| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
COMMENTARY |
Division of Geriatric Medicine, Saint Louis University, Missouri.
In a recent commentary (1), Dr. Goodwin raises a pertinent and complex clinical issue: We don't know enough about 80 year olds. The knowledge deficit is critical, since the 80 and older group is the most rapidly increasing portion of the population. Specifically, Goodwin analyzes the data (or lack thereof) about the treatment of hypertension in persons older than 80 years.
In a survey reported in the Journal, Hajjar and colleagues (2) found that 25% of surveyed physicians thought that treatment of hypertension in persons older than 85 years produced more harm than benefit. In an accompanying editorial, Aronow reviewed the treatment of hypertension and concluded that no age threshold exists beyond which hypertension should not be treated (3). Goodwin joins in on the side of the quarter of surveyed physicians who believe that treatment causes more harm than good.
Goodwin concedes that the rationale for treating persons with hypertension aged 65 to 79 years is established. He disagrees with Aronow on the treatment of hypertension for subjects 80 years of age and older. In support of his thesis, Goodwin references 2 epidemiological reports conducted in Tempere, Finland, and Leiden, the Netherlands (4,5). In the first paper, the greatest mortality in the very old occurred in persons with the lowest blood pressure. Similarly, in the second paper, all-cause mortality was highest in persons with blood pressures less than 200/100. However, these epidemiological studies do not give us a clear picture of the relationship between blood pressure and mortality. Persons who had low blood pressure in these reports differed from the general population in substantial ways. When other predictors of mortality, such as age, sex, type of residence, and serum albumin concentration, were added into the model in the Leiden study, the effect of a low blood pressure disappeared. Poor health status was more common in persons with low blood pressure, and after adjusting for health status, the effect of low blood pressure disappeared. Moreover, in the Leiden study, treatment of hypertension did not shorten life expectancy, and some benefit of treatment was seen for disability from stroke.
Should hypertension be treated in persons older than 80 years? Some clinical trials have shown a benefit of hypertensive treatment into very old ages. Compared with placebo, active treatment of hypertension significantly reduced the number of cardiovascular deaths, stroke morbidity, and stroke mortality. A significant reduction in all-cause mortality was observed in the active treatment group. The benefits of treatment were discernible up to age 84 years (6).
Both Goodwin and Aronow reference a meta-analysis of intervention trials that recruited patients aged 80 years or older (7). In this analysis, treatment of hypertension reduced strokes by 34%, major cardiovascular events by 22%, and heart failure by 39%. However, there was no treatment benefit for cardiovascular death, and a nonsignificant 6% (95% confidence interval -5,18) relative excess of death from all causes. In his table, Goodwin reports slightly different numbers, and calculates the all-cause mortality as approaching significance [relative risk 1.1 (95% confidence interval 1.0,1.3)]. Thus, the issue is whether treatment of hypertensive persons older than 80 years results in an excess all-cause mortality despite a clear reduction in nonfatal events.
The question is whether treatment is more beneficial than harmful. Treatment of hypertension in the very old exposes persons to adverse effects of medications and a potential decrease in quality of life. In an epidemiological survey of treated hypertension, mortality was highest in men older than 75 years who had a 5-mm fall in diastolic pressure after treatment compared with treated men whose diastolic pressure did not decrease. A systolic fall in men and a decrease in either diastolic or systolic in women was not associated with poorer survival after adjustment for baseline pressure (8). A diastolic blood pressure between 80 and 90 mmHg seems to be associated with a clear benefit in older patients, but the data in support of a systolic reduction below 140 mmHg are controversial (9).
As Goodwin points out, it is more difficult to demonstrate a benefit from therapy in the oldest old. The effect of classical risk factors for cardiovascular mortality is attenuated in the very old. Hypertension in this group is not a predictor of cardiovascular or stroke mortality. Likewise, hypercholesterolemia is not a risk factor (78). Explanations for this effect could simply be that elderly persons are the survivors in a population where significant mortality has already made its mark, eliminating those with the worst risk pattern. A second explanation is the axiom that all prevention strategies must fail, sooner or later. The effect of a chronic condition such as hypertension may eventually take its toll, no matter how effective the intervention.
A potential confounder in the treatment controversy is the "J-curve" phenomenon. All-cause mortality may increase as a result of blood pressure being lowered too much. Interestingly, in several clinical trials the excess mortality began at the same 5-mm reduction in diastolic pressure in treated patients, but occurred only in persons with an abnormal electrocardiogram (11). In large epidemiological populations it is not possible to control for undiagnosed underlying comorbid conditions. Given the higher prevalence of coronary artery disease in the very old, some of the adverse treatment effect may result from excessive lowering of blood pressure.
These epidemiological associations raise intriguing questions, but cannot solve the clinical dilemma. Translating the epidemiology to real people is difficult. It appears from these data that low blood pressure may be a marker for poor health status, that a treated reduction in diastolic pressure of 5 mm may be harmful but may only apply to higher risk patients, that lowering blood pressure may decrease cardiovascular events and stroke but could increase all-cause mortality, and that blood pressure can be lowered too much. What is needed is a randomized, placebo-controlled trial. Medical literature is replete with epidemiological associations that have been countered with clinical trial data. Fortunately, a trial is under way (12).
So, what should the clinician do while waiting for a randomized controlled trial? As Goodwin suggests, we should admit our current ignorance. For most patients, the presence of cardiovascular disease or stroke and the data suggesting improved outcomes with treatment makes the decision to use drug therapy less challenging. For patients who are currently receiving antihypertensive treatment, the data do not suggest a rationale for stopping treatment at age 80. For the very few patients who present with hypertension untreated at age 80, the best course of action is not clear. At the minimum, we should reassess our treatment goals in light of a potential J-shaped curve.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| All GSA journals | The Gerontologist |
| Journals of Gerontology Series B: Psychological Sciences and Social Sciences | |
