I I I I I I I B&B ImFoRmikriom & lm^=E M^m^ucmEmT 300 "Inac D=nnclln ROULXV Ulaimica M.%OtLaostot M.%Ft No 20772 0 USA 10 1301) 24@l 10 RESOUJTIO,N Or- THE JOINT COLNCIL SUBC(lt4I= ON CEPEBROVASCULAR DISEASE CONCERNING TREAIIENIT OF @ERT -aTSIO,'\T At its September 21, 1970 meeting, the Joint Council Subcommittee on Cerebro- vascular Disease, NHLI-NINIDS, adopted tJie following resolution: Several clinically oriented studies have suggested the effectiveness of antihypertensive drugs in reducing the frequency of stroke, and to a lesser degree, coronary heart disease. The recent publications of the Veterans Administration Cooperative Study present results of a uell-controlled investigation, employing random allocation of study subjects to experimental and placebo groups. Over 500 men were studied for periods up to 5 years (average about 20 months). In those persons with very high diastolic blood pressure, the results favored the treatment group so dramatically that the investigators considered that they must terminate the study for reasons of medical ethics. In the group with modest elevation of diastolic blood pressure, the study was allowed to run its course and also yielded results strongly in favor of the treated group. Blood pressure reductions achieved were substantial. The disease cate- gories most markedly affected by the treatment were congestive heart failure, accelerated hypertension, renal damage, and stroke. The re- sults with coronary artery disease are somewhat equivocal but perhaps show a genuine reduction in sudden deaths. The -results for stroke are summarized as follows: Treated Control All events 5 20 @lost serious events 1 12 Thus, not only did the treated group have fewer stroke events, but those that occur-red were much less likely to.be severe. Since this represents a dramatic and extremely hopeful development in the control of stroke, the Joint Council Subcommittee on Cerebrovascular Dis- ease, NHLI-NINDS, strongly recommends that further steps shouzd be taken as expeditiouszy and aggressively as @ossibze@ nonezy, that a committee be convened to design studies to @@ment the appzication of antihyper- tensive therapil on conmnitz es. ,y bas The Joint Council Subcommittee reconmnded that the above resolution be for- warded to the 'National Advisory Heart and Lung Council, the National Advisory Neurological Diseases and Stroke Council, and the National Advisory Council on Regional Medical Programs. Atiyust, 17, 19/-O THE JOURNAL of the American Medical Association Vol 213, No 7 Copyright 0 1970 by the American hledical Association 7il@ of L-L @at i7-Ti 0 Y." II. Results in Patients With Diastolic Blood Pressure Averaging cu'O Thr'ou(yh 1 lz,-, mm H' g Veterans Administration Cooperative Study Group on Antihypertensive A,-ents Three hundred and eighty n2.ale hypertensive patients ivith mm H- who had been random;zed diastolic blood pressures avera6in 690 to 114 mm Hgivere into a prospective double-blind trial 0 of active antihypertensi%,e druls -%,s -rando7n 13, assign ed to either active antII2.@,pei-tensive a cents placebos. Twenty-seven patient@s de- or placebos.The.es@i77zated7-islo o'dez)elopina a morbid event veloped assessable events in the over a fzve-year period was 7-educed from 55% to'18 /'c, by control group as compared to two patients in the @roup receivin- ac- treat7nen t. Tei-i??i?7atiiig morbid events occurred in 35 tive antihyp--rtensive a-ents. This Patients of the control crozip as compared to 9 patients in the striking result favoring treatment treated a7-oup. -ATin eteen deaths related to hypertension or was in a-reement with the results atherosclerosis occurred in, the con ti-ol group and 8 in the of other prospective triais7',' in ija- tients with hypertension of similar actively treated gi-ozip. In addition to morbid events, 20 severity. - control patiei2 ts developed persistent diastolic levels of 125 In hypertension of lesser se-%,erity, Ml ni Hg or hi,,-her. T2-eat7nent was more effective in p2-event- however, there are little or no con- trolled data available on the value in,c congestive heart failure and st7,o7.,e than ill preventing the of antiliype,-tensi%-e drii- therapy. complications of coronary artery disease. The de,ree of Resolution of this ques"i@,oii is of benefit ivas related to the level of prerandomiz-ation great importance not on@\, because blood pressure. of the large number of patients with mild hypertension but ilso because the potential benefits of dru- tr,-at- 0 ment have been questioned espe-. n a previous publication in this Group on Antihypertensive Agents cially in this -roup of li.N,I)ertensive journal' the Veterans Adniin- reported on the beneficial effects of patients.' The present report pre- istration ' Cooperative Study antiliypertensive drugs on morbidi- sents the results of a prospective, ty in patients 'with moderately controlled trial of dru,, treatment For complete list of participants, see severe hypertension. These were on morbidity and mortality in a page 1152. patients with initial diastolic blood (,roup of 3SO patients with i-nild or Iteprint I'C(ILICSt9 to 50 Irving St NW, 0 Washington, DC 20422 (Dr. Frois). - pressures averaging 115 through 129 moderate hypertension whose initial @JAMA, Aug 17, 1970 0 Vol 213, No 7 Morbidity in Hypertension 1143 diastolic blood pressure avera-ed observitioii period aN,era,,cd 90 diily. The lattc!r medication wis 90 tlirou,ii 114 mm H,,Y. tl)roti-i-i 129 mm H- were entered raised to 50 iii- three times cliily if ii@to't.lie triil, I)i-ovidin,- there were the diastolic blood prc-ssure re- Plan of lnvestic,,ation 1'10 other reasons for exclusion. mztiiied it 90 mm or Iii,,,Iic'r. The clinical triil included 52,, Blood pressure wis measured by a Obviously, I)r@icticilly ill of the physician with the patient in a patients in the I)Iicebo -1-otip lead m.ile veterans who, while not re- ceiviii- intili3,I)erteiisi%,e treatment, sitting position. their "doses" raised to this. level. exhibited diastolic blood pressures Other reisons for excluding pa- Provision wis mide for reduction a,.,er.i-iii- 90 tliroti,,,h 129 mm H,. tieiits from the triil, in @iddition to of doses if h,N potciisi%,e reactions or Riiidoiiiizition of patients be,-an diastolic blood pressure, are detailed other disturbin.- side ciTects oc- in April 1964. However, in I\,Iay in other reports.'-' Such reasons curred. Patients in the control -i-oup 1967, the study was terminated in included a Iiistoi:y of a se-%,ere 1-iy- received placebos identical in taste tl-ie subrroul) of 143 patients whose. pertensi%-e complicaton sucl-i as a and ipl)earance to the -icti%,e dru-s. diastolic blood pressures averaged cerebral or siibiraclinoid liei-nor- Indicated symptomatic treatment, 115 tlii-ou-i-i 129 mm Ha prior to rha-e, hypertensive neuroretinop iiicliidin,-, dru- other than antihy- ? -IS raddomiz,ition. Termination of the atliy, discectiiic,,, aneurysm, or renal pertensive agents, was permitted in studs, of this group as previously re- failure, but did not include athero- all patients. ported' was necessitated by the sclerotic complications such as cor- Postrandomization clinic visits bi,-,Ii incidence of morbid events in onary artery diseise-or cerel)rovas- were at monthly intervals for the the control as compared to the- cular thrombosis. Also excluded first two months and at bimonth]3, treated patients, demonstrating at were (1) patients with surgically intervals thereafter. Annual exami- a relatively early date a highly si@- curable hypertension, (2) with un- nations included takin- a history nificant (P < 0.001) effect of treat- related fatal diseases such as malia and a physical examination, roen@L- n-lent. Such a significant difrerei-Lce nant tuniors, (3) those unnvillina or cenocrram of the chest, electro- was not evident at the time, how- unable to return to clinic, and (4) cardiogram, pertinent chemical an- ever, in the patients whose diastolic poorly motivated or otherwise un- alyses of the blood, and renal func- blood pressures averaged below 11.5 cooperative or unreliable patients. tion tests. Additional interim visits mm H.- prior to randomization. The outpatiei-it prerandomizatioii could be scheduled when indicated. These latter patients were con- observation period provided a fur- Characteristics of Patients tinued in the randomized trial until ther opportunity to check 'on the 1969 and are the subject of the reliability of the patients. Ribo- Three hundred and ei.-lity pi- present communication. flavin, which produces bri-ht yellow tients with diastolic blood pressure@, The experimental design has been fluorescence of the urine, was incor- avera-ing 90 throu-h 114 mm HI- described in previous reports.'-@ porated in the placebos. At each were randomized into the trial. Of Initially all patients were hospital- clinic visit a urine specimen was this nui-nber, 186 received active ized for dia-nosis and evaluation examined under ultraviolet !i-ht. drugs while 194 were ai-,,en place.- of the severity of their hyl)erteii- In addition, pill counts were made bos. Tables 1 and 2 indicate tliit. sion. Patients whose diastolic blood at each clinic visit. No patient w,-s the two groups were comparable pressure averaged 90 through 129 accepted into the randomized trial according, to the indicated variables. mm H.- during the fourth through unless the urine exhibited fluores- The median ac.,es were 49.2 and sixth hospital day were accepted cence and the -pill counts were with- 48.1 years and the i-%,erLi.-e a-.es for further follow-up. Patients whose in a stipulated raii-e, at each of two were 52.0 and 50.5 years in the con- diastolic averi.-es fell below 90 mm successive visits durin.- the p,-eran- trol and treatment aroul)s, resi)ec- Hg or rose above 129 mm H- dur- domization observation period. ti-,,ely. N-e-ro pitients comprised in.a this period of hospitalization Accepted patients were then ran- 420' of the control aroup and 411,',, 0 were excluded. domly assi-ned double-blind to of the treated aroup. Blood pressure Followin-, hospitalization the pa- either active drugs or placebos. as measured in the clinic durinff tients entered a prerandomization Active drugs consisted of two types the I)osthospitilization ob-lervitioii observation period of two to four of tablets, one bein.- a combination period prior to randomization z;vcr- months' duration durin- which time tablet containin- 50 m- hydrochlo- a-ed 165.1/104.7 min H" in the they received I)Iicebos of antiliy- rotliiazide and 0.1 m-, reserpiiie control (Troul) iiid 162.1/103.8 min n pertensive igents. The patients which was -iven twice daily. The H.- in the treated patients. There whose diastolic blood pressures dur- other was @5 m.- of hydralazi,-ic were no si,.,niCicant ditlerences L-c- ing the last two clinic visits of the hydrochloride given three times tween the control and treated pa- 1144 JAMA, Aug 17, 1970 o Vol 213, No 7 Morbidity in Hypertension tieiits with rer,,Ird to findings from Table I.-Back@round of Randomized Patients: [Numeration Data renal function tests, fistiii.- blood Control Group Treatment Group sugar value, serum cholesterol value, 11 Characteristic No. % No. % Total uric icid level, and left ventricular 'rotal randomized 194 186 320 enl,ir.-ei-nent as assessed by x-ray -(\.ei.,ro 81 42 76 41 157 fili-ns and electrocardio,rapliv Ot'ier,@ 114 58 109 59 223 By Heart size by roentgenogram all factors measured the two rroups Ungerleider enlarged 42 22 53 29 S3 were comparable. Electrocardiograrr, Left ventricular hypertropt,.y 32 16 30 6 E2 Duration of Observation 'in addition to whites, this group includes four patients of Asiatic extraction, two in'the con- trol group and two in the treated group. Patients were entered into the trial f@om April 196-1 to September 1968, and the study was terminated Table 2.-Measurement Data Prior to Randomization in October 1969. Thus, the earliest Control Group Treatment Group Characteristic tAean ',ican entrants were observed for 5.5 yeal7S Age (yr) 52.0 50.5 and the latest entrants for a niini- AFe (median, yr) 49.2 S. i @ei@gtit, cm (ft, in) I (5. 9) 172.7-(5, Si mum of 1 year. The iveracrle poten- Weight, kg (lb) 82.0 (180.9) 79.3 (I 7ei.1 tial duration.of observation, disre- Duration known hypertension (yr) 4.4 4.6 garding losses and terminations, Average hospital Diastolic pressure (mm Hg) 101.3 100.2 v@rage hospital systolic pressure (mm @g 57.5 154.0 was 3.9 years for the control -rouj, Average (.Iin:c diastolic pressure (mm Hg) iO4.7 103.8 and 3.7 years for the treated pa- v@(,r@6 clinic systolic pressure (mm Hg) 16511 162.1 tients, However, because of the Total severity see. e 6.7 6.8 Renal score (0-4) 0.2 0.2 losses and terminations due to eIL- Cardiac score (0-4) 0.8 -6-. vated diastolic blood pressure de- CNST score (0-4) 0.3 0.3 Serum creatinine (mgl 100 cc) 2.26 1.24 scribed below, the actual duration B U N (mg/ I 00 cc) 15.6 16.2 of postrandomi-zation , observation Ser,um potassium (mEq/liter) 4.4 4.4 was 3.3 years for the control -roup PSP@, excretion (I,. in 2 hr) 58.8 60.0 Fasting blood glucose (mg/100 cc)- 96.5 100.4 and 3.2 years for the treated pa- Cholesterol (MF/"Co cc) 250.1 245.0 tients. I Uric acid (nirIIOO cc) 6.3 6.0 'Detailed criteria for grades 0 through 4 given in reference 6. Changes in Blood Pressure tCNS signifies central ner ous system. v I PSP signifies phenoisulfonphthatein. Systolic and diastolic blood pres- sure fell promptly and significantly the left into the "decrease" zone for laxis. Postmortem examination was in the' treated patients and re- the treated patients as compared to carried out in both of these patients. mained at reduced levels throu,-h- the control -roup. Also apparent is Losses Due to Dru.-I, Toxicity.- out the trial. The changes in blood the wide variation in individual Two patients in the treatment aroup pressure at the fourth month of responses particularly with regard developed reactions thought to be observation in the treated and con- to systolic blood pressure. due to dru- toxicity. The first pa- trol patients are depicted in Fi, 1. tient developed orbital edema w-,,h Losses Other Than Assessable The mean change in systolic blood Events fever and malaise. Roei-.t,-eno.@.rzini pressure was an increase of 4.2 mm of the chest revealed iifil@Lrites in Hg in the control group and a fall Deaths Due to Unrelated Condi- the lun-s. There wis no derii!aliti.@ 'of 27.2 mm H.- in the treated I)a- tions.-Four patients died of dis- or arthritis. Lupus cells were rol. tients froi-n the levels recorded dur- orders unrelated to hypertension. found in the blood alttiou.-h the in.- the prerandomization observa- Two of the patients were in the antinuclear antibody test. was pos- tion period. 'nie mean change in control group. One died of ,-eneral- itive. Protocol dru.-s were d;scon- diastolic blood pressure was a rise ized carcinomatosis demonstrated tinued because of the possibility, of of 1.2 mm Hg in the control pa- at autopsy and the other of ui-e-nia lupus syndroi-ne associated w,-,h tients and a fall of 17.4 mm H- in secondary to carcinoma of the uri- hydralazine. The second patient de- the treated group dlirin- this same niry bladder. One patient in the veloped purpuri one month interval. The distribution of the treated - -roup died of a subdural be-innin- active dru- treifm(,nt. changes in blood pressure as sho%vn hematoma followin- a skull fracture Findin,,s from eximin.-,tion in he in Fig 1 indicates a marked shift '#,O and another of pqrlicillin anapliy- hospital, including biopsy, were con- JAMA. Aug 17, 1970 o Vol 213, No 7 Morbidity in Hypertension 1145 x PLACEBOS 40 PLACEBO 30 30 x INCREASE DECREASE DECREASE 7,-@XCRr,@ SE 20 20 10 .10 0 0 40 ACTIVE DRUGS w 40 C', CL ACTIVE DRUGS 30 30 20 20 10 10 0 0 .64 -48 '32 -16 0 416 +32 +48 -48 -32 -16 0 +16 -'32 +4s 1. Changes in systolic (left) and diastolic'bzood pressure (right) after four months of treatment iii patients given placebos (top) and in patients treated with active drugs (bottom). Mean of changes (R). sistent with anapli3,lactoid purpura. complained of side effects prior to electrocardiographic si-ns of left The purptiric lesions cleared two droppin- out. Nine of these patients ventricular liypertrophy or of roent- weeks after protocol treatment was had been receiving drugs, and six geno-raphic evidence of cardiac en- discontinued and-reapl)eared within were taking placebos. Five patients lar,-ement, which will be reported in three days after administration of had psychiatric or alcoholic prob- a subsequent communication. All active drugs began again. Protocol lei-ns of such severity as to ip.,,ike available data pertaining to each treatment was, therefore, di--con- continued protocol treatment im- organic complication, except the tinued. practical. In the i-emainina patients type of protocol treatment and the Drop-Outs.-Fifty-six or 15% of the reason for drop-out could not be level of blood pressure, were pi-e- the 380 randomized patients were determined. It should be noted that sented to the reviewers and their classified as drop-outs during the three of the patients taking pla- decisions re.-ardinc., the occurrence course of the trial. Of this nui-nber cebos sustained nonterminatin.- and classification of an event ac- 27 had been randomized to -receive morbid events prior to their drop- cordinc, to the definitions -i,,-en n placebos and 29 to receive active ping out. the protocol (see -list of assessable drugs. The avera-e period of follow- events at the end of the coi-niuni- Assessable Morbid Events up prior to dropping out was 17.6 cation) were accepted as final. months with a range from less than The records of the patients re- Table 3 summarizes the assessa- 1 month to 49 months. Six patients ported as havin- assessable morbid ble events by major categories. Such -Aoved away from the area of the events were reviewed by two con- events occurred in 98 of the 380 linic. Two were lost from follow-up sultin.-I physicians who had not randomized patients, 76 in the co,-.- oecause of closure of one participat- participated in tile trial. All assessa- trol -roup and 22 in tle treated jiig clinic. Four returned to tile care ble events -were reviewed - except patients. Of this number 20 control or their private physicians. Fifteen those related to tile development of patients developed an increase in 1146 JAMA, Aug 17, 1970 o Vol 213, No 7 Morbidity in Hypertension diastolic blood - pressure to levels Table 3.-Sumniary of Assessable Events exceeding 124 iiiiii H.- on three Control Group Treated Group separate clinic visits and persisting A for 3 N@,eek-s or Ion-er. Since these No. % No. % Effectiveness, T minating morbid eventst 35 18.0 9 4.8 73 patients were removed from the N-o-,iterminatinE,, 6 events 22 1-3 trial only because of persistent Total morbid events 56 28.9 2 11.8 59 Terminated on account of blood pressure elevations and not elevated blood pressure 20 0 for an organic complication, they Total assessable events 76 39.2 22 1 1.8 70. .No. patients randomized 194 100.0 186 100.0 will not be included in the subse- 'See text. queiit assessment of effectiveness of tlncludes cardiovascular deaths, class A events, and treatment failures except those due to treatment in prdveiiting morbid dias,olic levels >124 nim Hg. events. The remaining 78 patients had organic coiiil)licitions subdivided as Table 4.-Causes of Death ,.'Ollows: 56 of 194 or 28.9% oil the Control Treated Cause Group Group control -roup and 22 of 186 or Deaths due to class A events 11.8% of the treated patients. The Cerebrovascular hernrrrliage 3 0 most strikina evidence -of benefit of Subarachrioid hemorrhage I 0 o Dissecting aneurysm I 0 treatment was manifested in the Deaths due to class B events count of class A events (hyperten- -.'myocardial infarc,ion 3 2 sive complications defined in the Sudc!en death 8 4 protocol which required removal of Cerebrovascular thrombosis 3 1 Ruptured 3the.,oscleo',ic aneurysm 0 1 the patient from the study.' There Total rcia'cd deaths' 19 8 - 'Does not include four unrelated deaths, two in the control group and two in the treated Nvere n(ine amon- the treated group (see text). pa tionts but 14 among the controls. These included five class A deaths (Table 4) plus nine other class A events (Table 5).'NNrhen other car- Table 5.-Terminating t..@orbid Events Other Than Death diovascular (class B) deaths and Control Treated Type of Event Group Group treatment failures were added, the Class A events comparisons were still impressive, Uncontrolled cardiac failure 5 0 Dissecting aortic aneurysm I 0 35 of 194 patients or 18.0% amoii-st Subarachnoid hemorrhage I 0 the controls and only 9 of 186 or Fundi, striate hemorrhages I 0 4.8'Yo in the treated group (Table Acute hypertensive encephalopathy I 0 3). rllie effectiveness of treatment Subtotal 9 0 of Treatment failures (differenc 'e in percent incidence Cerebrovascular thrombosis, severe 4 0 complications between control and Progressive azotemia 1 0 groups divided by the per- Fundi, one striate hemorrhage and ? early papilledema I 0 .cent incidence in the control group) Fundi, one striate hemorrhage and ? encephalopahy 1 0 in preventing termiiiatin.- or-anic Hypotension 0 1 complications was 73% (Table 3). Subtotal 7 The decision to discontinue the trial Total 16 Nvas based on this favorable evidence supplemented by the life-table an- :alyses described below which sug- control group and 8 in the treated associated predominantly with co.@- gested that the benefit of treatment patients (Table 4). Five deaths onary artery disease. Eleven pa- was continuing through time and associated with class A or hyper- tients in the placebo -roup and 6 in Nvas not solely concentrated in the tensive events (see list of assessable the treated group had either a docu- first year or two of treatment. events at the end of the communi- men'-ed myocardial infarci@on or a Terminating Events. - DPATHS cation) were cerebral hemorrhage "sudden death." Cerebrovascul@ar @ELATED TO CARDIOV,%SCULAN Dis- in four and dissectin- aortic an- thrombosis as opposed to hemor- PASE.-Twenty-seven patients died curysm in one, all occurrin(r in the rha-c was the cause of death in of hypertensive or atherosclerotic control group of patients. Deai.hs three control patients and in one complications, 19 occurring in the resulting from class B events '%vere treated patient. The remainin- OAMA, Aug 17, 1970 o Vol 213, No 7 Morbidity in Hypertension 1147 Table 6.-Nontermina"ing Class 6 Events col treatment (see list of isses,,;,ible events listed at 4lie end of the coni- Control Treated Type Group Group i-nunicition) . O,)jcc'tively deii,,on- CVA, thrombosis or TIA' 8 4 strable atherosclerotic coi-npl:lcitions no ;cFsfv'e-- II- 0 predominate is cliss B events, but m.. 2 5 - yocardial infarrtion Atri-al fibrillation 2 the cate.abry ilso includes coii-es- I tive heart failure rcsponsin-e to rou- Wca-,i.b-lock ru@m cT,-,, t nne, persistent, >2.0 mg/100 cc 1 0 tine tl)erapy other thin admiiiistra- prpte..-i 0 tioii of antiliypertensive driias and u r certiin less severe manij'estati'ons-of 'Cerebrovascular accident, either a thrombosis (clinical diagnosis) or transient ischemic at- tack with objective neurological signs. reiial diseases iControlled by administration of digitalis and short-term diuretics. Nonfatal class B events occurred in 21 of the control patients ind in Table 7.-Classification of Niorbid Events by Diagnostic Categories 13 of the treated patients il'Tible Total Events Terminating Events 6). Six patients developed con,,res- tive heart failure controllable by Control Treated Control Treated Diagnosis 20 5 2 1 diitilis and sl)ort-term administra- C@rebrovascular accident 0 c;ronary art-ery-dsease 13 6 tion of diuretics. It is noteworthy Congest-ive heart fai that all six of these patients were "Accelerated" hyperl 0 4 0 R damage 0 1 0 in the control croup. Also, the in- 0r:r 5 - cidence of nonterminatin- cere- th 2 2 otal 56 22 35 brovascular accidents was twice as great in the control as in the treated death in the treated group was ure." Four were associated with patients. However, nonfatal myo- r accidents diagnosed cardiil infarction occurred in five caused by a rupture of an ather- cerebrovascula o8clerotic aneurysm of the aorta. clinically as thrombosis rather.than of the treated patients as opposed OTHER CLAss A EVENTS.-Nine hemorrhage but which resulted in to two of the control aroup. The in- patients in the control group as such severe incapacity that the pq- cidence of atrial 'fibrillation and opposed to none in the treated tients were -unable to attend the conduction defects was essentially croups. group developed nonfatal class A clinic. Two additional control pa- the same in the two . events (Table 5). Five of the pa- tients were removed from the study Life-Table Analysis.-Tlie bene- tients had congestive heart failure because of the appearance of a fit of treatment is more precisely life-table netl)ods e 0 which could not be controlled by sin.-le striate retinal hemorrha.- ai-ialyzed usiii- a ssociated in one with symptoms (Fi- 2). This method has the fol- -dministration of digitalis, sodium a es: (1) it adjusts restriction,'and the intermittent ad- suggesting acute hypertensive en- lowing advanta.@ ministration of diuretics. In the four cephalopathy, and, in the other, for the fact that patients enter the remaining patients there was one with questionable early papill'ede- study at different times and thus instance of each of the following ma. The remaining control patient are observed for varying lengths of complications: dissecting aortic an- exhibited increisin- azoteniia. One time; (2) the method adjusts for eurysm, subarachnoid hemorrlia,,,e, patient in the treated group was any differences in losses to o,)serva- multiple striate retinal hei-norrhaes, removed from the study because of tion between the control and treated and acute hypertensive encepha- hypotension following a myocardial ,i-oups; and (3) most important, it lopathy with accompinyin.- neuro- infarctio'n which resulted in his in- determines whetl)er the benefit of lo,-ical signs. ability to tolerate the antihyperten- treatment occurs cirly or ]@ite or is OTI-Irrt TERNIINATT,G L@ vrNTs.- sive re-imen. It is noteworthy that continuing tlirouh time. The dis- Additional organic complications, of the 17 nonfatal tel-minatina tance sel),xratin,, the control and which did not fulfill the criteria for events (class A and others) 16 oc- treatment lines is a measure of the class A events bUt which were curred in the control group and derre6 of benefit. nevertheless of sufficient severity to only one in the treated patient It is clear from r@i,., 2 tliit the re nefit of treatment manifested it- .quire tern-iinatin,y protocol treat- (Tible 5). be IltilILied tliroti-liout merit occurred in ei,,Iit patients of Noiztet-iiiinatin.- (Class B) @, vents. self early and co n yeirs of follow-ul). which seven were in the control -Class B events include or.-ai-Lic the entire five group. These are listed in Table 5 complications which require no or The life-table iiizilysis of either tf2r- under the subtitle "treatment fail- only temporary suspension of proto- minatin- or all i-norbicl events inoi- Morbidity in Hypertension 1148 JAMA, Aug 17, 1970 Vol 213, No 7. Table 8.-Incidence of Morbid Even,ls With Respect to Level of Preranu@omization Blood P.,essure Control Group Treated Group Patients With Patients With Prerandomization "Morbid Event" "Niorbid Event" Blood Pressure, Patients Patients % mm Hg Randomized No. % Randomized No. % Effectiveness Systolic <165 98 15 15.3 108 10 9.3 40 Systolic 165+ 96 41 42.7 78 12 IS.4 64 Total 294 56 186 22 Diastolic 90-104 84 21 25.0 86 14 35 Diastolic 105-114 110 35 31.8 100 8 8.0 75 Total 194 56 186 22 Table 9.-Incidence of Morbid Events With Respec' , to Aa,e and Race Control Group Treated Group Patients With Patients With "IV.orbid Event" "Morbid Event" Patients Patients % Randomized No. % Randomized No. % Effectiveness Age (on admission) <50 yr 99 15 15.2 102 7 6.9 55 50 & over 95 41 43.2 84 15 17.9 59 Total 194 56 186 22 Race Negro 81 21 25.9 76 8 10.5 54 Other 113 35 31.0 110 14 12.7 59 Total. 194 56 186 22 Cates that the benefit increased with 18% yielded a t-value of 5.0 Nvhich sudden death) were nearly the time. For example, with respect to is highly si.-nificant. A crude esti- same in the two -roups, 13 in the de all morbid e\,ents" it may be seen mate -lave confidence limits of 49% control and 11 in the treated, al- tl)at at three years the estimated to 81% for the observed 67% eitec- though fatal coronary events were cumulative incidence of morbidity tiveness. somewhat greater in the control in the control group is twice as -Relationship- of Treatrnent to -roup. great a@ in the treated patients. Other Factors.-RFLATIO@NSHIP TO Rr@LATIONSI-ilp TO PRE-RA@-N-DO.%@@IZA- This suggests that treatment was DIAG,-,'OSTIC CATEGoriFs.-It is re- TioN BLOOD PIZE;SsurE.-The bene- about 50% effective at tl-iree years. vealin,, to examine the incidence of ficial effect of treatment was most At five years the spread between morbid events as related to treat- evident in the patients with hi-lier the two curves was substantially ment when the events are classified initial levels of blood pressure. Nvli;,.h greater indicatin- an increasiii.- accordin- to diagnostic cate-ories respect to diastolic blood pressv..-e degree of benefit with the passa.-e (Table 7). Thus, in the control vs the effectiveness of treatment was of time. Specifically, at five years the treated ,,roups, the prevalence 750/,-o in the patients with preran- the cumulative incidence rate of of congestive heart failure was 11:0, domization diastolic blood pressure events for the control -roup rises to of renal deterioration 3:0, and of averacrin- 105 tlirou-h 114 mm F@, 55%. By contrast, for the treated "accelerated" hypertension (hyper- as opposed to only 35% in the group the indicted incidence of tensive netiroretinol)atliy or enceph- group avera-in.- 90 throu-ii 104 ir.,n events at five years is only 18%. It alopathy) 4:0. The number of H- (Tible 8'. A sii-nilar ilthou.-Ii cifn be estimated, therefore, that cerebrovascular complications also somewhat less sti-ikin- trend was over a five-year period treatment seemed to be considerably influ- noted with respect to svstolic blood prevented 37% morbidity (55% enced by treatment since the ratio pressure, tl-ie effectiveness of trea@,- minus 18%), and this represents a of cerebrovascular events in - the ment bein- 64c/7 in pit;ents with 67% effectiveness (37/55). control vs treated patients was 20:5, initial systolic levels avel..i-in- 165 'ne standard errors at five years and, of the i-nore severe or terminat- mm H- and above as opposed to were 6.3% for the control patients in.-, cerebrovascular events, it was 40% in the -Irotip with lower initial and 4.0% for the treated groul). The 12:1. On the o'Llier'hand, assessable systolic blood pressure. significance of the difference be- events caused by coronary artery RF-LATIONSII.UL To AGE.-IL e Yra- tween the two rates of 55% and disease (myocardial infarction or jority of the patients de\,elopin- JAMA, Aug 17, 1970 9 Vol 213, No 7 Morbidity in Hypertension 1149 will be made in a subsequent coiii- niunication. The two patieii@ lost to protocol because of cIrLi- toxicity Control group have been (,Iescril)ed ibove. In addi- TERtiliNIATI NG tIORB ID EVENT tion, in tli(,@ present report oily those side effects rc-ciiirin,r rc,rro,,,,il of either reserpine O.- liydrocliloro- tliiazide from tl)e treatment,,re.-inlen will be considered. Treated group I Administration of either res(-@rpine or hydroclilorotiiiazide or tl-eir v@. A@ (,-,L, i placebos was withdra%vn because of side effect in 29 patients. Resc-i-I)iiie L and liydrochlo.-o@Iiiazide were ad- ministered combined in a sin-le -,oid losses to tablet. In order to a, protocol because of side c-@@@ects pre- Control group siimably caused by one or the other ALL MORBID EVENTS of the two agents, provision was made to permit substitution of a tablet which contained either resei7- pine or li3,drochlorotliiizide alone medica- and omitted the offendin tion. These special tablets %-,-ere Treated group made available on request of a participatin- physician. Similar- -e appearing, placebo tablets wei made available for the control pa- know whether the substitution rep- tieiits and the physician did not resented active drii-S or placebos. 2 Estiti?ated cu7nulative incide7ice of 7norbidity over In the majority of the 29 patients a five-year period as calculated by life-table method. substitution of the special tablet Terininating morbid events (top) and all niorbid events (bottom). was necessitated by presui-ned res- -rental erpine-indticed side effects. depression occurred in 12 patients. morbid events Nvere in the older a.-le cidence of morbid events was no However, only seven of these pi- group. Of the 56 control patients -reater in Ne-ro patients. In fact, tients had been receivin- active developin- morbid events 41 were in the control group the incidence dru,s while the remaining five l@ad 50 years of age or older at the time of events was sli.-litly lower in been randomly selected to receive of admission to the study, while LN-e.-roes, 25.9'yu as opposed to placebos. Ten patients cle%,f-l@)I)od -only 15 were below a-e 50. A sim- 31.0% of the other patients. A peptic ulcer of which six had been ilar distribution was found with similar relationship was noted in taking'active dru.-s and four I)Iacc-- respect to the treated patients. The the treated patients (Table 9). The bos. In two patients slbstitutioii percent effectiveness of treati-nent 'percent effectiveness of treatment was made because of impotence; was approximately the same in the was essentially the same in the two one of these two had been -raidoinlv younger and older -roups (Table racial -Iroups. selected to receive the placebo re.-i- 9). However, because of the lower men. The remainin@ six pit;ents all Side Effects number of@ events in the patients were receivin- active treatment. I)elow a-e 50 the estimated effec- In the treated -.roup of patients Their side efTects included sleel)- .iveness of treatment cannot be ac- dosage adjustments frequently were iness, severe nisil stuffiness, ut, cl'pted with the same degree of required because of hy@otensive and seizures presui-n,-,bly caused by hy- confidence as in the older patients. other symptoms. A complete analy- potension, and abnormal results RELATIONSHW To RACE.-The in- si's of these and other side effects from the -lucose tolerance test. 1150 JAMA, Aug 17,'1970 'o Vol 213, No 7 Morbidity in Hypertension with coroniry irtery diseise w.4s events ilso was difterei-it in the two Co@ment essentii]IN, the s@iiiie in the control c.troul)s of p,-itieits divided accord- The eP.ecti,%,oness of treatment and treated patients- iii@@ to level of initiil diastolic blood was cle,-irl@, demonstrated In the Bec@iuse of the -ridu,-Il pro,-Ires- pressure. Ii-i the' control pitieits patients with prorindoiiiz,-ition sys- sion o' atherosclerosis, the iie-,iti%,e with iiiitiil diastolic levels of 115 tolic blood pressures above 16.1 or result %@,itli re.-.ird to p,@e%,entloii of iliroul-li 129 i-nm H.- iccolera@tc-d diistolic pressure above 104 i-nm myocardial iiifirction and sudden hyl-)ez-teiision @vith hypertensive Hc,. The difference in the incidence deitli ciniiot be tiken as eN,idc.,nce neu@-orc-tinol)athv was iiic@ most f'rc,- of morbid events between control that treatment is ineffec4ive. Con- queiit coi-nplic,-itl:on. In the present and treited patients was less clear tinuatioii of the present study Nvas series cei-ebroviscular diseas'c,, con-. cut in the patients with blood pres- not justified because of the favor- gestive lico@rt failure, -ind co,,-o@ir3, sures below these le%,els. This may able evidence with i-e.-ard to pi-e- artery disease were the most frc- be due to the fact tli@it or,aanic com- veiition of hypertensive complica- queiit morbid events occtlrrin- in plicatioiis ipl)eir slowlv in inild tioiis. If follow-up had been ]on.-er, the control -roup. Of the four coil- .h trol patients who died in the pre- .@,pertension as indicated by the and if administration of -,ntihNper-, considerable, lo@@,ei- incidence of tensive dru.-s had been started at In viotisly reported series of patients such events in patients Nvitli blood earlier a,-e, a si,--nii-icant difference with Irish diastolic pressures, three pressures below 165/105 i-iim I-I-,. mi.-lit have been demonstrated. The deaths were caused by cliss(,ctiii.- As would be expected, a -Ircater avera.-e a-,e of the patients wis 51 or ruptured aortic aneurysm where- incidence of oi-,-.aiiic complications years and li3,pertensioh could liavc, as the most common causes of dE,,a-h occurred in the older than in the been present for many years prior in the series of patients with lower yotiii.-er patients. Of considerable to randomization. Atherosclerosis of diastolic pressures were strokes, importance is the observation thit the coronary arteries, therefore, myocardial infarcts, and sudden treati-neiit was found to be effective may have been well established at death. in reducin.- the number of such the time of entrance into the study. It should be empl-iasized that the coi-nplicitions in these older pa- Further trials are needed in a more prec-eiit study dealt with a selected tients. Although the indicated effec- selected population to determine population. @,lany uncooperative tiveness of treatment was essentially whether antihyl)ertensive treatment and unreliable patients were iden- the same in patients above and helps prevent coronary artery dis- tiried and eliminated from the trial below age 50 years, the results were ease. on the basis of pill counts, urine not as convincing in the youn-er It is of interest to compare the fluorescence test results, and irre- group be@-iiise of the low incidence results of the I)resent.series of pa- ulirity of clinic ittendaiice durin.- of i-norbid events in both the control tients whose initial diastolic blood -a proraiidomiz,-ition observation pe- -es a- -nezit obviously would and treated patients. It should be pressili N,era-ed 90 throu,,h 114 riod. Treati -mentioned, however, that in 'the mm H. with the results previously not have been as efiecti%,e in a groul) -of 20 control pitients, not reported in the patients whose dia- -roup of patients less cirefull@, se- counted ' as haviii.- morbid evei-its stolic blood pressures at the begin- lected with re---,ird to their desire but \vlio were removed from the nin.- of study averi.-ed 115 through to cooperate. The population was study beciuse of persistent eleva- 129 mm H.-.' The benefit of treit- further limited ii-i that it excluded tion of diastolic blood pressure ment was quickly manifested in the female pitients and pzit;eiits with greater than 124 i-nni Hg, 14 so re- latter series. Tliirty-ei.-ht percent of labile Hypertension whose diastolic moved were below 50 years of a-e. the control patients in thit series blood pressures i%-era-eci lower than Treatment was most effective in developed assessable events over an 0,0 mm H.- dui-in.- the fourth 'chrou,-h .preventative hypertensive coiiiplica- avera-e period-of only 15.7 months the sixth day of liosl)it,-ilization. LI tions and leist effective in prevent- of postraiidomization follow-up, Finally, the incidence of i-norbid in- atherosclerotic 0 p 0 complications, whereas such events occurred in events in the -rou bclo%@ a-e 50 -particularly those associated with only 3% of the treated ]),itients. A wis relatively low. Further studies coronary artery diseases Conipl,,ca- considerably longer period of fol- are needed to evaluate the eirecri%,e- tions such as congestive heart fail- low-up was required to demonstrate ness of treatment in libile Iiyper- ure, renal damage, cerebrovascular a si,-nificint benefit of treatment in tension and in the pi-e%,eiiiion of hemorrhage, and accelerated hyper- the presently reported series of pa- atherosclerotic complic@itions, par- tension occurred only in the control tients with lower levels of diastolic ticularly coroniry artery d;sease. group. On the other hand, the in- blood pressure. Such studies would seei-n to require cidence of complications associated The distribution as to type of larger nwnbers of younger patients JAMA, Aug 17, 1970 o Vol 213, No 7 Morbidity in Hypertension 1151 %N,Iio can be follonved up for Ion" Coziii: and N\(@.st Roxbury, 6. Acute hN,I)erteiisiN,e (..zi(-el)halol)a- periods of time. 13iostntisticiins: Riisscll I3. ToNvk.,;bui-y, thy requiriii- liosl)italizatioii. SLI), ariti T,,i%vi-enct@ N" Sli@i%v. Within the limits dei'ined by this Ceiitril 0,11!cc Coor'd.in.,itor:. Hir.olcl W. Treatment Failures study, liowe%,ei-, the present results SCIIIILl@ei-, %,II). COIISIllt.llliS: C@c-iiest, MD; 1',Iv I. Diastolic' I)Iooi' cxceeci- leave little doubt that intili in,, 124 iyim I-f,, on (,,i(,Il of three su'c- .N,pe.r- W. Gii'fc)i-(]. Jr., '.%ID; NVilter '@%l. l@ir;wn- tensive cirti.- treatment is I)encficiil. ci;ill. ',%ID; @%ID; D.,ivi(I NV. ce.;sive visits and 1)c-rs-'Istin, for three The present results to-ctlier with ll;c-liii-dson, @ID; and l@ol)ert W. NVilkiiis, weeks or longer. 0 @,@D. 2. Assessallc, organic cc)rnl-,Iicaliors those previously reported in 1),i- 'I'liL medications used in this in- not ftilfillin,, criteria for A c,%'ents tients with initial diastolic blood vestig.,ition I)rcl),ii-ed by NVilliiiii r!'.- but of sufficient sc.-@-erity to rc-qu;rc ciii- pressures of 115 throti-li 129 inm N\@-igiier, %,ID, of Ciba lliiirniaccittical Co., continuation of I)rotocol re,,imc-n. SLIllInlit, NJ. Hg' indicate clear]N, that the Iii-her Class B Events the level of blood pressure the References Cardiac ,Ireater the de.-ree of benefit of such 1. Effects of treatment on morbidity in I\Iyocar(liil ii-ifarction document- therapy. Certiin complications suc@ Results in patients %,.'itli diastolic blood I)resstircs .;tveri.-in iz 115 ed by characteristic clectrocirdio.-ran as coii-estivc heirt failure, hyper- ttirou,,I) 129 iiini Hg. N7etei-,iiis Adiiiiiiistri-t- or serum ezizvmc cliangL@. tensi%,e neuroretiiiopitliy,' strokes, tion Cooperative Study Gi-otil) on Anti- 2. Con&@csrl%le heart future control- and renal deterioration were re- hvl)ertei)cive , A-@nts. JAilfA 202:10f)8- lable b.N, routine therapy other than 163,1, 19C,7. antiliyj)ertensive a,,c-ii@'s, inclu(iin,, cl;"i- educed or essentially eliminated in 2. Haniiltoiii\,I: Seiection of p@itients for talis, restricted acti@,it,-,,, 'o%N, sa@ diet, the treated patients. In addition, antili),I)crtensive therapy, in Gross F (cd): and intermittent diuretics. .-Intih-,,pertensive Titcral@%,: Princil)lcs and 3. Atrial fibrillation or flutter or ven- treatment prevented elevation of Pract'ice, an International S.N,ti7posiuni. tricular tachn,carclia -,vithout evidence diastolic blood pressure to levels @New York, Springer-Verlig Inc, 196G, pp of quinidine or di, where the risk of de%,elopin- li%,Der- :Lg,3-211. italis intoxication. 3. Wolff FW, Lindeman RD: Effects of 4. Heart-block such as bundle- tensi-%,e complications is -reatly in- treatment in hypertension: Ilesults of a branch block, second or third cle.-ree creased. The effectiveness of the controlled stLi(IN,. J Chronic Dis 19:227-240, heart-block or first degree hc-art-b!ocl.- treatment in preventin- such pro- 19G6. with P-R inter%,al of 0.28 seconds or 4. Relniin AS: Comment on who needs ore. gression is indicated by the fact drugs for hypertension in lngelfin,-er FJ; m that while persistent elevation of Relm,-in AS; Fiiland M (eds): Controversy 5. Left ventricular enlargement by diastolic blood pressure exceedinr in Internal '.Icdicinc. Philadelphia, W B E- CG or roent-enogram. Saunders Co, 1966, I)p 101-102. 6. Pulmonary embolism or infarc- 124 mm H- occurred in approxi- 5. Freis ED: Or,-aiiizition of a long- tion. term i-nultielinic therapeutic trial on hyper- mately 10PIO of the control patients, r, i2tral Nervous Systei@z tension, in Gross F (ed): Antihyperteiisive %,e they were completely absent in the Tlierap3,: Principles and Practice, an ititer- 1. Cerebrovascular thrombosis or treated -roup. national Sviril)osiutr2, '.N'e%v York, Springer- embolism. Verlag Inc, 1966, pp 345-354. 6. A double-blind control study of anti- Transient ischemic attacks with Participants -ical cban-es durin- objecti%,e iieurolo. hypertensive agents: 1. Comparative effec- the attack. Permanent'.Nfeml)ers of the Study Group;- tiveness of reserpine and hyciralazine, and Aorta three -anglionic blocking agents, Veterans Nlassiino Calibresi.'@ll); C. Hilnion Castle, Adniinistrtiion Cooperative Study Group 1. Arteriosclerotic aneurysm. @-ID; Leo Elsoii, %II-); rd%vai-d D. Freis, on x,%ntihvpertensi%-e Agents. Arch Intern NID, Chairnian; Rudolph P. Freniont, %ID; &led 106:@1-96, 1960, Renal hlicliael A. Harris, @NID; David Littman, 1. Doublin- of BU.\' (but to beloiv MD; Eli A. Riiiiirez, NID; and J. R. Assessable Events 60 mg/100 c@) or creitinine levels to Thomas, @%ID. values abo%,e normal limits not due to Other Alcml)ers: Ltais A. A@iis, NID; Abbreviated definitions of terminat- primary renal disease. Mark L. Arinsti-on,-, NID; Alston W. in.- events (class A and treatment fail- u 2. Proteinuria (2+ or more in three 131ount, @%-ID; 'I'liomas A. Bruce, '.\,ID; Ovid res) and nozit-erminatin- (class B) P. Bush, Jr., '.%ID, deceased; Eu--ene C. events. or more specimens) in absenec- of con- Clark, NID; Annette Fitz,'@%ID, R. ',%I. Free- ,;estive heart failure, I)ri.,i-iary rt-nal dis- man, MD; D. Frolilicli, - %ID; Class A Events ease, or lower urinary tric, disease. Arthtii- Geir, '%ID; John D. l@yi-i,-tcopoulos, 1. Striate hemorrhages in more than 3. Persistent hc.,niaturia (> 5 ce!.'s iMD; Alin I;'. J,yon, NID; Gloria D. i'vlas- per high power field centrifuged sedi- @-iro, i\ID, Doiiil(I NlcCaugliin, %-ID; Jean o'ne retinal quadrant or cotton wool ment) not due to primary renal or Morgan, .,NID; Henry NV. Overbeck. 'vfD; exudates or I)apilleclema. @Iisco C. Ilerez-Stable, i%ID; Mitchell 2. Cerebral or subarachnoid hemor. lower urinary tract diseases. Perry, MD; Ilo.-er Sutton, IVID; and James rha.-e. Taguchi, @NID. 3. Dissecting aortic aneurysm. Participating Veterans Administration 4. Inability to control congestive hospitals Allen Pirk, Nlicliigan; Birmin,-- bam, Ala; Brooklyn, NY; Diyton, Ohio; heart failure without usin. antihyper- Iowa City; Jacl,.@on, Miss; \Iemphis; Nash- tensive agents. ville, Tenn-, Oklihom.,i City; Pittsbur.-ii; 5. Elevation of blood urea 'nitrogen Richmond, Va; Salt I-ike City; St. Louis; level (BUN-) by more than 50% of pre- San Juan, I'll; NVasliington, DC; West vious level and exceedin,., 59 m.-/IOO cc. 1152 JAMA, Aug 17, 1970 o Vol 213, No 7 Morbidity in Hypertension