Body composition of HIV-positive candidates for and recipients of a kidney transplant: comparative analysis between DEXA and anthropometric indices

Abstract

Objective: To determine body composition (BC) and the correlation, if any, between indices measured by anthropometry and dual-energy X-ray absorptiometry (DEXA). Design: Cross-sectional descriptive. Setting: National sample of HIV-positive patients on the ‘positive-to-positive’ kidney transplant programme, South Africa. Methodology: 34 participants categorised as (i) HIV-positive transplant recipients from an HIV-positive donor (n = 16); and (ii) HIV-positive transplant candidates on the waiting list to receive a kidney from an HIV-positive donor (n = 18). Pearson's coefficient was used to correlate anthropometry with DEXA. Outcome measures: Body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR) were compared with DEXA-derived percentage body fat (%BF), truncal fat (TF) and visceral adipose tissue (VAT). Mid arm muscle circumference (MAMC) was correlated with DEXA lean indices namely lean mass (LM), lean mass index (LMI) and appendicular lean mass index (ALMI). Results: Pearson's correlation coefficient between BMI and %BF was strong (r = 0.773, p < 0.001). The correlation between WC with TF (r = 0.799, p < 0.001) and VAT (r = 0.885, p < 0.001) was highly significant, as was the correlation for WHtR with TF and VAT (r = 0.778, p = 0.013 and r = 0.830, p < 0.001). MAMC best correlated with ALMI (r = 0.511, p = 0.011). Conclusion: BMI, WC, WHtR and MAMC are suitable indicators of overall and regional adiposity as well as musculature, based on correlations with DEXA derived %BF, TF, VAT and ALMI respectively. The findings support the use of these anthropometric indices for measurement of BC in this patient group as a cost-effective alternative to DEXA.