O-273 Risk of hypertensive disorders in pregnancy after fresh and frozen embryo transfer in assisted reproduction: A population-based cohort study with within sibship analysis

Study question

Is the risk of hypertensive disorders in pregnancy (HDP) increased following assisted reproductive technology (ART) with fresh or frozen cycles compared to conception without ART?

Summary answer

Pregnancies after frozen embryo transfer (FET) had a substantially higher risk of HDP, also after accounting for constant parental factors in within sibship comparisons.

What is known already

FET in assisted reproductive technology (ART) is increasingly common because of improved cryopreservation methods, blastocyst culture, and freeze-all practice. Conventional observational studies have raised concerns about treatment safety due to higher risk of HDP after FET compared to both non-ART and fresh embryo transfer (fresh-ET). Whether this is attributable to parental factors or treatment is unknown. Within sibship comparisons can provide unique insight by controlling for confounding from unmeasured, constant parental factors, like genetics, preconception lifestyle and health, as well as socioeconomic status.

Study design, size, duration

Population-based cohort study with within sibship comparison using health registry data from three Nordic countries. In total, 4,523,028 singleton pregnancies resulting in delivery between 1988 and 2015 were included, of which 78,300 were after fresh-ET, and 18,037 were after FET. We identified 33,209 sibships that were conceived using two or three different conception methods (fresh-ET, FET, and non-ART).

Participants/materials, setting, methods

Data were collected from several national health registries in Denmark, Norway, and Sweden, including ART, medical birth, and patient registries, and linked using the residents’ unique identification numbers. Adjusted odds ratios (aOR) of HDP with 95% confidence intervals (CI) were estimated using multilevel logistic regression, where random effects provided conventional population estimates, and fixed effects gave within sibship estimates. Main models included adjustment for maternal age, parity, birth year and country.

Main results and the role of chance

Risk of HDP was higher after FET compared to non-ART conception, both at the population level (7.4% vs 4.3%, aOR 1.74, 95% CI 1.61 to 1.89) and within sibships (aOR 2.02, 95% CI 1.72 to 2.39). For fresh-ET, the risk was similar to non-ART, both at the population level (5.9% vs 4.3%, aOR 1.02, 95% CI 0.98 to 1.07) and within sibships (aOR 0.99, 95% CI 0.89 to 1.09). Sensitivity analyses with adjustment for body mass index (BMI) and smoking, restriction to full siblings (same mother and father) and siblings born within a three-year interval, were all consistent with our main findings. Furthermore, the results were not driven by order of conception method or other ART procedures (intracytoplasmic sperm injection, culture duration or number of embryos transferred).

Limitations, reasons for caution

Despite the extra control for shared confounders provided by the within sibship analyses, we cannot exclude residual confounding from non-shared confounders, such as smoking and BMI, where confounder control was limited by a large proportion of missingness, and causes of infertility, which were largely unknown.

Wider implications of the findings

Our findings are important for the ongoing discussion about the freeze-all policy, and further investigation of which treatment factors might be involved is warranted.

Trial registration number

Not applicable