articleIncidence and risk factors of severe adverse events with nevirapine-based antiretroviral therapy in HIV-infected women . MTCT-Plus program , Abidjan , Côte d ' Ivoire

Background: In resource-limited settings where nevirapine-containing regimen is the preferred regimen in women, data on severe adverse events (SAEs) according to CD4 cell count are limited. We estimated the incidence of SAEs according to CD4 cell count and identify their risk factors in nevirapine-treated women. Methods: All HIV-infected women who initiated nevirapine-containing regimen in the MTCT-Plus operational program in Abidjan, Côte d'Ivoire, were eligible for this study. Laboratory and clinical (rash) SAEs were classified as grade 3 and 4. Cox models were used to identify factors associated with the occurrence of SAEs. Results: From August 2003 to October 2006, 290 women initiated a nevirapine-containing regimen at a median CD4 cell count of 186 cells/mm3 (IQR 124-266). During a median follow-up on treatment of 25 months, the incidence of all SAEs was 19.5/100 patient-years. The 24-month probability of occurrence of hepatotoxicity or rash was not different between women with a CD4 cell count >250 cells/mm3 and women with a CD4 cell count ≤250 cells/mm3 (8.3% vs. 9.9%, Log-rank test: p = 0.75). In a multivariate proportional hazard model, neither CD4 cell count >250 cells/mm3 at treatment initiation nor initiation NVP-based regimen initiated during pregnancy were associated with the occurrence of SAEs. Conclusion: CD4 cell count >250 cells/mm3 was not associated with a higher risk of severe hepatotoxicity and/or rash, as well as initiation of ART during pregnancy. Pharmacovogilance data as well as meta-analysis on women receiving NVP in these settings are needed for better information about NVP toxicity. Background At the end of 2007, more than three million people were receiving Highly Active Antiretroviral Therapy (HAART) in resource-limited settings [1]. The first-line regimen recommended by the World Health Organization (WHO) in resource-limited includes two nucleoside reverse transcriptase inhibitors (NRTI) with one nonnucleoside reverse transcriptase inhibitor (NNRTI) [2]. Nevirapine (NVP) is the preferred NNRTI in first-line antiretroviral regimens in pregnancy because of substantial clinical experience with pregnant women and its proven efficacy in reducing mother-to-child transmission [3,4]. The most frequent adverse events of NVP are hepatotoxicity and cutaneous rash. Studies from developed countries report an increased risk of severe hepatotoxicity and cutaneous rash in women with CD4 count >250 cells/mm3 [5-9]. Based on these findings, WHO recommends avoiding NVP in women, including pregnant women, with CD4 counts between 250 and 350 cells/ mm3, or if no other options exist as it is often the case in resource-limited settings, using it with caution and close monitoring during the first 12 weeks of therapy [2,8]. Although some cases have been reported in pregnant women, it is still not known whether pregnancy increases the risk of the occurrence of hepatotoxicity or rash [9-11]. * Correspondence: didier.ekouevi@gmail.com 1 Programme MTCT-Plus, ACONDA, BP: 1954 Abidjan 18, Abidjan, Côte d'Ivoire Full list of author information is available at the end of the article Coffie et al. BMC Infectious Diseases 2010, 10:188 http://www.biomedcentral.com/1471-2334/10/188 Page 2 of 10 In resource-limited settings, where NVP is widely used in women and during pregnancy whatever the level of CD4, there is little data on the occurence of hepatotoxicity and/ or rash according to CD4, and all except one study did not find this association [12-16]. We hypothesize that HAART initiation among pregnant HIV-infected women or HIV-infected women with CD4 count >250 cells/mm3 may be associated with an increased frequency of SAEs. Thus, we estimated in the MTCT-Plus program in Abidjan, Côte d'Ivoire, the incidence rate of all severe adverse events (SAEs) according to CD4 cell count and pregnancy status, especially hepatotoxicity and/or rash and investigated their risk factors in women initiating NVP-based HAART. Methods Study design and setting A prospective cohort study was conducted in Abidjan, Côte d'Ivoire between August 2003 and October 2006 among women enrolled in the MTCT-Plus program, which is built upon existing prevention of mother-tochild transmission (PMTCT) services and provides HIVinfected women, their partners, and their children, holistic family care with unrestricted access to HAART for eligible patients [17,18]. Study population All HIV-infected women were included in this study if they initiated NVP-based HAART according to the following eligibility criteria: WHO clinical stage 2 (until December 2004) or stage 3 and CD4+ T lymphocytes (CD4 cell count) <350 cells/mm3, or stage 4 regardless of CD4 cell count, or CD4 cell count <200 cells/mm3. The women had to meet one additional biological criterion before initiating HAART: aminotransferase and/or alaninetransferase levels no more than three times the upper limit of the normal range. All women with a CD4 cell count <350 cells/mm3 were also prescribed co-trimoxazole prophylaxis. All women were ARV-naive with the exception of prior PMTCT exposure. Ethical aspects The MTCT-Plus program was reviewed by the institutional review board (IRB) from Columbia University in 2000 (principal sponsor) and was not considered a research project but rather a demonstration program in the context of the ARV roll-out. As an operational HIV care and treatment program MTCT-Plus program was exempted from review by the IRB from the Ministry of health of Côte d'Ivoire. Enrollment and follow-up Maternal socio-demographic, clinical and biological characteristics were recorded at enrollment in the program and at initiation of HAART [18]. During weekly follow-up visits for the first two months and monthly visits thereafter, clinical signs and symptoms, drug intake, and tolerance were assessed. Between scheduled visits, women had free access to the study clinics for medical problems. Total blood cell count (MaxM, BeckmanCoulter, Miami, FL, USA), serum liver enzymes, and serum creatinine were monitored according to the following schedules: in women starting HAART after delivery at baseline, month 1, and every three months thereafter; in women starting HAART prepartum at baseline, week 2, month 1, and monthly thereafter during pregnancy. CD4 cell count were measured using a dualplatform flow cytometry technique with an automated blood cell counter (MaxM, BeckmanCoulter, Miami, FL, USA), at the screening visit and then every six months. Serology for hepatitis B was not routinely performed. All blood samples were processed in the same laboratory. Outcomes and definitions The outcomes of interest were the occurrence of SAEs defined as clinical and biological grade 3⁄4 adverse events according to the Agence Nationale de la Recherche sur le Sida et les hepatites virales (ANRS) table for grading the severity of adult adverse events [19]: at least one haemoglobinaemia measurement <70 g/l, one neutrophil count <750/mm3, one amino and/or alaninetransferase measurement more than five times the upper limit of the normal range and any extended papulovesicular or oozing eruption, palpable purpura (suggestive of vasculitis), polymorphous erythaema, small-size cutaneous or mucous ulcerations, any blistering cutaneous and/or mucosal lesions (Lyell or Stevens-Johnson), febrile erythrodermia, whether or not associated with other signs indicative of hypersensitivity, cutaneous necrosis requiring surgical excision. All SAEs were reported by clinicians during follow-up visits and subsequently validated by two expert clinicians. If the experts disagree, a third expert was called upon to avoid misclassification and to validate a final diagnosis. Statistical analysis Group comparisons used Student's t-test, non-parametric Mann-Whitney U test or variance analysis for continuous variables, and the Chi2 test or Fisher's exact test for categorical variables. The incidence rate of SAEs per 100 patient-months or years was estimated with its 95% confidence interval (CI). The Log-Rank test was used to compare the incidence rate between two groups. Univariable and then multivariable Cox regression analyses were performed to identify factors associated with the occurrence of these SAEs. All factors associated with the outcomes at a P value <0.25 were included in the multivariable analysis. The CD4 cell count and WHO clinical stage at Coffie et al. BMC Infectious Diseases 2010, 10:188 http://www.biomedcentral.com/1471-2334/10/188 Page 3 of 10 HAART initiation were kept in the final model. Adjusted hazard ratios (aHR) and their 95% CI are reported with two-sided p-values. All analyses were performed in intent-to-treat and on treatment [20] with SAS software version 9.1 (SAS Institute, Cary, NC, USA). Results Patients and follow-up From August 2003 to October 2006, 530 women were enrolled in MTCT-Plus and 290 (54.7%) women initiated a NVP-containing regimen and were included in this study. Their median age at HAART initiation was 29 years (inter-quartile range [IQR] 26-33) and median CD4 cell count was 186 cells/mm3 (IQR 124-266). Two-hundred and two women (70%) initiated HAART with a CD4 cell count ≤250 cells/mm3, 130 (44.8%) were at WHO clinical stage 3 or 4, 125 (43.0%) initiated HAART during pregnancy, and 287 (99.0%) started co-trimoxazole. During a median follow-up on HAART of 25 months (IQR 14-30), seven women (2.4%) were lost to follow-up, seven (2.4%) stopped HAART on their own initiative and 16 (5.5%) died. The baseline and follow-up characteristics of these women are summarized in Table 1. Antiretroviral therapy The first-line HAART regimen was ZDV/3TC/NVP in 265 women (91.4%) and stavudine (d4T)/3TC/NVP in 25 women (8.6%). In this cohort, 153 women (52.8%) were previously exposed to a PMTCT regimen during a previous pregnancy: short-course (sc) ZDV and lamivudine plus single dose nevirapine (sdNVP) (n = 84), scZDV plus sdNVP (n = 65) and scZDV alone (n = 2). The median interval between exposure to PMTCT and