Introduction
Several antiretroviral regimens have been shown to reduce mother-to-child transmission (MTCT) of HIV. In developing countries the most practical regimen is the use of a single oral dose of nevirapine (NVP) to the mother at the start of labor and to the infant shortly after birth. This regimen reduced MTCT of HIV by about 50% during the first 14-16 weeks of life in a breastfeeding population in Uganda [1]. Simplified regimens of zidovudine (ZDV) were also shown to be effective among breastfed infants in Africa [2-4].
In addition to measuring efficacy, monitoring the safety of these drugs in different populations is important. NVP is a non-nucleoside HIV reverse transcriptase inhibitor. It has some desirable properties such as potent antiviral activity, prolonged half-life, rapid absorption when given orally, quick passage through the placenta and ability to cross into breastmilk [1,5-6]. ZDV is a potent nucleoside analog HIV reverse transcriptase inhibitor with proven antiviral activity against HIV-1. In the newborn there is a rapid increase in ZDV clearance during the first weeks of life, reaching adult levels by 4-8 weeks [7]. The main laboratory abnormalities associated with ZDV are hematological changes [8] whereas with NVP, elevations in liver enzymes and hematological changes have been reported [9]. These adverse events are most common with long-term therapy. Information on the effect of short-term use of these drugs at time of birth is limited. Earlier neonatal hematological studies from sub-Saharan Africa predated the use of antiretrovirals and focused mainly on the effect of environmental and genetic factors [10-11].
In Malawi, southeast Africa, prevalence of HIV is about 30% among pregnant women [12] and the rate of MTCT of HIV is 28% [13,14]. HIV infection of the infant substantially contributes to both morbidity and mortality [15,16]. Therefore, affordable and practical measures to reduce MTCT of HIV are urgently needed. We conducted this safety study to determine levels of the liver enzyme alanine aminotransferase (ALT, formerly SGPT) and measure hematological parameters at birth and again at 6 weeks of age among infants born to HIV-positive mothers randomized to receive a neonatal post-exposure prophylaxis of NVP alone or ZDV in addition to NVP.
Methods
A MTCT of HIV prevention randomized controlled clinical trial is currently on-going in Blantyre, Malawi. Children born to two groups of women, early presenters and late presenters, are being enrolled. Early presenters are women who arrive the labor room 'early' (approximately 4 h before delivery), and therefore there is time to counsel, test for HIV and administer the drug NVP to the mother (200 mg single tablet) before delivery. Late presenters are women who arrive the labor room too late (within 4 h of estimated time of delivery) to be counseled, tested and provided NVP before delivery. The infants of both groups of women are randomized to receive orally (as soon as the infant is able to swallow fluids) either NVP alone (2 mg/kg weight single dose) or NVP plus ZDV (4 mg/kg weight orally twice daily for 1 week). Preterm/low birth weight or anemic (< 10 g/dl hemoglobin) infants are excluded from enrollment in the trial.
As part of this study and during the period May 2000 to July 2001, blood samples to measure levels of ALT and complete blood count (CBC) were collected at birth and again at 6 weeks of age. At birth capillary blood was collected by a heel prick. Collection of capillary blood, instead of cord blood, was necessary to make collection procedures consistent with other specimens which were required to test infants' HIV status at birth. Whole blood was collected using EDTA vacutainer tubes. In situations where it was not possible to obtain adequate blood using regular EDTA tubes (especially at birth), RAM microcollection tubes (RAM Scientific, Needham, Massachusetts, USA) or EDTA microcapillary tubes were used. Blood was allowed to flow freely from a heel prick (avoiding any squeezing). Specimens were collected at birth before the infant was administered any drug. ALT measurements on approximately 20 μl serum were performed using an ASCA AGII Chemistry System and ASCA liquid reagents (Landmark Scientific Inc, Greensboro, North Carolina, USA). For CBC measurements, whole blood was analyzed using a Coulter ACT Diff Hematology Analyzer (Coulter Corporation, Miami, Florida, USA). All tests were performed on fresh blood samples at the local laboratory in Malawi. At birth, infants were screened to exclude those who were anemic (hemoglobin < 10g/dl) from the intervention study by a quick method using a B-Hemoglobin Photometer (Hemacue AB, Angelholm, Sweden). The HIV status of the infant was determined at 6 weeks on dried blood spots using RNA PCR (Organon-Teknika Nuclisens assay, BioMerieux, Durham, North Carolina, USA).
Measurement of ALT at birth and at 6 weeks was performed on approximately the first 200 infants who were consecutively enrolled and randomized at the onset of the trial. Due to difficulties of collecting blood samples at birth, more infants were included at 6 weeks than at birth. As a comparison group (controls), we collected blood samples for measurement of ALT and CBC on about 200 infants born to HIV uninfected mothers. These control infants were consecutively screened and enrolled at the same clinics where the main trial was conducted. All infants were full-term and with normal birth weight. Due to ethical concerns, we were not able to enrol as a control group a sample of HIV-exposed infants who were not on any HIV antiviral therapy at the clinics where our trial is being conducted. At present, measurements of ALT and CBC on all control children have been completed. Measurements of ALT on infants participating in the trial have also been completed; CBC measurements on these infants will continue to the end of the trial. For this report, results of CBC on children enrolled in the on-going trial to mid July 2001 are included.
We present separately descriptive statistics for control infants and for infants randomized to receive NVP alone or NVP plus ZDV. The analyses are also stratified by whether the mother had received intrapartum NVP (early presenters) or not received intrapartum NVP (late presenters). To take account of skewness of the data, ALT measures were log10 transformed; the mean and standard deviation (SD) of the transformed data are presented. We also present the geometric mean (antilog of the mean of log10 transformed data). ALT and hematological values which were statistically significant were stratified by the HIV status of the infant at 6 weeks of age. We conducted analysis of variance (ANOVA) to test if there were statistically significant differences (P < 0.05) in the means of the parameters of interest among all groups (one control and four treatment groups) and only among the groups that received antiviral treatment (four groups). (ALT and hematological changes between birth and 6 weeks of age in each group were also examined; there were similar trends and the data are not presented here.)
To estimate severity, we graded ALT and hematological data from this study based on tables for Pediatric Adverse Experiences developed by the National Institutes of Health Division of AIDS (DAIDS) HIV Vaccine and Prevention Research Programs. In these tables, grades of ALT are defined as the upper limit of normal (ULN) multiplied by a specified range. For example, ALT grades are defined as ULN multiplied by the following ranges: grade 1, ULN × 1.1-4.9 U/l; grade 2, ULN × 5.0-9.9 U/l; grade 3, ULN × 10.0-15.0 U/l; and grade 4, ULN × > 15.0 U/l. For hemoglobin the ranges are as follows: grade 1, 8.5-9.4 g/dl; grade 2, 7.0-8.4 g/dl; grade 3, < 7.0 g/dl; and grade 4 is cardiac failure secondary to anemia.
Informed consent was obtained from mothers of all children included in this study. The study was approved by Institutional Review Boards in both Malawi (College of Medicine Research Committee) and the USA (Johns Hopkins University Bloomberg School of Public Health Committee on Human Research and the Department of Health and Human Services Office for Protection from Research Risks).
Results
Table 1 shows ALT values at birth; the geometric means ranged from 15.5 to 18.6 U/l among the five groups of infants (one control and four treated groups). There were no statistically significant differences in the means of ALT of these five groups of children. There were also no significant differences in means of the four groups of children at birth according to whether the mother had received or not received NVP intrapartum. ALT values at 6 weeks of age are shown in Table 2 these are the values after the infant has been dosed with NVP at birth and has completed a regimen of ZDV syrup, if randomized to receive this drug. The geometric mean values were 11.5 U/l for the control group and 16.2-19.1 U/l for the four treated groups. There were statistically significant differences among the five groups of children at 6 weeks of age (P < 0.0001): mean ALT values were higher among children who received treatment compared with control children. In addition, compared with mean ALT values at birth, the mean ALT values declined among the control group but increased in three of the four treated groups. However, among the four groups of children who received antiviral treatment there were no statistically significant differences in mean value of ALT at 6 weeks of age. Based on conventional tables currently used to assess severity of this parameter (see Methods), the increases in ALT values were mild (grade 1 only).
Tables 3 and 4 compare hematological measures, respectively, at birth and at 6 weeks of age among the control and the treated groups of infants. At birth, mean hemoglobin and hematocrit values ranged, respectively, from 18.9 to 19.5 g/dl and from 56.7 to 60.0% among the five groups of children with no statistical differences. There were no differences in white blood cell counts between any of the groups of children at birth. However, among control infants compared with those born to HIV-infected mothers, granulocyte percentages were higher (51.1% versus 53.9-56.1%;P = 0.05) and lymphocyte percentages were lower (40.7% versus 35.4-36.7%;P = 0.01) (Table 3). The differences in granulocyte and lymphocyte percentages among the four groups of infants born to HIV-positive mothers were not significantly different. At birth, platelet counts were comparable among all groups.
There were declines in hematological values between birth and 6 weeks of age (with the exception of platelets which increased slightly) (Tables 3 and 4). These changes were observed among control children as well as among treated groups of children. At 6 weeks of age, both hemoglobin and hematocrit values were significantly lower among antiviral drug-treated groups of infants. The hemoglobin mean value was 11.7 g/dl in the control group versus 10.4-10.9 g/dl among treated groups;P < 0.0001; hematocrit mean value was 34.2% in the control group versus 30.7-31.8% for the treated groups;P < 0.0001 (Table 4).
There were no significant differences among the four treated groups in all the hematological parameters. Although there were no differences in white blood cell count at 6 weeks among all groups of children, there were statistically significant differences in percentages of granulocytes (mean 31.6% among controls and 24.1-26.0% among treated groups;P < 0.0001) and lymphocytes (mean 58.5% among controls and 65.7-66.7% among treated groups;P < 0.0001). At 6 weeks of age, platelet counts were significantly lower among the treated groups of children compared with control children (mean 294.8-311.6 × 109/l for treated groups versus 366.1 × 109/l for the control group;P < 0.0001). The mean platelet counts among the four treated groups were not significantly different.
The hemoglobin values at 6 weeks were stratified by toxicity grade levels based on the DAIDS tables (see Methods). Overall there were no significant differences in the proportions of infants who were in grades 1-3 among the five groups of children (P = 0.50; χ2 test). In the control group, 82% of 133 infants had a normal hemoglobin level (≥ 9.5 g/dl) compared with 70-78% among the treated groups (75% of 109 infants born to mothers who received NVP and infant had received ZDV plus NVP; 77% of 120 infants born to mothers who received NVP and infant had received NVP only; 78% of 208 infants born to mothers who did not receive NVP and infant had received ZDV plus NVP; and 70% of 206 infants born to mothers who did not receive NVP and infant had received NVP only). More treated infants showed grades 1 and 2 severity compared with control infants: 12.8-15.1% among treated groups of infants compared with 9.8% among control infants had grade 1 severity; 5.3-9.7% among treated groups of infants compared with 3.8% among control infants had grade 2 severity. The proportion of infants with the lowest hemoglobin values (< 7.0 g/dl; grade 3) was comparable among the control group (4.5%) and treated groups (1.7-5.3%). No infant was admitted to the hospital with cardiac failure due to anemia (i.e., severity of grade 4).
To determine if changes in ALT or hematological measurements could be due to HIV infection of the infant, selected results are presented in Table 5 stratified by HIV infection status of the infant at 6 weeks of age. ALT measurements were highest when the infant was infected and the mother had received intrapartum NVP (geometric mean of 20.8 U/l) and lowest when the infant was not infected and mother had not received intrapartum NVP (geometric mean of 17.0 U/l). ALT values were intermediate when the infant was infected and the mother did not receive NVP (geometric mean of 18.5 U/l) or when the infant was not infected and mother had received intraprtum NVP (geometric mean of 17.4 U/l). These differences, however, were not statistically significant.
The hematological results showed trends similar to that of differences in ALT measurements (Table 5). For example, hemoglobin, hematocrit and platelet values were lowest among HIV-infected infants whose mothers received intrapartum NVP and highest among HIV-uninfected infants whose mothers did not receive intrapartum NVP. Lymphocyte values were highest among HIV-infected infants whose mothers received intrapartum NVP and lowest among HIV-uninfected infants whose mothers did not receive intrapartum NVP. Granulocyte geometric means were lowest if maternal NVP was received intrapartum irrespective of the infant's HIV status. Stratifying further the results in Table 5 by infants' treatment arm did not reveal clear trends or statistically significant differences.
Discussion
With the successes of short course neonatal antiretroviral regimens in reducing MTCT of HIV in breastfeeding populations in sub-Saharan Africa, major policy issues are now being discussed to make these drugs available and to support other programmatic issues such as counseling and testing [17]. A wider application of antiretroviral interventions, however, requires assurances that these drugs are safe and will not cause harm because clinical and laboratory infrastructure to monitor drug toxicity is inadequate in most developing countries.
High levels of the enzyme ALT are suggestive of acute liver damage due to exposure to hepatotoxic agents. At birth ALT levels are high [up to double the adult ULN (35 U/l)], and subsequently decline to adult levels at approximately 3 months of age [18]. A commonly cited upper level of reference value for infants is 54 U/l [19]. In our study, there were no significant differences in ALT measurements among the groups of children at birth. ALT mean values of the four groups of children born to HIV-infected mothers were not different either. Because early presenting mothers received intrapartum NVP and late presenting mothers did not receive NVP, this finding suggests that administering a single dose of NVP (200 mg tablet) to the mother during labor does not increase ALT levels of the neonate at birth.
Between birth and 6 weeks of age, however, and consistent with maturation of the liver, there was a decline in ALT level among the control infants (from 18.2 U/l to 11.5 U/l). In contrast, there were elevations in ALT levels in three of the four treated groups (Table 2). Because all treated children received some treatment (either single dose NVP alone or single dose NVP combined with ZDV for 1 week), our data suggest that short-term antiretroviral drugs could influence levels of ALT. Stratifying these findings by HIV status of the infant (Table 5), suggests that these increases could be due in part to immunological changes which accompany HIV infection because levels of ALT were highest among infected children. As we explained in Methods, we were not able to recruit an HIV-exposed group of children who did not receive antiretroviral treatment in our setting to investigate these differences further. More reassuring is that these hepatic changes were minimal and none of the infants developed a toxicity more severe than grade 1. Therefore these changes appear to have no clinical significance.
Similar to ALT results, the hematological values at birth (with the exception of granulocytes and lymphocytes) suggested that an intrapartum single dose of NVP did not influence neonatal hematological parameters because infants born to mothers who received or did not receive NVP (including controls) had similar results (Table 3). At 6 weeks of age, there were significant differences in several hematological indices. For example, compared with controls, hemoglobin, hematocrit, granulocyte, and platelet values decreased, and lymphocyte values increased (Table 4). This is not an unexpected finding, although in this study duration of use was brief, only 1 week for ZDV and a single dose for NVP. In the 076 ZDV trial reported hemoglobin concentrations at birth and 6 weeks of age were significantly lower among children in the ZDV group than children in the placebo group [8,20]. Similar observations were reported in other studies that used ZDV [21]. Mild hematologic changes with ZDV and NVP use have also been reported among Ugandan infants who received either ZDV for 1 week or NVP as a single dose in the 012 trial; comparable rates of anemia, neutropenia and thrombocytopenia were observed [1]. To-date, follow-up of infants who have received short-term perinatal antiretrovirals in most studies in the USA and other countries suggests that these neonatal hematological changes are transient and have no major clinical implications to warrant change of policy [8,22].
In this study, the proportion of infants at 6 weeks with the lowest hemoglobin values (< 7.0 g/dl) were comparable between control (4.5%) and treated groups (1.7-5.3%). In addition, the hematological changes were more noticeable among HIV-infected children (Table 5), with the exception of granulocyte percentage which was lowest among infants whose mothers had received intrapartum NVP. These data suggest that infection with HIV also contributes to these changes. Distinguishing between the adverse effects of short-term antiretroviral drugs and HIV infection of the infant could be difficult. This is a potential limitation of this study. No reference values are available for HIV-exposed but untreated infants and we were not able to recruit a comparison group for ethical reasons. Our data did not show significant differences among the four treated groups of children either at birth or at 6 weeks of age (Tables 3 and 4). Therefore, it is not possible to distinguish between effects due to NVP alone or to its combination with ZDV. This is further compounded by the fact that both NVP and ZDV have been reported to be associated with hematologic abnormalities [8,9].
It should be noted that despite potential differences in methodology and sample collection, there were similarities in hematological and ALT values between Malawian and Caucasian children. For example, at birth several hematological parameters among the control and treated groups of Malawian children were comparable to pediatric values reported from developed countries in hematology textbooks [23]. Of interest also, the median hemoglobin concentration at 6 weeks of age among infants who received ZDV in the 076 trial and in this study in Malawi were similar: 10.0 g/dl among children in the ZDV arm in the 076 trial, and 10.4-10.7 g/dl among the four treated groups of Malawian children. The upper limit of ALT values at 6 weeks of age among the Malawian control infants (range 3.0-50.1 U/l) was similar to reference values from developed countries (54 U/l). These similarities may imply that at this age there are no intrinsic differences between African and Caucasian infants; this is consistent with observations from previous hematological studies in Africa [11].
We have monitored some important safety parameters among Malawian infants. Although hepatic and hematological changes were observed with these ultra-short post-exposure prophylaxis antiretroviral regimens, these alterations were minimal and within expected limits based on previous studies. Assuming that most HIV infections of the neonate occur intrapartum or close to the time of delivery, it appears that both acquisition of HIV infection and antiretroviral drugs contribute to hepatic and hematological changes in the neonate. We will continue to monitor the clinical conditions and hematological parameters of these infants up to the age of 2 years. These results could guide the implementation of antiretroviral treatment programs in similar pediatric populations.
Acknowledgements
We thank the Ministry of Health of Malawi, the Malawi College of Medicine, and the staff of the Johns Hopkins University-Ministry of Health-College of Medicine Research Project in Blantyre, Malawi for their active and dedicated collaboration.
References
1. Guay LA, Musoke P, Fleming T. et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999, 354: 795-802.
2. Wiktor SZ, Ekpini E, Karon JM. et al. Short-course oral zidovudine for prevention of mother-to-child transmission of HIV-1 in Abidjan, Cote d'Ivoire: a randomized trial. Lancet 1999, 353: 781-785.
3. Dabis F, Msellati P, Meda N. et al. 6-month efficacy, tolerance, and acceptability of a short regimen of oral zidovudine to reduce vertical transmission of HIV in breastfed children in Cote d'Ivoire and Burkina Faso: a double-blind placebo-controlled multicentre trial. Lancet 1999, 353: 786-792.
4. Saba J. The results of the PETRA intervention trial to prevent perinatal transmission in Sub-Saharan Africa. Sixth International Conference on Retroviruses and Opportunistic Infections. Chicago, January 1999 [abstract S7].
5. Mirochnick M, Fenton T, Gagnier P. et al. Pharmacokinetics of nevirapine in human immunodeficiency virus type-1 infected pregnant women and their neonates. J Infect Dis 1998, 178: 368-374.
6. Musoke P, Guay LA, Bagenda D. et al. A phase I study of the safety and pharmacokinetics of nevirapine in HIV-1 infected pregnant women and their neonates. AIDS 1999, 13: 479-486.
7. Mirochnick M, Capparelli E, Connor J. Zidovudine pharmacokinetics in infants: a population analysis across studies. Clin Pharmacol Ther 1999, 66: 16-24.
8. Sperling RS, Shapiro DE, McSherry GD. et al. Safety of the maternal-infant zidovudine regimen utilized in the pediatric AIDS Clinical Trial Group 076 Study. AIDS 1998, 12: 1805-1813.
9. Bardsley-Elliot A, Perry CM. Nevirapine. A review of its use in the prevention and treatment of paediatric HIV infection. Paediatric Drugs 2000, 2: 373-407.
10. Scott-Emuakpor AB, Okolo AA, Omene JA, Ukpe SI. Normal hematological values of the African neonate. Blut 1985, 51: 11-18.
11. Mukiibi JM, Mtimavalye LAR, Broadhead RL. et al. Some hematological parameters in Malawian neonates. East Afr Med J 1995, 72: 10-14.
12. Taha TE, Dallabetta GA, Hoover DR. et al. Trends of HIV-1 and sexually transmitted diseases among pregnant and postpartum women in urban Malawi. AIDS 1998, 12: 197-203.
13. Biggar RJ, Miotti PG, Taha TE. et al. Perinatal intervention trial in Africa: effect of a birth canal cleansing intervention to prevent HIV transmission. Lancet 1996, 347: 1647-1650.
14. Semba RD, Kumwenda NI, Hoover R. et al. Human immunodeficiency virus load in breastmilk, mastitis, and mother-to-child transmission of human immunodeficiency virus type 1. J Infect Dis 1999, 180: 93-98.
16. Taha TE, Kumwenda NI, Broadhead RL. et al. Mortality after the first year of life among human immunodeficiency virus type 1-infected and uninfected children. Pediatr Infect Dis J 1999, 18: 689-694.
17. De Cock KM, Fowler MG, Mercier E. et al. Prevention of mother-to-child HIV transmission in resource-poor countries. Translating research into policy and practice. JAMA 2000, 283: 1175-1182.
18. Murray RL. Alanine aminotransferase. In Clinical Chemistry - Theory, Analysis, and Correlation. Edited by Kaplan LS, Pesce AJ. St. Louis: C.V. Mosby; 1984: 10901090.
19. Hicks JM, Boeckx RL (eds) Pediatric Clinical Chemistry. Philadelphia: W.B. Saunders; 1984: 283-284.
20. Connor EM, Sperling RS, Gelber R. et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994, 331: 1173-1180.
21. Lallemant M, Jordain G, Le Coeur S. et al. A trial of shortened zidovudine regimens to prevent mother-to-child transmission of human immunodeficiency virus type 1. N Engl J Med 2000, 343: 982-991.
22. Chotpitayasunondh T, Vanprapar N, Simonds RJ,
et al. Safety of late in utero exposure to zidovudine in infants born to human immunodeficiency virus-infected mothers: Bangkok. Pediatrics 2001,
107 (1). URL:http//
http://www.pediatrics.org/cgi/content/full/107/1/e5. 23. Brugnara C, Platt OS. In Nathan and Oski's Hematology of Infancy and Childhood, 5th edn. Edited by Nathan DG, Orkin SH. Philadelphia: W.B. Saunders; 1998: 19-46.
© 2002 Lippincott Williams & Wilkins, Inc.