Multiple frequency bioimpedance is an adequate tool to assess total and regional fat mass in HIV-positive patients but not to diagnose HIV-associated lipoatrophy: a pilot study

Introduction

HIV-associated lipodystrophy syndrome causes systemic metabolic alterations and psychological distress that worsen the quality of life of these patients. An early detection should be considered to efficiently treat it. Objective criteria or reference indices are needed for an early diagnosis. Bioelectrical Impedance Analysis (BIA) is an operator-independent, repeatable and non-invasive method of body composition evaluation that is less expensive than dual-energy X-ray absorptiometry (DXA) and/or CT scans. The aims of this pilot study were to validate the data obtained by BIA to measure fat mass in HIV-positive patients with/without lipoatrophy and to determine if BIA correctly diagnoses lipoatrophy in HIV-positive patients.

Methods

Thirty-nine participants were included in this preliminary study. Fourteen were HIV-negative (eight men) whereas 25 were HIV-positive patients (17 men). Eleven of the HIV-positive patients were classified as lipoatrophic according to subjective evaluation by the physicians. Total and regional body composition was measured in basal conditions by DXA and by BIA. To obtain abdominal CT scan fat values, transverse slices with 6-mm thickness were acquired at the L4-L5 intervertebral space.

Results

BIA measurements of total and regional body fat were significantly correlated with those obtained by DXA (p < 0.05 to <0.01) in HIV-positive patients. However, agreement between methods was poor as not very high ICC (intraclass correlation coefficient) values were observed. BIA and DXA showed higher ICC values in lipoatrophic patients. The visceral index obtained by BIA was correlated with total and visceral fat in L4 measured by CT scan (r = 0.607 and r = 0.617, respectively, p < 0.01) in HIV-positive patients. The Fat Mass Ratio (FMR) calculated by BIA did not correlate or agree with DXA values.

Conclusions

Multi-frequency BIA could be an effective method to evaluate the evolution of total and regional fat composition in HIV-positive patients with/without lipoatrophy. The correlations between BIA and DXA improved in lipoatrophic patients and in men, suggesting that its efficacy depends on fat mass, gender and probably other factors. The visceral index obtained by BIA seems to be a reliable indicator of abdominal obesity. However, BIA did not fulfil the need for easy quantitative diagnostic tools for lipoatrophy, and it did not provide sufficient diagnostic cut-off values for this syndrome.