Past studies [1–6] have documented a broad spectrum of neurocognitive impairments among perinatally HIV-infected infants and children, including lower full-scale intelligence quotients (FSIQs) and higher rates of learning disabilities than their noninfected peers. In the era prior to HAART, rates of HIV-related progressive encephalopathy (HPE), the most severe manifestation of central nervous system (CNS) HIV, ranged from 24 to 35% in this population [1,7–11]. However, a host of more subtle neurocognitive impairments have also been described [2,12]. Past studies [6,13] indicate that an early severe HIV course may be associated with poor neurocognitive outcomes in infants and early school-aged children. The degree to which these deficits persist into adolescence is unknown.
In addition to cognitive delay, recent studies describe a high prevalence of psychiatric illness and behavioral disorders in HIV-infected children and adolescents. The rates of mood disorders, attention deficit hyperactivity disorder (ADHD), psychotropic medication use, and psychiatric hospitalization are higher in perinatally HIV-infected youth than in the general adolescent population [5,14–18]. The pathogenesis of behavioral and psychiatric illness within this population is unknown, but may be related to early neurotrophic effects of the virus.
Examining the relationship between severity of HIV disease and neurocognitive and psychiatric outcomes may influence treatment decisions in infancy and early childhood. Studies [1–3] examining the effect of antiretroviral therapy (ART) on infants and children with known neurocognitive impairment have yielded varying results. The lack of consistent neurocognitive response to HAART may argue for earlier initiation of treatment in an attempt to prevent the development of potentially irreversible CNS complications. In areas of the world where the incidence of pediatric HIV remains high and the availability of ART is increasing, it is crucial to determine whether early treatment before the onset of AIDS can prevent long-term neurodevelopmental sequelae in infants and children. Finally, understanding the prevalence of developmental delay and psychiatric illness in HIV-infected adolescents allows practitioners to provide the best care and resources for this population as they transition to adulthood.
Study design and setting
A retrospective cohort study was conducted at the HIV clinic of the Children's Hospital of Philadelphia. The clinic is a multidisciplinary specialty care center, and patients receive integrated care from clinic physicians, neuropsychologists, and clinical psychologists. Patients are seen at a minimum of every 3 months and CD4+cell counts and percentages and HIV plasma viral RNA levels (viral load) are collected at each visit. Patients undergo age-appropriate neuropsychiatric testing on an annual basis. All patient encounters since 1989 were available for review. Our study was reviewed and approved by the hospital's Institutional Review Board with a waiver of informed consent.
Selection of participants
Perinatally HIV-infected patients born before 1 September 1995 (age ≥11 years) were eligible for inclusion in the study if they had at least one outpatient visit during 2 consecutive years from 1 January 1989 through 1 September 2006 and were currently enrolled in care at the study site. Patients with HIV acquired through sexual contact or blood transfusion were excluded. There was no maximum age limit for inclusion in the study, provided the patients were currently enrolled in care at the study site. Patients were included irrespective of timing of ART initiation.
Long-term survival was defined as living at least 11 years since perinatal HIV infection. Clinical and immunologic characteristics of patients were described using the Centers for Disease Control and Prevention (CDC) clinical classification system . Within this system, class N, A, and B denote asymptomatic, mildly symptomatic, and moderately symptomatic infection, respectively. Class C denotes a past history of an AIDS-defining illness such as HPE, an opportunistic infection, or recurrent invasive bacterial infections . HAART was defined as receipt of at least three antiretroviral agents from at least two classes. Immunologic status was monitored via CD4+ cell counts and percentages. CD4+ cell percentages may vary less than absolute counts based on age, medication use, and viral coinfections . The CDC guidelines define a percentage of at least 25% as no evidence of immune suppression, 15–24% as moderate suppression, and less than 15% as severe suppression .
FSIQ was measured annually for all patients using the Weschler Intelligence Scale for Children-IV (WISC-IV) or the Weschler Abbreviated Scale of Intelligence (WASI), and the most recent test result (2005–2006 data) was used in the study analysis. Both tests are well validated measures of childhood intelligence, with a scaled mean score of 100 and a standard deviation (SD) of 15. The WISC-IV has a reliability coefficient of 0.97 for the FSIQ score . Psychiatric illness was diagnosed at the time of presentation by clinic physicians on the basis of symptomatology, and diagnoses were confirmed by a clinical psychologist. No diagnoses were assigned retrospectively during the data review process. Mental health treatment was defined as receipt of individual or group mental health counseling. Learning disability was defined as a discrete impairment in at least one area of learning, with diagnosis confirmed with psychometric testing and validated by the clinic's neuropsychology staff. Educational learning support was defined as having a formal individualized educational plan through the school system or attending specialized classes. ADHD was defined on the basis of clinical symptoms and supported by caregiver and teacher Connor's Rating Scale questionnaires. HPE was defined as at least one of the following present for at least 2 months in absence of another illness that could explain the findings: failure to attain or loss of developmental milestones or loss of intellectual ability; acquired microcephaly, cerebral atrophy with or without basal ganglia calcifications on neuroimaging, or both; or acquired symmetric motor deficit .
Data collection and statistical analysis
The medical records of all eligible patients were reviewed by one of the authors. Data were collected from a clinical database and through medical record review. Data abstracted included demographics, dates of class C diagnoses, clinical and immunologic class, neuropsychiatric testing results, past and current antiretroviral use, psychiatric illness, psychotropic medication use, and episodes of mental health hospitalization.
Data were analyzed using STATA version 9.2 (Stata Corp., College Station, Texas, USA). Continuous variables were described using mean, median, and range or interquartile range (IQR) values. Categorical variables were described using counts and percentages. Univariate analysis was conducted to explore the association between a history of class C diagnosis and clinical, immunologic, neurocognitive, and psychiatric outcomes. Continuous variables were compared with the Wilcoxon rank sum test. Categorical variables were compared with the Fisher exact test. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to estimate the magnitude and precision of estimate of effect. Unadjusted ORs rather than relative risks were presented in this cohort study to facilitate comparison with the adjusted models. Adjusted analyses were performed using logistic regression for learning disability, psychiatric diagnoses, mood disorder, mental health treatment, psychiatric hospitalization, and psychotropic medication use. Linear regression was performed for FSIQ. All models were adjusted for age at initiation of ART as the timing of medication introduction is often related to severity of illness and impacts potential neurocognitive outcomes. Additional variables such as age, CD4+ cell percentage class, and current viral load were considered as confounders, although none remained in the final model because the adjusted OR or β-coefficient did not change by more than 10%. The lifetime duration of HAART was not included in the final model because of colinearity with age at antiretroviral initiation. A two-tailed P value of less than 0.05 was considered statistically significant. For variables with a prevalence of at least 35% in the control population, our study had 80% power (α = 0.05) to detect an OR of at least 3.5.
To determine whether the timing of initiation of HAART was associated with poor neurocognitive outcomes among the subset of patients with class C diagnoses, several additional analyses were performed. Class C patients were classified on the basis of whether they had initiated HAART before or after their class C diagnosis. We then examined the association between timing of HAART initiation and long-term neurocognitive outcomes using the Fisher's exact test for categorical variables and the Wilcoxon rank sum test for continuous variables.
Characteristics of the study population
One hundred seventy-two perinatally HIV-infected patients born before 1 September 1995 received outpatient care at the study site between January 1989 and September 2006. Of these patients, 23% died, (n = 39), 26% transferred care (n = 45), and 4% were lost to follow-up (n = 7) during the study period. The remaining 81 (47%) formed the study group. Dates of birth of deceased patients receiving care during the study period ranged from 1979 to 1995, 75% of these patients died prior to 1997 when HAART became widely available. The median age at death was 1 year. Dates of birth for transferred patients ranged from 1982 to 1995. The median age of transfer was 5.3 years.
The total eligible cohort consisted of 46.9% females (n = 38) and 71.6% African–American (n = 58). The median age was 15.2 years (range 11.1–23.8 years, IQR 13.2–17.2). There were no significant difference between the class C and nonclass C groups with respect to sex, race, or current age.
Clinical, immunologic and virologic status
The CDC clinical classification of the cohort at the end of the study period was 21% class N or A, 32% class B, and 47% class C. The median age at HIV diagnosis was 9 months. The median age at class C diagnosis was 3.1 years (IQR 0.9–8.1 years). Of those patients with a class C condition, 51% later developed at least one additional C diagnosis. The initial class C diagnoses of study patients were as follows: HPE (26%), Pneumocystis jiroveci pneumonia (18.4%), recurrent severe bacterial infections (16.2%), disseminated Mycobacterium avium complex infection (7.9%), Candida esophagitis (7.9%), disseminated cytomegalovirus infection (7.9%), toxoplasmosis (2.6%), wasting syndrome (2.6%), and herpes simplex esophagitis (2.6%). The current clinical and immunologic characteristics of the cohort with respect to class C status are detailed in Table 1. There was no significant difference between those with and without class C diagnoses with respect to most recent HIV plasma viral RNA level, CD4+ cell percentage, or CDC immunologic category. Furthermore, adolescents with past class C diagnoses were as likely as those without to have achieved an undetectable level of HIV plasma RNA while on HAART.
At the end of the study period, 93% of the cohort was receiving HAART. The median ages of antiretroviral and HAART initiation were 3.1 (IQR 1–6) and 6.5 years (IQR 4.9–10.3), respectively. There was no significant difference between age at HAART initiation in class C patients (median 6.4 years, IQR 5.1–10) vs. nonclass C patients (median 7.1 years, IQR 4.8–10.3) (Wilcoxon rank sum test, P = 0.83). Sixteen patients (19.7%) were started on HAART as their initial regimen vs. single or dual-drug therapy. There was no statistically significant difference between the proportion of patients initiated on HAART as their first-line regimen between the class C and nonclass C groups. The cohort was heavily treatment experienced, with patients receiving a lifetime median of five antiretroviral regimens (IQR 3–9). Patients with class C diagnoses had been treated with a greater number of antiretroviral regimens (P = 0.002) than nonclass C patients. Of the class C patients, 68.4% had initiated HAART prior to their AIDS diagnosis.
Neurocognitive and psychiatric outcomes
The median FSIQ of the full cohort was 87 (IQR 78–99). This is within one SD of the mean reference score for the WISC and WASI, and falls within the ‘average’ category of results within the normative descriptive system of the WISC-IV . Thirty-four cohort patients (42%) had been diagnosed with a learning disability and 29 (35.8%) were receiving specialized learning support in school. Fourteen patients (17%) had a lifetime history of HPE.
The results of the univariate analysis exploring the association between class C diagnosis and neurocognitive and psychiatric outcomes are displayed in Table 2. There was a significant association between class C diagnosis and lower FSIQ, presence of a learning disability, and receipt of specialized learning support. Class C patients had significantly lower median FSIQs than those with nonclass C patients (82 vs. 93). Although those nonclass C patients remained within the ‘average’ range of FSIQ, those with class C disease ranked in the ‘below average’ category of the WISC-IV normative descriptive system. Of those class C patients, 21% had a FSIQ of less than 70, placing them in the ‘lower extreme’ category on the WISC-IV scale . After excluding those patients with a past diagnosis of HPE (n = 14), the association between class C diagnosis and decreased FSIQ retained statistical significance (P = 0.009), and the 11-point FSIQ split remained constant. Those patients with known HPE were therefore included in the final models.
A high rate of psychiatric comorbidities was also identified within the cohort. Thirty-nine patients (48%) had a diagnosed psychiatric illness, with 15 (18.5%) having multiple psychiatric comorbidities. The lifetime prevalence of specific psychiatric diagnoses was as follows: mood disorder (30.8%), psychotic disorder (8.6%), ADHD (18%), non-ADHD behavioral disorder (13.6%), and eating disorder (2.5%). Twenty-six patients (32%) had ever received psychotropic medications and 13 (16%) had a lifetime history of mental health hospitalization. In the univariate analysis (Table 2), there was a significant association between class C diagnosis and a lifetime history of psychiatric illness, mood disorder, psychotic disorder, psychotropic medication usage, and psychiatric hospitalization. There was no association between class C status and diagnosis with ADHD or a behavioral disorder.
A multivariate logistic regression model was constructed to further explore the relationship between previous class C diagnosis and current neurocognitive and psychiatric status. The results of the multivariable analysis are detailed in Table 3. After adjusting for age at ART initiation, the association between a history of severe HIV disease and neurocognitive impairment remained significant. The association between severe HIV disease and psychiatric comorbidities, in particular having at least one psychiatric diagnosis, a mood disorder, or receiving mental health services also remained significant. However, the association between class C diagnosis and mental health hospitalization did not remain significant.
In order to determine whether implementation of HAART prior to AIDS diagnosis was associated with improved long-term neurocognitive outcomes, we conducted univariate analyses among the subset of class C patients comparing those who received HAART prior to their AIDS diagnosis to those who did not. We found no difference in FSIQ or rates of learning or psychiatric disorders between the two groups.
Our study is the first to expand on recent reports that have correlated early severe HIV disease with short-term neurocognitive and psychiatric impairments in infants and early school-aged children. We now extend those results to adolescents living with HIV since birth. Although the median FSIQ of nonclass C patients fell within the average range, the median score of class C patients placed them within the below average range, even after excluding patients with known HPE. This suggests that early severe HIV disease, even in the absence of a confirmed diagnosis of HPE, may be a marker for neurotoxicity secondary to high-level viral replication or immune activation leading to persistent neurocognitive dysfunction. These data expand the results of the Females and Infant Transmission study  that described an association between class C diagnoses and decreased scores in all cognitive domains for HIV-infected children aged 3–7 years, as well as other studies [22–24] documenting this association in HIV-infected infants.
Currently, treatment guidelines for pediatric ART vary with respect to the optimal time to implement HAART. The WHO global guidelines for ART of infants recommend ART if advanced or severe symptoms (WHO clinical stage 3 or 4) are present irrespective of CD4+ cell percentage and/or if CD4+ cells <25% or total lymphocyte count is <4000 cells/μl regardless of symptoms . In contrast, the Paediatric European Network for Treatment of AIDS (PENTA) guidelines  recommend initiating infant HAART if CDC clinical class B or C symptoms are present or if the CD4+ percentage falls below 35%. The USA-based Working Group on Antiretroviral Therapy and Medical Management of HIV-Infected Children recently changed their guidelines, partly to reflect interim results of the Children with HIV Early Antiretroviral Therapy (CHER) study, which found a 75% reduction in early mortality for infants started on HAART prior to meeting immune or clinical criteria [27,28]. The group now recommends initiating therapy in all children less than 12 months of age, irrespective of CD4+ cell percentage, viral load, or clinical symptoms .
Our data suggest that an association exists between early severe HIV disease and severe and long-lasting CNS sequelae. In light of our findings and those of the CHER study, it appears that physicians should consider treatment for all asymptomatic infants below 12 months of age. Although we did not find that initiating HAART prior to severe class C disease had an effect on neurocognitive outcomes, the number of patients with class C disease was small and we were underpowered to detect even modest associations in the subanalyses. However, the development of HPE and CNS impairment in HIV-infected children on HAART has been documented, suggesting that even early initiation of HAART may not be sufficiently neuroprotective [12,30]. This may be due to either poor CNS penetration of HAART, suboptimal adherence, or neurotoxic effects of antiretroviral medications. Further studies are warranted to elucidate the relationship between timing of HAART initiation and CNS outcomes.
This longitudinal survey of HIV-infected adolescents at a single site also provides a cross-sectional overview of the clinical and neuropsychiatric status of long-term survivors of perinatally acquired HIV. Our study confirms the overall high rates of psychiatric illness, behavioral disorders, ADHD, and psychiatric hospitalization within this population [4,5,9,15,16]. In addition, we identified an association between class C disease and psychiatric illness. Early studies of HIV-infected patients with hemophilia demonstrated a correlation between severity of psychobehavioral illness and the extent of immune suppression in HIV, suggesting a correlation between HIV progression and mental health diagnoses . Although early HIV-related CNS toxicity may be a cause of psychiatric illness, the emotional and social upheaval associated with severe HIV infection likely plays a significant role. Past research in survivors of other pediatric chronic illnesses such as childhood cancer have demonstrated high rates of psychiatric and behavioral illness as well . These data suggest that HIV-infected adolescents, regardless of the severity of HIV disease, may require significant behavioral, psychiatric, and educational support services as they transition into adulthood.
Finally, our study also found that immunologic recovery from early AIDS diagnosis, with respect to CD4+ cell counts and plasma viral RNA levels, has been excellent in this population. There was no significant difference at the end of the study period between those with and without class C diagnoses with respect to these clinical parameters. This finding is promising for resource-poor areas of the world where ART is often implemented late in the course of HIV disease.
There are several limitations to this study. We were unable to perform analyses comparing the adolescents in our study group with those surviving adolescents who had transferred care prior to the study period. Not all adolescents had received the full-length WISC-IV for their most recent neuropsychiatric testing, and therefore results from the WASI were also considered in study results. However, the correlation between the WASI and WISC-IV is excellent with respect to FSIQ score . Another limitation of the retrospective study design is the inability to ascribe causality to a sole factor for the discrepancy between the class C and nonclass C with respect to neuropsychiatric outcomes. Psychiatric diagnoses were based on symptomatology, and standardized symptom rating scales were not always employed. We were unable to control for factors such as prematurity, head circumference, maternal drug use or maternal antiretroviral use, and suboptimal adherence to HAART that may have contributed to neurocognitive impairment in the study population.
Finally, the total number of patients was relatively small; therefore, it is possible that we may have been underpowered to detect certain associations, for example, the relationship between early HAART implementation and long-term neurocognitive outcomes in the subset of patients with class C disease. Although we did not find a gross effect, a smaller significant one may have been present. As the rate of maternal-to-child transmission of HIV is decreasing due to effective preventive interventions, the number of patients available for long-term follow-up is decreasing. A recent multicenter study  examining this topic in young children described 117 perinatally infected patients over seven sites. Our study population of 81 patients is therefore relatively robust for a single site. Because of the decreasing prevalence of perinatally acquired HIV, future studies of this subject will require multicenter or international studies.
In summary, we report an association between severe HIV disease and persistent neurocognitive deficits in adolescent long-term survivors of perinatally acquired HIV. We also identified high rates of psychiatric illness and hospitalization within our cohort, which were associated with severe HIV disease. These associations suggest that early ART, initiated before onset of symptomatic HIV disease, may be warranted to protect the developing CNS in infants and children with HIV.
Financial support was provided by the University of Pennsylvania Center for AIDS Research (CFAR) (IP30AI45008-01 to R.M.R. and 5-P01-AI-045008-09 subaward 542904 to A.P.S.), the Robert Wood Johnson Foundation (Faculty Scholars Program to S.S.S.), and the Doris Duke Charitable Foundation (Doris Duke Clinical Research Fellowship to S.M.W.). These funding organizations had no role in design and conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the manuscript.
S.M.W. and R.M.R. conceptualized and designed the study, data analysis and interpretation were done by all authors, first draft of manuscript was prepared by all authors, critical manuscript revisions were done by all authors, research space and resources were provided by all authors.
There are no conflicts of interest.
1. Shanbhag MC, Rutstein RM, Zaoutis T, Zhao H, Chao D, Radcliffe J. Neurocognitive functioning in pediatric human immunodeficiency virus infection: effects of combined therapy. Arch Pediatr Adolesc Med 2005; 159:651–656.
2. Jeremy RJ, Kim S, Nozyce M, Nachman S, McIntosh K, Pelton SI, et al
. Neuropsychological functioning and viral load in stable antiretroviral therapy-experienced HIV-infected children. Pediatrics 2005; 115:380–387.
3. Lindsey JC, Malee KM, Brouwers P, Hughes MD. Neurodevelopmental functioning in HIV-infected infants and young children before and after the introduction of protease inhibitor-based highly active antiretroviral therapy. Pediatrics 2007; 119:e681–e693.
4. Frederick T, Thomas P, Mascola L, Hsu HW, Rakusan T, Mapson C, et al
. Human immunodeficiency virus-infected adolescents: a descriptive study of older children in New York City, Los Angeles County, Massachusetts and Washington, DC. Pediatr Infect Dis J 2000; 19:551–555.
5. Nozyce ML, Lee SS, Wiznia A, Nachman S, Mofenson LM, Smith ME, et al
. A behavioral and cognitive profile of clinically stable HIV-infected children. Pediatrics 2006; 117:763–770.
6. Smith R, Malee K, Leighty R, Brouwers P, Mellins C, Hittelman J, et al
. Effects of perinatal HIV infection and associated risk factors on cognitive development among young children. Pediatrics 2006; 117:851–862.
7. Cooper ER, Hanson C, Diaz C, Mendez H, Abboud R, Nugent R, et al
, Females and Infants Transmission Study Group. Encephalopathy and progression of human immunodeficiency virus disease in a cohort of children with perinatally acquired human immunodeficiency virus infection. J Pediatr 1998; 132:808–812.
8. Van Rie A, Harrington PR, Dow A, Robertson K. Neurologic and neurodevelopmental manifestations of pediatric HIV/AIDS: a global perspective. Eur J Paediatr Neurol 2007; 11:1–9.
9. Chiriboga CA, Fleishman S, Champion S, Gaye-Robinson L, Abrams EJ. Incidence and prevalence of HIV encephalopathy in children with HIV infection receiving highly active antiretroviral therapy (HAART). J Pediatr 2005; 146:402–407.
10. Tardieu M, Le Chenadec J, Persoz A, Meyer L, Blanche S, Mayaux MJ. HIV-1-related encephalopathy in infants compared with children and adults. French Pediatric HIV Infection Study and the SEROCO Group. Neurology 2000; 54:1089–1095.
11. Lobato MN, Caldwell MB, Ng P, Oxtoby MJ, Pediatric Spectrum of Disease Clinical Consortium. Encephalopathy in children with perinatally acquired human immunodeficiency virus infection. J Pediatr 1995; 126(5 Pt 1):710–715.
12. Martin SC, Wolters PL, Toledo-Tamula MA, Zeichner SL, Hazra R, Civitello L. Cognitive functioning in school-aged children with vertically acquired HIV infection being treated with highly active antiretroviral therapy (HAART). Dev Neuropsychol 2006; 30:633–657.
13. Loveland KA, Stehbens JA, Mahoney EM, Sirois PA, Nichols S, Bordeaux JD, et al
. Declining immune function in children and adolescents with hemophilia and HIV infection: effects on neuropsychological performance. Hemophilia Growth and Development Study. J Pediatr Psychol 2000; 25:309–322.
14. Mellins CA, Brackis-Cott E, Dolezal C, Abrams EJ. Psychiatric
disorders in youth with perinatally acquired human immunodeficiency virus infection. Pediatr Infect Dis J 2006; 25:432–437.
15. Gaughan DM, Hughes MD, Oleske JM, Malee K, Gore CA, Nachman S. Psychiatric
hospitalizations among children and youths with human immunodeficiency virus infection. Pediatrics 2004; 113:e544–e551.
16. Battles HB, Wiener LS. From adolescence through young adulthood: psychosocial adjustment associated with long-term survival of HIV. J Adolesc Health 2002; 30:161–168.
17. Mellins CA, Smith R, O'Driscoll P, Magder LS, Brouwers P, Chase C, et al
. High rates of behavioral problems in perinatally HIV-infected children are not linked to HIV disease. Pediatrics 2003; 111:384–393.
18. Wiener L, Battles H, Ryder C, Pao M. Psychotropic medication use in human immunodeficiency virus-infected youth receiving treatment at a single institution. J Child Adolesc Psychopharmacol 2006; 16:747–753.
19. CDC. 1994 Revised Classification System for human immunodeficiency virus infection in children less than 13 years of age
20. Panel on Antiretroviral Guidelines for Adult and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents
. Department of Health and Human Services; 2008. http://www.aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf
21. The Psychological Corporation. WISC-IV technical and interpretive manual.
London: The Psychological Corporation; 2003.
22. Nozyce M, Hittelman J, Muenz L, Durako SJ, Fischer ML, Willoughby A. Effect of perinatally acquired human immunodeficiency virus infection on neurodevelopment in children during the first two years of life. Pediatrics 1994; 94(6 Pt 1):883–891.
23. Msellati P, Lepage P, Hitimana DG, Van Goethem C, Van de Perre P, Dabis F. Neurodevelopmental testing of children born to human immunodeficiency virus type 1 seropositive and seronegative mothers: a prospective cohort study in Kigali, Rwanda. Pediatrics 1993; 92:843–848.
24. Drotar D, Olness K, Wiznitzer M, Schatschneider C, Marum L, Guay L, et al
. Neurodevelopmental outcomes of Ugandan infants with HIV infection: an application of growth curve analysis. Health Psychol 1999; 18:114–121.
25. WHO. Antiretroviral therapy of HIV-infection in infants and children in resource-limited settings: towards universal access; 2006
. Geneva: WHO; 2006.
26. Sharland M, Blanche S, Castelli G, Ramos J, Gibb DM. PENTA guidelines for the use of antiretroviral therapy, 2004. HIV Med 2004; 5(Suppl 2):61–86.
27. Violari A, Cotton MF, Gibb DM, Babiker AG, Steyn J, Madhi SA, et al
. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med 2008; 359:2233–2244.
28. Siva N. Treating HIV-infected infants early prolongs life. Lancet Infect Dis 2007; 7:573.
29. Working Group of Antiretroviral Therapy and Medical Management of HIV-Infected Children. Guidelines for the use of antiretroviral agents in pediatric HIV infection; 2008
30. Mitchell CD. HIV-1 encephalopathy among perinatally infected children: neuropathogenesis and response to highly active antiretroviral therapy. Ment Retard Dev Disabil Res Rev 2006; 12:216–222.
31. Nichols S, Mahoney EM, Sirois PA, Bordeaux JD, Stehbens JA, Loveland KA, et al
. HIV-associated changes in adaptive, emotional, and behavioral functioning in children and adolescents with hemophilia: results from the Hemophilia Growth and Development Study. J Pediatr Psychol 2000; 25:545–556.
32. Stam H, Hartman EE, Deurloo JA, Groothoff J, Grootenhuis MA. Young adult patients with a history of pediatric disease: impact on course of life and transition into adulthood. J Adolesc Health 2006; 39:4–13.