The articles by Von Roenn and colleagues [1] and Oster and associates [2] showed an important subjective benefit of megestrol acetate in patients with AIDS-related cachexia. However, no substantial increase in lean body mass or total body muscle mass was evident in either study, factors that alone can improve physical functionality (Karnofsky performance score) or robustness toward infection.

The authors need to show risk group data because lifestyle and nutritional habits differ, for example, between intravenous drug users and homosexuals or patients with hemophilia. More information on the treatment of patients with Kaposi sarcoma is warranted because potential systemic therapy [interferon- or cytotoxic agents] may produce anorexia and weight loss as side effects. Von Roenn and colleagues should provide data on body weight at study entry, information that is necessary to properly evaluate the benefit of therapy. How can patients with a body weight of 83 kg or 90.5 kg Table 4 in Oster and colleagues' study) be considered severely wasted and eligible for experimental weight gain therapy? If the duration of weight gain after the 12-week study period is to be analyzed Table 3 in Von Roenn and colleagues' study), information on the time of the last examination and on the number of patients continuing to receive study medication is crucial.

The high prevalence of oral candidiasis (34% in Oster and colleagues' trial) and the high incidence of candidal pneumonia (19 of 270 patients in Von Roenn and colleagues' study), a rather infrequent manifestation in AIDS, causes us to question whether usual prophylaxes were used.

Von Roenn and colleagues' conclusion that their data suggest no effect on immune functions or survival is unsubstantiated, given that virologic and immunologic variables were not assessed. Strikingly, patients in the 400-mg group with the highest CD4 cell counts had the highest mortality rates during the study (8 of 75 patients died).

The study by Oster and colleagues shows an obvious divergence between decreasing lean body mass (their Table 4 and the stated increase in muscle mass indices [their Table 5. This discrepancy warrants further explanation.

Because AIDS-related cachexia is primarily caused by the loss of lean body mass (particularly muscle protein), drugs with protein anabolic effects (anabolic steroids) represent an obvious treatment alternative. Consequently, we did a 30-week pilot study with the testosterone derivative oxymetholone in 30 patients with AIDS-related cachexia at the University of Essen in Germany [3]. We observed weight gain in 91% of the patients, with a mean weight gain (±SD) of 8.2 ±6.2 kg after a mean of 20 weeks. The Karnofsky performance score improved from 65% before treatment to 74% during treatment. Virus replication and T-cell counts were not affected. The cost of oxymetholone treatment was approximately $60 per month.

The wide use of megestrol acetate for extended periods (which causes storage of body fat without functional improvements) should not be recommended for patients with AIDS-related cachexia.