Measuring Blood Pressure and Treating Hypertension in (Frail) Older People

In the current issue of JAGS, Auseon et al.1 describe a cross-sectional, observational study of blood pressure measurements in 857 nursing home residents, with a mean age of 84 years, conducted in 45 nursing homes operated by a large national chain. Using a random-zero sphygmomanometer, blood pressure was measured supine and standing before and after breakfast and lunch, and patients were classified as to whether they had “hypertensive” levels of systolic blood pressure (> 160 mm Hg) on none, one, or two to four of the supine measurements. Blood pressure control was defined as supine blood pressure less than 160/95 mm Hg. Isolated diastolic hypertension (systolic blood pressure < 160 mm Hg and diastolic blood pressure > 95 mm Hg) was rare (0.9%). Because of this infrequency of diastolic hypertension and because systolic blood pressure is a stronger predictor of risk than diastolic blood pressure in older populations,2,3 the authors’ focus was appropriately on systolic blood pressure. One of the principal findings was that systolic blood pressure was significantly higher before breakfast than at any of the other three measurement times. The authors recommend basing the diagnosis of hypertension in frail older people on multiple within-day supine and standing blood pressure determinations, taken before and after meals, in order to avoid overly aggressive treatment if the diagnosis were based only on early morning readings or to avoid insufficient treatment if the early morning readings were not available. Calling attention to this pattern of variability is an important contribution of this report, especially when providers must interpret blood pressure measurements taken at various times in a nursing home or hospital setting. However, data proving that treatment decisions based on their recommendation would lead to improved outcomes are not presented. It must also be acknowledged that the definition of hypertension and blood pressure control in this paper is different from those in the Sixth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI), which is sitting (not supine) systolic blood pressure of 140 mm Hg and/or diastolic blood pressure of 90 mm Hg for both diagnosis and goal.4 Auseon and colleagues’ focus on systolic blood pressure of 160 mm Hg is, however, consistent with evidence from randomized, controlled trials of treatment of systolic hypertension in older people, all of which defined elevated (sitting) systolic blood pressure as > 160 mm Hg and used systolic blood pressure treatment goals of 150–160 mm Hg or less.5–8 Virtually all of the epidemiologic data relating levels of blood pressure to cardiovascular risk, and the randomized, controlled morbidity trials in hypertension in older persons that defined levels of blood pressure benefiting from antihypertensive drug treatment, were based on blood pressure measurements taken during the day without regard to meals. Since these studies were conducted in ambulatory outpatients, blood pressure was not measured on arising before breakfast. Auseon et al. report that nursing home residents with systolic blood pressure of 160 mm Hg or greater recorded at least twice during the four measurements were more likely to have orthostatic hypotension than those with systolic blood pressure less than 160 mm Hg on at least three of the measurements, regardless of whether treatment for hypertension was taking place. Falls within the previous 6 months were not significantly different, however, regardless of blood pressure control or whether treatment was received for hypertension. Thus, there seems to be insufficient evidence at this time to support the authors’ recommendation for multiple within-day measures of blood pressure. This is an important issue because of the added burden multiple within-day measurements would place on physicians and patients, especially for the majority of older people who do not reside in a nursing home. Whether better control of the early morning rise in blood pressure is associated with reduced cardiovascular events is one objective of the Controlled ONset Verapamil INvestigation of Cardiovascular Endpoints (CONVINCE) trial, which compares a unique long-acting preparation of verapamil given at bedtime (Covera HS), designed to exert its peak antihypertensive effect during early morning hours, with usual care, initial therapy with a diuretic or beta-blocker taken in the morning.9 CONVINCE is not scheduled to be completed, however, until 2002. See also p 285 It is worth emphasizing that, although falls within the previous 6 months were not different in residents whose blood pressure was normal or elevated or treated or not treated with antihypertensive medications, orthostatic hypotension was significantly less frequent if systolic blood pressure was less than 160 mm Hg on most or all determinations, whether receiving antihypertensive medications or not. Even in a situation where there is uncertainty about a nursing home patient's likelihood of receiving cardiovascular risk reduction from long-term antihypertensive treatment because the resident is considered to have a limited survival, avoidance of postural hypotension by control of systolic blood pressure with antihypertensive therapy is a reasonable short-term goal. It is also likely, however, that many such patients would also benefit from reduction in incidence of heart failure, stroke, and coronary events, as suggested by the early separation of Kaplan-Meier curves seen in the Systolic Hypertension in the Elderly Program.5,10,11 Auseon et al.1 also report that the use of angiotensin-converting enzyme inhibitors or calcium channel blockers, but not diuretics, is more common in nursing home residents with elevated systolic blood pressure on two to four (of four) determinations during the day. Residents receiving a diuretic, either as single-drug therapy or in combination with another antihypertensive agent, or a calcium channel blocker as single-drug therapy, had better systolic blood pressure control than residents receiving an angiotensin-converting enzyme inhibitor alone or in combination with a calcium channel blocker, or those receiving all three of these classes combined. In a multivariate regression analysis, use of angiotensin-converting enzyme inhibitors or calcium channel blockers was associated with elevated systolic blood pressure, and use of diuretics was associated with better blood pressure control. These differences in drug regimens may not reflect their relative antihypertensive efficacy, however, because of selection bias: patients with more severe or resistant hypertension may be treated preferentially with angiotensin-converting enzyme inhibitors or calcium channel blockers. Lesser antihypertensive efficacy of angiotensin-converting enzyme inhibitors has been demonstrated previously in older black Americans12,13 although Auseon et al.1 do not report the proportion of black residents in their study. Diuretics are the drug class of choice for treatment of hypertension in older people, but not only because they lower systolic blood pressure well. Either alone or in combination with other agents, they have been demonstrated to lower morbidity and mortality in older persons with hypertension in at least six major clinical trials.5,6,14–18 One randomized, placebo-controlled, clinical trial, the Systolic Hypertension in Europe (Syst-Eur) trial,8 reported in 1997 that the dihydro-pyridine calcium antagonist nitrendipine reduced strokes and cardiovascular events in older patients with isolated systolic hypertension. Beta-blockers, especially as single-drug therapy, have not reduced most cardiovascular events in hypertension morbidity trials in older people.6,18 Therefore, although long-acting dihydropyridine calcium antagonists are an acceptable alternative to diuretics in isolated systolic hypertension,4 the number of studies and patients studied, the confirmed benefit in reducing coronary events and heart failure in addition to stroke, the lower cost, and the proven benefit in combination regimens indicate a diuretic-based regimen should be preferred in hypertension in older individuals until further evidence is available to the contrary. There are several hypertension morbidity trials currently in progress that compare two or more antihypertensive agents. The largest of these is the National Heart, Lung and Blood Institute (NHLBI) sponsored Antihypertensive and Lipid Lowering to Prevent Heart Attack Trial (ALLHAT).19 Initiated in 1994, the hypertension portion of the study is a randomized controlled trial comparing a diuretic (chlorthalidone), a dihydropyridine calcium antagonist (amlodipine), an angiotensin-converting enzyme inhibitor (lisinopril), or an alpha1-blocker (doxazosin) in 42,000 high-risk older hypertensive patients. Unless ALLHAT is stopped early because of compelling differences in treatment groups, however, it is not due to be completed until 2001. The article by Auseon and colleagues in this issue of JAGS1 offers reassurance that control of systolic blood pressure below 160 mm Hg is associated with reduction in the prevalence of orthostatic hypotension in frail older residents of nursing homes and demonstrates that diuretic-based therapy is associated with better blood pressure control than other classes of drugs in this population. Until ALLHAT or other comparative morbidity trials in older hypertensive patients are completed, we will not know whether classes of drugs other than diuretics will reduce cardiovascular events in older individuals as well as or better than diuretics.