ACP J Club. 1997 Sep-Oct;127:A14. doi:10.7326/ACPJC-1997-127-2-A14
Update in General Internal Medicine is a regular feature of the American College of Physicians Annual Meeting and provides an opportunity to review clinically important advances. As in the past (1), we generated themes based on information from 13 general internists at the University of Washington and our own personal reading, deliberately choosing common and clinically relevant problems. We circulated our list for review among ACP Journal Club editors and solicited their suggestions.
The most important articles were grouped into 6 themes: HIV disease, cardiovascular disease, treatment of benign prostatic hyperplasia, efficacy of selective serotonin reuptake inhibitors (SSRIs), treatment of postmenopausal osteoporosis, and preventive medicine.
Two highly publicized papers (2, 3) showed that zidovudine alone was inferior to combination therapy with di-danosine or zalcitabine in persons with CD4+ cell counts of 200 to 500/mm3 (2) or under 350/mm3 (3). Combination therapy improved survival compared with zidovudine alone but was difficult to achieve, as over 50% of patients discontinued therapy prematurely in one study (3). A new class of drugs, protease inhibitors, has emerged. Saquinavir, in combination with other drugs, produced a greater decrease in viral load and improved CD4+ cell counts compared with combination nucleoside therapy alone (4). Protease inhibitor combinations are likely to lead to better patient outcomes.
This past year, HIV-RNA viral load was reported to be a better predictor of outcome than CD4+ cell counts (5). Treatment guidelines now include measurement of RNA viral load levels (6). Finally, Mycobacterium avium complex prophylaxis with clarithromycin in persons with CD4+ cell counts < 100/mm3 improved survival (7).
Thus, more efficacious treatments of HIV and AIDS have emerged in the past year and they likely prolong life, but treatment remains complex, difficult, and expensive.
Important papers continued to appear comparing treatments that open coronary vessels. Based on an analysis of a large community registry of > 12 000 patients with myocardial infarction (MI) (8), primary percutaneous transluminal coronary angioplasty (PTCA) did not appear to be better than thrombolysis in immediate or 3-year outcomes. Costs were higher with PTCA. In a randomized controlled trial (RCT) comparing PTCA with coronary artery bypass graft (CABG) for multivessel coronary artery disease (9), outcomes were generally equivalent, except in patients with diabetes, where CABG was superior. Consistent with other trials, fewer subsequent procedures occurred after CABG.
Evidence supporting the value of cholesterol lowering continued to appear. A large multicenter trial in patients with previous MI and average cholesterol levels showed that cholesterol lowering with pravastatin lowered recurrence rates at 5 years (10). Survival was not significantly improved.
For congestive heart failure (CHF), angiotensin-converting enzyme (ACE) inhibitors remain the mainstay of many treatment regimens. However, 2 interesting papers explored 1 new and 1 traditional treatment. Carvedilol, a drug with α- and β-adrenergic blocking properties, improved short-term mortality and reduced hospitalizations when carefully titrated in patients with ejection fraction < 35% who were receiving ACE inhibitors and diuretics (11). Digoxin was studied in an RCT of nearly 7000 patients with CHF in sinus rhythm on ACE inhibitors and diuretics (12). Fewer patients on digoxin were hospitalized, but 6-year mortality (about 35%) was similar in placebo and digoxin groups.
For hypertension, traditional first-step therapy with diuretics was analyzed as part of a follow-up of the well-known Systolic Hypertension in the Elderly Program (13). Diuretics for major cardiovascular disease events were associated with a relative risk of 0.66 for persons both with and without diabetes. Absolute risk reduction at 5 years was 101 events/1000 patients with diabetes and 51 events/1000 for patients without diabetes.
Another study raised concern about adverse outcomes from certain calcium channel blockers. In a multicenter trial of 883 patients who had hypertension with carotid stenosis (14), patients receiving isradipine were more likely to develop angina or any vascular event at 3 years compared with patients receiving hydrochlorothiazide.
The evidence for anticoagulants in patients with atrial fibrillation seems to improve each year. This year, we included a recent RCT (15) showing that adjusted-dose warfarin (international normalized ratio [INR], 2.0 to 3.0) was associated with fewer strokes or systemic emboli (1.9%/y) compared with low-intensity warfarin (INR, 1.2 to 1.5) plus aspirin (7.9%/y) without any increased risk for major bleeding. A case-control study (16) of persons with nonrheumatic atrial fibrillation also concluded that the lowest effective intensity of warfarin appears to be an INR of 2.0.
This year's update included a paper describing alternative medical treatments of obstructive symptoms caused by benign prostatic hyperplasia (17). Terazosin, an α-blocker, was compared with finasteride and placebo in men with mean estimated prostate volumes approximately 2 times the normal level. The terazosin group had prompt improvement of symptoms that remained at 1 year. Combination therapy with terazosin and finasteride increased side effects but did not improve symptoms further.
Several new studies document the relative efficacy of an extremely popular class of new drugs, SSRIs, compared with traditional tricyclic antidepressants. An RCT in a primary care setting (18) compared fluoxetine, imipramine, and desipramine as initial treatments for adults with depressive disorders. At 6 months, approximately 50% of all patients were in clinical remission, but more patients with fluoxetine continued their original therapy because few adverse events had occurred. In another trial of patients with dysthymia (19), a chronic mood disorder characterized by milder symptoms than those associated with major depression and increased use of health care services and sedatives, sertraline was compared with imipramine or placebo. At 12 weeks, imipramine and sertraline were somewhat more effective than placebo, but imipramine caused more side effects.
For women with postmenopausal osteoporosis, studies showed reduced fracture rates and improved bone density. Alendronate reduced the risk for vertebral fractures by about 50% at 3 years in 2 separate studies (20, 21) of women with decreased bone density (20) or decreased bone density and previous vertebral fractures (21). Cyclical use of slow-release sodium fluoride was studied in women with osteoporosis and ≥ 1 nontraumatic fracture (22). Compared with those receiving placebo, women with mild-to-moderate bone loss who received sodium fluoride had fewer vertebral fractures and increased vertebral and femoral neck bone mass. Women with severe bone loss did not appear to experience similar benefits.
Several important papers added to our knowledge of preventive medicine. A large, well-done analysis has shown the futility of using vaginal Papanicolaou smears after hysterectomy to screen for benign disease (23). An RCT of screening for chlamydia in young women at high risk for pelvic inflammatory disease showed a reduced relative risk of 0.44 (24). New studies (25, 26) addressed the relation of postmenopausal hormone replacement therapy (HRT) to venous thrombosis and showed that HRT is associated with slightly increased risk but that the absolute risk for thrombosis is still quite small.
We concluded our preventive medicine section citing studies examining antioxidants. A large, 12-year RCT of physicians (27) revealed that β-carotene had no effect on cancer or heart disease. Another RCT of β-carotene (28) and vitamin A in persons at high risk for lung cancer was stopped at 4 years when the group taking supplements had a higher rate of lung cancer and death. Finally, a study of vitamin E (29) in patients with confirmed coronary disease showed less nonfatal MI but no change in risk for cardiovascular death.
Comments and observations
Overall, we continue to be impressed by the ongoing changes in the evidence base of general internal medicine. The availability of multiple, well-done clinical trials provides conscientious professionals with the best evidence for making therapeutic decisions.
Several other observations emerge based on compiling updates in general internal medicine for 3 consecutive years. In some areas, our colleagues reported clinically important advances before convincing evidence was published. An example is the emerging trend in the treatment of HIV disease in which highly publicized presentations at scientific meetings affect treatment decisions before publication of results. Another is the widespread use of SSRIs for depressive disorders, a frequent appearance on our colleagues' list of important advances in past years that we could not find convincing published papers to support until the 1997 update.
What's next? We guarantee there will be more clinical trials, many of which will provide insights into prevention and treatments for chronic diseases. Important advances to monitor include discovery of breast cancer genes (30) and development of pegylated thrombopoietin (31). We also predict that changes in information technology will continue to have dramatic effects on how we practice medicine and especially how we communicate with our patients. We predict that given the growing crisis facing health care financing in all industrialized countries, one of our greatest challenges will be choosing among an increased array of proven, effective therapies while balancing what is best for individual patients with pressures to contain health care costs. Measures of relative treatment effectiveness will likely become more important in helping us make these choices.
1. Larson EB, McGee SR.Update in general internal medicine. ACP J Club. 1995 Sep-Oct;123:A12-4.