See “Conundrum of Growth and Childhood HIV Infection” by Chantry and Loomba-Albrecht on page 424.
Although the human immunodeficiency virus (HIV) was known historically as a wasting disease (1), the advent of highly active antiretroviral therapy (HAART) has resulted in improved nutritional status among many HIV-infected people, with a decreased overall prevalence of wasting in this population (2). HIV has become a chronic disease, instead of one of progressive deterioration.
Obesity and lipodystrophy, the redistribution of body fat to the trunk and wasting of fat at the extremities, are some of the most obvious phenotypic cardiovascular risk factors for both HIV-infected adults and youth (3–5). Obesity and abdominal adiposity are serious health concerns for non–HIV-infected youth worldwide with an increase in their prevalence during the past several decades (6–10). HIV-infected youth may be particularly susceptible to the adverse sequelae of obesity and abdominal adiposity (11–13) because of chronic inflammation and exposure to medications that can also affect cardiovascular disease risk. In turn, both obesity and abdominal adiposity are risk factors for diabetes, dyslipidemia, and cardiovascular disease in adults (14) with evidence that these associations hold true in youth (15–17).
A total of 69% of vertical transmission (ie, mother to child transmission) of HIV occurs in African Americans, a race that is particularly susceptible to obesity and adiposity (7,18). Thus, to ascertain the prevalence of obesity and adiposity in HIV, it is prudent to compare HIV-infected youth not only with national standards but also demographically similar populations. The extent to which HIV-infected youth follow national and international trends for obesity and abdominal adiposity has not been well reported.
The objectives of our study are to compare body mass index (BMI), the prevalence of overweight and obesity, and waist circumference (WC) of a cohort of youth with perinatally acquired HIV with a local cohort of youth without HIV and with the general US population based on the National Health and Nutrition Examination Survey (NHANES) data on BMI and WC for the years 2007 to 2010. In a subset analysis, we compared HIV-infected African Americans with an African American subset of NHANES. We also determined whether clinical factors such as race; HIV Centers for Disease Control and Prevention (CDC) stage; viral load; CD4 count; and HAART, protease inhibitor (PI), nucleoside reverse transcriptase inhibitor (NRTI), and non-nucleoside reverse transcriptase inhibitor (NNRTI) duration are associated with BMI or WC in our HIV cohort.
HIV-infected subjects followed in the Pediatric Special Immunology Program at the University of Miami were enrolled in a National Heart, Lung, and Blood Institute–sponsored longitudinal cohort study that evaluated cardiovascular disease risk in perinatally HIV-infected youth between August 2004 and March 2011. HIV infection was confirmed by repeated positive serum enzyme-linked immunosorbent assays and Western blot assays, repeated positive HIV RNA or DNA polymerase chain reaction assays, or HIV culture. All of these laboratory tests were performed in outpatient laboratories that were determined based on clinical care.
A total of 165 HIV-infected children and adolescents aged 10 to 19 years were seen in the Special Immunology Clinic in 2010. For this cross-sectional analysis, the most recent visit between 2007 and 2010 with complete anthropometric data (HIV-infected and HIV-uninfected youth) was used and constituted the dataset of 134 individuals. HIV-infected youth who did not acquire HIV from their mother (eg, horizontally infected) were excluded, as were those who were wheelchair bound because anthropometric data could not be accurately obtained on these individuals.
The HIV-uninfected comparison group was a convenience group of 75 uninfected youth. These youth were identified from the family members of the HIV-infected group or from an urban general pediatric outpatient program at the University of Miami that cares for youth of similar socioeconomic backgrounds. None of these individuals were known to be HIV infected. Uninfected individuals were excluded if they had any chronic illness or an acute infectious process.
The NHANES database (19) was used as an additional comparison group of children and adolescents. NHANES is a series of studies to assess the health and nutritional status of children and adults in the United States. Its data, including anthropometric measurements, are reported on a biannual basis. The NHANES uses probability sampling and weighting methodology (20). Ignoring its complex design can lead to biased population estimates and overstated significance levels. Thus, the present study followed NHANES guidelines (20) to account for the design features, including stratification, cluster sampling, and weighting. Demographic and anthropometric data were extracted from the NHANES database for the years 2007 to 2008 and 2009 to 2010 for all individuals aged 10 to 19 for a total of 3216 individuals as representative of a national sample of children and adolescents of this age group. We also identified a subset of 771 African American NHANES participants to compare anthropometrics with an African American HIV cohort.
The Human Subjects Research Office at the University of Miami approved the research protocol (protocol number 20030814), and written informed consent from the parent or guardian and assent from the patient (when appropriate) were obtained.
At the visit, the following clinical data were collected for both the HIV-infected group and the HIV-uninfected group: demographic information (age, sex, race), weight, height, and WC. For the HIV-infected group only Tanner stage; CDC HIV stage (21); CD4 T-lymphocyte count; HIV viral load; and HAART, PI, NRTI, and NNRTI present use and lifetime duration were obtained. The definition of HAART was guided by the DHHS/Kaiser Panel (DHHS/Kaiser 2004) guidelines and defined as ≥2 NRTIs in combination with at least 1 PI or 1 NNRTI, 1 NRTI in combination with at least 1 PI and at least 1 NNRTI, a regimen containing ritonavir and saquinavir in combination with 1 NRTI and no NNRTIs, or an abacavir- or tenofovir-containing regimen of ≥3 NRTIs in the absence of both PIs and NNRTIs, except for the 3-NRTI regimens consisting of: abacavir + tenofovir + lamivudine or didanosine + tenofovir + lamivudine (22). If CD4 count or viral load was not collected at the most recent visit, a value within 90 days of the visit was used.
Weight and standing height were measured by standardized procedures (23). BMI was calculated as weight (kg)/(height)2 (m2) (24). WC was measured using a nonstretchable plastic tape measure at the navel at the end of gentle exhalation (23). Weight, height, and BMI were standardized for age and sex using CDC-derived z scores (24). Each child was classified as underweight (BMI < 5%), normal weight (5% to <85%), overweight (85% to <95%), or obese (≥95%) (25).
We used linear mixed models with random family cluster–specific intercepts and heterogeneous residual variances for the HIV-infected and HIV-uninfected groups to determine whether any differences in height/height z score, weight/weight z score, BMI/BMI z score, or WC existed between our HIV-infected group and our HIV-uninfected group. Height, weight, BMI, and WC of the NHANES sample and NHANES African American subsample were also compared with our HIV-infected population using Welch t tests.
In a subanalysis using only HIV-infected children, race; CD4 count; viral load; HIV CDC stage; and HAART, PI, NRTI, and NNRTI duration were modeled as predictors of BMI z score in linear mixed models with random family cluster–specific intercepts. In a separate random intercept model, race; sex; age; CD4 count; viral load; HIV CDC stage; and HAART, PI, NRTI, and NNRTI duration were modeled as predictors of WC. Tanner stage was included in a separate analysis as a predictor of BMI z score and WC because 27% of Tanner stage data were missing from the sample.
Although primarily for descriptive purposes, a categorical analysis of BMI and HIV status was also performed. This was done to provide additional information regarding possible clinical interpretations. BMI categories were compared between the HIV-infected and HIV-uninfected groups using the Cochran-Armitage test. Categories were compared between HIV and NHANES using the χ2 test. Statistical analyses were performed with the SAS statistical software program, version 9.3 (SAS Institute, Cary, NC).
A total of 134 HIV-infected (including 103 African Americans), 75 HIV-uninfected, and 3216 NHANES (including 771 NHANES African Americans) children and adolescents were included in the analysis. HIV-specific characteristics of our affected population are shown in Table 1. A total of 37% of the group met criteria for the acquired immunodeficiency syndrome (CDC class C). The median CD4 count was 535 cells per cubic millimeter, and 20% of the HIV-infected population had a CD4 count <200 cells per cubic millimeter. The median viral load was 760 copies per milliliter, with 42% having an undetectable viral load (<400 copies per milliliter). A total of 90% of children were receiving a HAART regimen with a median duration of 110 months. A total of 93% of the youth on HAART had a PI as part of their HAART regimen.
Table 2 compares the demographic and anthropometric characteristics of the HIV-infected and HIV-uninfected groups. The HIV-infected group was older than the HIV-uninfected group (16.5 years vs 14.2 years, P < 0.0001) and was more likely to be African American (77% vs 52%, P = 0.002). After adjusting for race and accounting for biological relatedness, the HIV-infected group was shorter than the HIV-uninfected group (height z score −0.41 vs 0.10, P = 0.003) and, although not statistically significant, the BMI z score was lower in the HIV-infected group also (0.69 vs 0.87, P = 0.36). WC—adjusted for age, sex, and race, and accounted for biological relatedness—was not statistically significantly different between the HIV-infected and HIV-uninfected groups (81.9 cm vs 80.5 cm, P = 0.66).
Table 3 compares the demographic and anthropometric characteristics of the HIV-infected group and the NHANES sample. All anthropometric data were unadjusted except for z score data, which were intrinsically adjusted for age and sex, and WC, which was adjusted for age and sex. The HIV-infected group was older than the NHANES group (16.5 years vs 15.0 years, P < 0.001). Compared with the NHANES group, the HIV-infected group was shorter (height z score −0.48 vs 0.19, P < 0.001) and lighter (weight z score 0.27 vs 0.59, P = 0.01). The HIV-infected group had nonstatistically significant differences in BMI z scores (P = 0.84). WC was not statistically significantly different (P = 0.63).
Table 4 compares the demographic and anthropometric characteristics of the African American HIV-infected group and the African American NHANES subgroup. Similar to the racially diverse comparison, the African American HIV-infected group was older (16.4 years vs 15.0 years, P < 0.001), shorter (height z score −0.44 vs 0.24, P < 0.001), and lighter (weight z score 0.26 vs 0.82, P < 0.001) than the NHANES African American group. The HIV-infected group had a statistically significant difference in BMI z score (P = 0.04). WC was not statistically significantly different (P = 0.96, adjusted for age and sex).
Figure 1 shows the frequency of individuals in each sample categorized as underweight, normal, overweight, and obese. For the HIV-infected group, 13.4% were overweight compared with 10.7% in the HIV-uninfected group and 15.6% in the NHANES sample. A total of 22% of the HIV-infected group was obese compared with 32.0% in the HIV-uninfected group and 20.1% in the NHANES sample. There were no statistically significant differences in BMI categories between the HIV-infected and HIV-uninfected groups, the NHANES sample, and the NHANES African American sample (P > 0.08). There was a trend that the HIV-infected group had lower rates of overweight and obesity when compared with the NHANES African American sample (13.4% vs 16% overweight and 22% vs 26% obese, respectively).
Among the HIV-infected group alone, we examined factors associated with a higher BMI. In a random intercept linear model analysis, there were no statistically significant associations (all P ≥ 0.14) between race, HIV CDC stage, CD4 count, viral load, and duration of HAART and BMI z score. In a separate random intercept linear model analysis with WC as the outcome and age and sex added in the model as predictors, age was a statistically significant predictor (P = 0.046) of WC, but the other covariates were not (all P ≥ 0.19). A similar model using individual antiretroviral drug classes (ie, PI, NRTI, and NNRTI) instead of HAART in the model showed no associations with BMI z score (P > 0.15). Again, age was a statistically significant predictor (P = 0.03) of WC, but the other covariates were not (all P ≥ 0.20). When Tanner stage was included as a predictor of both BMI and WC, there were no statistically significant associations.
Our study shows differences in some anthropometric measures (heights and weights) among HIV-infected youth when compared with a local, uninfected group or with a multiethnic national sample of youth of similar ages. BMI z scores and WCs are similar, suggesting comparable rates of “adiposity” between these groups. Yet when these measures are compared between an exclusively African American subset of HIV-infected and African American NHANES youth, there are even greater differences in height and weight and significantly lower BMI z scores in HIV-infected youth. When BMI is categorized, we find that approximately 35% of the HIV-infected youth meet criteria for being overweight or obese. These percentages are similar to our contemporary group of uninfected youth and to national trends (7,8), yet lower than the national African American subsample (42%). We found that HIV-specific factors were not associated with BMI or WC.
Advances in antiretroviral therapy have led to improvements in growth of HIV-infected youth (2,26–30), yet our data show that these youth lag behind their racially matched peers in some anthropometric indicators. In essence, HIV-infected youth are “less obese” than their noninfected counterparts but plot along “normal” anthropometric measures. Some studies find lower BMIs in HIV-infected children compared with demographically similar uninfected children exposed to HIV in utero (5,26). Rates of obesity in children and adolescents differ by race, with African Americans reported to have higher rates of obesity than non-Hispanic whites (7). Our analyses show these differences because the BMI of the African American subsample of NHANES was higher than the BMI of the entire NHANES sample (BMI z score 0.49 vs 0.76). Therefore, despite improvements in growth with antiretroviral therapy and better viral control, HIV-infected youth continue to demonstrate lower anthropometric indicators when compared with a demographically similar group.
Approximately 33.2% of US HIV-uninfected children and adolescents ages 6 to 19 years are either overweight or obese (8). The prevalence of overweight and obesity in our HIV-infected population is similar to national trends, with 35.8% being overweight or obese. For the African American NHANES sample, 42.5% are overweight or obese. Although the difference in BMI categories between HIV-infected and NHANES African American youth does not reach statistical significance, our HIV-infected group has a lower prevalence of overweight and obesity compared with a racially similar national sample of NHANES African Americans (P = 0.08). These categorical findings further support the need to compare HIV-infected youth with a group with a similar racial composition.
The differences in BMI between HIV-infected youth and uninfected African Americans could be a result of persistent inflammation associated with chronic viral infection (31,32), because other disorders associated with chronic inflammation can affect growth in children (33–35). High viral load and antiretroviral medications (31,32) are associated with chronic inflammation as manifested by higher levels of biomarkers of inflammation in children with HIV infection compared with controls (31,36), which can, in turn, affect basal metabolic rate (37). Our study did not find an association between measures of HIV disease severity (CDC stage, viral load, CD4 counts) and anthropometric outcomes. There are data that also show persistent inflammation in the context of good HIV viral control (38). Studies also demonstrate that the anthropometric effects of HIV are apparent even in utero; here, infants prenatally exposed to HIV but were uninfected have lower birth lengths and weights compared with unexposed infants (39,40).
Although our data support a lower BMI in HIV-infected youth compared with their uninfected, racially alike peers, there is a high prevalence of obesity in these individuals. HIV-infected youth are invariably exposed to many of the same risks for overweight and obesity that are relevant to the general population. Low level of cardiorespiratory fitness (41) and dietary patterns that are more obesogenic are reported in HIV-infected youth (42). These lifestyle factors can contribute to high rates of overweight and obesity in HIV-infected children. Obesity (43) and HIV infection (4,11) are known independent risk factors for cardiovascular disease. Thus, although these youth may be “less obese” than their peers, the high prevalence of obesity that exists in these individuals and higher risk of cardiovascular disease caused by the virus make HIV infection in youth a public health concern.
There are limitations to our study. We did not have a perfectly matched control group. Our HIV population was drawn from a convenience sample of HIV-infected youth seen at our institution. Our comparison group consisted of youth who were either family members of our HIV-infected cohort or recruited from a nearby urban pediatric practice with an ethnically and socioeconomically similar population. This comparison population was younger than our sample population and had a smaller percentage of African Americans (52% vs 77% in HIV infected), and many were related to our HIV-infected group. Although we controlled for these factors in our analyses, this mismatch could affect our results. It is possible that a greater sample size in our local HIV-uninfected group could have resulted in better power to determine differences between the groups. In addition, 1 covariate that we were unable to control for in our studies is pubertal stage because these data were lacking in our uninfected comparison group and NHANES sample. Differences in Tanner stage between groups could influence BMI and WC and affect the results.
In conclusion, we found that the BMI and WC of HIV-infected youth approximate those of a nationally representative sample. There may be more subtle differences in BMI when HIV-infected youth are compared with demographically similar populations. We found that HIV-specific factors such as HIV CDC stage; viral load; CD4 count; and HAART, PI, NRTI, and NNRTI duration were not associated with BMI or WC in HIV-infected youth. Future studies should try to define factors that are associated with residual anthropometric differences in demographically matched HIV-infected youth and to determine effective interventions that can target those youth who are overweight and obese and who parallel national trends in obesity and carry additional risks for cardiometabolic disease.
1. Macallan D. Wasting in HIV infection and AIDS. J Nutr
2. Nachman SA, Lindsey JC, Moye J, et al. Growth of human immunodeficiency virus
-infected children receiving highly active antiretroviral therapy. Pediatr Infect Dis J
3. Lichtenstein KA, Ward DJ, Moorman AC, et al. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. AIDS
4. Miller TL, Orav EJ, Lipshultz SE, et al. Risk factors for cardiovascular disease in children infected with human immunodeficiency virus
-1. J Pediatr
5. Jacobson DL, Patel K, Siberry GK, et al. Body fat distribution in perinatally HIV-infected and HIV-exposed but uninfected children in the era of highly active antiretroviral therapy: outcomes from the Pediatric HIV/AIDS Cohort Study. Am J Clin Nutr
6. Daniels SR, Arnett DK, Eckel RH, et al. Overweight in children and adolescents: pathophysiology, consequences, prevention, and treatment. Circulation
7. Ogden C, Carroll M. Prevalence of obesity
among children and adolescents: United States, trends 1963–1965 through 2007–2008. National Center for Health Statistics; June 2010. www.cdc.gov/nchs/products/hestats.htm
. Accessed January 19, 2014.
8. World Health Organization. Global strategy on diet, physical activity and health: childhood overweight and obesity
. Accessed January 19, 2014.
9. Ogden CL, Kit BK, Flegal KM. Prevalence of obesity
and trends in body mass index
among US children and adolescents, 1999–2010. JAMA
10. Messiah SE, Arheart KL, Lipshultz SE, et al. Ethnic group differences in waist circumference
percentiles among U.S. children and adolescents: estimates from the 1999–2008 National Health and Nutrition Examination Surveys. Metab Syndr Relat Disord
11. Islam FM, Wu J, Jansson J, et al. Relative risk of cardiovascular disease among people living with HIV: a systematic review and meta-analysis. HIV Med
12. Duprez DA, Neuhaus J, Kuller LH, et al. Inflammation, coagulation and cardiovascular disease in HIV-infected individuals. PLoS One
13. Kelishadi R, Ardalan G, Gheiratmand R, et al. Paediatric metabolic syndrome and associated anthropometric indices: the CASPIAN Study. Acta Paediatr
14. Bray GA. Complications of obesity
. Ann Intern Med
15. Freedman DS, Katzmarzyk PT, Dietz WH, et al. Relation of body mass index
and skinfold thicknesses to cardiovascular disease risk factors in children: the Bogalusa Heart Study. Am J Clin Nutr
16. Maximova K, Kuhle S, Davidson Z, et al. Cardiovascular risk factor profiles of normal and overweight children and adolescents: insights from the Canadian Health Measures Survey. Can J Cardiol
17. Friedemann C, Heneghan C, Mahtani K, et al. Cardiovascular disease risk in healthy children and its association with body mass index
: systematic review and meta-analysis. BMJ
18. Centers for Disease Control and Prevention. HIV among African Americans. www.cdc.gov/hiv/pdf/risk_HIV_AAA.pdf
. Published February 2013. Accessed August 19, 2013.
19. Centers for Disease Control and Prevention. National Center for Health Statistics. National Health and Nutrition Examination Survey data, 2007–2011. http://www.cdc.gov/nchs/nhanes.htm
. Accessed January 19, 2014.
20. National Center for Health Statistics (NCHS). Analytic and reporting guidelines: the National Health and Nutrition Examination Survey (NHANES). www.cdc.gov/nchs/data/nhanes/nhanes_03_04/nhanes_analytic_guidelines_dec_2005.pdf
. Accessed January 19, 2014.
21. HAART: DHHS/Henry J. Kaiser Family Foundation Panel on Clinical Practices for the Treatment of HIV Infection. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. http://aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL10292004002.pdf
. Published October 2004. Accessed January 19, 2014.
22. Castro KG, Ward JW, Slutsker L, et al. Centers for Disease Control. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. http://www.cdc.gov/mmwr/preview/mmwrhtml/00018871.htm
. Published December 18, 1992. Accessed July 27, 2013.
23. National Health and Nutrition Examination Survey III Body Measurements (Anthropometry). http://www.cdc.gov/nchs/data/nhanes/nhanes3/cdrom/nchs/manuals/anthro.pdf
. Published October 1988. Accessed February 1, 2012.
24. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, et al. CDC growth charts: United States. Adv Data
25. Must A, Dallal GE, Dietz WH. Reference data for obesity
: 85th and 95th percentiles of body mass index
(wt/ht2) and triceps skinfold thickness. Am J Clin Nutr
26. Chantry CJ, Cervia JS, Hughes MD, et al. Predictors of growth and body composition in HIV-infected children beginning or changing antiretroviral therapy. HIV Med
27. Verweel G, van Rossum AM, Hartwig NG, et al. Treatment with highly active antiretroviral therapy in human immunodeficiency virus
type 1-infected children is associated with a sustained effect on growth. Pediatrics
28. Miller TL, Mawn BE, Orav EJ, et al. The effect of protease inhibitor therapy on growth and body composition in human immunodeficiency virus
type 1-infected children. Pediatrics
29. Kruzich LA, Marquis GS, Wilson CM, et al. HIV-infected US youth
are at high risk of obesity
and poor diet quality: a challenge for improving short- and long-term health outcomes. J Am Diet Assoc
30. McGrath CJ, Chung MH, Richardson BA, et al. Younger age at HAART initiation is associated with more rapid growth reconstitution. AIDS
31. Miller TL, Borkowsky W, DiMeglio LA, et al. Metabolic abnormalities and viral replication are associated with biomarkers of vascular dysfunction in HIV-infected children. HIV Med
32. Brown TT, Tassiopoulos K, Bosch RJ, et al. Association between systemic inflammation and incident diabetes in HIV-infected patients after initiation of antiretroviral therapy. Diabetes Care
33. Ahmed SF, Farquharson C, McGrogan P, et al. Pathophysiology and management of abnormal growth in children with chronic inflammatory bowel disease. World Rev Nutr Diet
34. Scaparrotta A, Di Pillo S, Attanasi M, et al. Growth failure in children with cystic fibrosis. J Pediatr Endocrinol Metab
35. Gerner P, Hörning A, Kathemann S, et al. Growth abnormalities in children with chronic hepatitis B or C. Adv Virol
36. Miller TL, Somarriba G, Orav EJ, et al. Biomarkers of vascular dysfunction in children infected with human immunodeficiency virus
-1. J Acquir Immune Defic Syndr
37. Watkins RD, Somarriba G, Orav EJ, et al. Vascular inflammation and body fat are associated with resting energy expenditure in HIV-infected children [abstract]. Presented at: 4th International Workshop on HIV Pediatrics; July 20–21; Washington, DC.
38. Pedersen KK, Pedersen M, Gaardbo JC, et al. Persisting inflammation and chronic immune activation but intact cognitive function in HIV-infected patients after long-term treatment with combination antiretroviral therapy. J Acquir Immune Defic Syndr
39. Neri D, Somarriba GA, Schaefer NN, et al. Growth and body composition of uninfected children exposed to human immunodeficiency virus
: comparison with a contemporary cohort and United States national standards. J Pediatr
40. Ibieta MF, Cano JM, Amador JT, et al. Growth of uninfected infants exposed to antiretrovirals born to HIV-infected woman. An Pediatr (Barc)
41. Somarriba G, Lopez-Mitnik G, Ludwig DA, et al. Physical fitness in children infected with the human immunodeficiency virus
: associations with highly active antiretroviral therapy. AIDS Res Hum Retroviruses
42. Sharma TS, Kinnamon DD, Duggan C, et al. Changes in macronutrient intake among HIV-infected children between 1995 and 2004. Am J Clin Nutr
43. Hubert HB, Feinleib M, McNamara PM, et al. Obesity
as an independent risk factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham Heart Study. Circulation