Assisted reproductive technology and the risk of unplanned peripartum hysterectomy: analysis using propensity score matching

Abstract

STUDY QUESTION: Is there an increased risk of unplanned peripartum hysterectomy in pregnancies with assissted reproductive technology compared to those without ART? SUMMARY ANSWER: Although the absolute risks are low, there is an almost five-fold increased risk of unplanned peripartum hysterectomy and 1.7 more unplanned peripartum hysterectomies occur per 1000 deliveries in pregnancies with ART compared to those without ART. WHAT IS KNOWN ALREADY: It has been reported that pregnancies with ART was associated with increased risk of peripartum hysterectomy in one case-control study and in one cohort study. STUDY DESIGN, SIZE, DURATION: A retrospective cohort study was conducted using a birth cohort from 2014 and 2015 in the United States, which includes more than 7 million births. Propensity score (PS) matching was used to control for confounding. PARTICIPANTS/MATERIALS, SETTING, METHODS: Subjects were divided into two groups: pregnancies with and without ART. We calculated PSs with demographic, clinical and socioeconomic variables, and subjects were matched using the PS with a 1:1 ratio. Subjects comprised 43 868 ART pregnancies and 43 868 non-ART pregnancies after PS matching. The primary outcome of interest was the risk of unplanned peripartum hysterectomy which was compared by evaluating the relative risk and the risk difference between the two groups after PS matching. MAIN RESULTS AND THE ROLE OF CHANCE: Baseline characteristics were similar between groups after PS matching. The risk of peripartum hysterectomy in women with ART was 4.947 times that of those without ART (0.0021 [94/43 868] vs 0.0004 [19/43 868]; 95% confidence interval [CI] 3.022-8.098). The risk difference between two groups was 0.0017 (95% CI 0.0012-0.0022). LIMITATIONS, REASONS FOR CAUTION: There is a possibility of bias due to unmeasured confounding such as fibroids, previous history of uterine surgery and intrauterine procedures. Misclassification of the exposure and/or the outcome could also influence the results. WIDER IMPLICATIONS OF THE FINDINGS: Although we found a five-fold increased risk of unplanned peripartum hysterectomy in pregnancies with ART compared to those without ART, the results should be interpreted with caution in a clinical context as the overall number and the absolute risk of unplanned peripartum hysterectomy are very low in either group (1/2325 in the non-ART group, and 1/468 in the ART group). However, it would be appropriate as future research agenda to explore mechanisms and/or etiology underlying this finding.