Pregnancy rate and birth outcomes among women receiving antiretroviral therapy in Burkina Faso: a retrospective cohort study

Introduction In Sub-Saharan Africa, few studies reported pregnancy incidence and outcomes in women taking antiretroviral therapy (ART). This survey aims to estimate the incidence and outcomes of pregnancy in a cohort of HIV positive women initiating ART in Bobo-Dioulasso, Burkina Faso. Methods We carried out a retrospective cohort study. We selected women in childbearing age initiating ART and followed up in Bobo-Dioulasso teaching hospital between January 2005 and June 2011. The incidence of pregnancies during follow-up was calculated. Childbirth was defined by the expulsion of a fetus after 22 weeks of amenorrhea. Before this term, it is an abortion. Childbirth is said premature if it occurs before 37 weeks of gestation, to term if it occurs between the 38th and the 42nd week. The annual age-standardized fertility rates were calculated using the baseline population from the 2010 demographic and health survey (DHS) in Burkina Faso. Results A total of 1,763 women of childbearing age under ART were included in the study. They ranged between 18 and 48 years old with a median of 35 years old. A total of 222 pregnancies were observed during 4639 women-years of follow-up, corresponding to an incidence density of 5 pregnancies for 100 women-years (95% CI: 4.2-5.5). Among the 222 pregnancies recorded, 9(4.0%) ended with abortion, 205(92.4%) with childbirth (including 15 premature childbirths); the outcome of 8(3.6%) pregnancies were unknown abortion. Live birth and stillborn rates were 94.0% (193/205) and 6.0% respectively. The standard fertility rate in our cohort was 45 live births for 1,000 women-years. The general decrease in fertility rates was 66.0% among women infected with HIV compared to the overall population Conclusion This study shows a low pregnancy incidence among women initiating ART as compared to their peers from the general population. Pregnancies that occurred during ART generally end with live births. Care packages for HIV infected women of childbearing age must include reproductive health services to better address this issue.


Introduction
Sub-Saharan Africa is the most HIV affected area, with women as the most vulnerable population. Indeed, 58% of people living with HIV/AIDS (PLWHA) in Sub-Saharan Africa was women of childbearing age [1]. Improving life expectancy and quality for PLWHA on antiretroviral therapy (ART) increases their desire for bearing children [2][3][4][5][6][7] even though HIV infection is associated with a 25 to 40% decrease in fertility among women of childbearing age [8][9][10]. Pregnancy is not likely to influence the progression of the infection [11]. Additionally, as long as the mother is in good health, HIV does not impact the course of the pregnancy [12]. The risks of complications are higher for the mother and her fetus when mother develops an HIV-related opportunistic disease and when the CD4 cells count is below 200/µl [13]. Miscarriages, abortions, ectopic pregnancies and stillborns delivery seem to be as recurrent among HIV positive women (some of whom were under ART) as among non infected women [8,[14][15][16]. According to the UNAIDS Report on Global AIDS Epidemic, 60% of HIV-infected people in Burkina Faso are women of childbearing age [17]. Besides, the desire for bearing children is quite high among women in general in Burkina Faso including HIV infected women [18]. However, there are little data on the frequency of pregnancies among HIV-infected women under antiretroviral treatment. We therefore deemed it necessary to study the incidence and outcome of pregnancies in Bobo-Dioulasso Day Hospital cohort. Knowing the status of women under ART who get pregnant will help provide appropriate care to countries with limited resources.  The annual age-standardized rates (indirect standardization) were calculated using the baseline population from the 2010 demographic and health survey (DHS) in Burkina Faso [10]. The agreed significance threshold was 5% for all analyses.

Methods
Ethical considerations: this observational study was conducted as part of the routine care services and therefore presented no risk to patients. For confidentiality purposes, the data collected were made anonymous. The frequency of live births and stillborns was respectively 94.0%

Discussion
This retrospective study has some limitations. Despite it is condemned by the law in Burkina Faso, pregnancy termination is still performed. Therefore, we cannot assert that all cases of pregnancy have been registered. However, this innovative study made it possible to assess the frequency of pregnancies among patients on ART in Burkina Faso. The incidence of pregnancy was 5 for 100 women-years. However, this number is below most of rates found in African studies [8,9,20-

What this study adds
 This study highlights a low pregnancy incidence among women under ART compared to the population in general.
 The standardized incidence ratios (SIR) are low and the overall decrease in fertility was 66.0% compared to the entire population.
 Women who became pregnant while under ART delivered live babies in most cases.

Competing interests
The authors have no competing interests to declare.

Acknowledgments
The authors thank all patients of the Day Care Unit of Bobo Dioulasso, the male nurses, and all non-medical staff of our department.