Arterial and venous complications after fertility treatment: A French nationwide cohort study

To determine whether the risk of thromboembolic complications is higher in women following unsuccessful fertility treatment (FT) and in pregnant women following successful FT, and whether the risk differs according to FT type.

This is an observational prospective cohort study. All French women aged 18-45 years who received FT between 2013 and 2015 were selected from the French health insurance claim database which registers healthcare consumption for the entire French population. All FT reimbursed over a 28-day period from the date of the first FT were considered to constitute one FT cycle. Each FT cycle was classified according to type: either simple ovulation induction (OI) or ovulation stimulation (OS). All hospitalisations with a diagnosis of venous thromboembolism (VTE), arterial thrombosis (AT) or ovarian hyperstimulation syndrome (OHSS) were identified for the selected women in the French hospital discharge database. Poisson regressions were used to estimate incidence rate ratios (IRR) by comparing i) the incidence of thromboembolic complications (i.e., VTE and AT) and OHSS following unsuccessful FT cycles with the incidence of these two diseases in all non-pregnant women of the same age range (i.e. non-pregnant control group), and ii) incidence of thromboembolic complications and OHSS in women who became pregnant following successful FT with the incidence in women of the same age range with spontaneous (i.e., no FT) pregnancies (i.e., pregnant control group (spontaneous pregnancy)).

During the study period, 277,913 women underwent FT, for a total of 788,007 FT cycles, with 82,821 FT-related pregnancies. Among unsuccessful FT cycles, 75 VTE and 43 AT were observed. OS treatment cycles but not OI were associated with a higher risk of VTE than in reference group (age-adjusted IRR 1.74, 95%CI [1.30-2.34]). Among FT-related pregnancies, 207 VTE and 35 AT were reported. VTE and AT incidence rates during the first trimester were higher after OS treatment cycles than in the pregnant control group (spontaneous pregnancy) after adjusting for age and twin/multiple pregnancies (IRR_VTE = 3.29, 95%CI [2.24-4.81]; IRR_AT = 2.63, 95%CI [1.06-6.51]).

Monitoring women undergoing FT, especially OS, irrespective of pregnancy status is crucial. The risk of thromboembolic complications in the first trimester for FT-related pregnancies seems to be higher than that for spontaneous pregnancies.