https://orcid.org/0000-0002-0073-873X
-
PDF
- Split View
-
Views
-
Cite
Cite
Juan Tiraboschi, Antonio Navarro-Alcaraz, Dolors Giralt, Carmen Gomez-Vaquero, Maria Saumoy, Arkaitz Imaz, Daniel Podzamczer, Changes in Body Fat Distribution in Antiretroviral-Naive HIV-Positive Individuals Initiating Current ART Regimens, The Journal of Clinical Endocrinology & Metabolism, Volume 104, Issue 3, March 2019, Pages 900–905, https://doi.org/10.1210/jc.2018-01489
Close - Share Icon Share
Abstract
To describe the changes in body fat distribution (BFD) occurring over 60 months in a group of antiretroviral therapy (ART)-naive individuals starting different antiretroviral regimens.
A prospective ongoing fat change assessment including clinical evaluation and dual X-ray absorptiometry scan is being conducted in all consecutive patients initiating ART from January 2008. Arm, leg, trunk, and total fat as well as fat mass ratio were determined.
A total of 146 patients were included (80% male, 40% MSM). Mean age was 44 years, HIV-1 RNA was 4.98 log10 copies/mL, and CD4 count was 254 cells/μL. The most common initial antiretroviral combination included non-nucleoside reverse transcription inhibitor (NNRTI) drugs followed by protease inhibitor (PI) and integrase strand transfer inhibitor (INSTI)-based regimens. At month 36, an increase was seen in the body mass index (BMI), total fat, trunk fat, and limb fat. The fat mass ratio (FMR) also showed a significant increase in both men and women (P = 0.001). In patients receiving NNRTI- or INSTI-based regimens (but not PIs), there was a marginal but statistically significant increase in the FMR (0.10 and 0.07, respectively; P = 0.01). Sixty-two subjects completed 60 months of follow-up. FMR showed a significant increase even in the PI group at this time point (P < 0.03).
We observed a significant increase in the fat and lean body mass in all compartments and treatment groups over 36 and 60 months. Clinically irrelevant differences were found in fat distribution regardless of the treatment group and baseline characteristics. The data suggest that current antiretroviral regimens have little impact on BFD during the first years of treatment.
The life expectancy and quality of life of individuals infected with HIV has improved significantly since the introduction of highly effective combined antiretroviral therapy (ART) (1–3). Moreover, the pill burden and side effects of antiretroviral drugs and treatment combinations have gradually decreased, and the treatment-related metabolic complications have become less prevalent. One of the most concerning and stigmatizing conditions related, at least in part, to ART is lipodystrophy, a diverse group of disorders characterized by changes in fat distribution and metabolic alterations (4–6). The changes that occur in body composition include lipoatrophy, which is complete or partial loss of the peripheral adipose tissue compartments, and lipohypertrophy, which is a pathological accumulation of central adipose tissue (7).
Although the use of a backbone of nonthymidine nucleoside analogs has greatly decreased the incidence of lipoatrophy, the effect of the third drug on fat redistribution [i.e., protease inhibitors (PIs), nonnucleoside reverse transcription inhibitors (NNRTIs), or integrase strand transfer inhibitors (INSTIs)] needs further clarification. Moreover, the role of current antiretrovirals in fat gain/loss is not well defined (8–12).
Concerns and uncertainties about fat distribution motivated us to perform a local, noninvasive, affordable long-term follow-up using annual dual-energy X-ray absorptiometry scan to collect data to reassure the low impact in fat changes with the use of newer antiretrovital agents. This was the reason to include this technique in our routine care of ARV-naive patients with HIV during several years.
The aim of this study was to describe the changes in body fat distribution (BFD) occurring over 60 months in a group of ART-naive individuals starting different antiretroviral regimens. We present data from patients with at least 36 months of follow-up and a subgroup reaching 60 months of follow-up.
Design and Methods
As part of our routine care of HIV-infected patients at a teaching hospital in Barcelona, Spain, we prospectively assessed body fat changes and clinical data before or within 1 month after initiating ART in patients with no clinically evident signs of lipodystrophy. Body fat composition was assessed by dual X-ray absorptiometry scanning (QDR 4500; Hologic, Marlborough, MA). Leg, trunk, and total fat, as well as the lean mass and fat mass ratio (FMR = %trunk fat/%leg fat) were determined in all participants at baseline and at months 12 and 36. Because this is an ongoing study, some participants had completed month 60 of follow-up by the time of this analysis. The fat-free mass index (FFMI) (kg/m2) was calculated as the ratio of FFM to height in meters squared. FFMI was classified as low (<16 kg/m2) or normal (>16 kg/m2). Fasting blood samples were taken as a part of the routine follow-up in the HIV unit. The densitometer technician followed a standardized protocol at each assessment, and dual X-ray absorptiometry scans were centrally read at the Rheumatology Service of Duran I Reynals Hospital, Barcelona.
Clinical and epidemiological data as well as the nearest to annual dual-energy X-ray absorptiometry scan routine blood determinations (CD4 count, HIV viral load, biochemistry, and full blood count) were exported from clinical records once yearly during the follow-up.
Statistical analysis
A normality test was used to assess homogeneity of variances. ANOVA or the Kruskal-Wallis test were used to compare continuous variables. Qualitative variables were compared using the χ2 or Fisher exact test. Comparisons between baseline and follow-up measures in each arm were done with ANOVA for repeated measures or with the Friedman test. Continuous variables are expressed as the mean and SD or the median and interquartile range. Qualitative variables are expressed as the number and percentage. For all statistical analyses, P values ≤0.05 were considered significant. Analyses were performed using SPSS v.19.0 (IBM, Chicago, IL).
Every patient enrolled in this study was analyzed as intention-to-treat, but on-treated analyses were performed in patients who completed the treatment at month 36 or 60.
Results
From January 2008 to January 2012, 146 patients with HIV participated in the study. Sixty-two patients completed the month 60 analysis. The mean age of the patients enrolled was 44 years, 80% (n = 117) were men, 40% were men who have sex with men (MSM) (n = 62), and 13.9% had AIDS (n = 20). Most patients were of Spanish (63%) or Latin American (26%) origin. The median time from HIV diagnosis to start of treatment was 11 months. Median BMI was 24. At baseline, median CD4 count was 254 (132 to 346) cells/mm, and median HIV viral load was 4.98 log10 copies/mL (4.3 to 5.35). None of the participants had clinical evidence of lipodystrophy according to the clinicians’ criteria and the patients’ self-perception.
The most common initial antiretroviral combination included NNRTI drugs (nevirapine, 23.3%; efavirenz, 17.2%) followed by protease inhibitor (PI)- and INSTI-based regimens (42.5% and 13%, respectively). The treatment regimen was considered PI-based when PIs were prescribed together with any other drugs. The most common backbone was a fixed-dose combination of tenofovir/emtricitabine prescribed in 76.7% of patients, followed by abacavir/lamivudine in 16.4%. Even though treatment was not randomly assigned, baseline characteristics between the groups defined by ART families were fairly balanced. FFMI at baseline was normal regardless of sex or treatment group.
At baseline, MSM had a significantly larger lean body mass than the other transmission groups. Similarly, patients from Spain had a significantly larger lean mass than Latin American patients. Baseline CD4 cell count <200 cells/mL was also associated with lower BMI and lean mass (Table 1). After 36 months of treatment, most patients remained on an NNRTI-based regimen (47.6%), followed by those receiving PIs (35.2%) and INSTIs (13.8%). By month 60, treatment distribution was as follows: NNRTI-based regimen 52.3%, PI 12.5%, and INSTI 22.7%.
| . | NNRTI (N = 65) . | PI (N = 62) . | INSTI (N = 19) . | TOTAL (N = 146) . | P Value . | ||||
|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | ||
| Age, y | 65 | 45.1 (10.5) | 62 | 42.8 (10.5) | 19 | 44.1 (7.3) | 146 | 44 (10.7) | 0.588a |
| Sex, % male | 53 | 81.5 | 51 | 82.3 | 13 | 68.4 | 117 | 80.1 | 0.388b |
| Risk practice | |||||||||
| MSM, % | 29 | 44.6 | 23 | 37.1 | 7 | 36.8 | 59 | 40.4 | 0.65a |
| Other, % | 36 | 54.4 | 39 | 62.9 | 19 | 63.2 | 87 | 59.6 | |
| AIDS, yes, % | 8 | 12.7 | 12 | 19.4 | 0 | 0 | 20 | 13.9 | 0.096b |
| CD4, cells/mL | 65 | 262 (173.9) | 62 | 233.1 (164.1) | 19 | 319.8 (112.5) | 146 | 257.23 (164.38) | 0.057a |
| Viral load, log10 | 65 | 5.45 (5.89) | 62 | 5.5 (5.8) | 19 | 5.1 (5.3) | 146 | 5.44 (5.81) | 0.391a |
| Total cholesterol, mmol/L | 65 | 4.4 (1) | 62 | 4.1 (1) | 19 | 4.2 (0.8) | 146 | 4.26 (0.99) | 0.218a |
| LDL, mmol/L | 64 | 2.7 (0.8) | 62 | 2.3 (0.9) | 19 | 2.6 (0.7) | 145 | 2.5 (0.8) | 0.051a |
| HDL, mmol/L | 65 | 1.1 (0.3) | 62 | 1 (0.4) | 19 | 1 (0.3) | 146 | 1.05 (0.35) | 0.189a |
| Ratio T Ch:HDL, mmol/L | 65 | 4.3 (1.6) | 62 | 4.4 (1.5) | 19 | 4.8 (3) | 146 | 4.42 (1.80) | 0.967a |
| TG, mmol/L | 65 | 1.5 (0.6) | 62 | 1.7 (1) | 19 | 1.3 (0.5) | 146 | 1.5 (0.8) | 0.622a |
| BMI | 65 | 25 (3.6) | 62 | 24.1 (3.9) | 19 | 24.8 (3.2) | 146 | 24.6 (3.7) | 0.413a |
| Time to first DXA scan, d, mean (95% CI) | 65 | 20.6 (14.5–26.7) | 62 | 24 (15.3–32.7) | 19 | 35.5 (17.8–53.2) | 146 | 24.0 (18.8–29.2) | 0.117a |
| Limb fat, g | 65 | 7016.7 (3138.8) | 62 | 6510.4 (3474.8) | 19 | 6467.6 (2740.3) | 146 | 6730.3 (3229.0) | 0.5021b |
| Trunk fat, g | 65 | 7572.8 (3958.7) | 62 | 7172.9 (4499.7) | 19 | 7464.6 (3670.5) | 146 | 7388.9 (4140.4) | 0.4871b |
| Total fat, g | 65 | 15,513.6 (6816.2) | 62 | 14,345.8 (7660.4) | 19 | 14,842 (6309.3) | 146 | 14,930.3 (7101.7) | 0.3781b |
| FMR | 65 | 1 (0.3) | 62 | 1 (0.3) | 19 | 1 (0.2) | 146 | 0.98 (0.26) | 0.8271b |
| Lean mass, g | |||||||||
| Trunk | 65 | 25,992.63 (3970.56) | 62 | 25,192.01 (4256.37) | 19 | 25,636.85 (4705.86) | 146 | 25,606.34 (4179.65) | 0.6232b |
| Limbs | 65 | 23,845.21 (4775.66) | 62 | 23,098.81 (4681.46) | 19 | 23,250.12 (5775.65) | 146 | 23,450.8 (4853.39) | 0.7212b |
| Total | 65 | 53,860.48 (8634.93) | 62 | 52,416.49 (9086.19) | 19 | 52,842.12 (10,530.63) | 146 | 53,114.75 (9048.51) | 0.7162b |
| FFMI | 65 | 18.40 (1.9) | 62 | 17.92 (2.5) | 19 | 18.25 (2.37) | 146 | 18.18 (2.25) | 0.49 |
| . | NNRTI (N = 65) . | PI (N = 62) . | INSTI (N = 19) . | TOTAL (N = 146) . | P Value . | ||||
|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | ||
| Age, y | 65 | 45.1 (10.5) | 62 | 42.8 (10.5) | 19 | 44.1 (7.3) | 146 | 44 (10.7) | 0.588a |
| Sex, % male | 53 | 81.5 | 51 | 82.3 | 13 | 68.4 | 117 | 80.1 | 0.388b |
| Risk practice | |||||||||
| MSM, % | 29 | 44.6 | 23 | 37.1 | 7 | 36.8 | 59 | 40.4 | 0.65a |
| Other, % | 36 | 54.4 | 39 | 62.9 | 19 | 63.2 | 87 | 59.6 | |
| AIDS, yes, % | 8 | 12.7 | 12 | 19.4 | 0 | 0 | 20 | 13.9 | 0.096b |
| CD4, cells/mL | 65 | 262 (173.9) | 62 | 233.1 (164.1) | 19 | 319.8 (112.5) | 146 | 257.23 (164.38) | 0.057a |
| Viral load, log10 | 65 | 5.45 (5.89) | 62 | 5.5 (5.8) | 19 | 5.1 (5.3) | 146 | 5.44 (5.81) | 0.391a |
| Total cholesterol, mmol/L | 65 | 4.4 (1) | 62 | 4.1 (1) | 19 | 4.2 (0.8) | 146 | 4.26 (0.99) | 0.218a |
| LDL, mmol/L | 64 | 2.7 (0.8) | 62 | 2.3 (0.9) | 19 | 2.6 (0.7) | 145 | 2.5 (0.8) | 0.051a |
| HDL, mmol/L | 65 | 1.1 (0.3) | 62 | 1 (0.4) | 19 | 1 (0.3) | 146 | 1.05 (0.35) | 0.189a |
| Ratio T Ch:HDL, mmol/L | 65 | 4.3 (1.6) | 62 | 4.4 (1.5) | 19 | 4.8 (3) | 146 | 4.42 (1.80) | 0.967a |
| TG, mmol/L | 65 | 1.5 (0.6) | 62 | 1.7 (1) | 19 | 1.3 (0.5) | 146 | 1.5 (0.8) | 0.622a |
| BMI | 65 | 25 (3.6) | 62 | 24.1 (3.9) | 19 | 24.8 (3.2) | 146 | 24.6 (3.7) | 0.413a |
| Time to first DXA scan, d, mean (95% CI) | 65 | 20.6 (14.5–26.7) | 62 | 24 (15.3–32.7) | 19 | 35.5 (17.8–53.2) | 146 | 24.0 (18.8–29.2) | 0.117a |
| Limb fat, g | 65 | 7016.7 (3138.8) | 62 | 6510.4 (3474.8) | 19 | 6467.6 (2740.3) | 146 | 6730.3 (3229.0) | 0.5021b |
| Trunk fat, g | 65 | 7572.8 (3958.7) | 62 | 7172.9 (4499.7) | 19 | 7464.6 (3670.5) | 146 | 7388.9 (4140.4) | 0.4871b |
| Total fat, g | 65 | 15,513.6 (6816.2) | 62 | 14,345.8 (7660.4) | 19 | 14,842 (6309.3) | 146 | 14,930.3 (7101.7) | 0.3781b |
| FMR | 65 | 1 (0.3) | 62 | 1 (0.3) | 19 | 1 (0.2) | 146 | 0.98 (0.26) | 0.8271b |
| Lean mass, g | |||||||||
| Trunk | 65 | 25,992.63 (3970.56) | 62 | 25,192.01 (4256.37) | 19 | 25,636.85 (4705.86) | 146 | 25,606.34 (4179.65) | 0.6232b |
| Limbs | 65 | 23,845.21 (4775.66) | 62 | 23,098.81 (4681.46) | 19 | 23,250.12 (5775.65) | 146 | 23,450.8 (4853.39) | 0.7212b |
| Total | 65 | 53,860.48 (8634.93) | 62 | 52,416.49 (9086.19) | 19 | 52,842.12 (10,530.63) | 146 | 53,114.75 (9048.51) | 0.7162b |
| FFMI | 65 | 18.40 (1.9) | 62 | 17.92 (2.5) | 19 | 18.25 (2.37) | 146 | 18.18 (2.25) | 0.49 |
Abbreviations: DXA, dual X-ray absorptiometry; T Ch, total cholesterol; TG, triglycerides.
Kruskal-Wallis test.
ANOVA.
| . | NNRTI (N = 65) . | PI (N = 62) . | INSTI (N = 19) . | TOTAL (N = 146) . | P Value . | ||||
|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | ||
| Age, y | 65 | 45.1 (10.5) | 62 | 42.8 (10.5) | 19 | 44.1 (7.3) | 146 | 44 (10.7) | 0.588a |
| Sex, % male | 53 | 81.5 | 51 | 82.3 | 13 | 68.4 | 117 | 80.1 | 0.388b |
| Risk practice | |||||||||
| MSM, % | 29 | 44.6 | 23 | 37.1 | 7 | 36.8 | 59 | 40.4 | 0.65a |
| Other, % | 36 | 54.4 | 39 | 62.9 | 19 | 63.2 | 87 | 59.6 | |
| AIDS, yes, % | 8 | 12.7 | 12 | 19.4 | 0 | 0 | 20 | 13.9 | 0.096b |
| CD4, cells/mL | 65 | 262 (173.9) | 62 | 233.1 (164.1) | 19 | 319.8 (112.5) | 146 | 257.23 (164.38) | 0.057a |
| Viral load, log10 | 65 | 5.45 (5.89) | 62 | 5.5 (5.8) | 19 | 5.1 (5.3) | 146 | 5.44 (5.81) | 0.391a |
| Total cholesterol, mmol/L | 65 | 4.4 (1) | 62 | 4.1 (1) | 19 | 4.2 (0.8) | 146 | 4.26 (0.99) | 0.218a |
| LDL, mmol/L | 64 | 2.7 (0.8) | 62 | 2.3 (0.9) | 19 | 2.6 (0.7) | 145 | 2.5 (0.8) | 0.051a |
| HDL, mmol/L | 65 | 1.1 (0.3) | 62 | 1 (0.4) | 19 | 1 (0.3) | 146 | 1.05 (0.35) | 0.189a |
| Ratio T Ch:HDL, mmol/L | 65 | 4.3 (1.6) | 62 | 4.4 (1.5) | 19 | 4.8 (3) | 146 | 4.42 (1.80) | 0.967a |
| TG, mmol/L | 65 | 1.5 (0.6) | 62 | 1.7 (1) | 19 | 1.3 (0.5) | 146 | 1.5 (0.8) | 0.622a |
| BMI | 65 | 25 (3.6) | 62 | 24.1 (3.9) | 19 | 24.8 (3.2) | 146 | 24.6 (3.7) | 0.413a |
| Time to first DXA scan, d, mean (95% CI) | 65 | 20.6 (14.5–26.7) | 62 | 24 (15.3–32.7) | 19 | 35.5 (17.8–53.2) | 146 | 24.0 (18.8–29.2) | 0.117a |
| Limb fat, g | 65 | 7016.7 (3138.8) | 62 | 6510.4 (3474.8) | 19 | 6467.6 (2740.3) | 146 | 6730.3 (3229.0) | 0.5021b |
| Trunk fat, g | 65 | 7572.8 (3958.7) | 62 | 7172.9 (4499.7) | 19 | 7464.6 (3670.5) | 146 | 7388.9 (4140.4) | 0.4871b |
| Total fat, g | 65 | 15,513.6 (6816.2) | 62 | 14,345.8 (7660.4) | 19 | 14,842 (6309.3) | 146 | 14,930.3 (7101.7) | 0.3781b |
| FMR | 65 | 1 (0.3) | 62 | 1 (0.3) | 19 | 1 (0.2) | 146 | 0.98 (0.26) | 0.8271b |
| Lean mass, g | |||||||||
| Trunk | 65 | 25,992.63 (3970.56) | 62 | 25,192.01 (4256.37) | 19 | 25,636.85 (4705.86) | 146 | 25,606.34 (4179.65) | 0.6232b |
| Limbs | 65 | 23,845.21 (4775.66) | 62 | 23,098.81 (4681.46) | 19 | 23,250.12 (5775.65) | 146 | 23,450.8 (4853.39) | 0.7212b |
| Total | 65 | 53,860.48 (8634.93) | 62 | 52,416.49 (9086.19) | 19 | 52,842.12 (10,530.63) | 146 | 53,114.75 (9048.51) | 0.7162b |
| FFMI | 65 | 18.40 (1.9) | 62 | 17.92 (2.5) | 19 | 18.25 (2.37) | 146 | 18.18 (2.25) | 0.49 |
| . | NNRTI (N = 65) . | PI (N = 62) . | INSTI (N = 19) . | TOTAL (N = 146) . | P Value . | ||||
|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | n . | Mean (SD) . | ||
| Age, y | 65 | 45.1 (10.5) | 62 | 42.8 (10.5) | 19 | 44.1 (7.3) | 146 | 44 (10.7) | 0.588a |
| Sex, % male | 53 | 81.5 | 51 | 82.3 | 13 | 68.4 | 117 | 80.1 | 0.388b |
| Risk practice | |||||||||
| MSM, % | 29 | 44.6 | 23 | 37.1 | 7 | 36.8 | 59 | 40.4 | 0.65a |
| Other, % | 36 | 54.4 | 39 | 62.9 | 19 | 63.2 | 87 | 59.6 | |
| AIDS, yes, % | 8 | 12.7 | 12 | 19.4 | 0 | 0 | 20 | 13.9 | 0.096b |
| CD4, cells/mL | 65 | 262 (173.9) | 62 | 233.1 (164.1) | 19 | 319.8 (112.5) | 146 | 257.23 (164.38) | 0.057a |
| Viral load, log10 | 65 | 5.45 (5.89) | 62 | 5.5 (5.8) | 19 | 5.1 (5.3) | 146 | 5.44 (5.81) | 0.391a |
| Total cholesterol, mmol/L | 65 | 4.4 (1) | 62 | 4.1 (1) | 19 | 4.2 (0.8) | 146 | 4.26 (0.99) | 0.218a |
| LDL, mmol/L | 64 | 2.7 (0.8) | 62 | 2.3 (0.9) | 19 | 2.6 (0.7) | 145 | 2.5 (0.8) | 0.051a |
| HDL, mmol/L | 65 | 1.1 (0.3) | 62 | 1 (0.4) | 19 | 1 (0.3) | 146 | 1.05 (0.35) | 0.189a |
| Ratio T Ch:HDL, mmol/L | 65 | 4.3 (1.6) | 62 | 4.4 (1.5) | 19 | 4.8 (3) | 146 | 4.42 (1.80) | 0.967a |
| TG, mmol/L | 65 | 1.5 (0.6) | 62 | 1.7 (1) | 19 | 1.3 (0.5) | 146 | 1.5 (0.8) | 0.622a |
| BMI | 65 | 25 (3.6) | 62 | 24.1 (3.9) | 19 | 24.8 (3.2) | 146 | 24.6 (3.7) | 0.413a |
| Time to first DXA scan, d, mean (95% CI) | 65 | 20.6 (14.5–26.7) | 62 | 24 (15.3–32.7) | 19 | 35.5 (17.8–53.2) | 146 | 24.0 (18.8–29.2) | 0.117a |
| Limb fat, g | 65 | 7016.7 (3138.8) | 62 | 6510.4 (3474.8) | 19 | 6467.6 (2740.3) | 146 | 6730.3 (3229.0) | 0.5021b |
| Trunk fat, g | 65 | 7572.8 (3958.7) | 62 | 7172.9 (4499.7) | 19 | 7464.6 (3670.5) | 146 | 7388.9 (4140.4) | 0.4871b |
| Total fat, g | 65 | 15,513.6 (6816.2) | 62 | 14,345.8 (7660.4) | 19 | 14,842 (6309.3) | 146 | 14,930.3 (7101.7) | 0.3781b |
| FMR | 65 | 1 (0.3) | 62 | 1 (0.3) | 19 | 1 (0.2) | 146 | 0.98 (0.26) | 0.8271b |
| Lean mass, g | |||||||||
| Trunk | 65 | 25,992.63 (3970.56) | 62 | 25,192.01 (4256.37) | 19 | 25,636.85 (4705.86) | 146 | 25,606.34 (4179.65) | 0.6232b |
| Limbs | 65 | 23,845.21 (4775.66) | 62 | 23,098.81 (4681.46) | 19 | 23,250.12 (5775.65) | 146 | 23,450.8 (4853.39) | 0.7212b |
| Total | 65 | 53,860.48 (8634.93) | 62 | 52,416.49 (9086.19) | 19 | 52,842.12 (10,530.63) | 146 | 53,114.75 (9048.51) | 0.7162b |
| FFMI | 65 | 18.40 (1.9) | 62 | 17.92 (2.5) | 19 | 18.25 (2.37) | 146 | 18.18 (2.25) | 0.49 |
Abbreviations: DXA, dual X-ray absorptiometry; T Ch, total cholesterol; TG, triglycerides.
Kruskal-Wallis test.
ANOVA.
There was an overall increase in the mean (SD) CD4 cell count and mean (SD) log10 decline in HIV viral load during follow-up: 358.6 cells/ml (226.7) and −5.44 (5.81), respectively (P < 0.001).
At 12 months of treatment, we observed a significant increase in the BMI, total fat, and lean mass in all groups, with mean (SD) changes from baseline to week 48 of 0.90 (1.97), 2082 g (4003), and 621.7 g (49.8), respectively. In addition, there was significant fat accumulation in the trunk and limbs and a significant increase in lean mass in the limbs (P = 0.01). The overall FMR changes were small but statistically significant in men; there were no differences in women. No associations were found between the FMR changes and any of the treatment regimens (Table 2).
| . | NNRTI . | PI . | INSTI . | Total . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | |
| BMI | 65 | 0.99 (2.3) | 0.001 | 62 | 1.48 (2.27) | <0.001 | 19 | 0.61 (1.34) | 0.084 | 146 | 1.13 (2.08) | <0.001 |
| Limb fat, g | 65 | 1399.7 (2009.8) | <0.001 | 62 | 1816.2 (2170.9) | 0.001 | 19 | 1865.2 (1425.3) | <0.001 | 146 | 1637.2 (2015.8) | <0.001a |
| Trunk fat, g | 65 | 1169.7 (2911.5) | <0.001b | 62 | 1309.2 (3093.8) | <0.001b | 19 | 2174.3 (2035.7) | 0.001b | 146 | 1359.7 (2896.0) | <0.001a |
| Total fat, g | 65 | 2649.3 (4444.1) | <0.001 | 62 | 3533.7 (5115.6) | <0.001 | 19 | 4289.4 (3051.2) | <0.001 | 146 | 3238.3 (4606.5) | <0.001a |
| FMR | 65 | 0.10 (0.31) | 0.001b | 62 | 0.00 (0.19) | 0.723b | 19 | 0.07 (0.11) | 0.013b | 146 | 0.06 (0.25) | <0.001a |
| Lean mass, g | ||||||||||||
| Trunk | 65 | −636.78 (2205.16) | 0.013 | 62 | 157.86 (2449.8) | 0.902 | 19 | −1526.97 (1325.79) | <0.001 | 146 | −415.18 (2283.35) | 0.006 |
| Limbs | 65 | 637.97 (1836.09) | 0.013 | 62 | 989.16 (2423.41) | 0.004 | 19 | 145.18 (1867.42) | 0.936 | 146 | 722.98 (2113.96) | <0.001 |
| Total | 65 | 183.8 (3540.92) | 0.688 | 62 | 1188.7 (4195.88) | 0.042 | 19 | −1247.35 (2048.96) | 0.011 | 146 | 424.29 (3755.47) | 0.569 |
| FFMI, kg/m2 | 65 | 0.06 (1.19) | 0.67 | 62 | 0.40 (1.44) | 0.032 | 19 | −0.42 (0.72) | 0.02 | 146 | 424.29 (3755.47) | 0.038 |
| . | NNRTI . | PI . | INSTI . | Total . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | |
| BMI | 65 | 0.99 (2.3) | 0.001 | 62 | 1.48 (2.27) | <0.001 | 19 | 0.61 (1.34) | 0.084 | 146 | 1.13 (2.08) | <0.001 |
| Limb fat, g | 65 | 1399.7 (2009.8) | <0.001 | 62 | 1816.2 (2170.9) | 0.001 | 19 | 1865.2 (1425.3) | <0.001 | 146 | 1637.2 (2015.8) | <0.001a |
| Trunk fat, g | 65 | 1169.7 (2911.5) | <0.001b | 62 | 1309.2 (3093.8) | <0.001b | 19 | 2174.3 (2035.7) | 0.001b | 146 | 1359.7 (2896.0) | <0.001a |
| Total fat, g | 65 | 2649.3 (4444.1) | <0.001 | 62 | 3533.7 (5115.6) | <0.001 | 19 | 4289.4 (3051.2) | <0.001 | 146 | 3238.3 (4606.5) | <0.001a |
| FMR | 65 | 0.10 (0.31) | 0.001b | 62 | 0.00 (0.19) | 0.723b | 19 | 0.07 (0.11) | 0.013b | 146 | 0.06 (0.25) | <0.001a |
| Lean mass, g | ||||||||||||
| Trunk | 65 | −636.78 (2205.16) | 0.013 | 62 | 157.86 (2449.8) | 0.902 | 19 | −1526.97 (1325.79) | <0.001 | 146 | −415.18 (2283.35) | 0.006 |
| Limbs | 65 | 637.97 (1836.09) | 0.013 | 62 | 989.16 (2423.41) | 0.004 | 19 | 145.18 (1867.42) | 0.936 | 146 | 722.98 (2113.96) | <0.001 |
| Total | 65 | 183.8 (3540.92) | 0.688 | 62 | 1188.7 (4195.88) | 0.042 | 19 | −1247.35 (2048.96) | 0.011 | 146 | 424.29 (3755.47) | 0.569 |
| FFMI, kg/m2 | 65 | 0.06 (1.19) | 0.67 | 62 | 0.40 (1.44) | 0.032 | 19 | −0.42 (0.72) | 0.02 | 146 | 424.29 (3755.47) | 0.038 |
Wilcoxon U test.
Student t test for repeated measures.
| . | NNRTI . | PI . | INSTI . | Total . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | |
| BMI | 65 | 0.99 (2.3) | 0.001 | 62 | 1.48 (2.27) | <0.001 | 19 | 0.61 (1.34) | 0.084 | 146 | 1.13 (2.08) | <0.001 |
| Limb fat, g | 65 | 1399.7 (2009.8) | <0.001 | 62 | 1816.2 (2170.9) | 0.001 | 19 | 1865.2 (1425.3) | <0.001 | 146 | 1637.2 (2015.8) | <0.001a |
| Trunk fat, g | 65 | 1169.7 (2911.5) | <0.001b | 62 | 1309.2 (3093.8) | <0.001b | 19 | 2174.3 (2035.7) | 0.001b | 146 | 1359.7 (2896.0) | <0.001a |
| Total fat, g | 65 | 2649.3 (4444.1) | <0.001 | 62 | 3533.7 (5115.6) | <0.001 | 19 | 4289.4 (3051.2) | <0.001 | 146 | 3238.3 (4606.5) | <0.001a |
| FMR | 65 | 0.10 (0.31) | 0.001b | 62 | 0.00 (0.19) | 0.723b | 19 | 0.07 (0.11) | 0.013b | 146 | 0.06 (0.25) | <0.001a |
| Lean mass, g | ||||||||||||
| Trunk | 65 | −636.78 (2205.16) | 0.013 | 62 | 157.86 (2449.8) | 0.902 | 19 | −1526.97 (1325.79) | <0.001 | 146 | −415.18 (2283.35) | 0.006 |
| Limbs | 65 | 637.97 (1836.09) | 0.013 | 62 | 989.16 (2423.41) | 0.004 | 19 | 145.18 (1867.42) | 0.936 | 146 | 722.98 (2113.96) | <0.001 |
| Total | 65 | 183.8 (3540.92) | 0.688 | 62 | 1188.7 (4195.88) | 0.042 | 19 | −1247.35 (2048.96) | 0.011 | 146 | 424.29 (3755.47) | 0.569 |
| FFMI, kg/m2 | 65 | 0.06 (1.19) | 0.67 | 62 | 0.40 (1.44) | 0.032 | 19 | −0.42 (0.72) | 0.02 | 146 | 424.29 (3755.47) | 0.038 |
| . | NNRTI . | PI . | INSTI . | Total . | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | n . | Mean (SD) . | P Value . | |
| BMI | 65 | 0.99 (2.3) | 0.001 | 62 | 1.48 (2.27) | <0.001 | 19 | 0.61 (1.34) | 0.084 | 146 | 1.13 (2.08) | <0.001 |
| Limb fat, g | 65 | 1399.7 (2009.8) | <0.001 | 62 | 1816.2 (2170.9) | 0.001 | 19 | 1865.2 (1425.3) | <0.001 | 146 | 1637.2 (2015.8) | <0.001a |
| Trunk fat, g | 65 | 1169.7 (2911.5) | <0.001b | 62 | 1309.2 (3093.8) | <0.001b | 19 | 2174.3 (2035.7) | 0.001b | 146 | 1359.7 (2896.0) | <0.001a |
| Total fat, g | 65 | 2649.3 (4444.1) | <0.001 | 62 | 3533.7 (5115.6) | <0.001 | 19 | 4289.4 (3051.2) | <0.001 | 146 | 3238.3 (4606.5) | <0.001a |
| FMR | 65 | 0.10 (0.31) | 0.001b | 62 | 0.00 (0.19) | 0.723b | 19 | 0.07 (0.11) | 0.013b | 146 | 0.06 (0.25) | <0.001a |
| Lean mass, g | ||||||||||||
| Trunk | 65 | −636.78 (2205.16) | 0.013 | 62 | 157.86 (2449.8) | 0.902 | 19 | −1526.97 (1325.79) | <0.001 | 146 | −415.18 (2283.35) | 0.006 |
| Limbs | 65 | 637.97 (1836.09) | 0.013 | 62 | 989.16 (2423.41) | 0.004 | 19 | 145.18 (1867.42) | 0.936 | 146 | 722.98 (2113.96) | <0.001 |
| Total | 65 | 183.8 (3540.92) | 0.688 | 62 | 1188.7 (4195.88) | 0.042 | 19 | −1247.35 (2048.96) | 0.011 | 146 | 424.29 (3755.47) | 0.569 |
| FFMI, kg/m2 | 65 | 0.06 (1.19) | 0.67 | 62 | 0.40 (1.44) | 0.032 | 19 | −0.42 (0.72) | 0.02 | 146 | 424.29 (3755.47) | 0.038 |
Wilcoxon U test.
Student t test for repeated measures.
At month 36, an increase was seen in the BMI, total fat, trunk fat, and limb fat. The FMR also showed a significant increase in both men and women (P = 0.001). In patients receiving NNRTI- or INSTI-based regimens (but not PIs), there was a marginal but statistically significant increase in the FMR (0.10 and 0.07, respectively; P = 0.01).
At this time point, a significant increase in the FMR and a decrease in the trunk lean mass occurred in the MSM population. Patients who had a lower CD4 cell count at baseline (<200 cell/mL) gained substantially more fat and lean mass in all compartments, maintaining the FMR with no significant changes at month 36 (data not shown). A small increase in the FMR was observed in patients with higher baseline CD4 counts (>200 cells/mL). FFMI changed differently between groups. No relevant differences were observed between NNRTI vs PI or NNRTI vs INSTI, but there was a marginal significant reduction in FFMI in subjects receiving INSTI compared with those in the PI group (mean/SD, −424.1/724 vs 401.6/1440.2; P = 0.04). BMI increased significantly in those receiving PI compared with INSTI (Table 2).
Sixty-two individuals completed 60 months of follow-up. Changes similar to those seen at months 12 and 36 occurred in the BMI, total fat, trunk fat, and limb fat. The FMR showed a significant increase even in the PI group at this time point (P < 0.03).
We observed a decline in the total lean body mass, led by the trunk fraction. These results were seen in all the treatment groups but were statistically significant in patients receiving NNRTI- and INSTI-based regimens. These results persisted at month 60.
Significant increases in total, HDL, and LDL cholesterol were also observed in all groups and at 12, 36, and 60 months. Triglycerides remained unchanged.
The mean (SD) total lean mass and limb lean mass were significantly higher in patients younger than 50 years compared with those older than 50 years: 54,085.9 g (8945.8) vs 50,354.4 g (8879.9) (P = 0.03) and 24,078.8 g (4750.5) vs 21,665.7 g (4756.5) (P = 0.01), respectively. At 36 months, the mean FMR increase was 0.14 (P = 0.001) in patients older than 50 years and 0.03 (0.18) (P = 0.057) in patients younger than 50 years.
In the on-treatment analysis at month 36 (patients who remained in the same treatment group, n = 101), only patients receiving PIs showed no significant increase in the FMR. No differences were observed regarding the NRTI backbone. The remaining markers showed changes similar to those in the intent-to-treat analysis.
Discussion
We compared the changes in body composition after initiation of ART in 146 individuals with HIV at 36 months and in a smaller group at 60 months. To our knowledge, this is one of the few cohorts in which long-term changes in BFD are reported after treatment initiation in the recent antiretroviral era. We observed a significant increase in the fat and lean body mass in all compartments and treatment groups over follow-up. Small and clinically insignificant differences were found in fat distribution regardless of the treatment group and baseline characteristics. Most of the initial changes seemed to be related with the “back to health effect” occurring with treatment. Even patients who initiated treatment with lower CD4 cell counts gained fat mass, and the redistribution of this component was balanced (FMR unchanged). The return to health in terms of the fat and BMI increase in this immunosuppressed population responding to ART is consistent with the decrease in mortality observed in a previous study by our group (12).
Recently published reports have suggested that INSTI may be associated with greater weight gain compared with other ARV families in ARV-experienced subjects (13). In our cohort of treatment-naive patients, initiating antiretroviral treatment of the first time, we did not observe this association, although the sample size of patients receiving INSTI was small. Moreover, those receiving PIs had a proportionally greater weight gain (BMI).
Some sex differences in fat distribution were observed, with men gaining proportionally more fat in trunk than in limbs. Sex-specific differences in the prevalence of different lipodystrophy syndromes have been described. In a study of 180 adults with HIV in Brazil, women were more likely to develop lipohypertrophy or mixed syndrome, whereas men showed a similar prevalence of all three lipodystrophy syndromes (lipoatrophy, lipohypertrophy, and mixed syndrome) (13). Other studies, using older antiretroviral drugs (13, 14), have reported differing fat distribution changes in men and women on ART, with women being more likely to develop fat accumulation around the abdomen. A more recent study has shown opposite results (similar to ours); that is, an increase in visceral adipose tissue was associated with male sex and age but not with ART exposure (15).
Ethnicity is also an important factor influencing lipodystrophy, with black women being more vulnerable to lipohypertrophy (14). In our cohort, Latin American patients had significantly less lean mass than the native Spanish population. However, no significant differences were observed over follow-up between these groups.
Aging is also a relevant factor associated with body composition changes, especially lean mass loss and trunk fat gain. This process is known to be accelerated in patients with HIV (16). In our cohort, patients younger than 50 years had a larger lean mass at baseline. At 36 months, the FMR had increased significantly in those older than 50 years. Some of these changes might be associated with this phenomenon.
Patients with HIV show significantly greater increases in lean mass and total fat than individuals without HIV, particularly those with a lower baseline CD4 T-cell count and a higher HIV-1 viral load. The fat gain in these patients is thought to be a consequence of treating HIV infection, which reduces inflammation and wasting (recovery effect). In our cohort, this notion was supported by the differences seen from baseline to months 36 and 60. Conversely, there was a significant decline in the total lean mass at months 36 and 60, driven by a decrease in the trunk lean mass. The FMR increased significantly during this period, which suggests that the trunk-to-limbs ratio was influenced by a gain in trunk fat. We found no significant relationship between HIV disease–related markers, such as CD4 T-cell count and HIV-1 viral load, and the changes in body composition. These results are consistent with a study in which HIV-infected persons gained more fat but lost lean mass after the first 96 weeks of ART compared with control subjects with HIV (17). The mechanism to explain why there is a gain in trunk fat but less lean mass after the first 36 months of ART remains unclear.
A persistent proinflammatory state in treated patients with HIV may contribute to the changes in body composition over the long term (18, 19). As was seen in another recently published study (12), we found no clear relationship between fat distribution changes and ART groups.
The older antiretroviral drugs thymidine analogs and first-generation PIs have been related to lipoatrophy and lipohypertrophy, respectively (20). One previous study has suggested that efavirenz is associated with peripheral fat loss (8). Lipodystrophy may be associated with the newer antiretroviral drugs, but the marginal changes make its presentation clinically unapparent. Although our study was not designed to detect minor metabolic changes, we believe that an impact at this level might be associated with subtle changes in fat distribution. Early recognition of these changes may help to prevent potential damage.
Our study has a number of limitations. First, treatment allocation was not randomized; all data were obtained from routine follow-up assessments in patients initiating ART. In addition, because few women were enrolled in our study, we are unable to establish firm conclusions for this population. Due to the time when follow-up was started, only a small percentage of patients had started on the newer INSTIs. Finally, despite the final objective of the study was to assess fat changes after 60 months of follow-up, this was achieved in a limited number of patients. A slow recruitment process as well as a strong need of data regarding this important issue motivated us to report our preliminary results (month 36) at this point together with data of a group of patients reaching 60 months. We believe this information may be useful and may help to shed some light in this regard.
Due to the relatively small number of patients in different subgroups analyzed, we should be cautious about the results and conclusions drawn in the current study.
The new antiretroviral drugs are clearly less toxic, and their impact on BFD seems to be low. However, irreversible changes in body shape and the potential impact of ART on metabolic syndrome remain of crucial interest, particularly in younger patients initiating lifelong treatment.
Well-designed, long-term studies with newer agents are needed to understand and minimize the impact of ART in HIV-infected individuals. Despite the aforementioned limitations, our data suggest that current antiretroviral regimens have a low impact on fat redistribution in the first 3 to 5 years of therapy.
Abbreviations:
- ART
antiretroviral therapy
- BFD
body fat distribution
- BMI
body mass index
- FFMI
fat-free mass index
- FMR
fat mass ratio
- INSTI
integrase strand transfer inhibitor
- MSM
men who have sex with men
- NNRTI
nonnucleoside reverse transcription inhibitor
- PI
protease inhibitor
Acknowledgments
We thank CERCA Programme/Generalitat de Catalunya for institutional support. We dedicate this paper to our colleague and friend Elena Ferrer, MD, who has recently died. She largely contributed to the design and implementation of this study. Thanks, Helen, for all your important contributions to our HIV Unit throughout more than two decades. We will miss you.
Disclosure Summary: J.T. has received financial compensation for lectures, consultancies, and educational activities,as well as research funding for from AbbVie, Gilead Sciences, Janssen-Cilag, Merck Sharp & Dome, and ViiV Healthcare. E.F. has received honoraria for advisories and/or conferences from Viiv, BMS, Abbott, Gilead, Janssen, and Merck. M.S. has received honoraria for advisories and/or conferences from Viiv, BMS, Abbott, Gilead, Janssen, and Merck. A.I. has received financial compensation for lectures, consultancies, and educational activities or research funding for from AbbVie, Gilead Sciences, Janssen-Cilag, Merck Sharp & Dome, and ViiV Healthcare. D.P. has received research grants and/or honoraria for advisories and/or conferences from Viiv, BMS, Abbott, Gilead, Janssen, and Merck.
