Impact of highly active antiretroviral therapy (HAART) on body composition and other anthropometric measures of HIV-infected women in a primary healthcare setting in KwaZulu-Natal : a pilot study
Background and objectives: An understanding of the effect of HAART on different aspects of health, including nutritional status, of HIV-infected individuals in South Africa is needed to ensure that appropriate population-specific guidelines and policies can be developed. This study aimed to investigate the impact of HAART on nutritional status, focusing on changes in anthropometric measures, and to explore the relationship between these measures and immunological and virological response to HAART. Methods: A prospective study of 30 adult females was carried out at a clinic in Cato Manor, KwaZulu-Natal. Anthropometric measurements, including weight, mid-upper arm circumference (MUAC), waist circumference, hip circumference, body mass index (BMI) and waist-to-hip ratio (WHR), were performed at baseline and 12 and 24 weeks after commencing HAART. Laboratory values, including CD4 lymphocyte count, viral load, albumin and haemoglobin as well as bioelectrical impedance analysis data, including lean body mass (LBM), fat mass (FM) and body fat percentage (BF%), were collected at baseline and after 24 weeks on HAART. Results: Overall, there was a statistically significant increase in all anthropometric measures, except WHR and LBM. The mean weight change was 3.4±5.8kg (p=0.006). Fifty percent of the subjects had a BMI above normal at baseline and mean BMI increased from 25.6±5.7kg/m2 to 27.3±5.6kg/m2 (p=0.007). Seventy percent of subjects gained weight, 18.5% had a stable weight and 11.1% lost weight. The weight gain in most subjects was attributable to a gain in FM while in subjects who lost weight, the loss consisted mainly of LBM. Some patients with stable body weight experienced changes in the relative proportions of fat and lean mass. Six patients showed evidence of disproportionate gains and losses in body circumference measurements which may be indicative of fat redistribution. Subjects with lower CD4 lymphocyte counts experienced greater increases in weight, BMI, FM and BF%. The strongest correlation was observed with FM (rs=-0.53; p=0.00). Greater increases in weight, BMI, MUAC, waist circumference, hip circumference, FM and BF% were seen in those with lower baseline haemoglobin. Baseline viral load and albumin did not correlate significantly with changes in any anthropometric variables. Change in CD4 count was only significantly associated with baseline MUAC (rs=0.40; p=0.04). Change in viral load was significantly correlated with baseline weight, LBM, FM, BF% and MUAC with the strongest correlation being with weight (rs=0.44; p=0.01). No significant association was found between anthropometric changes and changes in CD4 count and viral load between baseline and the 24-week visit. Conclusion: Overall, subjects experienced a significant increase in most anthropometric measures. There appears to be a relationship between some anthropometric and laboratory measures but this needs clarification. The findings of this study demonstrate the value of including circumference measurements and body composition techniques as part of nutritional status assessment and demonstrate the need for studies to determine the prevalence and significance of overweight and obesity in the HIV-infected population. Research is needed to determine the best methods of bringing about the most favourable anthropometric changes to enhance the health of patients on HAART.