Skip to main content
. Author manuscript; available in PMC: 2012 Jan 2.
Published in final edited form as: AIDS. 2011 Jan 2;25(1):37–47. doi: 10.1097/QAD.0b013e32833f9d02

Table 4.

Unadjusted and adjusted odds ratios (OR) for selected predictors of lipoatrophya

Model 1- NRTI-sparing
vs. other regimens
Model 2- Stavudine XR
vs. other regimens
Model 3- Zidovudine
vs. other regimens

Covariateb Unadjusted OR
(95% CI)
P-value Adjusted OR
(95% CI)
P-value Adjusted OR
(95% CI)
P-value Adjusted OR
(95% CI)
P-value
Randomized study arm
 Efavirenz 2.4 (1.2–5.4) 0.02
 Lopinavir 0.9 (0.4–1.9) 0.8
 Lopinavir-efavirenz 0.3 (0.1–0.9) 0.03 0.3 (0.1–0.9) 0.03
Selected NRTI
 Stavudine XR 4.8 (2.0–11.6) <0.001 5.2 (2.1–12.9) <0.001
 Zidovudine 1.5 (0.7–3.0) 0.3 1.4 (0.7–3.0) 0.3
 Tenofovir 0.3 (0.1–1.1) 0.06
Baseline extremity fat
(per kg increase)
1.1 (0.98–1.2) 0.13 1.1 (0.98–1.2) 0.13 1.1 (0.98–1.2) 0.13 1.1 (0.97–1.2) 0.18
mtDNA haplogroup I
(vs. all others)
3.9 (1.1–14.4) 0.04 3.0 (0.8–11.4) 0.11 3.7 (0.9–14.6) 0.06 3.5 (0.9–12.9) 0.07
a

Lipoatrophy defined as ≥20% extremity fat loss at 96 weeks.

b

Baseline age, CD4 T cells, HIV RNA, and sex were not significant predictors of lipoatrophy (data not shown) and were not included in adjusted models.

NRTI=nucleoside reverse transcriptase inhibitor; CI=confidence interval.