See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/51476600
Time to and Predictors of CD4+ T-Lymphocytes Recovery in HIV-Infected
Children Initiating Highly Active Antiretroviral Therapy in Ghana
Article  in  AIDS research and treatment · May 2011
DOI: 10.1155/2011/896040 · Source: PubMed
CITATIONS READS
10 27
6 authors, including:
Lorna Renner Meghan Prin
Korle Bu Teaching Hospital Columbia University
74 PUBLICATIONS   563 CITATIONS    17 PUBLICATIONS   168 CITATIONS   
SEE PROFILE SEE PROFILE
Fang-Yong Li Bamenla Q Goka
Yale University University of Ghana
95 PUBLICATIONS   1,047 CITATIONS    68 PUBLICATIONS   1,712 CITATIONS   
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
International Data to evaluate AIDS View project
Uganda USAID Feed the Future project View project
All content following this page was uploaded by Bamenla Q Goka on 05 June 2014.
The user has requested enhancement of the downloaded file.
Hindawi Publishing Corporation
AIDS Research and Treatment
Volume 2011, Article ID 896040, 9 pages
doi:10.1155/2011/896040
Clinical Study
Time to and Predictors of CD4+ T-Lymphocytes Recovery in
HIV-Infected Children InitiatingHighly Active Antiretroviral
Therapy in Ghana
Lorna Renner,1 Meghan Prin,2 Fang-Yong Li,3 Bamenla Goka,1
Veronika Northrup,3 and Elijah Paintsil4
1Department of Child Health, University of Ghana Medical School, Accra, Ghana
2University of Medicine and Dentistry of New Jersey, Piscataway, NJ 08854-8021, USA
3Yale Center for Analytical Sciences, Yale University School of Medicine, New Haven, CT 06510-3206, USA
4Departments of Pediatrics and Pharmacology, Yale University School of Medicine, 333 Cedar Street, New Haven,
CT 06510-3206, USA
Correspondence should be addressed to Elijah Paintsil, elijah.paintsil@yale.edu
Received 26 January 2011; Revised 24 February 2011; Accepted 3 March 2011
Academic Editor: Eric Daar
Copyright © 2011 Lorna Renner et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background. CD4+ T-lymphocyte monitoring is not routinely available in most resource-limited settings. We investigated pre-
dictors of time to CD4+ T-lymphocyte recovery in HIV-infected children on highly active antiretroviral (HAART) at Korle-Bu
Teaching Hospital, Ghana. Methods. Time to CD4+ T-lymphocyte recovery was defined as achieving percent CD4+ T-lymphocytes
of 25%. We used Cox proportional hazard models for identifying significant predictor variables. Results. Of the 233 children with
complete CD4+ T-lymphocyte data, the mean age at HAART initiation was 5.5 (SD = 3.1) years. The median recovery time
was 60 weeks (95% CL: 55–65). Evidence at baseline of severe suppression in CD4+ T-lymphocyte count adjusted for age, age
at HAART initiation, gender, and having parents alive were statistically significant in predicting time to CD4+ T-lymphocyte
recovery. Conclusions. A targeted approach based on predictors of CD4+ T-lymphocyte recovery can be a viable and cost-effective
way of monitoring HAART in HIV-infected children in resource-limited settings.
1. Introduction The therapeutic goal of highly active antiretroviral ther-
apy (HAART) is to suppress HIV viral replication and restore
HIV primarily targets CD4+ T-lymphocytes, and recent immune function (i.e., CD4+ T-lymphocyte recovery). The
studies have shown that during primary HIV infection, HIV standard of care for monitoring treatment in HIV-infected
viral replication in CD4+ T-lymphocytes in gut-associated children is the routine laboratory monitoring of CD4+ T-
lymphoid tissue (GALT) results in significant CD4+ T- lymphocyte percentage or count and HIV viral load at least
lymphocyte depletion [1, 2]. Chronic HIV infection affects every 3-4 months [6]. However, CD4+ T-lymphocyte and
both quantitative and qualitative function of CD4+ T- viral load testing, for monitoring the efficacy of HAART,
lymphocytes, and disease progression results from contin- are not routinely available in most resource-limited settings
uous depletion of these cells with a concomitant increase [7]. Though several low-cost and technically less complex
in risk for opportunistic infections, acquired immune defi- devices have been developed for CD4+ T-lymphocyte testing,
ciency syndrome (AIDS), and death [3–5]. Given the central there are still quite a number of centers in resource-limited
role of CD4+ T-lymphocytes in HIV pathogenesis, CD4+ T- countries with no access to reliable CD4+ T-lymphocyte
lymphocyte determination during the course of HIV disease testing [8, 9]. Furthermore, the cost of commercially avail-
is one of the most reliable predictors of prognosis [6]. able viral load testing is prohibitive (between $50 and $100
2 AIDS Research and Treatment
per test), making it unaffordable and inaccessible to many 2.2. Statistical Analysis. Patient characteristics were summa-
HIV treatment centers in resource-limited countries [10]. rized with N and percentage. Kaplan-Meier approach was
Several studies have demonstrated that the efficacy of used to describe the cumulative proportion of CD4+ T-
HAART in HIV-infected children in resource-limited coun- lymphocyte recovery as of particular time in the followup.
tries is comparable to that of children in resource-rich We used (1–Survival) to plot the trajectories of recovery
countries [11–19]. Given the empirical obstacles of using function over time. The median recovery time across the
the standard of care for monitoring HAART therapy in categories of specific patient characteristics was evaluated
resource-limited countries and the need to develop clinical with the log-rank test. We used unadjusted Cox proportional
practices that would reduce the overall cost of patient care, hazards (CPHs) models to evaluate independent associations
we investigated factors affecting treatment response among between each of the patient characteristics and the rate
HIV-positive children on HAART in a resource-limited of CD4+ T-lymphocyte recovery. The magnitudes of the
country. In particular, we examined the predictors of the associations were summarized with hazard ratios (HRs) and
time to CD4+ T-lymphocyte recovery using retrospective 95% confidence intervals (95% CI). Adjusted associations
longitudinal data. were obtained from multivariate CPH. Significance was
established with alpha of 0.05. All analyses were performed
2. Methods with SAS 9.2 (Cary, NC).
2.1. Study Cohort. This was a single center retrospective 3. Results
study at the Pediatric HIV/AIDS Care program at Korle-Bu
Teaching Hospital in Accra, Ghana. The study population 3.1. Characteristics of Study Population. Three hundred and
consisted of all HIV-infected children on HAART, since fifty-one HIV-infected children were started on HAART
pediatric antiretroviral therapy became available in 2004. All between 2004 and 2009 at the Pediatric HIV clinic at Korle-
the patients included in the analysis were on their first-line Bu Teaching Hospital. During the study period, 233 of the
regimen of nonnucleoside analog-based HAART consisting children on HAART had at least two CD4+ T-lymphocyte
of zidovudine (AZT) or stavudine (d4T) plus lamivudine counts and were included in our analysis. The followup time
(3TC), plus either nevirapine (NVP) or efavirenz (EFV). We ranged from 11 to 349 weeks, with the median followup of
analyzed longitudinal data extracted from medical records of 127 weeks. There were two deaths; one died after meeting
children on treatment with at least two CD4+ T-lymphocyte the study’s primary outcome (i.e., CD4+ T-lymphocyte
determinations between June 2004 and December 2009. The recovery), and the other died after 54 weeks without recovery.
study was reviewed and approved by the Ethics and Protocol Both patients were included in the analysis. Table 1 illus-
Review Committees of University of Ghana Medical School trates the clinical and demographic characteristics of study
and Yale School of Medicine. participants. About 96% of the children on HAART were a
The main study outcome was time to CD4+ T-lym- year of age or older, with an almost equal male-to-female
phocyte recovery, defined as achieving and maintaining a ratio. The mean age at initiation of HAART was 5.5 (SD =
target CD4 percentage of 25% after initiation of HAART. 3.1) years, and the majority were started on HAART at an
Because of well-known large natural decline and variation in advanced stage of their disease, corroborated by their WHO
absolute CD4+ T-lymphocyte numbers in early childhood, clinical staging and immune classification. About half of the
the percentage of CD4+ T-lymphocytes is preferentially used cohort had been diagnosed and treated for tuberculosis. Of
in the management of pediatric HIV infection; the target is note is the small percentage (1.3%) of PMTCT graduates
to achieve at least 25% of CD4+ T-lymphocytes according to who became infected with HIV and went on to HAART
age on treatment [20, 21]. treatment.
The predictor variables included gender, age at start The baseline CD4+ T-lymphocyte count of children
of therapy, WHO clinical staging and WHO immune between one and five years old was significantly higher
classification at baseline, mode of transmission, HIV and than those older than six years (P < .0001). The median
living status of parents, TB diagnosis and treatment, being baseline CD4+ T-lymphocyte count was 423 (IQR: 272–742),
3
a graduate of prevention of mother-to-child transmission 531 (IQR: 305–774), and 174 cells/mm (IQR: 35–371) for
(PMTCT) program, and adherence to treatment regimen. children less than one year old, between one and five, and six
Since 2004, CD4+ T-lymphocyte counts and percentages and older, respectively.
were done at least every 6 months. CD4+ T-lymphocyte
count and percentage were quantified by a dual-platform 3.2. Time to CD4+ T-Lymphocyte Recovery. One hundred
flow cytometry technology using an FACSCount system and seventy-two (73.8%) of study subjects achieved im-
(Becton-Dickinson, Franklin Lakes, NJ) at the clinical munological recovery, that is, achieved a target CD4+ T-
laboratory at Korle-Bu Teaching Hospital according to lymphocyte percentage of 25%, during the study period.
manufacturer’s instructions. Samples were processed within In the unadjusted analyses, the median recovery time for
four hours. The laboratory is certified by the South African all ages was 60 weeks (95% CL: 55–65). Children between
Public Health Reference Laboratory and they participate ages one and five years recovered faster, with median
in an external quality assurance testing program by the recovery time of 52 weeks (95% CL: 38–60 weeks), followed
South African Public Health Reference Laboratory. Viral load by children less than one year old (median: 66 weeks; 95%
testing is not available routinely. CL: 19–210 weeks), and then children six years or older
AIDS Research and Treatment 3
Table 1: Characteristics of 233 HIV-infected children on highly evidence of immune suppression at the start of HAART
active antiretroviral therapy (HAART) from 2004–2009, Accra, therapy and at least one parent alive (Table 2). We also
Ghana. observed trends for the following predictors of faster CD4+
Characteristics N (%) T-lymphocyte recovery: female gender (P = .06) and both
Age category parents with HIV positive status (.10).
Kaplan-Meier curves of the cumulative probability over
<1 year 10 (4.3)
time of CD4+ T-lymphocyte recovery stratified by age cate-
1–5 years 118 (51.0) gory at start of HAART, gender, WHO immune classification,
>6 years 105 (45.1) and parental living status are illustrated in Figure 1.
Gender
Female 112 (48.1)
3.3. Predictors of Time to Recovery of CD4+ T-Lymphocyte.
Male 121 (51.9)
In unadjusted analyses, starting HAART between 1–5 years
Immune recovery of age (HR = 1.64, P = .002), higher baseline CD4+ T-
No 61 (26.2) lymphocyte count (HR = 1.05 per 100-cell increase, P =
Yes 172 (73.8) .001), and having at least one parent alive (HR = 1.82,
WHO immune classification P = .016) were significantly associated with faster CD4+
No evidence of suppression 31 (13.3) T-lymphocyte recovery. We found a border line association
Evidence of moderate suppression 80 (34.3) between female gender and time to recovery (HR = 1.34,
Sever suppression 122 (52.4) P = .059). There was no statistically significant association
WHO clinical staging between mode of HIV transmission, parental HIV status,
I 33 (14.1) WHO clinical staging, treatment adherence self-report, pre-
vious TB diagnosis and treatment, and CD4+ T-lymphocyte
II 53 (22.7)
recovery (Table 3).
III 106 (45.5)
Because baseline CD4+ T-lymphocyte was significantly
IV 41 (17.6)
associated with age at the start of HAART therapy and
Previous TB diagnosis and treatment since WHO immune classification at the start of HAART
Yes 119 (51.1) incorporates the effect of child’s age, we retained WHO
No 114 (48.9) immune classification in the multivariate model instead
Graduate of PMTCT program of baseline CD4+ T-lymphocytes. In a multivariate model
Yes 3 (1.3) (Table 4), statistically significant predictor variables of time
No 203 (87.1) to recovery were parental living status (P = .025), gender
Do not know 27 (11.6) (P = .051), age at start of HAART (P = .014), and WHO
Mode of transmission immune classification at start of HAART (P < .0001). The
Vertical 199 (85.4) probability of recovering faster was two times greater in
children with at least one parent alive than those with both
Indeterminate 11 (4.7)
parents deceased (HR: 2.00, 95% CI: 1.18, 3.38). Females
Do not know 23 (9.9) were 1.37 times more likely to achieve faster recovery than
Self report of adherence males (HR: 1.37, 95% CI: 1.00, 1.87). Children who were
Poor 13 (5.6) started on HAART between one and 5 years of age had 1.59
Good 46 (19.8) times greater rate of recovery than children 6 years of age or
Excellent 144 (61.8) older.
Unknown 30 (12.9)
Parental HIV status 4. Discussion
Both parents HIV-positive 28 (12.0)
Only one parent HIV-positive 128 (54.9) Our findings corroborate with earlier, though few and
Both parents unknown status 75 (32.2) far in between, reports on the efficacy of NNRTI-based
Mother unknown and father HIV-negative 2 (0.9) HAART regimens in HIV-infected children in resource-
Parental living status limited countries. Moreover, we identified predictor variables
Both parents alive 109 (46.8) that could be useful in developing a targeted approach to
One parent known alive 79 (33.9) CD4+ testing and cost-effective monitoring of HAART with
CD4+ T-lymphocyte counts where resources are limited.
Both parents dead 23 (9.9)
Furthermore, children with correlated variables predictive of
Both parents unknown 8 (3.4) slow CD4+ T-lymphocyte recovery could be identified and
One parent alive and other unknown 14 (6.0) followed more frequently and carefully.
Seventy-four percent of study participants achieved
immunological recovery during the study followup period of
(median: 69 weeks; 95% CL: 57–84 weeks). The median mean duration of 110 (SD = 67.7) weeks. The magnitude
recovery time was shorter for children with no or moderate of CD4+ T-lymphocyte recovery is consistent with other
4 AIDS Research and Treatment
1 1
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 Median recovery time: 0.4
0.3 < 12 months = 66 weeks 0.3 Median recovery time:
0.2 1–5 years = 52 weeks 0.2 Female = 56 weeks
≥ 6 years = 69 weeks
0.1 = 0.1
Male = 67 weeks
log-rank Pvalue .004 log-rank Pvalue = .056
0 0
0 52 104 156 208 260 0 52 104 156 208 260
Followup time from HAART (weeks) Followup time from HAART (weeks)
Age category Gender
< 12 months ≥ 6 years Female
1–5 years Censored Male
Censored
(a) (b)
1 1
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 Median recovery time: 0.4
0.3 No evidence of suppression = 52 weeks 0.3 Median recovery time:
Evidence of suppression = 38 weeks
0.2 At least one parent alive = 59 weeksEvidence of severe suppression = 77 weeks 0.2 Both parents dead or “unknown” = 75 weeks
0.1 log-rank Pvalue < .0001 0.1 log-rank Pvalue = .014
0 0
0 52 104 156 208 260 0 52 104 156 208 260
Followup time from HAART (weeks) Followup time from HAART (weeks)
Who immunity status Parental status
Evidence of moderate suppression At least one parent alive
Evidence of severe suppression Both parents dead or “unknown”
No evidence of suppression Censored
Censored
(c) (d)
Figure 1: Kaplan-Meier survival curves of CD4+ T-lymphocyte recovery versus significant predictor variables of recovery for the 233 HIV-
infected children on highly active antiretroviral therapy (HAART) from 2004 to 2009. (a) Age category: the median recovery time for ages
<12 months, 1 to 6 years, and ≥6 years was 66, 52, and 69 weeks, respectively (logrank, P = .004). (b) Gender: the median recovery time
for female and male was 58 and 67 weeks, respectively (logrank, P = .056). (c) WHO immune classification: the median recovery time for
no evidence of suppression, evidence of moderate suppression, and evidence of severe suppression was 52, 38, and 77 weeks, respectively
(logrank, P = .004). (d) Parental living status: the median recovery time for at least one parent alive and both parents dead or “unknown”
was 59 and 75 weeks, respectively (logrank, P = .014).
reports on pediatric antiretroviral therapy in resource- to CD4+ T-lymphocyte recovery in our cohort (the median
limited countries [11, 12, 16, 22, 23]. In a study from Thai- recovery was 60 weeks) was shorter than most of the reports
land with a median followup of 168 weeks, the probability of from resource-rich countries but compared favorably with
achieving CD4% of greater or equal to 25% ranged from 65% that of studies from other resource-limited countries.
to 98% in children with baseline CD4% ranging from ≤5% Although studies from sub-Saharan Africa have demon-
to 24% [14]. In a report on efficacy of HAART in 14 resource- strated the efficacy of HAART in HIV-infected children,
limited countries (including 10 from sub-Saharan Africa), the predictors of immunological response have not been
62% of children attained CD4 of ≥25% after one year of investigated systematically [26]. In most centers, CD4+ T-
receiving antiretroviral therapy [23]. In a cohort study from lymphocyte monitoring is either unavailable or done infre-
Cambodia, median CD4% rose from 6% at baseline to quently due to resource constraints. We found it expedient
25% at 12 months [16]. In contrast, the PATCG -219 study for the same reasons to look at patient characteristics that
reported that only 26% to 49% of children on HAART with could predict immunological success in a resource-limited
baseline CD4% ranging from <5 to 24% achieved CD4% >2 setting. Statistically significant predictor variables of CD4+
after three years on therapy [24, 25]. Moreover, the time T-lymphocyte recovery in our study were age at starting
Cumulative probability of recovery Cumulative probability of recovery
(1-cumulative survival) (1-cumulative survival)
Cumulative probability of recovery Cumulative probability of recovery
(1-cumulative survival) (1-cumulative survival)
AIDS Research and Treatment 5
Table 2: Median CD4+ T-lymphocyte recovery time among 233 HIV-infected children on highly active antiretroviral therapy (HAART)
from 2004–2009, Accra, Ghana.
Median recovery time, weeks (95% CL) P value of Log rank
Age category .004
<1 year 66 (19, 210)
1–5 years 52 (38, 60)
>6 years 69 (57, 84)
WHO immune classification <.0001
No evidence of suppression 52 (35, 56)
Evidence of moderate suppression 38 (32, 53)
Severe suppression 74 (63, 84)
Gender .056
Female 56 (44, 60)
Male 67 (56, 80)
Graduate of PMTCT <.0001
Yes 16 (2, 35)
No 60 (56, 66)
Parental living status .014
At least one parent alive∗ 59 (52, 62)
Both parents died or Do not know 75 (53, 150)
Parental HIV status .099
Both parents with HIV 39 (26, 60)
One parent with HIV 62 (57, 71)
∗
Include those with “both parents alive”, “status known one alive”, and “status known one alive and another status unknown”.
N = 156. Exclude “both parents with unknown HIV status” (n = 75) and “mother unknown HIV status and father without HIV” (n = 2).
HAART, gender, WHO immune classification, and parental Consistent with previous reports from resource-rich
living status. WHO clinical staging, mode of transmission, settings [19, 31–33] and resource-limited countries [14, 28],
HIV status of parents, TB diagnosis and treatment, and baseline absolute CD4+ T-lymphocyte count correlated pos-
adherence to treatment regimen were not statistically signif- itively with the rate of immune reconstitution. HIV-infected
icant predictors. Being a graduate of the Korle-Bu Teaching children in the United States who began antiretroviral ther-
Hospital PMTCT program was a significant predictor of time apy when severely immunosuppressed (CD4+ T-lymphocyte
to recovery; however, this finding was based on a very small <15%) had a significantly shorter time to first-line regimen
sample (N = 3). It is worth mentioning that the very few switch than children who were immunocompetent or mod-
numbers of HIV-infected children from the program attest erately immunosuppressed at initiation [34]. Combining the
to the merit of the PMTCT programs. One can speculate effect of age and baseline CD4+ T-lymphocyte on recovery
that the faster recovery rate of graduates of PMTCT program raises the importance of early diagnosis and treatment to
who go on HAART may be due to good followup and strong achieve immune reconstitution and preservation of first-
patient compliance. line regimens. This is particularly pertinent to resource-
Our finding that children aged 1–5 years at the start of limited countries where second-line regimes are limited and
HAART achieve CD4+ T-lymphocyte recovery significantly expensive.
faster than older children confirms the results from previous Having at least one parent alive was found to predict
pediatric studies in both resource-limited and resource-rich rate of immune recovery. This relationship may be rather
countries [14, 24, 27]. We also found that gender is an complex and there may be other contributing factors, con-
indicator of faster CD4+ T-lymphocyte recovery, which is sidering the extended family structure and support in Africa.
consistent with a recent Thai study that reported female Interestingly, a study from Cambodia found that children
children having a better immunologic and virologic response who were orphaned were more likely to fail therapy, and the
than males [14]. There are no reports from sub-Saharan authors attributed this to poor adherence to treatment [16].
Africa on association between immune recovery and gender Among children with complete information on parental HIV
in pediatrics, with the exception of one study that found status (67%), having both HIV-positive parents contributed
male gender to be associated with virologic failure [28]. to faster recovery than having one parent with HIV, albeit
The reasons for this gender effect are not well understood. statistically not significant (P = .13). We are inclined to infer
Interestingly, in HIV-infected adults, gender differences in that this is really a surrogate for having one versus both
treatment outcome (i.e., immunologic and virologic) have parents alive, hence the stronger observed association of
received mixed reviews [29, 30]. parental living status with the outcome in our study.
6 AIDS Research and Treatment
Table 3: Unadjusted associations of patient characteristics and time to CD4+ T-lymphocyte recovery among 233 HIV-infected children on
highly active antiretroviral therapy (HAART) from 2004–2009, Accra, Ghana.
N (%)§ Hazard Ratio P-value
Age category at ARV start .005
<1 year 10 (4.3) 0.93 .864
1–5 years 118 (51.0) 1.64 .002
>6 years 105 (45.1) 1.00
preHAART CD4+ counts (per 102) 228 (97.9) 1.05 .001
WHO immune classification at ARV start <.0001
No evidence of suppression 31 (13.3) 2.22 .0004
Evidence of moderate suppression 80 (34.3) 2.08 <.0001
Severe suppression 122 (52.4) 1.00
Gender .059
Female 112 (48.1) 1.34
Male 121 (51.9) 1.00
Mode of transmission .928
Vertical transmission 199 (85.4) 0.88 .715
Unknown 23 (9.9) 0.91 .842
Indeterminate¶ 11 (4.7) 1.00
Graduate of PMTCT program <.0001
Yes 3 (2.2) 4.77‡
No 201 (86.3) 1.00
Parental living status .054
Known both alive 109 (46.8) 1.80 .024
Known one alive 79 (33.9) 1.86 .020
Both died or unknown 31 (13.3) 1.00 —
Parental HIV status .241
Both parents with HIV 28 (12.0) 1.46 .131
One parents known with HIV 128 (54.9) 1.00 .999
Both parents unknown 75 (32.2) 1.00 —
WHO clinical staging at ARV start .197
I 32 (13.7) 0.98 .95
II 53 (22.7) 0.70 .98
III 106 (45.5) 1.09 .71
IV 41 (17.6) 1.00
Adherence self report .078
Poor 13 (5.6) 1.80 .131
Good 46 (19.8) 1.24 .607
Excellent 144 (61.8) 1.00
Unknown 30 (12.9)
Previous Tb diagnosis and treatment .803
Yes 119 (51.1) 0.15
No 114 (48.9) 1.00
§% corresponds to proportion of the 233 subjects. The sum of % and N may not be up to 100% and 233, respectively, due to subjects with “unknown”
responses.
¶Subjects with history of possible maternal transmission, blood transfusion, or sexual abuse.
‡(95% CL: 2.90, 18.36).
CD4+ T-lymphocyte count is a significant parameter to operational constraints. Despite that, recent studies in
in monitoring ART in HIV-infected individuals [6, 35– children from resource-limited countries have demonstrated
37]. In most resource-limited countries, treatment failure that immunologic and virologic outcomes of antiretroviral
is defined by immunologic criteria due to the lack of therapy is comparable to that of children in resource-rich
routine viral load monitoring. Unfortunately, not all HIV countries [11–18]. In these studies, on average the baseline
centers have access to CD4+ T-lymphocyte monitoring due CD4+ T-lymphocyte count doubled after six months on
AIDS Research and Treatment 7
Table 4: Adjusted associations of patient characteristics and time to CD4+ T-lymphocyte recovery in 233 HIV-infected children on highly
active antiretroviral therapy (HAART) from 2004–2009, Accra, Ghana.
Predictors Hazard ratio (95% CL) P-value
Parental living status .025
Both parents alive 1.98 (1.18, 3.35)
One parent known to be alive 2.00 (1.18, 3.38)
Both parents dead or unknown 1.00
Gender .051
Female 1.37 (1.00, 1.87)
Male 1.00
Age category .014
<12 months 0.98 (0.44, 2.20)
1–5 years 1.59 (1.15, 2.19)
>6 years 1.00
<12 months∗ 0.62 (0.28, 1.37)
1–5 years 1.00
WHO immune classification <.0001
No evidence of suppression 2.31 (1.48, 3.63)
Evidence of moderate suppression 2.04 (1.46, 2.86)
Severe suppression 1.00
∗
Small sample size in children <12 months.
therapy. The rate of increase remained the same over the 5. Conclusion
second 6 months and slowed thereafter. In one study, there
was no significant increase in CD4+ T-lymphocyte count Our study demonstrates that a clinically meaningful and
at 12 and 24 months [18]. This is consistent with studies robust immune recovery in HIV-infected children on
in HIV-infected adults where CD4+ T-lymphocyte recovery HAART is attainable in resource-limited countries and that
plateaued after 4 to 5 years of HAART despite complete there are patient characteristics that can predict time to
viral suppression [25, 33, 38]. From these studies, one can immune recovery. Based on these predictors, a followup
surmise that frequent (e.g., every three to four months) analysis can examine how well we can predict a time to
determination of CD4+ T-lymphocyte count may not be CD4+ T-lymphocyte recovery for an individual child so that
necessary. Our findings of predictors of CD4+ T-lymphocyte a targeted approach to CD4+ T-lymphocyte testing could be
recovery further argue for the adoption of a more targeted adopted as a viable and potentially cost-effective approach
approach or “personalized” CD4+ T-lymphocyte testing, in monitoring HAART in HIV-infected children in resource-
particularly in resource-limited countries, rather than one limited countries.
based on a general time interval. The testing interval could
be based on an individual’s pre-HAART risk of having a poor
immune recovery rate. Acknowledgments
Our study, like other retrospective studies, has inherent
limitations and also additional limitations due to operational The authors are grateful to the Pediatric AIDS Care Program
constraints of resource-limited settings. It is a single center team at Korle-Bu Teaching Hospital, and Mr. Kakra Adjei
study, and, therefore, one has to be cautious in generalizing for managing the data. They thank the children and families
our findings. Moreover, data on viral loads were not for their participation. M. Prin was supported by the
available, which precluded us from examining the effect of Doris Duke Charitable Foundation International Medical
HAART on viral load and the correlation between viral Students Fellowship. F. Y. Li and V. Northrup were supported
load and CD4+ T-lymphocytes. Our findings have to be by CTSA Grant no. UL1 RR024139 from the National
validated with effect of HAART on viral load decline, since Center for Research Resources (NCRR), a component of
it is not uncommon to have discordance in the gain of the National Institutes of Health (NIH), and NIH Roadmap
CD4+ T-lymphocytes and decline in viral load. Furthermore, for Medical Research. E. Paintsil was supported by grants
the data on adherence to therapy was by self-report and from the National Institute of Allergy and Infectious Disease
not validated, so we did not use it in the final model. (KO8AI074404). The Pediatric AIDS Care Program is sup-
However, with these caveats, our findings are consistent with ported by funding from the Korle-Bu Teaching Hospital, the
previous studies, while our proposal of targeted approach to Ministry of Health and the National AIDS Control Program
CD4+ T-lymphocyte testing is worth further investigation through the Global Fund for AIDS, TB, and Malaria. L.
and validation. Renner and M. Prin contributed equally to this work.
8 AIDS Research and Treatment
References atric Infectious Disease Journal, vol. 28, no. 6, pp. 488–492,
2009.
[1] S. M. Schnittman, H. C. Lane, J. Greenhouse, J. S. Justement, [15] N. Kumarasamy, K. K. Venkatesh, B. Devaleenol, S. Poongu-
M. Baseler, and A. S. Fauci, “Preferential infection of CD4+ lali, S. N. Moothi, and S. Solomon, “Safety, tolerability and
memory T cells by human immunodeficiency virus type 1: effectiveness of generic HAART in HIV-Infected children in
evidence for a role in the selective T-cell functional defects South India,” Journal of Tropical Pediatrics, vol. 55, no. 3, pp.
observed in infected individuals,” Proceedings of the National 155–159, 2009.
Academy of Sciences of the United States of America, vol. 87, no. [16] B. Janssens, B. Raleigh, S. Soeung et al., “Effectiveness of
16, pp. 6058–6062, 1990. highly active antiretroviral therapy in HIV-positive children:
[2] R. S. Veazey, M. DeMaria, L. V. Chalifoux et al., “Gastroin- evaluation at 12 months in a routine program in Cambodia,”
testinal tract as a major site of CD4+ T cell depletion and viral Pediatrics, vol. 120, no. 5, pp. e1134–e1140, 2007.
replication in SIV infection,” Science, vol. 280, no. 5362, pp. [17] E. George, F. Noël, G. Bois et al., “Antiretroviral therapy
427–431, 1998. for HIV-1-infected children in Haiti,” Journal of Infectious
[3] D. Vlahov, N. Graham, D. Hoover et al., “Prognostic indicators Diseases, vol. 195, no. 10, pp. 1411–1418, 2007.
for AIDS and infectious disease death in HIV- infected [18] C. Bolton-Moore, M. Mubiana-Mbewe, R. A. Cantrell et
injection drug users: plasma viral load and CD4+ cell count,” al., “Clinical outcomes and CD4 cell response in children
Journal of the American Medical Association, vol. 279, no. 1, pp. receiving antiretroviral therapy at primary health care facilities
35–40, 1998. in Zambia,” Journal of the American Medical Association, vol.
[4] P. Nishanian, J. M. G. Taylor, B. Manna et al., “Accelerated 298, no. 16, pp. 1888–1899, 2007.
changes (inflection points) in levels of serum immune acti-
[19] E. Paintsil, M. Ghebremichael, S. Romano, and W. A. Andi-
vation markers and CD4+ and CD8+ T cells prior to AIDS
man, “Absolute CD4+ T-lymphocyte count as a surrogate
onset,” Journal of Acquired Immune Deficiency Syndromes and
marker of pediatric human immunodeficiency virus disease
Human Retrovirology, vol. 18, no. 2, pp. 162–170, 1998.
progression,” Pediatric Infectious Disease Journal, vol. 27, no.
[5] G. Pantaleo, C. Graziosi, and A. S. Fauci, “Mechanisms of dis-
7, pp. 629–635, 2008.
ease: the immunopathogenesis of human immunodeficiency
[20] A. M. Wade and A. E. Ades, “Incorporating correlations
virus infection,” The New England Journal of Medicine, vol.
between measurements into the estimation of age-related
328, no. 5, pp. 327–335, 1993.
reference ranges,” Statistics in Medicine, vol. 17, no. 17, pp.
[6] J. W. Mellors, A. Muñoz, J. V. Giorgi et al., “Plasma viral
1989–2002, 1998.
load and CD4+ lymphocytes as prognostic markers of HIV-
1 infection,” Annals of Internal Medicine, vol. 126, no. 12, pp. [21] D. S. Stein, J. A. Korvick, and S. H. Vermund, “CD4+
946–954, 1997. lymphocyte cell enumeration for prediction of clinical course
[7] S. Diagbouga, C. Chazallon, M. D. Kazatchkine et al., “Suc- of human immunodeficiency virus disease: a review,” Journal
cessful implementation of a low-cost method for enumerating of Infectious Diseases, vol. 165, no. 2, pp. 352–363, 1992.
CD4+ T lymphocytes in resource-limited settings: the ANRS [22] M. W. Kline, S. Rugina, M. Ilie et al., “Long-term follow-up of
12-26 study,” AIDS, vol. 17, no. 15, pp. 2201–2208, 2003. 414 HIV-infected Romanian children and adolescents receiv-
[8] F. Mandy, G. Janossy, M. Bergeron, R. Pilon, and S. Faucher, ing lopinavir/ritonavir-containing highly active antiretroviral
“Affordable CD4+ T-cell enumeration for resource-limited therapy,” Pediatrics, vol. 119, no. 5, pp. e1116–e1120, 2007.
regions: a status report for 2008,” Cytometry B, vol. 74, [23] D. P. O’Brien, D. Sauvageot, D. Olson et al., “Treatment
supplement 1, pp. S27–S39, 2008. outcomes stratified by baseline immunological status among
[9] B. O. Taiwo and R. L. Murphy, “Clinical applications and young children receiving nonnucleoside reverse-transcriptase
availability of CD4+ T cell count testing in sub-Saharan inhibitor-based antiretroviral therapy in resource-limited set-
Africa,” Cytometry B, vol. 74, supplement 1, pp. S11–S18, 2008. tings,” Clinical Infectious Diseases, vol. 44, no. 9, pp. 1245–
[10] D. Katzenstein, M. Laga, and J. P. Moatti, “The evaluation 1248, 2007.
of the HIV/AIDS drug access initiatives in Côte D’ivoire, [24] C. H. Soh, J. M. Oleske, M. T. Brady et al., “Long-term effects
Senegal and Uganda: how access to antiretroviral treatment of protease-inhibitor-based combination therapy on CD4 T-
can become feasible in Africa,” AIDS, vol. 17, supplement 3, cell recovery in HIV-1-infected children and adolescents,”
pp. S1–S4, 2003. Lancet, vol. 362, no. 9401, pp. 2045–2051, 2003.
[11] R. Song, J. Jelagat, D. Dzombo et al., “Efficacy of highly active [25] R. D. Moore and J. C. Keruly, “CD4+ cell count 6 years
antiretroviral therapy in HIV-1-infected children in Kenya,” after commencement of highly active antiretroviral therapy
Pediatrics, vol. 120, no. 4, pp. e856–e861, 2007. in persons with sustained virologic suppression,” Clinical
[12] A. Reddi, S. C. Leeper, A. C. Grobler et al., “Preliminary Infectious Diseases, vol. 44, no. 3, pp. 441–446, 2007.
outcomes of a paediatric highly active antiretroviral therapy [26] C. G. Sutcliffe, J. H. van Dijk, C. Bolton, D. Persaud, and W.
cohort from KwaZulu-Natal, South Africa,” BMC Pediatrics, J. Moss, “Effectiveness of antiretroviral therapy among HIV-
vol. 7, p. 13, 2007. infected children in sub-Saharan Africa,” The Lancet Infectious
[13] T. Puthanakit, A. Oberdorfer, N. Akarathum et al., “Efficacy of Diseases, vol. 8, no. 8, pp. 477–489, 2008.
highly active antiretroviral therapy in HIV-infected children [27] A. S. Walker, K. Doerholt, M. Sharland, and D. M. Gibb,
participating in Thailand’s National Access to Antiretroviral “Response to highly active antiretroviral therapy varies with
Program,” Clinical Infectious Diseases, vol. 41, no. 1, pp. 100– age: the UK and Ireland Collaborative HIV Paediatric Study,”
107, 2005. AIDS, vol. 18, no. 14, pp. 1915–1924, 2004.
[14] T. Puthanakit, S. J. Kerr, J. Ananworanich, T. Bunupuradah, [28] M. R. Kamya, H. Mayanja-Kizza, A. Kambugu et al., “Predic-
P. Boonrak, and V. Sirisanthana, “Pattern and predictors of tors of long-term viral failure among ugandan children and
immunologic recovery in human immunodeficiency virus- adults treated with antiretroviral therapy,” Journal of Acquired
infected children receiving non-nucleoside reverse transcrip- Immune Deficiency Syndromes, vol. 46, no. 2, pp. 187–193,
tase inhibitor-based highly active antiretroviral therapy,” Pedi- 2007.
AIDS Research and Treatment 9
[29] E. Nicastri, C. Angeletti, L. Palmisano et al., “Gender dif-
ferences in clinical progression of HIV-1-infected individuals
during long-term highly active antiretroviral therapy,” AIDS,
vol. 19, no. 6, pp. 577–583, 2005.
[30] A. L. Moore, A. Mocroft, S. Madge et al., “Gender differences
in virologic response to treatment in an HIV-positive popu-
lation: a cohort study,” Journal of Acquired Immune Deficiency
Syndromes, vol. 26, no. 2, pp. 159–163, 2001.
[31] D. W. Notermans, N. G. Pakker, D. Hamann et al., “Immune
reconstitution after 2 years of successful potent antiretroviral
therapy in previously untreated human immunodeficiency
virus type 1-infected adults,” Journal of Infectious Diseases, vol.
180, no. 4, pp. 1050–1056, 1999.
[32] S. Staszewski, V. Miller, C. Sabin et al., “Determinants of
sustainable CD4 lymphocyte count increases in response to
antiretroviral therapy,” AIDS, vol. 13, no. 8, pp. 951–956, 1999.
[33] F. Garcı́a, E. De Lazzari, M. Plana et al., “Long-term CD4+ T-
cell response to highly active antiretroviral therapy according
to baseline CD4+ T-cell count,” Journal of Acquired Immune
Deficiency Syndromes, vol. 36, no. 2, pp. 702–713, 2004.
[34] S. Brogly, P. Williams, G. R. Seage III, J. M. Oleske, R. Van
Dyke, and K. McIntosh, “Antiretroviral treatment in pediatric
HIV infection in the United States: from clinical trials to
clinical practice,” Journal of the American Medical Association,
vol. 293, no. 18, pp. 2213–2220, 2005.
[35] R. M. Gulick, J. W. Mellors, D. Havlir et al., “Treatment with
indinavir, zidovudine, and lamivudine in adults with human
immunodeficiency virus infection and prior antiretroviral
therapy,” The New England Journal of Medicine, vol. 337, no.
11, pp. 734–739, 1997.
[36] S. M. Hammer, M. S. Saag, M. Schechter et al., “Treatment
for adult HIV infection: 2006 Recommendations of the Inter-
national AIDS Society-USA panel,” Journal of the American
Medical Association, vol. 296, no. 7, pp. 827–843, 2006.
[37] W. A. O’Brien, P. M. Hartigan, E. S. Daar, M. S. Simberkoff,
and J. D. Hamilton, “Changes in plasma HIV RNA levels
and CD4+ lymphocyte counts predict both response to
antiretroviral therapy and therapeutic failure. VA Cooperative
Study Group on AIDS,” Annals of Internal Medicine, vol. 126,
no. 12, pp. 939–945, 1997.
[38] G. R. Kaufmann, H. Furrer, B. Ledergerber et al., “Charac-
teristics, determinants, and clinical relevance of CD4+ T cell
recovery to ¡500 cells/μL in HIV type 1-infected individuals
receiving potent antiretroviral therapy,” Clinical Infectious
Diseases, vol. 41, no. 3, pp. 361–372, 2005.
View publication stats