Use of a Test That Evaluates How the Body Handles Insulin and Glucose

The purpose of this study is to find out if testosterone gel affects how the body handles insulin and glucose.

Specific anti-HIV treatments may increase a patient's risk of certain diseases by causing metabolic problems such as reduced sensitivity to insulin. This substudy will examine how testosterone affects insulin sensitivity. If testosterone increases insulin sensitivity in the patients, then giving testosterone to HIV-infected patients may allow the continuation of anti-HIV treatments without increasing the patients' risk of disease. The test used to determine insulin sensitivity will be a modified frequently sampled intravenous glucose tolerance test (FSIVGTT), which can accurately measure insulin sensitivity.

Assessment of insulin sensitivity is an important objective of this substudy. Extrapolation from earlier studies suggests that insulin resistance will be associated with low testosterone levels at baseline in HIV-infected men with visceral obesity and that physiologic testosterone replacement in these individuals is likely to increase insulin sensitivity. However, direct assessment of insulin sensitivity under carefully controlled conditions is crucial to understanding the full metabolic impact of physiologic testosterone replacement in the context of A5079. The FSIVGTT, using the Bergman minimal model, has become a widely used method for the assessment of insulin sensitivity in epidemiological and intervention studies. Based on these considerations, the modified FSIVGTT was chosen as the optimal modality to assess insulin sensitivity in A5112s.

This study is a substudy of A5079. Patients enrolled in A5079 are eligible for this substudy but receive no study treatment other than that which they receive on A5079. Insulin sensitivity is measured by FSIVGTT performed at entry and at Weeks 24 and 48. In the modified FSIVGTT, 15 serial blood specimens for glucose are drawn over a 245-minute interval. The FSIVGTT assesses whether patients given testosterone supplements have more sensitivity to insulin than patients given a placebo.