Background:  The volume and anatomical distribution of facial fat depletion in lipodystrophy have not been quantified previously using objective measurements. 

Methods:  We studied HIV-infected patients with clinically defined moderate to severe lipodystrophy without wasting (n = 20), clinically defined wasting (> 10% weight loss and recent opportunistic infection) with no previous report of lipodystrophy (n = 15), and controls without lipodystrophy or wasting (n = 15). Facial MRI scans were performed, areas of fat were manually traced bilaterally in all individual 3-mm image slices, and summed to calculate volume in anatomically defined regions of interest. The area of the temporalis and masseter muscles were quantified using a similar approach.

Results:  Patients with lipodystrophy had lower fat volume in the temporal region (9.0 ± 9.0 vs 20.5 ± 7.2 mL; p < 0.001), cheek region (25.2 ± 23.1 vs 55.5 ± 15 mL; p < 0.001), and the buccal fat pad (12.6 ± 8.4 vs 21.8 ± 9.8; p = 0.003) compared to controls. Patients with wasting had temporal, cheek, and buccal fat pad volumes (10.4 ± 6.7 mL, 34.0 ± 14.8 mL, and 13.1 ± 4.6 mL, respectively) that were lower than controls (all p < 0.001) but similar to lipodystrophy patients (all p > 0.7). Patients with lipodystrophy had similar temporalis muscle volume (57.3 ± 2.2 vs 60.3 ± 13.4 mL; p = 1.0) and masseter muscle volume (46.4 ± 10.1 vs 49.1 ± 12.4 mL; p = 1.0) compared to controls, whereas patients with wasting had significantly lower temporalis muscle (45.3 ± 11.1) and masseter muscle volume (35.6 ± 6.8) than lipodystrophy or control patients (all p < 0.02). The correlation between the fat area in a single slice measurement taken at 2.5 cm below the zygomatic arch and total superficial (temporal and cheek) fat volume was 0.94 (p < 0.001).

Conclusions:  Facial fat depletion in lipodystrophy is substantial (approximately 50% volume loss) and involves superficial and deep (buccal) fat. The distribution and volume of change is similar to that seen in wasting and is therefore non-diagnostic. Depletion of facial muscle occurs in wasting but not lipodystrophy. Multi-slice or single slice MRI may be a useful method for quantifying facial fat changes of lipodystrophy in clinical trials.

Keywords: lipodystrophy; wasting; MRI