Blood pressure response with fixed-dose combination therapy: comparing hydrochlorothiazide with amlodipine through individual-levelmeta-analysis

Background:Although fixed-dose combination drug therapy is commonly used to treat hypertension, the efficacy of head-to-head comparisons of dual fixed-dose combinations has not been well described. We hypothesized that when used in combination with an angiotensin receptor blocker (ARB) olmesartan medoxomil, hydrochlorothiazide (HCTZ) will be as effective as the dihydropyridine calcium channel blocker (CCB) amlodipine to lower both clinic and 24-h ambulatory blood pressure (BP). Furthermore, we hypothesized that response to ARB along with HCTZ or ARB along with CCB may be heterogeneous depending on clinical characteristics.Methods:An individual-level meta-analysis was performed among 559 individuals treated with dual combination therapy in five trials. A forced titration scheme was used in each of these trials and blood BP was measured both in the clinic and outside using 24-h ambulatory BP monitors.Results:The mean age was 62 years, 55% were men, 46% had diabetes mellitus, 17% were black, clinic BP averaged 159.5/89.5mmHg and 24-h ambulatory BP 145.0/82.5mmHg. Overall, baseline-adjusted lowering of mean 24-h ambulatory BP was 22.0/11.7mmHg. BP reductions were similar between ARB along with HCTZ and ARB along with CCB groups. However, clinic BP was lowered 4.3/1.8mmHg more with ARB along with CCB combination (28.4/13.0mmHg drop) than with ARB along with HCTZ combination (24.1/11.2mmHg drop). The white coat effect (WCE) was therefore mitigated 3.8/1.7mmHg more with ARB along with CCB combination. Heterogeneity in ambulatory BP response was noted. Compared with men, women had a greater ambulatory and clinic BP lowering with either combination. ARB along with HCTZ produced a greater BP-lowering effect among men, elderly, nonobese and nondiabetic. On the contrary, ARB along with CCB produced a greater BP-lowering effect among women, young, obese and diabetic individuals. This heterogeneity in response was often undetectable with clinic BP measurements. In multivariable analysis, sex and diabetes mellitus remained independent measures of heterogeneity.Conclusion:Overall, the combination of olmesartan and HCTZ is as effective as olmesartan and CCB in lowering 24-h, daytime, and night-time ambulatory BP. However, greater lowering is noted with the olmesartan and CCB combination for clinic BP. Thus, out-of-office BP monitoring is necessary to provide better assessment of overall BP and response to treatment. Women and diabetic individuals may have slightly better 24-h ambulatory BP response with the olmesartan and CCB combination therapy.