Comments on ALLHAT and Other Papers in This Issue of the JCH


  • Marvin Moser MD, Editor in Chief

In this issue of The Journal of Clinical Hypertension, the results of blood pressure control in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) are reviewed. In addition, articles discussing the use of home blood pressure measurements, sexual dysfunction in patients with hypertension and its implications, and whether or not responses to a commonly used antihypertensive agent (hydrochlorothiazide) are reproducible are also presented. Finally, this issue includes papers on the effectiveness of electronic medical records in improving treatment outcome in hypertension, and the evidence for benefit of angiotensin-converting enzyme (ACE) inhibitors in protection against renal disease progression in diabetics.


The ALLHAT study represents the largest single study of hypertensive individuals ever undertaken. In this report in the JCH (page 393), results in 33,000 participants with hypertension and at least one other coronary heart disease risk factor are reviewed by Dr. Cushman and the ALLHAT investigators. The patients, who included a large number of diabetics, were randomly assigned, after all previous medications were stopped, to treatment with a diuretic (chlorthalidone), a calcium channel blocker (CCB) (amlodipine), or an ACE inhibitor (lisinopril). Doses could be increased and additional drugs from other classes could be added in efforts to achieve goal pressure. As noted in the paper, the additional drugs, however, were limited to nonstudy drugs, i.e., if a patient had not reached goal pressures on increasing doses of a diuretic, a β blocker, clonidine, methyldopa, reserpine, or hydralazine could be added. If patients failed to respond to the ACE inhibitor, a β blocker or other drugs could be added, but a diuretic or a CCB could not. Thus, there were some limitations in the approach to progression of treatment that might not parallel what physicians would do in their practices.

Patients were in the older age group, with a mean age of more than 65 years. The study included a higher percentage of diabetics (36% of total) and black patients (35% of total) than are found in the general US public. At the beginning of the study fewer than 30% of hypertensive patients had been controlled at goal pressures of <140/90 mm Hg or below, despite the fact that more than 90% of them were receiving some kind of antihypertensive therapy prior to entry into the study.

After 5 years of follow up, more than 90% of patients had attained goal diastolic pressures of <90 mm Hg and 67% had achieved systolic blood pressures <140 mm Hg. These results were achieved by following a fixed protocol that included a careful follow up, and demonstrated that hypertensive patients can be better controlled in a practice setting than national statistics suggest. Most previous data have suggested control of diastolic pressures in 60%–70% of patients with a lesser degree of control of systolic pressure. Fewer than 30%–40% of patients were controlled below 140/90 mm Hg. The ALLHAT data strongly suggest that more careful attention to achievement of goal pressures, even with the use of possibly less than optimal add-on medication, will control blood pressure at goal levels in a relatively high percentage of patients—an important message.

This initial paper from the ALLHAT investigators did not review the specific outcomes with the study drugs, i.e., CCBs, ACE inhibitors, or diuretics. Those results will be published shortly and may help to clarify a major issue. Is there much difference in outcome with regimens based on different medications? Is a diuretic-based regimen as, more, or less effective in reducing cardiovascular events than a CCB or ACE inhibitor? Does it make a difference how blood pressure is lowered?

At present there are abundant data suggesting that 1) a diuretic-based regimen, oftentimes with a β blocker, reduces morbidity and mortality; 2) the use of a CCB-based treatment program in the elderly, especially in those with isolated systolic hypertension, reduces morbidity and mortality both in diabetics and nondiabetics; and 3) the use of ACE inhibitors or angiotensin receptor blockers (ARBs), usually with a diuretic, will decrease cardiovascular end points in hypertensive individuals with diabetes and renal disease. Some trials report that the use of a regimen that includes an ACE inhibitor or an ARB may be more effective in certain population groups in reducing myocardial infarction and the occurrence of heart failure than CCBs.

We await the next ALLHAT paper describing outcomes with various interventions. In the meantime, based on the study reported in this issue of the JCH, targets for blood pressure control should be reset with the expectation of better results than have been achieved in the past. The ALLHAT investigators offer compelling evidence that blood pressure control rates can be increased to at least two thirds of the treated hypertensive population.

It is incumbent upon all physicians to look at these results, compare them to the results they are obtaining in their private practices or in clinics, and to rethink approaches to treatment. If goal blood pressures are established and the use of presently available drugs is more intelligently applied, management results in hypertension and reduction of cardiovascular end points will be considerably improved.


Physicians and patients are increasingly using home blood pressure measurements as a guide to therapy. The alleged advantage is that numerous blood pressures can be taken in between doctor visits and guidance provided for therapy. In this issue of The Journal of Clinical Hypertension, Bachmann and colleagues explore the possible advantages of memory-stored home blood pressures compared to home blood pressures taken and recorded directly by the patient. How many readings are discarded because they aren't “good”? Will knowledge that all blood pressures have been “stored” change outcomes?

The investigators concluded that in individual patients who are aware that their blood pressures are stored in memory, there are fewer “fictional or manipulated” entries. However, they conclude that, overall, this approach may not make a major difference in planning therapy in a majority of patients. Perhaps it is enough to advise patients to report all of their home blood pressure readings without having to resort to more expensive approaches to monitoring.


Drs. Ferrrario and Lang review the problem of sexual dysfunction in hypertensive subjects who are taking antihypertension medications. This has been a controversial subject for years. Many poorly designed studies have reported a higher incidence of sexual problems with some medications (diuretics and β blockers), but placebo-controlled trials have not reported major differences among the most commonly used classes of antihypertensive drugs. These investigators review the physiology of sexual dysfunction and, based on recent data from studies with ARBs, suggest that these medications do not adversely affect sexual activity and may also have a positive effect on male sexual performance.

It should be remembered that a majority of male hypertensive patients are in an age group where libido and sexual performance are decreasing but expectations may not be; some of the symptoms in these men of 60–80 years of age may have nothing to do with medication. It should also be remembered, however, that careful studies have indicated that perhaps as many as 25% of hypertensive men have some evidence of sexual dysfunction prior to therapy.

Although this problem does exist in women, it is far less common.

If the ARBs do prove to have advantages in terms of sexual function, this may be another possible reason to select them for therapy.


Drs. Finkielman and colleagues explore the reproducibility of blood pressure responses to hydrochlorothiazide and conclude that, although blood pressure responses as measured as a “group mean response” appear to be reproducible, individual responses may be unpredictable. These findings should be kept in mind when planning therapy or reviewing the results of therapy that report findings on average.


Does the use of interactive electronic medical records result in better detection and control of hypertension? Is this more complicated and expensive approach necessary to improve outcome, or do the “pen and paper” records suffice? Kinn and colleagues examine this question in another paper in this issue of the JCH.

It appears that there may be an improvement in documentation of the rates of hypertension. But although there also was a significantly greater use of antihypertensive therapy in the electronically monitored group, one might question whether the increase from 90%–94% is of clinical importance.

It does appear that there is a difference in outcome with a considerably greater number of patients achieving goal pressure of <140/90 mm Hg in the monitored group compared to a control cohort. Perhaps the reminder system helps to focus physicians on the task of changing therapy if goal blood pressures are not reached. There are limitations to this particular study, but the area of more careful monitoring of physician treatment patterns appears to be one that merits further attention.


Finally, a provocative paper by Bakris and Weir examines the question of the effectiveness of the use of ACE inhibitor-based treatment regimens in reducing renal disease progression in patients with type 2 diabetes. While most physicians believe that the evidence is clear, these experienced investigators note that, at present, there are no long-term data from a controlled trial confirming the prevention of end-stage renal disease in type 2 diabetics when these agents are used. Despite lack of a definitive study, however, it is probably reasonable to assume that the benefits noted in type 1 diabetics and in other patients with evidence of renal disease when ACE inhibitors are given can be extrapolated to type 2 diabetic subjects. The use of an ARB-based regimen (usually with a diuretic) has been shown to reduce renal disease progression in type 2 diabetics and logic suggests that if and when a large study were done with ACE inhibitors, the results would be similar. Not all experts, however, will agree with this supposition.