SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

J Clin Hypertens (Greenwich). 2012;14:773–778. ©2012 Wiley Periodicals, Inc.

Postmenopausal women are at greater risk for hypertension-related cardiovascular disease. Antihypertensive therapy may help alleviate arterial stiffness that represents a potential modifiable risk factor of hypertension. This randomized controlled study investigated the difference between an angiotensin receptor blocker and a calcium channel blocker in reducing arterial stiffness. Overall, 125 postmenopausal hypertensive women (age, 61.4±6 years; systolic blood pressure/diastolic blood pressure [SBP/DBP], 158±11/92±9 mm Hg) were randomized to valsartan 320 mg±hydrochlorothiazide (HCTZ) (n=63) or amlodipine 10 mg±HCTZ (n=62). The primary outcome was carotid-to-femoral pulse wave velocity (PWV) changes after 38 weeks of treatment. Both treatments lowered peripheral blood pressure (BP) (−22.9/−10.9 mm Hg for valsartan and −25.2/−11.7 mm Hg for amlodipine, P=not significant) and central BP (−15.7/−7.6 mm Hg for valsartan and −19.2/−10.3 mm Hg for amlodipine, P<.05 for central DBP). Both treatments similarly reduced the carotid-femoral PWV (−1.9 vs −1.7 m/s; P=not significant). Amlodipine was associated with a higher incidence of peripheral edema compared with the valsartan group (77% vs 14%, P<.001). BP lowering in postmenopausal women led to a reduction in arterial stiffness as assessed by PWV measurement. Both regimens reduced PWV to a similar degree after 38 weeks of treatment despite differences in central BP lowering, suggesting that the effect of valsartan on PWV is mediated through nonhemodynamic effects.

Cardiovascular disease is the most frequent cause of death in women, with the risk rising exponentially with age.1,2 Postmenopausal women with hypertension significantly outnumber hypertensive men and show a steep age-related increase in arterial stiffness3–5 leading to increases in systolic blood pressure (SBP) and left ventricular hypertrophy.6,7 Arterial stiffness measured by pulse wave velocity (PWV) is an independent risk factor for cardiovascular events and is a strong predictor of cardiovascular disease–associated mortality in hypertensive patients.8–10 Drugs that have the potential to reduce arterial stiffness in addition to lowering blood pressure (BP) can effectively reduce the risk of cardiovascular outcomes in postmenopausal women.

Angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) are both effective antihypertensive agents with similar effects on the renin-angiotensin system (RAS) and both have potential in reducing arterial stiffness in hypertensive patients. Several clinical studies have demonstrated that both ACE inhibitors and ARBs increase arterial compliance and distensibility, thereby reducing arterial stiffness even independent of BP lowering.11–13 The present study assessed the differential effects of valsartan (RAS antagonist) vs amlodipine (a calcium channel blocker [CCB] or non-RAS antagonist) on PWV changes as a marker of large artery stiffness in postmenopausal women. Data suggest that CCBs and, more specifically, amlodipine treatment, increase peripheral sympathetic basal tone as well as RAS activity.14 Thus, the purpose of this study was to compare the effects of two antihypertensive agents with similar BP reductions on changes in PWV. To our knowledge, this study is the first to compare the efficacy of a RAS vs a non-RAS antihypertensive agent on arterial stiffness in postmenopausal women with hypertension.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

Study Design

This was a 42-week, single-center, randomized, controlled, double-blind study. Groups were stratified by baseline statin use, as statins have been shown to improve endothelial function. The present study comprised (1) a 4-week screening period (a 2-week washout period and a 2-week placebo run-in phase), and (2) a 38-week double-blind treatment period. All eligible patients were randomized (1:1) to receive valsartan 160 mg or amlodipine 5 mg. After 4 weeks of treatment, patients were force-titrated to valsartan 320 mg and amlodipine 10 mg until the end of the study. From week 12 onwards, open-label HCTZ 12.5 mg was added to each treatment arm if patients did not achieve target BP of <140/90 mm Hg.

Patients

Postmenopausal women (aged 50–75 years) with mild to moderate hypertension (SBP ≥140 mm Hg, DBP <110 mm Hg, and pulse pressure [PP] ≥50 mm Hg) were eligible for enrollment. Patients were recruited by primary care physicians and addressed for exhaustive inclusion assessment at our angiology department. Exclusion criteria included BP above the safety limit of SBP ≥180 mm Hg and/or DBP ≥110 mm Hg before or at any point during the study. Similarly, patients were discontinued from the study if SBP was <100 mm Hg.

Patients with low-density lipoprotein >4.1 mM (>164 mg/dL) who were not taking any antihyperlipidemic medication and patients with a history or current use of oral or topical hormone replacement therapy at the entry of study were excluded. Patients with a history of type 1 or 2 diabetes, Raynaud disease, atrial fibrillation or other arrhythmia, evidence of secondary form of hypertension, cerebrovascular accident, transient ischemic cerebral attack or myocardial infarction, congestive heart failure, clinically significant valvular heart disease, history of malignancy including leukemia and lymphoma, or with life-threatening disease were also excluded. Patients were also excluded if they had known hypersensitivity or contraindications to valsartan, other ARBs, thiazide diuretics, amlodipine or other CCBs, and glycerin trinitrite.

The study was performed according to the good clinical practice guidelines and in compliance with the Declaration of Helsinki (2002) of the World Medical Association. The study was initiated and designed by the academic principal investigator (DH). Financial support was provided by Novartis Pharmaceuticals. The study was approved by the appropriate institutional review board and all patients provided written informed consent.

Efficacy Outcomes

The primary efficacy outcome was change in carotid-femoral PWV from baseline (week 1) to study endpoint (week 38). Secondary efficacy outcomes were changes from baseline in peripheral and central BP at weeks 12 and 38.

Safety Outcome

Safety outcomes were defined as any clinical adverse events (AEs) or any laboratory values falling outside the predetermined ranges. AEs were assessed at each clinical visit, and laboratory examination was performed at weeks 1, 12, and 38.

Measurements

All measurements and procedures were performed in fasting patients. BP was measured using a standard sphygmomanometer with the appropriate cuff size in accordance with the American Heart Association Committee Report on BP determination.15 All BPs were measured 3 times at 1-minute intervals while the patient was sitting for a minimum of 5 minutes. PP was determined as the difference between mean SBP and DBP values. Carotid-to-femoral PWV was determined from transcutaneous Doppler flow recordings and the foot-to-foot method triggered by simultaneous electrocardiography (ECG).16 Two simultaneous Doppler flow tracings were taken at the left common carotid and the right femoral artery in the groin with a linear array probe (L12-5 MHz-ATL Ultrasound, Inc, Bothell, WA) using HDILab software.17 The time delay (t) was measured between R wave of the ECG and the base of the flow waves recorded at these different points and averaged over 10 beats. The distance (D) traveled by the pulse wave was measured over body surface as the distance between two recording sites and, when measured from the carotid artery, the distance from the suprasternal notch to the carotid artery was subtracted. PWV was calculated as PWV=D/t.

Central BP was measured via applanation tonometry recordings of the common carotid artery18 and calibration was performed according to the Kelly and Fitchett methods.19

Statistical Analysis

A sample size of 78 patients per treatment group was required based on the primary variable change from baseline and assuming a dropout rate of 15%. This would ensure 90% power at the 5% level (2-sided) to detect a statistically significant difference in PWV between valsartan 320 mg±HCTZ and amlodipine 10 mg±HCTZ of 0.4 m/s (equivalent to 5 years of vascular aging). All statistical analyses were performed with SAS version 8.2 (SAS Institute Inc, Cary, NC) using 2-sided tests at a .05 significance level on the intent-to-treat (ITT) population and 95% confidence intervals (CIs) were calculated for treatment differences. Groups were compared for demographic and baseline characteristics by descriptive statistics. The change in primary and secondary outcomes between valsartan and amlodipine from baseline to week 12 and week 38 were analyzed using an analysis of covariance model with age and baseline central SBP as covariates. The least-squares mean changes from baseline, treatment difference, 95% CI for the treatment difference, and P value were determined.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

Patient Demographics and Baseline Characteristics

Of the 164 patients who entered the screening phase, 125 were randomized to receive valsartan 320mg±HCTZ 12.5 mg (n=63) or amlodipine 10 mg±HCTZ 12.5 mg (n=62). Patient demographics were similar between the treatment groups, and are reported in Table I. Patients in the valsartan group were slightly older (62.3±5.8 years vs 60.4±5.1 years; P=not significant) than in the amlodipine group, with 35% of women older than 65 years in the valsartan group vs 19% in the amlodipine group. More patients were previously treated in the valsartan group (87.3% vs 75.8%; P<.05). The mean body mass index of the randomized population was 27.3 kg/m2. Twenty-seven patients (43%) in the valsartan group and 13 patients (21%) in the amlodipine group required add-on HCTZ therapy at week 12, 20, or 28. The majority of patients (103 [82.4%]) completed the study. The reasons for discontinuation are shown in Figure 1. At the end of the study, the overall percentage of the patients reaching target office BP (<140/90 mm Hg) was identical in both groups (71.7% vs 71.4%; P=not significant).

Table I.   Demographic and Baseline Characteristics of the Patient Population at Randomization
 Valsartan (n=63)Amlodipine (n=62)
  1. Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; PP, pulse pressure; PWV, pulse wave velocity; SBP, systolic blood pressure; SD, standard deviation. aP<.05.

Age ≥65 y, mean±SD (%)62.3±5.8 (35)60.4±5.1 (19)
Weight, kg72.3±14.571.5±12.9
BMI, kg/m227.5±5.227.0±4.8
Prior antihypertensive medication, %87.375.8a
Duration of hypertension, y6.8±78.3±6.4
Statins, No. (%)18 (28.6)16 (25.8)
Carotid femoral PWV, m/s14.0±3.413.6±3.2
SBP, mm Hg157.8±10.2158.0±12.3
DBP, mm Hg91.7±8.792.8±8.8
PP, mm Hg65.9±11.165.2±12.2
Mean arterial pressure, mm Hg101.3±9.799.5±7.2
Central SBP, mm Hg136.7±14.8136.3±13.2
Central DBP, mm Hg86.4±1187.4±7.8
Heart rate, beats per min77.2±1176.1±11
Glucose, mM5.6±0.65.5±0.6
Total cholesterol, mg/dL228±32228±28
LDL-C, mg/dL132±28128±24
HDL-C, mg/dL74±2077±16
Triglycerides, mg/d119±54115±54
image

Figure 1.  Overall patient disposition during the study. BP indicates blood pressure; HCTZ, hydrochlorothiazide.

Download figure to PowerPoint

Primary and Secondary Efficacy Outcomes

At the study endpoint, both treatment groups showed a reduction in PWV, with a slightly greater decrease in the valsartan group compared with the amlodipine group (change from baseline, −1.9±0.3 vs −1.7±0.3 m/s; P=not signficant). However, there was no statistically significant difference observed between the two treatment regimens (Figure 2). Secondary efficacy variables at week 12 and week 38 are presented in Table II. At week 12, the mean reduction in brachial SBP was significantly greater in the amlodipine group when compared with the valsartan group (−23.7 vs −19.4 mm Hg; P<.05). At week 38, the reduction in brachial SBP/DBP was slightly greater in the amlodipine group than the valsartan group (−25.2/−11.7 mm Hg vs −22.9/−10.9 mm Hg; P=not signficant). The reductions in central SBP/DBP were higher in the amlodipine group compared with the valsartan group (−19.2/−10.4 mm Hg vs −13.6/−5.9mm Hg, P<.01) at week 12. This difference decreased at week 38 (−19.2/−10.3 mm Hg vs −15.7/−7.6mm Hg, P<.05 only for central DBP). PP was similarly reduced in both groups. There was no significant correlation between HCTZ use/non-use and PWV, BP, and central BP at the end of the study.

image

Figure 2.  Least-square mean (LSM) change in pulse wave velocity (PWV) at week 12 and at the end of the study (EOS).

Download figure to PowerPoint

Table II.   Change From Baseline in the Efficacy Variables at Week 12 and at Endpoint (Week 38)
 Valsartan (n=56)Amlodipine (n=53)
Week 12Week 38Week 12Week 38
  1. Abbreviations: DBP, diastolic blood pressure; PP, pulse pressure; SBP, systolic blood pressure. Values are expressed as least-square means±standard error from an analysis of covariance model with treatment and statin use as factors and baseline measurement and age as covariates, except for mean arterial pressure (MAP) and heart rate (mean±standard deviations [SDs]). aP<.05. bP<.01.

SBP, mm Hg−19.4±1.5−22.9±1.6−23.7±1.5a−25.2±1.6
DBP, mm Hg −9.4±1.1−10.9±1.0−10.2±1.1−11.7±1.0
PP, mm Hg −9.9±1.1−13.5±1.2−11.9±1.2−13.4±1.3
MAP mm Hg±SDb−12.8±9.8−15.1±9.5−15.1±7.8−16.5±7.3
Central SBP, mm Hg−13.6±1.5−15.7±1.5−19.9±1.5b−19.2±1.5
Central DBP, mm Hg −5.9±1.0 −7.6±1.0−10.4±1.0b−10.3±1.0a
Heart rate, beats per min±SDb78.2±15.875.2±9.279.2±11.776.3±10.3

Safety Results

The most frequent AEs (≥1% in either group) are presented in Table III. The overall incidence of AEs was higher in the amlodipine treatment group (n=56 [90.3%]) compared with the valsartan treatment group (n=49 [77.8%]) and the majority of AEs were mild in severity. A much higher percentage of AEs were related to the study drug in the amlodipine treatment group (77.4%) as compared with the valsartan group (25.4%). The most frequently reported AE was peripheral edema, affecting 77.4% and 14.3% patients in the amlodipine and valsartan treatment groups, respectively. The majority of AEs for peripheral edema occurred in the first 12 weeks of the study for both valsartan (78%) and amlodipine (73%); however, the incidence of peripheral edema was much higher at every time point during the study (weeks 4, 12, and 38) for amlodipine. There were 4 patients (2 [3.2%] in each group) who experienced serious AEs (SAEs). No deaths were reported during the study. Discontinuations due to AEs were 12.7% and 12.9% in the valsartan and amlodipine treatment groups, respectively. AEs requiring dose adjustment or interruption of the study drug were 15.9% in the valsartan group compared with 29% in the amlodipine treatment group.

Table III.   Patients With Most Frequent Adverse Events
Adverse EventsValsartan 320 mg, No. (%) (n=63)Amlodipine 10 mg, No. (%) (n=62)
  1. aA patient with multiple occurrences of an adverse event while taking one treatment is counted only once.

Adverse events, No. (%)a49 (77.8)56 (90.3)
Peripheral edema9 (14.3)48 (77.4)
Bronchitis8 (12.7)0 (0.0)
Headache6 (9.5)5 (8.1)
Asthenia5 (7.9)2 (3.2)
Hypotension5 (7.9)0 (0.0)
Insomnia4 (6.3)1 (1.6)
Vertigo4 (6.3)2 (3.2)
Back pain3 (4.8)0 (0.0)
Hot flush3 (4.8)1 (1.6)
Arthralgia2 (3.2)1 (1.6)
Dizziness2 (3.2)2 (3.2)
Epistaxis2 (3.2)1 (1.6)
Muscle spasms2 (3.2)4 (6.5)
Nasopharyngitis2 (3.2)1 (1.6)
Neck pain2 (3.2)0 (0.0)
Osteoarthritis2 (3.2)0 (0.0)
Rash2 (3.2)0 (0.0)
Syncope2 (3.2)0 (0.0)
Abdominal pain1 (1.6)2 (3.2)
Abdominal pain upper1 (1.6)2 (3.2)

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

The major finding of the present study is that despite a larger brachial and central BP reduction in the amlodipine group than in the valsartan group, a larger but not significant carotid-to-femoral PWV reduction was observed with the RAS blockade regimen. The other clinically interesting finding of the study was the observation of a very high percentage of edema in patients taking the CCB to reach target BP as specified by the study design in this postmenopausal hypertensive female population.

Comparison of CCB With ARB

Why should the carotid-to-femoral PWV be different for similar BP reduction between the two groups? We hypothesized that two different drug regimens reducing mean brachial and central BP to the same magnitude could influence aortic stiffness in a pressure-independent manner due to more pronounced reflex sympathetic activation in the CCB group. Indeed, several markers of sympathetic activation have been observed in patients taking CCBs. An increase in sympathetic nerve activity without evident changes in heart rate has been reported previously, although newer long-acting CCBs are less prone to stimulate the sympathetic nervous system in the resting position.20 In patients taking CCB therapy, 24-hour mean heart rate has been shown to be slightly increased compared with in those taking a RAS blocker.14 A positive correlation between heart rate and carotid-to-femoral PWV was observed in this study, which was consistent with previous studies.21–23 However, in the present study, HR was not monitored over 24 hours and we did not observe any significant difference in HR between treatment groups at week 12 and week 38 at the time of PWV assessments in our patients in a supine position. Recently, de Champlain et al reported significant changes in plasma catecholamine levels in patients treated with amlodipine (10 mg/d) compared with patients taking valsartan (160 mg/d). The difference between the two treatment groups was more pronounced when the patients were standing.14 The role of catecholamine in resistant hypertension was confirmed by the substantial and sustained BP reduction following renal denervation therapy (RDN).24 The high prevalence of resistant hypertension, despite multiple therapies (including a diuretic) and the results of RDN strongly support the importance of the compensatory sympathetic efferent nerves overactivity.

Catecholamines have been shown to inactivate nitric oxide (NO) and impair endothelial vasomotion via enhanced reactive oxygen species production and possibly via endothelin secretion.25,26 Whether local differences in NO concentration may have played a role in the observed changes in PWV between the two regimens remains unclear. The implication of endothelium-derived NO as a putative mediator of PWV attenuation has been demonstrated repeatedly.23,27 There are data suggesting that such a mechanism may be involved in large arteries. Indeed, when hypertensive patients were treated with valsartan 80 mg/d to 160 mg/d or amlodipine 5 mg/d to 10 mg/d for 1 year, significant differences in endothelial function were observed between the groups. The 12-month treatment with valsartan induced an increased flow-mediated dilation whereas amlodipine did not modify endothelial function.28

Side Effects of CCB vs ARB in Postmenopausal Hypertensive Women

More than 70% of the participants reached target office BP (<140/90 mm Hg) by study end with both drug regimens. This result was achieved at a cost of a very high proportion of patients presenting significant ankle edema (defined as the necessity to change shoes or feeling uncomfortable with the regular shoes at the end of the day). This proportion was extremely high in the CCB groups, reaching more than 75% despite diuretic use vs <15% in the ARB group. This is also illustrated by the need to reduce the drug dosage in only 15.9% in the valsartan group compared with 29% in the amlodipine treatment group.

Limitations

This study has some limitations. Because it was the first study addressing PWV changes in postmenopausal woman taking antihypertensive agents, the calculation of sample size was limited to hypothesis. A larger number of patients should have been included to assess true therapeutic effect difference between the treatments. The very high prevalence of edema in the amlodipine group may have unblinded the study, but the measurements were performed by researchers unaware of the clinical data and of the treatment delivery. Another limitation is the difference in the number of patients taking diuretics at the end of the study (43% for valsartan and 21% for amlodipine). This is due to the choice of amlodipine as a comparator. Indeed, amlodipine has been steadily demonstrated to be more potent in BP reduction than ARBs when used as monotherapy. The effect of this difference in the use of diuretics on PWV is not measurable but is probably limited because BP reduction was similar in both groups at end of study. Moreover, HCTZ has not been demonstrated to have any significant influence on vascular stiffness independent from its BP effect. Finally, ambulatory BP monitoring to fully assess changes in systemic BP and possible differences in day/night BP control as well as differences in heart rate outside the examination period could be interesting and allow calculation of the ambulatory arterial stiffness index which may have helped confirm findings with PWV measures.

Acknowledgments and disclosures:  The authors gratefully thank Drs Shivali Arora and Lakshmi Deepa of Novartis Pharmaceuticals for writing assistance and comments. The study was initiated and designed by the principal academic investigator (DH). Novartis Pharmaceuticals provided financial support. DZ, IYB, and AK are employees of Novartis Pharmaceuticals.

Conclusions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

At 38 weeks, both valsartan and amlodipine treatment regimens reduced PWV in hypertensive postmenopausal women to a similar degree, despite differences in central BP lowering suggesting that the effect of RAS blockade to influence PVW may partly be independent of central BP with a slightly larger PWV decrease in the valsartan treatment group.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References