Growth and body composition of adolescents and young adults exposed to HIV and antiretroviral therapy in utero: a systematic review and meta-analysis

doi:10.21203/rs.3.rs-3411427/v1

Introduction: Antiretroviral therapy (ART) is a highly effective measure to prevent mother-to-child transmission of the human immunodeficiency virus (HIV) and to decline mortality allowing children to reach adolescence and adulthood. However, studies evaluating in utero exposure to HIV/ART and adolescent growth are scarce. Thus, this study aims to summarize the evidence of observational studies about the growth and body composition of adolescents and young adults exposed to HIV and ART in utero.

Methods: We conducted a systematic review using PubMed/MEDLINE, CENTRAL, Lilacs, EMBASE, CINAHL, and Web of Science databases. We included cohort studies that evaluated adolescents and young adults who were exposed to HIV and ART in utero. A meta-analysis was conducted to estimate the prevalence of low height-for-age (HAZ) and low BMI-for-age (BAZ). Also, meta-analyses for the mean values of HAZ and BAZ in the population after the follow-up period and for the mean change of HAZ and BAZ after the follow-up were conducted.

Results: In total, twelve studies were included in this systematic review, 11 of those evaluated adolescents with perinatally acquired HIV. We estimated a prevalence of low HAZ of 26% (P = 0.26; 95%CI 95% CI 0.23–0.29) and low BAZ of 7% (P = 0.07; 95%CI95% CI 0.05–0.09). In this population, the mean value of HAZ was − 1.58 (95%CI95% CI -1.90; -1.27), and that for the BAZ was − 0.44 (95%CI95% CI -0.67; -0.21). We also identified an increase of 0.55 (MD: 0.55, 95% CI: 0.07;1.03) and 0.12 (MD: 0.12, 95% CI: -0.75; 0.99) in the mean values of the HAZ and BAZ, respectively, after the follow-up period.

Discussion: Adolescents exposed to HIV/ART in utero continue to fail to thrive. However, BAZ found mean values within the normal limits recommended by the WHO, demonstrating that although delayed weight gain is common among adolescents with prenatally acquired HIV, BMI is less affected.

Conclusions: Adolescents exposed to HIV/ART in utero continue to fail to thrive. However, the limited number of studies and their limitations have restricted the ability to draw consistent conclusions about the effect and magnitude of the exposure on anthropometric outcomes. Further studies are needed to better describe these relationships.

PROSPERO Number: CRD42022372837

HIV maternal

growth

body composition

nutrition

systematic review

children development

Highly effective measures to prevent mother-to-child transmission of the human immunodeficiency virus have been available for several years, leading to impressive declines in vertical HIV transmission, which can be as high as 45% in the absence of actions. It is estimated that coverage of antiretroviral medicines provided to pregnant women living with HIV to prevent transmission to their offspring rose to 85% [1]. However, vertical transmission is still occurring, and rates can range from 4–21% in the 22 global priority countries [2].

Due to the introduction of antiretroviral therapy (ART), mortality has declined, allowing children with perinatally acquired HIV to reach adolescence and even adulthood [3]. However, information on the health status of this adolescent population is scarce, with some studies evidencing an increased risk of all-cause mortality and cancer compared to their uninfected peers, with the excess of cancer deaths driven by lymphomas [4]. Additionally, little is known about the growth patterns of perinatally infected adolescents, with most of the studies describing children with a median age of 10 years [3, 5].

The use of antiretroviral therapy (ART) during pregnancy contributes substantially to reducing the rates of mother-to-child HIV transmission; consequently, 15.4 million HIV-exposed uninfected (HEU) children are born in the world annually [6, 7, 8]. Evidence of the effects of ART on offspring throughout the life course is scarce and conflicting. ART during pregnancy has been associated with an increased risk of birth outcomes such as preterm births and low birth weight [9]. Children who were HEU had higher mortality, morbidity and stunting and modest impairments in early child development compared to children who were HIV-unexposed [10].

Studies evaluating in utero exposure to HIV and antiretroviral therapy and adolescent growth are scarce, vary by age group and HIV status and have presented mixed results [11, 12, 13]. Thus, this study aims to summarize and critically appraise the evidence of observational studies about the growth and body composition of various stages of life (adolescents and young adults) exposed to HIV and antiretroviral therapy in utero.

This systematic review and meta-analysis was based on recommendations from the Cochrane Guidelines for Systematic Reviews of Interventions [14] and was written according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [15]. The review protocol was registered at PROSPERO (CRD42022372837).

Search strategy

To identify observational studies evaluating the effects of exposure to HIV and antiretroviral therapy in utero on the growth and body composition of adolescents and young adults, we searched five independent databases to perform a sensitive literature search: MEDLINE (by PubMed), Embase, Central (by Cochrane Library), Latin American and Caribbean Health Science Information (LILACS) and Web of Science.

There was no language, date, document type, publication status or geographic restriction for inclusion of records. The search was conducted in 24th November 2022. Descriptors were identified in Medical Subject Headings (MeSH), Descritores em Ciências da Saúde (Decs) and Embase Subject Headings (Emtree). We used the following terms to search: “HIV”; “HIV Infections”; “HIV-1”; “HIV-2”; “Anti-Retroviral Agents”; “Infants”; “Child”; “Adolescents”; “Young Adult”; “Growth and Development”; “Growth”; “Body Size”; “Body composition”; “Body Fat Disturbance”; “Obesity”; “Overweight”. The search strategy was adapted based on descriptors in each database and is presented in the Supplementary material [Supporting Information].

Outcomes

The primary outcome was growth evaluated by height-for-age, weight-for-height/length and weight-for-age. The secondary outcome was the occurrence of underweight and obesity evaluated by BMI, % body fat, % lean mass, and waist circumference.

Eligibility criteria

We included cohort studies that evaluated adolescents and young adults who were exposed to HIV during their uterine life. Reviews, in vitro studies, experimental studies, interventional studies, and editorials were excluded.

Study selection and data extraction

Electronic search results from defined databases were uploaded to the Rayyan Qatar Computing Research Institute. Study selection and data extraction were independently performed by two investigators. A third reviewer resolved any disagreements. Authors initially screened titles and abstracts. Subsequently, they assessed each study to determine whether it met the inclusion criteria.

We extracted data on study design (methods, location, setting, inclusion/exclusion criteria, duration and number of participants in each group), participant characteristics (sample size, common population demographics and baseline characteristics), outcome information (association measures, mean, standard deviation, median, interquartile interval, p value, confidence interval), methods of outcome data collection (weight, height and waist circumference methods), statistical analysis, study limitations, and main conclusions.

Quality assessment

Two investigators independently assessed the risk of bias in the selected studies according to the Joanna Briggs Institute Critical Appraisal checklist for cohort studies. The included studies were classified as having “high risk of bias” when the study's “yes” score was between 0% and 49%; “moderate risk of bias” when the study was scored “yes” between 50% and 69%; and “low risk of bias” when the study 'yes' score was ≥ 70% [16]. Judging was also performed by two reviewers independently, and disagreements were resolved by a third reviewer. The risk of bias results were presented descriptively and tabularly.

Statistical analysis

For the data synthesis, we adopted a narrative review for the qualitative data. For the combinable studies, we undertook three different types of meta-analyses. One used the random effects model conducted with the Stata metaprop command to estimate the prevalence of low height-for-age (HAZ) and low BMI-for-age (BAZ). It allows you to compute 95% confidence intervals using the statistical score and the exact binomial method and incorporates the Freeman-Tukey arcsine double proportions transformation. This method also allows us to model the intra study variability using the binomial distribution. That is, it makes the data distribution normal and stabilizes the variances [17]. The inverse function of the double arcsine transformation was also derived in the literature to recover the original proportion scale after data aggregation [17], maintaining the interpretability of the final result. Thus, the summary prevalence of low HAZ and low BAZ was generated, as well as its respective 95% confidence interval.

Meta-analyses for the mean of HAZ and BAZ in the population after the follow-up period and for the mean change of HAZ and BAZ after the follow-up (by subtracting the final mean from the baseline mean) were calculated converting the median, first and third quartiles, and the number of subjects into mean (by the Luo et al., 2016 [18] transformation-based approach) and standard deviation (adopting the Wan et al., 2014 estimation [19]). Then, we estimated random effects pooled means and differences between means and their respective 95% confidence intervals (CIs) for the HAZ and BAZ. For the other nutritional measurements (fat and lean mass, weight-for-age, and waist circumference), the meta-analyses could not be estimated, considering the lack of combinable studies.

The heterogeneity and consistency of the studies were tested by the Cochran Q-test and quantified by the I² test [14, 20]. The heterogeneity between the studies varies from 0 to 100%, and values close to zero suggest that there is no heterogeneity, considering that the dispersion can be attributed to random error; close to 25% indicates low heterogeneity; >50% indicates moderate heterogeneity; and > 75% indicates high heterogeneity [14, 20].

Considering the small number of studies included in the meta-analysis, we could not conduct meta-regression and subgroup analysis to investigate the causes of the heterogeneity. For the same reason, publication bias was not evaluated by visual inspection of the funnel plot and calculating the Egger test [20]. Analyses were performed in Stata for Mac, version 16.0.

Search results

Our search strategy retrieved 11,017 documents. After excluding 1,744 duplicates, 9,273 titles and abstracts were analyzed. Full-text articles for the remaining 87 records were retrieved, of which 68 were excluded (12 were not accessed for retraction or because it has not been located, without a DOI number). Of the 19 studies previously included for extraction, 7 were excluded in the textual reading due to part of the sample having HIV acquired through transfusion (n=2); sample was not exposed to HIV (n=1); case‒control (n=2); and no information on how they acquired HIV (n=2). In total, twelve studies were selected to compose this systematic review: Anderson, Muloiwa and Davies [21], 2019; Bakeera-Kitaka et al., 2015 [22]; Bellavia et al., 2017 [23]; Boettiger et al., 2016 [24]; Bunupuradah et al., 2016 [25]; Bounora et al., 2008 [26]; Crichton et al., 2019 [27]; Desai, Mullen and Mathur, 2008 [28]; Fabiano et al., 2013 [29]; Filteau et al., 2022 [30]; Mwambenu et al., 2022 [31]; Slogrove, Judd and Leroy, 2016 [32]. The PRISMA flowchart of the screening process is illustrated in Figure 1.

Ten studies included only adolescents with prenatally acquired HIV [21, 22, 24, 25, 26, 27, 28, 29, 31, 32]; one study evaluated positive and negative HIV adolescents [23]; and one study included only negative HIV adolescents exposed to HIV in utero [30].

Study and patient characteristics

The main characteristics of the included studies are summarized in Table 1. All included studies had cohort designs published between 2008 and 2022. Four studies were conducted in a multicenter fashion involving two or more countries [24, 25, 27, 32]; two in South Africa [21, 31]; two in the United States of America [23, 28]; one in Brazil [26]; one in Uganda [22]; one in Zambia [30]; and one in Italy [29].

In total, 46,936 volunteers were evaluated. Of the eleven studies that provided information by sex, 26,180 were men (or 55.7%). Eight studies evaluated the adolescent population (n=39,776) [22, 23, 24, 25, 26, 29, 30, 32], while four studies evaluated adolescents and adults (n=7.187) [21, 27, 28, 31].

Of the studies that followed adolescents and adults, only Anderson's study had a total follow-up time equal to 12 years [21], and the others did not fix an average follow-up time [27, 28, 31]. Of the studies that followed only adolescents [22, 23, 24, 25, 26, 29, 30, 32], the average follow-up time was 10.1 years. During the time evaluated, only the study undertook by Filteau et al. (2022) [30] described adolescents and adults who were exposed but did not become infected with HIV. All the other studies evaluated populations who were exposed in utero and were infected with HIV.

Table 1. Characteristics of the studies included in the systematic review.

Author, year	Country	Population	Sample	Age (years)		Follow-ap times	HIV+ sample?	Antiretroviral use (time of use)	Used antiretroviral
Author, year	Country	Population	Sample	Age range	Average or median	Follow-ap times	HIV+ sample?	Antiretroviral use (time of use)	Used antiretroviral
Anderson, Muloiwa e Davies, 2019	South Africa	Adolescents and young adults	127 (62 men; 65 women)	10-22	Median = 15,1 (13 – 17,7)	12 years	Yes	Yes (10 years)	Nevirapine; nucleoside reverse transcriptase inhibitor; lopinavir; ritonavir; efavirenz; atazanavir.
Bakeera-Kitaka et al., 2015	Uganda	Adolescents	118 (42 men; 76 women)	10-19	Average = 13,6	2 years	Yes	Yes	Zidovudina or estavudina, lamivudina combined with nevirapina or efavirenzFour patients were on a protease inhibitor-based regimen with lopinavir/ritonavir, in combination with AZT or D4T, plus 3TC.
Bellavia et al., 2017	USA	Adolescents	1.017 (526 men; 491 women)	10-12	-	14 years	Yes and No	-	-
Boettiger et al., 2016¹	Multicentric	Adolescents	534 (234 men; 300 women)	10-19	Median = 11,8 (10,7 – 13,2)	5 years	Yes	Yes	Lamivudine, zidovudine, efavirenz, stavudine, nevirapine, tenofovir.
Bunupuradah et al., 2016²	Multicentric	Adolescents	273 (109 men; 164 women)	11-18	-	7 years	Yes	Yes (7,3 years)	Nevirapine, efavirenz and ritonavir-boosted lopinavir. Nucleoside reverse transcriptase inhibitors were lamivudine, zidovudine and stavudine.
Bounora et al., 2008	Brazil	Adolescents	108 (47 men; 61 women)	10,5-19,5	Average = 12,7	8 years	Yes	Yes	-
Crichton et al., 2019³	Multicentric	Adolescents	463	10-16	Average = 1,66 ± 8,7 (men); 1,58 ± 6,9 (women)	9,1 years	Yes	Yes (9,1 years)	-
Desai, Mullen and Mathur, 2008	USA	Adolescents	35	10-15	-		Yes	Yes (2,4 years)	Protease inhibitors.
Fabiano et al., 2013	Italy	Adolescents	24	12-20	-	8 years	Yes	Yes (8 years)	Tenofovir; efavirenz.
Filteau et al., 2022	Zambia	Adolescents	88 (35 men; 53 women)	-	Average = 13.3 (SD=2)	4 years	No	No	-
Mwambenu et al., 2022	South Africa	Adolescents	288 (149 men; 139 women)	13-18	Median = 15.8 (14,2 – 17,6)	5 years	Yes	Yes (9 years)	Children > 3 years and 10 kg started on a nonnucleoside reverse transcriptase inhibitor.
Slogrove, Judd e Leroy, 2016⁴	Multicentric	Adolescents	37.614 (18.591 men; 19.023 women)	-	Median = 12.4 (11 – 14,4)	10 years	Yes	Yes	-

¹Cambodia, India, Indonesia, Malaysia, Thailand and Vietnam; ²Cambodia, India, Malaysia, Vietnam, Thailand; ³Thailand, United Kingdon/Irland; ⁴Countries from all continents.

Results of the Studies

Weight-For-Age Z-Score (WAZ)

Five studies evaluated the WAZ mean/median in positive HIV individuals. Boettiger et al. (2016) [24], Mwambenu et al. (2022) [31], Anderson, Muloiwa and Davies (2019) [21] and Bakeera-Kitaka et al. (2015) [22] calculated the median WAZ and found median values of -2.6 (IQR: -3.6; -1.4), -1.5 (IQR: -2.5; - 0.8), -1,97 (IQR: -3,23; -0,66) and -2,61 (IQR: -3,93; -1,67) at baseline, respectively, and Bakeera-Kitaka et al. (2015) [22] also calculated the median WAZ at the follow-up (one year later) and found a value of -1,26 (IQR: -2,5; -0,4). Bounora et al. (2008) [26] identified a mean value of -0,84 (SD: 1,42) at baseline and -1,15 (DP: 1,32) at follow-up. Additionally, Mwambenu et al. (2022) [31] and Bounora et al. (2008) [26] identified a prevalence of WAZ ≤-2 of 36,7% and 17,6%, respectively, at baseline, and Bounora et al. (2008) [26] found a prevalence of 29,2% at follow-up.

Height-For-Age Z-Score (HAZ)

Ten studies evaluated the HAZ mean/median at baseline and in adolescence or young adult life. Eight of them included positive HIV individuals [21, 22, 24, 25, 26, 27, 31, 32], one included negative HIV individuals exposed to HIV in intrauterine life [30], and Bellavia et al. (2017) [23] included both negative and positive HIV individuals in their sample.

Studies including positive HIV individuals

The median HAZ identified by Crichton et al. (2019) [27] was -1.2 (IQR: -2.3, -0.2) at baseline, and the median age was 6.4 (IQR: 2.8, 9.0) years. At age 16 years, the mean (standard deviation) heights for boys and girls were 166 (8.7) cm and 158 (6.9) cm, respectively, significantly shorter than the WHO reference mean height of 173 (7.8) cm for boys and 163 (6.8) cm for girls (both P < 0.001). Boettiger et al. (2016) [24], Bunupuradah et al. (2016) [25], Slogrove, Judd & Leroy (2016) [32], Anderson, Muloiwa and Davies (2019) [21], Bakeera-Kitaka et al. (2015) [22], and Mwambenu et al. (2022) [31] estimated a median value of HAZ of -2,3 (IQR: -3,6; -1,4), -2,2 (IQR: -3,2; -1,4), -1,54 (IQR: -2,06; -0,72), -2,92 (IQR: -4,09; -1,95), -2,69 (IQR: -3,57; -1,78), and -2,2 (IQR: -3,1; -1,3), respectively, in the baseline; and -1,6 (IQR not presented), -1,5 (IQR: -2,2; -0,9), -1,60 (IQR: -2,46; -0,73), -1,52 (IQR: -2,22; -0,79), -2,58 (IQR: -3,3; -1,6), and -1,1 (IQR: -1,8; -0,6), respectively, in the follow-up. A mean value of HAZ was calculated by Bounora et al. (2008) [26], who found -0,94 (DP1,23) at baseline and -1,20 (DP1,18) at follow-up.

Study including negative HIV individuals

Filteau et al. (2022) [30] evaluated the relationship between intrauterine exposure and HAZ in young adults using a linear regression model. They found a negative association (β: -, 95% CI %: -2.6; -0.3), but it was not statistically significant.

Study including positive and negative HIV individuals

Bellavia et al. (2017) [23] calculated the median HAZ in their population and found -0,66 (IQR:-1,06; -0,27) (girls) and -0,51 (IQR: -0,91; -0,10) (boys) at baseline and -0,56 (IQR: -0,99; -0,13) (girls) and -0,75 (IQR: -1,15; -0,39) (boys) at follow-up.

BMI

Studies including positive HIV individuals

Nine studies included only positive HIV individuals in their samples. Four of them evaluated the BMI-by-age z score (BAZ) in adolescents, and one evaluated the BMI in adults. Anderson, Muloiwa and Davies (2019) [21] and Mwambenu et al. (2022) [31] calculated prevalences of BAZ<-2 and found 13.3% and 9,1% at baseline and 5.5% and 6,7% at follow-up, respectively. Additionally, Anderson, Muloiwa and Davies (2019) [21], Bakeera-Kitaka et al. (2015) [22], Slogrove, Judd and Leroy (2016) [32], and Mwambenu et al. (2022) [31] calculated the median of the BAZ in their populations and found 0.2 (IQR: -0.78; 1.25), -1.61 (IQR: -2.49; -0.81), -0.54 (IQR: -1.26; 0.13) and -0.2 (IQR: -1.0; 0.6), respectively, at baseline and -0.16 (IQR: -1.04; -0.56), -0.68 (IQR: -1.3; 0.1), -0.68 (IQR: -1.46; 0.09), and -0.6 (-1.4; 0.1), respectively, at follow-up.

Fabiano et al. (2013) [29] identified a BMI mean of 18.9 kg/m2 (CI95% 17.9-19.8) at baseline and 21 kg/m2 (CI95% 19.8-22.3) at follow-up.

Study including negative HIV individuals

One study included only negative HIV individuals exposed or unexposed to HIV in intrauterine life. Filteau et al. (2022) [30] evaluated the relationship between HIV exposure and BAZ using a linear regression model and identified a negative association (ß: -0.26; CI95% -0.57; 0.05), but it was not statistically significant.

Study including positive and negative HIV individuals

One study included HIV-positive and HIV-negative individuals. Bellavia et al. (2017) [23] calculated the median BAZ in their population and found -0.12 (IQR:-0.50; 0.27) (girls) and -0.02 (IQR: -0.39; 0.34) (boys) at baseline and -0.13 (IQR: -0.52; 0.26) (girls) and -0.19 (IQR: -0.61; 0.22) (boys) at follow-up.

Waist Circumference

Two studies evaluated waist circumference (WC). Fabiano et al. (2013) [29], including positive HIV individuals in their sample, identified a median value of 68 cm (IQR: 65-70) at baseline and 74 cm (IQR: 67-80) at follow-up. Filteau et al. (2022) [30], including only negative HIV individuals, evaluated the relationship between HIV exposure and WC using a linear regression model and found a negative association (ß: -1,5; CI95% -3.4; 0.5), but it was not statistically significant.

Fat Mass

Studies including positive HIV individuals

Three studies evaluated fat mass. Two of them included positive HIV individuals in their samples [28, 29], and only one included negative HIV individuals [30]. Desai, Mullen and Mathur (2008) [28] assessed total body fat by bioelectrical impedance, and the results reported in body fat percentage (%BF) at baseline were 34% (range 23.9 to 45; SD=7.4) and 38% (range from 31.3 to 47.3; SD=5.1) after 18 months of follow-up. Additionally, among young adults with lipodystrophy, the prevalence of high %BF (≥30%) and low %BF (<30%) was 84.6% and 15.4%, respectively. On the other hand, in those who did not have lipodystrophy, the prevalence of high %BF (≥30%) and low %BF was 27.3% and 72.7%, respectively. Fabiano et al. (2013) [29] found a prevalence of total high %BF of 16.2% (95% CI 13.4 to 19.0) at baseline and a linear increase of 0.6% (0.2 to 1.0)/year (p= 0.005); arm fat at baseline was 8.1% (7.5 to 8.6) and remained stable at follow-up (p=0.5); leg fat was 42.8% (39.6 to 45.8) at baseline and linearly decreased by 1.1% (1.5 to 0.7)/year (p<0.001) at follow-up; trunk fat percentage was 49.1% (46.2 to 52.1) at baseline and increased linearly by 1.2% (0.6 to 1.6)/year (p<0.001).

Studies including negative HIV individuals

Filteau et al. (2022) [30], evaluating negative HIV individuals, identified the association between exposure to HIV and body composition. The results showed that young adults exposed to HIV in intrauterine life had arm circumference (β: –0.8, 95% CI: –1.6; 0.0), hip circumference (β: -2.3, 95% CI %: -4.5; -0.1), triceps skinfold (β: -1.5, 95% CI: -2.9; -0.2), subscapular skinfold (β: -1.4, 95% CI %: -2.6; -0.3), and suprailiac skinfold (β: -1.2, 95% CI: -2.5; 0.2) lower than adults unexposed to HIV in intrauterine life after controlling for age and sex.

Lean Mass

Regarding lean mass, only Filteau et al. (2022) [30] evaluated the relationship between intrauterine exposure to HIV and lean mass composition in young adults and found a nonsignificant negative association (β: -0.22, 95% CI: -0.66; 0.23).

Meta-analysis

The prevalence of low HAZ (z-score<-2) and low BAZ (z-score<-2) were calculated for the combinable studies and are presented in Figures 2 and 3, respectively. Combinable studies were considered when they included similar population (positive HIV adolescents) and they presented the precalculated prevalence rate or provided the number of adolescents with low HAZ or BAZ. Thus, four studies were included in the meta-analysis aiming to evaluate the prevalence of HAZ in positive HIV adolescents. We found that 26% (P=0.26; 95% CI 0.23-0.29) of positive HIV adolescents exposed to HIV in utero and antiretroviral therapy had low HAZ (Figure 2). For BAZ, three studies were combined and we identified a global prevalence of 7% (P=0.07; 95% CI 0.05-0.09) underweight in positive HIV adolescents exposed to HIV in utero and antiretroviral therapy (Figure 3). For the other nutritional outcomes, it was not possible to estimate summary prevalence for the population due to the absence of combinable studies.

For the studies that evaluated the mean values of HAZ and BAZ, we also calculated pooled estimation for these anthropometric parameters, considering the combinable studies. The mean value of HAZ for positive HIV adolescents exposed to HIV in utero and in antiretroviral therapy was -1.58 (95% CI -1.90; -1.27) (Figure 4). For BAZ, the pooled mean value for positive HIV adolescents exposed to HIV in utero and in antiretroviral therapy was -0.44 (95% CI -0.67; -0.21) (Figure 5).

We also calculated the changes in the mean values of HAZ and BAZ after the follow-up period by subtracting the final mean from the baseline mean. We identified an increase of 0.55 (MD: 0.55, 95% CI: 0.07;1.03) and 0.12 (MD: 0.12, 95% CI: -0.75; 0.99) in the mean values of the HAZ (Figure 6) and BAZ (Figure 7), respectively, after the follow-up period of individuals exposed to HIV and ART in utero.

Quality assessment

Considering the 12 studies included in this systematic review, 66.66% presented a low risk of bias, and 33.33% presented a moderate risk of bias. The items that contributed to increasing the risk of bias between studies were lack of clarity, nonapplicability or nondeclaration of strategies to address confounding factors in the analyses performed; low clarity regarding follow-up, that is, whether it was fully followed up and, if not, whether the studies presented and explored the reasons for loss of follow-up; or even low clarity with strategies to address incomplete follow-up. More details on the risk of bias are found in Table 2.

Table 2 – Results of the risk of bias analysis for the included studies.

Questões

Anderson, Muloiwa e Davies, 2019

Bakeera-Kitaka et al., 2015

Bellavia et al., 2017

Boettiger et al., 2016

Bunupuradah et al., 2016

Bounora et al., 2008

Crichton et al., 2019

Desai, Mullen and Mathur, 2008

Fabiano et al., 2013

Filteau et al., 2022

Mwambenu et al., 2022

Slogrove, Judd e Leroy 2016

Were the two groups similar and recruited from the same population?

Y

Were the exposures measured similarly to assign people

Y

To both exposed and unexposed groups?

Y

Was the exposure measured in a valid and reliable way?

Y

Were confounding factors identified?

Y

U

Y

U

NA

U

NA

Y

NA

Were strategies to deal with confounding factors stated?

Y

N

Y

U

NA

U

NA

Y

NA

Were the groups/participants free of the outcome at the start of the study (or at the moment of exposure)?

Y

Were the outcomes measured in a valid and reliable way?

Y

U

Y

Was the follow up time reported and sufficient to be long enough for outcomes to occur?

Y

Was follow up complete, and if not, were the reasons to loss to follow up described and explored?

Y

N

Y

U

Y

U

Y

N

U

N

Were strategies to address incomplete follow up utilized?

NA

Y

U

NA

U

NA

U

NA

Y

U

Y

Was appropriate statistical analysis used?

Y

U

Y

Classification

Low risk of bias

Moderate risk of bias

Low risk of bias

Moderate risk of bias

Low risk of bias

Moderate risk of bias

Low risk of bias

Moderate risk of bias

Low risk of bias

Abbreviations: Y= Yes; N= Not; U= Unclear; NA= Not applicable.

In this systematic review, we systematized the studies that evaluated adolescents' and young adults' growth and body composition exposed to HIV and antiretroviral therapy in utero. We observed that adolescents in the uterus exposed to HIV and antiretroviral therapy continue to fail to thrive. In our meta-analysis, including prenatally acquired HIV adolescents, approximately 26% still had stunted growth (HAZ <-2dp), and 7% were underweight. The studies that reported the BMI-by-age z score found mean values within the normal limits recommended by the WHO, demonstrating that although delayed weight gain is common among adolescents with prenatally acquired HIV, BMI is less affected. One study showed that young adults exposed to HIV but not infected in intrauterine life had less fat mass than the unexposed group.

This systematic review supports the growing body of literature confirming that despite weight improvement, growth failures still occur among adolescents and young adults in utero exposed to HIV and antiretroviral therapy. Growth failure could be due to several underlying mechanisms depending upon individual HIV status and age at ART start [31, 33, 34, 35], including but not limited to the infection itself, virological and immunological control, opportunistic diseases, endocrine alterations, nutrient intake [36], nutrient absorption [37] and psychosocial factors [38, 39]. Among adolescents with prenatally acquired HIV, antiretroviral therapy substantially reduces morbidity [35] and is associated with recovery of weight and height-for-age z score, especially when started early in asymptomatic children and still without weight-height impairment [7, 40]. Virologic control is also highly associated with sustained growth. However, it is not enough to reach the averages of the general population [31, 35, 41].

Only one study in our systematic review evaluated the growth of adolescents who were HIV-exposed uninfected and found that this population has less fat mass. They also observed a nonsignificant negative association between HIV exposure and BAZ and HAZ. Evidence has shown that HEU children have higher morbidity and mortality than HIV/ART-unexposed uninfected children [32]. Desmonde and collaborators (2016) [42] described growth impairment among HIV-exposed uninfected children. A potential explanation for the growth failure in this population could be due to higher susceptibility to infectious diseases they experience during life [43]. However, this hypothesis needs to be better investigated since other factors, including residual confounding, can play a role in this relationship.

The present systematic review with meta-analysis has limitations, especially regarding the methodological fragilities of some studies included. One-third of the studies showed a moderate risk of bias, especially due to the lack of clarity in identifying confounders and which strategies were adopted to address confounding factors in the analyses undertaken in addition to the lack of clarity regarding follow-up strategies to address losses and if they were identified, described and considered in the analysis. In addition, we could not evaluate the causes of the high heterogeneity found in the meta-analyses, either through meta-regression analysis or subgroup analysis due to the small number of studies included in the meta-analysis [14].

In addition, the limited number of studies and their limitations such as lack of comparison group have restricted the ability to draw consistent conclusions about the effect and magnitude of exposure to HIV and ART in utero on anthropometric outcomes. The combinable studies of this review included only positive HIV individuals, which makes the comparison with the growth pattern of exposed children not infected by the virus unfeasible. More research is critically needed to better understand these relationships and contribute to informing prevention and intervention strategies to reduce low height in infected children.

This review identified the mean values of important anthropometric indicators, the changes in these values after the follow-up, and the prevalence of underweight and low height that affect adolescents exposed to HIV and ART in utero. However, further study is needed to describe the effect of HIV on the anthropometry of this population in greater detail, mainly among HIV-exposed uninfected adolescents. This information is needed for planning targeted interventions to meet the special needs that HIV-in utero exposure has in adolescence and young adult lives.

Antiretroviral therapy (ART)

human immunodeficiency virus (HIV)

height-for-age (HAZ)

BMI-for-age (BAZ)

Ethics approval and consent to participate

Not applicable.

Consent for Publication

Not applicable.

Availability of data and material

Data are available upon reasonable request, contacting the author PRFC ([email protected]).

Competing interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

E.S.P. is funded by the Wellcome Trust 225925/Z/22/Z.

Authors’ contributions

P.R.F.C., N.S.G., C.R.N.L., L.O.L., R.C.R.S., M.L.B. and E.S.P. performed the research and designed the research study. P.R.F.C. analysed the data. P.R.F.C., N.S.G., R.C.R.S., M.L.B. and E.S.P. wrote the paper. P.R.F.C., N.S.G., C.R.N.L., L.O.L., R.C.R.S., M.L.B. and E.S.P. revised the manuscript critically for important intellectual content.

Acknowledgements

Authors thank Wellcome Trust for funding this article.

Supporting Information

Supporting Information file 1: Pubmed Search Strategy

World Health Organization. Global health sector strategies 2022-2030. Available at: <https://www.who.int/teams/global-hiv-hepatitis-and-stis-programmes/hiv/prevention/mother-to-child-transmission-of-hiv>.
Goga A, Singh Y, Jackson D, Pillay Y, Bhardwaj S, Chirinda W, Hayashi C, Essajee S, Idele P. Is elimination of vertical transmission of HIV in high prevalence settings achievable? BMJ. 2019 Mar 26;364:l687. https://doi:10.1136/bmj.l687
Dollfus C, Chenadec J, Faye A, Blanche S, Briand N, Rouzioux C, Warszawski J. Long-Term Outcomes in Adolescents Perinatally Infected with HIV-1 and Followed Up since Birth in the French Perinatal Cohort (EPF/ANRS CO10), Clinical Infectious Diseases 2010; 51(2): 214–224. https://doi.org/10.1086/653674
Chhabra S, Fidler S, Ayers S, Bower M, Lyall H, Foster C. Malignancy and all-cause mortality; incidence in adolescents and young adults living with perinatally acquired HIV. J Virus Erad. 2020 Feb 20;6(1):30-33.
Guillén S, Ramos JT, Resino R, Bellón JM, Muñoz MA. Impact on weight and height with the use of HAART in HIV-infected children. Pediatr Infect Dis J. 2007 Apr;26(4):334-8. https://doi:10.1097/01.inf.0000257427.19764.ff
Wedderburn CJ, Weldon E, Bertran-Cobo C, Rehman AM, Stein DJ, Gibb DM, Yeung S, Prendergast AJ, Donald KA. Early neurodevelopment of HIV-exposed uninfected children in the era of antiretroviral therapy: a systematic review and meta-analysis. Lancet Child Adolesc Health. 2022 Jun;6(6):393-408. https://doi:10.1016/S2352-4642(22)00071-2
Rough K, Seage GR 3rd, Williams PL, Hernandez-Diaz S, Huo Y, Chadwick EG, Currier JS, Hoffman RM, Barr E, Shapiro DE, Patel K; PHACS and the IMPAACT P1025 Study Teams. Birth Outcomes for Pregnant Women with HIV Using Tenofovir-Emtricitabine. N Engl J Med. 2018 Apr 26;378(17):1593-1603. https://doi:10.1056/NEJMoa1701666
Sturt AS, Dokubo EK, Sint TT. Antiretroviral therapy (ART) for treating HIV infection in ART-eligible pregnant women. Cochrane Database Syst Rev. 2010 Mar 17;(3):CD008440. https://doi:10.1002/14651858.CD008440
Uthman OA, Nachega JB, Anderson J, Kanters S, Mills EJ, Renaud F, Essajee S, Doherty MC, Mofenson LM. Timing of initiation of antiretroviral therapy and adverse pregnancy outcomes: a systematic review and meta-analysis. Lancet HIV. 2017 Jan;4(1):e21-e30. https://doi:10.1016/S2352-3018(16)30195-3
Evans C, Jones CE, Prendergast AJ. HIV-exposed, uninfected infants: new global challenges in the era of pediatric HIV elimination. Lancet Infect Dis 2016; 16: e92–107
Toledo G, Landes M, van Lettow M, Tippett Barr BA, Bailey H, Thorne C, Crichton S. No Difference in Growth Outcomes up to 24 Months of Age by Duration of Exposure to Maternal Antiretroviral Therapy Among Children Who Are HIV-Exposed and Uninfected in Malawi. Front Pediatr. 2022 Jun 20;10:882468. https://doi:10.3389/fped.2022.882468
Ramokolo V, Kuhn L, Lombard C, Jackson D, Goga AE. Impact of antenatal antiretroviral drug exposure on the growth of children who are HIV-exposed uninfected: the national South African Prevention of Mother to Child Evaluation cohort study. BMC Infect Dis. 2022 Dec 6;22(1):908. https://doi:10.1186/s12879-022-07847-9
Nyemba DC, Kalk E, Madlala HP, Malaba TR, Slogrove AL, Davies MA, Boulle A, Myer L, Powis KM. Lower birth weight-for-age and length-for-age z scores in infants with in-utero HIV and ART exposure: a prospective study in Cape Town, South Africa. BMC Pregnancy Childbirth. 2021 May 4;21(1):354. https://doi:10.1186/s12884-021-03836-z
Higgins, J.P.T, Thomas, J., Chandler, J. et al. (editors). (2021). Cochrane handbook for systematic reviews of interventions version 6.2 (updated February 2021). Cochrane. Available at: <www.training.cochrane.org/handbook>.
Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021 Mar 29;372:n71. https://doi.org/ 10.1136/bmj.n71
Réus, J.C., Honnef, L.R., Massignan, C., Stefani, C.M., Canto, G.D.L. Análise da qualidade metodológica de estudos observacionais (coorte, caso-controle e transversal) com as ferramentas do Joanna Briggs Institute (JBI). In: Graziela de Luca Canto, Cristine Miron Stefani, Carla Massignan. Risco de viés em revisões sistemáticas: Guia prático. 1.ed. Curitiba: Brazil Publishing, 2021.
JEONG, J. Domain of Inverse Double Arcsine Transformation. v. 1, n. 1, p. 2–5, 2018.
Luo, D., Wan, X., Liu, J., & Tong, T. (2016). Optimally estimating the sample mean from the sample size, median, mid-range, and/or mid-quartile range. Statistical Methods In Medical Research, 27, 1785–1805.
Wan, X., Wang, W., Liu, J., & Tong, T. (2014). Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Medical Research Methodology, 14(1), 135.
Deeks JJ, Higgins JPT, Altman DG, eds. Analyzing data and undertaking meta-analyses. In: Higgins JPT, Thomas J, Chandler J, et al. eds. Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Barcelona: Centro Cochrane Iberoamericano; 2019.
Anderson K, Muloiwa R, Davies MA. Treatment outcomes in perinatally infected HIV-positive adolescents and young adults after ≥10 years on antiretroviral therapy. S Afr Med J. 2018 Dec 13;109(1):27-34. https://doi:10.7196/SAMJ.2018.v109i1.13230
Bakeera-Kitaka S. McKellar M, Snider C, Kekitiinwa A, Piloya T, Musoke P, Ronald A, Javanbakht M, Colebunders R. Antiretroviral therapy for HIV-1 infected adolescents in Uganda: Assessing the impact on growth and sexual maturation. Journal of Pediatric Infectious Diseases 3 (2008) 97–104.
Bellavia A, Williams PL, DiMeglio LA, Hazra R, Abzug MJ, Patel K, Jacobson DL, Van Dyke RB, Geffner ME, International Maternal Pediatric Adolescent AIDS Clinical Trials (IMPAACT) P219/219C Study, and the Pediatric HIV/AIDS Cohort Study (PHACS). Delay in sexual maturation in perinatally HIV-infected youths is mediated by poor growth. AIDS. 2017 Jun 1;31(9):1333-1341. https://doi:10.1097/QAD.0000000000001486
Boettiger DC, Sudjaritruk T, Nallusamy R, Lumbiganon P, Rungmaitree S, Hansudewechakul R, Kumarasamy N, Bunupuradah T, Saphonn V, Truong KH, Yusoff NKN, Do VC, Nguyen LV, Razali KAM, Fong SM, Kurniati N, Kariminia A. Non-Nucleoside Reverse Transcriptase Inhibitor-Based Antiretroviral Therapy in Perinatally HIV-Infected, Treatment-Naïve Adolescents in Asia. J Adolesc Health. 2016 Apr;58(4):451-459. https://doi:10.1016/j.jadohealth.2015.11.006
Bunupuradah T, Kariminia A, Aurpibul L, Chokephaibulkit K, Hansudewechakul R, Lumbiganon P, Vonthanak S, Vibol U, Saghayam S, Nallusamy R, Van Nguyen L, Yusoff NK, Sohn AH, Puthanakit T; TREAT Asia Pediatric HIV Observational Database. Final Height and Associated Factors in Perinatally HIV-infected Asian Adolescents. Pediatr Infect Dis J. 2016 Feb;35(2):201-4. https://doi:10.1097/INF.0000000000000961
Buonora S, Nogueira S, Pone MV, Aloé M, Oliveira RH, Hofer C. Growth parameters in HIV-vertically infected adolescents on antiretroviral therapy in Rio de Janeiro, Brazil. Ann Trop Paediatr. 2008 Mar;28(1):59-64. https://doi:10.1179/146532808X270699
Crichton European Pregnancy and Pediatric HIV Cohort Collaboration (EPPICC) study group. Height and timing of growth spurt during puberty in young people living with vertically acquired HIV in Europe and Thailand. AIDS. 2019 Oct 1;33(12):1897-1910. https://doi:10.1097/QAD.0000000000002294
Desai N, Mullen P, Mathur M. Lipodystrophy in pediatric HIV. Indian J Pediatr. 2008 Apr;75(4):351-4. https://doi:10.1007/s12098-008-0037-2
Fabiano V, Giacomet V, Viganò A, Bedogni G, Stucchi S, Cococcioni L, Mora S, Zuccotti GV. Long-term body composition and metabolic changes in HIV-infected children switched from stavudine to tenofovir and from protease inhibitors to efavirenz. Eur J Pediatr. 2013 Aug;172(8):1089-96. https://doi:10.1007/s00431-013-2018-3.
Filteau S, Kasonka L, Wells JCK, Munthali G, Chisenga M, Rehman AM. Anthropometry, body composition, early growth and chronic disease risk factors among Zambian adolescents exposed or not to perinatal maternal HIV. Br J Nutr. 2022 Jun 13;129(4):1-12. https://doi:10.1017/S0007114522001775
Mwambenu B, Ramoloko V, Laubscher R, Feucht U. Growth and the pubertal growth spurt in South African adolescents living with perinatally acquired HIV infection. PLoS One. 2022 Jan 25;17(1):e0262816. https://doi:10.1371/journal.pone.0262816
Slogrove A, Judd A, Leroy V and Collaborative Initiative for Pediatric HIV Education and Research (CIPHER) Global Cohort Collaboration The epidemiology of perinatally HIV-infected adolescents: a CIPHER cohort collaboration global analysis. 21st International AIDS Conference 18–22 July 2016, Durban, South Africa.
Omoni AO, Ntozini R, Evans C, et al. Pediatr infectar Dis J. Setembro de 2017; 36(9):869–876
Feucht UD, Van Bruwaene L, Becker PJ, Kruger M. Growth in HIV-infected children on long-term antiretroviral therapy. Trop Med Int Health. 2016 May;21(5):619–29
Jesson J, Saint-Lary L, Revegue MHDT, O'Rourke J, Townsend CL, Renaud F, Penazzato M, Leroy V. Safety and efficacy of abacavir for treating infants, children, and adolescents living with HIV: a systematic review and meta-analysis. Lancet Child Adolesc Health. 2022 Oct;6(10):692-704. doi: 10.1016/S2352-4642(22)00213-9.
Lane C, Bobrow EA, Ndatimana D, Ndayisaba GF, Adair LS. Decreased consumption of common weaning foods is associated with poor linear growth among HIV-exposed infants participating in the Kigali antiretroviral and breastfeeding assessment for the elimination of HIV (Kabeho) study. Am J Hum Biol. 2019 Nov;31(6):e23308. https://doi:10.1002/ajhb.23308
Semba RD, Miotti P, Chiphangwi JD, Henderson R, Dallabetta G, Yang LP, Hoover D. Maternal vitamin A deficiency and child growth failure during human immunodeficiency virus infection. J Acquir Immune Defic Syndr Hum Retrovirol. 1997 Mar 1;14(3):219-22. https://doi:10.1097/00042560-199703010-00004
Sevenoaks T, Wedderburn CJ, Donald KA, Barnett W, Zar HJ, Stein DJ, Naudé PJW. Association of maternal and infant inflammation with neurodevelopment in HIV-exposed uninfected children in a South African birth cohort. Brain Behav Immun. 2021 Jan;91:65-73. https://doi:10.1016/j.bbi.2020.08.021
Sirajee R, Brophy J, Conroy AL, Namasopo S, Opoka RO, Rai U, Forgie S, Salami BO, T Hawkes M. Biomarkers of Growth Faltering and Neurodevelopmental Delay in Children who are HIV-Exposed but Uninfected: A Systematic Review. Curr HIV Res. 2023 May 5. https://doi:10.2174/1570162X21666230505152846
McGrath CJ, Diener L, Richardson BA, Peacock-Chambers E, John-Stewart GC. Growth reconstitution following antiretroviral therapy and nutritional supplementation: systematic review and meta-analysis. AIDS. 2015 Sep 24;29(15):2009-23. https://doi:10.1097/QAD.0000000000000783
Achan J, Kakuru A, Ikilezi G, Mwangwa F, Plenty A, Charlebois E, Young S, Havlir D, Kamya M, Ruel T. Growth Recovery Among HIV-infected Children Randomized to Lopinavir/Ritonavir or NNRTI-based Antiretroviral Therapy. Pediatr Infect Dis J. 2016 Dec;35(12):1329-1332. https://doi:10.1097/INF.0000000000001318
Desmonde S, Goetghebuer T, Thorne C, Leroy V. Health and survival of HIV perinatally exposed but uninfected children born to HIV-infected mothers. Curr Opin HIV AIDS. 2016 Sep;11(5):465-476. https://doi:10.1097/COH.0000000000000300
Ruck C, Reikie BA, Marchant A, Kollmann TR, Kakkar F. Linking Susceptibility to Infectious Diseases to Immune System Abnormalities among HIV-Exposed Uninfected Infants. Front Immunol. 2016 Aug 19;7:310. https://doi:10.3389/fimmu.2016.00310.

No competing interests reported.

SearchStrategy.pdf

Growth and body composition of adolescents and young adults exposed to HIV and antiretroviral therapy in utero: a systematic review and meta-analysis

Status:

Version 1

Abstract

Figures

Introduction

Methods

Search strategy

Outcomes

Eligibility criteria

Study selection and data extraction

Quality assessment

Statistical analysis

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1