Abstract Study question Are gestational carriers at higher risk of severe maternal (SMM) or neonatal (SNM) morbidity, compared to patients using unassisted conception or in vitro fertilization (IVF)? Summary answer A higher risk of SMM and adverse pregnancy outcomes is seen among gestational carriers, without evident neonatal morbidity. What is known already The use of gestational carriers (GC) also known as “surrogate pregnancy” is increasing. Gestational carriage may be associated with adverse outcomes linked to both IVF techniques and exposure to foreign fetal antigens. Adverse outcomes of GC pregnancies are largely unknown. Study design, size, duration Retrospective population-based cohort study of 937,938 births, derived from existing administrative health datasets. Data were linked using unique for birthing woman and child identifiers and analyzed at ICES Ontario. The study excluded participants with untreated infertility, those utilizing ovulation induction or intrauterine insemination, and those with multifetal gestation. Participants/materials, setting, methods Singleton births at > 20 weeks gestation in Ontario, 2012-2021. Pregnancy characteristics were obtained from the Ontario Birth Registry database. Exposure groups were i) unassisted conception (comparison group 1), ii) IVF (comparison group 2), or iii) GC. Primary outcomes included validated composite outcomes of i) SMM and ii) SNM. Secondary outcomes included hypertensive disorders of pregnancy, cesarean delivery, preterm birth, and postpartum hemorrhage. Modified Poisson regression models generated relative risks adjusted for birthing woman’s characteristics (aRR). Main results and the role of chance Of all singleton births, 937,938 (97.7%) were from unassisted conception, 20,958 (2.2%) from IVF, and 956 (0.1%) from GC. GC were more likely to be parous, reside in a lower-income area, and have higher rates of chronic hypertension. GC were more likely than unassisted conception women to be older and be non-smokers, with some opposite trends when compared to IVF recipients. Respective rates of SMM were 2.4%, 4.6% and 7.1%. The aRR were 3.07 (95% CI 2.44-3.87) comparing GC to unassisted conception, and 1.88 (95% CI 1.46-2.42) comparing GC to IVF. Respective rates of SNM were 6.0%, 9.1% and 6.5%, generating aRR of 1.21 (95% CI 0.95-1.54) for GC vs. unassisted conception, and 0.78 (95% CI 0.60-1.00) for GC vs. IVF. Respective rates of hypertensive disorders were 6.6%, 11.6% and 13.9%. The aRR were 1.89 (1.61 - 2.21) comparing GC to unassisted conception, 1.42 (1.19 - 1.69) comparing GC to IVF. Respective rates of postpartum hemorrhage were 5.7%, 10.5% and 13.9%, generating aRR of 2.87 (2.43 - 3.38) for GC vs. unassisted conception, and 1.40 (1.17 - 1.67) for GC vs. IVF. Limitations, reasons for caution The databases did not include certain baseline details, including the reasons for opting for a GC and the sources of gametes. There is a chance that the measured association may be distorted or confounded by the unmeasured variables. Wider implications of the findings GC appear to be at an elevated risk of maternal morbidity, without evident risk of neonatal morbidity. Further studies are needed to understand the mechanisms involved. Eligibility for GC should be determined using consistent selection criteria, and consideration should be given to specialized obstetrical care to promote optimal pregnancy outcomes. Trial registration number Not applicable
Read full abstract