Clinical Trials in Hypertension

In his text Clinical Trials in Hypertension, Dr. Henry Black provides an intriguing historical review of the progression of medical opinion and knowledge of hypertension from the decade of the 1960s to the pres-ent. He presents 24 trials in chapter format, written by the authors who participated in the original studies. A thorough exegesis of each trial is provided to include the criteria for inclusion or exclusion of participants, the pharmacologic regimen with intent-to-treat analysis, the statistical assumptions, and the primary and some unexpected secondary results.

This textbook is not written for casual reading and in some cases requires advanced mathematical and statistical training to fully understand the assumptions made in designing the trials, in patient selection, and in the conclusions drawn from the results yielded by the randomized groups of patients. As Dr. Black explains in his Preface, the text “will be a valuable resource for those in the field,” and I would think he means the clinical trialists and epidemiologists. Those faint of heart in statistics, such as myself, can still find much to enjoy in the presentation of medical opinion that forms the background of each study and provides many amusing glimpses of the medical bias of each period. I would bet that if Dr. Paul D. White were able, he would bury the statement that he made on Australian national radio in 1958: “Much high blood pressure is not serious; it is benign.”

Dr. Black has selected the major pharmacologic hypertensive trials of their times, all of which had some effect on enriching or otherwise changing the medical knowledge of the disease hypertension. In the 1st chapter, we find that the Veterans Administration Cooperative Study on Anti-hypertensive Drugs, a trial begun in 1965, showed that even the primitive drugs reserpine, hydrochlorothiazide, and hydralazine were effective in reducing morbid events in the mildly hypertensive population with diastolic pressures of 90 to 114 mmHg. This countered the prevailing thought at that time that the morbidity and mortality associated with mild diastolic hypertension was due to underlying arterial disease, and not to the hypertension. Later trials using beta blockers and diuretics—MAPHY, EWPHE, STOP, and SHEP^* showed that treatment of both diastolic and systolic hypertension favorably reduced morbid events in both younger and older hypertensive patients. A surprising result of the Department of Veterans Affairs Cooperative Study—which compared monotherapy of hydrochlorothiazide with that of the beta-blocker atenolol, the angiotensin-converting enzyme-inhibitor captopril, the central-alpha-agonist clonidine, and the calcium-channel blocker diltiazem—was that the greatest regression of left ventricular mass was achieved by hydrochlorothiazide. Later, there arose the question of whether too much blood-pressure lowering was dangerous, because it led to a “J shaped curve,” with higher event rates at lower achieved pressures. This was resolved by the HOT^* study, which showed that stepped therapy to achieve 140/85 mmHg was effective to prevent events, but that lowering to 120/70 mmHg caused neither additional benefit nor risk. This study also implied that aspirin in hypertensive men would prevent coronary events. Later trials would concentrate on the effectiveness of various agents in the reduction of hypertension.

In the last chapter, Bruce Neal and Stephen Mac-Mahon provide a prospective view for designing a mega trial (WHO-ISH^*) by incorporating multiple (35 to 37) ongoing studies in order to determine the effectiveness of individual treatment regimens in reducing stroke and coronary heart disease. This will be a truly daunting task involving about 268,000 patients with 1.1 million patient-years of follow-up.

More intimidating for the average physician will be reading and understanding the final reports. The average reader is frequently “blinded” by the terminology of the status of groups in a randomized control trial (RCT) and cannot judge the quality of the study. ¹ Certainly, then, the reader cannot be expected to follow the tracking of huge cohorts of patients, from initial enrollment through intervention, and the analyses of intent-to-treat and follow-up data. One hopes that some agreement on format, such as that found in the recently revised CONSORT^* recommendations, ² will help physicians delve into the numbers and thereby understand the effectiveness of the treatments. In this present text, Dr. Black dealt with this very problem obliquely by having the authors follow a somewhat similar format for the presentation of their trials. Other medical editors should insist upon some common format for RCT reports, so that the average physician can sort through the many trials, served up in their alphabet soup of acronyms, and make sound decisions in choosing treatments for patients.