Endocrine disease and pregnancy loss - a nationwide cohort study

To investigate associations between endocrine diseases and pregnancy loss, including associations according to the number of pregnancy losses, temporal patterns, and familial clustering independent of the woman's own endocrine diagnosis.

Nationwide population-based cohort study.

366,539 women born in Denmark between 1977 and 1993 with documented pregnancies in the Danish national registers.

Presence of endocrine disease diagnosed before or after pregnancy. Subgroup analyses of major endocrine diseases, including hypothyroidism, hyperthyroidism, polycystic ovary syndrome (PCOS), type 1 and 2 diabetes, and gestational diabetes.

Pregnancy loss, defined as the number of pregnancy losses before 22 weeks of gestation (categorized as 0, 1, 2, or ≥3 losses) in association with endocrine disease. Multivariable multinomial logistic regression models calculated odds ratios (ORs) with 95% confidence interval adjusted for maternal age and year of pregnancy. Additional analyses examined associations according to timing of endocrine disease diagnosis relative to the first pregnancy and evaluated familial clustering.

Endocrine disease was diagnosed in 56,618 women (15%). In analyses that did not account for temporal order, endocrine disease showed progressively stronger associations with increasing number of pregnancy losses: OR 1.15 (95% CI: 1.12-1.17) for one, OR 1.30 (95% CI: 1.24-1.37) for two, and OR 1.81 (95% CI: 1.70-1.93) for three or more (p<0.001). All major endocrine diseases were significantly associated with pregnancy loss. Women diagnosed with an endocrine disease after their first pregnancy showed a stronger association with pregnancy loss in the first pregnancy (OR 1.24 [95% CI: 1.20-1.29]) than those diagnosed before the first pregnancy (OR 1.11 [95% CI: 1.07-1.15]). Familial endocrine disease in a parent or sister was also positively associated with pregnancy loss (OR 1.07 [95% CI: 1.05-1.09] and OR 1.08 [95% CI: 1.03-1.13], respectively), independent of the woman's own endocrine diagnosis.

Endocrine diseases and pregnancy loss were strongly associated regardless of temporal order, with progressively stronger associations across increasing numbers of pregnancy losses, with significant familial clustering independent of personal diagnosis, suggesting shared genetic or environmental factors. Our findings support consideration of targeted endocrine evaluation in women with pregnancy loss. Further research should determine optimal screening strategies and timing.