Body fat abnormality in HIV-infected children and adolescents living in Europe: prevalence and risk factors

Abstract

Objectives: To estimate the prevalence of and identify risk factors for lipodystrophy syndrome (LS) and body fat abnormality in a population of HIV-infected children and adolescents.

Design: Cross-sectional observational study.

Methods: HIV-infected subjects aged 2-18 years were recruited from 15 HIV centers in Belgium, Italy, and Poland between January 2007 and December 2008. Standardized assessments by the patient's long-term clinician were performed to establish the presence of abnormality. Risk factors were explored in logistic regression models for fat abnormality outcomes and LS (abnormality plus dyslipidemia).

Results: Among 426 subjects (70% white), median age was 12.2 years (interquartile range: 9.0-15.0 years) and median duration of antiretroviral therapy was 5.2 years (interquartile range: 2.2-8.8 years). Prevalence was 57% (n = 235) for LS and 42% (n = 176) for fat abnormality; 90 subjects with abnormality were affected in ≥3 locations. Lipoatrophy occurred in 28% (n = 117) of subjects and lipohypertrophy in 27% (n = 115), most commonly in the face and trunk, respectively. In multivariable analysis, white ethnicity, body mass index, ritonavir/lopinavir, and nonnucleoside reverse transcriptase inhibitors were each associated with an increased risk of LS (P < 0.05). White ethnicity, history of Centers for Disease Control and Prevention-defined disease, and stavudine were associated with risk of lipoatrophy (P < 0.05). Increased risk of lipohypertrophy was associated with body mass index and prior HIV disease.

Conclusions: Fat abnormality was prevalent in close to half of children and adolescents, who had accumulated long treatment durations. Risk of fat abnormality was associated with specific drugs, including stavudine and ritonavir, and other variables. Our results underline the importance of continued surveillance of children treated with antiretroviral therapy.

Figure 1. Frequencies of body fat redistribution (n = 422)

*Includes 2 subjects known to have lipohypertrophy but lipoatrophy status was unknown. **Includes 1 subject known to have lipoatrophy but lipohypertrophy status was unknown.

Figure 2. Presence and distribution of abnormality by body site among the 176 subjects with body fat abnormality

Percentages are given as a proportion of the number of subjects with body fat redistribution. Missing data: arms; 1, leg; 1, buttocks; 5, neck; 1, breasts; 1