Abstract
Background: Associations between abnormal body fat distribution and clinical variables are poorly understood in pediatric HIV disease.
Objective: Our objective was to compare total body fat and its distribution in perinatally HIV-infected and HIV-exposed but uninfected (HEU) children and to evaluate associations with clinical variables.
Design: In a cross-sectional analysis, children aged 7–16 y in the Pediatric HIV/AIDS Cohort Study underwent regionalized measurements of body fat via anthropometric methods and dual-energy X-ray absorptiometry. Multiple linear regression was used to evaluate body fat by HIV, with adjustment for age, Tanner stage, race, sex, and correlates of body fat in HIV-infected children. Percentage total body fat was compared with NHANES data.
Results: Males accounted for 47% of the 369 HIV-infected and 51% of the 176 HEU children. Compared with HEU children, HIV-infected children were older, were more frequently non-Hispanic black, more frequently had Tanner stage ≥3, and had lower mean height (−0.32 compared with 0.29), weight (0.13 compared with 0.70), and BMI (0.33 compared with 0.63) z scores. On average, HIV-infected children had a 5% lower percentage total body fat (TotF), a 2.8% lower percentage extremity fat (EF), a 1.4% higher percentage trunk fat (TF), and a 10% higher trunk-to-extremity fat ratio (TEFR) than did the HEU children and a lower TotF compared with NHANES data. Stavudine use was associated with lower EF and higher TF and TEFR. Non-nucleotide reverse transcriptase inhibitor use was associated with higher TotF and EF and lower TEFR.
Conclusion: Although BMI and total body fat were significantly lower in the HIV-infected children than in the HEU children, body fat distribution in the HIV-infected children followed a pattern associated with cardiovascular disease risk and possibly related to specific antiretroviral drugs.
INTRODUCTION
HIV-infected children have benefited from the success of HAART5, yet problematic side effects pose a continuing challenge. Altered body composition, lipid abnormalities, and dysregulation of glucose metabolism (1–16)—all factors that lead to increased global risk of cardiovascular disease (17–19)—are complications of HIV disease and specific ARV drugs.
Patterns of abnormal fat distribution or lipodystrophy with HIV infection vary from peripheral fat wasting (lipoatrophy) in the face, extremities, and/or buttocks to central fat accumulation (lipohypertrophy) in the abdomen, dorso-cervical spine regions (buffalo hump), and/or breasts. Lipoatrophy and lipohypertrophy may occur alone or in combination (20, 21) and may be difficult to assess in children because changes in body fat occur normally as children progress through puberty (22–25). The prevalence of lipodystrophy in HIV-infected children is usually observed or evaluated by clinical examination or self-report and may be as high as 32% (2, 10, 26). Imaging can detect clinically unrecognized variation in fat depots (27). DXA is used to quantify total, trunk, and limb fat. Computed tomography and magnetic resonance imaging are used to differentiate between abdominal VAT and SAT. Several observational studies show decreased total and extremity fat and greater trunk-to-extremity fat ratio in HIV-infected compared with uninfected children by DXA (4, 28, 29). Others have shown no differences by HIV status in VAT, SAT, or VAT-SAT ratio by computed tomography (3, 30). The type of comparison group varies across studies, and few include perinatally HEU children (31, 32) who may be more comparable with HIV-infected children in terms of sociodemographic and other characteristics.
Less extremity fat in HIV-infected children is associated with more advanced HIV disease (26, 27, 31) and NRTI use (26, 27, 29). PI use is associated with increased trunk-to-extremity fat ratio (26, 28). Many studies of body fat in HIV-infected children are underpowered to evaluate underlying risk factors, such as sociodemographic characteristics, clinical status, puberty, and different ARV exposures. Thus, little is known about the effects of lifetime exposure to individual ARV drugs on body fat distribution. In the era of effective HAART, an understanding of factors associated with metabolic complications could inform treatment decisions and thereby decrease long-term risk of cardiovascular outcomes as HIV-infected children age into adulthood. The objectives of this study are to compare total body fat and its distribution in perinatally HIV-infected children with similar data in HEU children and with NHANES norms and to evaluate associations of body fat with lifetime ARV exposure in infected children. Our hypothesis was that lipodystrophy is related to HIV infection and to specific ARV drugs.
SUBJECTS AND METHODS
Population
The AMP of the PHACS is an ongoing prospective cohort study conducted at 15 US sites. It was designed to define the effect of HIV infection and ARVs on preadolescents and adolescents with perinatal HIV infection. HEU children were enrolled as a comparison group. Enrollment of children from 7 y of age to their 16th birthday began in March 2007 and ended in December 2009 when the targeted number of HIV-infected and HEU children was reached. Participants in AMP attended clinic visits semiannually for the assessment of numerous clinical and behavioral outcomes. A DXA scan was performed at the entry visit for all HIV-infected children and for HEU who were at least 12 y old. For HEU children who were not yet 12 y of age at entry, a DXA was performed when they turned 12 or in the fourth year of PHACS if they were aged <12 y. We also used norms from NHANES on children as a reference group for some measures described below. This analysis includes children who had a DXA scan, anthropometric measurements, and other relevant covariates as of 1 April 2010. The Institutional Review Boards at all clinical sites and the Harvard School of Public Health (Statistical and Data Management Center) approved the protocol, and informed consent from the parent(s) or guardian(s) and assent from the participants (when appropriate) were obtained.
Sociodemographic and clinical history
Through a chart review, the use of data from prior study participation, and an interview we collected sociodemographic information; clinical events; current, ever, and lifetime duration of ARV use by class and individual agents; concomitant medications; CDC clinical disease category (N/A, B, or C); current and nadir CD4+ T lymphocytes (percentages); and current and peak HIV viral load (copies/mL). Systolic and diastolic blood pressures were measured in all participants, and z scores were computed on the basis of age, sex, and height z score. Fasting lipids (total, LDL, and HDL cholesterol), insulin, and glucose were measured annually in HIV-infected children according to standard techniques.
Measurements
Anthropometric measures
A registered dietitian experienced in anthropometric measurements conducted training sessions at the annual PHACS meeting to standardize measurement techniques across study sites. Weight (kg), height (m), and BMI [weight (kg)/height2 (m)] were expressed as z scores for age and sex by using CDC 2000 growth charts (33), and suprailiac skinfold thicknesses were measured. For children in the age group of our study, these norms are based on data from 1963 to 1974. Waist and hip circumferences (cm) were measured with a nonstretchable plastic tape measure according to standard methods. Waist-hip ratio was calculated as the waist circumference divided by the hip circumference (34). To determine the prevalence of obesity, we calculated the percentage of children with a BMI z score ≥95th percentile (35). We defined 2 variables of wasting (weight z score <−2.0 SD and BMI z score <−2 SD) and one for stunting (height z score <−2 SD).
Tanner stage
Tanner staging was assessed by site clinicians who routinely evaluated growth and development in the children and who received standardized training from the same pediatric endocrinologist (MEG). Tanner stage was ascertained by inspection of breasts and pubic hair for females and of genitalia and pubic hair for males at semiannual visits until Tanner stage 5 was reached. For boys and girls, the more advanced stage of the 2 respective pubertal components was used for classification if there was discordance between the examined body sites (eg, between breast and pubic hair in females).
DXA
Total-body DXA scans including the head were performed on a Lunar (General Electric Health Care) or Hologic (Hologic Inc) scanner, and the data were sent to the Body Composition Analysis Center at Tufts University School of Medicine for analysis. All scans were analyzed by using the same respective software version (Hologic QDR for Windows XP v12.3, APEX v2.3.1, and Lunar GE enCORE 2005 v9.00.219) with standardization of scans across sites. The total body Bio-Imaging Variable Composition Phantom instrument was circulated and scanned at each clinical site to cross-calibrate DXA scanners. DXA measurements included total body mass (kg), fat mass (kg), and lean mass (kg). The body was divided into regions to measure extremity fat (kg) and trunk fat (kg). The trunk is segmented by placing the regions of interest between the bottom of the chin and the top of the shoulders, through the gleno-humeral joint to separate the arms, and directly above the iliac crest to separate the legs.
Body fat measurements by DXA
Percentage body fat was calculated by dividing total body fat (kg) by total body mass (kg) and then multiplying by 100. Regional percentage fat measurements were calculated as the amount of fat in the region (kg) divided by total body fat (kg) and then multiplying by 100. These regional measurements included percentage extremity fat, percentage trunk fat, percentage upper extremity fat, and percentage lower extremity fat. We evaluated percentage upper and lower extremity fat separately to compare our results with those of Arpadi et al (29), who found the percentage of fat to be lower in the lower extremities and higher in the upper extremities and trunk of HIV-infected than of -uninfected children. These uninfected children were community controls and not known to be exposed to HIV. Trunk-to-extremity fat ratio was calculated as trunk fat (kg) divided by extremity fat (kg).
Statistical analysis
Sociodemographic and anthropometric characteristics of HIV-infected and HEU children were compared by using Fisher's exact test for categorical variables and Wilcoxon's rank-sum tests for continuous variables. One-sample t tests were performed in HIV-infected and HEU children separately to test whether z scores for height, weight, and BMI differed from the standard norm of 0 based on CDC 2000 norms (33). NHANES measured percentage body fat by DXA from 1999 to 2004 in a representative cross-sectional sample of the US population aged 8 to >80 y (36). We plotted the means and SEs for percentage body fat by group (NHANES, HIV-infected, and HEU) within subgroups of age (8–11 and12–15 y of age), sex (males and females), and race-ethnicity (Hispanic and non-Hispanic black). Hispanics in AMP were compared with Mexican Americans in NHANES because NHANES did not report means (±SEs) overall for Hispanics. We compared HEU and HIV-infected children with NHANES children of the same age group, sex, and race-ethnicity using a 2-sample t test and a 2-sided P value. We did not plot values or perform t tests for percentage total fat for non-Hispanic white/other because there were too few HIV-infected and HEU children in each category for meaningful representation.
Comparison of body-composition outcomes by HIV status
The raw means and SDs for each body fat measurement were computed for HIV-infected and HEU children overall and separately by sex. With the use of general linear regression for each outcome, we estimated the difference between HIV-infected and HEU children, adjusted for age (y), race (non-Hispanic black, Hispanic, and non-Hispanic white/other), sex, and Tanner stage (1–5). The ratio of trunk-to-extremity fat and waist-hip ratio were log10 transformed for analysis, but adjusted differences between HIV-infected and HEU was exponentiated for presentation. The interpretation for trunk-to-extremity fat, for example, is the ratio of trunk-to-extremity fat in the HIV-infected divided by the trunk-to-extremity fat ratio in the HEU. Thus, if the ratio is 1.1, it can also be interpreted that the HIV-infected children have a 10% higher trunk-to-extremity fat ratio than the HEU children. For every model, we examined the distribution of the Studentized residuals for extreme values (<−3, >3). Models were refit without extreme values. If no qualitative difference in the variance of the variables between the models with and without outliers existed, all data points were retained. We plotted the Studentized residuals against age, as a continuous variable, to evaluate linearity. Two-way interactions (effect modification) were fit for HIV status, with race-ethnicity, sex, and Tanner stage for each outcome. If an interaction term was significant at P < 0.05, the term was kept in the model; otherwise no interaction was reported.
Differences in body composition by HIV disease severity and ARV treatment
Among HIV-infected children, we evaluated potential correlates of each DXA body fat measurement. The correlates tested were highest lifetime HIV viral load (copies/mL), lowest lifetime CD4%, CDC category, wasting, stunting, “ever use” of each ARV class, and each ARV ever drug used by ≥10% of HIV-infected children. We also evaluated ever use of ritonavir and ever use of indinavir (used by 5.7% of children).
We evaluated “ever use” rather than “current use” because a child may have been switched to the current regimen due to previous toxicities such as lipodystrophy. First, for each outcome, we tested each covariate in a separate model (basic models). Second, covariates that were associated with the outcome at P < 0.20 in a basic model were tested together in a multivariable model. Third, the final model (full model) included covariates with a P ≤ 0.10 or covariates that were confounders of other variables in the model. We adjusted all of the models described above for age (y), sex, race-ethnicity, and Tanner stage. For the ARVs included in each final model, we also examined the association of body fat measurement with lifetime exposure (y) to that ARV. We categorized lifetime exposure as never, <2, 2–4, and >4 y. If inspection showed a linear trend with increasing or decreasing duration of exposure, duration was kept in the model. Otherwise, ever versus never use was reported. In a secondary analysis, we examined whether age at initiation of ARVs in the final models was associated with body fat measurements. Thus, among those taking a particular ARV, we included a term for age at initiation of use of that ARV (<5, 5–9, 10–13, and ≥14 y of age) rather than including ever use. We examined a categorical variable (<1996, 1996–1998, and 1999 onward) for year of first use of ARV, other than neonatal prophylaxis. We did not evaluate cardiovascular disease risk factors in these models because they are, in general, an outcome of body fat abnormalities. All models were tested for extreme values and linearity as described previously. All analyses were performed in SAS version 9.2 (SAS Institute). No adjustments were made for multiple comparisons.
In HIV-infected children, we evaluated correlations of each body fat measurement by DXA with fasting lipids (total cholesterol, triglycerides, LDL, and HDL), the homeostatic model assessment of insulin resistance or HOMA-IR [(fasting insulin in μU/mL × fasting glucose in mmol/L)/22.5], and blood pressure (diastolic and systolic) z scores. We tested these associations with Spearman partial correlations adjusted for age, sex, race-ethnicity, and Tanner stage. We additionally performed Spearman correlations between percentage trunk and extremity fat with suprailiac skinfold thickness in all HIV-infected and HEU children, unadjusted for other variables.
RESULTS
Characteristics of HIV-infected and HEU children
A total of 545 children in AMP (369 HIV-infected and 176 HEU) with a baseline DXA were included in this analysis (Table 1). The proportion of males was similar in the HIV-infected and HEU groups. The HIV-infected children were more likely to be black non-Hispanic, older, and more advanced in Tanner stage than were the HEU children. The HIV-infected children had lower mean (±SD) z scores for height, weight, and BMI than did the HEU children. Compared with population norms, the HIV-infected cohort had significantly lower height z scores and slightly above average weight and BMI z scores. The HEU had height, weight, and BMI z scores higher than the population norms. The HEU children had a higher prevalence of obesity than did the HIV-infected children. These z scores were not adjusted for race-ethnicity. The means (±SDs) for cardiovascular disease risk factors among HIV-infected children are shown in Table 1. Markers of disease severity (CD4, viral load, and CDC category) and ARV use (current, ever, and duration) are reported in Table 2.
TABLE 1.
Characteristics of HIV-infected and HEU children1
| Characteristic | HIV-infected (n = 369) | HEU (n = 176) | P value2 |
| Male sex [n (%)] | 173 (47) | 89 (51) | 0.42 |
| Age (y) | 12.2 ± 2.63 | 10.9 ± 2.3 | <0.01 |
| Race-ethnicity [n (%)] | 0.02 | ||
| Hispanic | 98 (27) | 66 (38) | |
| Non-Hispanic black | 241 (65) | 93 (53) | |
| Non-Hispanic white/other | 29 (8) | 17 (10) | |
| Unknown | 1 (0) | 0 (0) | |
| Height z score | −0.32 ± 1.124 | 0.29 ± 1.094 | <0.01 |
| <−2 SD | 27 ± 7 | 2 ± 1 | 0.003 |
| Weight z score | 0.13 ± 1.235 | 0.70 ± 1.404 | <0.01 |
| <−2 SD | 18 ± 5 | 5 ± 3 | 0.26 |
| BMI | |||
| BMI z score | 0.33 ± 1.114 | 0.63 ± 1.364 | 0.01 |
| BMI ≥95th percentile [n (%)] | 46 (12) | 49 (28) | <0.01 |
| Tanner stage [n (%)] | <0.01 | ||
| 1 | 95 (26) | 64 (36) | |
| 2 | 76 (21) | 57 (32) | |
| 3 | 62 (17) | 19 (11) | |
| 4 | 77 (21) | 19 (11) | |
| 5 | 59 (16) | 17 (10) | |
| Metabolic6 | |||
| Total cholesterol (mg/dL) | 170 ± 39 | — | |
| HDL cholesterol (mg/dL) | 52 ± 15 | — | |
| LDL cholesterol (mg/dL) | 89 ± 32 | — | |
| Triglycerides (mg/dL) | 108 ± 62 | — | |
| HOMA-IR | 2.4 ± 3.8 | — | |
| Blood pressure | |||
| Systolic z score | 0.42 ± 0.94 | — | |
| Diastolic z score | 0.14 ± 0.66 | — |
HEU, HIV-exposed but uninfected.
P value for difference between HIV-infected and HEU children by using Fisher's exact test for categorical variables and the Wilcoxon's test for continuous variables.
Mean ± SD (all such values).
P < 0.01 from a one-sample t test comparing z scores in HIV-infected or HEU children with a z score of standard normal of 0 based on CDC 2000.
P < 0.05 from a one-sample t test comparing z scores in HIV-infected or HEU children with a z score of standard normal of 0 based on CDC 2000.
Fasting values.
TABLE 2.
Immunologic, virologic, and treatment-specific characteristics of HIV-infected children1
| Characteristic | HIV-infected (n = 369) |
| CD4 T cell (%) | |
| Current | 31.8 (1–57)2 |
| Nadir | 19.7 (0–50) |
| HIV viral load [n (%)] | |
| Current | |
| ≤400 copies/mL | 247 (67) |
| 401–5000 copies/mL | 62 (17) |
| >5000 copies/mL | 60 (16) |
| Peak | |
| ≤400 copies/mL | 31 (8) |
| 401–5000 copies/mL | 36 (10) |
| 5000–10,000 copies/mL | 27 (7) |
| >10,000 copies/mL | 275 (75) |
| CDC clinical category [n (%)] | |
| N/A | 178 (49) |
| B | 105 (28) |
| C | 86 (23) |
| Current ARV [n (%)] | |
| PI-based HAART | 259 (70) |
| Non-PI-based HAART | 16 (16) |
| Non-HAART ARV | 25 (7) |
| No ARV | 22 (6) |
| Never ARV | 4 (1) |
| Individual ARV [n (%)]3 | |
| PI | |
| Atazanavir | |
| Current | 40 (11) |
| Ever | 54 (15) |
| Nelfinavir | |
| Current | 55 (15) |
| Ever | 222 (60) |
| Ritonavir-lopinavir | |
| Current | 202 (55) |
| Ever | 252 (68) |
| Saquinavir | |
| Current | 10 (3) |
| Ever | 39 (11) |
| NRTI | |
| Abacavir | |
| Current | 52 (14) |
| Ever | 135 (37) |
| Lamivudine | |
| Current | 196 (53) |
| Ever | 335 (91) |
| Didanosine | |
| Current | 83 (22) |
| Ever | 252 (68) |
| Stavudine | |
| Current | 96 (26) |
| Ever | 285 (77) |
| Tenofovir | |
| Current | 77 (21) |
| Ever | 90 (24) |
| Zidovudine | |
| Current | 105 (28) |
| Ever | 315 (85) |
| NNRTI | |
| Efavirenz | |
| Current | 65 (18) |
| Ever | 122 (33) |
| Nevirapine | |
| Current | 27 (7) |
| Ever | 129 (35) |
| Lifetime use (mo) | |
| PI | 95 (1–155) |
| NRTI | 123 (11–181) |
| Stavudine | 75 (0.3–158) |
| NNRTI | 44 (0.03–138) |
ARV, antiretroviral; HAART, highly active antiretroviral therapy; NNRTI, non-nucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
Median; range in parentheses (all such values).
Percentages are reported for agents ever used by ≥10% of HIV-infected children. Timpranavir, indinavir, saquinavir, darunavir, fosamprenavir, and amprenavir were used by <10% of children. Percentages represent the use of the ARV in a single or multidrug combination.
Percentage total fat by DXA in HIV-infected, HEU, and NHANES children by sex, age, and race-ethnicity
The mean (95% CI) percentage total body fat for Hispanics and non-Hispanic blacks by age and sex are plotted in Figure 1. Among 8–11-y-olds, all HEU children had percentage total body fat similar to that of children in NHANES, regardless of sex or race-ethnicity subgroup (Figure 1A). In this same age group, HIV-infected Hispanic and non-Hispanic black males and non-Hispanic females had a significantly lower percentage total body fat than did NHANES children. However, HIV-infected Hispanic females in this age range did not differ significantly from NHANES.
FIGURE 1.
Percentage total body fat in HEU and HIV-infected children compared with NHANES norms by age, sex, and race-ethnicity by age group: 8–11 y (A) and 12–15 y (B). HEU and HIV-infected children were compared, separately, with NHANES children within subgroups (age, sex, race-ethnicity) by using t tests. **,***Significantly different from NHANES: **P< 0.01, ***P < 0.001. At the bottom of each graph is the number of children in the bar for mean and 95% CI above in the graph. HEU, HIV-exposed but uninfected; Hisp., Hispanic.
Among 12–15-y-olds, HEU males had percentage total body fat similar to that of NHANES children in each race-ethnicity subgroup (Figure 1B). Whereas HEU non-Hispanic black females did not differ from NHANES children in the same age group, HEU Hispanic girls had significantly lower percentage total body fat than did NHANES children in this group. HIV-infected males aged 12–15 y had lower percentage total body fat than did NHANES children in each race-ethnicity subgroup. HIV-infected non-Hispanic black females also had a significantly lower percentage total body fat than did NHANES children, but Hispanic females in this age group did not differ from NHANES children.
Differences in body composition between HIV-infected and HEU children
For each body fat measurement, the raw means and SDs for HIV-infected and HEU overall and by sex are shown (Table 3) with adjusted differences (95% CI) between groups. In the adjusted models, HIV-infected children had a lower percentage total fat and extremity fat and a higher percentage trunk fat and trunk-to-extremity fat ratio compared with HEU children. Percentage arm fat did not differ significantly between HIV-infected and HEU children, but percentage leg fat was lower in the HIV-infected children than in the HEU children. The waist and hip circumferences were lower in HIV-infected than in HEU children. Waist-to-hip ratio did not differ significantly between HIV-infected and HEU children. For all body fat measures, the differences between HIV-infected and HEU were similar across sex, race-ethnicity, and Tanner stages (no interaction).
TABLE 3.
Comparison of body fat measures in HIV-infected and HEU children1
| Raw values |
Adjusted difference for comparison of HIV-infected with HEU2 |
|||
| Fat distribution | HIV-infected (n = 369) | HEU (n = 176) | Difference (95% CI) | P value |
| %Total fat by DXA | ||||
| All | 22.1 ± 10.3 | 27.2 ± 11.8 | −5.0 (−6.8, −3.1) | <0.01 |
| Male | 18.1 ± 10.0 | 24.1 ± 11.2 | — | |
| Female | 25.7 ± 9.2 | 30.4 ± 11.5 | — | |
| %Extremity fat by DXA | ||||
| All | 50.1 ± 6.6 | 52.9 ± 4.7 | −2.8 (−3.9, −1.7) | <0.01 |
| Male | 49.9 ± 6.6 | 52.9 ± 4.6 | — | |
| Female | 50.3 ± 6.5 | 52.9 ± 4.9 | — | |
| %Arm fat by DXA | ||||
| All | 8.8 ± 2.6 | 9.4 ± 2.6 | −0.4 (−0.8, 0.1) | 0.12 |
| Male | 8.5 ± 2.5 | 9.2 ± 2.7 | — | |
| Female | 9.1 ± 2.7 | 9.6 ± 2.5 | — | |
| %Leg fat by DXA | ||||
| All | 41.3 ± 6.6 | 43.4 ± 4.8 | −2.4 (−3.5, −1.3) | <0.01 |
| Male | 41.4 ± 6.5 | 43.7 ± 5.0 | — | |
| Female | 41.1 ± 6.7 | 43.2 ± 4.6 | — | |
| %Trunk fat by DXA | ||||
| All | 42.0 ± 7.6 | 39.8 ± 6.5 | 1.4 (0.1, 2.7) | 0.04 |
| Male | 41.1 ± 7.6 | 39.0. ± 6.7 | — | |
| Female | 42.8 ± 7.4 | 40.7 ± 6.2 | — | |
| Waist circumference (cm) | ||||
| All | 70.3 ± 12.6 | 71.8 ± 15.8 | −4.1 (−6.5, −1.8) | <0.01 |
| Male | 70.1 ± 12.9 | 69.0 ± 14.8 | — | |
| Female | 70.4 ± 12.5 | 74.7 ± 16.3 | — | |
| Hip circumference (cm) | ||||
| All | 79.2 ± 13.5 | 81.0 ± 16.3 | −5.7 (−8.0, −3.3) | <0.01 |
| Male | 78.3 ± 13.3 | 78.3 ± 14.1 | — | |
| Female | 80.0 ± 13.7 | 83.7 ± 18.0 | — | |
| Trunk-to-extremity fat ratio | ||||
| All | 0.87 ± 0.303 | 0.77 ± 0.183 | 1.1 (1.0, 1.1)4 | <0.01 |
| Males | 0.86 ± 0.293 | 0.75 ± 0.193 | — | |
| Female | 0.89 ± 0.303 | 0.78 ± 0.183 | — | |
| Waist-to-hip ratio | ||||
| All | 0.89 ± 0.103 | 0.89 ± 0.103 | 1.0 (1.0, 1.0)4 | 0.09 |
| Males | 0.90 ± 0.123 | 0.88 ± 0.093 | — | |
| Female | 0.88 ± 0.083 | 0.90 ± 0.113 | — | |
All raw values are means ± SDs adjusted for age (y), sex, race-ethnicity, and Tanner stage. DXA, dual-energy X-ray absorptiometry; HEU, HIV-exposed but uninfected.
Values are ratios of the ratio (eg, mean trunk-to-extremity ratio in HIV-infected children/mean trunk-to-extremity ratio in HEU children), mean waist-to-hip ratio in HIV-infected children/mean waist-to-hip ratio in HEU children. For the trunk-to-extremity ratio, the result is interpreted as the HIV-infected children having a 1.1 times greater or 10% higher trunk-to-extremity fat ratio than do the HEU children.
Values are the mean fat ratios.
Values are ratios (95% CIs).
Multivariable models of each DXA body fat measure in HIV-infected children
Sociodemographic characteristics
The 4 final multivariable models, one for each body fat measurement by DXA, among HIV-infected children are shown in Table 4. Percentage total fat and percentage trunk fat were lower in males than in females, but percentage extremity fat and the trunk-to extremity fat ratio did not differ by sex. Percentage total fat and percentage extremity fat differed significantly by race-ethnicity. The following pairwise comparisons indicate which groups differed significantly from each other. Hispanics had significantly higher percentage total body fat than did non-Hispanic blacks and non-Hispanic white/other children. Hispanic children had lower percentage extremity fat than did non-Hispanic black children.
TABLE 4.
Multivariable models of 4 DXA-derived measures of body fat in HIV-infected children: associations with sociodemographic, disease severity, and ARV drugs1
| Outcome variable: body fat measure2 |
||||||||
| Percentage total fat3 |
Percentage extremity fat3 |
Percentage trunk fat3 |
Trunk-to-extremity fat ratiob |
|||||
| Correlate | Estimate (95% CI) | P value | Estimate (95% CI) | P value | Estimate (95% CI) | P value | Estimate (95% CI) | P value |
| Male sex | −8.0 (−9.9, −6.0) | <0.01 | −0.5 (−1.8, 0.8) | 0.44 | −1.9 (−3.4, −0.4) | 0.01 | 1.0 (0.9, 1.0) | 0.23 |
| Age (y) | 0.4 (−0.3, 1.0) | 0.31 | −0.2 (−0.3, 0.7) | 0.44 | 0.5 (0.05, 1.0) | 0.08 | 1.0 (0.98, 1.0) | 0.56 |
| Race-ethnicity | <0.01 | 0.04 | 0.11 | 0.08 | ||||
| Hispanic | 4.2 (0.3, 8.1)b | −0.7 (−3.3, 1.9)a | 1.2 (−1.8, 4.1) | 1.0 (0.9, 1.2) | ||||
| Non-Hispanic black | −1.1 (−4.7, 2.4)a | 1.1 (−1.3, 3.6)b | −0.6 (−3.4, 2.1) | 1.0 (0.9, 1.1) | ||||
| Non-Hispanic white/other | Referencea | Referencea,b | Reference | Reference | ||||
| Tanner stage | 0.14 | 0.08 | 0.38 | 0.14 | ||||
| 4–5 | −1.4 (−5.9, 3.1) | −2.5 (−5.6, 0.5) | 2.4 (−1.1, 5.9) | 1.1 (1.0, 1.3) | ||||
| 2–3 | 1.1 (−2.0, 4.2) | 0.3 (−2.5, 1.8) | 1.1 (−1.4, 3.5) | 1.0 (0.9, 1.1) | ||||
| 1 | Reference | Reference | Reference | Reference | ||||
| ARV, ever-use4 | ||||||||
| PI | — | — | 1.4 (−1.1, 3.8) | 0.27 | — | |||
| Stavudine | — | −3.3 (−4.9, −1.7) | <0.01 | 2.1 (0.2, 3.9) | 0.03 | 1.1 (1.1, 1.2) | <0.01 | |
| NNRTI | 1.6 (−0.4, 3.5) | 0.12 | 1.8 (0.4, 3.2) | 0.01 | — | 0.9 (0.87, 0.98) | 0.02 | |
| CDC disease category | 0.38 | 0.16 | 0.16 | |||||
| C | — | −1.2 (−2.8, 0.5) | 1.8 (−0.1, 3.7) | 1.1 (1.0, 1.2) | ||||
| B | — | −0.5 (−2.1, 1.0) | 0.3 (−1.5, 2.0) | 1.0 (0.9, 1.1) | ||||
| NA | — | Reference | Reference | Reference | ||||
| Wasting/stunting | ||||||||
| Weight z score <−2 | −8.4 (−12.8, −3.9) | <0.01 | — | −3.3 (−7.3, 0.6) | 0.09 | — | ||
| Height z score <−2 | — | −5.7 (−8.3, −3.2) | <0.01 | 4.4 (1.1, 7.8) | <0.01 | 1.2 (1.1, 1.4) | <0.01 | |
Groups with different superscript letters are significantly different, P < 0.05. ARV, antiretroviral; DXA, dual-energy X-ray absorptiometry; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
There were 4 final models, one for each body fat measure (the outcome in each model). All models include age (y), sex, race-ethnicity, and Tanner stage. Additional variables included in each model were either significant at P ≤ 0.1 or were confounders of other variables in the model. Dashed lines indicate that the variable was not included in that model. P values are reported to 2 digits. The P values for each outcome are from the F test of any difference across subgroups. If the overall P value from the F test was <0.05, pairwise comparisons were performed.
The estimates (95% CIs) and P values under each outcome are for each correlate included in the model. The estimates for percentage total, extremity, and trunk fat are differences from the reference group. For example, males have a 7.9% lower percentage total fat than females. For trunk-to-extremity fat, it is the ratio of the ratio (eg, mean trunk-to-extremity ratio in HIV-infected/mean trunk-to-extremity ratio in HIV-exposed but uninfected children). For example, ever-users of stavudine have a 1.1 times greater or 10% higher trunk-to-extremity fat ratio than never-users of stavudine.
Each ARV class and individual ARV listed in Table 2 and indinavir and ritonavir were tested for inclusion in each model of body fat measurement.
ARV use and disease severity
HIV-infected children who had ever used stavudine had lower percentage extremity fat and higher percentage trunk fat and a trunk-to-extremity fat ratio than never users. Children who had ever used NNRTIs had higher percentage total fat and percentage extremity fat and a lower trunk-to-extremity fat ratio than did those who had never used NNRTIs. Ever use of a PI was not a significant independent predictor of any body fat measurement even after adjustment for NNRTI use. PI was included as a confounder in the model of percentage trunk fat. CDC disease category was associated with trunk-to-extremity fat ratio. In a pairwise comparison, children with a history of CDC category C had a higher trunk-to-extremity fat ratio than did those with a history of category B and N/A. CDC category was included as a confounder in the models of percentage extremity fat and percentage trunk fat, but was not a significant independent predictor in either model.
Wasting and stunting
Children with a weight z score <−2 SD had less percentage total body fat, whereas those with a height z < −2 SD had lower percentage extremity fat and higher percentage trunk fat and trunk-to-extremity fat ratio. A BMI z score < −2 SD was also tested in models but was less significant than weight z score, so it was not included.
Other variables were tested in the models but were not included (other individual ARVs in Table 2, ritonavir ever use, age at first ARV, CD4%, and viral load) because they were not associated with the body fat measurements at P ≤ 0.1 and were not confounders. Finally, no evidence suggested that body fat measurements were increasing or decreasing with greater lifetime exposure to any of the ARVs included in the above models. There was also no clear pattern suggesting that earlier or later age at initiation of these ARVs was associated with body fat measurements. Thus, these variables were not included in any model.
Partial correlations between body fat by DXA and cardiovascular disease risk factors and abdominal skinfold-thickness measurement
Higher percentage total fat was associated with higher LDL cholesterol (r = 0.13, P = 0.03), HOMA-IR (r = 0.34, P < 0.0001), and systolic blood pressure (r = 0.13, P = 0.01). Lower percentage extremity fat was associated with higher triglycerides (r = −0.21, P = 0.0002) and lower HDL cholesterol (r = 0.12, P = 0.03). Higher percentage trunk fat was associated with higher triglycerides (r = 0.18, P = 0.001) and systolic blood pressure (r = 0.13, P = 0.01) and lower HDL (r = −0.16, P = 0.005). Finally, a higher trunk-to-extremity fat ratio was associated with higher triglycerides (r = 0.20, P = 0.0004) and systolic blood pressure (r = 0.10, P = 0.05) and lower HDL cholesterol (r = −0.15, P = 0.008). Suprailiac skinfold thickness was positively and strongly associated with percentage trunk fat (r = 0.38, P < 0.0001) and negatively and weakly associated with percentage extremity fat (r = −0.09, P = 0.04).
DISCUSSION
We evaluated anthropometric- and DXA-measured body fat distribution in a large cohort of perinatally HIV-infected and HEU preadolescents and adolescents. Whereas infected children were of normal weight on average, they had less total body fat than did the HEU children and NHANES norms and a greater proportion of trunk to extremity fat than did the HEU children. HEU children had a high rate of obesity. HIV-infected children who had ever used stavudine had a lower percentage extremity fat and higher percentage trunk fat and trunk-to-extremity fat ratio. Ever having received PI was not significantly associated with any measure of fat distribution. The proportion of extremity fat was higher and the trunk-to-extremity fat ratio was lower in those who had ever received an NNRTI. This pattern of excess fat in the trunk relative to the extremities in infected children may reflect increased cardiovascular disease risk as in the general population.
HIV-infected children have made improvements in linear growth and weight because of the introduction of PIs and HAART (8, 31, 32, 37). Although they may lag in height compared with HIV-uninfected and HEU children, HIV-infected children catch up on weight (32), as we observed. Whereas 12% of HIV-infected children were obese on the basis of CDC growth charts, their weight and BMI were still lower than those of the HEU children. This finding is supported by most (27, 28, 31, 38), but not all (29), current studies. In our study, HIV-infected children with wasting had less total fat and those with stunting had more fat in the trunk and less in the extremities. Wasting and stunting may have long-term effects on fat distribution or it may be due to the fact that HIV-infected children have shorter limbs. In addition, the 28% prevalence of obesity in the HEU children on the basis of CDC growth charts was greater than the more current 19% rate in US adolescents overall in 2007–2008 (35) but was closer to rates in US Hispanic boys (27%) and non-Hispanic black girls (26%). This high rate may reflect the growing trend of overweight/obesity related to lifestyle factors in this sociodemographic group (39), although a contributory role of fetal exposure to HIV and/or ARV drug use remains possible. The HEU children in our study appeared to be a good comparison group for our HIV-infected children because they had similar sociodemographic characteristics, and the HEU children had rates of obesity and body fat similar to those of the general population.
Our findings of lower total and extremity body fat concur with those of other DXA studies (4, 28, 29), 2 of which reported similar lean mass in HIV-infected and -uninfected children (28, 4). In contrast, Arpadi et al (29) observed similar total fat, trunk fat, and percentage total fat in HIV-infected and -uninfected children, but lower leg and higher arm fat in infected children. Because of our cross-sectional study design, we could not determine the temporal sequence of changes in each body fat compartment.
Our findings are critically important because excess trunk fat in children and adolescents is linked to cardiovascular disease risk factors. In the Bogalusa Heart Study, children with greater trunk fat had higher blood pressure (40) and lipid, lipoprotein (41), and insulin levels (42), even after levels of peripheral fat and obesity were accounted for. However, peripheral fat provided no additional information over central fat on associations with lipid measures (41). HIV-uninfected children with greater central obesity had more adverse levels of cardiometabolic risk factors, even those of normal weight (43). Thus, the associations we observed between trunk fat and cardiovascular disease risk factors (lipids, HOMA-IR, blood pressure) may place HIV-infected children at higher risk of cardiometabolic complications over time (44).
Healthy female children and young adults have greater increases in total fat relative to lean mass than do males as they progress through puberty (22, 23): the trunk-to-leg fat ratio increases with advancing Tanner stage in females (22). Our data suggest that differences in fat measures between HIV-infected and HEU children remain constant across these factors and that, overall, percentage total fat was similar between the HEU and NHANES children but was lower in HIV-infected children. Among HIV-infected children, differences in trunk fat differed by sex and race, but not by Tanner stage, adjusted for age. Longitudinal studies are needed to understand changes in body fat in infected children throughout adolescence.
HIV disease severity (26, 27, 31) and specific ARVs (4, 12, 45) may affect acquisition or loss of fat. In our study, children who ever used stavudine had lower percentage extremity fat and higher percentage trunk and trunk-to-extremity fat ratio even after adjustment for wasting or stunting. No other NRTI was associated with body fat measures. In children and adults, stavudine use is strongly associated with lipoatrophy (26, 27). Both stavudine and zidovudine can inhibit mitochondrial DNA polymerase-γ and lead to mitochondrial DNA depletion and/or impairment in the production of mitochondrial enzymes and proteins involved in oxidative phosphorylation (46, 47). Additionally, mitochondrial peripheral blood mononuclear cell 8-oxo-deoxyguanine concentrations improved in adults who switched from stavudine to tenofovir, and extremity fat improved after discontinuation of stavudine or zidovudine (48–50). We found no clear effect of age at stavudine initiation or with duration of use. These findings are difficult to interpret because stavudine may be discontinued because of toxicity, or persistent effects may remain after discontinuation or switching to another NRTI or ARV class (45, 51).
PIs may decrease adipocyte differentiation (52); however, we did not find a strong association between PI use and body fat. Most of our infected children had PI exposure, which may have limited our ability to detect differences. PIs have been associated with lipodystrophy (26). Aldrovandi et al (28) found lower extremity fat, similar trunk fat, and a higher trunk-to-extremity fat ratio in children taking PIs than in HIV-uninfected children. Also, children who were naive to PIs had lower extremity fat, a trend toward lower trunk fat, and a similar trunk-extremity ratio compared with HIV-uninfected children. The findings of switch studies are less clear. Moyle et al (50) found increases in visceral fat in HIV-infected adults who switched from PIs or NNRTIs to abacavir, but decreases in those who switched from stavudine and PIs or NNRTIs to abacavir and zidovudine. In children, changes in body measurements did not differ between children who discontinued and those who remained on PIs (53).
Efavirenz, an NNRTI, can suppress lipogenic pathways of adipocytes in vitro (54). We found significantly higher percentage extremity fat and a lower trunk-to-extremity fat ratio in HIV-infected children who ever used NNRTIs. The estimated percentage body fat of those taking NNRTIs was lower than that of children in NHANES of comparable race-ethnicity. This finding may be related to improvements in lipodystrophy after the switch to NNRTIs (51) or to an effect of NNRTIs on fat deposition. Gynecomastia has been reported in HIV-infected men on NNRTI therapy (55).
Our cross-sectional study had some important limitations. Temporal relations between ARVs and body fat could not be discerned, and no information was available on treatment choices or DXA before treatment initiation. Children may have switched medication because of lipodystrophy. We could not differentiate between subcutaneous and visceral fat with the use of DXA, but trunk fat by DXA correlates well with visceral fat by magnetic resonance imaging (56). The positive correlation of suprailiac skinfold thickness with DXA percentage trunk fat and the negative and weak association with percentage extremity fat suggests increases in SAT in the abdomen. However, more sensitive techniques would be required to confirm this. Whereas birth weight may be a confounder of body composition, this variable was not available on most of the children. Finally, although we did not have an HIV-unexposed group, we were able to compare some measures with CDC growth charts and more recent data on pediatric population norms.
Despite having a lower BMI and lower percentage body fat than the HEU children, the HIV-infected children in our study had altered fat distribution characterized by greater central relative to extremity fat compared with the HEU children, which was associated with specific ARV drugs and disease-severity indicators. This finding warrants careful monitoring given the co-occurrence of elevated rates of dyslipidemia.
Acknowledgments
We thank the children and families for their participation in the PHACS AMP and the individuals and institutions involved in the conduct of PHACS AMP. The following institutions, clinical site investigators, and staff participated in conducting PHACS AMP in 2010, in alphabetical order: Baylor College of Medicine (William Shearer, Norma Cooper, and Lynette Harris), Bronx Lebanon Hospital Center (Murli Purswani, Mahboobullah Baig, and Anna Cintron), Children's Diagnostic & Treatment Center (Ana Puga, Sandra Navarro, and Doyle Patton), Children's Hospital Boston (Sandra Burchett, Nancy Karthas, and Betsy Kammerer), Children's Memorial Hospital (Ram Yogev, Kathleen Malee, Scott Hunter, and Eric Cagwin), Jacobi Medical Center (Andrew Wiznia, Marlene Burey, and Molly Nozyce), St Christopher's Hospital for Children (Janet Chen, Elizabeth Gobs, and Mitzie Grant), St Jude Children's Research Hospital (Katherine Knapp, Kim Allison, and Patricia Garvie), San Juan Hospital/Department of Pediatrics (Midnela Acevedo-Flores, Heida Rios, and Vivian Olivera), Tulane University Health Sciences Center (Margarita Silio, Medea Jones, Cheryl Borne, and Patricia Sirois), University of California San Diego (Stephen Spector, Kim Norris, and Sharon Nichols), University of Colorado Denver Health Sciences Center (Elizabeth McFarland, Emily Barr, Carrie Chambers, and Jennifer Dunn), University of Maryland Baltimore (Douglas Watson, Nicole Messenger, and Rose Belanger), University of Medicine and Dentistry of New Jersey (Arry Dieudonne, Linda Bettica, and Susan Adubato), University of Miami (Gwendolyn Scott, Lisa Himic, and Elizabeth Willen), and Body Composition Analysis Center at Tufts University (Andrea Desilets and Justin Wheeler).
The authors’ responsibilities were as follows—DLJ, KP, RBVD, and TLM: designed the research; RBVD, JSC, EJM, WB, and MS: conducted the research (hands-on data collection); MEG, RAF, and SS: provided the materials; DLJ: analyzed the data; DLJ, KP, GKS, LAD, and TLM: provided significant advice or consultation; DLJ and TLM: wrote the manuscript; and DLJ: had primary responsibility for the final content. All authors read and approved the final manuscript. None of the authors had a conflict of interest.
Footnotes
Abbreviations used: AMP, Adolescent Master Protocol; ARV, antiretroviral medication; DXA, dual-energy X-ray absorptiometry; HAART, highly active antiretroviral therapy; HEU, HIV-exposed but uninfected; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PHACS, Pediatric HIV/AIDS Cohort Study; PI, protease inhibitor; SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue.
REFERENCES
- 1.Carter RJ, Wiener J, Abrams EJ, Farley J, Nesheim S, Palumbo P, Bulterys M. Perinatal AIDS Collaborative Transmission Study- HIV Follow-up after Perinatal Exposure (PACTS-HOPE) Group. Dyslipidemia among perinatally HIV-infected children enrolled in the PACTS-HOPE cohort, 1999-2004: a longitudinal analysis. J Acquir Immune Defic Syndr 2006;41:453–60 [DOI] [PubMed] [Google Scholar]
- 2.Jaquet D, Levine M, Ortega-Rodriguez E, Faye A, Polak M, Vilmer E, Levy-Marchal C. Clinical and metabolic presentation of the lipodystrophic syndrome in HIV-infected children. AIDS 2000;14:2123–8 [DOI] [PubMed] [Google Scholar]
- 3.Bitnun A, Sochett E, Babyn P, Holowka S, Stephens D, Read S, King SM. Serum lipids, glucose homeostasis and abdominal adipose tissue distribution in protease inhibitor-treated and naive HIV-infected children. AIDS 2003;17:1319–27 [DOI] [PubMed] [Google Scholar]
- 4.Brambilla P, Bricalli D, Sala N, Renzetti F, Manzoni P, Vanzulli A, Chiumello G, di Natale B, Vigano A. Highly active antiretroviral-treated HIV-infected children show fat distribution changes even in absence of lipodystrophy. AIDS 2001;15:2415–22 [DOI] [PubMed] [Google Scholar]
- 5.Ergun-Longmire B, Lin-Su K, Dunn AM, Chan L, Ham K, Sison C, Stavola J, Vogiatzi MG. Effects of protease inhibitors on glucose tolerance, lipid metabolism, and body composition in children and adolescents infected with human immunodeficiency virus. Endocr Pract 2006;12:514–21 [DOI] [PubMed] [Google Scholar]
- 6.Innes S, Levin L, Cotton M. Lipodystrophy syndrome in HIV-Infected children on HAART. South Afr J HIV Med 2009;10:76–80 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Krause JC, Toye MP, Fisher DJ, Stechenberg BW, Reiter EO, Allen HF. Metabolic abnormalities in human immunodeficiency virus-infected children: two-year follow-up. J Pediatr Endocrinol Metab 2009;22:345–51 [DOI] [PubMed] [Google Scholar]
- 8.Miller TL, Mawn BE, Orav EJ, Wilk D, Weinberg GA, Nicchitta J, Furuta L, Cutroni R, McIntosh K, Burchett SK, et al. The effect of protease inhibitor therapy on growth and body composition in human immunodeficiency virus type 1-infected children. Pediatrics 2001;107:E77. [DOI] [PubMed] [Google Scholar]
- 9.Mulligan K, Harris DR, Monte D. Stoszek S, Emmanuel P, Hardin DS, Kapogiannis BG, Worrell C, Meyer WA 3rd, Sleasman J, et al. Obesity and dyslipidemia in behaviorally HIV-infected young women: Adolescent Trials Network study 021. Clin Infect Dis 2010;50:106–14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sánchez Torres AM, Munoz Muniz R, Madero R, Borque C, Garcia-Miguel MJ, De Jose Gómez MI. Prevalence of fat redistribution and metabolic disorders in human immunodeficiency virus-infected children. Eur J Pediatr 2005;164:271–6 [DOI] [PubMed] [Google Scholar]
- 11.Taylor P, Worrell C, Steinberg SM, Hazra R, Jankelevich S, Wood LV, Zwerski S, Yarchoan R, Zeichner S. Natural history of lipid abnormalities and fat redistribution among human immunodeficiency virus-infected children receiving long-term, protease inhibitor-containing, highly active antiretroviral therapy regimens. Pediatrics 2004;114:e235–42 [DOI] [PubMed] [Google Scholar]
- 12.Viganò A, Mora S, Testolin C, Beccio S, Schneider L, Bricalli D, Vanzulli A, Manzoni P, Brambilla P. Increased lipodystrophy is associated with increased exposure to highly active antiretroviral therapy in HIV-infected children. J Acquir Immune Defic Syndr 2003;32:482–9 [DOI] [PubMed] [Google Scholar]
- 13.Beregszaszi M, Dollfus C, Levine M, Faye A, Deghmoun S, Bellal N, Houang M, Chevenne D, Hankard R, Bresson JL, et al. Longitudinal evaluation and risk factors of lipodystrophy and associated metabolic changes in HIV-infected children. J Acquir Immune Defic Syndr 2005;40:161–8 [DOI] [PubMed] [Google Scholar]
- 14.Bockhorst JL, Ksseiry I, Toye M, Chipkin SR, Stechenberg BW, Fisher DJ, Allen HF. Evidence of human immunodeficiency virus-associated lipodystrophy syndrome in children treated with protease inhibitors. Pediatr Infect Dis J 2003;22:463–5 [PubMed] [Google Scholar]
- 15.Fontana M, Zuin G, Plebani A, Bastoni K, Visconti G, Principi N. Body composition in HIV-infected children: relations with disease progression and survival. Am J Clin Nutr 1999;69:1282–6 [DOI] [PubMed] [Google Scholar]
- 16.Melvin AJ, Lennon S, Mohan KM, Purnell JQ. Metabolic abnormalities in HIV type 1-infected children treated and not treated with protease inhibitors. AIDS Res Hum Retroviruses 2001;17:1117–23 [DOI] [PubMed] [Google Scholar]
- 17.McGill HC, Jr, McMahan CA, Gidding SS. Preventing heart disease in the 21st century: implications of the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) study. Circulation 2008;117:1216–27 [DOI] [PubMed] [Google Scholar]
- 18.Morrison JA, Friedman LA, Gray-McGuire C. Metabolic syndrome in childhood predicts adult cardiovascular disease 25 years later: the Princeton Lipid Research Clinics Follow-up Study. Pediatrics 2007;120:340–5 [DOI] [PubMed] [Google Scholar]
- 19.Tresaco B, Moreno LA, Ruiz JR, Ortega FB, Bueno G, Gonzalez-Gross M, Warnberg J, Gutierrez A, Garcia-Fuentes M, Marcos A, et al. Truncal and abdominal fat as determinants of high triglycerides and low HDL-cholesterol in adolescents. Obesity (Silver Spring) 2009;17:1086–91 [DOI] [PubMed] [Google Scholar]
- 20.Jacobson DL, Knox TA, Spiegelman D, Skinner S, Gorbach SL, Wanke C. Prevalence, evolution and risk factors for fat atrophy and fat deposition in a cohort of HIV-infected men and women. Clin Infect Dis 2005;40:1837–45 [DOI] [PubMed] [Google Scholar]
- 21.Bachetti P, Gripshover B, Grunfeld C, Heymsfield S, McCreath H, Osmond D, Saag M, Scherzer R, Shlipak M, Tien P. Fat distribution in men with HIV infection. J Acquir Immune Defic Syndr 2005;40:121–31 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Goulding A, Taylor RW, Gold E, Lewis-Barned NJ. Regional body fat distribution in relation to pubertal stage: a dual-energy X-ray absorptiometry study of New Zealand girls and young women. Am J Clin Nutr 1996;64:546–51 [DOI] [PubMed] [Google Scholar]
- 23.Taylor RW, Grant AM, Williams SM, Goulding A. Fat distribution in men with HIV infection. [Google Scholar]
- 24.Mihalopoulos NL, Holubkov R, Young P, Dai S, Labarthe DR. Expected changes in clinical measures of adiposity during puberty. J Adolesc Health 2010;47:360–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Vink EE, van Coeverden SC, van Mil EG, Felius BA, van Leerdam FJ, Delemarre-van de Waal HA. Changes and tracking of fat mass in pubertal girls. Obesity (Silver Spring) 2010;18:1247–51 [DOI] [PubMed] [Google Scholar]
- 26.European Paediatric Lipodystrophy Group Antiretroviral therapy, fat redistribution and hyperlipidaemia in HIV-infected children in Europe. AIDS 2004;18:1443–51 [DOI] [PubMed] [Google Scholar]
- 27.Arpadi SM, Cuff PA, Horlick M, Wang J, Kotler DP. Lipodystrophy in HIV-infected children is associated with high viral load and low CD4+ -lymphocyte count and CD4+ -lymphocyte percentage at baseline and use of protease inhibitors and stavudine. J Acquir Immune Defic Syndr 2001;27:30–4 [DOI] [PubMed] [Google Scholar]
- 28.Aldrovandi GM, Lindsey JC, Jacobson DL, Zadzilka A, Sheeran E, Moye J, Borum P, Meyer WA 3rd, Hardin DS, Mulligan K, et al. Morphologic and metabolic abnormalities in vertically HIV-infected children and youth. AIDS 2009;23:661–72 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Arpadi SM, Bethel J, Horlick M, Sarr M, Bamji M, Abrams EJ, Purswani M, Engelson ES. Longitudinal changes in regional fat content in HIV-infected children and adolescents. AIDS 2009;23:1501–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Miller TL, Orav EJ, Lipshultz SE, Arheart KL, Duggan C, Weinberg GA, Bechard L, Furuta L, Nicchitta J, Gorbach SL, et al. Risk factors for cardiovascular disease in children infected with human immunodeficiency virus-1. J Pediatr 2008;153:491–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Chantry CJ, Cervia JS, Hughes MD, Alvero C, Hodge J, Borum P, Moye J, Jr; PACTG 1010 Team Predictors of growth and body composition in HIV-infected children beginning or changing antiretroviral therapy. HIV Med 2010;11:573–83 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Newell ML, Borja MC, Peckham C, European Collaborative Study Height, weight, and growth in children born to mothers with HIV-1 infection in Europe. Pediatrics 2003;111:e52–60 [DOI] [PubMed] [Google Scholar]
- 33.Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS, Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL. CDC growth charts: United States. Adv Data 2000;314:1–27 [PubMed] [Google Scholar]
- 34.US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics NHANES III reference manuals and reports. Hyattsville, MD: National Center for Health Statistics, 1996 [Google Scholar]
- 35.Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007-2008. JAMA 2010;303:242–9 [DOI] [PubMed] [Google Scholar]
- 36.Borrud LG, Flegal KM, Looker AC, Everhart JE, Harris TB, Shepherd JA. Body composition data for individuals 8 years of age and older: U.S. population, 1999-2004. Vital Health Stat 11 2010;250:1–87 [PMC free article] [PubMed] [Google Scholar]
- 37.McGrath CJ, Chung MH, Richardson BA, Benki-Nugent S, Warui D, John-Stewart GC. Younger age at HAART initiation is associated with more rapid growth reconstitution. AIDS 2011;25:345–55 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Miller TL, Easley KA, Zhang W, et al. Maternal and infant factors associated with failure to thrive in children with vertically transmitted human immunodeficiency virus-1 infection: the prospective, P2C2 human immunodeficiency virus multicenter study. Pediatrics 2001;108:1287–96 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Skelton JA, Cook SR, Auinger P, Klein JD, Barlow SE. Prevalence and trends of severe obesity among US children and adolescents. Acad Pediatr 2009;9:322–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Shear CL, Freedman DS, Burke GL, Harsha DW, Berenson GS. Body fat patterning and blood pressure in children and young adults. The Bogalusa Heart Study. Hypertension 1987;9:236–44 [DOI] [PubMed] [Google Scholar]
- 41.Freedman DS, Srinivasan SR, Harsha DW, Webber LS, Berenson GS. Relation of body fat patterning to lipid and lipoprotein concentrations in children and adolescents: the Bogalusa Heart Study. Am J Clin Nutr 1989;50:930–9 [DOI] [PubMed] [Google Scholar]
- 42.Freedman DS, Serdula MK, Srinivasan SR, Berenson GS. Relation of circumferences and skinfold thicknesses to lipid and insulin concentrations in children and adolescents: the Bogalusa Heart Study. Am J Clin Nutr 1999;69:308–17 [DOI] [PubMed] [Google Scholar]
- 43.Mokha JS, Srinivasan SR, Dasmahapatra P, Fernandez C, Chen W, Xu J, Berenson GS. Utility of waist-to-height ratio in assessing the status of central obesity and related cardiometabolic risk profile among normal weight and overweight/obese children: the Bogalusa Heart Study. BMC Pediatr 2010;10:73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Lawlor DA, Benfield L, Logue J, et al. Association between general and central adiposity in childhood, and change in these, with cardiovascular risk factors in adolescence: prospective cohort study. BMJ 2010;341:c6224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Viganò A, Aldrovandi GM, Giacomet V, Merlo M, Martelli L, Beretta S, Luraschi P, Rombola G, Mora S. Improvement in dyslipidaemia after switching stavudine to tenofovir and replacing protease inhibitors with efavirenz in HIV-infected children. Antivir Ther 2005;10:917–24 [PubMed] [Google Scholar]
- 46.Gerschenson M, Shiramizu B, LiButti DE, Shikuma CM. Mitochondrial DNA levels of peripheral blood mononuclear cells and subcutaneous adipose tissue from thigh, fat and abdomen of HIV-1 seropositive and negative individuals. Antivir Ther 2005;10(suppl 2):M83–9 [PubMed] [Google Scholar]
- 47.Dagan T, Sable C, Bray J, Gerschenson M. Mitochondrial dysfunction and antiretroviral nucleoside analog toxicities: what is the evidence? Mitochondrion 2002;1:397–412 [DOI] [PubMed] [Google Scholar]
- 48.John M, McKinnon EJ, James IR, Nolan DA, Herrmann SE, Moore CB, White AJ, Mallal SA. 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 [DOI] [PubMed] [Google Scholar]
- 49.Carr A, Workman C, Smith DE, et al. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy: a randomized trial. JAMA 2002;288:207–15 [DOI] [PubMed] [Google Scholar]
- 50.Moyle GJ, Baldwin C, Langroudi B, Mandalia S, Gazzard BG. A 48-week, randomized, open-label comparison of three abacavir-based substitution approaches in the management of dyslipidemia and peripheral lipoatrophy. J Acquir Immune Defic Syndr 2003;33:22–8 [DOI] [PubMed] [Google Scholar]
- 51.Viganò A, Brambilla P, Cafarelli L, Giacomet V, Borgonovo S, Zamproni I, Zuccotti G, Mora S. Normalization of fat accrual in lipoatrophic, HIV-infected children switched from stavudine to tenofovir and from protease inhibitor to efavirenz. Antivir Ther 2007;12:297–302 [PubMed] [Google Scholar]
- 52.Caron M, Auclair M, Sterlingot H, Kornprobst M, Capeau J. Some HIV protease inhibitors alter lamin A/C maturation and stability, SREBP-1 nuclear localization and adipocyte differentiation. AIDS 2003;17:2437–44 [DOI] [PubMed] [Google Scholar]
- 53.McComsey G, Bhumbra N, Ma JF, Rathore M, Alvarez A, First Pediatric Switch Study Impact of protease inhibitor substitution with efavirenz in HIV-infected children: results of the First Pediatric Switch Study. Pediatrics 2003;111:e275–81 [DOI] [PubMed] [Google Scholar]
- 54.Rodríguez de la Concepcion ML, Yubero P, Domingo JC, Iglesias R, Domingo P, Villarroya F, Giralt M. Reverse transcriptase inhibitors alter uncoupling protein-1 and mitochondrial biogenesis in brown adipocytes. Antivir Ther 2005;10:515–26 [PubMed] [Google Scholar]
- 55.Mira JA, Lozano F, Santos J, Ramayo E, Terron A, Palacios R, Leon EM, Marquez M, Macias J, Fernandez-Palacin A, et al. Gynaecomastia in HIV-infected men on highly active antiretroviral therapy: association with efavirenz and didanosine treatment. Antivir Ther 2004;9:511–7 [PubMed] [Google Scholar]
- 56.Siegel MJ, Hildebolt CF, Bae KT, Hong C, White NH. Total and intraabdominal fat distribution in preadolescents and adolescents: measurement with MR imaging. Radiology 2007;242:846–526 [DOI] [PubMed] [Google Scholar]