Randomized Double-Blind Comparison of a Calcium Antagonist and a Diuretic in Elderly Hypertensives
Abstract—Although diuretics are recommended for the treatment of hypertension, decreased diuretic use and increased calcium antagonist use necessitate a comparison of the efficacy of these drugs in preventing cardiovascular events. Patients ≥60 years of age with systolic blood pressure of 160 to 220 mm Hg and diastolic blood pressure <115 mm Hg were enrolled. Patients were randomly assigned to 20 mg of sustained-release nicardipine hydrochloride twice daily or 2 mg of trichlormethiazide once daily by the double-dummy method and followed up for 5 years. A total of 414 patients were analyzed: 204 in the nicardipine group and 210 in the diuretic group. Blood pressure at entry was 172/94 mm Hg and 173/93 mm Hg, respectively, and decreased to 147/81 mm Hg and 147/79 mm Hg, respectively. Cardiovascular morbidity rates per 1000 persons per year were similar in the nicardipine and diuretic groups (27.8 and 26.8, respectively; P=0.923). The sex- and age-adjusted risk ratio for the nicardipine group was 0.973 (95% confidence interval, 0.514 to 1.839, P=0.932). The calcium antagonist and diuretic groups had a similarly decreased rate of cardiovascular events.
The National Intervention Cooperative Study in Elderly Hypertensives (NICS-EH) was initiated in 1989 to compare the effects of a diuretic and a calcium antagonist in preventing cardiovascular complications in elderly hypertensive patients in Japan.
The efficacy of treating hypertension in the elderly was reported by Kuramoto et al,1 the European Working Party on High Blood Pressure in the Elderly trial,2 the Systolic Hypertension in the Elderly Program (SHEP),3 the STOP-Hypertension study,4 and the Medical Research Council (MRC) study.5 These studies used diuretics or β-blockers, which were the first-line drugs at the time, in treating hypertension. In recent years, the Shanghai Trial of Nifedipine in the Elderly in China (STONE),6 the Systolic Hypertension in Europe (Syst-Eur),7 and the China (Syst-China)8 trials have shown that calcium antagonists are useful for the prevention of cardiovascular complications in elderly hypertensive patients.
Although diuretics are recommended for the treatment of hypertension, a decrease in diuretic use and increase in calcium antagonist use necessitate a comparison of the efficacy of these agents in preventing cardiovascular events.
The details of the design of and methods used in the NICS-EH have been published previously.9 In brief, patients were ≥60 years of age with systolic blood pressure (SBP) 160 to 220 mm Hg and diastolic blood pressure (DBP) <115 mm Hg after 4 weeks of placebo administration, with no history of cardiovascular complications. Patients were randomly assigned by the controller in the study coordinating office to 20 mg sustained-release nicardipine hydrochloride twice daily (nicardipine group) or 2 mg trichlormethiazide once daily (diuretic group) by the double-dummy method and followed up for 5 years. When the blood pressure response was insufficient in the judgment of the investigator, doubling of the dose was permitted. All other antihypertensive drugs were prohibited.
Major end points were cardiovascular complications. For patients who had any end point, the attending physician’s judgment was assessed blindly by the Steering Committee and the diagnosis was confirmed.
For statistical analysis, the baseline characteristics of the 2 groups were compared by use of the χ2 test or Mann-Whitney test for categorical variables and the 2-sample t test for continuous variables. Event-free curves were generated by use of the Kaplan-Meier method, and a log-rank test was used to assess differences between the 2 treatment groups. Risk ratios were calculated with proportional hazards regression analysis; prognostic factors for event occurrence were analyzed with the same method. Per-protocol analysis was used to analyze the results of this trial.
Ambulatory patients were recruited from October 1989 to April 1992. Patients were randomly assigned to the nicardipine group (215) or the diuretic group (214). Eight and 4 patients did not attend the hospital after the observation period and were not treated; therefore they were withdrawn from the nicardipine and diuretic groups, respectively. Two patients were disqualified because of their age, and 1 patient in the nicardipine group was withdrawn on the basis of primary disease (primary aldosteronism). Thus 204 patients were analyzed in the nicardipine group and 210 in the diuretic group.
Patient characteristics as recorded during the observation period are listed in Table 1⇓. Systolic hypertension was observed in 25% of patients in each group. The proportion of women was 67% overall but 60% in the nicardipine group and 74% in the diuretic group (P=0.03). No other significant differences between the groups were noted. Median follow-up was 1666.5 days in the nicardipine group and 1440.5 days in the diuretic group.
Changes in Blood Pressure/Pulse Rate
The changes in blood pressure are shown in Figure 1⇓. Blood pressure had fallen at 4 weeks and by the 8th week had fallen to 152±16/85±9 mm Hg in the nicardipine group and 153±17/85±10 mm Hg in the diuretic group. From 24 weeks onward, blood pressure stabilized at 147 to 150/79–83 mm Hg in both groups; at 240 weeks, blood pressure was 147±15/81±8 mm Hg in the nicardipine group and 147±16/79±9 mm Hg in the diuretic group, representing a significant reduction in both groups (P=0.000). The percentage of patients in the nicardipine and diuretic groups, respectively, who required a double dose of the study medication to control blood pressure was 13.7% and 16.4% at 3 months, 18.5% and 24.8% at 6 months, 29.7% and 23.6% at 1 year, and 30% to 41% thereafter.
The pulse rate varied from 73 to 76 bpm in the nicardipine group and from 72 to 75 bpm in the diuretic group (not significant) throughout the study. The number of patients also taking other antihypertensive agents during the period was 13 (6.4%) in the nicardipine group and 12 (5.7%) in the diuretic group.
Cardiovascular End Points
The fatal and nonfatal cardiovascular end points that occurred during the study period are listed in Table 2⇓. The total number of cardiovascular complications during the 5-year period was 21 (10.3%) in the nicardipine group and 18 (8.6%) in the diuretic group, and the estimated morbidity rate per 1000 persons per year calculated for the study period was 27.8 in the nicardipine group and 26.8 in the diuretic group. Kaplan-Meier continuation curves are shown in Figure 2⇓; the groups showed similar continuation rates throughout (P=0.923, log-rank test). According to the proportional hazards model, the sex- and age-adjusted risk ratio in the nicardipine group was 0.973 (95% confidence interval, 0.514 to 1.839, P=0.932) compared with the diuretic group. There was no difference between the groups in terms of the number of patients with cardiovascular complications who exhibited left ventricular hypertrophy on ECG during the observation period (P=0.975, log-rank test).
Unfavorable Blood Pressure Changes
Eight patients in the nicardipine group and 16 in the diuretic group had the minor end point of blood pressure in excess of 200/100 mm Hg despite a doubling of the antihypertensive agent dose. In addition, 3 patients in the nicardipine group and 6 in the diuretic group showed excessive reductions in blood pressure during treatment. No difference between the 2 groups in terms of the cumulative continuation rate (P=0.160) by Kaplan-Meier analysis was found among the patients who had cardiovascular end points, blood pressure increase, or excessive blood pressure reduction.
Other Noncardiovascular End Points
Patients withdrawn because of noncardiovascular end points are listed in Table 3⇓. Four patients were diagnosed with diabetes with no previous history, all in the diuretic group. Cancer developed in 5 patients in each group; 2 patients in the diuretic group died. The site of cancer was the large intestine in 3 patients, stomach in 2 patients (1 died), and lung, liver, breast, uterus (died), and bladder in 1 patient each; this pattern is often seen in the elderly. No instances of gastrointestinal hemorrhage were observed.
The number of patients who discontinued treatment because of side effects such as clinical disorders and abnormal laboratory values totalled 6 in the nicardipine group and 9 in the diuretic group. Headache, rash, and liver disorders were observed in both groups. The gastrointestinal disorders were diarrhea and heartburn in the nicardipine and diuretic groups, respectively. All these side effects improved on withdrawal from the study. Side effects that did not lead to patients being withdrawn from the study occurred in 35 (17.2%) patients in the nicardipine group and 38 patients (18.1%) in the diuretic group (P=0.897).
Laboratory Test Data
Table 4⇓ lists changes in laboratory test values over the course of the study, together with the percentage of patients who had abnormal values during treatment despite being within normal ranges during the observation period. The mean serum sodium values were significantly lower, whereas uric acid levels were significantly higher in the diuretic group at the end of treatment than at baseline. Glucose and blood urea nitrogen levels also tended to increase. Serum total cholesterol levels were unchanged, and HDL-cholesterol levels increased in both groups. Significantly more patients in the diuretic group exhibited abnormal blood urea nitrogen and aspartate transaminase levels.
Four patients in the diuretic group received potassium supplementation. Two patients in the nicardipine group and 7 in the diuretic group required uric acid–lowering agents during treatment.
The NICS-EH study demonstrated that the dihydropyridine calcium antagonist nicardipine exhibits an effect against cardiovascular complications in elderly hypertensive patients that is equivalent to that of the diuretic trichlormethiazide. Cardiovascular complication rates were 10.3% and 8.6% in the nicardipine and diuretic groups, respectively, and the estimated morbidity rate per 1000 persons per year was 27.8 and 26.8, respectively. The age- and sex-adjusted risk ratio for the nicardipine group was 0.973 (0.514 to 1.839, P=0.932) compared with the diuretic group, and the incidence of cardiovascular end points was similar (P=0.923).
The NICS-EH study is the first study to make a direct comparison of a dihydropyridine calcium antagonist and a diuretic. A proper comparison between the two has also become desirable because of decreasing diuretic and increasing calcium antagonist use.10 In this study, a placebo comparison was ruled out by the Steering Committee because of ethical considerations based on the known efficacy of diuretics in the elderly. During the study, results of further placebo-controlled trials indicating that diuretics and calcium antagonists are useful in elderly hypertensive patients became available.3 4 5 6 7 8
The cardiovascular complication rate per 1000 persons per year in the treated groups was 21.0 in the MRC study,5 33.5 in the STOP-Hypertension study,4 23.3 in the Syst-Eur study,7 and 21.4 in the Syst-China study8 ; the values of 27.8 (nicardipine group) and 26.8 (diuretic group) obtained here are similar to these. Although a direct comparison is difficult because the incidence of cardiovascular complications is affected by patient age, sex, blood pressure, atherosclerotic complications before the study, and frequency of left ventricular hypertrophy, the similarities with the values from the previous studies suggest that the incidence of cardiovascular complications in Japan is not different from that in Europe, the United States, and China. The ratio of stroke to myocardial infarction in the NICS-EH study was 4, whereas it was 0.8 to 1.3 in Europe4 5 7 and the United States3 and 5 to 8 in China.6 8 The ratio was higher in Japan and China than in Western countries.
The efficacy of calcium antagonists in preventing cardiovascular complications in elderly hypertensive patients has been reported in the STONE study,6 the Syst-Eur study,7 and the Syst-China study.8 Staessen et al11 showed the usefulness of treatment with calcium antagonist monotherapy in the Syst-Eur study reanalysis. These studies used twice-daily long-acting calcium antagonists and demonstrated that calcium antagonists and diuretic treatment have a similar preventive effect against cardiovascular events; there was greater protection against stroke than against myocardial infarction.
Reports by Furberg et al12 and Psaty et al13 suggested that calcium antagonists increase the myocardial infarction mortality rate but were limited to short-acting agents; a reduction in myocardial infarction rates has been observed for long-acting calcium antagonists.14 Furthermore, despite retrospective case-control studies indicating an increase in cancer rates in patients treated with calcium antagonists,15 16 the incidence of cancer in the NICS-EH study was no different in the calcium antagonists and diuretic groups. Similarly, the incidence of cancer in the STONE and Syst-Eur studies was not significantly different in the treatment and placebo groups. Finally, the NICS-EH study did not show the increased incidence of gastrointestinal hemorrhage indicated by Pahor et al17 ; similarly, no such increase was found in the STONE or Syst-Eur studies.
In the MIDAS study,18 which primarily compared the antiatherosclerotic effects of a calcium antagonist (isradipine) and a diuretic (hydrochlorothiazide), major vascular complications were found in 5.65% of the isradipine group and 3.17% of the hydrochlorothiazide group over 3 years of treatment (P=0.07). Reasons contributing to the difference between these results and those of the NICS-EH study may be the greater decrease in SBP in the diuretic group than in the isradipine group in MIDAS.
Four cases of onset of diabetes and decreased serum sodium levels and increased uric acid and blood urea nitrogen levels that were significantly different from those in the nicardipine group were observed in the diuretic group. In the 3-year SHEP study,19 laboratory values that showed greater variation in the diuretic than the placebo group were blood sugar, triglycerides, uric acid, creatinine, potassium, and HDL-cholesterol. Although these differences between the groups were statistically significant, the low degree of variation indicates that long-term administration of low-dose diuretics produces minor metabolic side effects.
Insufficient numbers of patients were recruited to the NICS-EH study and therefore individual conditions such as stroke and ischemic heart disease could not be analyzed, although no differences in absolute numbers were found. However, in terms of the use of calcium antagonists and diuretics in the elderly, it is believed that the results are valuable because they indicate that there are no clear differences in cardiovascular complications between these drugs. These results should be confirmed in a further large trial. Therefore the results support the observations of JNC-VI, which advocates the use of calcium antagonists as well as diuretics in elderly hypertensive patients.
- Received March 17, 1999.
- Revision received April 16, 1999.
- Accepted June 25, 1999.
Amery A, Birkenhager W, Brixko P, Bulpitt C, Clement D, Deruyterre M, De Schaepdryver A, Dollery C, Fagard R, Forette F, Forte J, Hamdy R, Henry JF, Joossens JV, Leonetti G, Lung-Johansen P, O’Malley K, Petrie J, Strasser T, Tuomilehto J, Williams B. Mortality and morbidity results from the European Working Party on High Blood Pressure in the Elderly Trial. Lancet. 1985;1:1349–1370.
MRC Working Party. Medical Research Council trial of treatment of hypertension in older adults: principal results. BMJ. 1992;304:405–412.
Staessen JA, Fagard R, Thijs L, Celis H, Arabidze GG, Birkenhäger WH, Bulpitt CJ, de Leeuw PW, Collery CT, Fletcher AE, Forette F, Leonetti G, Nachev C, O’Brien ET, Rosenfeld J, Rodicio JL, Tuomilehto J, Zanchetti A, for the Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Randomised double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet. 1997;350:757–764.
Kuramoto K, and the National Intervention Cooperative Study Group. Treatment of elderly hypertensive in Japan: National Intervention Cooperative Study in Elderly Hypertensives. J Hypertens. 1994;12(suppl 6):S35–S40.
Psaty BM, Koepsell TD, Yanez ND, Smith NL, Manolio TA, Heckbert SR, Borhani NO, Gardin JM, Gottdiener JS, Rutan GH, Siscovick DS, Furberg CD. Temporal patterns of antihypertensive medication use among older adults, 1989 through 1992: an effect of the major clinical trials on clinical practice? JAMA. 1995;273:1436–1438.
Staessen JA, Thijs L, Fagard RH, Birkenhager WH, Arabidze G, Babeanu S, Gil-Extremera B, Bulpitt CJ, Davidson C, de Leeuw PW, Efstratopoulos AD, Fletcher AE, Fogari R, Jaaskivi M, Kawecka-Jaszcz K, Nachev C, Petrie JC, Seux ML, Tuomilehto J, Webster J, Yodfat Y. Calcium channel blockade and cardiovascular prognosis in the European Trial on isolated systolic hypertension. Hypertension. 1998;32:410–416.
Furberg CD, Psaty BM, Meyer JV. Nifedipine dose-related increase in mortality in patients with coronary heart disease. Circulation. 1995;92:1326–1331.
Pahor MP, Guralnik JM, Salive ME, Corti MC, Carbonin P, Havlik RJ. Do calcium channel blockers increase the risk of cancer? Am J Hypertens. 1996;9:695–699.
Savage PJ, Pressel SL, Curb JD, Schron EB, Applegate WB, Black HR, Cohen J, Davis BR, Frost P, Smith W, Gonzalez N, Guthrie GP, Oberman A, Rutan G, Probstfield JL, Stamler J, for the SHEP Cooperative Research Group. Influence of long-term, low-dose, diuretic-based, antihypertensive therapy on glucose, lipid, uric acid, and potassium levels in older men and women with isolated systolic hypertension: the Systolic Hypertension in the Elderly Program. Arch Intern Med. 1998;158:741–751.