230 Reprinted front the Archives of Intertsal Med@ce Azlgust 1962, Vol. 110, pp. 230-236 Copyright 1962, by Avnerican Medical AssociatioFc Control Study of i VETERANS ADMINISTRATION = I Antihypertensive Agents I COOPERATIVE STUDY ON ANTIHYPERTENSIVE AGENTS= Es I z WASHINGTOh', D. C. E III. Chlorothiazide Alone and in Combination with I Other Agefits; Preliminary Results 8 E E E s The previous 2 reports of the Veterans `fbc criteria used for selecting patients and for classifying severity have been described in a pre- vious communication.' The Group A patients enter- ing the study for the first time were treated as follows : Administration Cooperative Study on Anti- hypertensive Agents described the results ob- tained with reserpine, hydralazine, and 3 ganglion blocking drugs.1*2 The present in- vestigation is concerned with the evaluation of chlorothiazide both alone and in combina- tion with the agents listed above and with cryptenamine, an alkaloid derived from Ve- ratrum viride. Although these latter studies are still in progress, the results seemed of sufficient interest to warrant a preliminary report at this time. Methods This investigation used 2 separate protocols for each of 2 different groups of patients. The first (Group A) were untreated cases entering the study for the first time. The second (Group B) were patients who had been under treatment in the co- operative study for at least 1 year,l*' after which chlorothiazide was added to the regimens of the patients taking active medications. Submitted for publication Feb. 26, 1962. * Participants : Mark L. Armstrong, M.D. ; Mas- simo Calabresi, M.D.; Loyal L. Conrad, M.D.; Harold Dodge, M.D. ; E. E. Eddelman, M.D. ; Edward D. Freis, M.D. (Chairman) ; Rudolph E. Fremont, M.D.; Walter M. Kirkendall, M.D.; David Littman, M.D. ; Clifford G. Pilz, M.D. ; Eli A. Ramirez, M.D. ; David W. Richardson, M.D., and John H. Williams, Jr., B.A. These included the following Veterans Adminis- tration Hospitals : Birmingham, Ala. ; Brooklyn ; Chicago West Side; Iowa City; Oklahoma City; Richmond, Va. ; San Juan, P.R. ; Seattle; Wash- ington, D.C.; West Haven, Conn., and West Rox- bury, Mass. Mild and moderately severe cases were assigned one of the following 4 regimens using a double blind, randomization method : (1) chlorothiazide 500 mg. with reserpine 0.25 mg. combined in a single tablet twice daily plus hydralazine 25 mg. 4 times daily for 1 week followed by hydralazine 50 mg. 4 times daily thereafter ; (2) chlorothiazide without reserpine plus hydralazine, both drugs given as above ; (3) chlorothiazide without reserpine plus placebo of hydralazine, given as above ; (4) placebos of chlorothiazide-reserpine and of hydralazine, given as above. Severe cases were assigned one of the following regimens : ( 1) chlorothiazide 500 mg. combined with reserpine 0.25 mg. in a single tablet twice daily: in addition, hydralazine 12.5 mg. (r/i tablet) 3 times daily were given and the doses raised by 12.5 mg. increments to a total dose of 50 mg. 3 times daily or until the diastolic pressure fell below 100 mg. Hg, or else side-effects supervened ; (2) same as above except that cryptenamine (an alkaloid de- rived from Veratrum viride) was substituted for hydralazine. Each tablet of cryptenamine contained 3 mg., and the initial dose was 1.5 mg. (l/i tablet) 3 times daily. The tablets of hydralazine and the Veratrum compound were made up to appear and taste the same. Chlorothiazide, reserpine, and hydralazine also were prepared to resemble their respective placebos. As in the previous study1 3 different code numbers were assigned to each preparation including the placebos. The Group B patients were those who had been under treatment for more than 1 year with either reserpine, reserpine plus hydralazine, placebos, or reserpine plus ganglion blocking agents.`va The 134 .s'TUDY OF ANTIHYPLRTENSlVk AGENTS 231 TABLE I.--Croup A: Cllange in Blood Pressure from Pretreatment Hospital Averaac to Third-Muntlt Hnnle Pressure by Regimen and Severity Severity :XKI RegimeD Mild and modwltely SeTwe Cl~lorotbi:~zide+resergine+ hydr:&lazine Chlon,thi;lzide+hydralazine Chlorothiazide rlcitcetm Severe ClllorotIliilzide+resergine+ cryptenxmine Chlorotlii;lzide+reserpine+ hydr:rlazine No. of Cases 23 37 34 32 15 25 Systolic Dinstolic rretre:lt. Level DifTel-enw xt 3 MO. _--- 0 90 Confidence AV Limit 163 Ii9 186 -17.2 -10.4 to -24.0 -11.2 - 5.7 to -16 7 - 3.2 + 0.8 to - 7.2 - 0.9 + 5.2 to - 7.0 -17.9 - 6.2 to -29.6 -22.7 -14.2 to -31.2 cases had been divided so that all patients with All of the patients with severe hypertension in mild and half of those with moderately severe hy- prior treatment Group B and half of those with pertension received the rescrpine, hytlralazine, and moderately severe hypertension had been treated placebo regimens. The addition of chlorothiazide with reserpine 0.25 mg. twice daily plus 1 of 3 was accomplished in these cases by substituting for ganglion blocking agents : mecamylamine, chlori- the 0.25 mg. reserpine tablet a specially prepared sondamine, or pentolinium tartrate."' Chlorothiazide tablet containing reserpine 0.25 mg. pius chloro- was added to their regimens by substituting a tab- thiazide 500 mg. A similar-appearing placebo also let containiq 500 mg. of chlorothiazide and 0.25 was manufacturetl. These tablets were given the mg. of reserpine for the prior 0.25 mg. of reserpine. same code number identifications as used in the This tablet was administered twice daily. At the prior study except that the letter "C" preceded same time the dose of the ganglion blocking agent the series of digits. By substituting the "C" series was decreased to half the prechlorothiazide level medication for the prior "A" series those patients because of the well-known enhancement of the taking either reserpine plus hydralazine or reserpine antihypertensive effect of blocking drug induced by plus placebo of hydralazine had chlorothiazide 500 chlorothiazide. Further adjustments of the doses mg. twice daily added to their regimens, while those of the ganglion blocking agent were then made from patients who were not treated with active prepara- this reduced level as seemed indicated fpr control tions had only placebo of chlorothiazide added. of the blood pressure. CHLOROTHIAZIDE + RESERPINE + HYDRALAZINE PERCENT OF CASES CHLOROTHIAZIDE + HYDRALAZINE 30 r Difference nt 3 MO. Pretrent. 0.90 Confidence Level AV. Limit 108 -14.3 - 9.8 to -18.8 105 -11.X - 7.9 to -15.7 103 - 1.9 + 1.2 to - 5.0 105 0.0 + 3.7 to - 3.7 119 -16.4 - 9.3 to -23.5 118 -20.9 -15.8 to -26.0 PERCENT OF CASES CHLOROTHIAZIDE PLACEBO 30 r f Fig. l.-Percentage distribution curves of changes in systolic blood pressure at the end of 3 months of treatment in patients with mild and moderate hypertension. Changes in systolic l,ressure from pretreatment values are listed on the abscissae and percentage of patients on the ordinates. 135 Veterans Administration 232 ARCN1I'E.C OF INTERNAL MEDICINE CHIDROTHIAZIDE + RESERPINE + HYDRALAZINE C!iLOROTHIAZIDE PERCENT or CASES PERCEHT OF CASES ?? CHLOROTHlAZlDE o PLACEBO Fig. Z.-Percentage tlistriblltion curves of changes in diastolic blood pressure. Other notations . are as-in Figure 1. Results Group A (Previously Untreated) Patients. Chlorothiazide Regimens in Mild and Mod- erate Hypertension: The results in the mild and moderately severe groups suggested that the various agents have additive effects. The average reduction from the hospital basal pretreatment, systolic/diastolic blood pres- sure l to the average home blood pressure was 7.9/3.8 mm. Hg in the chlorothiazide-treated patients and 0.9,`O.O in the placebo-treated group (Figs. 1 and 2). The patients receiving chlorothiazide plus hydralazine exhibited an average reduction of 10.2/11.8 mm. Hg, and CHLOROTHIAZIDE + RESERPINE + CRYPTENAMINE PERCENT OF CASES 3o I- c DECREASE those taking chlorothiazide, hydralazine, and reserpine showed an average fall of 17.2i14.3 mm. Hg. None of the 23 patients receiving the 3-drug regimen (chlorothiazide, hydrala- zine, and reserpine) exhibited an increase in tliastolic blood pressure (Fig. 2). Chlorothiazide and Reserpine with Hy- clralazine or Veratrum in Severe Hyperten- sion: In the Group A patients with severe hypertension, the 25 patients receiving re- serpine and chlorothiazide plus a titrated dose of hydralazine exhibited a reduction of blood pressure averaging 22.7120.9 mm. Hg (Figs. 3 and 4). These values are not significantly different from those which had been previ- CHLOROTHIAZIDE + RESERPINE + CRYPTENAMINE PERCENT OF CASES 30 1 i INCREASE DECREASE 20 INCREASE 20 t IO IO n 0 , CHLOROTHIAZIDE + RESERPINE + HYDRALAZINE CHLOROTHIAZIDE + RESERPINE + HYDRALAZINE 44 36 28 20 12 404 12 20 28 36 44 IO 81 0 ?3 I L I I I I I I I I I I I I I I I I , I I I I 44 36 28 20 12 404 12 20 28 36 4: a4 36 28 20 I2 4 04 12 20 28 36 44 Fig. 3.-Percentage distribution curves of changes in systolic blood pressure at the end of 3 months of treatment in patients with severe hypertension. Fig 4.-Percentage distribution curves of changes in diastolic blood pressure in patients with severe hypertension. Vol. 110, Aug., 1962 136 STUDY OF ANTIHYPERTENSIVE AGENTS. T.?BLE 2.-Grouj B: Effect of Addition of Chlorothiaside After One Year of Treatment mtlz ReseTpine, Reserpine plus Hydralazine or Reserpinc plus Ganglion Blocking Age& 233 Antihyp. Regimen Chlorothiazide added to Reserpine Reserpine dr hydrnlazinr Reswpine & gang. block. n&mts "A" plncebo to "C" placebo Av. Blood Pressure, Mm. Hg p-w-- - No of Prior to T.nst MO. Refore 3 Ma. After Casea Any Treat ~l~lo~ot~~i~zi~le C~~lorothia~ide Xl l59/102 154196 145/N il 158/102 153/92 14?/P6 68 I80/116 IA1 19; 151 I93 36 m/101 164/104 lfi4/105 I_- ously obtained with reserpine and ganglion thiazide was 5/10 mm. Hg. During the 3 blocking agents without clliorothiazide in months after addition of chlorothiazide there severe hypertension.2 The somewhat greater was a further reduction of 1116 mm. Hg or a reduction of blood pressure in the severe as total average fall of 16/16 mm. Hg from the compared to the mild and moderate groups mean pretreatment value. The post-treatment probably is related to the fact that the pre- average diastolic pressure was 86 mm. Hg. treatment levels of blood pressure were In 70 patients receiving reserpine 0.25 mg. higher in the severe cases.`" The distribution twice daily without hydralazine, a mean re- curve in this reserpine-, ~lorothiazide-, and duction of S/6 mm. IIg from pretreatment hydralazine-treated group showed no patients basal values before chlorothiazide decreased with a rise in diastolic blood pressure (Fig. further to 14/l 1 after the addition of chloro- 4). The average reduction of blood pressure thiazide, an additional reduction of 9/5 mm. in the severe patients treated with chlorothia- Hg. The average pressure of the 36 placebo- zide, reserpine, and cryptenamine was 17.9,' treated patients was unaffected by the addi- 16.4 mm. Hg. tion of chlorothiazide placebo. Group B ~P~e~.ou~l~ Twoted) Patients.- The reduction of diastolic blood pressure Addition of Chlorothiazide to Mild and after the administration of ChIorothiaz~de was Moderate Cases Previously Treated with correlated with the degree of elevation of Reserpine, Hydralazine : Chlorothiazide or these readings prior to chlorothiazide. The its placebo was added to the regimens of 177 patients with elevations of diastolic above 90 mild and moderately severe Group B patients mm. Hg exhibited the greatest response; who had been treated with reserpine, hydrala- those between 80 and 90, a lesser response, zine, or placebo regimens for one year or and the few below 80 mm. Hg showed essen- longer. I>2 In 71 patients receiving reserpine tially no change in average diastolic pressure 0.25 mg. twice daily and hydralazine 50 mg. after chlorothiazide was added. 4 times daily, the average reduction from Addition of Chlorothiazide to Moderate pretreatment "basal" levels prior to chloro- and Severe Cases Previously Treated with TABLE 3.-Group B: Average Change in Diastolic Blood Pressure, According to Average Diustolic Presmre Prior to Change Chlorothiazid~ Added to Diastolic Pressure Prior to Change, Mm. Hg 90 or higher 80 to QO Less than 80 AIX cases Reserpine No. of AP. cmnge. cvm?s Mm. Hg 47 -7.4 15 -5.4 8 +3.7 70 -5.7 Re~rpine~~~d~~tzi~e h-0. of Av. Change, CCLSCAS Mm Hg 36 -9.1 27 -2.3 8 t-1.0 71 -5.5 137 Veterans Adnzinistratiora 234 ARCHIVES OF INTERNAL MEDICINE TABLE 4.-Azrerage Laboratory Values Before Therapy and at 3 and 12 Months, for Patients Receiving Chlorothiazide and Control Patients Weight, Na, K CL co 1, BUN, Uric Acid, FBS, Lb. MEq/L MEq/L MEq/L MEq/L Mg. Yo Mg. % -clg. % Patients on chlorothiazide No. of patients tested Pretrwtment levels Three-month levels -No. of patients tested Pretreetment levels Twelve-month levels Patients on placebo No. of patients tested Pretreatment levels Three-month levels No. of patients tested Pretreatment levels Twelve-month levels (203) (216) (209) (217) (211) (214) (65) (41) 172 139 4.5 104 27 16 5.5 100 172 139 4.2 103 2i 17 5.6 103 (95) (101) (101) (100) (95! (102) (22) (8) 168 139 45 104 27 16 5.6 102 170 139 4.2 103 27 18 6.2 103 (51) (52) (521 (50) (49) (53) (11) (11) 175 139 45 104 27 15 5.2 108 174 140 4.6 103 27 5.4 95 (23) (24) (24) (23) (23) (2:: (9) (6) 176 138 4.6 105 26 14 5.3 I14 174 140 4.5 103 27 13 5.8 9i Reserpine and Ganglion Blocking Drugs: Data were available on the addition of chloro- thiazide to the regimens of 68 patients taking reserpine 0.25 mg. twice daily plus 1 of 3 ganglion blocking agents-mecamylamine, chlorisondamine, or pentolinium tartrate. Since the previous study failed to reveal sig- nificant differences in the antihypertensive effectiveness of the 3 blocking drugs, the 68 patients are reported as a single group. The average blood pressure in these patients was reduced from a basal pretreatment value of 180/116 to 161/97 after 1 year of treatment with reserpine and ganglion blocking drugs. The addition of chlorothiazide resulted in a further average reduction of 11 mm. systolic and 5 mm. diastolic, to 151/93 mm. Hg. In addition, doses of the blocking drug were reduced to one-half or less of the prechloro- thiazide levels in approximately 45% of these patients. Changes in Body Weight and Certain Blood Constituents The average values for body weight, serum sodium and chloride, blood COz, and BUN in approximately 210 patients were essen- tially unchanged from pretreatment values when tested at the end of 3 months of therapy with chlorothiazide. There also were no sig- nificant changes in these parameters after 1 year of continuous treatment as determined in approximately 100 patients. The group treated with placebos also showed no signifi- cant changes. The average values of serum potassium were unchanged in the placebo group, but in the chlorothiazide-treated pa- tients, the average serum potassium level fell from 4.5 pretreatment to 4.2 mEq. per liter both at 3 months and at 1 year after begin- ning administration of chlorothiazide. Serum potassium levels also were determined after TABLE S.-Percentage of Patients Showing Abnormal Laboratory Values Before and Afer 3 Months of Treatment with Chlorothiazide or Placebo Serum Potassium, MEq/L Serum Sodium, - Serum Chloride, Serum Uric Acid, F&sting Blood Sugar, <135 MEq/L <3.5 <3.0 <2.5 i.Q Mg % > 120 Mg % Patients on chlorothiazide Number of patients tested (216) (209) (209) (209) (211) (65) (41) `% of patients with ab- 14 2 2.6 0.5 0.0 0.9 12.3 ti.6 normal levels pretreatment % at 3 months 10.6 8.7 0.5 0.0 0.5 12.6 17 0 Patients 011 plncebo Number of patients tested (52) (53) 1.9 (j3) (53) (53) (14) (14) % of p:ttients with ab- 15.3 0.0 0.0 0.0 7.2 21.4 normal levels pretreatment 70 at 3 months 13.5 0.0 0.0 0.0 1.9 7.2 23.1 Vol. 110, Aug., 1962 138 STUDY 0); ANTlHYPEKTENSlVE AGENTS 235 1 week of chlorothiazide treatment in 97 pa- tients. The serum potassium in these 97 patients averaged 4.50 pretreatment, 4.06 at one week, and 4.06 mEq. per liter at 3 months, indicating that the reduction oc- curred during the first week of treatment. The laboratory data also were analyzed for the percentage of patients exhibiting abnor- mal serum concentrations of sodium, chlo- ride, and potassium. There was no significant difference between the number of patients exhibiting serum sodium levels below 13.5 mEq. per liter and serum chloride values below 90 mEq. per liter at 3 months after chlorothiazide as compared to before treat- ment. The incidence of serum potassium values below 3.5 mEq. per liter increased from 2.6% at pretreatment to 8.7% at 3 months, whereas in the placebo group the incidence decreased from 1.9% to 0.0%. There was no increase in the number of pa- tients exhibiting potassium levels below 3.0 mEq. per liter, and there were no patients with levels below 2.5 mEq. either before or after treatment. Changes in serum uric acid and in fasting blood sugar were determined in a smaller number of patients. The results obtained to date must be regarded as being preliminary. The average values for serum uric acid were essentially unchanged in 65 patients at the end of 3 months of treatment. At 1 year, the mean values increased in 22 patients from 5.6 to 6.2 mg. %. The average values for the fasting blood sugar in 41 patients were not significantly altered either at 3 months or at 1 year. The incidence of patients with levels above 120 mg. $%, however, increased from 6.6% in the pretreatment period to 17.0% at 3 months. Data on a larger series of cases will be required to determine the significance of these trends. Comment Although the numbers of patients in some of the treatment groups are still too small to provide a final estimate of the comparative effectiveness of certain of the treatment regimens, particularly chlorothiazide alone 130 Veterans Administratio+a and the combination of chlorothiazide and hydralazine, certain features seemed worthy of reporting at this time. There was a distinct trend indicating the additive effects of com- bining various drugs. Chlorothiazide plus reserpine resulted in a greater reduction of arterial pressure than either chlorothiazide or reserpine alone. The chlorothiazide, reser- pine, and hydralazine combination was quite effective and, in fact, resulted in as great a reduction of blood pressure in the severe cases as reserpine plus ganglion blocking agents.2 In the less severe groups this combi- nation reduced the average diastolic blood pressure to normal. The chlorothiazide, re- serpine, and hydralazine regimen did not require the careful dosage regulation needed with blocking agents, nor did it produce as high an incidence of disturbing side-effects. Although not quite as effective as the chlo- rothiazide, reserpine, hydralazine regimen, it is of interest that the Veratrum compound plus the first 2 agents resulted in a significant reduction of both systolic and diastolic blood pressure. Since patients of comparable se- verity were not given chlorothiazide and reserpine without other agents, it is not possi- ble to state how much of the antihypertensive effect was contributed by cryptenamine. The results obtained with this regimen will be reported in more detail, including incidence of nausea and vomiting and treatment fail- ures, as more data become available. The blood chemistry values indicated that the induction of hyponatremia or hypo- chloremia by chlorothiazide was rarely en- countered in hypertensive patients. Mild to moderate hypokalemia was common, but serum potassium levels below 3.0 mEq. per liter were unusual. Potassium supplements were prescribed in only a few cases in this series, although many patients were advised to drink 1 glass of orange juice daily. Individual patients often react differently from the average, the blood pressures of some being controlled on a regimen contain- ing 1 or 2 antihypertensive drugs. Consider- ing the extra cost and increased incidence of toxic effects as additional agents are pre- scribed, it would seem desirable in clinical 236 ARCHIVES OF INTERhrAL MEDICINE practice, in managing mild and moderate forms of hypertension requiring treatment, to begin with a single agent and add other drugs one at a time until the desired level of blood pressure is obtained. Summary Preliminary results of the effects on ele- vated blood pressure of chlorothiazide alone and in combination with various other agents are the subject of this report. Mild to Moderate Hypertension.-Chloro- thiazide in a dose of 500 mg. twice daily resulted in a reduction of blood pressure averaging 7.9/3.8 mm. Hg (systolic/diastolic) as compared to no change in the placebo- treated group. The average reduction of blood pressure in previously untreated patients given chloro- thiazide plus hydralazine 50 mg. 4 times daily was 1 l/12 mm. Hg, and in those taking chlo- rothiazide, hydralazine, and reserpine in a maintenance dose of 0.25 mg. twice daily, it was 17/11 mm. Hg. Chlorothiazide also was added to the regi- mens of patients who had been treated previ- ously with reserpine alone or in combination with hydralazine for 1 year. The reduction from average pretreatment levels of blood pressure was 14/l 1 mm. in the chlorothiazide plus reserpine group and 16/16 mm. in the patients receiving these 2 agents plus hy- dralazine. Moderate and Severe Hypertension.- Chlorothiazide, reserpine, and hydralazine resulted in an average fall of 23/21 mm. Hg. This was not significantly different from the results previously obtained with reserpine and ganglion blocking agents in patients of similar severity. Treatment with chlorothiazide, reserpine, and cryptenamine (an alkaloid of Veratrum viride) was associated with an average fall of B/16 mm. Hg. The addition of chlorothiazide after 1 year to the regimens containing reserpine plus ganglion blocking drugs resulted in an addi- tional average reduction of 10/4 mm. Hg. In 45% of these patients, the doses of the block- ing agents were reduced to half or less of the prechlorothiazide levels. We wish to thank the representatives of Merck, Ciba, Irwin Neisler & Co., and Wyeth Laboratories who collaborated in the manufacture, labeling, an(l distribution of the special preparations of antihyper- tensive drugs and placebos used in this study. Edward D. Freis, M.D., Chairman, Antihyperten- sive Drug Study, Veterans Administration Hospital, 2650 Wisconsin Ave., N.W., Washington 7, D.C. REFERENCES I. Veterans Administration Cooperative Study on Antihypertensive Agents : A Double Blind Control Study of Antihypertensive Agents : I. Compara- tive Effectiveness of Reserpine, Reserpine and Hy- dralazine, and 3 Ganglionic Blocking Agents. Chlorisondamine, Mecamylamine, and Pentoliniun- Tartrate, Arch. Tntern. Med. 106 :81, 1960. 2. Veterans Administration Cooperative Study on Antihypertensive Agents : A Double Blind Control Study of Antihypertensive Agents, Arch. Intern. Med. this issue, page 222. Printed and Published in the United States of America 140