely to explain arterial stiffening and thus the increase of PWV, it seems more reasonable to suggest that in the present study drug therapy may have directly affected baroreceptor function independently from structural changes. In the Rotterdam Study5 no distinction between treated and untreated subjects was made and no data on the relation of PWV with SBP or DBP are provided; because less than 25% of the participants were under drug treatment,5 it is possible that this differentiation between the two groups did not have a major impact on the overall result. The present data should be cautiously interpreted because other potential physiologic mechanisms which are associated to postural BP changes, such as inadequate venous return, decreased ventricular compliance, and stroke volume,18,19 were not assessed in the present study. However, as previously described,20 a common underlying pathway may be responsible for arterial and ventricular stiffening, contributing synergistically to postural BP changes.
Conclusion In conclusion, we showed that over a wide range of age, in untreated hypertensive subjects without pronounced arterial stiffening, increased PWV is related with a deterioration of the ability to maintain SBP during postural changes, independently from age, gender, MBP, and diabetes mellitus, even in subjects without orthostatic hypotension, implying a causal association, potentially via disturbed baroreflex sensitivity. Because this relation appears to be consistent in repeated measurements it might have potential clinical application in predicting the predisposition of untreated hypertensive subjects to develop orthostatic hypotension.
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377
References 1. Asmar R, Benetos A, Topouchian J, Laurent P, Pannier B, Brisac AM, et al. Assessment of arterial distensibility by automatic pulse wave velocity measurement: validation and clinical application studies. Hypertension 1995;26:485–90. 2. Blacher J, Asmar R, Djane S, London GM, Safar ME. Aortic pulse wave velocity as a marker of cardiovascular risk in hypertensive patients. Hypertension 1999;33: 1111–17. 3. Mattace-Raso FU, van der Cammen TJ, Hofman A, van People NM, Boes ML, Schlekamp MA, et al. Impaired stiffness and risk of coronary heart disease and stroke: the Rotterdam Study. Circulation 2006; 113:657– 63. 4. Blacher J, Protogerou AD, Safar ME. Cardiovascular risk and the macrocirculation. In: Safar ME, editor. Macro- and microcirculation in hypertension. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2005:83–97. 5. Mattace-Raso FUS, van der Cammen TJM, Knetsch AM, van den Meiracker AH, Schalekamp MADH, Hofman A, et al. Arterial stiffness as the candidate underlying mechanism for postural blood pressure changes and orthostatic hypotension in older adults: the Rotterdam Study. J Hypertens 2006;24:339 – 44. 6. Mattace-Raso FUS, van den Meiracker AH, Bos WJ, van der Cammen TJ, Westerhof BE, Elias-Smale S, et al. Arterial stiffness, cardiovagal baroreflex sensitivity and postural blood pressure changes in older adults: the Rotterdam Study. J Hypertens 2007;25:1421–26. 7. Monahan KD, Dinenno FA, Seals DR, Clevenger CM, Desouza CA, Tanaka H. Age-associated changes in cardiovagal baroreflex sensitivity are related to central arterial compliance. Am J Physiol Heart Circ Physiol 2001;281:H284 – 89. 8. Monahan KD, Tanaka H, Dinenno FA, Seals DR. Central arterial compliance is associated with age and habitual exercise-related differences in cardiovagal baroreflex sensitivity. Circulation 2001;104:1627–32. 9. Boddaert J, Tamim H, Verny M, Belmin J. Arterial stiffness is associated with orthostatic hypotension in elderly subjects with history of falls. J Am Geriatr Soc 2004;52:568 –72. 10. Blacher J, Protogerou AD, Safar ME. Large artery stiffness and antihypertensive agents. Curr Pharm Des 2005;11:3317–26.
377
11. O’Brien E, Pickering T, Asmar R, Myers M, Parati G, Staessen J, et al, Working Group on Blood Pressure Monitoring of the European Society of Hypertension. Working Group on Blood Pressure Monitoring of the European Society of Hypertension International Protocol for validation of blood pressure measuring devices in adults. Blood Press Monit 2002;7:3–17. 12. Mancia G, de Backer D, Dominiczak A, Cifkoya R, Fogord R, Germano G, et al. 2007 Guidelines for the Management of Arterial Hypertension. The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2007;25:1105– 87. 13. Consensus statement on the definition of orthostatic, pure autonomic failure, and multiple system atrophy. The Consensus Committee of the American Autonomic Society and the American Academy of Neurology. Neurology 1996;46:1470. 14. Laurent S, Cockcroft J, van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, et al, European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27:2588 – 605. 15. Van Bortel LM. Is arterial stiffness ready for daily clinical practice? J Hypertens 2006;24:281– 83. 16. Berdeaux A, Giudicelli JF. Antihypertensive drugs and baroreceptor reflex control of heart rate and blood pressure. Fundam Clin Pharmacol 1987;1:257– 82. 17. Carretta R, Fabris B, Bardelli M, Muiesan S, Fischetti F, Vran F, et al. Arterial compliance and baroreceptor sensitivity after chronic treatment with indapamide. J Hum Hypertens 1998;2:171–75. 18. Casadei B, Meyer TE, Coats AJ, Conway J, Sleight P. Baroreflex control of stroke volume in man: an effect mediated by the vagus. J Physiol 1992;448: 539 –50. 19. Caird FI, Andrews GR, Kennedy RD. Effect of posture on blood pressure in the elderly. Br Heart J 1973;35:527–30. 20. Ikonomidis I, Aznaouridis K, Protogerou A, Stamatelopoulos K, Markomihelakis N, Papamichael C, et al. Arterial wave reflections are associated with left ventricular diastolic dysfunction in Adamantiades-Behçet’s disease. J Card Fail 2006;12:458 – 63.
Research Article
Arterial stiffness and orthostatic blood pressure changes in untreated and treated hypertensive subjects Athanase D. Protogerou, MD*, George S. Stergiou, MD, Panagiota Lourida, MD, and Apostolos Achimastos, MD Hypertension Center, 3rd University Department of Medicine, Sotiria Hospital, University of Athens, Athens, Greece Manuscript received December 20, 2007 and accepted March 26, 2008
Abstract Carotid-femoral pulse wave velocity (PWV), an integrated marker of segmental aortic stiffness, was recently proposed as one of the underlying mechanisms inducing orthostatic hypotension in the elderly with marked arterial rigidity. We examined the relationship between PWV (Complior; Colson, Paris, France) and orthostatic blood pressure (BP) changes, measured repeatedly, over a wide range of age and arterial stiffness. Sixty-nine hypertensive subjects (age, 37 to 76 years; 39 untreated and 30 treated) were studied. BP, in both sitting and erect position, was measured at two occasions a few weeks apart, and in between PWV was assessed by means of pulse wave analysis. In untreated hypertensive subjects, the orthostatic alterations in systolic, but not in diastolic blood pressure (DBP), were inversely related to PWV, independently from age, gender, mean BP, and diabetes mellitus. The greater the aortic stiffness the larger was the systolic blood pressure (SBP) decrease during upraises. On the contrary, no such association was found between PWV and orthostatic changes of BP in treated hypertensive subjects. These results suggest the presence of a pathophysiological association between arterial stiffening and BP postural changes. Antihypertensive drug treatment, as well as other factors that have not been evaluated in the present study, might have modulated this association. However, it might be argued that a causal association between arterial stiffness – disturbed baroreflex sensitivity – postural BP changes, even in subjects without pronounced vascular aging or orthostatic hypotension, is implied. J Am Soc Hypertens 2008;2(5): 372–377. © 2008 American Society of Hypertension. All rights reserved. Keywords: Aortic stiffness; pulse wave velocity; orthostatic hypotension; mean blood pressure.
Introduction Large artery (aortic) stiffness may be easily and reproducibly assessed by simple automatic devices that measure pulse wave velocity (PWV).1 Carotid-femoral PWV, an index of segmental aortic stiffness, is an independent predictor of overall and cardiovascular mortality in hypertensive subjects and also in the general population.2,3 The increase of PWV reflects not only the aging process but is also further accelerated by various cardiovascular risk factors including hypertension, diabetes mellitus, smoking, and hypercholesterolemia.4 Therefore it may be considered as an integrated marker of the arterial wall damage. Conflict of interest: none. *Corresponding author: Athanase D. Protogerou, MD, Hypertension Center, 3rd University Department of Medicine, 152 Mesogeion Av., Athens 115 27, Greece. Tel: 0030 210 7763117; fax: 0030 210 7719981. E-mail: [email protected]
It was recently suggested that aortic stiffness (carotidfemoral PWV) might be one of the underlying mechanisms inducing orthostatic hypotension in the elderly.5 Arterial wall stiffness in barosensitive regions (e.g., aorta) may reduce baroreceptor sensitivity by restricting their ability to stretch and relax in response to blood pressure (BP) changes.5–9 Data from repeated recordings of orthostatic changes of BP and their association with arterial stiffness in younger subjects are lacking. Because the association of arterial stiffness and baroreflex sensitivity has been described even in healthy subjects,7 we sought to determine the relation between aortic stiffness and postural BP changes in a wide range of age and PWV. Because antihypertensive drug treatment not only lowers peripheral resistance, and thereby brachial BP, but also affects large arteries’ function and structure,10 we analyzed untreated and treated hypertensive subjects separately, in order to investigate potential differences.
1933-1711/08/$ – see front matter © 2008 American Society of Hypertension. All rights reserved. doi:10.1016/j.jash.2008.03.011
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377 Table 1 Demographics of untreated and treated hypertensive subjects
Age (y) Gender (male) Weight (kg) Smokers (current) Diabetes mellitus Cardiovascular disease
Untreated Subjects (n ⫽ 39)
Treated Subjects (n ⫽ 30)
52.6 ⫾ 8.7 28 87.4 ⫾ 20.1 22 2 1
54.2 ⫾8.9 17 85.9 ⫾ 14.1 7 4 4
Data are n or mean ⫾ SD.
Methods Study Population and Study Protocol This is a retrospective study involving untreated and treated hypertensive subjects who were examined as outpatients in the hypertension clinic between January 2004 and December 2004. Inclusion criteria were: 1) at least two clinic visits (visit 1 and 2) for the assessment of office BP within a period of two months, without medication for untreated or with steady medication for at least one month for treated subjects, and 2) measurement of the arterial stiffness between visit 1 and 2. Subjects with cardiac arrhythmia were excluded from the study. Brachial BP measurements were performed by a mercury sphygmomanometer by three meticulously trained physicians who have been successfully tested for accuracy and agreement according to the international protocol.11 The setting of the BP evaluation was the day clinic of the Hypertension Center which is used for clinical research purposes.
Hemodynamic Parameters Triplicate brachial BP measurements were performed by mercury sphygmomanometer after a five-minute sitting rest,
373
using the first and the fifth Korotkoff sounds for the assessment of systolic blood pressure (SBP) and diastolic blood pressure (DBP), respectively, according to the international guidelines.12 The average of the second and third measurements was used for the analysis. Immediately after one minute in the erect position, SBP and DBP were measured. According to the consensus committee of the American Autonomic Society and the American Academy of Neurology, a reduction in SBP ⱖ 20 mm Hg and/or in DBP ⱖ 10 mm Hg, in any of the two visits, was used in order to define orthostatic hypotension.13 Carotid-femoral (aortic) PWV was measured in the supine position automatically using the Complior apparatus (Colson, Paris, France).1 Its determination is based on the simultaneous recording of the pulse wave in the common carotid and femoral arteries by two transducers. It is calculated as the distance separating the two transducers divided by the time delay between the onset (foot) of the two recorded waves. PWV was measured twice and the mean value was used for data analysis. The reproducibility of these measurements has been previously published.1
Statistical Analysis Bivariate linear correlation (Pearson’s correlation coefficient) was used in order to investigate the association between PWV and BP. Linear regression analysis was applied in order to study the independent effect of PWV on postural BP changes in the presence of confounding factors (age, gender, mean blood pressure [MBP], and pulse pressure [PP]). Bland-Altman plots were used in order to assess the reproducibility of SBP postural changes between visit 1 and 2. Statistical analysis was performed using SPSS 11.5 version software (SPSS Inc, Chicago, IL).
Table 2 Blood pressure parameters in seated and erect positions, orthostatic changes, and pulse wave velocity Untreated Subjects
sSBP (mm Hg) sDBP (mm Hg) sPP (mm Hg) eSBP (mm Hg) eDBP (mm Hg) ePP (mm Hg) orthSBP (mm Hg) orthDBP (mm Hg) orthPP (mm Hg) PWV (m/sec)*
Treated Subjects
Visit 1 (n ⫽ 39)
Visit 2 (n ⫽ 31)
Visit 1 (n ⫽ 30)
Visit 2 (n ⫽ 24)
154.3 ⫾ 7.5 99.5 ⫾ 4.7 54.8 ⫾ 7.3 151.7 ⫾ 17.6 99.6 ⫾ 9.8 52.1 ⫾ 17.2 ⫺2.6 ⫾ 9.1 0.3 ⫾ 5.5 ⫺2.7 ⫾ 10.2
147.5 ⫾ 16.4 98.3 ⫾ 13.8 51.1 ⫾ 15.4 146.4 ⫾ 18.1 97.0 ⫾ 8.8 49.3 ⫾ 16.0 ⫺0.7 ⫾ 7.7 ⫺1.8 ⫾ 10.2 ⫺0.7 ⫾ 7.5
158.2 ⫾ 18.1 100.1 ⫾ 10.4 58.1 ⫾ 15.7 154.2 ⫾ 19.4 98.1 ⫾ 12.2 56.1 ⫾ 18.0 ⫺4.1 ⫾ 6.7 ⫺2.1 ⫾ 6.7 ⫺1.9 ⫾ 6.6
159.8 ⫾ 18.9 99.7 ⫾ 11.1 60.1 ⫾ 16.4 155.8 ⫾ 20.8 98.1 ⫾ 12.8 57.7 ⫾ 18.,2 ⫺4.0 ⫾ 7.1 ⫺1.7 ⫾ 6.3 ⫺2.3 ⫾ 6.5
9.9 ⫾ 1.9
10.7 ⫾ 2.2
BP, blood pressure; DBP, diastolic blood pressure; orth, orthostatic changes, PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure; SD, standard deviation. Arterial blood pressure in the seated (s) and the erect (e) position, as well as orth of BP and PWV, in untreated and treated hypertensive subjects (mean values ⫾ SD). * Values of PWV correspond to n ⫽ 39 and n ⫽ 30, for untreated and treated hypertensive subjects, respectively.
374
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377 a
25
orthSBP difference between visits 1 & 2 (mmHg)
20 15 10 5 0 -5 -10 -15 -20 -25 -20
-15
-10
-5
0
5
10
15
20
Mean orthSBP of visit 1 & 2 (mmHg)
orthSBP difference between visits 1 & 2 (mmHg)
b
25 20 15 10 5 0 -5 -10 -15 -20 -25 -20
-15
-10
-5
0
5
10
15
20
Mean orthSBP of visit 1 & 2 (mmHg)
Figure 1. Bland-Altman plots of orthSBP (mm Hg) in (A) untreated and (B) treated hypertensive subjects between visit 1 and visit 2. The dotted lines in panel A represent the mean difference between orthSBP at visit 1 and visit 2 (mean, ⫺1.1 and 0.48 mm Hg, for untreated and treated subjects, respectively) and in panel B, ⫾ 2 SD (SD, 8.1 and 7.6 mm Hg, for untreated and treated subjects, respectively). orthSBP, orthostatic systolic blood pressure; SD, standard deviation.
Results Sixty-nine hypertensive subjects (age, 37 to 76 years; 39 untreated and 30 treated) fulfilled the inclusion criteria. Their characteristics are presented in Table 1. Eleven subjects were treated with beta-blockers, 14 with calcium channel blockers, 10 with thiazide diuretics, seven with angiotensin-converting enzyme (ACE) inhibitors, and seven with angiotensin receptor blockers (ARB). All quantitative parameters had a normal distribution. PWV (range, 5.2 to 17.8 m/sec) and BP parameters for both untreated and treated subjects, for visit 1 and 2, are presented in Table 2. Five (13%) untreated subjects and four (13%) treated subjects had orthostatic hypotension. In both untreated and treated subjects, the reproducibility of orthostatic SBP change between visits 1 and 2 was good as shown by Bland-Altman plots (Figure 1). The linear correlation between PWV and seated blood pressure (sBP), erect blood pressure (eBP), BP and orthostatic change (orth) of BP (orthBP; ⫽ sBP ⫺ eBP) in untreated hypertensive subjects is presented (for visit 1 and 2) in Table 3. PWV correlated with seated SBP (sSBP) and seated pulse pressure (sPP) but not with DBP. PWV did not correlate with eBP. On the other hand, PWV had an inverse association with orthSBP (Figure 2A and Table 3), and
Table 3 Linear correlation between pulse wave velocity and blood pressure, and orthostatic changes
sSBP sDBP sPP eSBP eDBP ePP orthSBP orthDBP orthPP
Untreated Subjects (Visit 1 and 2; n ⫽ 70)
Treated Subjects (Visit 1 and 2; n ⫽ 54)
0.300* ⫺0.105 0.381† 0.126 ⫺0.147 0.220 ⫺0.320† ⫺0.024 ⫺0.270*
0.204 ⫺0.238* 0.493† 0.245 ⫺0.273* 0.520† 0.154 ⫺0.068 0.235
BP, blood pressure; DBP, diastolic blood pressure; orth, orthostatic changes; PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure. Linear correlation (expressed by Pearson’s r correlation coefficient) between PWV and BP in seated (s) and erect (e) position, as well as orth, in treated and untreated hypertensive subjects. * P ⬍ .05. † P ⬍ .001.
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377
375
Figure 2. Linear correlation between PWV (m/sec) and (A) orthoSBP (mm Hg) (r ⫽ ⫺0.320; P ⬍ .001), and (B) orthPP (r ⫽ ⫺0.270; P ⬍ .005), in untreated hypertensive subjects at visits 1 and 2. orthPP, orthostatic pulse pressure; orthoSBP, orthostatic systolic blood pressure; PWV, pulse wave velocity.
orthPP (Figure 2B and Table 3), but not with orthDBP. This association was consistent even when visit 1 and visit 2 were analyzed separately (data not shown). In treated hypertensive subjects (Table 3), PWV had no association with sSBP, but it was correlated with sDBP and sPP. eBP and orthBP were not correlated with PWV. PWV was inversely correlated with orthSBP independently from age, gender, and MBP (Table 4). When PP was entered in the multivariate model, PWV was still the only independent predictor of orthSBP (data not shown). On the contrary, when PP was used instead of PWV in the multivariate model, PP was not an independent predictor of orthSBP. After exclusion of the subjects with diabetes mellitus, similar results were found.
Discussion In this study, we investigated the relationship between large artery stiffness and orth in BP in both untreated and
treated hypertensive subjects. We showed that in the untreated hypertensive subjects, the orthostatic changes found in SBP, but not in DBP, were inversely related to carotid-femoral PWV, independently from age, gender, MBP, and diabetes mellitus. The higher the aortic stiffness, the more SBP decreased during the uprise. On the contrary, we found no such association between PWV and the postural changes in BP in treated hypertensive subjects. These findings imply the presence of a pathophysiological association between arterial stiffening and BP postural changes, but it seems that this association is blurred in the presence of antihypertensive drug treatment. We measured aortic stiffness by means of carotidfemoral PWV, which is considered to be the “gold standard” for the assessment of segmental arterial stiffness.14,15 The distance traveled by the pulse wave was assessed directly by the line connecting the carotid artery to the femoral artery. This type of calculation is suggested to overestimate PWV by at least 2 meters per
376
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377
Table 4 Pulse wave velocity predicts postural changes in systolic blood pressure independently from age, gender, and mean blood pressure in untreated subjects Variables in the Model
P
Unstandardized Coefficient 
95% Confidence Intervals for 
PWV (m/sec) Age (y) Gender (female) MBP (mm Hg)
.037 .904 .962 .627
⫺1.228 0.029 ⫺0.118 0.092
⫺2.658 ⫺0.286 ⫺5.083 ⫺0.273
⫺0.088 0.265 4.845 0.166
MBP, mean blood pressure; ortho, orthostatic changes; PWV, pulse wave velocity; SBP, systolic blood pressure. PWV predicts postural changes in SBP independently from age, gender, and MBP, in untreated subjects (multivariate linear regression analysis; dependent variable, ortho in SBP).
second,5,15 introducing a systemic bias in our results. This type of systemic “error” is also present in the study of Mattace-Raso et al,5 which is the first large epidemiological study in 3,362 subjects (the Rotterdam Study) demonstrating that high arterial stiffness (PWV) is related with the presence of orthostatic hypotension, independently from age, gender, and MBP relation. Therefore the results of the present study may be directly compared with the results of the Rotterdam Study. In the present study, relatively young subjects have been included (mean age, 52 to 54 years vs. 71 to 73 years — all subjects ⬎55 years old — in the Rotterdam Study), and therefore, our sample had lower mean PWV (9.9 - 11.0 m/sec vs. 13.3 - 14 m/sec, respectively) than in the Rotterdam Study. Therefore, the studied population had lower prevalence of orthostatic hypotension (13% vs. almost 20% in the Rotterdam Study). Nevertheless, even in these younger untreated hypertensive subjects, increased aortic stiffness was consistently associated with a higher decrease of SBP during postural change, independently from age, gender, and MBP relation. Postural changes of BP were assessed on two occasions (a few weeks apart) with good repeatability as assessed by Bland-Altman plots (Figure 1), and PWV was assessed in a different setting in the interval between the two clinic visits. The presence of a consistent relationship of PWV with postural changes of SBP implies that this relationship might not be simply based on mathematical associations, and further reinforces the proposed theory suggesting a role of impaired baroreflex sensitivity due to, or associated with, increased vascular wall stiffness.5–9 Finally, in this study there was no relation between PWV and postural changes of SBP (and thus PP) in treated hypertensive subjects. Lowering of BP by drug treatment is known to act through the reduction of peripheral resistance and/or cardiac output, and may have direct or indirect effects on large artery stiffness and pressure wave reflections,10 but also on baroreceptor sensitivity.16,17 These effects appear to modulate the relation of PWV with SBP and DBP as depicted in Tables 3 and
4. In both untreated and treated hypertensive subjects, PWV is associated with PP. However, in untreated subjects this association appears to be mainly driven by a close relation of PWV with SBP, whereas in treated subjects the negative relation with DBP appears to be the dominant trait. Because potentially irreversible structural alterations in the aorta are more likely to explain arterial stiffening and thus the increase of PWV, it seems more reasonable to suggest that in the present study drug therapy may have directly affected baroreceptor function independently from structural changes. In the Rotterdam Study5 no distinction between treated and untreated subjects was made and no data on the relation of PWV with SBP or DBP are provided; because less than 25% of the participants were under drug treatment,5 it is possible that this differentiation between the two groups did not have a major impact on the overall result. The present data should be cautiously interpreted because other potential physiologic mechanisms which are associated to postural BP changes, such as inadequate venous return, decreased ventricular compliance, and stroke volume,18,19 were not assessed in the present study. However, as previously described,20 a common underlying pathway may be responsible for arterial and ventricular stiffening, contributing synergistically to postural BP changes.
Conclusion In conclusion, we showed that over a wide range of age, in untreated hypertensive subjects without pronounced arterial stiffening, increased PWV is related with a deterioration of the ability to maintain SBP during postural changes, independently from age, gender, MBP, and diabetes mellitus, even in subjects without orthostatic hypotension, implying a causal association, potentially via disturbed baroreflex sensitivity. Because this relation appears to be consistent in repeated measurements it might have potential clinical application in predicting the predisposition of untreated hypertensive subjects to develop orthostatic hypotension.
A.D. Protogerou et al. / Journal of the American Society of Hypertension 2(5) (2008) 372–377
References 1. Asmar R, Benetos A, Topouchian J, Laurent P, Pannier B, Brisac AM, et al. Assessment of arterial distensibility by automatic pulse wave velocity measurement: validation and clinical application studies. Hypertension 1995;26:485–90. 2. Blacher J, Asmar R, Djane S, London GM, Safar ME. Aortic pulse wave velocity as a marker of cardiovascular risk in hypertensive patients. Hypertension 1999;33: 1111–17. 3. Mattace-Raso FU, van der Cammen TJ, Hofman A, van People NM, Boes ML, Schlekamp MA, et al. Impaired stiffness and risk of coronary heart disease and stroke: the Rotterdam Study. Circulation 2006; 113:657– 63. 4. Blacher J, Protogerou AD, Safar ME. Cardiovascular risk and the macrocirculation. In: Safar ME, editor. Macro- and microcirculation in hypertension. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins, 2005:83–97. 5. Mattace-Raso FUS, van der Cammen TJM, Knetsch AM, van den Meiracker AH, Schalekamp MADH, Hofman A, et al. Arterial stiffness as the candidate underlying mechanism for postural blood pressure changes and orthostatic hypotension in older adults: the Rotterdam Study. J Hypertens 2006;24:339 – 44. 6. Mattace-Raso FUS, van den Meiracker AH, Bos WJ, van der Cammen TJ, Westerhof BE, Elias-Smale S, et al. Arterial stiffness, cardiovagal baroreflex sensitivity and postural blood pressure changes in older adults: the Rotterdam Study. J Hypertens 2007;25:1421–26. 7. Monahan KD, Dinenno FA, Seals DR, Clevenger CM, Desouza CA, Tanaka H. Age-associated changes in cardiovagal baroreflex sensitivity are related to central arterial compliance. Am J Physiol Heart Circ Physiol 2001;281:H284 – 89. 8. Monahan KD, Tanaka H, Dinenno FA, Seals DR. Central arterial compliance is associated with age and habitual exercise-related differences in cardiovagal baroreflex sensitivity. Circulation 2001;104:1627–32. 9. Boddaert J, Tamim H, Verny M, Belmin J. Arterial stiffness is associated with orthostatic hypotension in elderly subjects with history of falls. J Am Geriatr Soc 2004;52:568 –72. 10. Blacher J, Protogerou AD, Safar ME. Large artery stiffness and antihypertensive agents. Curr Pharm Des 2005;11:3317–26.
377
11. O’Brien E, Pickering T, Asmar R, Myers M, Parati G, Staessen J, et al, Working Group on Blood Pressure Monitoring of the European Society of Hypertension. Working Group on Blood Pressure Monitoring of the European Society of Hypertension International Protocol for validation of blood pressure measuring devices in adults. Blood Press Monit 2002;7:3–17. 12. Mancia G, de Backer D, Dominiczak A, Cifkoya R, Fogord R, Germano G, et al. 2007 Guidelines for the Management of Arterial Hypertension. The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2007;25:1105– 87. 13. Consensus statement on the definition of orthostatic, pure autonomic failure, and multiple system atrophy. The Consensus Committee of the American Autonomic Society and the American Academy of Neurology. Neurology 1996;46:1470. 14. Laurent S, Cockcroft J, van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, et al, European Network for Non-invasive Investigation of Large Arteries. Expert consensus document on arterial stiffness: methodological issues and clinical applications. Eur Heart J 2006; 27:2588 – 605. 15. Van Bortel LM. Is arterial stiffness ready for daily clinical practice? J Hypertens 2006;24:281– 83. 16. Berdeaux A, Giudicelli JF. Antihypertensive drugs and baroreceptor reflex control of heart rate and blood pressure. Fundam Clin Pharmacol 1987;1:257– 82. 17. Carretta R, Fabris B, Bardelli M, Muiesan S, Fischetti F, Vran F, et al. Arterial compliance and baroreceptor sensitivity after chronic treatment with indapamide. J Hum Hypertens 1998;2:171–75. 18. Casadei B, Meyer TE, Coats AJ, Conway J, Sleight P. Baroreflex control of stroke volume in man: an effect mediated by the vagus. J Physiol 1992;448: 539 –50. 19. Caird FI, Andrews GR, Kennedy RD. Effect of posture on blood pressure in the elderly. Br Heart J 1973;35:527–30. 20. Ikonomidis I, Aznaouridis K, Protogerou A, Stamatelopoulos K, Markomihelakis N, Papamichael C, et al. Arterial wave reflections are associated with left ventricular diastolic dysfunction in Adamantiades-Behçet’s disease. J Card Fail 2006;12:458 – 63.