Safety of Late In Utero Exposure to Zidovudine in Infants Born to Human Immunodeficiency Virus-Infected Mothers: Bangkok
Background. Short-course zidovudine (ZDV) given in the late antenatal period can reduce mother–infant human immunodeficiency virus (HIV) transmission by one half. Because this intervention is being implemented in developing countries, evidence of its safety is needed.
Methods. In a randomized, double-blinded, placebo-controlled trial in Bangkok, HIV-infected pregnant women received either ZDV (300 mg twice daily from 36 weeks' gestation until labor, then every 3 hours until delivery) or an identical placebo regimen. Infants were evaluated at birth and at 1, 2, 4, 6, 9, 12, 15, and 18 months of age. Growth, clinical events, and hematologic and immunologic measurements were compared between treatment groups.
Results. Of the 395 children born (196 in ZDV group and 199 in placebo group), 330 were uninfected, 55 were infected, and 10 had indeterminate infection status. Overall, 319 children (81%) completed 18 months of follow-up, and 14 (4%) died before 18 months of age. Among uninfected children, the mean hematocrit was lower in the ZDV group at birth (49.1% vs 51.5%) but not at later ages; mean weight, height, head circumference, and CD4+ and CD8+ T lymphocyte counts were similar in both groups at all ages. Five uninfected children in the ZDV group but only one in the placebo group had a febrile convulsion. No other signs suggestive of mitochondrial dysfunction and no tumors were observed. Among infected children, an estimated 62% in the ZDV group and 77% in the placebo group survived free of Centers for Disease Control and Prevention class C disease during the 18-month follow-up.
Conclusions. No significant adverse events were associated with short-course ZDV during 18 months of follow-up in this population.
The use of zidovudine (ZDV) from 36 weeks' gestation until delivery has been found to reduce the risk for mother–infant human immunodeficiency virus (HIV) transmission by 50% when used with infant formula to replace breastfeeding.1 This intervention, or modifications of it, is now being implemented throughout Thailand and in other developing countries.2,3
The US/French AIDS Clinical Trials Group (ACTG) 076 trial previously studied a longer regimen of ZDV: ∼3 months during pregnancy, intravenous treatment during labor, and 6 weeks of treatment for infants. The only short-term adverse event in children attributed to ZDV in this study was mild anemia that resolved rapidly without intervention.4,5 Moreover, after an additional 4 years of follow-up of 122 ZDV-exposed children in this trial, no additional adverse events clearly associated with ZDV treatment were identified.6 Nonetheless, concerns have been raised that prenatal ZDV exposure may cause cancer, mitochondrial toxicity, or other problems.7,8 In addition, in utero exposure to ZDV has been postulated to increase the risk for rapid disease progression among infected infants.9,10
Because of the rapid, wide-scale implementation of short-course ZDV in Thailand and elsewhere, there is an urgent need to evaluate its safety. We report the 18-month follow-up of children enrolled in the Bangkok trial, comparing growth, immunologic status, morbidity, and mortality of children born to HIV-infected women who received short-course ZDV with that of children born to women who did not.
We conducted a randomized, double-blinded, placebo-controlled trial in 2 large Bangkok hospitals to determine the safety and efficacy of a short course of late antenatal ZDV to prevent mother–infant HIV transmission. Women were enrolled at 36 weeks' gestation and randomly assigned to receive either ZDV (300 mg twice daily from 36 weeks' gestation until labor onset, then every 3 hours until delivery) or an identical placebo regimen. No study drug was given to infants, and infants were not breastfed. The details of the trial procedures and efficacy results are published.1 Both children of twin births were included in this analysis.
Children were scheduled for study visits at birth and at 1, 2, 4, 6, 9, 12, 15, and 18 months of age. At each visit, the child's weight, height, head circumference, physical examination findings, illness history, hospitalization and outpatient visit history, and medication use were recorded along with a summary of the child's diagnoses. Venous blood specimens were taken at birth (≤72 hours of life) and at 2, 6, 12, and 18 months of age for complete blood count, lymphocyte immunophenotyping, and viral load measurement, according to published methods.1
Each child's infection status was evaluated using HIV DNA polymerase chain reaction (PCR) testing performed at birth and at 2 months and 6 months of life. Children were considered uninfected if their last test at 2 months of age or older was negative, infected if any test was positive, or of indeterminate infection status if neither of these conditions was met. HIV antibody results from all 317 children tested at 18 months of age concurred with the infection status determined by PCR. HIV-1 subtype was determined by peptide enzyme immunoassay serotyping of maternal plasma using synthetic peptides derived from the consensus sequence of the crown of the V3 loop of gp120.11Nearly all infants received trimethoprim-sulfamethoxazole for prophylaxis against Pneumocystis cariniipneumonia.12 In general, antiretroviral treatment was started for children with signs of HIV infection.
Most analyses compared findings between treatment groups, after stratifying by infection status. We used the χ2or Fisher's exact test to compare categorical variables, and we used Student's t test to compare continuous variables. We used the Centers for Disease Control and Prevention classification to categorize children's HIV disease and immunosuppression status.13 We estimated survival and disease-free survival using the Kaplan-Meier method, compared survival curves using the log-rank test, and adjusted survival estimates using Cox proportional hazards regression.
Weight for age, height for age, and weight for height zscores were calculated using Epi Info, Version 6.14 We compared the mean z score and proportion with z score <−2.0 (lower 2.3% of reference population) at each time point for each treatment group. The average growth rate on each z score was summarized for each infant by a slope estimated from a linear regression, and the mean slopes for the treatment groups were compared.
Children were prospectively monitored for signs of drug-related toxicities, which were graded according to the ACTG adverse events monitoring guide.15 To further evaluate the possibility of mitochondrial toxicity related to in utero ZDV exposure,8we retrospectively reviewed all recorded physical examination findings, clinical diagnoses, hospitalizations, and causes of death among uninfected children. This review was restricted to uninfected children to avoid inclusion of confounding HIV-related conditions. After excluding infectious conditions, rashes, trauma, surgical conditions (eg, hernia), and several miscellaneous diagnoses probably not related to mitochondrial dysfunction (eg, colic and gastroesophageal reflux), we compared the frequency of the remaining diagnoses between the treatment groups. For statistical comparisons, we limited diagnoses to those reported in >5 children (convulsions, developmental delay, hyperbilirubinemia, poor weight gain, hepatomegaly, or splenomegaly).
Between June 1996 and February 1998, 395 children were born to 393 women in the trial, 196 in the ZDV group and 199 in the placebo group. Of these children, 330 were uninfected, 55 were HIV-infected, and 10 had indeterminate infection status. One set of twins was born in each treatment group; all 4 twins were uninfected. Characteristics of the mothers and children are shown in Table 1. Three hundred nineteen children (81%) completed 18 months of follow-up, 14 (4%) died before 18 months of age, and 62 (16%) were lost to follow-up. The total observation time was 3114 child months in the ZDV group and 3268 child months in the placebo group.
Congenital anomalies were noted in 8 children: 5 in the ZDV group (3 cardiac anomalies, 1 pyloric stenosis, and 1 hydrocephalus) and 3 in the placebo group (1 microcephaly, 1 pyloric stenosis, and 1 cleft palate). At birth, 2 children in the ZDV group and 1 in the placebo group had grade 3 anemia (hematocrit, <36%), and 1 infected child in the ZDV group had grade 4 granulocytopenia (<400 cells/μL); all blood abnormalities resolved within 1 week without intervention. Cancer was not diagnosed in any child.
Of the 330 uninfected children, 171 were in the ZDV group and 159 were in the placebo group. The follow-up observation times were similar in the 2 groups (Table 2). The mean hematocrit was lower in the ZDV group at birth (49.1% vs 51.5%;P < .001) but not at later ages. The mean weight, height, and head circumferences were similar between groups at all study visits (Fig 1A). There was also no statistically significant difference in the proportion of children who were <2 standard deviations for weight or height at any follow-up visit and there was no difference in mean growth velocity overall (data not shown). No significant differences were seen in CD4+ or CD8+ counts or percents between treatment groups (Fig 1B). Overall, 2 uninfected children (.6%) died, 1 presumably of sepsis at 4 months of age (ZDV group) and the other of fever and respiratory distress at 17 months of age (placebo group). Both children died outside study hospitals and neither had an autopsy. There was one death among the 330 uninfected children (3 deaths/1000 live births).
No evidence of cardiomyopathy, encephalopathy, significant developmental delay, or other diagnoses suggestive of mitochondrial dysfunction was found in any uninfected child by clinical history or routine examination. Thirteen children (6 in ZDV group and 7 in placebo group) had developmental delays noted on physical examination, all of which were mild delays in reaching age-appropriate gross motor function. Six children had convulsions; 5 in the ZDV group and 1 in the placebo group (P = .21). One child, who was born to a mother who took ZDV for 46 days, had 2 episodes of convulsions. At birth, the child had low birth weight (2400 g), asphyxia, hypoglycemia, hypocalcemia, and possibly sepsis. On the first day of life, the child had a convulsion that resolved after treatment with glucose, calcium, antibiotics, and anticonvulsants. At 12 months of age, the child had a simple febrile convulsion. The child's development was normal; no other clinical problems were noted. Another child in the ZDV group with convulsions was the child who died at 4 months of age. This child had a convulsion associated with high fever during the hospitalization in which the child died, presumably of sepsis. The other 4 children with convulsions had single episodes of simple febrile convulsions.
Twenty uninfected children had poor weight gain noted during outpatient visits, 14 in the ZDV group and 6 in the placebo group (P = .09). There was no significant difference between groups in the frequency of neonatal hyperbilirubinemia (8 in ZDV group and 7 in placebo group) or hepatosplenomegaly (2 in ZDV group and 6 in placebo group). Fewer uninfected children in the ZDV group than in the placebo group had lymphadenopathy (0 vs 7; P = .005).
Of the 55 infected children, 18 were in the ZDV group and 37 were in the placebo group. Peptide serotyping performed on specimens from 54 of the mothers showed 47 infections (87%) with subtype E and 6 (11%) with subtype B; 1 infection could not be typed. Twenty-two children (40%) had positive HIV PCR test results in the first 72 hours of life; 9 (50%) in the ZDV group and 13 (35%) in the placebo group (P = .30).
There was no significant difference between the 2 treatment groups in mean weight, height, CD4 count, CD8 count, viral load, or proportion who were hospitalized or who started on antiretroviral therapy (Table 3). Seven children developed class C HIV disease during the first 18 months of follow-up; 2 (11%) in the ZDV group and 5 (14%) in the placebo group. Seventeen children developed severe (class 3) immunosuppression; 5 (28%) in the ZDV group and 12 (32%) in the placebo group. Ten children died; 5 (28%) in the ZDV group and 5 (14%) in the placebo group (Fig 2). The causes of death and status before death of the 10 children who died are shown in Table 4.
By Kaplan-Meier analysis, the estimated proportions of infected children surviving to 12 and 18 months of age for the ZDV group were 81% and 68%, respectively, and for the placebo group were 89% and 86%, respectively (P = .28). The estimated proportions remaining alive and free of class C disease at 12 and 18 months of age were 75% and 62% in the ZDV group and 83% and 77% in the placebo group (P = .49). Because 18-month disease-free survival differed somewhat by result of the birth PCR test (63% if positive and 77% if negative; P = .30), we adjusted the survival estimates for the early PCR results. After adjustment, the difference remained nonsignificant (P = .49).
Children With Indeterminate Infection Status
Of 10 children with indeterminate infection status, 7 were in the ZDV group and 3 were in the placebo group. Two died: 1 in the ZDV group (no HIV PCR performed) of multiple cardiac anomalies on the day of birth, and 1 in the placebo group (negative HIV PCR at birth; missed 2-month visit) of pneumonia at 3 months of age.
Our trial previously demonstrated that a short course of antenatal ZDV can significantly reduce the risk for mother–infant HIV transmission, thus prompting the widespread use of short-course ZDV in Thailand and other developing countries.1–3 Eighteen months of follow-up of children now shows no major adverse events associated with short-course ZDV.
Our trial design, which included a concurrent, randomly assigned control group, allowed us to compare children who were exposed to short-course ZDV with a control group of children who were not. The only statistically significant difference in adverse events observed between groups was in the mean hematocrit, which was lower in the ZDV group at birth only, but was not clinically significant. Stratified by infection status, other indicators—death, growth, immunologic status, and number of hospitalizations—were similar between the 2 groups.
Several concerns have been raised about the safety of ZDV treatment during pregnancy. First, as with all drugs taken during pregnancy, the potential for teratogenesis must be considered; however, the incidence of birth defects in ZDV-exposed children does not seem to be increased,5,16 and the ZDV regimen in our trial began after completion of organogenesis. Second, ZDV can be integrated into infant DNA, and animal data suggest the possibility that in utero ZDV exposure can cause tumors.17,18 However, follow-up of >700 HIV-exposed infants in the United States has failed to identify any tumors in the first few years of life.6,19 Third, the potential for mitochondrial toxicity in infants exposed to nucleoside analog antiretroviral medications has been raised recently.8 Review of several hundred deaths of ZDV-exposed children in the United States, however, has not identified deaths that seem to be related to mitochondrial dysfunction.20,21Fourth, ZDV exposure can cause anemia,4,5 and a direct effect of ZDV on T cells has been postulated.22,23Finally, although other unidentified effects could affect the growth and development of ZDV-exposed children, no such effects were seen after 4.5 years of follow-up of children in the ACTG 076 study.6
Because of concerns raised about the potential effect on mitochondria of prenatal ZDV exposure,8 we specifically searched for clinical findings that might suggest mitochondrial dysfunction. Our retrospective review of clinical diagnoses found no evidence of significant cardiomyopathy, encephalopathy, or other organ dysfunction. The incidence of some less severe conditions differed between the groups. For instance, febrile convulsions and reported poor weight gain were more common in the ZDV group; lymphadenopathy and hepatosplenomegaly were more common in the placebo group. The importance of these nonspecific findings is not clear. Although it is possible that they might indicate a subtle effect of ZDV exposure, it is also possible that these findings were attributable to chance alone.
Two reports from uncontrolled observational studies have suggested that children who became infected despite maternal ZDV treatment may have faster disease progression than infected children not exposed to ZDV in utero. The Italian Registry for HIV Infection in Children reported a significantly higher probability of death, severe HIV disease, or severe immunosuppression at 3 years of age among 38 ZDV-exposed children compared with 178 unexposed children.9 A US study of 325 infected children (82 ZDV-exposed and 243 unexposed) reported that maternal ZDV use was associated with a 1.8-fold increase in risk for AIDS or death within 1 year.10 In contrast, no increased rate of progression was seen among the 57 infected children (14 ZDV-exposed and 43 unexposed) born in the ACTG 076 study.24
In our study, we found a slightly higher risk for disease progression among ZDV-exposed, HIV-infected children during the 18-month follow-up period, although this difference was not statistically significant. Other outcomes related to HIV disease progression—CD4 count, viral load, and frequency of starting antiretroviral treatment—were not significantly different at the end of the 18-month follow-up.
Although our study is greatly strengthened by its use of randomization and high rates of follow-up, it has some limitations. First, our sample size was chosen to determine the efficacy of short-course ZDV and to identify adverse events that occur with high frequency; it was not large enough to detect an increased rate of adverse events that occur rarely. Also, the number of infected children is too small and follow-up too short to draw conclusions about disease progression related to ZDV exposure. Second, our review of clinical conditions potentially associated with mitochondrial dysfunction was retrospective and could not evaluate the incidence of subtle clinical findings or laboratory abnormalities that might suggest mitochondrial dysfunction. Third, as in other studies to date, our follow-up period of 18 months is too short to enable us to evaluate the incidence of conditions, such as cancer, that may take many years to develop.
Whether adverse events will appear with more widespread use of this regimen or after many years of follow-up remains unknown, and monitoring for late-appearing adverse events will be challenging. Nonetheless, our data suggest that short-course ZDV is safe for infants and that the incidence of adverse effects does not seem, so far, to outweigh the benefits of this regimen in preventing HIV infection in children, particularly in Thailand.
Other members of the collaborative study group, not listed as coauthors: Faculty of Medicine Siriraj Hospital, Department of OB-GYN, S. Neungton, P. Chaisilwattana, A. Roongpisuthipong, A. Chalermchokcharoenkit, K. Sirimai, P. Phopong, C. Bhadrakom, P. Chaiyakul, P. Rattananikhom, and R. Prechanont; Faculty of Medicine Siriraj Hospital, Department of Pediatrics, M. Tuchinda, S. Pichitchaichan, W. Boonyavit, and S. Chearskul; Faculty of Medicine Siriraj Hospital, Department of Microbiology, C. Wasi; Rajavithi Hospital Department of OB-GYN, P. Chinayon, W. Siriwasin, S. Asavapiriyanont, B. In-neam, S. Supatosa, C. Kannasot, S. Sangkasuwan, S. Leampojara, and P. Pramukhakul; Rajavithi Hospital Laboratory, S. Singhanati and G. Kaewchaiyo; Rajavithi Hospital Department of Nursing, J. Sawakwan and N. Montasewee; Queen Sirikit National Institute for Child Health, S. Horpaopan, V. Sangtaweesin, P. Na Chiengmai, R. Kulvisuthpravit, B. Phasukdee, and P. Sojirat; and The HIV/AIDS Collaboration, K. Limpakarnjanarat, W. Supapol, A. Bennetts, N. Chantharojwong, T. Naiwatanakul, J. Laosakkitiboran, P. Yuentrakul, C. Manopaiboon, A. Teeraratkul, and N. Skunodom.
We thank the dedicated field work of the project study nurses and social workers: K. Neeyapun and B. Jetsawang (team leaders); S. Bhengsri, S. Henchaichon, S. Jalanchavanapate, K. Klumthanom, R. Krajangthong, C. Prasert, W. Sanyanusin, W. Suwannapha, S. Sorapipatana, S. Suwanmaitre, W. Triphanitchkul, and C. Yuvasevee.
We also thank T. Chaowanachan and the HIV/AIDS Collaboration laboratory staff for laboratory specimen processing and testing, and the study participants and their families for volunteering for this study.
- Received April 28, 2000.
- Accepted July 10, 2000.
Reprint requests to (R.J.S.) HIV/AIDS Collaboration, DMS 6 Building, Ministry of Public Health, Tivanon Rd, Nonthaburi, 11000, Thailand. E-mail:
- ZDV =
- zidovudine •
- HIV =
- human immunodeficiency virus •
- ACTG =
- AIDS Clinical Trials Group •
- PCR =
- polymerase chain reaction
- Thaineua V,
- Sirinirund P,
- Tanbanjong A,
- Lallemant M,
- Soucat A,
- Lamboray JL
- Centers for Disease Control and Prevention. Public Health Service task force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. MMWR Morb Mortal Wkly Rep. 1998;47(RR-2):1–30
- Kuhn L,
- Abrams EJ,
- Weedon J,
- et al.
- 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(RR-12):1–10
- Dean AG, Dean JA, Coulombier D, et al. Epi Info, Version 6: A Word Processing, Database, and Statistics Program for Public Health on IBM-Compatible Microcomputers. Atlanta, GA: Centers for Disease Control and Prevention; 1995
- National Institute of Allergy and Infectious Diseases, Division of AIDS Regulatory Operations Center (US). Serious Adverse Experiences Reporting Manual. Rockville, MD: National Institute of Allergy and Infectious Diseases; 1998
- Garcia PM, Watts DH, Fox HE, et al. Assessing the teratogenic potential of antiretroviral drugs: data from the Antiretroviral Pregnancy Registry. Proceedings of the Seventh National Conference on Retroviruses and Opportunistic Infections; January 31–February 2, 2000; San Francisco, CA. Abstract 68
- Olivero OA,
- Anderson LM,
- Diwan BA,
- et al.
- Smith ME, the US Nucleoside Safety Review Working Group. Ongoing nucleoside safety review of HIV exposed children in US studies. Proceedings of the Second Conference on Global Strategies for the Prevention of HIV Transmission From Mothers to Infants; September 1–6, 1999; Montreal, Canada. Abstract 096
- Hanson C, Frederick M, McIntosh K. Evaluation of living uninfected children for mitochondrial defects: Women and Infants Transmission Study. Proceedings of the Seventh National Conference on Retroviruses and Opportunistic Infections; January 31–February 2, 2000; San Francisco, CA. Abstract 665
- Copyright © 2001 American Academy of Pediatrics