 |
|
|
| |
Limb Fat Loss Must Reach 30% Before Patients and Physicians See It
|
| |
| |
15th Conference on Retroviruses and Opportunistic Infections
February 3-6, 2008
Boston
Mark Mascolini
(Poster online at http://www.retroconference.org/2008/PDFs/941.pdf.)
An intriguing substudy of the ABCDE trial suggests that antiretroviral-treated people must lose almost a third of their limb fat before they or their physicians see enough loss to call it lipoatrophy [1]. On the other hand, absolute limb fat loss did not reliably mirror clinically diagnosed lipoatrophy in this study of previously untreated people randomized to d4T/3TC/efavirenz or abacavir/3TC/efavirenz [2]. The results could inform interpretation of finished and future lipoatrophy studies.
ABCDE trial researchers analyzed limb fat loss by DEXA scans in 54 study participants at study entry and after 48 and 96 weeks. Patients and physicians also figured whether people had lipoatrophy at those points by eyeballing subcutaneous fat loss in the arms and legs, face, or buttocks. The investigators defined clinically evident lipoatrophy as moderate or severe changes versus no or mild changes. By this rating system, 13 people ended up with clinically evident lipoatrophy and 41 did not.
Among the 13 people with clinically diagnosed lipoatrophy, DEXA scans recorded a steady 5-to-3-kg decline in absolute limb fat from baseline to week 96. But DEXA saw no limb fat change in the 41 people without visually evident lipoatrophy (P = 0.002). That result was not surprising. But when the researchers compared the sensitivity of absolute limb fat loss versus percent limb fat loss in predicting clinical lipoatrophy diagnosis, they found a much better correlation with percent loss:
· 30% limb fat loss: sensitivity (true positive) 0.846
· More than 1.5 kg absolute limb fat loss: sensitivity 0.769
Among people with clinically diagnosed lipoatrophy, absolute limb fat fell about 2 kg through 96 weeks of follow-up (P = 0.002 versus baseline), while median percentage limb fat fell about 45% from baseline in this group (P < 0.001). Most people with clinically apparent lipoatrophy lost at least 30% of limb fat regardless of pretreatment weight or limb fat.
If the investigators set the DEXA percent cutoff for clinically apparent lipoatrophy at 20% instead of 30%, more than one third of people without clinically apparent lipoatrophy would get a lipoatrophy diagnosis from DEXA scanning (false-positive rate 36.6%). With the 30% DEXA cutoff the false-positive rate still came to a hefty 26.8%. Upping the cutoff to 35% limb fat loss would improve the false-positive rate to 24.4%, but that would lower the true-positive rate from 84.6% (with a 30% cutoff) to 69.2%.
Analysis of limb fat changes in ACTG study 5142 confirmed the key ABCDE finding [3]. This trial randomized previously untreated people to efavirenz plus two nucleosides, lopinavir plus two nucleosides, or efavirenz plus lopinavir without nucleosides. People taking d4T or AZT or (for still-unexplained reasons) efavirenz lost more limb fat than those taking other nucleosides or lopinavir/ritonavir rather than efavirenz. Men versus women, older people versus younger people, and people with less fat at study entry ran higher risks of lipoatrophy at 96 weeks.
Setting a DEXA-measured 20% limb fat loss as the threshold for lipoatrophy, the ACTG team found that lipoatrophy did not correlate with self-reported lipoatrophy by study participants. The ABCDE analysis suggested this lower 20% cutoff for DEXA-defined lipoatrophy may have overestimated lipoatrophy in the ACTG enrollees. It would be interesting to see if a 30% DEXA-determined cutoff for lipoatrophy would change predictors of lipoatrophy in the ACTG trial--or if a similar cutoff analysis in the larger ACTG study would back up ABCDE results.
References
1. Podzamczer D, Ferrer E, Martinez E, et al. How much fat loss is needed for lipoatrophy to become clinically evident? 15th Conference on Retroviruses and Opportunistic Infections. February 3-6, 2008. Boston. Abstract 941.
2. Podzamczer D, Ferrer E, Sanchez P, et al. Less lipoatrophy and better lipid profile with abacavir as compared to stavudine: 96-week results of a randomized study. J Acquir Immune Defic Syndr. 2007;44:139-147.
3. Haubrich R, Riddler S, DiRienzo G, et al. Clinical associations of extremity fat loss: ACTG 5142, a prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for ART of naive, HIV-1-infected subjects. 15th Conference on Retroviruses and Opportunistic Infections. February 3-6, 2008. Boston. Abstract 935.
|
| |
|
|
|
|
|
