| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||
COMMENTARY |
University of Pittsburgh School of Medicine, Pennsylvania.
Goodwin's article (1) focuses on mortality as a critical outcome in relationship to blood pressure. This focus on mortality distracts from the weight of evidence that clearly documents the importance of systolic blood pressure as a risk factor for stroke and congestive heart failure, two costly and severely disabling conditions in older people. Systolic blood pressure is a major risk factor for these serious cardiovascular outcomes to at least age 90 (2).Emerging data suggest that treatment of hypertension may preserve brain integrity and delay the onset of dementia (3).
For patients who are well into their late 80s, the benefits of treating hypertension for the prevention of stroke and heart failure are substantial and well documented in meta-analyses of the oldest participants in hypertension clinical trials (4). Treatment of hypertension would not be expected to affect noncardiovascular mortality, and the association of treatment with higher total mortality was only of borderline significance in one meta-analysis. Focusing on a questionable risk of mortality when benefits for preventing serious morbidity are so clear may foster therapeutic nihilism. It is already too easy in geriatrics to forgo treating the oldest old. For patients like those enrolled in trials to date treatment of systolic hypertension is beneficial.
The paradox of high levels of risk factors becoming "protective" in very late old age is important and is neither widely recognized nor fully understood. This paradox is not unique to blood pressure. Both higher cholesterol levels and higher body weight have been shown to predict lower mortality in late old age. In men and women aged over 85 years in the Netherlands, a 1-mmol/L unit higher cholesterol level was associated with a 15% lower risk of death over 10 years' follow-up (5). The nadir of the risk for mortality related to body mass index actually increases from about 25 to 27 with age, and the relative risk of obesity is greatly attenuated for total and cerebrovascular disease mortality with advancing age (6).
Several potential explanations for a higher survival in those with higher blood pressure are cogently discussed in Goodwin's article. Most of the difficulties in interpretation relate to the comparison group in any observational study of very old people. It often turns out that there is no real control group, as virtually everyone has developed at least some subclinical vascular disease by age 80 (7). As discussed, careful adjustment for other aspects of health tend to eliminate this paradoxical association of a lower mortality with higher blood pressure. Goodwin points out that this can be viewed as a sign of physiologic vigor relative to the remainder of the population of the same age. A related but more specific explanation might be that those with the highest pressure may represent the last group in a given birth cohort to have developed age-related vascular stiffness and its consequences. Systolic blood pressure increases in old age because of a loss of elasticity of the large arteries resulting in a reflection of the systolic pressure wave in systole rather than diastole (8). This change in the pattern of flow maintains perfusion and is certainly adaptive, yet this adaptation has well-known adverse consequences including an increase in cardiac afterload. The diminished elastic recoil of the stiff aorta in combination with the absence of augmentation from the reflected pressure wave results in a lower diastolic pressure with the potential to reduce coronary filling. While systolic hypertension is a necessary adaptation for maintaining perfusion in very old people, these same very old people also happen to have extremely high rates of poor outcomes. Thus it might appear that higher systolic pressure is good for old people because the only alternatives are worse.
In spite of being proven to protect against stroke and heart failure, current treatments for hypertension were not specifically designed to address the pathophysiology of vascular stiffness. Direct measures of vascular stiffness are associated with higher levels of homocysteine, visceral fat, and lower physical activity (9,10). Perhaps treatments designed to more directly target the specific pathophysiology of vascular stiffness could result in improved outcomes.
Older adults with very mild elevations of systolic blood pressure represent the largest proportion of untreated hypertensives in the United States (11). It is often only a matter of time before the question of treatment of systolic hypertension will come up and need to be addressed in an individual patient. The lifetime risk for middle-aged adults is 90% (12), and 50% of older adults with high normal blood pressure hypertension will develop definite hypertension within 4 years of follow-up (13). When the pressure finally reaches the threshold for treatment for the first time after age 80, how likely is it that a physician can do better than the patient has already done on his or her own?
The number of older adults reaching late old age is increasing every day, and while few over 80 years of age were enrolled in studies in the past, it is becoming increasingly urgent to obtain data specifically in the oldest old.
Longitudinal studies have the greatest potential to address the problems of heterogeneity by classifying not only the presence of hypertension but the onset and duration as well. Possibly there is a point at which it is too late to forestall the adverse effects of hypertension. Perhaps, as Goodwin proposes, if we can better sort out subsets in the design of clinical trials, we can better determine whether the individual at the doctor's office is one who can do even better with treatment or if it is truly too late. For now, if an 80 or 85 year old is well enough to participate in a clinical trial and sign a consent form, they probably should be treated. The biggest problem for future research in the oldest old is that those who are eligible for clinical trials end up being only a small subset of the patients we wonder about. For the practicing physician, it is important to keep in mind that research results may not pertain to all patients, as the patients in these studies are not often encountered in the day-to-day practice of geriatrics.
Acknowledgments
Address correspondence to Anne B. Newman, MD, MPH, University of Pittsburgh School of Medicine, 3520 Fifth Avenue, Suite 300, Keystone Building, Pittsburgh, PA 15213. E-mail: anewman{at}pitt.edu
REFERENCES
| ||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
