| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
LETTER TO THE EDITOR |
Division of Geriatrics Department of Internal Medicine Palmetto Health Richland and University of South Carolina School of Medicine Columbia, South Carolina
It is great to see that the study of physicians' opinions about hypertension management in the elderly (1) has stimulated a debate about blood pressure management in the very old patient. Our study did not address the relationship between blood pressure and health outcome. However, this debate further supports our findings; in particular, the notion that treating hypertension in those 85 years or older has more risks than benefit. This notion was carried even further by a recent article stating that clinicians should be "happy if their 88-year-old patient has a blood pressure of 200/100" (2). This is of concern for 2 reasons.
First, the observation that, in some surveys, lower blood pressure is associated with an increased mortality has been used as a justification for the concern that treating hypertension in the very old patient has more risk. The debate about this issue lies in the interpretation of the evidence; namely, whether low blood pressure, defined arbitrarily by different studies, is a risk factor or marker for worse survival. Population surveys are usually cited as evidence to support the "risk factor" hypothesis (2). A major issue has to be considered when interpreting these data. The association between lower blood pressure and worse survival in many instances disappears after adjusting for other confounding variables. For example, in the cited survey done in Leiden, the Netherlands (3), the association between lower blood pressure and increased mortality was not significant when the authors of this study adjusted for "all information that predicts mortality." Other studies have found similar results (4,5). For example, Satish and colleagues, using the Established Populations for Epidemiologic Studies of the Elderly, found that the association between lower systolic and diastolic blood pressure was not significant after adjusting for covariables: All-cause mortality was higher with lower systolic blood pressure (odds ratio [OR] 1.5; 95% CI [confidence interval] 1.3–1.7) and diastolic blood pressure (OR 1.3; 95% CI 1.1–1.5), whereas, after adjusting for factors known to be associated with an increased mortality, the association was not significant for systolic blood pressure (OR 1.1; 95% CI 0.95–1.4) and for diastolic blood pressure (OR 1.4; 95% CI 1.0–1.4) (5). Recently, a pooled analysis of 61 prospective observational studies, including 12.7 million person-years, showed that the association between higher systolic and diastolic blood pressure and increased stroke, ischemic heart disease, and other vascular mortality rate decreased with advanced age but remained significant in those 80 years or older (hazard ratio [HR] for stroke mortality was 0.67; 95% CI 0.63–0.71; ischemic heart disease mortality, HR 0.67; 95% CI 0.65–0.75 for systolic blood pressure 115 mmHg or greater). A similar result was found for diastolic blood pressure and for nonvascular mortality (HR 0.88; 95% CI 0.87–0.89) (6). The debate question is now: Should these population surveys be used as support to not treat the 88-year-old patient with blood pressure of 200/100 mmHg?
Second, the evidence for a decrease in stroke or cardiovascular disease events from lowering blood pressure using antihypertensives in the very old patient is agreed upon, including the author of the aforementioned article (2,7). The debate lies in whether treating very elderly patients with hypertension leads to an increased survival. In other words, is hypertension treatment in this population justifiable if the evidence for a decrease in overall mortality is not robust, but there is evidence for decrease morbidity? None of the cited studies have shown an increase in mortality with antihypertensive treatment in the very old, but rather they failed to show an improved overall survival (8–10). However, they did show a decrease in stroke and cardiovascular events. Two explanations can be provided for this finding. First, most studies cited as evidence for lack of survival benefit including the studies used in the subgroup meta-analysis were not specifically designed or powered to measure a difference in overall mortality in those 80 years or older. Second, the concepts of "non-ignorable missing covariates" (11) and "adjustment for background mortality" (12) are important methodological issues when interpreting survival data in the very old. Stated differently, clinical trials that measure overall mortality and include very old patients need to consider, in their methodology, ways to account for the nonmeasured covariates that are highly correlated with this outcome and to account for competing mortality causes (13). Moreover, they need to adjust for the background mortality rates, which are high in this group of the population, as pointed out in the review article (2). In fact, none of these studies have accounted for these issues and make the interpretation of the results difficult. The debate then in the 88-year-old patient with a blood pressure of 200/100 mmHg can be stated as follows: Should this patient be denied antihypertensive medications due to the lack of evidence that such treatment can add years to the patient's life even if it decreases related cardiovascular morbidity?
Blood pressure management in the very old is a common problem facing the clinician. A clear understanding and interpretation of the evidence are important in the care of the very old patient. The recent debate supports the results of the physicians' survey that there is no consensus about treating hypertension in the very old. Until studies that are appropriately designed and powered to detect an overall survival benefit in this segment of the population are conducted, an 88-year-old patient with high blood pressure should be considered for treatment unless there are compelling reasons for a no treatment approach.
Acknowledgments
Address correspondence to Ihab Hajjar, MD, MS, Division of Geriatrics, Department of Internal Medicine, Palmetto Health Richland and USC, 9 Medical Park Drive, Suite 230, Columbia, SC 29203. E-mail: ihab.hajjar{at}palmettohealth.org
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 | |
