Influence of hormonal status and oral intake on phase angle in HIV-infected men

Phase angle alpha can be easily obtained from bioelectrical impedance analysis. In the literature, this angle is the single best predictor of survival in patients with the human immunodeficiency virus (HIV). The aim of our study was to detect nutritional and biochemical parameters that influence the phase angle.

One hundred men with HIV were studied. In all patients, anthropometric and biochemical nutritional evaluations were performed. Basal concentrations of testosterone and somatomedin C (insulin-like growth factor-1) were measured. Tetrapolar bioelectrical impedance analysis was used to determine body composition. All patients received instruction on keeping 24-h written food records. Albumin, prealbumin, transferrin, CD4 count, and total lymphocytes were measured.

Patients' mean age was 39.1 +/- 9.9 y, mean body weight was 66.8 +/- 11.2 kg, and mean body mass index was 22.5 +/- 3.5 kg/m2. Patients' average fat-free mass was 58.05 +/- 8.7 kg and average fat mass was 9.17 +/- 4.9 kg. Average phase angle alpha was 8.21 +/- 0.88 degrees. Patients were assigned to one of two groups according to phase angle: those with a phase angle alpha smaller than 8.2 degrees (group 1) and those with a phase angle alpha of at least 8.2 degrees (group 2). Weight, body mass index, transferrin, and somatomedin C were significantly higher in group 2 (P < 0.05 for all). There were significant positive correlations between phase angle alpha and somatomedin C (r = 0.3, P < 0.05) and protein intake (r = 0.2, P < 0.05). In the multivariate analysis (adjusted for number of antiretroviral drugs prescribed, age, disease stage, and energy and protein intakes), only somatomedin C remained in the model (F = 4.5, P < 0.05), with an increase in phase angle alpha of 0.56 degrees (95% confidence interval, 0.04 to 1.08) with each 1-nM/L increase in somatomedin C.

Somatomedin C and protein intake influenced phase angle alpha, but only somatomedin C remained as an independent factor in the multivariate analysis.