Evaluation of the Metabolic Complications of HAART. The evaluation of metabolic function and fat redistribution in clinical trials

Regional Body Fat Distribution in HIV-Infected Patients with Lipodystrophy

BACKGROUND

Objective criteria for the assessment of patients with lipodystrophy syndrome in human immunodeficiency virus infection (LDHIV) have not emerged.

METHODS

We compared regional body fat changes in 13 men with severe LDHIV on protease inhibitor-inclusive antiretroviral therapy with 13 control HIV-infected men using anthropometry, dual-energy X-ray absorptiometry (DEXA), and whole-body magnetic resonance imaging (MRI).

RESULTS

LDHIV patients, compared with control subjects, had thinner gluteal, suprailiac, and triceps skinfolds (p < .01) and increased waist circumference (98 +/- 5 cm vs 86 +/- 9 cm, respectively; p = .0008). DEXA studies revealed reduced lower extremity fat (12 +/- 5% vs 22 +/- 9%; p = .0006), increased head and neck fat (18 +/- 3% vs 16 +/- 1%; p = .01), and increased proportion of total body fat in the trunk (65 +/- 7% vs 53 +/- 8%; p = .0005). MRI analysis revealed reduced thigh fat (12 +/- 5% vs 22 +/- 12%; p = .01), increased dorsocervical fat depth (47 +/- 24 mm vs 19 +/- 7 mm; p = .0009), and nearly significant increase in intra-abdominal fat (218 +/- 90 cm2 vs 157 +/- 70 cm2; p = .057). Interestingly, control subjects showed a positive relationship between intra-abdominal and dorsocervical fat (r= .57, p = .04), but the LDHIV patients showed a negative relationship (r= -.55, p = .05), suggesting a novel split phenotype among LDHIV patients of either dorsocervical or intra-abdominal fat accumulation.

CONCLUSIONS

We conclude that MRI provides the best tools for definition of LDHIV syndrome and reveals variable phenotypes among LDHIV patients.

Psychometric properties of a lipodystrophy scale

Prior HIV lipodystrophy (LD) scales have been unidimensional, often combining changes of fat loss with fat accumulation with retinoid changes. Our objective was to determine the psychometric properties of a LD scale used in an outpatient LD subspecialty clinic at a university hospital HIV clinic. A patient and clinician self-administered LD scale was used to evaluate 107 patients referred to the Owen LD Clinic from March of 1999 to August of 2002. Sixteen measures of the LD syndrome were assessed on a severity scale ranging from 0 (none) to 3 (severe). SPSS was used to perform a reliability and factor analysis. A total of 150 paired patient and clinician assessments were performed. Thirty-one patients had repeat assessments. Reliability analysis of the patient and clinician self-administered LD scale revealed an alpha of 0.76 and 0.73, respectively. Factor analysis of patient self-assessment of LD identified 3 factors, which clustered in the following order: fat loss, fat accumulation, and retinoid changes. Factor analysis of clinician self-assessment of LD identified 4 factors. The first factor clustered around items of fat loss. A second cluster consisted of retinoid changes. The final 2 factors clustered around items consistent with fat accumulation. The patient and clinician self-administered Owen Clinic LD scale was found to be highly reliable and useful in the assessment of LD. Our analysis confirms clinical intuition that the evaluation of LD is a multidimensional construct and that 3 subscales may be used to evaluate dimensions of fat loss, fat accumulation, and retinoid changes.