Shlay, Judith C MD, MSPH*; Sharma, Shweta MS†; Peng, Grace MS†; Gibert, Cynthia L MD‡; Grunfeld, Carl MD, PhD§ ; for the Terry Beirn Community Programs for Clinical Research on AIDS and the International Network for Strategic Initiatives in Global HIV Trials
Multiple lines of evidence have implicated certain nucleoside reverse transcriptase inhibitors (NRTIs) in the development of lipoatrophy, with associations identified with cumulative duration of NRTI exposure and the current use of thymidine analogues, particularly stavudine (d4T).1-14 Insufficient information is available comparing long-term regional body composition changes with the use of thymidine analogue-sparing regimens to those with thymidine analogue-containing regimens.10,13,15,16 This comparison is clinically important in antiretroviral-naive patients, because recent results have demonstrated a modest improvement in fat mass among patients changing from a thymidine analogue-containing regimen to a thymidine analogue-sparing regimen.13,17-19
The prevalence of HIV-related lipodystrophy has varied as a result of different methods for assessing and defining lipodsytrophy.20 Recent studies have demonstrated regional differences in the morphologic changes associated with HIV infection, with the predominant changes being peripheral and central lipoatrophy affecting subcutaneous fat.11,12,21,22 Many of these studies use dual-energy x-ray absorptiometry (DEXA) scanning, which is not able to assess specific types of regional fat changes, or the studies have used costly radiologic tests (eg, computed tomography [CT], magnetic resonance imaging [MRI]). Furthermore, none of the studies have assessed how initiation of specific antiretroviral therapy (ART) regimens affects long-term regional subcutaneous tissue. Thus, the purpose of this study was to assess the long-term changes in subcutaneous tissue among ART-naive persons initiating 1 of 3 NRTI-containing regimens: d4T plus lamivudine (d4T+3TC) versus zidovudine plus lamivudine (ZDV+3TC) versus abacavir plus lamivudine (ABC+3TC).
METHODS
The metabolic study was a substudy of a long-term randomized clinical trial23 conducted by the Community Programs for Clinical Research on AIDS (CPCRA).24 Participants who were naive to antiretroviral drugs and enrolled in the Flexible Initial Retrovirus Suppressive Therapies (FIRST; CPCRA 058) study were offered coenrollment in the metabolic study (CPCRA 061). The FIRST study compared 3 initial treatment strategies for clinical and immunologic outcomes, with the goal of determining whether it was better to initiate ART with a 2-class strategy (protease inhibitor [PI] + NRTI or nonnucleoside reverse transcriptase inhibitor [NNRTI] + NRTI) than with a 3-class strategy (PI + NNRTI + NRTI) and of determining which of the 2-class strategies was better for initial therapy. The FIRST study demonstrated that the 3-class strategy was not superior to the 2-class strategy and was associated with more treatment-limiting toxicity. The study also demonstrated superior virologic response with the NNRTI strategy compared with the PI strategy but no differences in immunologic or clinical outcomes.24
The study design, data collection, and results of the metabolic study were published previously.25 This long-term clinical trial assessed changes in metabolic parameters and body composition among ART-naive participants randomized to 3 highly active ART strategies. The study demonstrated similar changes in total and regional fat, with no differences by ART strategy. Differential effects on lipid metabolism by strategy were seen, with overall increases in insulin and insulin resistance for all 3 strategies.
Eligibility criteria for the metabolic study included documented HIV infection and being at least 13 years of age. Participants were excluded if they were pregnant or breast-feeding, had any prior use of PIs or NNRTIs, or had a cumulative total of more than 4 weeks of NRTI use or more than 1 week of 3TC use. The metabolic study opened for enrollment in August 1999 and closed for follow-up in September 2005. A consent form approved by the institutional review board of each site was signed by each participant.
At enrollment, a baseline history and targeted physical examination were completed, including demographic characteristics, current medications, and prior AIDS-related diagnoses.26 A CD4 lymphocyte count was determined at local laboratories, and plasma HIV RNA measurement (Roche Amplicor 1.0, Roche Diagnostics, Basil, Switzerland) was performed at a central CPCRA laboratory.
Body composition measurements were performed by staff members who received extensive centralized training and were certified at study initiation and annually.27 Initially, anthropometric measurements were obtained at baseline, at every 4 months during the first year, and annually afterward. The protocol was subsequently modified to have anthropometric measurements performed at each follow-up visit instead of only annually to maintain competency in performing the measurements. Anthropometric measurements were obtained after an 8-hour fast and included measurements of 5 skinfold thicknesses (triceps, subscapular, abdomen, suprascapular, and thigh) using a Lange caliper,28 all measured twice on the right side and averaged for analyses. Four of the measurements (triceps, subscapular, abdomen, and thigh) were performed as previously described.28 The suprascapular measurement was used to assess changes at the back of neck.29
The study sample for this report included individuals enrolled in the metabolic study who were prescribed 1 of 3 NRTI combinations (d4T+3TC, ZDV+3TC, or ABC+3TC) at baseline, had no missing baseline skinfold measurements, and had complete data for the 5 skinfold measurements on 2 or more follow-up visits through month 36. There were 308 participants who met these criteria and comprised a nonrandomized sample.
Statistical Analysis
Summary statistics of the demographic and HIV disease measures were calculated and compared. Anthropometric measures at baseline were summarized and compared adjusting for age, gender, race, smoker status, prior injection drug use (IDU), prior AIDS, and HIV RNA level. Intent-to-treat analyses were performed to understand the long-term effects of using each of the 3 NRTI regimens. For the selected body composition measurements, the average change from baseline at each follow-up visit (months 4 through 36) was calculated by baseline prescribed NRTI regimen. Using repeated measures analyses30 with random intercepts and random slopes, unadjusted linear regressions were fitted to the follow-up data. The slopes of the regression lines were used as estimates for the rates of change of the body composition measurements. Comparisons of the rates for the 3 regimens were performed after adjusting for the following baseline variables: age, gender, race, smoker status, prior IDU, prior AIDS, HIV RNA level, NNRTI status (yes/no), and PI status (yes/no). Nominal P values are presented for pairwise comparisons of the rates for the NRTI regimens and should be <0.017 to be significant at the 0.05 level for a Bonferroni adjustment.
Regressions were performed on data from all follow-up visits (months 4 through 36), from the early follow-up period (months 4 through 12) and the late follow-up period (months 16 through 36). For a comparison of the rates between the early and late periods of follow-up, only participants with at least 2 visits in both the early and late periods were included in the analysis. Differences between the rates for the early and late periods for the NRTI regimens were assessed by ANOVA models. The rates of change in the early and late periods were adjusted for the same variables as listed previously.
Mean percent change from baseline over all follow-up (months 4 through 36) was also calculated by NRTI regimen for the 5 skinfolds. This measure provided a uniform platform for a comparison of the rates of change of the 5 skinfolds within a specific NRTI group. Similar regressions as described previously were fitted using all follow-up data. ANOVA that accounted for within-person variability was used to compare the rates of percent change in the 5 skinfolds. The Fisher least square difference (Student t) test was used for multiple comparison testing in the rates for the skinfolds.
An on-treatment analysis (not presented) yielded similar results as the intent-to-treat analysis. All statistical analyses were performed using SAS (version 8.2, SAS Institute, Cary, NC).
RESULTS
Baseline Characteristics
Between August 1999 and January 2002, 422 (30%) of 1397 participants in the FIRST study were coenrolled into the metabolic study. Among the 308 participants who met the criteria for this report of having complete baseline skinfold data and at least 2 follow-up measures for each of the 5 skinfolds, 63 were prescribed d4T+3TC, 192 were prescribed ZDV+3TC, and 53 were prescribed ABC+3TC at study entry.
Overall, the mean age of the participants was 38 years; 21% were female; and 72% were nonwhite, with 59% black and 11% Latino (Table 1). At baseline, the mean CD4 lymphocyte count was 208 cells/mm3, the mean log10 HIV RNA level was 5.0 copies/mL, 35% of participants reported a prior AIDS-defining event, 7% were diagnosed with hepatitis B infection by hepatitis B surface antigen testing, and 22% were diagnosed with hepatitis C infection by antibody testing.
The participants in the 3 NRTI regimens were balanced with respect to many of the baseline characteristics (see Table 1), with the exception that fewer participants with a history of IDU were prescribed the ABC+3TC regimen (P = 0.05). An imbalance between the 3 regimens was also present concerning the prevalence of prior AIDS and baseline anthropometric measures. At baseline, the d4T+3TC cohort had the highest prevalence of prior AIDS and the lowest averages for weight, body mass index (BMI), and the 5 skinfolds. Each of the 7 body composition measurements were negatively correlated with prior AIDS (results not shown). In the regression models used to adjust for confounding variables, the term prior AIDS was included to control for the potential effects of this imbalance when comparing the results of the 3 NRTI regimens.
Table 2 summarizes the initial antiretroviral regimens prescribed to study participants within each NRTI regimen. Nelfinavir was the most commonly prescribed PI, particularly for the d4T+3TC and ZDV+3TC regimens, whereas a higher percentage of participants in the ABC+3TC regimen were prescribed indinavir compared with the d4T+3TC or ZDV+3TC regimens. Nevirapine was more commonly prescribed with ABC+3TC, whereas no significant differences were seen among the regimens in efavirenz prescription.
During the 3 years of follow-up, the median length of exposure to d4T+3TC was 2.2 years (interquartile range [IQR]: 1.0 to 2.9 years); to ZDV+3TC, it was 2.8 years (IQR: 1.0 to 3.0 years); and to ABC+3TC, it was 2.3 years (IQR: 0.8 to 3.0 years) (Table 3). The median time to the first change in an NRTI regimen was 2.1 years for d4T+3TC, 2.2 years for ZDV+3TC, and 1.6 years for ABC+3TC. No statistically significant differences were seen between the 3 regimens in the percentage of participants who changed their NRTI regimen during follow-up. At the time of the first change of an NRTI, there were no differences in the percentage of participants discontinuing both NRTIs. For the d4T+3TC cohort, the median time to the first change in NRTI combination (2.1 years) and the median time of total exposure to the NRTI combination (2.2 years) were nearly equal, suggesting that among the 55% who stopped the regimen, relatively few restarted the regimen at a later time. For the ZDV+3TC regimen, the 2 estimates were 2.2 years and 2.8 years, suggesting that among the 54% who stopped these NRTIs, a sizable number must have restarted this combination. For the ABC+3TC regimen, the median time of exposure was 2.3 years, which was longer than the median time to first change in NRTI of 1.6 years, indicating that a large number of participants must have restarted this regimen.
Mean Change in Subcutaneous Tissue
In Figure 1, the mean changes from baseline of skinfold measurements at each follow-up visit for the 5 skinfold sites by NRTI regimen are presented. For the subscapular skinfold, all the mean changes were positive, with some being significantly different from 0 (see Fig. 1A). For the 3 regimens, the rates of change were positive but not significantly different from 0 and not significantly different from each other.
For the suprascapular site (see Fig. 1B), the mean changes at each follow-up visit were positive but generally not significantly different from 0. The mean changes decreased over time, resulting in negative estimates for the rates of change. Only the rate for the ZDV+3TC regimen was significantly different from 0. None of the rates differed by NRTI regimen.
The mean changes for the abdominal skinfold (see Fig. 1C) tended to increase and were positive before the second year of follow-up. Subsequently, the mean changes for the d4T+3TC and ZDV+3TC regimens decreased, whereas for the ABC+3TC regimen, the mean changes increased with longer follow-up. For the 3 NRTI regimens, the rates were significantly different from 0. The rate for the ABC+3TC regimen was positive and significantly different from the negative rates for the d4T+3TC and ZDV+3TC regimens. No difference was seen between the rates for the d4T+3TC and ZDV+3TC regimens.
The thigh skinfold results (see Fig. 1D) were similar to the abdominal skinfold results, with the mean changes for each of the 3 NRTI regimens initially being positive and increasing and then decreasing after the first year of follow-up for the d4T+3TC and ZDV+3TC regimens. For the d4T+3TC and ZDV+3TC regimens, the rates of change over the 3-year follow-up period were negative and significantly different from 0. For the pairwise comparisons, only the rate for the ZDV+3TC regimen was significantly different from that for the ABC+3TC regimen, whereas the difference between the d4T+3TC and ABC+3TC regimens was borderline significant.
For the triceps skinfold (see Fig. 1E), the rates of change were negative for the 3 regimens, with the rate for the ZDV+3TC regimen being significantly different from 0 and significantly different from the rate for the ABC+3TC regimen. The mean changes at the follow-up visits for the ZDV+3TC regimen beyond the first year were negative, with most being significantly different from 0.
Percent Change in Subcutaneous Tissue
The results of the comparison of the rates of change expressed as percent change of the baseline measurement are given in Figure 2. In each figure, the rates and their P values are listed for each skinfold site and significant differences from the pairwise comparisons of the 5 rates are noted.
For the d4T+3TC regimen (see Fig. 2A), the rates of percent change for the subscapular and suprascapular sites were positive and significantly greater than the negative rates for the abdomen and thigh skinfolds. The rate for the subscapular site was also significantly greater than that for the triceps, which was negative. There was no evidence that the rates for the subscapular and suprascapular sites differed from each other or that the abdomen, thigh, and triceps rates differed from each other.
For the ZDV+3TC regimen (see Fig. 2B), only the subscapular rate was positive, with the rates for the other skinfolds being negative and mostly significantly different from 0. All but 1 of the pairwise comparisons between the rates of change for the 5 sites were significant. The rates for the abdomen and thigh did not differ from each other, whereas the rates for the subscapular, suprascapular, and triceps skinfolds were different from each other and differed from those for the abdomen and thigh sites.
For the ABC+3TC regimen (see Fig. 2C), the 5 rates of percent change were positive. Those for the subscapular and abdomen skinfolds were significant. The rate of percent change for the abdomen was significantly different from the rates for the other 4 sites. The rate for the subscapular skinfold differed significantly from the rates for the suprascapular and triceps skinfolds.
Early versus Late Changes in Rates
The 3-year follow-up period was divided into 2 time periods, with the early time period defined as 12 months or less and the late time period defined as beyond 12 months. The rates of change for weight, BMI, and the 5 skinfolds were determined for each time period and compared with 0. Comparisons were also performed between the rates for the 2 time periods (Table 4).
The rates of change during the first year were generally positive but not significantly different from 0, except for weight and BMI of the ZDV+3TC and ABC+3TC regimens, which were positive and significantly different from 0. During the late time period, the rates for the d4T+3TC and ZDV+3TC regimens were negative (with 1 exception) and significantly different from 0 for the abdomen and thigh skinfolds (d4T+3TC and ZDV+3TC) and triceps skinfold (ZDV+3TC).
In every case, for the d4T+3TC and ZDV+3TC regimens, the rates for the late time period were less than the corresponding rates in the early time period. Most of the early versus late differences were significant. For the ABC+3TC regimen, the differences between the rates of the early and late time periods were only significant for weight, BMI, and the triceps skinfold, with the rates of change for the late period being smaller than those for the early period for these measures.
DISCUSSION
In this evaluation of long-term NRTI use, we assessed differences between 3 NRTI regimens on changes in subcutaneous tissue. Subcutaneous tissue losses were similar for the d4T+3TC and ZDV+3TC regimens, whereas use of the ABC+3TC regimen resulted in subcutaneous tissue gains. These findings were particularly apparent for the abdomen and thigh skinfolds, where significant declines were seen for the d4T+3TC and ZDV+3TC regimens, whereas for the ABC+3TC regimen, increases were seen, especially in the abdomen.
Temporal differences were demonstrated, with the rates in the early period (up to 12 months) for the d4T+3TC and ZDV+3TC regimens being positive, suggesting initial recovery after the initiation of ART followed by declines in the late period, representing the long-term treatment effect of d4T and ZDV on body composition changes. These effects were not seen with the ABC+3TC regimen, indicating differences in the effects of long-term use of the various NRTIs on subcutaneous tissue.
Some prior reports have documented lipoatrophy with the use of d4T,7,8,11-14,16 with others documenting similar changes with the use of ZDV.31,32 Some reports have demonstrated that the use of ZDV with 3TC has been associated with limb fat increases.15,33 Others have suggested that the risk of lipoatrophy is greater with the use of d4T compared with ZDV.6-8,31,34 In our study, use of ZDV(+3TC) or d4T(+3TC) was associated with similar losses in subcutaneous tissue. Variations in the assessment of body composition changes (ie, use of DEXA vs. anthropometry) potentially explain some of the inconsistencies in published reports on the effects of ZDV and d4T. Most studies, however, conclude that long-term use of either of these thymidine analogue NRTIs is associated with the development of lipoatrophy.
Previous studies have demonstrated that with the initiation of ART, there is an improvement in numerous clinical parameters (ie, weight gain, increases in body composition measures) associated with undetectable viral loads and increases in CD4 lymphocyte counts.6,15,22,25 When comparing early versus late changes, our study demonstrated a general trend in the early time period of increases in various clinical parameters (BMI and body composition), irrespective of the NRTI regimen used.
The use of thymidine analogue NRTIs has been associated with the occurrence of HIV-associated lipoatrophy, particularly with the use of d4T.5-8,10,14,16,35 In our study, use of both of the thymidine analogue-containing regimens (d4T+3TC or ZDV+3TC) and the thymidine analogue-sparing regimen (ABC+3TC) in ART-naive participants initiating therapy resulted in initial increases in subcutaneous tissue. These initial increases were followed by a progressive loss of subcutaneous tissue with increasing cumulative thymidine analogue exposure, whereas no declines were seen with use of the thymidine analogue-sparing regimen (ABC+3TC), as previously reported.6,14,16,36 These findings were most apparent in the abdomen and thigh skinfolds and were present irrespective of background PI or NNRTI use. Our findings are corroborated by recent reports of improvement in lipoatrophy after switching from a thymidine analogue regimen to a thymidine analogue-sparing regimen,17,18,31,37 suggesting that the use of d4T and ZDV should be avoided if other NRTIs are available.
In this study, we also compared the rates of percent change over time for the 5 skinfolds for participants on each regimen. Differential rates of change in regional subcutaneous tissue were identified. For the d4T+3TC regimen, these differences were apparent when comparing the subscapular and suprascapular skinfolds with the abdomen and thigh skinfolds. Similar findings were seen with ZDV+3TC. Analogous conclusions were drawn in the study of Fat Redistribution and Metabolic Change in HIV Infection (FRAM), a cross-sectional analysis of HIV-positive and HIV-negative men that used more sophisticated methods (eg, DEXA and MRI scans) for assessing body fat. Leg fat was much lower in HIV-infected men compared with the controls, whereas upper trunk fat was relatively spared.11
In our recent report of the metabolic study findings that assessed changes in metabolic parameters and body composition among ART-naive participants randomized to 3 highly active ART strategies (PI, NNRTI, PI+NNRTI), focusing on regional body composition, we found increases in nonsubcutaneous tissue areas of the midarm, midthigh and waist (visceral tissue area) with a loss in subcutaneous tissue areas in all regional areas regardless of strategy.25 The accumulation of trunk fat, particularly visceral fat, and the loss of lower peripheral fat in the thigh have been associated with the development of insulin resistance and adult-onset diabetes in non-HIV-infected populations.38-40 Few studies have evaluated the effects of upper trunk fat, but a recent study of HIV-infected subjects and controls found that visceral fat and upper trunk fat were independently associated with insulin resistance.41 Thus, the changes in body composition noted in this study may contribute to the gradual development of insulin resistance.
The strength of the results includes the large numbers of participants followed for the 3-year time period. For the presented analyses, the 192 participants who were initially prescribed ZDV+3TC had a median of 2.8 years of use. Although there were smaller numbers of participants on d4T+3TC (n = 63) and ABC+3TC (n = 53), the length of exposure time to the NRTIs was significant (2.2 years to d4T and 2.3 years to ABC), allowing an assessment of the long-term impact of these medications on body composition changes. On-treatment analyses yielded similar findings, albeit with a smaller sample size.
There are a number of limitations to this study. First, analysis of the effect of an NRTI regimen does not allow us to examine the individual impact of the NRTIs on the development of body composition changes, which would help to account for NRTI regimen changes and their impact on body composition. Second, at the time of the study's design and implementation, nelfinavir and indinavir were the primary PIs available. Newer PIs may have different long-term effects on body composition when combined with the NRTI regimens used in this study. Third, although there were slight differences in the use of other ART between the 3 NRTI regimens, we doubt that these differences affected the outcomes. For example, although there was more indinavir use in the ABC+3TC regimen, this could not explain the greater subcutaneous fat measured in the thigh, because indinavir has been associated with lipoatropy.11 To examine the potential impact of differential indinavir use, we substituted the PI status term in our regression models with a term denoting indinavir use. The results from these models were similar to our initial findings; as such, indinavir use was not adjusted for. Fourth, there is the potential for bias because of the nonrandomized treatment comparison and the noncomparability of groups with respect to certain baseline characteristics (eg, IDU). To account for this, all possible influential baseline parameters were adjusted for in the regression analyses. Finally, although more recent studies have used DEXA scanning to assess body composition changes with therapy, anthropometry performed by well-trained individuals using standardized measurement protocols and tools,29 as in our study, can be used as a reliable tool for measuring body composition in HIV-infected participants.42,43 Studies using both methods have shown correlations between DEXA and anthropometry.22,44
In conclusion, this long-term nonrandomized evaluation of the effects of 3 different NRTI regimens in the presence of similar PI and/or NNRTI therapy identified significant subcutaneous tissue differences depending on the NRTI regimen used. All 3 NRTI regimens were initially associated with clinical improvements, but with prolonged use of d4T+3TC or ZDV+3TC, similar subcutaneous tissue losses were seen, whereas with prolonged use of ABC+3TC, subcutaneous tissue gains were seen, reflecting differences in the long-term clinical effects of these medications. The selection of an ART regimen should consider the associated toxicities (lipoatrophy with prolonged use of a thymidine analogue-based regimen), which can have a negative impact on the success of the regimen selected.
ACKNOWLEDGMENT
We gratefully acknowledge the hard work and dedication of Glenn Bartsch, ScD.
REFERENCES
1. Carr A, Workman C, Smith D, et al. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy: a randomized trial. JAMA. 2002;288:207-215.
2. Lichtenstein KA, Ward DJ, Moorman AC, et al. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. AIDS. 2001;15:1389-1398.
3. Bogner JR, Vielhauer V, Beckmann R, et al. Stavudine versus zidovudine and the development of lipodystrophy. J Acquir Immune Defic Syndr. 2001;27:237-244.
4. Saint-Marc T, Partisani M, Poizot-Martin I, et al. Fat distribution evaluation by computed tomography and metabolic abnormalities in patients undergoing antiretroviral therapy: preliminary results of the LIPCO study. AIDS. 2000;14:37-49.
5. Bernasconi E, Boubaker K, Junghans C, et al. Abnormalities of body fat distribution in HIV-infected persons treated with antiretroviral drugs: the Swiss HIV Cohort Study. J Acquir Immune Defic Syndr. 2002;31:50-55.
6. Mallon PWG, Miller J, Cooper D, et al. Prospective evaluation of the effects of antiretroviral therapy on body composition in HIV-1-infected men starting therapy. AIDS. 2003;17:971-979.
7. Mallal SA, John M, Moore C, et al. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS. 2000;14:1309-1316.
8. Joly V, Flandre P, Meiffredy V, et al. Increased risk of lipoatrophy under stavudine in HIV-1-infected patients: results of a substudy from a comparative trial. AIDS. 2002;16:2447-2454.
9. Lichtenstein KA, Delaney K, Armon C, et al. Incidence of and risk factors for lipoatrophy (abnormal fat loss) in ambulatory HIV-1-infected patients. J Acquir Immune Defic Syndr. 2003;32:48-56.
10. Shlay J, Visnegarwala F, Bartsch G, et al. Body composition and metabolic changes in antiretroviral-naive patients randomized to didanosine and stavudine versus abacavir and lamivudine. J Acquir Immune Defic Syndr. 2005;38:147-155.
11. Bacchettti P, Gripshoever B, Grunfeld C, et al. Fat distribution in men with HIV infection. J Acquir Immune Defic Syndr. 2005;40:121-131.
12. Study of Fat Redistribution and Metabolic Change in HIV Infection. Fat distribution in women with HIV infection. J Acquir Immune Defic Syndr. 2006;42:562-571.
13. Gallant J, Staszewski S, Pozniak A, et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA. 2004;292:191-201.
14. Saint-Marc T, Partisani M, Poizot-Martin I, et al. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS. 1999;13:1659-1667.
15. Dube M, Parker R, Tebas P, et al. Glucose metabolism, lipid, and body fat changes in antiretroviral-naive subjects randomized to nelfinavir or efavirenz plus dual nucleosides. AIDS. 2005;19:1807-1818.
16. 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. 2006;44:139-147.
17. Martin A, Smith D, Carr A, et al. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX extension study. AIDS. 2004;18:1029-1036.
18. Moyle G, Baldwin B, Landroudi B, et al. A 48-week, randomized, open-label comparison of 3 abacavir-based substitution approaches in the management of dyslipidemia and peripheral lipoatrophy. J Acquir Immune Defic Syndr. 2003;33:22-28.
19. Tien P, Schneider M, Cole S, et al. Relation of stavudine discontinuation to anthropometric changes among HIV-infected women. J Acquir Immune Defic Syndr. 2007;44:43-48.
20. Domingo P, Labarga P, Palacios R, et al. Improvement of dyslipidemia in patients switching from stavudine to tenofovir: preliminary results. AIDS. 2004;18:1475-1478.
21. Tien P, Cole S, Williams C, et al. Incidence of lipoatrophy and lipohypertrophy in the Women's Interagency HIV Study. J Acquir Immune Defic Syndr. 2003;35:461-466.
22. Mulligan K, Parker R, Komarow L, et al. Mixed patterns of changes in central and peripheral fat following initiation of antiretroviral therapy in a randomized trial. J Acquir Immune Defic Syndr. 2006;41:590-597.
23. MacArthur R, Chen L, Mayer D, et al. The rationale and design of the CPCRA (Terry Beirn Community Programs for Clinical Research on AIDS) 058 FIRST (Flexible Initial Retrovirus Suppressive Therapies) trial. Control Clin Trials. 2001;22:139-141.
24. MacArthur R, Novak R, Peng G, et al. A comparison of three highly active antiretroviral treatment strategies consisting of non-nucleoside reverse transcriptase inhibitors, protease inhibitors, or both in the presence of nucleoside reverse transcriptase inhibitors as initial therapy (CPCRA 058 FIRST study): a long-term randomised trial. Lancet. 2006;368:2125-2135.
25. Shlay J, Bartsch G, Peng G, et al. Long-term body composition and metabolic changes in antiretroviral-naive persons randomized to protease inhibitor-, nonnucleoside reverse transcriptase inhibitor-, or protease inhibitor plus nonnucleoside reverse transcriptase inhibitor-based strategy. J Acquir Immune Defic Syndr. 2007;44:506-517.
26. Centers for Disease Control and Prevention. 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Morb Mort Wkly Rep. 1992;41(RR-17):1-19.
27. Wang J, Bartsch G, Raghavan S, et al. Reliability of body composition and skinfold measurements by observers trained in groups. International Journal of Body Composition Research. 2004;2:31-36.
28. Lohman TG, Roche A, Martorell R. Anthropometric Standard Reference Manual. Champaign, IL: Human Kinetics; 1988.
29. Wang J, Thorton J, Kolesnik S, et al. Anthropometry in body composition: an overview. Ann N Y Acad Sci. 2000;904:317-326.
30. Laird NM, Ware J. Random-effects models for longitudinal data. Biometrics. 1982;38:963-974.
31. John M, McKinnon E, James I, et al. Randomized, controlled, 48-week study of switching stavudine and/or protease inhibitors to Combivir/abacavir to prevent or reverse lipoatrophy in HIV-infected patients. J Acquir Immune Defic Syndr. 2003;33:29-33.
32. Hansen A, Lindegaard B, Obel N, et al. Pronounced lipoatrophy in HIV-infected men receiving HAART for more than 6 years compared with the background population. HIV Med. 2006;7:38-45.
33. Dube M, Komarow L, Mulligan K, et al. Long-term body fat outcomes in antiretroviral-naive participants randomized to nelfinavir or efavirenz or both plus dual nucleosides: dual X-ray absorptiometry results from A5005S, a substudy of Adult Clinical Trials Group 384. J Acquir Immune Defic Syndr. 2007;45:508-514.
34. Amin J, Moore A, Carr A, et al. Combined analysis of two-year follow-up from two open-label randomized trials comparing efficacy of three nucleoside reverse transcriptase inhibitor backbones for previously untreated HIV-1 infection: OzCombo 1 and 2. HIV Clin Trials. 2003;4:252-261.
35. Martinez E, Mocroft A, Garcia-Viejo MA, et al. Risk of lipodystrophy in HIV-1 patients treated with protease inhibitors: a prospective cohort study. Lancet. 2001;357:592-598.
36. Galli M, Ridolfo A, Fulvio A, et al. Body habitus changes and metabolic alterations in protease inhibitor-naive HIV-1-infected patients treated with 2 nucleoside reverse transcriptase inhibitors. J Acquir Immune Defic Syndr. 2002;29:21-31.
37. Moyle G, Sabin C, Cartledge J, et al. A randomized comparative trial of tenofovir DF or abacavir as replacement for a thymidine analogue in persons with lipoatrophy. AIDS. 2006;20:2043-2050.
38. Snijder M, Dekker J, Visser M, et al. Associations of hip and thigh circumferences independent of waist circumference with the incidence of type 2 diabetes: the Hoorn Study. Am J Clin Nutr. 2003;77:1192-1197.
39. Snijder M, Dekker J, Visser M, et al. Trunk fat and leg fat have independent and opposite associations with fasting and postload levels. Diabetes Care. 2004;27:372-377.
40. Snijder M, Visser M, Dekker J, et al. Low subcutaneous thigh fat is a risk factor for unfavorable glucose and lipid levels, independently of high abdominal fat. The Health ABC Study. Diabetologia. 2005;48:301-308.
41. Grunfeld C, Rimland D, Gibert C, et al. Association of upper trunk and visceral adipose tissue volume with insulin resistance in control and HIV-infected subjects in the FRAM study. J Acquir Immune Defic Syndr. 2007;46;283-290.
42. Wang J, Kotler D, Russell M, et al. Body-fat measurement in patients with acquired immunodeficiency syndrome: which method should be used? Am J Clin Nutr. 1992;56:963-967.
43. Knox T, Zafonte-Sanders M, Fields-Gardner C, et al. Assessment of nutritional status, body composition, and human immunodeficiency virus-associated morphologic changes. Clin Infect Dis. 2003;36(Suppl 2):S63-S68.
44. He Q, Wang J, Engelson E, et al. Ability of an anthropometric model to track changes in subcutaneous adipose tissue area. FASEB J. 2002;16: A1024-A1025.
© 2008 Lippincott Williams & Wilkins, Inc.