Update in Geriatrics

Common Office Problems

The management of medications in elderly patients is an everyday office issue. Polypharmacy is a common problem, and therapy with most, but not all, drugs can apparently be safely discontinued. Benzodiazepines are especially problematic and may actually increase the prevalence of insomnia. Isolated systolic hypertension continues to be a difficult problem, and appropriate therapy decreases symptomatic episodes of heart failure. Some data suggest the interesting possibility that treatment of hypertension may actually mitigate orthostatic hypotension. Finally, an ancient herbal remedy was found to improve the course of some patients with dementia, and the recognition of the role of various micronutrients in maintaining health was enhanced.

Therapy with Most Drugs Can Be Safely Stopped

Graves T, Hanlon JT, Schmader KE, et al. Adverse events after discontinuing medications in elderly outpatients. Arch Intern Med. 1997; 157:2205-10.

Patients in their eighties are commonly prescribed 8 to 12 medications, and the potential dangers of adverse effects with such polypharmacy are well known. Elderly persons taking more than 5 medications simultaneously have about a 35% chance of developing an adverse drug reaction per year. About 70% of these persons will require a physician visit, and 10% will be hospitalized. But is stopping drug therapy any safer than continuing it?

Graves and colleagues assessed the safety of discontinuing therapy with various drugs in office practice. For 1 year, they followed 124 ambulatory elderly patients who had stopped taking at least one medication. All of these volunteers were older than 65 years of age, were taking at least five medications, and were patients in a general Veterans Affairs medicine clinic. After 1 year, charts were carefully reviewed.

Discontinuation of therapy with 238 drugs resulted in 72 adverse effects among 38 patients. Table 1 lists the implicated drugs. Adverse drug withdrawal events occurred most commonly in patients who stopped taking cardiovascular (42%) and central nervous system (18%) drugs. Eighty-eight percent of adverse drug withdrawal events represented exacerbations of underlying disease, and the remainder were physiologic withdrawal events. About 36% of these events resulted in hospitalization or some type of acute care visit to a physician. Adverse drug withdrawal events occurred as long as 4 months after cessation of drug therapy. No problems occurred in 74% of drug cessation actions, and 80% of the drugs whose use was discontinued were not restarted or replaced.

These findings suggest that when a drug is being given for an uncertain indication, physicians must have the courage to discontinue its use. However, they should be aware of the substantial time lag for development of some adverse drug withdrawal events, primarily those related to drugs that act on the circulatory system (for example, drugs for angina, hypertension, and congestive heart failure). Finally, caution is needed in extrapolation of these findings to other practice situations because patterns of illness and medication use among patients in the Veterans Affairs system may differ from patterns in other practice settings.

Controlling Systolic Hypertension Decreased Episodes of Congestive Heart Failure

Kostis JB, Davis BR, Cutler J, et al. Prevention of heart failure by antihypertensive drug treatment in older persons with isolated systolic hypertension. SHEP Cooperative Research Group. JAMA. 1997; 278:212-6.

The incidence of heart failure is steeply increasing in elderly persons; this event now causes more than 1 million hospitalizations annually. Among Medicare recipients, it is by far the most common reason for hospitalization. One of the key risk factors for heart failure is inadequately treated hypertension. For at least 10 years, significant information on the importance of appropriate therapy for isolated systolic hypertension has come from the Systolic Hypertension in the Elderly Program (SHEP) study. Isolated systolic hypertension, defined as systolic blood pressure of at least 160 mm Hg, occurs in about 5% of persons in their sixties, 12% of those in their seventies, and 25% of those in their eighties.

The SHEP trials assessed the effect of therapy for systolic hypertension on several key clinical outcomes [1]. Step-care treatment, beginning with low-dose thiazides and proceeding to β-blockers, reduced the incidence of stroke by about 36% and the incidence of major cardiovascular events by 32%. Moreover, it defined the clinical importance of systolic blood pressure control in elderly persons.

In an analysis of the original SHEP data, Kostis and colleagues sought to determine whether step-care therapy for isolated systolic hypertension would decrease the incidence of heart failure in addition to reducing the incidence of stroke. The study group consisted of 4736 persons older than 60 years of age who had a systolic blood pressure of 160 to 219 mm Hg (mean, 170 mm Hg) and a diastolic blood pressure less than 90 mm Hg. Step 1 treatment was chlorthalidone (12.5 to 25 mg/d) or placebo, and step 2 treatment was atenolol (25 to 50 mg/d) or placebo. After 4.5 years, heart failure occurred in 2.3% of patients receiving treatment and 4.4% of those receiving placebo (relative risk reduction, 49%; number needed to treat [NNT] to prevent one episode, 48).

Like the findings seen with stroke and overall heart disease, the results of Kostis and colleagues' study represent a major advance, both for individual patients and public health. Some subgroups had even more impressive results. For example, among patients with a history of myocardial infarction, the risk for heart failure declined by 81% when hypertension was treated (NNT, 15). Treatment was also more effective in older patients and in men.

The management of patients at risk for heart failure has become a major focus for reducing costs of hospitalization, and control of isolated systolic hypertension is a critical element. However, a remaining question is what to do with the 10% of patients whose hypertension does not respond to this step-care program. This question remains unanswered, but ambulatory case management may offer promise.

One outcome not yet shown to improve with therapy is life expectancy; indeed, these treatments may not extend life. However, freedom from episodes of heart failure implies great improvement in quality of life and functionality.

Orthostatic Hypotension Was Associated with Morning Hypertension

Ooi WL, Barrett S, Hossain M, et al. Patterns of orthostatic blood pressure change and their clinical correlates in a frail, elderly population. JAMA. 1997; 277:1299-304.

Orthostatic hypotension is a common problem among elderly persons that contributes to syncope, falls, and organ ischemia [2]. Its prevalence in the elderly ranges from 5% to 30%. Ooi and colleagues asked why that range was so wide and whether blood pressure varies during the day and with various activities.

They recruited 911 nursing home residents and obtained blood pressure and heart rate measurements while the residents were supine and after they had stood for 1 and 3 minutes. These measurements were done before breakfast, 30 to 60 minutes after breakfast, before lunch, and 30 to 60 minutes after lunch. Orthostatic hypotension was defined as a 20-mm Hg decline in systolic blood pressure or a 10-mm Hg decline in diastolic blood pressure 1 or 3 minutes after the resident's position had switched from supine to standing. Patients were asked about symptoms and were observed for signs.

Orthostatic hypotension occurred at some point in more than 50% of volunteers. It was most prevalent before breakfast and least prevalent after lunch. One of the most astonishing findings was that the single most predictive factor for orthostatic hypotension was the extent to which systolic blood pressure in the supine position was elevated before breakfast (mean pressure, 139 mm Hg). These elevations did not necessarily meet the criteria for systolic hypertension. In contrast, however, use of antihypertensive medications did not correlate with orthostatic hypotension. Other associations included dizziness on standing, low body mass index, being male, use of medication for Parkinson disease, and greater independence in activities of daily living (this last was assumed to be due to increased mobility and postural changes).

This study was particularly interesting because it not only confirmed the high prevalence of orthostatic hypotension but also found that prebreakfast high systolic blood pressures were excellent predictors. In fact, 21% of patients with orthostatic changes had isolated systolic hypertension before breakfast, whereas isolated systolic hypertension was present in only 12% of patients after breakfast. Moreover, treatment of hypertension, if anything, improved orthostatic blood pressure changes.

For clinical practice, these findings mean that problems associated with orthostasis are most likely to occur in the morning, a very busy time in nursing homes. In addition, one might hypothesize that aggressive control of isolated systolic hypertension may alleviate this problem, although the issue needs further study. Finally, because of the change in blood pressure throughout the day, physicians should not base treatment decisions on the blood pressure measured at only one point during the day.

Insomnia Was Associated with Benzodiazepine Use

Newman AB, Enright PL, Manolio TA, et al. Sleep disturbance, psychosocial correlates, and cardiovascular disease in 5201 older adults: the Cardiovascular Health Study. J Am Geriatr Soc. 1997; 45:1-7.

Sleep problems occur in about 40% of older adults and are a common reason why patients seek medical treatment. Specific conditions, such as the restless legs syndrome and sleep apnea, are often considered, but these are infrequently the cause of insomnia. More often, patients offer nonspecific symptoms, such as “I just can't get to sleep,” “I get to sleep but I wake up,” and “I nap during the day.”

The Cardiovascular Health Study is a cross-sectional study of 5201 elderly persons living in four communities in North Carolina, Maryland, Pennsylvania, and California. Although this study has focused on the demographic characteristics associated with heart failure and respiratory function in older persons, the participants were also asked several questions about sleep.

The prevalence of sleep disturbances in these persons is shown in Table 2. Independent factors highly associated with poor sleep initiation included depression (odds ratio, 3.9) and benzodiazepine use (odds ratio, 2.2). Both of these findings can help in daily medical practice. First, sleep disturbance may be the initial clue to depression or poor self-perception. Second, although the association between benzodiazepine use and insomnia does not identify cause and effect, it does raise the possibility that benzodiazepine use may actually aggravate the problems of insomnia. Thus, a remedy for sleep problems is implied: Stop benzodiazepine therapy. Physicians often see patients who have been using benzodiazepines to help them sleep and then present with a request for “stronger” medicine because the benzodiazepine is no longer working as well.

The seriousness of the problem of insomnia and its treatment is highlighted by the results of another study of 224 734 older Canadian drivers, 2.5% of whom were involved in automobile accidents over a 4-year period [3]. Long-acting benzodiazepine use increased the chance of an accident by 45% during the first week of use and 26% thereafter. Short-acting benzodiazepines were not associated with accidents. Thus, even brief use of long-acting benzodiazepines, which are often prescribed for insomnia, can be dangerous. A logical conclusion is that careful removal of benzodiazepines from a patient's drug regimen may improve the quality of sleep and safety. Weaning patients from long-term benzodiazepine use can be difficult. If a patient is receiving a long-acting benzodiazepine, therapy should shift to a short-acting one. After this change, complete tapering of the medication dose may take several weeks to months.

In a 4-month clinical trial of 43 healthy older adults, sleep quality improved when participants engaged in the equivalent of 30- to 40-minute sessions of brisk walking four times a week [4].

Gingko Extract Slowed the Progression of Dementia

Le Bars PL, Katz MM, Breman N, et al. A placebo-controlled, double-blind, randomized trial of an extract of Ginkgo biloba for dementia. North American EGb Study Group. JAMA. 1997; 278:1327-32.

Alzheimer disease is another illness common to every internist's practice; the prevalence of the disease is 11% in persons 85 years of age or older. The disease places a great burden on both patient and caregiver, and even small improvements in cognitive function are extremely helpful.

The tree Ginkgo biloba, named for its bi-lobed leaf, is indigenous to Asia but has been transplanted around the world. One common medicinal extract, called EGb 761, has been part of Chinese medicine for about 1000 years. This extract is also widely used in Europe for various cognitive and circulatory disturbances. It is one of the most commonly prescribed drugs in Germany and is primarily given for memory loss.

Surprisingly little is known about the effects of Ginkgo biloba extract, but it is presumed that a synergistic action of the extract's constituents creates an antioxidant compound that is thought to be a potent scavenger of free radicals.

Because of European claims of efficacy, Le Bars and colleagues assessed the efficacy and safety of EGb 761 in patients with mild to severe Alzheimer disease and multi-infarction dementia. They randomly assigned 302 patients to receive EGb 761 (40 mg three times daily before meals) or placebo. Patients were evaluated at 12, 26, and 52 weeks. Primary outcome measures were standardized cognitive scales used in most dementia research.

Cognitive performance and social functioning improved modestly. According to the cognitive scales, the difference between the two groups was only about 2.5%; in addition, the study has been criticized because of the dropout rates (53% in the EGb 761 group and 63% in the placebo group). Because the analysis was done by intention to treat, however, the positive effects seen were probably minimized by the dropout rate.

In clinical terms, progression of symptoms was delayed by approximately 6 months; this delay is similar to that seen in patients who respond to donepezil. Moreover, the patients' caregivers reported substantial relief of the burden associated with this role.

In summary, this fairly small study, marked by a high dropout rate, reported that an extract of Ginkgo biloba did show some promise in slowing the progression of Alzheimer disease and easing the burden on caregivers. No conclusions can be drawn about long-term benefits, but this study does demonstrate that more studies on the efficacy and safety of alternative therapies are needed.

Should physicians now recommend Ginkgo biloba extract to patients with dementia? I do not believe so. But if patients say, “I don't care what you think, I'm going to take it anyway,” physicians should at least be aware that it is probably a safe medication that may have modest therapeutic benefit.

Vitamin E Improved Immune Responses

Meydani SN, Meydani M, Blumberg JB, et al. Vitamin E supplementation and in vivo immune response in healthy elderly subjects. A randomized, controlled trial. JAMA. 1997; 277:1380-6.

It is well known that aging may be associated with relative immunodeficiency. Even healthy elderly persons have alterations in delayed-type hypersensitivity, decreased lymphocyte response to antigenic stimulation, and low rates of antibody response to antigen. In addition, it has been postulated that malnutrition worsens these problems for many older persons. As a result, much interest has focused on immune modulators, such as vitamin E.

In Meydani and colleagues' clinical trial, 88 healthy elderly persons were randomly assigned to receive placebo or one of three vitamin E regimens (60, 200, or 800 mg/d) for 235 days. Outcome measures were delayed-type hypersensitivity to skin tests, antibody response to common vaccines, and presence of autoantibodies to DNA and thyroglobulin; these measures were determined before and after treatment.

The results were striking. With 200 mg/d, the dosage that seemed to have the greatest effect, there were a 65% enhancement in delayed-type hypersensitivity, a sixfold increase in the titer of antibody to hepatitis B virus, significant increases in response to tetanus, no great effect on total immunoglobulin levels, and no effect on autoantibody titers.

Although this study was small and did not measure long-term clinical outcomes, it was carefully done and did demonstrate that vitamin E supplementation may influence physiologic responses in older healthy adults. A dosage of 200 mg/d seemed to be ideal in terms of antibody response.

Many older persons probably do not obtain adequate vitamin E in their diet through vegetable oils, nuts and seeds, and wheat germ. Indeed, previous studies have shown that one third of all elderly persons may have deficiencies of vitamins and trace elements.

One caveat is that although the dosage ranges used in this study (60 to 800 mg/d) were not associated with toxicity, it is known that much higher doses of vitamin E can antagonize the function of other fat-soluble vitamins. The results are a decrease in bone mineralization, reduced hepatic storage of vitamin A, and disorders of coagulation. This study is intriguing in the wake of another study published in 1997 that showed improved cognition in older persons who were taking vitamin A and vitamin C [5]. Clearly, more research on this entire area of micronutrient supplementation will be forthcoming.

Hospital Medicine

Physicians in practice are often told that what they are doing is wrong. In hospital practice, however, much is being done right. The rate of hospitalization for heart failure is increasing, but so is survival. Patients with pneumonia need antibiotics quickly, and most get them, although use varies greatly by region of the United States. Compliance with national pneumonia guidelines is low, but these guidelines may not be ideal for elderly, community-dwelling patients.

Hospitalizations for Congestive Heart Failure Continue To Increase

Croft JB, Giles WH, Pollard RA, et al. National trends in the initial hospitalization for heart failure. J Am Geriatr Soc. 1997; 45:270-5.

As mentioned above, heart failure is now the most common reason for initial and repeated hospitalization in elderly persons. In 1986, about 1.2 million hospitalizations per year were attributed to heart failure, more than double the number in 1973. Most of that increase has been among adults older than 75 years of age, and this group of elderly persons makes up more than 90% of those who die of heart failure. In the past decade, however, the treatment of heart failure has been improved through the use of vasodilators, especially angiotensin-converting enzyme (ACE) inhibitors.

In a study of 804 000 Medicare claims, Croft and colleagues described the characteristics of patients with heart failure whose initial hospitalization was for heart failure. The outcomes in these patients were compared with those in a similar group studied in 1986.

Overall, the rate of hospitalization increased by about 10% between 1986 and 1993, to 24.7 hospitalizations per 1000 person-years. Figure 1 shows that the burden is heaviest among the very old, the fastest-growing age group in the United States. The in-hospital mortality rate decreased to 10.3% by 1993, a 21% decline over 7 years. However, the rate of discharges to skilled nursing facilities increased by 49% and the rate of discharges to home decreased by 15%.

These data reveal much, but they also create new and difficult questions. First, heart failure could be called an emerging epidemic because a 10% increase represents about 60 000 more hospitalizations annually. Second, because more patients are being discharged to nursing homes, one might wonder whether these patients are sicker, whether we are discharging them from hospitals too soon, or whether we lack adequate subacute or transitional care units. Third, an increasing number of these episodes of heart failure was associated with hypertension as a secondary diagnosis (23% in 1993 compared with 17% in 1986); this finding should prompt clinicians to repeatedly assess their patients' reversible causes of heart failure, such as those shown in Table 3, before hospitalization becomes necessary.

The starting point for therapy is to accurately determine whether a patient has a depressed left ventricular ejection fraction [6]. If he or she does (and most patients do), the treatment is diet, diuretics, and ACE inhibitors. Digoxin can be added if symptoms continue. Isosorbide dinitrate also has an important secondary role, as does the very selective use of β-blockers. Even among patients with diastolic dysfunction, the usual initial therapy is a low-sodium diet and an ACE inhibitor.

Rapid Antibiotic Initiation Improved Pneumonia Outcomes

Meehan TP, Fine MJ, Krumholz HM, et al. Quality of care, process, and outcomes in elderly patients with pneumonia. JAMA. 1997; 278:2080-4.

Pneumonia accounts for 600 000 hospital discharges and 6 million days of hospitalizations. It is second to congestive heart failure in terms of frequency of hospitalization in elderly persons. Despite the availability of good antibiotic therapy, pneumonia remains the sixth leading cause of death in the United States. The fact that pneumonia has remained a major killer despite mostly effective antibiotics raises questions about the relation between the process of care and outcomes in patients with pneumonia.

This association was the focus of Meehan and colleagues' retrospective cohort study of 14 069 Medicare-eligible patients admitted to 3555 hospitals with community-acquired pneumonia. The researchers sought to determine whether the process of care helps identify patients who might have lower 30-day mortality from pneumonia. By reviewing medical records, Meehan and colleagues examined length of time from initial clinical encounter to initiation of antibiotic therapy, whether and when blood cultures were obtained, and whether oxygenation was assessed.

The results are summarized in Table 4. Simply initiating antibiotic therapy within 8 hours of the initial encounter resulted in a 15% decrease in the mortality rate. When antibiotics were given within 6 hours, the mortality rate decreased by 50%. The researchers believe that the higher association between mortality rate and oxygenation assessment reflects the fact that the testing was done in sicker patients.

The disturbing aspect of this study is the regional variation. Throughout the United States, about 75% of patients received antibiotics within 8 hours, but the geographic range was 49% to 90%. Not surprisingly, the mortality rate ranged from 9% to 24% according to geographic location. For the clinician, this study confirms the value of acting quickly. The American Thoracic Society has urged for years that action to treat pneumonia should be taken quickly, usually on a clinical basis, and even without chest radiography if that would delay initiation of therapy.

Pneumonia Guidelines Were Costly in Elderly Patients

Gleason PP, Kapoor WN, Stone RA, et al. Medical outcomes and antimicrobial costs with the use of the American Thoracic Society guidelines for outpatients with community-acquired pneumonia. JAMA. 1997; 278:32-9.

The American Thoracic Society guidelines for community-acquired pneumonia have been widely disseminated [7], but the medical outcomes and costs associated with adherence to these guidelines have not been assessed. These guidelines have suggested that physicians should be relatively conservative in prescribing antibiotics for community-acquired pneumonia. In higher-risk patients such as the elderly, however, a broader spectrum of antibiotic choice is generally recommended.

Gleason and colleagues studied the care of 864 immunocompetent outpatients with a diagnosis of community-acquired pneumonia, 318 of whom were at least 60 years of age. Antimicrobial therapy was classified as being consistent or inconsistent with the American Thoracic Society guidelines. Outcomes were mortality, hospitalization, complications, symptom resolution, and antimicrobial costs.

Only about 17% of patients were treated according to the guidelines. Among younger patients, outcomes were better with guideline-recommended conservative treatment, such as the use of erythromycin. Antibiotic costs were also lower ($5.43 compared with $18.51). For older adults, the guidelines recommend the use of a second-generation cephalosporin or β-lactam and β-lactamase inhibitor with or without a macrolide. However, the only effect found when the guidelines were followed in these adults was a 10-fold higher cost for the more complex antibiotic regimens compared with the simpler regimen recommended for younger patients ($73.50 versus $7.50); there was also a trend toward higher mortality in older patients. However, as the authors pointed out, this subgroup of elderly patients was small.

The lessons for treatment of pneumonia derived from research in 1997 are to treat rapidly and, when working on local guideline committees, to avoid placing well-intentioned barriers (such as insistence on chest radiography) in the way of a patient's need for rapid treatment.

Dying Patients Often Suffered

Lynn J, Teno JM, Phillips RS, et al. Perceptions by family members of the dying experience of older and seriously ill patients. SUPPORT Investigators. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. Ann Intern Med. 1997; 126:97-106.

Although every physician must deal with death, little research has examined just how people die-their experience, not their physiology. One of the few observations was made by William Osler, who, in reporting on 486 consecutive deaths, said that 90 patients suffered bodily pain or distress and 11 showed mental apprehension but that most gave no sign one way or another. For most patients, death appeared to be a sleep and a forgetting [8].

To a large degree, this somewhat detached view has carried over to the present. But patients do experience the process of dying, and they often do so in the presence of family members and other loved ones. Lynn and colleagues, recognizing the unique viewpoints of the survivors, asked them about their observations of the dying process.

In a five-hospital study, the researchers identified more than 400 patients older than 80 years of age who died and for whom a surrogate decision maker could complete an interview about the experience. These decision makers, 89% of whom were close family members, reported that 55% of the dying patients were conscious and could communicate with loved ones during the last 3 days of life, most until the hour of death.

A disturbing finding was that about 70% of patients, sometimes through surrogates, sought comfort care measures but that their wishes were not honored in 20% of cases. Inadequately treated severe pain, severe fatigue, and dysphoria were common, and most patients showed signs of difficulty in tolerating these physical or emotional symptoms. Families felt that care providers frequently had not adequately addressed these problems.

This study demonstrates that serious problems remain with the approach of physicians to end-of-life decision making and care. The opportunities for improving our approach to end-of-life decision making is one of the great challenges to internal medicine as the new millennium approaches.