Sage Journals: Discover world-class research

Abstract

Background:

Children with HIV infection are often reported to be short. The aim of this study was to assess the prevalence of HIV-associated short stature in HIV endemic setting.

Methods:

Data were obtained by retrospective review of the electronic medical records. Patients were grouped into various clinical categories. For each category, the proportion of patients with height-for-age Z score of less than −2 standard deviation [SD] and of less than −3 SD was determined.

Results:

The prevalence of short stature (less than −2 SD) was 28.4%. Severe short stature (less than −3 SD) is more likely with percentage of CD4 <15% (odds ratio [OR]: 3.30, confidence interval [CI]: 1.51-7.09, P = .002) and with males (OR: 1.49, CI: 1.19-1.87, P = .001). Severe short stature is more likely with viral load >400 copies/mL (OR 2.64, CI 1.27-5.38, P = .008) and poor adherence (<95%; OR 1.72, CI 1.03-2.05, P = .037).

Conclusion:

In Botswana, short stature affects a quarter of HIV-infected children and severe short stature is associated with poor adherence to antiretroviral treatment, severe immunosuppression, and virologic failure.

Keywords

short stature HIV infection viral load CD4 virologic failure

Introduction

HIV infection is a leading cause of morbidity and mortality worldwide.¹ Globally, an estimated 34 million people were living with HIV in 2011, and about 330 000 children became infected with HIV in 2011.¹ The number of children who acquired HIV decreased from 430 000 in 2009 to 330 000 in 2011, and more than 90% of them live in sub-Saharan Africa.¹ In heavily affected countries like Botswana, HIV is the underlying reason for more than one-third of all deaths among children less than 5 years.¹ Indeed, without antiretroviral treatment (ART), the progression of HIV in children is particularly aggressive and many die at a young age.² Unlike many other countries in sub-Saharan Africa, the Botswana government introduced a free universal program for highly active antiretroviral therapy (HAART) for all citizens since 2001.³

Growth failure is reported to be common in children with untreated HIV,⁴ and children with rapidly progressive disease have both stunting and wasting and are more severely affected.⁵ HIV-infected children who die early demonstrate more severe stunting, wasting, and malnutrition than do infected children who survive.² It has also been found that infected children born to HIV-positive women have early and sustained stunting and malnutrition.⁵ Determinants of growth failure and wasting in HIV-infected children are poorly understood and are likely to be multifactorial.⁶ Endocrine derangements have been suspected in association with the growth impairment observed in pediatric HIV disease, but available data are limited and inconclusive.⁶ The growth hormone (GH) axis is affected in some HIV-infected patients with reports of GH deficiency and reduced insulin-like growth factor 1, particularly in those with failure to thrive.⁶

Several studies have found an association between short stature, cognitive impairment, and an increased likelihood of infections, particularly diarrhea.⁷ In one study, it was found that the onset of stunting before 6 months of age and persistence of the stunting to 8 years were related to cognitive function at 8 and 11 years.⁸ The link between stunting and poor development may have significance both to the individuals concerned and to the national development. Therefore, stunting may represent an enormous waste of potential in millions of children in developing countries, and there is a need for intervention to prevent it.⁷

Highly active antiretroviral therapy may preserve or restore growth in HIV-infected children.⁴ The prevalence and associated clinical variables of short stature among HIV-infected children have never been reported in a country such as Botswana, which has a high prevalence of HIV infection but also has good socioeconomic status with prevalence of underweight, stunting, and wasting in children younger than 5 years of age reported at 13%, 23%, and 5%, respectively.⁹ In addition to the early initiation of HAART, stunted and underweight children in Botswana are regularly followed up by the dietician and they enroll in a feeding program with nutritious high-caloric and high-protein supplements. All children who attend the clinic are screened for opportunistic comorbidities, like tuberculosis (TB), on a regular basis and the appropriate medical intervention is made where necessary. The objectives of this study were to determine the prevalence of short stature among a cohort of HIV-infected children at HIV Clinical Centre of Excellence (COE) in Botswana and to determine the associations between short stature and clinical variables relevant to HIV, namely, viral load (VL), CD4 count, ART, and adherence to ART.

Methods

A retrospective review of all patients with HIV infection and patients between 1 year and 18 years attending the COE as of December 2010 was performed by examining the electronic medical records (EMRs). The patients may have enrolled in the clinic at any time from July 2003 to December 2010. The most recent height and weight recorded by the clinic staff were collected for the study. Height was measured by the Harpenden stadiometer and infantometer (Seca GmbH & Co.KG).

Short stature was defined as height-for-age Z scores (HAZs) of less than −2 standard deviation (SD) for sex and chronologic age. The World Health Organization (WHO) Global Database on Child Growth and Malnutrition recommends a cutoff Z score of less than −2 to classify low height for age (stunting) as moderate and a Z score of less than −3 SD to define severe/profound stunting.¹⁰ A Z score of less than −2 indicates that a child’s HAZ is 2 SD below the age and gender-specific medians for the normal population.¹⁰

According to Botswana National Antiretroviral (ARV) guidelines, at the time of the study, a VL of less than 400 copies/mL following treatment with HAART for a period of not less than 6 months is regarded as undetectable VL.³ Viral load values of more than 400 copies/mL following 6 months of HAART are referred to as treatment failure (virologic failure).³ The standard HAART regimen is made up of a backbone of 2 nucleoside reverse transcriptase inhibitor (NRTI) and a non-NRTI (NNRTI) or a protease inhibitor.³ The study was approved by the Human Research and Development Committee, Ministry of Health, Botswana, and Baylor College of Medicine Institutional Review Board.

Statistical Analysis

Data analysis involved the transfer of all patients’ data from the EMR into the SPSS version 17.0.1. Patients who were on HAART, not on HAART, whose CD4 percentage was <15%, whose adherence was <95%, and those with detectable VL were classified accordingly. The patients’ HAZs were determined using the WHO Anthro software for all the children of younger than 5 years of age and using WHO Anthroplus software for all the children of more than 5 years of age.^11,12 Statistical comparisons were performed using Fisher exact test for categorical variables and the Wilcoxon rank-sum test for continuous variables. All analyses were conducted using STATA version 11.1 (College Station, Texas, USA). A P value <.05 was considered statistically significant.

Results

Demographics and Overall Prevalence

A total of 1604 children attended the COE during the study period (Table 1). The median height SD score (HtSDS) for age and gender of these children was −1.40, and the range was −6.0 to 2.4. Of these, 114 (7.1%) children had an HtSDS of less than −3, while 455 (28.4%) had an HtSDS of less than −2. Therefore, the overall prevalence of short stature (HtSDS < −2) among the HIV-infected children was 28.4%. The majority of these patients were in the age-group 8 to 13 years of age (46.6%).

Table 1.

Demographics of the Study Population at Botswana-Baylor COE.

Parameter	n (%)
Age (n = 1604)
1-2 years	7 (0.4)
2-8 years	331 (20.6)
8-13 years	748 (46.6)
13-18 years	518 (32.3)
Gender
Male	840 (52.4)
Female	764 (47.6)

Abbreviation: COE, Clinical Centre of Excellence.

Short Stature in Children on HAART

Of the study cohort, 1441 (90%) children were on HAART with a median HtSDS of −1.36 and a range of −5.6 to 2.4. Of these children, 100 (6.9%) children had an HtSDS of less than −3, while 413 (28.7%) had an HtSDS of less than −2 (Table 2). No statistically significant difference (P value = .49) was observed in the prevalence of short stature in children on HAART compared to those not on HAART (28.7% versus 25.8%, respectively). There were 163 children who were not on HAART; their median HtSDS was −1.40, and the range was −6.0 to +2.2. Of these, 14 (8.6%) children have an HtSDS of less than −3, while 42 (25.8%) had an HtSDS of less than −2 (Table 2).

Table 2.

Summary of the Association Between Short Stature and Clinical Variables at Botswana-Baylor COE.

	Total number	Height < −2 SD	Prevalence, %	OR	P value	CI	Height < −3 SD	Prevalence, %	OR	P value	CI
Study participants	1604	455	28.37				114	7.11
Males	838	270	32.22	1.49	.001	1.19-1.87	68	8.11	1.38	.13	0.92-2.07
Females	764	185	24.20	0.67	.001	0.54-0.84	46	6.02	0.73	.13	0.48-1.09
On HAART	1441	413	28.66	1.16	.49	0.79-1.70	100	6.94	0.79	.54	0.43-1.49
Not on HAART	163	42	25.77	0.86	.49	0.59-1.27	14	8.59	1.26	.54	0.67-2.33
CD4 percentage <15%	52	22	42.30	1.84	.04	1.02-3.34	10	19.22	3.32	.002	1.51-7.09
Virologic failure (VL > 400 copies/mL)	69	24	34.78	1.37	.28	0.80-2.33	11	15.94	2.64	.008	1.27-5.38
Adherence <95%	214	70	32.72	1.27	.15	0.92-1.75	23	10.76	1.72	.037	1.03-2.85

Abbreviations: COE, Clinical Centre of Excellence; CI, confidence interval; HAART, highly active antiretroviral therapy; VL, viral load; OR, odds ratio; SD, standard deviation.

Underweight and Wasting in All the Children in the Study

Of the 603 children aged between 1 and 10 years, 572 were on HAART. Of these children on HAART, 87 (15%) were underweight compared to 5 (16%) who were underweight in the group not on HAART. Overall, the prevalence of underweight in this age-group was 92 (15.3%) of 603 and severe underweight in 16 (2.7%) of 603. Wasting was defined as body mass index (BMI) of less than −2 SD while severe wasting was defined as a BMI of less than −3 SD. A total of 288 (18%) of 1604 patients had wasting and 70 (4%) of 1604 patients had severe wasting. There was no statistically significant difference in the prevalence of severe wasting in the patients on HAART, 67 (4.6%) of 1441, compared to those not yet on HAART, 3 (1.8%) of 163 (P = .14).

Short Stature in Children with Poor Adherence, Virologic, and Immunologic Failure

In all, 69 children had a detectable VL (virologic failure) following at least 6 months of HAART (Table 2). Of these, 11 (15.9%) children had an HtSDS of less than −3, while 24 (34.8%) had an HtSDS of less than −2. Severe short stature (HtSDS < −3) was more likely associated with virologic failure (VL > 400 copies/mL) than with VL suppression (VL < 400 copies/mL; odds ratio [OR]: 2.64, confidence interval [CI]: 1.27-5.38, P = .008) and with poor adherence (<95%) than when it was good (95%-105%; OR: 1.72, CI: 1.03-2.05, P = .037). Severe short stature (< −3 SD) was more likely with severe immunosuppression (CD4 percentage < 15%) than when the percentage of CD4 is >15% (OR: 3.30, CI: 1.51-7.09, P = .002) and stunting was more likely with males than with females (OR: 1.49, CI: 1.19-1.87, P = .001).

Discussion

This was the first study to assess the prevalence of short stature and its associated clinical variables in Botswana, which is economically ranked top in sub-Saharan Africa and is also ranked as one of the countries with the highest prevalence of HIV-related illness. Many of the affected children were born during pre-HAART era as universal HAART in Botswana started only in 2001. As compared to studies performed in other settings, the prevalence of short stature (HAZ < −2) was not high as in other settings such as southern India, where the prevalence of short stature in treatment-naive HIV-infected children of younger than 18 years of age was 58%, the majority of whom were in WHO clinical stage 3.¹³ In Tanzania, the prevalence of stunting in treatment-naive HIV-infected children younger than 15 years of age was 52%, and majority of them were also in WHO clinical stage 3.¹⁴ According to WHO clinical guidelines, any patient with a WHO clinical stage 3 is eligible for HAART initiation, and in our study population all patients in this clinical stage were already on HAART; hence, this may account for our lower prevalence rate of short stature than that quoted from Tanzania and Southern India.^13,14 According to Botswana Central Statistics office survey of the year 2000, the prevalence of short stature in children (HIV status unknown) younger than 5 years of age stood at 23%.⁹ The overall prevalence in our clinic was higher than this national figure. However, caution has to be exercised when comparing these figures as the majority of our study patients were older than 8 years of age compared to the national survey that involved children younger than 5 years of age.⁹

The prevalence of underweight and wasting in this cohort was 15% and 18%, respectively. The prevalence of underweight is comparable to Botswana national prevalence of 12.5% and is far less than that found in studies from India and Tanzania, which was 63% and 40%, respectively.^9,13,14 However, there was a high prevalence of wasting in our study group, which was 18% compared to our Botswana national average of 5%.⁹ This could be a reflection of acute malnutrition secondary to HIV infection and some preservation of the linear growth.

Our study did not reveal any statistically significant differences in the prevalence of short stature in the children who are on HAART compared to those who are not yet on HAART. This observation can be explained by the fact that most of the patients who were not on HAART did not meet the clinical criteria for the initiation of HAART, hence most of them were clinically well. Stunting is also one of the clinical criteria to initiate HAART; hence as soon as these children were noted to be stunted, they would be initiated on HAART.

There was a statistically significant association between poor adherence and virologic and immunologic failure. Poor adherence to HAART often leads to a detectable VL, which in turn results in immunological failure that predisposes to opportunist infections. Virologic failure may be secondary to virologic resistance to HAART or due to poor adherence to HAART, and it has been associated with stunting in some studies.^4,5 The causes of HIV-related growth retardation are not completely understood and may include alteration in gastro-intestinal (GI) function, chronic infections, and endocrine dysfuntion.¹⁵ It has been noted from our study that in children with virologic failure, the prevalence seems to be the highest among the age-group 13 to 18 years (Figure 1), as compared to the younger age-groups. These teenagers represent a group of children who would have started HAART late, having survived for years without it, as HAART became universally available only in 2001. This may support the long known association of short stature and chronic diseases.

Figure 1.

Height SD score (HtSDS) of all the HIV-infected children who were on HAART (A) and those not on HAART (B) distributed according to the age-group. HAART indicates highly active antiretroviral therapy.

The limitation of this study was that some of the children in the study population were orphaned, and it was not possible to calculate the genetic midparental height and the estimated target height for these children. As such, the study had limitations on whether the perceived short stature was purely a result of chronic illness, especially untreated HIV/AIDS or whether there was a hereditary component to it. The other confounding factors in this study were chronic malnutrition, chronic coinfections with HIV like TB prior to enrollment in our clinic, and also the duration of HAART. Some ARV medications have been implicated in some studies to cause short stature, thereby the combination of HAART that the patient was taking could be a confounder in this study.¹⁵ As a limitation to this study, we were not able to establish the relationship between short stature, chronic malnutrition, chronic coinfection, and the duration and combination of HAART in this study cohort.

The study also had limitations that are intrinsic to retrospective study design as the anthropometric measurements were collected by the clinic staff during the routine clinical follow-up at the clinic. Despite the fact that our clinic staff are highly trained in taking routine clinical vital signs including the anthropometric measurements, the margin of error could be minimized by standardized prospective study design.

Conclusion

In Botswana, short stature is encountered in a quarter of children infected with HIV, and severe short stature is associated with poor ART adherence, severe immunosuppression, and virologic failure.

Footnotes

Acknowledgments

We would like to thank the Botswana-Baylor Children’s Clinical Centre of Excellence and Baylor International Pediatric AIDS Initiative for the support they gave us through the Paediatric Endocrinology Clinical Fellowship training and also for allowing us to use their patients’ databases.

Authors’ Note

Dipesalema R. Joel designed the study, analyzed the results, and he was mainly responsible for writing the report. S. Faisal Ahmed supervised and mentored Dipesalema R. Joel in the design of the study, data analysis, and the full writing of the report. Jerry Makhanda and Vincent Mabikwa contributed to data analysis and the writing of the report. Gabriel Anabwani and Michael Adam Tolle gave Dipesalema R Joel the permission to carry out the study at Botswana-Baylor Children's Clinical Centre of Excellence and they also contributed to the writing of the report.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the European Society for Paediatric Endocrinology and the World Diabetes Foundation grant number WDF 07-263 through the Paediatric Endocrinology Training Centers in Africa program.

References

UNAID. 2012 Report on the global AIDS epidemic. http://www.unaids.org/en/resources/publications/2012/name,76121,en.asp. Accessed September 7, 2013.

UNAIDS. Report on the global AIDS epidemic Geneva, December 2004, UNAIDS/WHO-AIDS epidemic update. UNAIDS 44 E-WHO/CDS/CSR/EDC/2000.9. Geneva, Switzerland: UNAIDS. December 2004. www.who.int/hiv/epi-update2005_en.pdf. Accessed November 16, 2009.

Botswana National HIV/AIDS Treatment Guidelines; 2008 version. November 01st 2008 edition. Gaborone, Botswana: Department of HIV/AIDS prevention and care, Ministry of Health; 2008. Accessed April 9, 2014, from www.gov.bw/global/MOH/PC_MOH_03.PDF.

Watson

Counts

. Growth hormone deficiency in HIV-infected children following successful treatment with highly active antiretroviral therapy. J Pediatr. 2004,145(4):549–551.

Bobat

Coovadia

Moodley

. Growth in early childhood in a cohort of children born to HIV-1-infected women from Durban, South Africa. Ann Trop Pediatr. 2001;21(3):203–210.

Chantry

Frederick

Meyer

. Endocrine abnormalities and impaired growth in human immunodeficiency virus infected children. Pediatr Infect Dis J. 2007;26(1):53–60.

Grantham-McGregor

. Linear growth retardation and cognition. Lancet. 2002;359(9306):542.

Mendez

Adair

. Severity and timing of stunting in the first two years of life affect performance on cognitive tests in late childhood. J Nutr. 1999;129(8):1555–1562.

Botswana Central Statistics Office. Botswana Multiple Indicators Survey – 2000, Census I–III: 2000. Last updated November 30, 2007.

10.

Who Multicentre Growth Reference Study Group. WHO Child Growth Standards Based on Length/Height, Weight and Age. Acta Pædiatric. 2006;(suppl 450):76–85.

11.

WHO AnthroPlus for Personal computers Manual: Software for assessing growth of the world’s children and adolescent. Geneva, Switzerland: WHO; 2009. Accessed April 9, 2014, htt://www.who.int/childgrowth/software/en/.

12.

WHO Anthro for Personal computers, version 3.2.2, 2011: Software for assessing growth and development of the world’s children: Geneva, Switzerland: WHO; 2010. Accessed April 9, 2014, htt://www.who.int/growthref/tools/en/.

13.

Padmapriyadarsini

Pooranagangadevi

Chandrasekaran

. Prevalence of underweight, stunting, and wasting among children infected with Human Immunodeficiency Virus in South India. Int J Pediatr. 2009;2009:837627.

14.

Mwiru

Spiegelman

Duggan

. Growth among HIV-infected children receiving antiretroviral therapy in Dar es Salaam, Tanzania [published online January 6, 2014]. J Trop Paediatr. 2014. doi:10.1093/tropej/fmt104.

15.

Majaliwa

Mohn

Chiarelli

. Growth and Puberty in Children with HIV infection. J Endocrinol Invest. 2009;32(1):85–90.

The Prevalence and Determinants of Short Stature in HIV-Infected Children