Treatment of Octogenarians: Should We and How?

THE ARGUMENT FOR—Robert A. Phillips, MD, PhD, Lenox Hill Hospital and New York University School of Medicine, New York, NY

The weight of the current evidence is that octogenarians should be treated. This assertion is based on pathophysiology, epidemiology, trial data, and clinical experience. Nevertheless, there are several controversial issues surrounding hypertension in octogenarians. These include questions about benefits of treatment, the use of ambulatory blood pressure (BP) monitoring to determine who should be treated, the goals of therapy, and the choice of antihypertensive agents.

Pathophysiology of Hypertension in Octogenarians

The first issue to be settled is what is the pathophysiology of hypertension in an octogenarian? While many octogenarians have combined systolic and diastolic hypertension, isolated systolic hypertension with low diastolic BP is the predominant form of hypertension in the elderly. This pattern is due to a combination of increased systemic vascular resistance and decreased vascular compliance that is caused by replacement of elastin by collagen in the arterial wall of the conduit vessels. The mechanism for the elevation in systolic BP is obvious: with each stroke volume there is an exaggerated rise in systolic BP because of limited expansion of the noncompliant conduit vessel. But why does diastolic BP go down even in the face of increased systemic vascular resistance? Due to increasing stiffness, the aorta and conduit vessels lose their capacitance function, which results in run‐off of blood volume into the periphery during systole, and therefore there is less blood in the arterial tree during systole and lower diastolic BP. ¹ Hence, paradoxically, in the elderly, higher systolic pressure, in combination with lower diastolic BP, is more reflective of vessel pathophysiology. ² Because of this, in patients over age 65 years, and especially in octogenarians, it has become the habit in my practice not to report the diastolic BP because the lower number provides a sense of false security.

Trials of BP Lowering in the Elderly

Given that the elevated BP represents pathology, how do we know we should lower BP in octogenarians? Several randomized, placebo‐controlled trials that included patients over age 80 years give us insight into this issue. The Systolic Hypertension in the Elderly Program (SHEP) ³ found that treatment of hypertension had equivalent reduction in stroke incidence among octogenarians as it did for younger patients. Similarly, the Swedish Trial in Old Patients with Hypertension (STOP‐ Hypertension), ⁴ demonstrated an effect of antihypertensive treatment on cardiovascular morbidity and mortality, as well as total mortality, that was discernible up to age 84 years. In the Study of Cognition and Prognosis in the Elderly (SCOPE), ⁵ in which patients were included up to age 89 years, the group with the lower BP had fewer non‐fatal strokes. In the pilot study of Hypertension in the Very Elderly Trial (HYVET), ⁶ which is an ongoing placebo‐ controlled trial evaluating efficacy of treating hypertension in persons older than age 80 years, antihypertensive treatment was associated with a significant reduction in stroke, but a nonsignificant trend toward increased mortality.

Another line of evidence to support treating octogenarians is that compared with placebo, Antihypertensive treatment leads to an equal relative risk reduction from stroke across all age groups, but a linear increase in absolute benefit in stroke reduction as a function of the age of the study participant. MRC I=Medical Research Council Trial I; MRC II=Medical Research Council Trial II; AUST=Australian National Blood Pressure Study; SHEP=Systolic Hypertension in the Elderly Program; EWPHE=European Working Party Trial on High Blood Pressure; Coope=Coope and Warrender; STOP=Swedish Trial in Old Patients with Hypertension. Adapted with permission from J Hypertens. 1995;13:571‐5797 antihypertensive treatment leads to an equal relative risk reduction from stroke across all age groups, but a linear increase in absolute number of strokes prevented as a function of the age of the study participant (Figure). ⁷ Therefore, there is no adverse effect observed in treating the elderly (i.e., relative risk reduction is equivalent to treating a young person), and more strokes are prevented because the risk for stroke in the elderly is so high if BP is left untreated.

Figure.

Antihypertensive treatment leads to an equal relative risk reduction from stroke across all age groups, but a linear increase in absolute benefit in stroke reduction as a function of the age of the study participant. MRC I=Medical Research Council Trial I; MRC II=Medical Research Council Trial II; AUST=Australian National Blood Pressure Study; SHEP=Systolic Hypertension in the Elderly Program; EWPHE=European Working Party Trial on High Blood Pressure; Coope=Coope and Warrender; STOP=Swedish Trial in Old Patients with Hypertension. Adapted with permission from J Hypertens. 1995;13:571–579. ⁷

Is Treatment Associated With Adverse Events?

One of the concerns in treating elderly persons is that they will have increased falls, fractures, and orthostatic symptoms, or even decreased cerebral blood flow and reduced cognitive function. In SHEP, where the average age of the patients was 72 years, BP reduction was not associated with more hip fractures; nor did it lead to more orthostatic symptoms. ³ While treatment in SHEP did not lower the incidence of dementia, in the Systolic Hypertension in Europe (Syst‐Eur) study, ⁸ treatment of isolated systolic hypertension, with a long‐acting dihydropyridine calcium channel blocker as the initial therapy, was associated with a 55% reduction in risk of dementia.

This brings us to what I believe are three debatable points: Should we be basing treatment decisions on ambulatory BP monitoring? How low should BP be lowered in the elderly? Is it the BP lowering that matters, or are some drugs more effective in treating the elderly?

Ambulatory BP Monitoring

In the Syst‐Eur study, it is perhaps no surprise that subjects with white coat hypertension did not get benefit from treatment. However, it is remarkable that no benefit was seen among treated subjects who had office hypertension and clearly elevated daytime systolic BP between 140–159 mm Hg. ⁹ The only patients who clearly benefited from treatment were those who not only had white coat hypertension, but whose daytime systolic BP was &160 mm Hg. Based on these results and a wealth of other data, I am comfortable recommending that we do not treat elderly patients with white coat hypertension without end organ damage or diabetes. However, until more studies corroborate a lack of benefit associated with treating an ambulatory systolic BP between 140–159 mm Hg, I would continue to treat patients in this group.

Treatment Goals

How far should BP be lowered in an octogenarian? Two lines of evidence support a systolic goal of <150 mm Hg as sufficient. In SHEP, reduction in systolic BP from 170 mm Hg to < 160 mm Hg lowered stroke rate by 33%, and if BP was lowered to <150 mm Hg, stroke rate was reduced an additional 5%. ¹⁰ However, in SHEP, there was no further reduction in stroke rate in those subjects who attained a systolic BP <140 mm Hg. Furthermore, in SHEP, relative risk for a cardiovascular event appeared to increase in those who achieved a diastolic BP <65 mm Hg. ¹¹ While SHEP May have been underpowered to detect a benefit in subjects who achieved a systolic BP <140 mm Hg, and lower achieved diastolic BP in isolated systolic hypertension most likely represents sicker vessels, we do not have good evidence for lowering BP in octogenarians with isolated systolic hypertension to <150/65 mm Hg. What about an octogenarian with elevated systolic and diastolic hypertension? The target BP for the ongoing HYVET trial, which is recruiting patients aged ≥80 years with these parameters, is <150/80 mm Hg. Because active treatment to these levels in the pilot trial was associated with stroke reduction, this goal is clearly an appropriate target.

Treatment

What treatments should we be using for octogenarians? It is never too late for lifestyle modification. Regular aerobic exercise at comfortable heart rates (72% maximal) in previously sedentary elderly men can significantly improve endothelial function. ¹² The Dietary Approaches to Stop Hypertension (DASH) diet is very effective for lowering BP even in patients with isolated systolic hypertension. ¹³ Sodium restriction in the elderly clearly lowers BP.

Which drugs should we be using in the elderly? My strategy is based on analyzing the multitude of trials from the perspective of whether the patient population that was enrolled had complicated or uncomplicated hypertension. In studies of uncomplicated hypertension, such as the Swedish Trial in Old Patients with Hypertension‐2 (STOP 2), ¹⁴ the Nordic Diltiazem (NORDIL) study, ¹⁵ and the International GITS Study: Intervention as a Goal in Hypertension Treatment (INSIGHT), ¹⁶ all drug classes appear to be equally effective in reducing events. Calcium channel blockers in these trials are consistently associated with higher incidence of congestive heart failure, but appear to confer protection against stroke and dementia. In this context, I view the Second Australian National BP Study (ANBP2) as a trial in uncomplicated hypertension in the elderly, because <10% of the population had prior cardiovascular disease. ¹⁷ The fact that an angiotensin‐ converting enzyme (ACE) inhibitor was superior to diuretic in ANBP2 is not consistent with all of the other trials of uncomplicated hypertension. Rather than view results of ANBP2 as a reason to use ACE inhibitors in lieu of diuretics in the elderly, I see it as support for using ACE inhibitors as one of the multiple drugs that is often needed in the elderly to control BP.

On the other hand, in elderly hypertensive patients with cardiovascular disease several lines of evidence suggest that, despite the benefits associated with ACE inhibitors and β blockers, diuretics should be part of the antihypertensive regimen. In the Antihypertensive and Lipid‐Lowering to Prevent Heart Attack Trial (ALLHAT), ¹⁸ where nearly 50% of the population had a prior cardiovascular event, diuretics were equally effective in reducing composite cardiovascular events when compared with ACE inhibitors and calcium channel blockers. In ALLHAT, diuretics demonstrated superiority with regard to prevention of congestive heart failure when compared with the calcium channel blocker and were superior to the ACE inhibitor for prevention of heart failure and stroke. In this context, an ACE inhibitor was only shown to be effective in reducing recurrent stroke when it was combined with a thiazide diuretic (the Perindopril protection against Recurrent Stroke Study [PROGRESS]). ¹⁹ And in the recently reported International Verapamil‐Trandolapril (INVEST) trial, ²⁰ which enrolled a large number of patients older than age 70 years with concomitant hypertension and coronary artery disease, an atenololhydrochlorothiazide‐based strategy was equally effective as the verapamil‐trandolapril strategy in preventing death and cardiovascular outcomes in this high‐risk population. Similarly, even in the setting of left ventricular hypertrophy, diabetes, or renal disease, where it appears that treatment of hypertension with blockade of the renin‐angiotensin aldosterone system is beneficial with either an ACE inhibitor or an angiotensin II receptor blocker, careful analysis of the studies shows that the overwhelming majority of patients required diuretics for BP control. ²¹ ' ²² Of course, it is always important to maintain normokalemia, since in the SHEP study patients with diuretic‐induced hypokalemia did not have a reduction in cardiovascular events, ²³ and many elderly patients are intolerant of diuretics and should have BP controlled with other agents. The emerging debate on the degree to which diuretics cause diabetes should not take our eyes off of the prize—in randomized, controlled clinical trials, diuretic‐based therapy consistently lowers cardiovascular risk and therefore should be part of the regimen for treatment unless there are compelling contraindications.

THE ARGUMENT AGAINST—John B. Kostis, MD, Department of Medicine, University of Medicine and Dentistry of New Jersey‐Robert Wood Johnson Medical School, New Brunswick, NJ

The question of whether to treat octogenarians with hypertension with pharmacologic therapy is not easy to answer because of the lack of large controlled clinical trials addressing the issue and its ramifications. There is, however, relevant information that can be reviewed in an attempt to answer questions pertaining to this problem.

Are There Enough Octogenarians to Make This Question an Important Health Care Issue?

It is widely known that life expectancy at birth has increased markedly. In the past 150 years it has increased from less than 50 years to more than 75 years. This increase in average life expectancy is in large part due to markedly lower infant mortality and fewer deaths from infectious diseases, factors that do not necessarily lead to an increase in the number of very old persons. However, an increase in the number of octogenarians has also occurred. According to census data, there are more than 9 million octogenarians in the United States (about one out of 30 persons). It is estimated that the number will rise to over 15 million (4.5% of the population) in the year 2025 and to more than 33 million in the year 2050 when this group will make up 8% of the population. ¹

Is Hypertension Prevalent Among Octogenarians?

Data from National Health and Nutritional Examination Survey (NHANES) and other sources have shown that although diastolic blood pressure (BP) decreases after age 55 or 60 years, there is a continuous increase in systolic BP with age through the 80s. This occurs in both men and women, in African Americans, Mexican Americans, and whites. The residual lifetime risk of hypertension inpersons aged 65 years is more than 90%. ²

Does Hypertension Impose Health Risks in Octogenarians?

The mere presence of the large number of octogenarians with hypertension does not necessitate treatment unless hypertension is harmful in this age group. Data from epidemiologic studies including more than a million persons indicate a continuous relationship of BP and increasing risk for stroke and coronary heart disease mortality starting with a systolic BP of 110 mm Hg. This relationship pertains to all age groups including those over age 80 years. Actually, in octogenarians, the absolute increase in risk for a given increase in systolic BP is much higher than in younger persons because of the increased baseline risk in older persons. ³ The situation is different when total mortality is considered. Two European studies carried out in Tampere, Finland and in Leyden, The Netherlands showed an inverse relationship between BP and mortality in persons aged 85 years and older. ⁴ , ⁵ In the United States, the Established Populations for Epidemiological Studies of the Elderly (EPESE) trial also showed that higher systolic BP was associated with better survival in men aged 85 years and older ⁶ (Table). We can only speculate about the causes underlying this relationship. It is possible that high BP is a marker of physiologic vigor or that a normal BP in some persons May be due to lowering of BP as a result of cardiovascular or other disease. This cannot be the only explanation because statistical adjustment for comorbidities does not eliminate the relationship.

Is There Clinical Trial Evidence That Antihypertensive Drug Therapy Benefits the Very Old?

Clinical trials have included only a small number of octogenarians. The total number of participants older than age 80 years in published reports of placebo‐controlled clinical trials is lower than 2000. The results in patients aged ≥80 years are consonant with findings of the epidemiologic studies mentioned above. In these studies, antihypertensive drug therapy was associated with a decreased rate of stroke, heart failure, and coronary events, but with inconclusive results for total mortality (6% increase), ⁷ , ⁸ as shown in the Systolic Hypertension in the Elderly Program (SHEP) (Figure 1). When only double studies were considered, a statistically significant 14% increase in total mortality was observed ⁸ (Figure 2). These data are different from those in the younger age groups where a decrease in fatal and nonfatal events as well as in total mortality has been observed. The reasons for the excess mortality with antihypertensive therapy in clinical trials in the very old are not well understood. It is possible that the (rather small) number of persons studied were not a representative group of the very old or that the findings are due to chance. It is also possible that competing risk from cancer and other diseases, which is much higher in octogenarians than in younger persons, obscured the benefit of decreasing cardiovascular death. It has also been proposed that persons susceptible to cardiovascular disease resulting from hypertension had died at a younger age, leaving a selected population who could tolerate their high BP well. It is also possible that the very old exhibit higher level of serious toxicity to the drugs used for treating hypertension or that they require a higher BP for perfusion of vital organs. Treating hypertension in the very old requires careful attention. The medications available to us are effective in lowering BP. However, impaired cardiac function, impaired renal function, decreased baroreceptor susceptibility resulting in orthostatic hypotension, subjective side effects, and polypharmacy with increased risk of drug interactions should be considered in prescribing for the very old. In the Hypertension in the Very Elderly Trial (HYVET), orthostatic fall in systolic BP by ≥20 mm Hg occurred in 7.7% of subjects. ⁹

Figure 1.

Effect of antihypertensive therapy on the occurrence of heart failure, by age group, in the Systolic Hypertension in the Elderly Program (SHEP). The absolute risk reduction is more pronounced in those aged ≥80 years. Rx=prescription. Adapted from JAMA. 1997;278:212–216. ⁷

Figure 2.

Effect of treating hypertension in persons aged ≥80 years in double‐blind clinical trials. The y‐axis shows the relative risk of the active treatment group compared with placebo. While cardiovascular events are decreased with antihypertensive therapy, all‐cause mortality is higher. The 95% confidence intervals are depicted by the vertical lines. HF=heart failure. Adapted from Lancet. 1999;353:793–968 and J Gerontol. 2003;58A:653–658. ¹²

Are All Octogenarians the Same With Respect to Hypertension and Its Treatment?

Chronological age does not reflect well the physiological age of the cardiovascular system or other organs. Persons age at different rates as a result of differences in heredity, lifestyle, and disease. Variation in these factors creates differences in health status that are magnified in the very old. A large percentage of octogenarians are independent, active, in relative good health, and able to lead productive and satisfying lives whereas others suffer from disease, infirmity, and dependence.

Is It Too Late to Treat Hypertension in the Very Old?

One May consider that in this age group hypertension has already resulted in severe vascular damage that is now irreversible or alternatively that persons who have tolerated hypertension to a very old age have developed cardiovascular adaptations to this condition and a high BP May be necessary for continuing good health. In the absence of data, we can only speculate about this question and extrapolate from what we know in younger age groups. The history of clinical trials in hypertension shows that when a question arose about the advisability of treating hypertension in a given subset (e.g., treating isolated systolic hypertension or treating women) the answer was always in the affirmative. Accumulating data show the significant degree of plasticity in many functions in the older age group.

Is It Worth It to Treat Hypertension in the Very Old?

Financial considerations and economic analyses are not a primary role of physicians whose duty it is to do the best they can for all their patients. However, in the current health care environment these questions are asked. Increasing demand for services because of aging of the population and the development of new expensive therapeutic and diagnostic modalities have resulted in high health care expenditures that consume a significant percentage of the gross domestic product. A common approach in evaluating the cost benefit of therapeutic interventions is to calculate life years gained or quality adjusted life years (QALY) gained. In such calculations, there is more benefit in treating younger persons than the very old who have a short life expectancy. However, when one expresses the life years gained or QALYs gained as a percentage of remaining life expectancy, the differences are attenuated and the gain in QALYs is higher for the very old. ¹⁰ In addition, the benefit to cost ratio of antihypertensive therapy is high in any case because cardiovascular events such as stroke or heart failure carry such a great lifetime expense that prevention of one allows enough savings to allow treatment for many.

More than in the young it is important to individualize therapy and consider patient preference in octogenarians. Lowering BP decreases the risk of stroke, ischemic events, and heart failure, complications of hypertension that are feared by many older persons. They would prefer to be treated with as few side effects as possible. Accepting a suboptimal BP lowering in these persons is better than not treating at all. In both clinical decisions and the design of randomized clinical trials, it is important to consider the great variability among the very old and to limit the questions to more narrowly defined cohorts. ¹¹ , ¹² HYVET, now in progress, is a double‐blind, placebo‐controlled trial designed to assess the benefits of treating very elderly patients with hypertension.