|Home | About | Journals | Submit | Contact Us | Français|
An increasing number of HIV-infected women become pregnant while receiving efavirenz (EFV). We compared the pregnancy outcomes of women exposed to EFV and to nevirapine (NVP) during the first trimester.
A retrospective study in four HIV care centers participating to clinical trials and international cohort collaboration. All HIV-infected pregnant women who conceived on EFV or NVP-based antiretroviral therapy (ART) between 2003 and 2009 were included. Pregnancy outcomes were: abortion (voluntary termination), miscarriage (unwanted termination <20 weeks of amenorrhea [WA]), stillborn (death ≥20 WA), preterm delivery [PTD] (live-birth <37 WA) and low birth weight [LBW] (<2,500 grams).
Overall, 344 HIV-infected pregnant women conceived on ART (213 on EFV and 131 on NVP). Median age was 29 years and median CD4 count 217 cells/μl at ART initiation. The overall proportion was 11.7% for abortion, 5.2% for miscarriage, 6.7% for stillborn, 10.8% for PTD and 20.2% for LBW. There was no difference between EFV and NVP exposure, except for abortion (14.3% vs 7.3%; p=0.05). No external and visible congenital malformation was observed neither in women exposed to EFV nor in women exposed to NVP.
Among women exposed to EFV, no significant increased risk of unfavorable pregnancy outcome was reported except for abortion.
In resource-limited settings, non-nucleoside reverse transcriptase inhibitors (NNRTI) are recommended for first-line antiretroviral therapy (ART) (1). Efavirenz (EFV) is the preferred NNRTI because of its lower toxicity compared to nevirapine (NVP), its possible use in case of co-infection of HIV with tuberculosis and in women with CD4 count above 250 cells/μl. However these benefits need to be carefully weighed against its potential risk of birth defects, especially in the settings where contraceptive and family planning services are limited. Indeed, EFV was classified by the US Food and Drug Administration (FDA) in category D (positive evidence of risk of teratogenicity) after birth defects were reported in some infants exposed to EFV during the first trimester of pregnancy (2-4).
In sub-Saharan Africa, 60% of HIV-infected patients are women, most of them of childbearing age (5-7). Thus, an increasing number of pregnancies is occurring in women initiating ART. An incidence of 9.0 new pregnancies per 100 women on ART was recently reported in the MTCT-Plus network (8). In Côte d’Ivoire, in the ANRS Ditrame-Plus cohort where the contraceptive methods were “free of charge”, the risk of a new pregnancy was 16.5% at 24 months (9). In Botswana, this probability was 7.9 per 100 person-years among women on EFV-based ART (10).
To date, few data are available on the pregnancy outcomes of women exposed to EFV during the first trimester. Data from the international Antiretroviral Pregnancy Registry (APR) (11) and cohort studies (12, 13) have found no evidence of increased prevalence of overall congenital abnormalities after first-trimester EFV exposure (2.8/100 live births), compared to second or third-trimester EFV exposure (3.6/100 live births), to first-trimester NVP exposure (2.1/100 live births) or to the general population (2.7/100 live births). Detailed data on these pregnancy outcomes with a breakdown by miscarriage, preterm delivery (PTD), stillbirth and low birth weight (LBW) are hardly reported (10, 12, 14, 15). Our objective was to describe and compare the pregnancy outcomes between women who were exposed to EFV and those exposed to NVP during the first trimester in Côte d’Ivoire, West Africa.
A retrospective survey was conducted in Abidjan within four adult HIV clinical centers that are participating in the IeDEA (International epidemiological Databases to Evaluate AIDS) West Africa collaboration (http://www.iedeawestafrica.org) and in two Agence Nationale de Recherches sur le Sida et les Hépatites Virales (ANRS) randomized trials , ANRS 1269 Trivacan (16) (registered number NCT 00158405) and ANRS 12136 Temprano (registered number NCT 00495651).
The occurrence of new pregnancy was based on the last menstruation date at each follow-up visit and/or the urine test. New pregnancy was systematically recorded in the standardized forms (for the two trials) or in the clinical forms.
We searched information on all women on EFV who became pregnant in the data base of the two trials and in the computerized information system of the participating clinics. Therefore all HIV-infected pregnancy women who conceived on EFV or NVP-based ART between January 2003 and July 2009 were eligible for this study. Only the first recorded pregnancy in the given period was taken into account for each woman. The gestational age was estimated by the history of last menstruation or an ultrasound examination. Standardized forms were developed to collect information on age, pretreatment CD4 count, date of starting ART (NVP or EFV), drug regimen, switch and reason of switching to other ARV drugs than NVP or EFV, gestational age at diagnosis of pregnancy, last menstruation date, outcome of pregnancy and newborn anthropometric characteristics. Women exposed to EFV were advised by the physicians about the potential risk of congenital abnormalities and received free of charge the contraceptive methods available.
Five outcomes were considered: (1) abortion defined as a voluntary termination of pregnancy (not based on the detection of congenital abnormality by obstetrical ultrasound), (2) miscarriage as an unwanted termination of pregnancy <20 weeks of amenorrhea, (3) stillborn as any death reported in children between 20 weeks of amenorrhea and delivery, (4) PTD as any live birth occurring <37 weeks of amenorrhea, LBW as a live born child weighing <2,500 grams, and (5) congenital abnormalities as all visible and external abnormalities observed or diagnosed in the first six weeks of age.
The protocol of the IeDEA West Africa collaboration and of the ANRS 1269 and 12136 trials were approved by the national ethics committees of Côte d’Ivoire. Each IeDEA-affiliated clinical center has a federal-wide assurance number identification.
Group comparisons were performed using Student’s t test or non-parametric Mann-Whitney U test for continuous variables, Chi-square test or Fisher’s exact test when appropriate for categorical variables. The proportion of birth outcomes was reported with its 95% confidence interval (95% CI). All analyses were performed with SAS software version 9.1 (SAS Institute, Cary, NC, USA).
From January 2003 to July 2009, a total of 344 HIV-infected women conceived while receiving ART: 213 (61.9%) conceived on EFV-based ART and 131 (38.1%) on NVP-based ART. Overall, the median age at ART initiation was 29 years (inter-quartile range [IQR], 26-32), the median CD4 count was 217 cells/μl (IQR, 146-280) and 193 (56.1%) women were at WHO stage 3 or 4. The two nucleoside drug combinations associated with EFV or NVP were zidovudine-lamivudine (59.9%), stavudine-lamivudine (37.2%), tenofovir-emtricitabine (2.3%), and didanosine-stavudine (0.6%). The baseline characteristics of women were comparable in the EFV and NVP-based ART groups, except for the ART regimen, with more women receiving ZDV-lamivudine with EFV than with NVP (Table 1).
During pregnancy, 18 (5.2%) women were lost to follow-up and 192 (55.8%) switched their initial treatment. The proportion of women who were lost to follow-up before delivery and whose pregnancy outcome could not be ascertained was not statistically different between women exposed to EFV and those exposed to NVP (4.7% vs 6.1%; p=0.57) (Table 2). As for the treatment modification, 190 (89.2%) of the 213 women exposed to EFV switched to either a protease inhibitor (PI) (n=25) or to a NVP-based regimen (n=165) because of the diagnosis of pregnancy. In those exposed to NVP, two women switched to a PI-based regimen because of severe adverse events. The median duration of ART exposure after conception was 52 days [IQR, 37-75] in the EFV group and 264 days [IQR, 222-285] in the NVP group (p <0.001).
Overall, the pregnancy outcomes were known in 94.8% of women: 203 exposed to EFV and 123 to NVP. Of these 326 women, 11.7% had an abortion (95% CI: 8.4-15.7), 5.2% had a miscarriage (95% CI: 3.1-8.2) and 6.7% experienced a stillbirth (95% CI: 4.3%-10.0%). The proportion of abortion was higher in the EFV group than in the NVP group (14.3% vs 7.3%; p=0.05). The proportion of miscarriages and stillbirths was similar in both groups (Table 2). A total of 249 live births (76.4%) were reported, including 30 (12.0%) via caesarean section (Table 2). Data on birth weight were available for 223 infants (89.6%) and median birth weight was 2800 grams [IQR 2500-3250]. PTD was reported in 27 (10.8%) live born infants, with no statistical difference between those exposed to EFV and to NVP (9.5% vs. 12.7%; p=0.76). A LBW was measured in 45 infants (20.2%), without any statistical difference between those exposed to EFV and to NVP (17.2% vs. 24.2%; p=0.20). No external congenital malformation was reported in the clinical charts of the 249 live born infants, neither those exposed to EFV (upper limit of the 95% confidence interval: 2.5%) nor those exposed to NVP (upper limit of the 95% confidence interval: 3.6%) (Table 2).
Our study conducted in Côte d’Ivoire is the largest single study to-date evaluating HIV-infected women exposed to EFV during the first trimester of pregnancy. In this survey, no increased risk of pregnancy outcomes was reported following EFV exposure compared to NVP exposure during the first trimester, except for induced abortion. In general, our results are comparable to those observed in previous studies in terms of frequency of pregnancy outcomes after exposure to EFV during the first trimester. For abortion, the frequency was higher after EFV exposure (14.3%) compared to NVP exposure (7.3%). Similar results were found in two studies conducted in Africa. The first report included 37 women and 15% of these chose to have an early termination of pregnancy (17). The second study included 117 women and the frequency of abortion was 33.7% (28/83) after EFV-based ART exposure and 5.9% (2/34) after a non EFV-ART (18). This high frequency of abortion after early exposure to EFV could be related by the fear of warning given by clinicians about the potential risk of congenital abnormalities. Another explanation is, at enrolment; the women did not want children and therefore were eligible for EFV
In our sample of women who became pregnant while receiving ART, no case of congenital abnormalities was reported in infants exposed to EFV (upper limit of the 95% confidence interval: 2.5%) or to NVP (upper limit of the 95% confidence interval: 3.6%), although the median time of exposure covered the critical period of organogenesis (52 days with EFV and 264 days with NVP). These results should be taken with caution, as they were obtained through a retrospective survey, with moderate sample size, and we could only focus on the visible and external congenital abnormalities during the first six weeks of age. However, our findings are compatible with data already reported where the frequency of congenital abnormalities ranged between 0.0% and 4.5% and where no increased risk of congenital abnormalities was found with EFV-based ART compared to a non-EFV-based ART (10-13, 19). In the US-based prospective antiretroviral pregnancy registry, congenital abnormalities had been observed as of July 2009 in 14 of 501 (2.8%, 95% CI: 1.5%-4.7%) live born infants following first-trimester exposure to EFV and in two of 55 (3.6%, 95% CI: 0.4%-12.5%) live births with second-or third-trimester EFV exposure (11). A similar observation was made in a study in South Africa where the proportion of congenital abnormalities was not significantly different between live births exposed to EFV during the first trimester (3.3%) and second/third trimester (2.6%). In this report, the proportion of congenital abnormalities was 3.0% in live births exposed to NVP-based ART during the first trimester (12).
For the other pregnancy outcomes, miscarriage, stillborn, PTD and LBW, the frequencies following first trimester EFV exposure found in our study were also comparable to other studies reported in Africa (10, 12, 18) or in Europe (13-15). As reported also in our study, there was no increased risk of these pregnancy outcomes after EFV exposure during the first trimester, compared to second/third trimester exposure with EFV or to a non-EFV-based ART (12, 18).
We also found that approximately 90% of HIV-infected women exposed to EFV during the first trimester switched to NVP or to LPV/r once pregnancy was diagnosed (median time of exposure of 52 days after conception). In South Africa, EFV was substituted with NVP in all the women, once the diagnosis of pregnancy was made (median time of exposure of 43 days) (10). However, in another study in the same country, EFV was substituted with NVP or PI in 32% only (63/195) of women who conceived on EFV-based ART with no increased risk of congenital abnormalities compared to second/third EFV exposure (12). It is thus likely that in EFV-treated HIV-infected women, EFV switches is only necessary if the diagnosis of pregnancy is made before the complete closure of the neural tube. This event occurs approximately 28 days in median after conception. This situation will thus be rarely encountered in Africa where antenatal care is usually initiated much later in pregnancy. In conclusion, our study is the largest evaluation made in a single setting to date on EFV exposure during the first trimester of pregnancy. It provides important additional information to the EFV teratogenicity debate and has already been integrated in the recently published meta-analysis (20). Indeed, Ford et al concluded that there was no increased risk of overall birth defects between 35/1132 EFV-exposed women with live births compared to 289/7163 women with live births exposed to other antiretroviral drugs in the first trimester of pregnancy (20). Although our data seem reassuring as the previous reports and the meta-analysis, it is clear that the limited sample size remains a weakness. The 2009 revision of the World Health Organization guidelines recommends that EFV be avoided during the first trimester of pregnancy or that it should not be used by women of childbearing age in the absence of an effective contraceptive method (21). In addition, it states that if a woman already on EFV is diagnosed as pregnant before 28 days of gestation, EFV should be stopped and substituted with NVP or a PI. If a woman is diagnosed as pregnant after 28 days of gestation, EFV should be continued. Finally, there is no indication for abortion in women exposed to EFV in the first trimester of pregnancy (21). More collaborative studies and pharmacovigilance systems capturing routine data should be put in place especially in sub-Saharan Africa to improve the evidence informing public health guidelines. In between, models may investigate the cost-benefit ratio between a small additional risk of severe malformations in the infants and an improvement in survival for a small number of mothers receiving ART using EFV compared to NVP.
We acknowledge Dr Renaud Becquet and Mrs Delphine Gabillard for their advice and contribution to the preparation of the study. The International epidemiological Database to Evaluate AIDS in West Africa (IeDEA West Africa) is supported by: the National Cancer Institute (NCI), the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) and the National Institute of Allergy And Infectious Diseases (NIAID) as part of the International Epidemiologic Databases to Evaluate AIDS (IeDEA) (grant no. 5U01AI069919-03). The Trivacan and Temprano trial are sponsored by the French Agence Nationale de Recherches sur le Sida et les hepatites virales (ANRS). The content of this publication is solely the responsibility of the authors and does not necessarily represent the official views of any of the institutions mentioned above.
The IeDEA West Africa pediatric working Group is constituted as follows:
IeDEA West Africa primary investigators: Pr F. Dabis* (INSERM U897, ISPED, Bordeaux, France), E. Bissagnene* (SMIT, CHU de Treichville, Abidjan, Côte d’Ivoire). Coordinator HIV Clinical care in Abidjan, Côte d’Ivoire: CEPREF, Abidjan: Dr E. Messou; MTCT Plus Network, Abidjan: Dr C Amani-Bosse; CIRBA: Dr Olivier Ba-Gomis, CNTS: Dr Albert Minga.
Bordeaux ISPED staff: E. Balestre, A. Doring, Dr DK. Ekouévi, Dr A. Jaquet, Dr V. Leroy, Dr C. Lewden, K. Malateste, E. Rabourdin, Dr R. Thiebaut.
Programme PAC-CI, CHU de Treichville (IeDEA West Africa Regional office), Abidjan, Côte d’Ivoire: G. Allou, JC Azani, Dr PA. Coffie, H. Djétouan, B. Kouadio, A. Kouakou
The ANRS 1269 Trivacan trial group is constituted as follows:
Principal Investigators: Emmanuel Bissagnene, Roger Salamon
Trial coordination team: Xavier Anglaret (Coordinator), Christine Danel (Coordinator), Raoul Moh (trial monitor), Souleymane Sorho (data manager). Delphine Gabillard (trial statistician), Investigators Clinical care in Abidjan, Côte d’Ivoire: SMIT: Serge Eholie, CIRBA : Henri Chenal, CNTS : Albert Minga, USAC : Constance Kanga, CePReF: Eugene Messou Biology in Abidjan, CeDReS (Centre de Diagnostic et de Recherches sur le SIDA) Francois Rouet
The ANRS 12136 Temprano trial group is constituted as follows:
Principal Investigators: Xavier Anglaret, Serge Eholie
Trial coordination team: Christine Danel (Coordinator), Raoul Moh (Monitoring Coordinator), Franck Bohoussou (data manager), Anani Badjé (trial monitor), Jean Baptiste N’takpé (trial monitor),, Eric Ouattara (trial monitor), Jerome Lecarrou (trial monitor), Delphine Gabillard (trial statistician), Investigators Clinical care in Abidjan, Côte d’Ivoire: SMIT: Emmanuel Bissagnene, CIRBA : Olivier Ba-Gomis, CNTS : Albert Minga, USAC : Stéphane Koulé, CePReF: Amani Anzian, Hôpital Général d’Abobo nord: Emmanuel Kouamé, Formation Sanitaire Urbaine Anonkoua kouté: O Makaïla, La pierre angulaire: Madeleine Kadio-Morokro
Biology in Abidjan, Côte d’Ivoire, CeDReS (Centre de Diagnostic et de Recherches sur le SIDA) Hervé Menan
Administrative team PACCI: Bertin Kouadio, Adrienne Kouakou, Hughes Djetouan, Gérald, Kouamé, Célestin N’Chot.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.