At the end of 2002, UNAIDS estimated that 42 million people worldwide, including 3.2 million children, had HIV/AIDS. Côte d'Ivoire is the West African country with the highest prevalence of HIV: 9.7% of those aged 15–49 years are HIV positive. In Côte d'Ivoire, approximately 690 000 adults and 80 000 children (under 15 years) are infected with HIV . Most of these cases are undiagnosed. However, even when diagnosed, a vast majority of patients do not have access to highly active antiretroviral treatment (HAART) except in some capital cities and research programmes.
In the absence of treatment, the mortality rate amongst HIV-infected children living in Africa is very high. Prospective studies have shown that 30–50% of infected infants die before the age of 15 months [2–4]. Few precise mid- and long-term evaluations have been carried out and few data are available for HIV-infected African children over 5 years of age. Two studies focused on children surviving until over 5 years of age and with few symptoms [5,6]. Most of the available data have come from hospital-based or cross-sectional studies describing the symptoms or signs of paediatric infection .
Several studies in Europe and the United States have demonstrated the efficacy of antiretroviral treatments and that they substantially reduce child mortality [8–12]. Although a large number of studies have shown the feasibility and efficacy of antiretroviral multitherapy in adults [13,14], including African adults [15–18], no data have been published concerning the treatment of children in Africa.
Following the 2000 recommendations for Africa , HAART has been available for a significant proportion of HIV-infected children living in Côte d'Ivoire and recruited in the Children's Programme [Agence Nationale de Recherches sur le SIDA (ANRS)] 1244 cohort from October 2000. The UNAIDS/Ministry of Health Drug Access Initiative , with the support of the International Therapeutic Solidarity Fund, provided free antiretroviral drugs to HIV-infected children. After registration and clinical and biological examinations for eligibility to HAART, the prescription and supply of therapy was through the Paediatric Department of the Centre Hospital of Yopougon, located in Abidjan, which was the only public sector department authorized for the prescription and supply of HAART to children in Côte d'Ivoire.
The present study describes the clinical and biological effects of antiretroviral drugs in HIV-infected children in Abidjan; tolerance, compliance and secondary effects; and the feasibility of using antiretroviral multitherapy in an African country.
The study population included all HIV-infected children treated with HAART in the ANRS 1244 cohort, who were followed from October 2000 to September 2003 in the day hospital (located in the only public sector clinic in Yopougon, the main commune of Abidjan). Children who did not receive HAART during follow-up period or who received therapy with only two drugs were excluded from this study. Most of this population first received HAART after inclusion in the ANRS 1244 cohort. This cohort was initially launched to improve knowledge on the natural history of HIV/AIDS in HIV-infected children and to establish care-management guidelines for these children. The study protocol was approved by the Côte d'Ivoire Ethics Committee on AIDS.
At baseline, each child underwent a clinical and psychological evaluation and a chest radiograph. Clinical evaluations were carried out every 3 months during the follow-up period. Children were also seen for any intercurrent diseases as necessary. All diseases and health-related events during the follow-up were recorded on a standardized form. Consultations at the outpatient's clinic and the day hospital, treatments and tests were free of charge. In accordance with the Harare African Consensus elaborated in 2000, all children with a CD4 cell percentage < 25% were prescribed daily cotrimoxazole prophylaxis .
CD4 cell counts were measured at baseline and every 6 months thereafter at the Centre de Diagnostic et de Recherches sur le SIDA (CeDReS) laboratory in Abidjan. This laboratory used a double-platform technique method. Percentages of CD4 cells were measured by flow cytometry (FACScan, Becton Dickinson, Aalst-Erembodegem, Belgium). Absolute CD4 lymphocyte counts were determined by using an automated blood cell counter (MaxM coulter, Beckman, Paris, France).
Plasma HIV-1 RNA assays (HIV-1 RNA 3.0 assay, Quantiplex, East Walpole, Massachusetts, USA), using 0.2 ml of plasma, were carried out at baseline and every 6 months thereafter at the CeDReS laboratory. The detection threshold of this assay was 250 copies/ml (i.e., 2.4 log10 copies/ml) .
The criteria for eligibility for HAART were having AIDS stage infection [Center for Diseases Control and Prevention (CDC) stage C; ] or having a CD4 cell percentage < 15%.
Antiretroviral drugs were obtained from the Public Health Pharmacy, the only official route available for acquiring antiretroviral drugs in Côte d'Ivoire from 1998 . If drugs were not available, all antiretroviral drugs were simultaneously stopped.
Overall compliance to treatment was evaluated subjectively by direct individual interviews conducted by psychologists, who were in a better situation to obtain information on compliance from the children and the family members. They were direct questions on treatment and how children took the drugs. Living conditions and drug administration at home and associated problems were recorded during home visits. Families participated in monthly group meetings run by a non-governmental organization for people living with HIV. Teenagers were particularly encouraged to attend specific meetings  that addressed the subjects of treatments, their value and difficulties.
Weight and height for age were calculated. The height-for-age and weight-for-age Z-scores were calculated by subtracting the median weight or height of the reference population at the child's age from the child's weight or height and dividing by the standard deviation of the reference population at that age [Epinut 6.04cfr software; CDC, Atlanta, Georgia USA and World Health Organization (WHO), Geneva, Switzerland].
The weight and height for age, incidence of pneumonia, incidence of acute diarrhoea, CD4 lymphocyte counts and viral load before and after initiation of treatment were compared by Student's t-test or, when appropriate, Kruskall–Wallis test. As the CD4 lymphocyte count changed with age, the percentage of CD4 cells was used in the comparisons.
The probability of survival was estimated using the Kaplan–Meier survival technique. Point estimates of summary statistics are reported with 95% confidence intervals (CI). Stata 6.0 software (Stata Corp., College Station, Texas, USA) was used for statistical analyses and EpiInfo 6.04cfr (CDC and WHO) was used to compare incidence densities.
Characteristics of study population
Of the 159 HIV-infected children regularly followed in this observational cohort, 81 were excluded from the analysis (one infected with HIV-2, seven treated with two drugs only and 73 who did not need HAART at the time). They were 78 (49.0%) children (44 boys and 34 girls) who started HAART between January 2000 and September 2002. With the exception of five children who died in the first 3 months, all children received HAART for at least 4 months. The average duration of HAART was 21 months. Sixty-two children who had never received antiretroviral drugs previously started HAART after inclusion in the cohort (41 between inclusion and month 6; 13 between month 6 and month 12; and eight after month 12). The mean pre-treatment follow-up period in the cohort for these 61 children was 5.5 months (348 children-months; range, 0–19.5). Sixteen children were already receiving HAART at the time of recruitment. The total follow-up period for these 78 children, with and without HAART, was 1927 children-months. The global follow-up period under HAART was 1660 children-months.
Fifty children were referred to the clinic by the paediatric department or other health structures; nine came from PMTCT programmes and 19 came from the network of People Living With AIDS associations.
Mean age at treatment initiation was 7.2 years (median 6.5; range, 0.7–15.2). Only four children were under 2 years of age at initiation of treatment. Twenty-three children were asymptomatic but had CD4 cell percentages < 15%. Fifty-five children were symptomatic and 10 of these children were at stage C. The baseline characteristics of these 78 HIV-1-infected children before HAART initiation are presented in Table 1.
All of the children were from families with limited resources, but 29 of them (37.2%) came from very poor families (income < 30€ a month, which is half of the official minimal wage and is insufficient for life in good conditions if the only resources of the family). Thirty children had lost one of their parents (mother or father) and four had lost both. These children (43.6%) all lived with the surviving parent or with members of the extended family (grandparents, uncles and aunts).
Trademark drugs were always used with the exception of stavudine (30 mg), which was replaced by a generic drug (CIPLA, Mumbai, India) for a few months at the end of 2000 and early in 2001. During the first semester of 2001, there was a shortage of some drugs (didanosine and lamivudine) in the Public Health Pharmacy. Therefore, treatment was totally interrupted in eight children for a total of 204 days (median and mean, 25.5). These involuntary treatment interruptions occurred only once during the follow-up among these children and were 4–60 days long.
Multitherapy was prescribed according to drug availability at the time of treatment initiation or the necessity to avoid interactions with antituberculosis drugs. Two nucleoside reverse transcriptase analogues were given with either a protease inhibitor (nelfinavir) (in 61 children) or a non-nucleoside reverse transcriptase inhibitor (efavirenz) (in 17 children) (Table 2). Thirteen of the youngest children received the drugs as syrups (mean age 3.21 years; median, 2.65; range, 0.9–6.1). All the others received adult formulations. Drugs doses were calculated according to the child's weight, as recommended by each manufacturer.
For the 60 children where anthropometric measurements were available before and after treatment, the mean weight-for-age and height-for-age Z-scores before treatment were −2.02 and −2.03, respectively. After an average of 620 days of treatment, the mean weight-for-age and height-for-age scores were −1.39 and −1.83, respectively. They were significant differences between the before and during treatment scores for weight-for-age Z-score but not for height-for-age Z-score (Kruskall–Wallis test; P < 0.01 and P = 0.51, respectively).
The incidence of pneumonia and acute diarrhoea was significantly lower during treatment than during the pretreatment period: 22 episodes of pneumonia and 46 of acute diarrhoea during the 348 child-months without treatment and 41 of pneumonia and 68 of acute diarrhoea during the 1447 child-months of treatment. The incidence of pneumonia fell from 0.063 to 0.028/child-month (incidence density ratio, 2.23; 95% CI, 1.33–3.75; P = 0.018) and that of acute diarrhoea fell from 0.132 to 0.047/child-month (incidence density ratio, 2.81; 95% CI, 1.93–4.09; P < 0.0001). If these events were analysed among children in two groups, under median age and children at median age or older, the results changed. Incidence of pneumonia and diarrhoea decreased dramatically for children < 6.5 years: incidence of diarrhoea 0.12 before HAART to 0.048 with HAART (incidence density ratio, 2.48; P < 0.001) and incidence of pneumonia from 0.072 to 0.0255 (incidence density ratio, 2.82; P < 0.002). However, there was no statistically significant change in children ≥ 6.5 years: incidence of diarrhoea 0.043 before HAART and 0.032 with HAART (incidence density ratio, 1.32; P = 0.31) and incidence of pneumonia 0.014 before HAART and 0.022 with HAART (incidence density ratio, 0.64; P = 0.40).
Nine children died while taking antiretroviral therapy, after a mean treatment duration of 182 days (range, 16–539). All but one had < 4% CD4 cells at recruitment. Causes of death were lower respiratory tract infections (three), cachexia (one), atypical mycobacterial infection (one), metabolic disorders secondary to pancreatitis (one), cryptococcosis (one), encephalitis of unknown cause (one) and severe malaria (one). Among these children, the six children from whom biological data were collected after initiation of treatment showed a good virological response to the treatment, but the CD4 percentage was only restored to > 15% in one.
Probability of survival with HAART was 92.3% at 6 months (95% CI, 83.7–96.5), 91.0% at 12 months (95% CI, 82.1–95.6) and 88.1% at 18 and 24 months (95% CI, 78.3–93.6).
As 38% of the children were very severely immunocompromised (CD4 cell percentage < 5%), this cut-off was chosen to analyse survival in two groups. For children with < 5% CD4 cells before the initiation of treatment, the probability of survival under treatment was 80.0% at 6 and 12 months and 72.8% (95% CI, 52–85) at 18 and 24 months. For children with ≥ 5% CD4 cells before the initiation of treatment, the probability of survival was 100% at 6 months and 97.8% (95% CI, 85.2–99.7) from 8 to 24 months of follow-up (Fig. 1). The log rank test was statistically significant (P < 0.01).
Drug changes and tolerance of drugs
All 78 children received cotrimoxazole prophylaxis before and during antiretroviral treatment.
Twelve children had one change and five had two changes in antiretroviral drug prescription during the follow-up. Among the 12 children who changed prescription once, the reasons for this change were tuberculosis treatment (two), shortage of drugs (two), prescription simplification (five) and adverse side effects (three). Among the children who changed twice, the reasons for these changes were poor response to antiretroviral multitherapy (one) and adverse side effects (four).
Twenty-three children suffered from 28 manifestations that were considered to be possible or certain adverse drug effects (the suspected drug given in parentheses). However, only seven children (three with one change and four with two changes) had their treatment changed because of adverse effects: one child had life-threatening pancreatitis possibly related to the drugs (didanosine); one had persistent diarrhoea (nelfinavir); two had anaemia (zidovudine and lamivudine); one had cutaneous allergy (efavirenz); and two had increased liver enzyme activity (didanosine and lamivudine). The other adverse effects were less severe and transient: eight occurrences of diarrhoea (nelfinavir) (8); three with increase in amylase (lamivudine); two with peripheral neuropathies (lamivudine); one with mild anaemia (zidovudine); one with vomiting (didanosine); one with stomach pains (stavudine); four with allergic symptoms such as dermatitis (zidovudine, didanosine and efavirenz); and one with conjunctivitis (stavudine). The overall morbidity related with the drugs was 0.018/child-month of follow-up.
According to the families, adherence to treatment was globally good. This was confirmed by the counsellors and psychologists from People Living With AIDS. Difficulties were identified for 16 children. For four, these compliance problems were associated with conflicts between the child and the parents (one), with the prolonged duration of treatment (two) and the fact that several adults were involved in supervising pills (one). For the 12 others, the problems were transient: failure to renew the prescription in time by parents (five), travelling (two) or the misunderstanding of the prescription (five). In these 12, compliance generally improved after changing and simplifying the treatment [for example using Combivir® (lamivudine 150 mg/zidovudine 300 mg tablets) instead of zidovudine plus lamivudine or efavirenz instead of nelfinavir] and further interviews with members of the team.
CD4 percentage and plasma HIV-1-RNA viral load were measured before and after treatment initiation (Fig. 2 and 3, respectively): 72 of the 78 children taking HAART were tested for HIV-1 RNA and CD4 cell count before antiretroviral treatment initiation (median 52 days before initiation).
HIV-1 RNA viral load and CD4 cell counts were obtained for 73 children for an average 339 days (median, 328) after the initiation of HAART. In 36 (49.3%), the viral load was below the detection limit (< 250 copies/ml; < 2.4 log10 copies/ml); in nine (12.3%) it was between 2.4 and 3.0 log10 copies/ml (250 and < 1000 copies/ml). HIV-1 RNA viral load and CD4 cell counts were obtained for 50 children for an average 756 days (median, 718) after the initiation of HAART. In 25 (50%), the viral load was below the detection limit (< 250 copies/ml; < 2.4 log10 copies/ml); in five (10%), it was between 2.4 and 3.0 log10 copies/ml (250 and < 1000 copies/ml).
As age of initiation of HAART could be important, the proportion of children with < 3.0 log10 copies/ml was compared in two groups: ≤ 5 (21) and > 5 (54) years of age. These proportions were, respectively, 67% and 56% (P = 0.54). In addition, there were no differences in evolution of the mean CD4 cell percentage between the two groups: from 7.7% to 24% among the youngest versus 7.3 to 22% among the oldest children.
This study on HIV-infected children is one of the first to analyse the feasibility and results of antiretroviral multitherapy in African children. Overall, our findings indicate that it is possible to treat African children and that this treatment is as effective in this poor-resource setting as it is in developed countries.
These encouraging results were obtained on a pilot site and such good results could be harder to obtain in even poorer parts of Africa, but we did not introduce any drug or care not available in Abidjan and the follow-up was, apart from viral load measurements, quite simple. The results were obtained mainly because staff members were correctly trained, supervised and motivated, which is possible in many places in Africa.
Under HAART, HIV RNA viral load was below the detection limit in about 50% of children, and 60% had < 1000 copies/ml (3 log10 copies/ml). This is similar to what is generally observed in Europe and the United States . These results were obtained in an observational cohort, which is more difficult conditions than a clinical trial, and also despite periods of drug shortage and the inclusion of some children at a very advanced disease stage when treatment was started. Immunological improvements were also of the same magnitude and this leads to clear clinical benefits. The low mortality rate observed is difficult to interpret in this type of observational cohort. All but one of the children who died had extremely severe immunodepression (< 5% CD4 cells) at the beginning of treatment and most deaths occurred during the first 6 months of follow-up. The dramatic difference in probability of survival between children with < 5% and > 5% CD4 cells highlights the necessity to identify and to treat HIV-infected children at an earlier disease stage than is currently the case in Africa.
The follow-up period was short and the population of children was not homogeneous (although all but two were infected by mother-to-child transmission), containing both older children, in whom the evolution to AIDS is slow, and young children, who progressed to symptomatic disease and immunosuppression more rapidly. Although we could not see any differences in disease evolution with HAART between the two groups of children, ≤ 5 and > 5 years, the follow-up period was short. A longer follow-up period would be needed to study effectively the differences of evolution between these two populations of children.
The incidence of pneumonia and diarrhoea was nearly threefold lower under treatment among the youngest children. Although the incidence of these events may decrease naturally with age and the follow-up period was insufficient to detect the significance of this, these reductions could still be related to HAART. There was no statistically significant change in older children, which may reflect lack of power in the analysis owing to the small number of events.
The weight-for-age Z-score improved during treatment, which is a good indicator of the clinical efficacy of the antiretroviral treatments. The height-for-age Z score did not change significantly. However, we expect that, with a longer period of observation, this indicator would also improve as in developed countries [25,26]. Finally, treatment was well accepted and tolerated by most children. The main side effects requiring treatment interruption were one life-threatening case of pancreatitis, possibly related to HAART (although it is always difficult to ascertain such a relationship), anaemia and an increase in amylase. Adverse drug effects led to withdrawal or modification of the treatment in only a small number of children. Compliance was considered to be good and the problems encountered were similar to those observed in industrialized countries.
Some ‘involuntary treatment interruptions’ occurred because of stock exhaustion of drugs at the central level. Even if this did not have a major effect on the treatment regimen in this study, it is an important problem and drug distribution must be better organized.
Adherence to treatment by parents and children was good, and the clinical benefits were rapidly apparent to parents, motivating them to ensure continued treatment. A long-term evaluation of compliance is, nevertheless, needed.
Children enrolled in this cohort benefited from a well-structured medical and psychological environment even if it was not a clinical trial, which undoubtedly influenced the results. Management of even trivial infections was optimal and there was support to encourage compliance. In an environment lacking these elements, the results of antiretroviral therapy may be very different.
As stated in a recent review concerning adults , many questions remain about the use of antiretroviral treatment in settings with poor resources: for example when to start a multitherapy or whether there is a need for a ‘first-line treatment’ strategy as in tuberculosis. The scarcity of data concerning children poses additional problems. However, these very encouraging results support the idea that antiretroviral multitherapy in African children is feasible on a large scale. It will be a long time before antiretroviral treatment becomes generally available for children in Africa. Nevertheless, treatment of HIV should cover the whole family, including the children. We hope that our study will help to eliminate the feeling of inevitability that surrounds this disease in children in Africa.
We wish to thank all staff from Le Programme Enfant Yopougon for their compassionate and dedicated care, as well as the staff from the University Hospital Paediatric Department. Special thanks to AMEPOUH, CHIGATA and Sol En SI for supporting us from the very beginning and made this programme possible.
Sponsorship: The ANRS 1244 project/Programme Enfant Yopougon was funded by the French Agence Nationale de Recherches sur le SIDA and the French Cooperation (Action Coordonnée 12). There is no conflicting interest.
Note: Some of the data from this study were presented at the Second International AIDS Society Conference on HIV Pathogenesis and Treatment. Paris, July 2003 [abstract 31]. (Published in Antiviral Ther 2003, 8:S10).
1. UNAIDS, WHO. Report on the Global HIV/AIDS Epidemic.
Geneva: UNAIDS and World Health Organization; December 2002: http://www.unaids.org
2. Spira R, Lepage P, Msellati P,
van de Perre P, Leroy V, Simonon A, et al
. Natural history of human immunodeficiency virus type 1 infection in children: a five-year prospective study in Rwanda.Pediatrics
3. Taha T, Graham S, Kumwenda N,
Broadhead RL, Hoover DR, Markatis D, et al.Morbidity among human immunodeficiency virus-infected and uninfected African children.Pediatrics
4. Dabis F, Elenga N, Meda N,
Leroy V, Viho I, Manigart O, et al.18-Month mortality and perinatal exposure to zidovudine in West Africa.AIDS
5. Lepage P, van de Perre P, van Vliet G,
Nsengumuremyi F, Van Goethem C, Kestelyn P, et al.Clinical and endocrinological manifestations in perinatally HIV-1 infected children aged five years or more.Am J Dis Child
6. Kabore A, Diagbouga S, Manigart O, Hien H, de Oliveira MA, Cassel-Bereau AM, et al.Absence de progression clinique de l'infection par le VIH chez des enfants au Burkina Faso.XII International Congress on AIDS and Sexually Transmitted Diseases in Africa.
Ouagadougou, Burkina Faso, December 2001 [abstract PT3-250].
7. Vetter K, Djomand G, Zadi F, Diaby L, Brattegaard K, Timité M, et al.Clinical spectrum of human immunodeficiency virus disease in children in a West African city.Pediatr Infect Dis J
8. Gortmaker S, Hughes M, Cervia J, M Brady M, Johnson GM, Seage GR, et al.Effect of combination therapy including protease inhibitors on mortality among children and adolescents infected with HIV-1.N Eng J Med
9. van Rossum A, Scherpbier H, van Lochem EG, Pakker NG, Slieker WAT, Wolthers KC, et al.Therapeutic immune reconstitution in HIV-1-infected children is independent of their age and pretreatment immune status.AIDS
10. van Rossum A, Geelen S, Hartwig N,
Wolfs TFW, Weemaes CMR, Scherpbier HJ, et al
. Results of 2 years of treatment with protease-Inhibitor-containing antiretroviral therapy in Dutch children infected with human immunodeficiency virus type 1.Clin Infect Dis
11. Essajee S, Kim M, Gonzalez C,
Rigaud M, Kaul A, Chandwani S, et al.Immunologic and virologic responses to HAART in severely immunocompromised HIV-1 infected children.AIDS
12. Teglas J, Mayaux M, Blanche S. Antiretroviral therapy and mortality among children with perinatal HIV infection.JAMA
13. van Rossum AM, Fraaij PL, de Groot R. Efficacy of highly active antiretroviral therapy in HIV-1 infected children.Lancet Infect Dis
14. Chêne G, Binquet C, Moreau JF, Neau D, Pellegrin I, Malvy D, et al.
Changes in CD4 cell count and the risk of opportunistic infection or death after highly active antiretroviral treatment. AIDS
15. Badri M, Wilson D, Wood R. Effect of highly active antiretroviral therapy on incidence of tuberculosis in South Africa: a cohort study.
Lancet 2002, 359
16. Weidle PJ, Malamba S, Mwebaze R, Sozi C, Rukundo G, Downing R, et al.Assessment of a pilot antiretroviral drug therapy programme in Uganda: patients’ response, survival and drug resistance.Lancet
17. Laurent C, Diakhaté N, Gueye NN,
Touré MA, Sow PS, Faye MA, et al.The Senegalese government's highly active antiretroviral therapy initiative: an 18-month follow-up study.AIDS
18. Djomand G, Roels T, Ellerbrock T, Hanson D, Diomande F, Monga B, et al.Virologic and immunologic outcomes and programmatic challenges of an antiretroviral treatment pilot project in Abidjan, Côte d'Ivoire.AIDS
(suppl 3): S5–S16.
19. Anon. Use of Antiretroviral Drugs in the Management of HIV-infected Persons.
Dakar: ANRS/IMEA/IRD/SAA/UNAIDS/PNLS Sénégal/PNLS CI, IAS; 2000.
20. UNAIDS/WHO. Provisional WHO/UNAIDS secretariat recommendations on the use of cotrimoxazole prophylaxis in adults and children living with HIV/AIDS in Africa.
Geneva: UNAIDS and World Health Organization; 2000 http://www.unaids.org/publications/documents/care/general/recommendation-eng.pdf
21. Rouet F, Montcho C, Rouzioux C, Leroy V, Msellati P, Kottan JB, et al.Early diagnosis of paediatric HIV-1 infection among African breast-fed children using a quantitative plasma HIV RNA assay.AIDS
22. Centers for Disease Control and Prevention. 1994 revised classification systems for human immunodeficiency virus infection in children less than 13 years of age.MMWR
23. Juillet A, Malaval C, Msellati P, Moatti J. The impact of the UNAIDS Initiative on HIV/AIDS treatments in Côte d'Ivoire.XIII International AIDS Conference
. Durban, July 2000 [late breakers].
24. Aka Dago-Akribi LH, Dossou R. Psychological support to HIV-infected pre-teens and teenagers.XIV International Conference on AIDS.
Barcelona, July 2002 [abstract F12392].
25. Nachman SA, Lindsey JC, Pelton S, Mofenson L, McIntosh K, Wiznia A, et al.Growth in human immunodeficiency virus-infected children receiving ritonavir-containing antiretroviral therapy.Arch Pediatr Adolesc Med
26. Verweel G, van Rossum AM, Hartwig NG, Wolfs TF, Scherpbier HJ, de Groot R. Treatment with highly active antiretroviral treatment in human immunodeficiency virus type 1-infected children is associated with a sustained effect on growth.Pediatrics
27. Rabkin M, El-Sadr W, Katzenstein DA, Mukherjee J, Masur H, Mugyenyi P, et al.Antiretroviral treatment in resource-poor settings: clinical research priorities.Lancet