Pregnancy and birth outcomes in a Danish nationwide cohort of women with autoimmune hepatitis and matched population controls.

Background: Many patients with autoimmune hepatitis are women of fertile age. Some concerns of these patients are related to pregnancy.
Aim: To conduct a nationwide study on risk of miscarriage, birth rate, and birth outcomes in women with autoimmune hepatitis.
Methods: From Danish healthcare registries, 1994-2015, we identified 179 births in 103 women with autoimmune hepatitis, 70 of which were first-time singleton births, and 1623 births in 1051 age-matched women (population controls), 662 of which were first-time singleton births. We calculated the risk of miscarriage and the birth rate after autoimmune hepatitis diagnosis in women with autoimmune hepatitis and controls. We used logistic regression to compare the odds of adverse birth outcomes (preterm birth, small for gestational age, congenital malformations and stillbirth) between women with autoimmune hepatitis and controls, adjusting for maternal age and smoking habits.
Results: The risk of miscarriage was similar in women with autoimmune hepatitis and controls: risk ratio 1.17 (95% confidence interval 0.81-1.68). The first-time birth rate, including singleton and multiple births, per 1000 person-years in women with autoimmune hepatitis was 37 (95% confidence interval 29-46), in controls 32 (95% confidence interval 30-35). Age at first-births was similar in women with autoimmune hepatitis and controls. Women with autoimmune hepatitis had an increased risk of preterm birth (adjusted odds ratio 3.19, 95% confidence interval 1.53-6.64) and small for gestational age babies (adjusted odds ratio = 3.20, 95% confidence interval 0.33-31.29), but not of congenital malformations (adjusted odds ratio = 1.27, 95% confidence interval 0.48-3.34) or stillbirth. Birth outcomes did not differ in autoimmune hepatitis patients on or off immunosuppression, and with or without cirrhosis.
Conclusions: In Danish women with autoimmune hepatitis, fertility was unaffected. They had an increased risk of preterm birth and small for gestational age children, but not of congenital malformations or stillbirth.
(© 2018 John Wiley & Sons Ltd.)