SYSTEMIC HYPERTENSION Efficacy of Nadolol Alone and Combined with Bendroflumethiazide and Hydralazine for Systemic Hypertension VETERANS ADMINISTRATION COOPERATIVE STUDY GROUP ON ANTIHYPERTENSIVE AGENTS Nadolol (N) titrated from 80 to 240 mg or ben- droflumethiazide (B) 5 to 10 mg, or the combination (B+N), were randomly assigned double-blind to 365 men with pretreatment diastolic blood pressures (BP) of 95 to 114 mm Hg. After 12 weeks of treat- ment, a diastolic BP of <90 mm Hg was achieved in 49% who received N, 46% who received B and 85% who received B+N. With N, the diastolic BP decreased more in whites than in blacks; with B, this racial trend was reversed. Side effects were infre- Beta-adrenergic blocking agents differ with respect to cardioselectivity, intrinsic sympathomimetic activity and membrane-stabilizing effects.1 Nadolol (N) does not exhibit any of these properties,2-? but it has 2 char- acteristics that are important for the treatment of sys- temic hypertension. The first is long duration of action. This permits once-daily dosage with a consequent gain in compliance. The second is that in contrast to other beta-adrenergic blocking agents, nadolol is not associ- ated with a decrease in renal blood flow, a desirable feature especially in patients with hypertension. The present study assesses the relative effectiveness of 3 regimens: N alone, bendroflumethiazide (B) alone®? and B+N combined. In addition, the effectiveness of adding hydralazine was assessed in patients whose blood pressure (BP) was not controlled with one or the other of these regimens. Methods Four hundred eighty men, aged 20 to 69 years, were evalu- ated for randomization out of 809 patients screened, of whom 365 were eventually randomized (Fig. 1). The untreated sitting From the Cooperative Studies Program, Medical Research Service of the Veterans Administration. Supported by a grant from E. R. Squibb & Sons, Inc. Princeton, New Jersey. Manuscript received June 6, 1983; revised manuscript received August 26, 1983, accepted August 30, 1983. Address for reprints: Edward D. Freis, MD, Senior Medical Investi- gator, Veterans Administration Medical Center, 50 Irving Street, N.W., Washington, D.C. 20422. 1230 quent; the most common were impotence, lethargy, weakness and postural dizziness, which occurred more often with B than with N. Addition of hydrala- zine, 25 to 100 mg twice daily, controlled diastolic BP at a level of <90 mm Hg in approximately 60 % of those previously uncontrolled. N, and especially B-tN, provided an efficacious once-daily treatment for systemic hypertension, and addition of hydral- azine was effective in most nonresponders. (Am J Cardiol 1983;52: 1230-1237) diastolic BP (Korotkoff phase V) had to be 95 to 114 mm Hg inclusive. Patients were excluded who had major cardiovas- cular complications, serious systemic diseases or who had preexisting conditions that would interdict the use of the test drugs (see Appendix A). Prerandomization placebo period: The nature of the study was explained to the patient and written informed consent was obtained. In the patients who met the age and diastolic BP criteria for entry and had no exclusion factors, antihypertensive therapy, if any, was discontinued for at least 2 weeks up to a maximum of 8 weeks, depending on the type of drug taken. A history was taken that included volunteered complaints, and a physical examination was performed. The following laboratory studies were obtained: a chest x-ray (if not taken within the previous 3 months), an electrocardio- gram, complete blood cell count, urinalysis, serum potassium fasting blood glucose, uric acid, cholesterol, triglycerides, creatine, alkaline phosphatase, serum glutamic oxaloacetic transaminase, fluorescent antinuclear antibodies and serum bilirubin. Systolic and diastolic (Korotkoff, phase V) BP readings were taken 3 times in the sitting position at each clinic visit and once in the standing position. Readings were taken in the right arm using a standard mercury sphygmomanometer. The median of 3 determinations of BP with the patient sitting was used for analysis. The patient qualified for randomization if the median diastolic BP on 2 successive weekly visits was 95 to 114 mm Hg and if 80 to 110% of the prescribed number of tablets had been taken as estimated by pill counts. A maxi- mum of 4 weekly visits was allowed to fulfill these require- ments. The patient was excluded from the study at any clinic visit if the diastolic BP was >119 mm Hg. Patients were also December 1, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 52 1231 Patients Screened Entering Pre- randomization FIGURE 1. Flow chart of phase A showing numbers of patients screened, entering prerandomization and ran- domized double-blind to the 3 regimens of bendroflumethiazide, nadolol and the 809 480 combined drugs. Also shown are the number terminated, completing the trial and the percentage controlled at dia- stolic blood pressure <90 mm Hg. Phase B (hydralazine) is not shown. 8 - Bendroflumethiazide N - Nadolol GBP - Goal Blood Pressure Completed Randomized B Terminated Study % at GBP 81 13 68° 46% N 365 132 28 104 49% B+N 152 16 136 F— 85% ‘includes one patient terminated at last clinic visit Phase A terminated if the diastolic BP was <95 mm Hg or >114 mm Hg on each of the 2 successive visits. The patient was given 2 bottles that contained placebos and was instructed to take 1 tablet daily from each. He also was requested to return the bottle of remaining tablets to the clinic each visit. A check list of known side effects associated with the administered drugs was reviewed at each visit before as well as after randomization. One hundred fifteen patients were dropped during the prerandomization phase: 62 because the diastolic BP was below the acceptable range (<95 mm Hg) and 7 because the diastolic BP was above the acceptable range (>114 mm Hg); 30 patients were noncompliant, of whom 20 failed to return to the clinic; and 16 patients were dropped for miscellaneous reasons. Postrandomization period (phase A): Of the 365 patients who were randomized into the study, 308 completed phase A. Recruitment goals were met or exceeded in most of the clinics; the hospital with the lowest number of randomizations achieved 94% of its quota. The study was a randomized, double-blind trial in which patients were assigned to 1 or 3 regimens: B plus placebo of N (81 patients), N plus placebo of B (132 patients) or B+N (152 patients). The reason for the unequal randomization is as follows: the patients eligible for entering phase B (addition of hydralazine) were those whose diastolic BP failed to de- crease to <90 mm Hg on N alone, B alone or the combination of B+N. We estimated that the combination would be the most effective in reducing BP and, therefore, would provide fewer patients eligible to receive hydralazine. Consequently, more patients were randomized to the 2-drug regimen so as to provide approximately equal numbers of eligibles for entry into phase B. It was estimated that for phase A only, in order to provide 90% power and a type I error of alpha = 0.05/2 for the 2 com- parisons, a sample size of 60 patients per group would be needed. This was based on the assumption that 50% of pa- tients receiving B or N and 80% receiving the combination would attain the goal diastolic BP of <90 mm Hg. However, larger sample sizes were chosen because of the need to provide sufficient patients for entry into phase B. With an additional allowance for dropouts the number of patients required for randomization was estimated to be 350, or 50 patients per hospital. The patients were assigned to the 3 treatment groups using simple randomization in a ratio of 3:5:6. The randomization was blocked after every 14 patients within each hospital and also across hospitals, i.e., each 2 consecutive patients across 7 hospitals equalled the block of 14. More patients were ran- domized to N than to B to gain more experience with the former drug. The placebos, which appeared identical to the active drugs, were used to maintain the double-blind nature. The initial dose were 80 mg of N and 5 mg of B, each given once daily before breakfast in the morning. Patients were seen in 1 week and were managed as follows: If the diastolic BP was >75 mm Hg, B or its placebo was increased to 10 mg, which dose was continued througout the study; if the diastolic BP fell to <75 mm Hg, the patient was removed from the trial. N was titrated TABLE! Baseline Characteristics of 365 Randomized Patients Bendroflu- methiazide (B) Nadolol (N) B+N No. of pts 81 132 152 Age (yr) 50.94 1.1 49.44 1.0 51.14 0.8 Black 65% 62% 57% White 35% 38% 43% Weight (kg) 196.7441 197843.7 1920427 Blood pressure (mm Hg) (Standing) Systolic 146.741.7 1455414 148.94 1.4 Diastolic 103.7408 1033406 1049406 Blood Pressure (mm Hg) (Sitting) Systolic 146.9416 144.7412 148.4413 Diastolic 1018406 1013404 1018+0.4 Heart rate 77.14 1.1 76.2+0.9 76.4 + 0.9 (beats/min) Uric acid (mg/dl) 6.0 + 0.2 6.4+0.1 6.2 + 0.2 Serum potassium 4.2+0.0 42+ 0.0 4.3 + 0.0 (mEq/liter) Creatinine (mg/dl) 1.2+0.0 1.140.0 1240.0 Fasting blood sugar 99 + 2.0 100 + 2.0 97 + 1.0 (mg/dl) Cholesterol (mg/dl) 233 + 7.0 220 + 4.0 223 + 4.0 Triglycerides 171+ 14.0 176411.0* 147+ 8.0* (mg/dl) * Significance of difference <0.05. Values are mean + standard error of the mean. 1232 NADOLOL IN HYPERTENSION TABLE Il Mean Changes in Sitting Blood Pressure (BP) and Heart Rate in Blacks and Whites After 12 Weeks of Treatment Variable Bendro Nadolol Combination Significance Number (blacks/whites) 68 (42/36) 104 (61/43) 136 (75/61) Attained goal blood pressure 46% 49% 85% B—C?, N—C? Blacks 46% 31% 84% B—ci, N-C? Whites 46% 77% 85% B-Nt, B—Ct p Value NS <0.001 NS Baseline systolic BP (mm Hg) 146.8 + 1.7 144.14 1.4 147.74 1.3 Blacks (178) 148.6 + 2.2 145.14 1.8 149.64 1.9 Whites (130) 143.8 + 2.6 142.74 2.1 145.5419 Change systolic BP (mm Hg) —17.4417 —10.5+4 1.6 —26.3 + 1.4 B—-N‘, B—C?, N—C Blacks —19.94+2.4 —5.8+4 2.1 —27.342.1 B—-N?, B—C*, N—CT Whites —13.3 + 2.0 —17.242.3 —22.941.7 N—-C*, B-Ct Baseline diastolic BP (mm Hg) 10104 0.6 101.44+ 0.4 10164 0.4 Blacks (178) 101.2+0.8 101.24 0.5 101.9 + 0.6 Whites (130) 100.5 + 0.8 101.6 + 0.7 101.2 + 0.6 Change diastolic BP (mm Hg) —11.6+ 1.2 —12.140.8 —17.9+40.7 B—ct, N—Ct Blacks —12.4+ 1.5 —96140.9 —18.1+ 1.0 B—C?, N—Ct Whites —10.2 + 1.7 —15.6+ 1.2 —17.74+08 B-NT, B—ct Baseline pulse rate (beats/min) 76.3 + 1.1 75.7 + 1.0 75.440.8 Change pulse rate (beats/min) 0.8 + 1.4 —16.1+4 1.0 —15.8+0.8 B-N!, B—C? *p <0.05. Tp <0.01. tp <0.001. B-N = significance of the difference between bendroflumethiazide (B) and nadolol (N); B—C = significance of the difference between B and combination (C); N-C = significance of the difference between N and C; NS = not significant. as necessary biweekly until goal diastolic BP, defined as <90 mm Hg, was achieved. The once-daily doses of N or its placebo were increased from 80 to 160 to 240 mg. After attaining goal diastolic BP, each regimen was then continued at the same dosage until the 12th week after randomization. If the diastolic BP was >119 mm Hg at any clinic visit or >104 mm Hg at 2 successive clinic visits during this phase of the study, the pa- tient was terminated from the study. These patients were removed from the trial and were treated openly. They did not enter phase B. The duration of phase A was 12 weeks and in- cluded initially 4 visits at 1-week intervals followed by 4 bi- weekly visits. Postrandomization period (phase B): The effects of adding hydralazine to the treatment regimens of patients who failed to achieve the goal diastolic BP of <90 mm Hg during phase A was assessed at completion of phase B, Hydralazine was added in an initial dose of 25 mg twice daily, but was in- creased to 50 mg and then 100 mg twice daily until either the diastolic BP fell to <90 mm Hg or intolerable side effects su- pervened. The duration of phase B was 9 weeks. Patients were seen at 1 week for the first week only and then were scheduled for biweekly visits. TABLE Ill Terminations During Phase A Treatment Period Termination Gause Bendro* Nadolol Combination DBP elevated’ 9 9 0 Dropouts 3 9 7 Lapse in treatment 0 4 3 Drug intolerance 0 2 4 Cardiovascular complication 1 1 1 All other 0 3 1 Total 13 28 16 No. randomized 81 132 152 Percent terminations 7 21 11 * Bendroflumethiazide, 5 to 10 mg/day. T DBP >119 mm Hg at any visit, DBP 114-119 mm Hg on 2 suc- cessive weekly visits during titration or > 104 mm Hg on 2 successive visits after maximum titration. DBP = diastolic blood pressure. Characteristics of randomized patients: The mean BP at the time of randomization was 146.7/101.6 mm Hg and did not differ significantly among the treatment groups (Table I). The mean age was 50.4 years. The racial distribution was 61% black and 39% white. There were no significant differ- ences in these characteristics among treatment groups or in heart rate and the various blood chemistry values except tri- glycerides, which averaged lower (p <0.05) in the group that received both drugs (Table I). Statistical analysis of results was carried out using the 2- sample ¢ test to compare mean values between independent samples. The comparison of percentages between independent samples was accomplished using the Z test based upon the normal distribution. Comparison of changes within patients was done using the paired t test. Results Changes in blood pressure during phase A: The percentage of patients who achieved goal BP (defined as a diastolic BP <90 mm Hg) at the last or 12th week of treatment was determined (Table II, Fig. 1). In the patients treated with N alone who either completed the 12-week treatment period or else were terminated be- cause of elevated diastolic BP, 49% were controlled, 44% TABLE IV_ Leading Side Effects in Phase A* % Complaining Complaint Any Visit Bendro Nadolol Both Drugs Weakness 3 0 1 Lethargy 6 2 6 Impotence 9 4 2 Postura! dizziness 3 0 2 Insomnia 0 3 1 * Complaint made on at least 2 visits during Phase A but not during placebo baseline period. Bendro = bendroflumethiazide. December 1, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 52 1233 were not controlled and 7% had to be terminated for elevated BP during the treatment period. With B, 46% were controlled, 43% were not controlled and 11% had to be terminated for BP above the acceptable range. The combination of the 2 drugs was significantly more effective (p <0.001) than either of the single drug regi- mens, with 85% controlled, only 15% uncontrolled and no terminations because of high BP. In the patients receiving N alone, 31% achieved goal BP with 80, 10% with 160 and 13% with 240 mg/day. In the group receiving the combination, which included 10 mg of B, 46% attained goal BP with the 80-mg dose of N, 29% with 160-mg dose and 10% with the 240-mg dose. Forty-three percent of the patients who responded to B alone did not achieve goal blood pressure immediately after taking the 10-mg dose, but required several more weeks before this dose decreased the diastolic BP to <90 mm Hg. The average changes in BP, which includes only the patients who completed phase A of the trial, were as follows: of the 104 patients assigned to N, the BP aver- aged 144.1/101.4 during the prerandomization period and 133.6/89.3 mm Hg by the end of the 12-week treatment period, a reduction of 10.5/12.1 mm Hg (Table II). The average BP of the 68 patients who re- ceived B decreased from 146.8/101.0 mm Hg before randomization to 129.4/89.4 mm Hg at the end of the treatment period, a reduction of 17.4/11.6 mm Hg. The reduction in systolic but not diastolic BP was signifi- cantly greater with B than with N (p <0.01). With the combination of B+N, the BP averaged 147.7/101.6 mm Hg before treatment and 122.4/83.7 mm Hg at the end of treatment, an average reduction of 25.3/17.9 mm Hg. The reductions in both systolic and diastolic BP were significantly greater with the combination than with either of the drugs given alone (p <0.001/p <0.001). A racial difference was observed in the response to N (Table IT). In white persons the average decrease in di- astolic BP was 15.6 mm Hg, significantly (p <0.001) greater than the 9.6 mm Hg average reduction attained in black persons. In the patients treated with N alone, 77% of whites achieved a diastolic BP <90 mm Hg, compared with only 31% of blacks (p <0.001). By con- trast, the diastolic BP response of the blacks to the thiazide diuretic was somewhat, but not significantly, greater than the response of the whites (12.4-mm Hg reduction in blacks and 10.2-mm Hg in whites). The greater reduction of systolic BP to B in blacks (19.9 mm Hg) compared with whites (>13.3 mm Hg) was almost significant (p = 0.055). There was essentially no racial difference in the response of diastolic BP to the com- bination of the 2 drugs, with average reductions of 17.7 mm Hg in whites and 18.1 mm Hg in blacks. The degree of reduction of diastolic BP was correlated with the height of the baseline diastolic BP in that the higher the baseline diastolic BP, the greater the de- crease. For example, in patients with a pretreatment diastolic BP of 95 to 99 mm Hg, the reduction of dia- stolic BP averaged 8.8 mm Hg with N, 8.2 mm Hg with B and 16.8 mm Hg with the combination. In contrast, the reductions of diastolic BP in patients with baseline levels of 110 to 114 mm Hg averaged 19.3 mm Hg with N, 23.5 mm Hg with B and 24.1 mm Hg with the com- bination. To assess the effects of age in the response to the various regimens, the patients were subdivided into 2 age groups, those age 50 years or less and those older than 50 years. The mean reductions in diastolic BP were almost identical in the 2 groups. Pulse rate did not change significantly with B alone; the average pulse rate increased, but only by 0.8 beats/ min. The average pulse rate decreased significantly from baseline, by 16.1 beats/min with N alone and by 15.8 TABLE V_ Changes in Serum Chemistry Values After 12 Weeks of Treatment Serum Chemistry Bendro Nadolol Both Drugs Potassium (mEq/liter ) No. of patients 68 99 134 Baseline 4.26 + 0.05 4.26 + 0.04 4.28 + 0.03 Change —0.57 + 0.06 0.08 + 0.04 —0.44 + 0.05 Uric acid (mg/dl) No. of patients 63 97 126 Baseline 6.7 + 0.2 6.4+0.1 6.5+0.1 Change 1.7 + 0.2t 0.4 + 0.1* 1940.17 Fasting glucose Number of patients 67 97 133 Baseline 100.6 + 2.0 103.0 + 1.9 97.2413 Change +6.1+ 2.17 +2.4+ 1.8 +7.44 1.17 - Cholesterol No. of patients 60 88 121 Baseline 234.9 + 8.0 223.6 + 5.0 227.2 + 4.9 Change 11.54 4.3t —1.54+ 3.9 3.5 + 3.6 Triglycerides No. of patients 66 94 132 Baseline 169.6 + 14.2 172.3 + 11.6 149.34 8.3 Change 34.6 + 14.8* 38.7 + 13.2* 67.8 + 11.9 Values are mean + standard error of the mean. Significant changes from baseline: *p <0.01. Tp <0.001. Bendro = bendroflumethiazide. 1234 NADOLOL IN HYPERTENSION beats/min with the combination (p <0.001). Body weight decreased on the B and B+N regimens. At the second visit after randomization, when most patients had received their maximal dose, the mean reductions were 3.0 and 2.8 pounds body weight, respectively, for B alone and for the B+N. Body weight did not change in the patients taking N alone. Terminations: Thirteen (16%) of the randomized patients receiving B, 28 (21%) of those receiving N and 16 (11%) of those receiving both drugs were terminated from the study (Table IIT). Nine patients receiving each of the single drug regimens were terminated because of an elevated diastolic BP. None of the patients receiving both drugs were terminated for this reason. Two pa- tients receiving N, 4 receiving both drugs and none re- ceiving B were terminated because of suspected drug intolerance. Side effects: The only complaints for each patient that were considered as possibly drug-related were those that were not manifest during the prerandomization placebo period. To be counted as a side effect, the complaint also had to be registered on more than 1 clinic visit during the drug treatment period. With these cri- teria, subjective side effects were relatively few (Table IV). The most frequently noted complaint was sexual impotence. This occurred with all regimens, but most frequently with B alone (9% of patients). Also, the complaints of weakness, lethargy and postural dizziness, while relatively infrequent were associated mostly with the thiazide-containing regimens. An exception was insomnia, which occurred in 3% of the patients receiving N, none receiving B and 1% of patients receiving B+N. Serum chemistries: Changes in serum chemistries reflected primarily those usually associated with the thiazide diuretics (Table V). Serum potassium de- creased significantly and serum uric acid increased significantly (both p <0.01) with B and B+N, but they remained essentially unchanged with N alone. Fasting serum glucose increased by an average of 6.0 and 7.4 mg/dl on B and B+N, respectively, and remained es- sentially unchanged after treatment with N alone. There was no significant change in serum creatinine with any of these regimens. Baseline triglyceride levels averaged 149.3 mg/dl in patients receiving the combination of drugs, compared with 169.6 and 172.3 mg/dl for the B and N groups, re- spectively. Serum triglycerides increased significantly, by 34.6 mg/dl (20%) with B, 38.7 mg/dl (23%) with N and 67.8 mg/dl (45%) with the combination of drugs. Serum cholesterol increased 11.5 mg/dl (4.9%) after B alone. Serum cholesterol averaged 1.5 mg/dl lower in patients receiving N alone. With B+N, it increased 3.5 mg/dl (1%). Changes during phase B, addition of hydralazine: The number of nonresponders (failure to achieve a di- astolic BP <90 mm Hg) during phase A who entered phase B, when hydralazine was added, included 30 re- ceiving B, 40 receiving N and 19 receiving the combined drugs (Table VI). The addition of hydralazine resulted in similar decreases in diastolic BP, averaging 7.5 mm Hg with either B or N alone and 7.7 mm Hg with the combination. The percentage of these previously un- controlled patients who attained a diastolic BP of <90 mm Hg after the addition of hydralazine was 57% in the patients receiving B alone, 68% in the N-treated pa- tients and 58% of those receiving the combination of these drugs. Thus, hydralazine was effective in more than half of the previously incompletely controlled patients. In contrast to the diuretic and beta blocker, there were no racial differences in the response to hy- dralazine. Heart rate increased by an average of 5.6, 2.4 and 4.1 beats/min after hydralazine was added to B, N and the combination, respectively (Table VI). Terminations from the study because of side effects were few. One patient receiving hydralazine with B re- quested discontinuation because of impotence. Four patients receiving N with hydralazine and 1 patient receiving all 3 drugs were terminated because of head- ache. There were no other terminations associated with drug intolerance. Discussion In designing the trial, care was taken to minimize known sources of bias. The double-blind nature was maintained as much as possible, with each drug and its placebo identical in appearance. Possible carryover effects from the prior regimen that may occur with crossover designs were avoided by using parallel treat- ment groups. The randomization procedure was suc- cessful in preventing significant differences between treatment groups with respect to any of the important prerandomization characteristics. The sample size quotas were met on time and with no great differences in recruitment among the various hospitals. All of the randomized patients were tested for compliance, and on the basis of tablet counts, all ingested >80% of the placebos prescribed during the prerandomization pe- riod. The results reported, therefore, may be better than the general experience because identified noncompliant patients were excluded. The 2 drugs given as single entities had approxi- mately the same effectiveness. However, the combina- tion of the 2 drugs was considerably more efficacious than either drug used alone. The percentage of patients whose diastolic BP was controlled <90 mm Hg (goal diastolic BP) was significantly greater with the com- bined drugs than with either agent used alone. In the group receiving the combination, 85% achieved goal diastolic BP, compared with 49% with N alone and 46% with B. This result is similar to that in a previous trial by our group.’ In that study, propranolol alone was compared with propranolol plus hydrochlorothiazide in patients with mild hypertension. Propranolol con- trolled the diastolic BP in 52% of these patients, whereas with the combination, 81% attained goal diastolic BP. The impressive results using the combined drugs should not negate the fact that N alone controlled BP in half of the patients, indicating that it is a highly effective treatment for hypertension, although not as effective as the N-diuretic combination. B produced a somewhat greater fall in systolic BP than N. This greater effect of the diuretic compared with the beta blocker on systolic BP was also found in December 1, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 52 TABLE VI 1235 Mean Changes in Sitting Blood Pressure (BP) and Heart Rate After 9 Weeks of Added Hydralazine Hydralazine, 25-100 mg b.i.d., plus Variable Bendro Nadolol Combination Significance No. of patients 30 Systolic BP (mm Hg) prehydralazine Change systolic BP (mm Hg) Diastolic BP (mm Hg) prehydralazine Change diastolic BP (mm Hg) Pulse rate beats/min prehydralazine Change pulse rate (beats/min) * p <0.05. t p <0.001. Fp <0.01. = lo lo abe NG FS hk ek —_ J ~ ce C2O OM HH HEHE OEE ier 1 3 3 5 2 6 = lap © hd a nel eel a ad RO MoO NO a 40 19 B+H*,N+H,B+N =n 4N ou = lo lo AN NO DW =n NO ©O® HH HH HH Pw om: NN =O OF Bt+Ht,N+H,B+N =20 bt ot B+ Ht HH OHH OEE BA oO oD — B + H = significance of change after adding hydralazine to bendroflumethiazide (Bendro); N + H = significance of change after adding hydralazine to nadolol; B + N + H = significance of change after adding hydralazine to the combination of B + N. a previous study.® Although the reason has not been clarified, it seems likely that the reduction in plasma volume and tendency to a somewhat low cardiac output may be important factors. Although the population was predominantly black, the randomized group included 142 white patients, which was sufficient to make valid black-white com- parisons in responsiveness to the various treatments. N was significantly more effective as an antihyperten- sive agent in whites than in blacks. The reverse was found with B, which was somewhat, although not sig- nificantly, more effective in blacks than in whites. Similar black-white differences in the antihypertensive response to beta blockers and to diuretics were found in the Veterans Administration Cooperative Study comparing propranolol with hydrochlorothiazide.9 Seedat!° also observed that a diuretic was more effective than a beta blocker in blacks.!° Hypertensive blacks are said to exhibit higher plasma volumes and lower plasma renin activities than hypertensive whites,!!:!2 although other investigators have disputed these claims.!3:!4 Laragh postulated that patients with high plasma vol- umes have a “volume-dependent” hypertension that will respond to reduction of extracellular and plasma volume with diuretics, while those with high plasma renin activity and low plasma volumes should respond to beta blockers.16 Few side effects were noted with either drug in this trial. The most frequent complaints were impotence, lethargy, weakness and postural dizziness. These side effects were encountered more often with B than with N, although they were uncommon with both drugs. In the prior trial of propranolol versus hydrochlorothia- zide,® subjective side effects also were uncommon. The most frequent hydrochlorothiazide-associated com- plaints in that study were diarrhea, impotence, consti- pation and numbness, and the most frequent side effects among propranolol-treated patients were in- somnia, swelling of the hands and vivid dreams. How- ever, as in the present trial, the number of possibly drug-related complaints was relatively small. Addition of hydralazine to the patients who failed to reach goal diastolic BP with B, N or both resulted in a similar decrement of blood pressure in each of the 3 treatment groups. The average additional reduction was about 7.5 mm Hg in all 3 treatment groups. This re- sponse is similar to that achieved in a previous Veterans Administration Cooperative Study.!® In the latter study, hydralazine was added to the regimen of patients who failed to achieve a diastolic BP of <90 mm Hg with hydrochlorothiazide alone. These patients had an ad- ditional average decrease in diastolic BP of 8.8 mm Hg 3 months after adding hydralazine. Although these re- ductions may seem small, the BP had already been re- duced, although not to goal levels, by the original ther- apy. Furthermore, the lower the level of BP the less will be the reduction following an antihypertensive drug. For example, in phase A of the present trial the reduc- tion in diastolic BP after B averaged 8.2 mm Hg in pa- tients with pretreatment baseline levels of 95 to 99 mm Hg, 11.6 mm Hg with 100 to 104 mm Hg diastolic BP prerandomization and 18.6 mm Hg with entry diastolic BP of 105 to 114 mm Hg. If the diastolic BP had not already been partially reduced by the initial treatment, the decrease associated with hydralazine might have been considerably greater. Side effects during hydralazine administration were not impressive. The most frequent side effect was moderately severe to severe headache, which occurred in 4 patients receiving N and 1 receiving the combina- tion. Headache of this severity was not noted in the groups receiving hydralazine and B alone. Although the incidence of headache was too low to make firm con- clusions, these results suggest that thiazide diuretics may prevent hydralazine-induced headache, possibly by reducing plasma and extracellular volume. Except for headache, the side effects complained of most fre- quently during hydralazine treatment were the same as those present before the drug, including lethargy, weakness and impotence. Elevation of serum triglyceride levels after either thiazide diuretics of beta blockers have been noted previously by other investigators.!718 The increase was especially marked after the combined drugs, when the increase averaged 47% above baseline values. The clinical importance of this change is not clear, however, 1236 NADOLOL IN HYPERTENSION because of the role of triglycerides in the pathogenesis of atherosclerosis is not well defined. Serum cholesterol increased modestly after B but not after N. These ob- servations with respect to serum cholesterol are also similar to those reported by others.!7:19 In conclusion, as judged by the results of this trial, N appears to be a safe and effective antihypertensive agent. Its long action permits once-daily dosage, which should facilitate compliance and offer an advantage over shorter-acting beta-adrenergic blocking drugs. The present results suggest that approximately half of the patients with mild and moderate hypertension will achieve a diastolic BP of <90 mm Hg with N alone, a further one-third with the addition of B and 10% more with the addition of hydralazine. Thus, approximately 85% of patients can be controlled by a relatively simple step-care regimen involving once-daily doses of 1 or 2 agents, with the third drug, hydralazine, reserved for the small percentage of nonresponders. Also, because of the differing racial response to these 2 agents, it would ap- pear advisable to initiate treatment with B in blacks and with N in whites. References 1. McDevitt DG. Position paper: differential features of beta-adrenoreceptor blocking drugs for therapy. In: Laragh JH, Buhler FR, Seldin DW, eds. Frontiers in Hypertension Research. New York: Springer-Verlag, 1981: 473-481. 2. Heel RC, Brogden RN, Pakes GE, Speight TM, Avery GS. Nadolol: a review of its pharmacological properties and therapeutic efficacy in hypertension and angina pectoris. Drugs 1980;20:1-23. 3. Frishman WH. Drug therapy. Nadolol: a new G-adrenoreceptor antagonist. N Engl J Med 1981;305:678-682. 4. Hollenberg NK, Adams DF, McKinstray DN, Williams GH, Borucki LJ, Sullivan JM. B-adrenoreceptor-blocking agents and the kidney: effect of nadolol and propranolol on the renal circulation, Br J Clin Pharmacol 1978;7:suppl 2:219S-225S, 5. Textor ST, Fouad FM, Bravo EL, Tarazi RC, Vidt DG, Gifford RW Jr. Re- distribution of cardiac output to the kidneys during oral nadolol adminis- tration. N Engl J Med 1982;307:60 1-605. 6. Medical Research Council Working Party on Mild to Moderate Hypertension. Adverse reactions to bendroflumethiazide and propranolol for the treatment of mild hypertension. Lancet 1981;2:539-543. 7. Berglund G, Anderson O. Beta-blockers or diuretics in hypertension? A six-year follow-up of blood pressure and metabolic side effects. Lancet 1981;1:744-747. 8. Veterans Administration Cooperative Study Group on Antihypertensive Agents. Propranolol in the treatment of essential hypertension. JAMA 1977;237:2303-2310. 9. Veterans Administration Cooperative Study Group on Antihypertensive Agents. Comparison of propranolol and hydrochlorothiazide for the initial treatment of hypertension. Results of short-term titration with emphasis on radical differences in response. JAMA 1982;248: 1996-2003. 10. Seedat YK. Trial of atenolol and chlorthalidone for hypertension in black South Africans. Br Med J 1980;281:124 1-43. 11. Chrysant SC, Danisa K, Kem DC, Dillard BL, Smith WJ, Frohlich ED. Racial differences in pressure volume and renin interrelationships in essential hypertension. Hypertension 1979; 1: 136-141. 12. Mitas JA Il, Holle R, Levy SB, Stone RA. Racial analysis of the volume-renin relationship in human hypertension. Arch Intern Med 1979; 139:157- 160. 13. Messerli FH, DeCarvalho JG, Christie B, Frohlich ED. Essential hyper- tension in black and white subjects. Hemodynamic findings and fluid volume state. Am J Med 1979;67:27-31. 14. Holland OB, Gomez-Sanchez C, Fairchild C, Kaplan NM. Role of renin classification for diuretic treatment of black hypertensive patients. Arch Intern Med 1979; 139:1365- 1370. 15. Laragh JH. Vasoconstriction-volume analysis for understanding and treating hypertension: the use of renin and aldosterone profiles. Am J Med 1973; 55:261-274. 16. Veterans Administration Cooperative Study Group on Antihypertensive Agents. Comparison of prazosin with hydralazine in patients receiving hy- drochlorothiazide. A randomized, double-blind clinical trial. Circulation 1981;64:772-779. 17. Ames RP, Hill P. Elevation of serum lipid levels during diuretic therapy of hypertension. Am J Med 1976;61:748-757. 18. Shaw J, England JD, Hau AS. Beta-blockers and plasma triglycerides (letter). Br Med J 1978; 1:986. 19. Grimm RH Jr, Leon AS, Hunninghake DB, Leng K, Hannan P, Blackburn H. Effects of thiazide diuretics on plasma lipids and lipoproteins in mildly hypertensive patients: a double-blind controlled trial. Ann Intern Med 1981;94:7-11. Appendix Exclusions 1. Known adverse reactions to hydrochlorothiazide, beta- blocking agents or hydralazine 2. Malignant hypertension including hypertensive neuro- retinopathy 3. Hypertensive retinopathy (K-W scale) greater than grade II . Acute hypertensive encephalopathy . Cerebral or subarachnoid hemorrhage . Atherosclerotic stroke within the past six months . Myocardial infarction within 6 months or angina pectoris greater than New York Heart Association class II 8. Patients currently taking ‘“digitalis-like” preparations 9. Patients with primary valvular heart disease (e.g., rheu- matic or congenital) 10. Atrial fibrillation 11. Heart block greater than 1st degree or Wolff-Parkin- son-White syndrome or, if not currently receiving beta- blocking agent, sinus bradycardia (<60 beats/min) 12. Patients with Raynaud’s disease or symptomatic and objective peripheral vascular disease 13. Asthma 14. Cor pulmonale due to obstructive lung disease Obstructive lung disease with asthmatic wheezes 15. Diabetes requiring treatment other than diet 16. Collagen vascular disease 17. Surgically curable forms of hypertension—pheochro- mocytoma, primary aldosteronism, Cushing’s disease or renovascular hypertension 18. History or evidence of psychiatrically documented nonsituational, clinically important mental depression 19. Malignancy including leukemia and lymphoma 20. Drug abuse, severe organic brain damage or severe alcohol abuse , 21. Patients on adrenergic augmenting psychotropic drugs including monoamine oxidase inhibitors, amphetamine . and its derivatives 22. Patients regularly using transcendental meditation, biofeedback relaxation and/or similar techniques 23. Serum creatinine >2.0 mg/dl 24. Congestive heart failure as evidenced by at least 2 of the following: A. Recent dyspnea or orthopnea not of pulmonary origin B. Ventricular diastolic gallop (S3) C. Basal pulmonary rales D. Cardiothoracic ratio greater than 0.5 on x-ray 25. Patient unreliable 26. Patient unable or unwilling to participate or refuses to sign the informed consent NO oe Participants Co-Chairmen: Edward D. Freis, MD (Washington, D.C.) and J. R. Thomas, MD (Memphis, TN) Principal Investigators: Frederick N. Talmers, MD (Allen Park, MI); William C. Cushman, MD (Jackson, MS); Harold Schnaper, MD (Birmingham, AL); Thomas J. White, MD (Memphis, TN); Orlando Fernandez, MD (Miami, FL); Eli A. Ramirez, MD (San Juan, PR); Ibrahim Khatri, MD (Washington, D.C.) Nurses: Barbara Gregory, RN, and Madeline Metcalfe, RN (Washington, D.C.); Chris Grant, RN, and Julie Pawelak, RN (Allen Park, MI); Pauline Derrington, RN (Jackson, MS); December 1, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 52 1237 Anita McKnight, RN, Susan Reece, RN, and Kristina Grossman, RN (Memphis, TN); Mary H. Smith, RN, and Eileen Haran, RN (Miami, FL); Maria Natal, RN (San Juan, PR); William Hackett, RN, and Donald Quinn, PA (Bir- mingham, AL) Biostatistician: Thomas J. Tosch, PhD Forms Reviewers: Janice Ivie (Memphis, TN); Mary Ellen Vitek and Jane Foregger (Hines, IL) Central Research Pharmacist: Larry Young, RPh Operations Committee: Walter Kirkendall, MD, James C. Gunnels, MD, C. Mort Hawkins, DS Consultants: Barry J. Materson, MD, John C. Alexander, MD, Joseph Meyer, PhD, Dionisio L. Caloza, Jr, MD Hines Cooperative Studies Program Coordinating Center Human Rights Committee: Jennie McKoy; Patrick Moran; Mary Davidson, PhD, Kenneth Elmer, Rev. Martin Feldbush Cooperative Studies Program Central Administration: James A. Hagans, MD, PhD, and Ping Huang, PhD (VA Central Office, Washington, D.C.); William G. Henderson, PhD, Janice Ivie, Mary Ellen Vitek (Cooperative Studies Program Coordinating Center, Hines, IL); Mike Sather, RPh, MS (Cooperative Studies Program Research Pharmacy Coordinating Center, Albuquerque, NM) Reprinted from the December 1 issue of The American Journal of Cardiology, A Yorke Medical Journal. Published by Technical Publishing, a Division of Dun-Donnelley Publishing Corporation, a Company of The Dun & Bradstreet Corporation, 875 Third Avenue, New York, N.Y. 10022. Copyright 1983. All rights reserved. Printed in the U.S.A.