Skip Navigation

CD4 Cell Response to Antiretroviral Therapy in Children with Vertically Acquired HIV Infection: Is It Associated with Age at Initiation?

  1. European Collaborative Studya
  1. Reprints or correspondence: Prof. Marie-Louise Newell, Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health UCL, 30 Guilford St., London WC1N 1EH, United Kingdom (m.newell{at}ich.ucl.ac.uk)
 
Next Section

Abstract

BackgroundConsiderable uncertainty remains as to whether early initiation of antiretroviral therapy (ART) in children with vertically acquired human immunodeficiency virus (HIV) infection increases the benefit in terms of immunological response

MethodsThe association between immunological outcome and early initiation of and/or more-potent ART was investigated, using age-standardized z scores for CD4 cell counts (hereafter, “CD4 z scores”), in 131 HIV-infected children enrolled in the European Collaborative Study, a birth cohort study

ResultsMedian age at initiation of the most-potent ART was 4 years (range, 0.1–15.5 years). Initiation of treatment after 5 months of age resulted in nonsignificantly lower CD4 z scores 6 months after initiation. Time to a 20% increase in CD4 z score was associated with age at initiation of the most-potent ART (adjusted hazard ratios [AHRs], 0.37 [P<.01] and 0.43 [P=.05] for 5 months–5 years of age and >5 years of age, respectively, compared with <5 months of age), ethnicity (AHR, 0.48 [P=.01], for black vs. white), and highly active ART (HAART) with or without prior ART (AHRs, 3.16 [P<.01] and 3.95 [P<.001], vs. mono or dual ART, respectively). The risk of subsequent deterioration of CD4 z score was similar for children who initiated ART in different age groups (χ2=0.824; P=.82)

ConclusionsWe confirm the effectiveness of HAART with respect to the recovery of CD4 cell count and suggest a benefit of initiating ART before the age of 5 months. Age at initiation of the most-potent ART was not associated with the likelihood of sustaining the recovery of CD4 cell count

Vertically acquired HIV infection has become a chronic disease, and growing numbers of perinatally infected children are surviving into adolescence and beyond [1–4]. Results from birth cohort studies of vertically infected children suggest that 70%–80% of untreated children survive to 5 years of age [3, 5, 6]; with increasing and earlier use of highly active antiretroviral therapy (HAART), survival is improving

However, despite nearly 15 years of experience in the treatment of HIV-infected children [7], considerable uncertainty remains as to when to start ART. The benefits of early initiation (prevention of disease progression) need to be balanced against the costs and drawbacks (quality of life, lifelong therapy, the need for good adherence, resistance, adverse effects, and limited therapeutic options), because a complex relationship exists among ART regimen, duration, and adherence

Although policies regarding the initiation of ART in children vary across Europe, clinical outcomes do not significantly differ by center policy [7]. However, the effect of age at initiation of ART on clinical outcome is difficult to assess, given that disease progression is slow in children receiving ART [5, 8]. Alternative approaches include the use of virological or immunological endpoints, but in young children the considerable age-related variation in HIV RNA load and CD4 cell count complicates such assessments [9, 10]. To avoid this problem, CD4 cell count can be adjusted for age. Here, we investigate whether early initiation and/or more-intense ART among HIV-infected children enrolled in the prospective European Collaborative Study (ECS) was associated with immunological or clinical outcome by use of z scores for CD4 cell counts (hereafter, “CD4 z scores”), which standardize the counts in relation to age

Previous SectionNext Section

Methods

Since 1986 [5, 11], data on infected children born to HIV-1–infected women at 11 pediatric centers in 9 European countries have been prospectively collected from birth, using a standard clinical and laboratory protocol. Follow-up was frequent during the first 2 years of life and subsequently was done at least every 6 months. Parental consent was obtained, and the ECS was approved by local ethics committees

A child was considered to be infected after the detection of virus in at least 2 separate blood samples or after the persistence of antibody beyond 18 months of age. Laboratory tests, including HIV RNA polymerase chain reaction (PCR) and assays for CD4 cell count, and were done locally, with the assays used recorded. HIV RNA copy number was usually assessed by either nucleic acid sequence–based amplification (Nuclisens; Organon Teknika) or the Roche Amplicor Monitor Test (versions 1.0 and 1.5; Roche Diagnostic Systems). CD4 cell counts were based on flow cytometry

We assigned Centers for Disease Control and Prevention (CDC) clinical and immunological categories at each follow-up visit on the basis of information on current status, which enabled reclassification in light of changing clinical status, such that children could be moved to a less-serious clinical category if their clinical or immunological status improved [5]. Clinical categories were as follows: asymptomatic (CDC clinical stage N), mildly symptomatic (stage A), moderately severe symptoms (stage B), and severe symptoms (stage C) [5, 12]. Immunological categories were as follows: normal (CDC immunological stage 1), moderate immune suppression (stage 2), and severe immune suppression (stage 3) [5, 12]. To be included in the analyses, children had to have had a CD4 cell count measured within 2 months of either side of the initiation of ART. CD4 cell counts were measured before the initiation of ART in 122 children and after the initiation of ART in 9 children; the characteristics of these 9 children were similar to those of the remaining 122. At each follow-up visit, information was collected on ART; the date of first report of the most-potent ART was considered to be the date of initiation, which in most cases would be close to the actual initiation date, especially for young children with frequent follow-up. The type of most-potent ART was classified as mono or dual ART, HAART with prior ART, and HAART without prior ART

For statistical analyses, SD scores (z scores) were obtained after modeling log10-transformed CD4 cell counts by use of the LMS method [13], using 10,330 measurements from 2106 uninfected children in the ECS cohort as the reference population [14]. By use of this method, z scores are no longer age dependent, because each measurement is compared with the standard for that age. The standardized normal distribution was used to convert CD4 z scores to percentiles

Time to an event was calculated from the date of initiation of the most-potent ART to the date of reaching a 20% increase in CD4 z score and was assessed using Kaplan-Meier and Cox regression analysis (with log-rank tests) as well as unadjusted and adjusted hazard ratios (HRs) [15, 16]. A 20% increase was deemed to be evidence of successful treatment on the basis of the percentage decrease in median CD4 z score between initiation and 6 months later. Event-free children were right censored at the time of their last visit. This analysis used an intention-to-treat approach, with all children included from the time of initiation of their most-potent ART, regardless of subsequent modifications. Recursive partitioning and regression-tree methods were used to assess which age categories had the highest degree of association with the probability of immunological response [17]; the optimal cutoffs derived from this data-driven process were included in subsequent analyses. The assumption of proportional hazards in Cox regression models was formally assessed using Therneau and Grambsch’s zph test [18]

An ordinary least-squares regression model was used to investigate the effect of neonatal characteristics and type and timing of treatment on CD4 z scores 6 months after initiation of the most-potent ART, using data from 120 vertically infected children for whom data on all variables were available. Univariable and multivariable logistic regression was used to investigate risk factors associated with deterioration of CDC clinical stage after initiation of the most-potent ART and to obtain unadjusted and adjusted odds ratios, respectively. After the multivariable model was adjusted for CDC clinical stage at initiation of the most-potent ART, the addition of subsequent variables and any improvements to the model were assessed by analysis of the deviance of the 2 models

Data entry and management were done using Microsoft Access XP (2002) and SAS (version 9.1; SAS Institute). Statistical analyses were done using R (version 2.1.0) [19]

Previous SectionNext Section

Results

Study Population

ART was initiated in 131 HIV-infected children (table 1), of whom 42 (32%) were born before 1990, 83 (63%) were born from 1990 to 1999, and only 6 (5%) were born after 1999 (reflecting the decreasing vertical-transmission rate over time [20]). Of the 131 children, 39 (30%) began receiving and only ever received mono or dual ART (median age at initiation, 21 months; range, 0.75–97 months); 55 (42%) began receiving mono or dual ART (median age at initiation, 30 months; range, 0.25–141 months) and subsequently changed to HAART (median age at initiation, 90 months; range, 17–185 months); and 37 (28%) first began receiving HAART without any prior ART (median age at initiation, 6 months; range, 1–175 months)

Figure 1
View larger version:
Figure 1

A Time to attaining a 20% increase in z score for CD4 cell count from the time of initiation of the most-potent antiretroviral therapy (ART). B Time to attaining a 20% increase in z score for CD4 cell count from the time of initiation of most-potent ART. HAART, highly active ART

Figure 2
View larger version:
Figure 2

Time to a decrease in z score for CD4 cell count to less than the initial 20% increase after initiation of the most-potent antiretroviral therapy

Table 1
View larger version:
Table 1

Characteristics of children with vertically acquired HIV infection at the time of initiation of the most-potent antiretroviral therapy (ART) (n = 131)

Immunological and Virological Status

Nearly half of the children were classified as CDC immunological stage 1 at initiation of their most-potent ART (table 1). However, at initiation, 47 children (36%) had CD4 z scores below the first percentile, 41 (31%) had CD4 z scores between the first and 15th percentiles, and 37 (28%) had CD4 z scores above the 15th percentile; 85% (40/47), 66% (27/41), and 68% (25/37), respectively, showed subsequent immunological improvement. The median CD4 z score was −2.01 (approximately the second percentile) at initiation, which improved by almost 20% to −1.64 (approximately the fifth percentile) after 6 months of ART

In children with a detectable viral load at initiation of their most-potent ART, the median HIV RNA load was 4.43 log10 copies/mL (range, 2.04–6.09 log10 copies/mL) (n=51; data were not available for 67 children, and 13 had an undetectable viral load, of whom 4 were naive and 9 had received dual ART). At 6 months after initiation, the median viral load was reduced to 3.89 log10 copies/mL (range, 1.75–6.70 log10 copies/mL) (n = 50; data were not available for 54 children, and 27 had an undetectable viral load). Of the 55 children with data on viral loads available at initiation and 6 months later, 32 (58%) had a decreased viral load (median decrease, 4.40 log10 copies/mL; range, 2.18–6.09 log10 copies/mL), 10 (18%) remained stable (with undetectable loads at both measurements), and 13 (24%) had an increased viral load (median increase, 3.63 log10 copies/mL; range, 2.06–5.86 log10 copies/mL). Achievement of an undetectable viral load was not significantly associated with the type of most-potent ART (P=.32) (although there were only 2 children receiving dual ART and all others were receiving HAART), with ethnicity (P=.48), or with age at initiation of the most-potent ART (P=1), but statistical power was limited. Viral load at initiation and CD4 z score 6 months after initiation were inversely but not highly correlated (correlation coefficient, −0.32; P<.05); similarly, the correlation coefficients for both were −0.29 (P<.05) at initiation and −0.35 (P<.01) at 6 months after initiation for the 55 children and was −0.41 for all children with both data on viral load and CD4 cell count available at 6 months after initiation

Most children were asymptomatic at initiation of the most-potent ART (table 1); 28 had improved clinical status 12 months after initiation (14 from CDC clinical stage A to N, 9 from B to N, 1 from C to A, 2 from C to B, and 2 from B to A), 20 deteriorated clinically (5 from N to A, 8 from N to B, 3 from N to C, 2 from A to B, and 2 from B to C), and 5 died (1 of whom was classified as A at initiation, 2 of whom were classified as B, and 2 of whom were classified as C). The clinical status of the remaining children was unchanged

Time to Improvement in CD4 z Score after ART

Kaplan-Meier analysisAn estimated 59% (95% confidence interval [CI], 49%–67%) of children overall would have achieved a 20% increase in CD4 z score 12 months after initiation of the most-potent ART, with white European children having a substantially and significantly increased likelihood of attaining a 20% relative increase in CD4 z score than black children (χ2=4.7; 1 df; P=.03). Differences by ethnicity only became apparent after ∼8 months of ART: at 8 months after initiation, ∼40% of both white and black children would have achieved a 20% increase in CD4 z score, but, at 24 months after initiation, the percentages would be 83% (95% CI, 71%–83%) for white children and 57% (95% CI, 33%–72%) for black children; at 36 months, the percentages would be 88% (95% CI, 75%–95%) and 68% (95% CI, 41%–83%), respectively

Time to a 20% increase in CD4 z score was not significantly associated with CDC clinical stage at the time of initiation of the most-potent ART (χ2=4.8; 1 df; P=.19), and there was only a marginally statistically significant association with CD4 z score percentile category at the time of initiation (χ2=5.0; 2 df; P=.08); children in the lowest percentile category were more likely to achieve a 20% increase

However, significant differences did exist by type of most-potent ART (χ2=19.9; 2 df; P<.0001) and age at initiation of most-potent ART (χ2=12.1; 2 df; P<.01) (figure 1A and 1B). Children who received mono or dual ART but never received HAART were considerably less likely to achieve a 20% increase in CD4 z score than were children who had received HAART with or without prior ART. Estimated, data-driven, optimal age cutoffs reflecting the changing associations between the type of most-potent ART and CD4 z score after initiation were <5 months, 5 months–5 years, and >5 years; in these age groups, 15 (79%), 13 (23%), and 9 (16%) children, respectively, received HAART as their first treatment regimen. There was some evidence that children who initiated ART between 5 months and 5 years of age were less likely to attain a 20% increase in CD4 z score than were those who initiated ART before 5 months or after 5 years of age. The estimated median time from initiation of ART to a 20% increase in CD4 z score was 5 months (95% CI, 2–21 months) for children who initiated ART before 5 months of age, was 17 months (95% CI, 10–29 months) for children who initiated ART between 5 months and 5 years of age, and was 8 months (95% CI, 6–11 months) for those who initiated ART after 5 years of age

Cox proportional-hazards analysisThese findings were confirmed in univariable Cox regression analyses (table 2): children who initiated their most-potent ART between 5 months and 5 years of age were almost 60% less likely to attain a 20% increase in CD4 z score at any time, compared with children who initiated their most-potent ART before 5 months of age. Children who initiated their most-potent ART after 5 years of age were 16% less likely to attain this increase than were children who initiated their most-potent ART before 5 months of age, but this difference was not statistically significant. However, in multivariable analysis controlling for ethnicity, type of most-potent ART, and CD4 z score, age, and CDC clinical stage at initiation, the association with age at initiation was statistically significant for both age categories, with a 20% increase in CD4 z score being ∼60% less likely in children who initiated ART after 5 months of age than in those who initiated ART before that age. Furthermore, a 20% increase in CD4 z score was approximately half as likely in black children than in white children; was 3 to nearly 4 times more likely in children receiving HAART than in those receiving mono or dual ART (regardless of history of previous ART); and was half as likely in children whose CD4 z scores were above the 15th percentile at initiation of the most-potent ART than were those whose scores were below the first percentile or between the first and 15th percentiles. Children who initiated their most-potent ART at a clinical stage of C (AIDS) were 84% less likely to achieve a 20% increase in CD4 z score than were those who initiated at a clinical stage of B, A, or N (no or mild/moderate symptoms)

Table 2
View larger version:
Table 2

Factors associated with improved immunological response (20% increase in z score for CD4 cell count) after initiation of the most-potent antiretroviral therapy (ART) (n=123)

Limiting the analysis to the 97 children who were below the 25th CD4 z score percentile at initiation of most-potent ART, we examined the time to a favorable immunological response (defined as crossing the 25th CD4 z score percentile) by use of a Kaplan-Meier analysis, censoring those children who had not reached this point at 24 months after initiation or at the time of death. An estimated 25% (95% CI, 15%–33%) of children would have crossed the 25th CD4 z score percentile 12 months after initiation

Because the actual date of initiation of the most-potent ART could have occurred before the date of first report, the analysis was repeated, taking the date of initiation as the midpoint between the date of first report of the most-potent ART and the previous follow-up visit. There were no substantial differences in the association with the type of most-potent ART (χ2=17.2; P<.0001) or with age at initiation of the most-potent ART (χ2=8.2; P=.01), and the size and direction of the adjusted HRs from Cox regression analyses were similar (data not shown)

CD4 Cell Response to ART 6 Months after Initiation

A linear regression of CD4 z scores 6 months after initiation of the most-potent ART was used to clarify the association between immunological status and ART and to quantify the associated changes in CD4 cell count. CD4 z score percentile category and CDC clinical stages B and C at initiation of the most-potent ART and HAART without prior ART were the only factors significantly associated with immunological response to treatment (table 3). HAART without prior ART was associated with a significant increase of 1.08 in z score 6 months after initiation, whereas HAART with prior ART was associated with a nonsignificant increase of 0.23 in CD4 z score 6 months after initiation, compared with mono or dual ART. Initiation of ART in children 5 months–5 years of age and those >5 years of age was associated with nonsignificant decreases of 0.60 and 0.10 in CD4 z score 6 months after initiation, respectively, compared with children who initiated ART before 5 months of age. Ethnicity was not significantly associated with CD4 z score 6 months after initiation. With CD4 z score at initiation of the most-potent ART as a continuous rather than a categorical variable, a 1 U increase in z score at initiation was associated with a highly significant increase of 0.41 in CD4 z score 6 months later

Table 3
View larger version:
Table 3

Factors associated with z score for CD4 cell count 6 months after initiation of the most-potent antiretroviral therapy (ART) (n=120; all variables are included)

ART-Associated Immunological Response and Clinical Status

To assess whether age at initiation of the most-potent ART was associated with immunological outcome, as measured by CD4 z score 6 months after initiation, in children whose HIV disease was not advanced, the multivariable analysis was repeated excluding children classified as CDC clinical stage C at the time when ART was initiated. The correlation coefficients for age were only marginally different and remained not statistically significantly associated with CD4 z score; children who were <5 months of age at the initiation of most-potent ART did better in terms of subsequent CD4 z score

To assess risk factors for the deterioration in CDC clinical stage or death 12 months after initiation, a univariable and multivariable logistic regression was performed on 112 children, with data on all variables included (data not shown). Neither type nor age at initiation of most-potent ART was significantly associated with the likelihood of clinical deterioration, but the number of events was small

Sustainability of CD4 Cell Response to ART

Treating infants with HAART before 5 months of age (which is like treating primary infection) was thus associated with a better response in terms of CD4 z score than starting treatment later, but it remained unclear whether this response was sustained. In a subanalysis of 84 children who attained at least a 20% increase in CD4 z score and with data on subsequent CD4 cell counts available, 38 (45%) experienced a reduction in their CD4 z scores to below the initial 20% increase. Kaplan-Meier analysis showed that the risk of subsequent deterioration of CD4 z score was similar for children who initiated treatment in different age groups (χ2=0.824; 2 df; P=.82) (figure 2). There was also no significant difference in risk between black and white children (χ2=0.1; 1 df; P=.72) or between children who initiated ART at different CD4 z score percentile categories (χ2=4.7; 2 df; P=.10)

Previous SectionNext Section

Discussion

The timing of the initiation of ART in HIV-infected children, particularly in those followed from birth and, thus, identified during primary infection, has been found to be key to their clinical management, but it remains a contentious issue [21–23]. In the present European cohort, many children were clinically and immunologically well at the time they were reported to have initiated their most-potent ART (HAART, in 70% of cases). Because of the substantial natural variation in CD4 cell count in children of different ages [24], we used age-adjusted CD4 z scores [14]. We found that both early initiation (before 5 months of age) and starting with HAART were strongly associated with a more-rapid improvement in CD4 cell count, independent of immunological and clinical status at initiation

The benefits of HAART with respect to immunological and clinical disease progression in infected children has been well described [5, 25–27]. We have confirmed that HAART is effective in terms of the recovery of CD4 cell count: HAART with or without prior ART was associated with a 3–4-fold increased likelihood of a 20% increase in CD4 z score, compared with mono or dual ART. Six months after initiation of the most-potent ART, children with prior mono or dual ART had a substantially lower (although nonsignificant) CD4 z score than did those who initiated HAART with or without prior ART, implying that the latter group responded faster to therapy

We have previously shown in the ECS cohort immunological differences between white European children and black children, regardless of infection status [28]. We describe here ethnic differences in immunological improvement after initiation of the most-potent ART, with black children (predominantly born to sub-Saharan African mothers) following a trajectory similar to that of white children during the first 8 months of therapy, with the former showing reduced improvement over time in unadjusted analyses. However, black children were more likely than white children to initiate HAART without prior receipt of ART and to do so before 5 months of age, and, in adjusted analyses, CD4 z score 6 months after initiation was not significantly lower in black children than in white children. Similarly, black children were not more likely to show serious clinical deterioration than were white children 12 months after initiation, although statistical power was limited

HAART has brought enormous benefits in terms of improved survival, health, and quality of life [29]—albeit with drawbacks, including adverse effects, the need for good adherence to avoid treatment failure, and the development of resistance. In this cohort of vertically infected children, immunological response to treatment was more rapid when ART was initiated early. In unadjusted analyses, white children who initiated ART before 5 months of age were equally likely to attain a 20% increase in CD4 z score as were children who initiated ART at >5 years of age, although this was likely to be due to different reasons. Children <5 months of age have a primary infection, and thymic productivity and, hence, immunological response is greater in this group than in older children [8, 30, 31]. Children ⩾5 months of age had more-severe HIV disease at the time of initiation and were also more likely to initially receive less-effective regimens. Lower CD4 cell count at initiation was associated with an increased likelihood of a 20% increase in CD4 z score. In adjusted analyses, immunological response in children initiating ART at >5 years of age became similar to that in children initiating ART between 5 months and 5 years of age. The age-related response pattern was different for black children; those who started before 5 months of age responded more rapidly than did either of the 2 older groups, and the oldest category did worst

Our results suggest that there is benefit in initiating ART early, soon after HIV diagnosis, although without treatment only about a fifth of infected children would be expected to progress rapidly to AIDS [5, 32]; our findings also suggest that the likelihood of sustaining an immunological response was not dependent on age at initiation. Predicting on clinical grounds which children would benefit from early initiation is, and is likely to remain, problematic [7, 33]. Our conclusions are based on data from a prospective observational cohort, which raises the problems of potential selection bias and temporal changes in the therapeutic management of HIV disease. For example, the youngest children were not only more likely to receive HAART but were also more likely to receive it as their first regimen. Although we adjusted for the timing of therapy and prior ART use, our findings are limited by the relatively small numbers of children in the youngest age group. However, in the absence of randomized trials to evaluate the effectiveness of earlier versus later initiation of ART, our findings contribute to the evidence base

Previous SectionNext Section

European Collaborative Study Collaborators

C. Giaquinto, O. Rampon, R. D’Elia, and A. De Rossi (Universita degli Studi di Padova, Italy); I. Grosch-Wörner (Charite Virchow-Klinikum, Berlin, Germany); J. Mok (Royal Hospital for Sick Children, Edinburgh, Scotland); I. Bates, I. de José, F. Hawkins, M. C. Garcia-Rodriguez, C. Ladrón de Guevara, J. Ma Pena, J. Gonzalez Garcia, and J. R. Arribas Lopez (Hospital Infantil La Paz, Madrid, Spain); F. Asensi-Botet, M. C. Otero, D. Pérez-Tamarit, A. Orti, M. J. San Miguel (Hospital La Fe, Valencia, Spain); H. Scherpbier, M. Kreyenbroek, M. Godfried, F. J. B. Nellen, and K. Boer (Academisch Medisch Centrum, Amsterdam, The Netherlands); A. B. Bohlin, E. Belfrage, and L. Navér (Karolinska University Hospital, Huddinge, Sweden); J. Levy, M. Hainaut, A. Peltier, and P. Barlow (Hospital St. Pierre, Brussels, Belgium); A. Ferrazin (Infectious Diseases Clinic, University of Genoa, Italy); A. De Maria (Department of Internal Medicine, University of Genoa, Italy); C. Gotta (Department of Obstetrics, Gynecology-Neonatology Unit, University of Genoa, Italy); A. Mûr, M. A. López-Vilchez, A. Payà, and R. Carreras (Hospital del Mar, Universidad Autonoma, Barcelona, Spain); N. H. Valerius (Hvidovre Hospital, Denmark); T. Niemieç (National Research Institute of Mother and Child, Warsaw, Poland); and M. Marczynska (Medical University of Warsaw, Poland)

Previous SectionNext Section

Acknowledgments

We thank all of the mothers and their children for their participation in the European Collaborative Study and L. Chieco-Bianchi, F. Zacchello, E. Ruga, A. M. Laverda, S. Cozzani, C. Cattelan, A. Mazza, B. Grella, A. R. Del Mistro, and S. Oletto (Padua, Italy); C. Feiterna and R. Weigel (Berlin, Germany); S. Burns, N. Hallam, P. L. Yap, and J. Whitelaw (Edinburgh, Scotland); B. Sancho and G. Fontan-Casanego (Madrid, Spain); A. Gonzalez Molina, M. Gobernado, J. L. Lopez, and J. Cordoba (Valencia, Spain); A. van der Plas (Amsterdam, The Netherlands); P. Bolme and U. Ewald (Sweden); G. Di Siena, E. Pontali, M. F. Pantarotto, G. Mantero, P. Dignetti, A. Nuzzolese, R. Rosso, S. Ferrando, and J. Nicoletti (Genoa, Italy); A. Hottard, M. Poncin, S. Sprecher, B. Lejeune, G. Zissis, and N. Clumeck (Belgium); and M. Guxens and P. Martinez (Barcelona, Spain). We also gratefully acknowledge the support of Colin Newell in the development and maintenance of the database

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported

    Financial support: European Commission (concerted action grant QLK2-CT-2000-00002); United Kingdom Medical Research Council Sexual Health and HIV Strategy Committee (support to the European Collaborative Study Coordinating Centre). The views expressed here are those of the authors

  • ↵a Prepared by Marie-Louise Newell (Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, University College London, London, United Kingdom), Deven Patel (Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, University College London), Tessa Goetghebuer (Hôpital St Pierre, Brussels, Belgium), and Claire Thorne (Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, University College London). European Collaborative Study collaborators are listed at the end of the text

  • Received August 2, 2005.
  • Accepted October 27, 2005.
Previous Section
 

References

  1. ↵
    1. Thorne C,
    2. Newell ML,
    3. Asensi-Botet F
    . Older children and adolescents surviving with vertically acquired HIV infection. European Collaborative Study. J Acquir Immune Defic Syndr 2002;29:396-410.
    MedlineWeb of Science
    1. de Martino M,
    2. Tovo PA,
    3. Galli L,
    4. et al
    . Puberty in perinatal HIV-1 infection: a multicentre longitudinal study of 212 children. AIDS 2001;15:1527-34.
    CrossRefMedlineWeb of Science
  3. ↵
    1. de Martino M,
    2. Tovo PA,
    3. Balducci M,
    4. et al
    . Reduction in mortality with availability of antiretroviral therapy for children with perinatal HIV-1 infection. JAMA 2000;284:190-7.
    Abstract/FREE Full Text
  4. ↵
    1. Frederick T,
    2. Thomas P,
    3. Mascola L,
    4. 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;19:551-5.
    CrossRefMedlineWeb of Science
  5. ↵
    1. European Collaborative Study
    . Fluctuations in symptoms in HIV-infected children: the first 10 years of life. Pediatrics 2001;108:116-22.
    Abstract/FREE Full Text
  6. ↵
    1. French Pediatric HIV Infection Study Group and the European Collaborative Study
    . Morbidity and mortality in European children vertically infected by HIV-1. J Acquir Immune Defic Syndr Hum Retrovirol 1997;14:442-50.
    MedlineWeb of Science
  7. ↵
    1. European Collaborative Study
    . Management of vertically HIV-infected children in Europe. Acta Paediatrica 2003;92:246-50.
    MedlineWeb of Science
  8. ↵
    1. Walker AS,
    2. Doerholt K,
    3. Sharland M,
    4. Gibb DM,
    5. Collaborative HIV Paediatric Study (CHIPS) Steering Committee
    . Response to highly active antiretroviral therapy varies with age: the UK and Ireland Collaborative HIV Paediatric Study. AIDS 2004;18:1915-24.
    CrossRefMedlineWeb of Science
  9. ↵
    1. European Collaborative Study
    . Level and pattern of HIV-1 RNA viral load over age: differences between girls and boys. AIDS 2002;16:97-104.
    CrossRefMedlineWeb of Science
  10. ↵
    1. European Collaborative Study
    . Are there gender and race differences in cellular immunity patterns over age in infected and uninfected children born to HIV-infected women. J Acquir Immune Defic Syndr 2003;33:635-41.
    CrossRefMedlineWeb of Science
  11. ↵
    1. European Collaborative Study
    . Height, weight, and growth in children born to mothers with HIV-1 infection in Europe. Pediatrics 2003;111:e52-60.
    Abstract/FREE Full Text
  12. ↵
    1. Centers for Disease Control and Prevention
    . 1994 Revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morb Mortal Wkly Rep 1994;43:1-10.
    Medline
  13. ↵
    1. Cole TJ,
    2. Green PJ
    . Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 1992;11:1305-19.
    MedlineWeb of Science
  14. ↵
    1. Bunders M,
    2. Cortina-Borja M,
    3. Newell ML
    . Age-related standards for total lymphocyte CD4+ and CD8+ T cell counts in children born in Europe. European Collaborative Study. Pediatr Infect Dis J 2005;24:595-600.
    CrossRefMedlineWeb of Science
  15. ↵
    1. Kaplan EL,
    2. Meier P
    . Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81.
    CrossRefWeb of Science
  16. ↵
    1. Cox DR,
    2. Oakes D
    . Analysis of survival data. London: Chapman and Hall; 1984.
  17. ↵
    1. Leblanc M,
    2. Crowley J
    . Relative risk trees for censored survival data. Biometrics 1992;48:411-25.
    CrossRefMedlineWeb of Science
  18. ↵
    1. Therneau TM,
    2. Grambsch PM
    . Modelling survival data: extending the Cox model. New York: Springer-Verlag; 2000.
  19. ↵
    1. R Development Core Team
    . A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2005.
  20. ↵
    1. European Collaborative Study
    . Mother-to-child transmission of HIV infection in the era of highly active antiretroviral therapy. Clin Infect Dis 2005;40:458-65.
    Abstract/FREE Full Text
  21. ↵
    1. Van de Perre P
    . Managment of HIV-infected children in Europe. Acta Paediatrica 2003;92:139-40.
    CrossRefMedlineWeb of Science
    1. Thuret I
    . When to initiate antiretroviral therapy in HIV-infected children. Arch Pediatr 2004;11:1521-4.
    CrossRefMedlineWeb of Science
  23. ↵
    1. Yogev R
    . Balancing the upside and downside of antiretroviral therapy in children. JAMA 2005;293:2272-4.
    FREE Full Text
  24. ↵
    1. European Collaborative Study
    . Are there gender and race differences in cellular immunity patterns over age in infected and uninfected children born to HIV-infected women. J Acquir Immune Defic Syndr 2003;33:635-41.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Luzuriaga K,
    2. Krogstad P,
    3. Robinson J,
    4. et al
    . Combination treatment with zidovudine, didanosine, and nevirapine in infants with human immunodeficiency virus type I infection. N Engl J Med 1997;336:1343-9.
    CrossRefMedlineWeb of Science
    1. Luzuriaga K,
    2. McManus M,
    3. Mofenson LM,
    4. Britto P,
    5. Graham BS,
    6. Sullivan JL
    . A trial of three antiretroviral regimens in HIV-1-infected children. N Engl J Med 2004;350:2471-80.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Soh C-H,
    2. Oleske JM,
    3. Spector SA,
    4. et al
    . Long-term effects of protease-inhibitor-based combination therapy on CD4 T-cell recovery in HIV-1-infected children and adolescents. Lancet 2003;362:2045-51.
    CrossRefMedlineWeb of Science
  28. ↵
    1. European Collaborative Study
    . Maternal factors and levels of total lymphocyte, CD4 and CD8 counts in uninfected children born to HIV-1-infected mothers. AIDS 2005;19:1071-9.
    MedlineWeb of Science
  29. ↵
    1. Foster CA,
    2. Lyall EGH
    . Children with HIV: improved mortality and morbidity with combination antiretroviral therapy. Curr Opin Infect Dis 2005;18:253-9.
    MedlineWeb of Science
  30. ↵
    1. Douek DC,
    2. Koup RA,
    3. McFarland RD,
    4. Sullivan JL,
    5. Luzuriaga K
    . Effect of HIV on thymic function before and after antiretroviral therapy in children. J Infect Dis 2000;181:1479-82.
    Abstract/FREE Full Text
  31. ↵
    1. Vigano A,
    2. Vella S,
    3. Saresella M,
    4. et al
    . Early immune reconstitution after potent antiretroviral therapy in HIV-infected children correlates with the increase in thymus volume. AIDS 2000;14:1-11.
    CrossRefMedlineWeb of Science
  32. ↵
    1. Faye A,
    2. Le Chenadec J,
    3. Dollfus C,
    4. et al
    . Early versus deferred antiretroviral multidrug therapy in infants infected with HIV type 1. Clin Infect Dis 2004;39:1692-8.
    Abstract/FREE Full Text
  33. ↵
    1. Italian Register for HIV Infection in Children
    . Italian guidelines for antiretroviral therapy in children with human immunodeficiency virus-type 1 infection. Acta Paediatrica 1999;88:228-32.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. J Infect Dis. (2006) 193 (7): 954-962. doi: 10.1086/500842
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most