From the American College of Physicians
Dr McCue is Medical Director, Tuolumne General Hospital, Sonora, Calif.
PHILADELPHIA?In 16 update sessions at the 2006 annual session of the American College of Physicians, prominent physicians presented synopses of what they considered the most important recent published trials in their discipline. The following discussion is a selection of these presentations.
New Vaccine Prevents Postherpetic Neuralgia
Herpes zoster virus (HZV) infection is a sequela of infection with varicella zoster virus, which also causes chickenpox. HZV can occur at any age, but it mostly occurs in older adults, with a peak incidence after age 75. Complications occur in about 50% of the affected elderly, the most common and debilitating being postherpetic neuralgia. The frequency and severity of pain rise with increasing age and can be prolonged and disabling.
Despite the declining incidence of chickenpox in children as a result of immunization with the varicella vaccine, the incidence of HZV has not declined. Antiviral therapy with valacyclovir (Valtrex) or famcyclovir (Famvir) reduces its severity and duration but does not prevent the neuralgia, which is often refractory to treatment. Corticosteroids offer modest benefit at the risk of complications such as bone loss, especially in the elderly.
Hope that HZV infection can be prevented or its symptoms reduced came from a large study of older adults who were immunized with a varicella vaccine that contains about 20 times as much live virus as the vaccine given to children. The good news is that it works; the bad news is that it is not yet available in the United States.
N Engl J Med
Study synopsis. The hypothesis tested was that varicella vaccination would decrease incidence, severity, or both of HZV and postherpetic neuralgia in older adults (. 2005;352: 2271-2284). Almost 40,000 adults aged ≥60 years who had a history of varicella or lived in the United States for >30 years were given a live attenuated HZV vaccine in a randomized, double-blind, placebo-controlled trial. More than 40% of participants were women, and >95% completed the study (median, 3.12 years). Vaccine recipients had significantly fewer confirmed cases of herpes zoster (n = 315) than placebo recipients (n = 642) and of postherpetic neuralgia (27 vs 80, respectively).
Commentary. The vaccine reduced the HZV incidence by 51% and the incidence of postherpetic neuralgia by 66%, both significant reductions. Injection-site reactions were generally mild. When it becomes available, this vaccine should be administered universally to our older patients. It has the potential for preventing a common disease, as well as of improving the quality of life for the millions who are at risk for postherpetic neuralgia.
Can Controlling BP Control Headache?
N Engl J Med
I was taught that mild-to-moderate blood pressure (BP) elevation does not cause headache, and I have consistently disavowed the notion that patients could tell when their BP was not under control. A recent review of chronic daily headache (. 2006;354:158-165) does not list hypertension as a cause or BP control as a preventive measure, nor do observational studies indicate that elevated BP can cause headache. Although this study does not answer the question with conviction, it offers considerable support to our patients' "unscientific" beliefs.
Study synopsis. This meta-analysis of 94 randomized placebo-controlled trials with 24,000 participants investigated whether antihypertensive drugs can prevent headache (. 2005; 112:2301-2306). Trials using thiazide diuretics, beta-blockers, angiotensin-converting-enzyme inhibitors, or angiotensin receptor antagonists in fixed doses reported data on headache. About three fourths of all the patients received antihypertensives and one fourth placebo; systolic and diastolic BPs were lowered by 9.4 and 5.5 mm Hg in the treatment groups. On average, one third fewer patients reported headache in the treated groups than in the placebo groups, a significant reduction for each of the 4 classes of drugs.
Commentary. BP-lowering drugs appear to also prevent a significant proportion of headaches. Because all 4 classes of drugs reduced headache, this meta-analysis suggests that the reduction in BP is responsible for the benefit. As geriatricians know well, elderly patients were systematically excluded from most trials of antihypertensive therapy, and applying the conclusion of this study to geriatric patients is a bit of a stretch. Nevertheless, >50% of patients aged >60 years have hypertension; if the meta-analysis is correct, the elderly might benefit most. So despite the uncertainty over cause and effect, this study gives us one more argument to encourage older patients to spend their money on antihypertensives. It also suggests the need to look again at the assumption that mild-to-moderate hypertension is asymptomatic.
Subclinical Hypothyroidism Should (Probably) Be Treated
Thyroid-stimulating hormone (TSH) tests are routinely ordered in the elderly for a variety of indications?dementia, depression, delirium, heart failure, or failure to thrive?and are supported more by wishful thinking than good clinical evidence. Results are usually normal, but even when elevated, the free thyroxine level is often normal.
Subclinical hypothyroidism (a free thyroxine level in the reference range and an elevated TSH level) is especially tricky. The chance that the patient will benefit must be balanced against the risk of precipitating symptoms (or worse) of ischemic heart disease and the problem of adding another medication to the ever-growing pharmacopeia our elderly patients take every day. Lurking behind the well-intentioned impulse to treat a probably asymptomatic elderly patient with subclinical hypothyroidism with thyroxine is the hope that it will decrease the risk of symptomatic cardiovascular disease (CVD).
The association of overt hypothyroidism with CVD is widely accepted, along with the assumption that CVD risk rises as the severity of hypothyroidism increases. Moreover, several mechanisms can be invoked as pathophysiologic support for the hypothesis that subclinical hypothyroidism might increase CVD risk. So far, results of clinical studies have been negative or, at best, mixed. These 2 studies provide support for clinicians who favor treatment, as well as for those who prefer to watch and wait.
Arch Intern Med
Study 1. Of 2730 patients aged 70 to 79 years, 12.4% had subclinical hypothyroidism (. 2005;165: 2460-2466). In those with a TSH level ≥7.0 mIU/L, congestive heart failure (CHF) events occurred significantly more often than in those with TSH levels of 4.5 to 6.9 mIU/L. CHF events occurred more than twice as often. Of the 2555 patients with no CHF at baseline, the hazard ratio for CHF events was 2.3 in those with a TSH of ≥7.0 mIU/L, but the subclinical hypothyroidism was not associated with increased risk for coronary heart disease (CHD), stroke, peripheral artery disease, or CVD-related or total mortality rates.
Arch Intern Med
Study 2. Serum thyrotropin and free thyroxine concentrations were measured in 2108 archived serum samples from a 1981 Australian health survey (. 2005;165:2467-2472). Investigators examined the 20-year risk of cardiovascular (CV) mortality and CHD in patients who had no CHD at baseline. Although relatively young (mean age, 58 years), the subclinical hypothyroidism group was significantly older than the euthyroid control group (mean age, 49 years). CHD prevalence in those with subclinical hypothyroidism was significantly higher than in euthyroid persons. In addition, those with subclinical hypothyroidism had 21 CV deaths (9.5 expected) and 33 CHD events (14.7 expected). The increased risk of CHD events remained significant after adjusting for recognized CV risk factors.
Commentary. Subclinical hypothyroidism may be an independent risk factor for CHD and CHF, especially if the TSH level is above 7 mIU/L, but apparently is not a risk factor for overall mortality. Despite encouraging data on the effect of thyroid replacement on left ventricular dysfunction and lipid profile, these studies do not answer the question about the benefit of treating subclinical hypothyroidism. Taken together, they suggest that the incidence of CV disorders may be increased in patients with subclinical hypothyroidism, particularly those with higher TSH levels. Treating patients with high-normal TSH levels would probably not be beneficial in preventing CVD. Nevertheless, all else being equal, a TSH level below 5 mIU/L is best for elderly patients, and if uncertainty exists, the goal is 2.5 mIU/L.
ChEIs: How Great the Benefits?
Beyond symptomatic treatment and attention to preventing functional decline, physicians have had little to offer patients with dementia. Diagnostic searches for potentially treatable conditions have a low (and falling) yield of only about 0.6% (. 2005;331:321-327). The hope that cholinesterase inhibitors (ChEIs) would improve, halt, or at least slow dementia was countered with a concern that they would be used indiscriminately. Indeed, they are now being prescribed at all stages of dementia and for all types of dementia in an effort to improve cognition, prevent cognitive and functional decline, and address behavioral complications. As the bloom wears off, appropriate questions are being asked as to whether pharmaceutical companies have overrated and overhyped efficacy and cost-effectiveness of these drugs. ChEIs certainly contribute to polypharmacy, are expensive, have adverse effects that may go unrecognized, and once started tend never to be discontinued.
Arch Intern Med
Study synopsis. A meta-analysis examined all published, double-blind, placebo-controlled, randomized trials of treatment with donepezil HCl (Aricept), rivastigmine tartrate (Exelon), or galantamine HBr (Reminyl) in patients with Alzheimer's disease (. 2003;163:2219-2229). The authors concluded that benefits of ChEIs were minimal and have been overstated and the adverse effects understated. There is no reason to believe that memantine HCl (Namenda), a second-line treatment for dementia, is any more effective or less toxic.
Commentary. In the treatment or prevention of the progression of dementia, ChEIs have marginal effectiveness that may not last beyond 1 year and may have no benefit or may even be harmful for individual patients. Clinicians cannot rely on published trials to provide clear guidance on how to identify which patients may respond to treatment with ChEIs, whether the cost?benefit assessment for that particular patient is favorable, and when treatment should be discontinued. It is important to recognize that deterioration of a disease process after stopping a medication does not prove that it was benefiting the patient, nor does it mean that restarting the medication would be beneficial. One is left with the unsettling conclusion that the scientific basis for recommending ChEIs for treating Alzheimer's disease is weak.
Anemia in the Elderly Correlates with Poor Survival
About 10% of the elderly have below-normal hemoglobin concentrations, but whether this has an independent negative impact on survival is unclear.
Arch Intern Med
Study synopsis. This analysis correlated baseline hemoglobin concentrations with mortality in 5888 community-dwelling men and women aged ≥65 years (. 2005;165:2214-2220). At 11 years of follow-up, 20.5% of participants were in the lowest hemoglobin quintile, and 8.5% were anemic (<13 g/dL for men; <12 g/dL for women). Mortality was increased for patients in the lower quintile of hemoglobin concentrations. The fully adjusted hazard ratio for mortality was 1.38 in those who were anemic. Adjusting for renal function, inflammation, or frailty did not reduce the risks.
Arch Intern Med
Commentary. Anemia appears to have an independent deleterious influence on morbidity, mortality, and quality of life (. 2005;165:2187-2189). But what is a normal hemoglobin level for elderly patients? The universally accepted criteria were established 37 years ago, and hemoglobin norms have changed. Still, even by those conservative criteria, as many as 20% of people aged >85 years are anemic (. 2004;104: 2263-2268), and this is despite the lack of evidence that the capacity for erythropoiesis declines with age. A third of anemia cases in the elderly remain unexplained, even when nutrition, chronic inflammation, and renal function are taken into account.
Anemia itself can worsen the underlying illness causing it, as well as have its own independent adverse effects. But the big unanswered question for clinicians is whether anemia is a mediator or a marker in an individual patient?that is, can correction of anemia be beneficial?
The answer will not be known for years, but trials of recombinant erythropoietin therapy in elderly anemic patients with heart failure, reduced renal function, or impaired mobility are ongoing. Will correcting anemia in the elderly with recombinant erythropoietin slow disease progression, reduce morbidity, improve quality of life, and prolong survival? And of great interest to your friends at Medicare who are paying for erythropoietin therapy in millions of elderly patients with renal failure, will there be a favorable cost?benefit ratio?
For the physician who sees minimally decreased hemoglobin concentrations frequently, these studies are cautionary. Although anemia workups are usually negative, and the efforts to raise the hemoglobin with dietary supplements may fail, the mere presence of a low hemoglobin concentration appears to be a predictor of mortality and morbidity. Until the evidence is there, it is prudent to follow these patients more carefully.