Jump to
Abstract
Day-by-day home blood pressure (BP) variability (BPV) was reported to be associated with increased cardiovascular risk. We aimed to test the hypothesis that the angiotensin II receptor blocker/calcium-channel blocker combination decreases day-by-day BPV more than the angiotensin II receptor blocker/diuretic combination does and investigated the mechanism underlying the former reduction. We enrolled 207 hypertensive subjects treated with olmesartan monotherapy for 12 weeks. The subjects were randomly assigned to treatment with hydrochlorothiazide (n=104) or azelnidipine (n=103) for 24 weeks. Home BP was taken in triplicate with a memory-equipped device in the morning and evening, respectively, for 5 consecutive days before each visit. Visits occurred at 4-week intervals. Home BPV was defined as within-individual SD of the 5-day home BP. Arterial stiffness was assessed by aortic pulse wave velocity at baseline and 24 weeks later. The reductions in home systolic BP were similar between the 2 groups, whereas the SD of home systolic BP decreased more in the azelnidipine group than in the hydrochlorothiazide group during the follow-up period (follow-up mean: 6.3 versus 7.1 mm Hg; P=0.007). In the azelnidipine group, the change in aortic pulse wave velocity was independently associated with the change in SD of home systolic BP (regression coefficient±SE=0.79±0.37; P=0.036). This study demonstrated that the angiotensin II receptor blocker/calcium-channel blocker combination improved home BPV in addition to home BP reduction and that the reduction in home BPV was partly attributable to the arterial stiffness reduction by this combination.
- home blood pressure
- day-by-day home blood pressure variability
- calcium channel blocker
- thiazide diuretic
- angiotensin II receptor blocker
- pulse wave velocity
Introduction
See Editorial Commentary, pp 1091–1093
There is a growing body of evidence that visit-to-visit blood pressure (BP) variability (BPV) in a clinical setting was significantly associated with target organ damage1–6 and that it could predict cardiovascular events independent of mean office BP level.7–10 One drawback of the application of visit-to-visit office BPV for the routine clinical management of hypertension is that obtaining it requires several office visits over a period of time. One possible way to solve this is to evaluate day-by-day home BPV, because this self-measurement makes it possible to obtain multiple BP values in a relatively short period under well-controlled conditions.11 It was reported recently that the increased day-by-day home BPV, calculated as the SD of home BP, was associated with the severity of target organ damage in patients with hypertension12 or diabetes mellitus13 and with cardiovascular mortality in a community-dwelling population.14 After treatment has been initiated, home BP monitoring can be a practical tool with which to assess the efficacy of antihypertensive drugs on day-by-day BPV. However, no study has investigated the effects of antihypertensive medication on day-by-day home BPV.
The Avoiding Cardiovascular Events Through Combination Therapy in Patients Living with Systolic Hypertension Trial has demonstrated that the combination of angiotensin-converting enzyme inhibitor/calcium channel blocker (CCB) is more effective than the angiotensin-converting enzyme inhibitor/thiazide diuretic for decreasing cardiovascular events in hypertensive patients.15 In an Avoiding Cardiovascular Events Through Combination Therapy in Patients Living with Systolic Hypertension Substudy, 24-hour BP levels were not significantly different between the 2 regimens.16 Thus, these results15,16 raised speculation that the difference in cardiovascular events that favored the CCB regimen was not explained by BP differences but, rather, by the favorable effects of the combination of CCB and renin-angiotensin system (RAS) blocker on central BP17 or BPV.18 Rothwell et al19 have demonstrated that the benefits of CCB in the prevention of cardiovascular events may be attributed partly to the reduction in visit-to-visit office systolic BPV. However, it remains to be determined whether the addition of CCB to RAS blockers decreases day-by-day home BPV more than does the addition of a diuretic to RAS blockers.
We hypothesized that the angiotensin II receptor blocker (ARB)/CCB combination would decrease day-by-day home BPV more than the ARB/diuretic combination would. Using data from the Japan Combined Treatment With Olmesartan and a CCB Versus Olmesartan and Diuretics Randomized Efficacy Study,17 we compared the change in day-by-day home BPV from baseline between the 2 combination groups. We showed recently that arterial stiffness was independently associated with office BPV3 and home BPV.12 With this information as background, we explored the mechanism underlying the reduction in home BPV using data on aortic pulse wave velocity (aPWV) in each combination group.
Methods
Study Subjects
The study participants were recruited from the Outpatient Department of Internal Medicine (Mishima Clinic, Hagi, Japan). The entry period was from May 2006 to October 2007. We initially enrolled consecutively treated or untreated hypertensive subjects who agreed to participate in this study. During the 12-week run-in period, the subjects received a once-daily 20-mg dose of olmesartan monotherapy. Those who were already being treated were instructed to switch their regimen of any current antihypertensive medications to treatment with olmesartan only. At the end of the run-in period, the subjects with an office BP ≥140 and/or 90 mm Hg were eligible for the study (Figure S1, available in the online-only Data Supplement). This study was approved by the institutional review board of Jichi Medical University, and written informed consent was obtained from all of the participants. A detailed description of the study participants appears in a previous report.17
Study Design
The Japan Combined Treatment With Olmesartan and a CCB Versus Olmesartan and Diuretics Randomized Efficacy Study was a 24-week, prospective, randomized, open-label, blinded-end point study with 2 treatment arms.17 Assessment of the effects of treatment on home BP was prespecified as a secondary outcome measure in this trial. Briefly, eligible subjects were randomly assigned to receive either azelnidipine 16 mg or hydrochlorothiazide (HCTZ) 12.5 mg as an add-on to olmesartan 20.0 mg. Both treatments were taken as a fixed dose daily after breakfast for 24 weeks. Dose titration was not permitted. Other drugs that could potentially interfere with the efficacy of the study medications were not allowed.
BP Measurement
At each office visit, office BP was recorded as the average of triplicate measurements taken at intervals of 1 minute using a validated oscillometric device (HEM-907; Omron Healthcare, Kyoto, Japan)20 after an initial 5 minutes of seated rest. Office BP was taken by the nurses in the morning.
Home BP was measured in a sitting position 3 times each morning and 3 times each evening for 5 consecutive days before each office visit (total of 7 visits and total of 30 BP measurements; Figure S1). The subjects were instructed to place the cuff on the nondominant arm, rest for 5 minutes before the first reading, and take a 15-second interval between readings. Morning BP was measured within 1 hour after waking, after urination, and before breakfast. Evening BP was measured just before going to bed and ≥60 minutes after taking a bath. This protocol is based on Japanese home BP guidelines.21 The device used to monitor home BP was a validated oscillometric device (HEM-747IC; Omron Healthcare), which incorporates an integrated circuit memory and a clock to store BP and heart rate (HR) readings and measurement times.22 It was shown that HEM-907 and HEM-747IC had the same algorithm for deriving systolic BP (SBP) and diastolic BP from the oscillometric readings.20,22 The arm circumferences of the subjects ranged between 22 and 32 cm, so a standard arm cuff could be used for BP measurement in all of the cases. Home BP was defined as the average of morning BP and evening BP for each day. The day-by-day home BPV and HR variability (HRV) were defined as the SD of the daily BP/HR average (average of 6 readings) of 5 consecutive days (SD of 5 day-by-day readings).14,23,24 Similarly, the day-by-day morning BPV/HRV and evening BPV/HRV were defined as the SD of the daily BP/HR average in the morning or in the evening (average of 3 readings) of 5 consecutive days.14,23,24
Arterial Stiffness Measurement
After the subject rested for 5 minutes in the supine position, the aPWV was measured with a SphygmoCor device (version 7.0; AtCor Medical, Sydney, New South Wales, Australia). The aPWV was measured from sequentially recorded ECG-gated carotid and femoral artery waveforms. This measurement was made in duplicate, and the mean values were used in the subsequent analysis. Detailed descriptions of the aPWV measurement and its reproducibility were reported previously.17
Statistical Analyses
Statistical analyses were performed based on an intention-to-treat principle. All of the data were expressed as the mean±SD or as a percentage. Differences between the groups at baseline were analyzed using the unpaired t test for continuous variables or the χ2 test for categorical variables. ANCOVA was performed to compare the changes in variability indices between the treatment groups, with each baseline value as a covariate. The follow-up means and mean differences (least-squares means) were estimated by repeated-measures ANOVA using all of the follow-up data. Pearson correlation coefficients were used to explore the univariate associations between the examined variables. Forced-entry multivariate linear regression analysis was performed to explore the independent associations between the SD of home SBP and clinical parameters at baseline, as well as those between the change in the SD of home SBP and the changes in hemodynamic parameters in the 2 groups, separately. The independent variables entered into the multivariate models were those that were significantly associated (P<0.05) in univariate analyses, as well as those variables that were known to affect or were previously associated with home BPV from published observations.12,23,24 The variance inflation factor was calculated to assess potential multicollinearity in multivariate models. We confirmed that all of the variance inflation factors were <2.5. Two-sided values of P<0.05 were considered to indicate statistical significance. All of the statistical analyses were performed with SPSS version 16.0 (SPSS Inc, Chicago, IL).
Results
Baseline Data
The flow of subjects through each stage of the study was described in our previous article.17 The average number of home BP measurements per subject during the 5-day period was 29.8±0.7 at baseline, 29.6±1.4 at 1 month, 29.5±1.4 at 2 months, 29.3±1.7 at 3 months, 29.4±1.5 at 4 months, 29.4±1.5 at 5 months, and 29.2±1.8 at 6 months. The baseline characteristics including BP were similar between the groups (Tables 1, 2, and S1). The percentages of subjects with white-coat hypertension (home BP <135/85 mm Hg and office BP ≥140/90 mm Hg) were 25% in the HCTZ group and 23% in the azelnidipine group. At baseline, age and aPWV were significantly correlated with SD of home SBP, whereas these independent associations were not observed in the multivariate analyses (Table 3). On the other hand, body mass index, home SBP, and the SD of home HR were independently associated with SD of home SBP. When the SD of home SBP was divided into morning and evening components, these associations were almost the same (Table S2).
Baseline Characteristics of the Study Subjects
Comparison of Baseline Values and Adjusted Changes After 6 Months in Home BP Variability and HR Variability Between the HCTZ and Azelnidipine Groups
Univariate and Multivariate Linear Regression Analyses of Factors Associated With SD of Home Systolic BP at Baseline (n=207)
Changes in BP and HR
The office BP and home BPs decreased significantly from baseline in both treatment groups, and the follow-up means of office BP/home BPs were similar between the groups, except for office diastolic BP (Figures 1 and S2). On the other hand, the home HRs decreased significantly from baseline only in the azelnidipine group, and the follow-up means of home HRs were significantly smaller in the azelnidipine group than in the HCTZ group (Figure S3).
Effects of 24 weeks of treatments on office and home blood pressure (BP). *P<0.001 for comparisons from baseline. SBP indicates systolic blood pressure; DBP, diastolic blood pressure. A, The follow-up means of office SBP/DBP were 134.2/73.7 mm Hg in the hydrochlorothiazide (HCTZ) group and 132.1/70.9 mm Hg in the azelnidipine group. B, The follow-up means of home SBP/DBP were 133.2/73.8 mm Hg in the HCTZ group and 134.7/73.5 mm Hg in the azelnidipine group. ●, +HCTZ (SBP); ■, +HCTZ (DBP); ○, +azelnidipine (SBP); ☐, +azelnidipine (DBP).
Changes in Home BP Variability and HR Variability
As shown in Figures 2 and S4, the follow-up means of the SD of home SBPs/diastolic BPs were significantly smaller in the azelnidipine group than in the HCTZ group. On the other hand, the follow-up means of the SD of home HRs were similar between the groups (Figure S5). After 6 months of treatment, the reductions in the SD of home BP and the SD of home HR were greater in the azelnidipine group than in the HCTZ group (Table 2). As shown in Table S1, the results of morning and evening BPV/HRV were largely confirmatory. When we similarly compared the reductions in the SD of home SBP between the groups after excluding subjects with white-coat hypertension, the results were the same as in the total population (data not shown). When coefficient of variation, defined as SD divided by mean BP level, was used instead of SD to measure variability, the results were essentially the same (data not shown).
Effects of 24 weeks of treatments on SD of home blood pressure (BP). *P<0.05 and †P<0.001 for comparisons from baseline. SBP indicates systolic blood pressure; DBP, diastolic blood pressure. The follow-up means of SD of home SBP/DBP were 7.1/3.7 mm Hg in the hydrochlorothiazide (HCTZ) group and 6.3/3.4 mm Hg in the azelnidipine group. ●, +HCTZ (SBP); ■, +HCTZ (DBP); ○, +azelnidipine (SBP); ☐, +azelnidipine (DBP).
Association With Change in Home SBP Variability in the 2 Treatment Groups
In univariate analyses, the change in aPWV was significantly correlated with the change in the SD of home SBP in the azelnidipine group, whereas this significant correlation was not observed in the HCTZ group (Table 4). As shown in Figure S6, the change in home SBP was not significantly correlated with the change in the SD of home SBP in both groups. In the multivariate linear regression analyses, the change in aPWV was independently associated with the change in the SD of home SBP in the azelnidipine group. In the HCTZ group, the change in the SD of home HR was independently associated with the change in the SD of home SBP (Table 4). The results showing independent associations with changes in morning and evening BPV were largely confirmatory (Table S3).
Univariate and Multivariate Linear Regression Analyses of Factors Associated With Change in SD of Home Systolic BP in the HCTZ and Azelnidipine Groups
Discussion
The present study demonstrated that the addition of CCB to ARB decreased the day-by-day home BPV more effectively than did the addition of a diuretic to ARB, whereas the reductions in home BP level were similar in both regimens. Only in the CCB group, the arterial stiffness reduction was independently associated with the reduction in home BPV. To the best of our knowledge, this is the first study showing the differential effects of antihypertensive combination therapies on day-by-day home BPV and the first to clarify the mechanism underlying the reduction in home BPV.
We found that the addition of CCB significantly reduced the day-by-day home BPV, independent of the reduction in mean home BP level. This finding is in agreement with previous reports that, compared with other antihypertensive drugs, CCB lowered long-term BPV as assessed by visit-to-visit office BPV18,19 or short-term BPV as assessed by ambulatory BP monitoring.25,26 Although the methods of measuring BPV differ, the decreases in SD compared with the mean decrease in SBP with antihypertensive medication are roughly similar to our results.19,26 Because of the limited capacity of peripheral vasodilation, BPV by pressor stimulation may be augmented in the presence of small-artery remodeling that occurs through aging and hypertension.27 Therefore, the greater reduction in BPV among subjects treated with CCB may be best explained by their more profound effects on the relaxation of peripheral muscular arteries. Another study suggested that CCB could decrease BPV by attenuating the myogenic response of vascular smooth muscle cells.28 On the other hand, because BPV is controlled mainly by arterial baroreflex sensitivity (BRS),29 one can speculate that CCB's effect on day-by-day home BPV is mediated by improving the BRS. Because it was reported that azelnidipine, a long-acting dihydropyridine-type CCB, could improve the BRS significantly and more than could amlodipine in hypertensives,30 the add-on treatment of azelnidipine might have reduced the home BPV partly through the improvement of the BRS.
In the CCB group of the present study, we found that the arterial stiffness reduction was independently associated with the reduction in home BPV. Cross-sectional studies have demonstrated that the long-term office BPV3 and the short-term BPV assessed by ambulatory BP monitoring31 were significantly associated with large-artery stiffness. This may be explained by the fact that there may be less buffering of BP changes when the elastic recoil of the large artery is impaired, resulting in a wider SBP oscillation for any given change in stroke volume.32 Another possible explanation is that increased arterial stiffness, through the impairment of the BRS, might favor an increase in BPV.33 Furthermore, it was reported that the change in BPV caused by CCB amlodipine was significantly associated with the change in PWV.25 Indeed, it is possible that a reduction in BPV by CCB could also contribute to an improvement in arterial stiffness. Considering the pharmacological action of azelnidipine,17,30 however, it is probable that the CCB-induced reduction in arterial stiffness would lead to reduced home BPV, both directly and through the improvement of baroreflex modulation.
We have demonstrated that aPWV decreased more in the ARB/CCB group than in the ARB/HCTZ group and that the aPWV reduction was an independent determinant of the reduction in central SBP.17 In addition, the present study provides evidence that the aPWV reduction is an independent determinant of the reduction in home BPV in the ARB/CCB group. Thus, the destiffening of the large arteries by ARB/CCB treatment might be a common underlying mechanism of the significant reductions in central SBP and home BPV, each of which was reported to be an independent predictor of cardiovascular events.14,34 Thus, the Japan Combined Treatment With Olmesartan and a CCB Versus Olmesartan and Diuretics Randomized Efficacy Study could provide a key to understanding why the combination of RAS blocker/CCB was more effective than that of RAS blocker/diuretic for decreasing cardiovascular events in hypertensive patients.15
In the present study, the ARB/HCTZ combination treatment did not change the SD of home BP significantly from baseline. This result may be explained by previous reports showing that the use of RAS inhibitors was associated with increased BPV,10,18 whereas the use of thiazide diuretics was associated with lower BPV but less so than was the case with CCB.18,19 We have shown that plasma aldosterone concentration increased significantly from baseline only in the ARB/HCTZ group.35 Monahan et al36 have demonstrated that aldosterone impaired BRS in humans. Thus, aldosterone excess may contribute to less improvement of home BPV in the ARB/HCTZ group through its detrimental effect on BRS. Furthermore, in this group, the change in home HRV, particularly in the morning, was independently associated with the change in home systolic BPV. These results are somewhat similar to those of a previous report26 that the change in 24-hour BPV by a thiazide diuretic was independently associated with the change in nighttime HRV. Although the mechanisms underlying these associations remain to be clarified, the regulation of BRS and autonomic nervous system during ARB/diuretic treatment might contribute to the significant association between these changes.
Study Limitations
This study has some limitations that require consideration. First, because this study was not carried out according to a double-blind design, this may have affected the main results. However, home BP was recorded on a memory-equipped device. Furthermore, both treatments in this study were given as a fixed dose throughout study period, and the researchers did not know the detail of home BP data until final analysis. Therefore, a substantial information bias may not contribute to the measurement and evaluation of home BPV. Secondly, the 5-day home BP measurement period may have been somewhat too short to accurately estimate day-by-day home BPV, and this may have affected the results. On the other hand, even when the visit-to-visit office BPV was estimated using BPs at only 3 time points, this BPV had significant prognostic value.10 Future studies should determine the optimal number of home BP measurements needed to obtain reliable and valid estimates of the day-by-day home BPV. Third, the present data represent the results of a post hoc analysis of the data collected during the Japan Combined Treatment With Olmesartan and a CCB Versus Olmesartan and Diuretics Randomized Efficacy Study. Thus, our findings on the possible different effects of different treatments on home BPV should be verified by other properly designed studies in which patients are randomly assigned to the treatment groups considered. Fourth, the relatively limited sample size of the present study might have affected the statistical analyses, leading to chance findings. Fifth, the reproducibility of day-by-day home BPV is unknown. However, a recent report suggested that visit-to-visit office BPV is reproducible and not a random phenomenon.37 Further research is needed to clarify the reproducibility of home BPV in subjects without antihypertensive medication. Finally, evening home BP was measured 60 minutes after a bath, before the patient went to bed. Therefore, we cannot exclude an effect of food or alcohol on these evening values.
Perspectives
In clinical practice, self-home BP monitoring will be more appropriate for the repeated assessment of home BPV in the long-term follow-up of hypertensives than both office and ambulatory BPV measurements. It may also help the practicing physician to optimize antihypertensive treatment at every clinic visit, because information needed for the calculation of home BPV can be collected relatively quickly rather than after months or years. Because an exaggerated day-by-day home BPV was associated with cardiovascular events, the ARB/CCB combination treatment may have additional beneficial effects on cardiovascular protection by reducing this BPV in addition to significantly reducing the mean levels of home BP. However, our study was unable to determine whether the reduction of home BPV by antihypertensive medication would lead directly to better outcomes. Further studies are needed to test this hypothesis.
Sources of Funding
The Japan Combined Treatment With Olmesartan and a Calcium-Channel Blocker Versus Olmesartan and Diuretics Randomized Efficacy Study was supported by a grant from Jichi Medical University School of Medicine. Home blood pressure monitors were provided by Omron Healthcare, Co, Ltd (Kyoto, Japan).
Disclosures
M.F.O. is a founding director of AtCor Medical, the manufacturer of the systems used for analyzing arterial pulse.
Acknowledgments
We thank the nurses (Akemi Taguchi and Akira Fukuma) for their assistance with the study.
Footnotes
Continuing medical education (CME) credit is available for this article. Go to http://cme.ahajournals.org to take the quiz .
This trial has been registered at www.clinicaltrials.gov (identifier NCT00607035).
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.111.189217/-/DC1.
- Received December 6, 2011.
- Revision received December 29, 2011.
- Accepted April 1, 2012.
- © 2012 American Heart Association, Inc.
References
This Issue
Jump to
Article Tools
Share this Article
- Combined Effect of Angiotensin II Receptor Blocker and Either a Calcium Channel Blocker or Diuretic on Day-by-Day Variability of Home Blood Pressure
