Thyroid dysfunction in pregnancy - a retrospective observational analysis of a Maltese cohort

Thyroid dysfunction is known to adversely affect pregnancy. This study evaluates the prevalence of thyroid disorders and explores their association with pregnancy complications/comorbidities and modes of delivery in the Maltese pregnant population over a ten year period.

A population based observational study.

We analysed data from the National Obstetrics Information Service of the Department of Health Informations and Research (NOIS) for all births delivered in Malta between 2006 and 2016. Cases identified and recorded by NOIS to have had some form of thyroid dysfunction during pregnancy were confirmed by cross-referencing  with laboratory results found in patients' medical records and/or iSOFT® database system. Using the Statistical Package for the Social sciences (SPSS®)  demographic data, past obstetric and medical history and obstetric outcomes were analysed for pregnancies with thyroid dysfunction and compared to data pertaining to pregnancies in euthyroid patients, that is those with no recorded thyroid dysfunction on NOIS. Chi square/Fisher's exact test were used to compare categorical variables while ANOVA/Mann-Whitney U test was used to compare continuous variables. Statistical significance was defined by a two-sided p value <0.05.

Data was available for 46,283 women (mean [SD] age = 29.2 [5.4] years). 587 pregnancies (1.3%) suffered from thyroid dysfunction. Of these, 67.3% were hypothyroid, 3.2% had hyperthyroidism, 28.3% had isolated hypothyroxinaemia (IHT) while 1.2% had a history of thyroid carcinoma. Patients with IHT and hypothyroidism were older than euthyroid patients (p < 0.001).  IHT and hypothyroid patients had a statistically significant higher body mass index (BMI) than euthyroid women (p=0.001 for hypothyroid women, p = 0.035 for IHT). Hypothyroid  and IHT women were more likely to have had a previous lower segment caesarean section  (p=0.043,  and 0.006 respectively). Type 1 diabetes and gestational diabetes  p = 0.012) were more common associated comorbidities in hypothyroid pregnancies. Offspring of patients with IHT had a higher birth weight than those born to euthyroid patients (p=0.009). Patients with hyperthyroidism were found to have a significantly increased risk of early preterm delivery before 34 weeks of gestation and were also more likely to have suspected intrauterine growth restriction and low mean birth weight. We report no significant differences in past history of obstetric loss, antenatal complications, mode of delivery, gestational age at delivery and postpartum haemorrhage rates across thyroid categories.

Available evidence suggests that thyroid dysfunction is more likely in the setting of older age, and higher body mass index. Moreover, it impacts on neonatal birth weight, rates of early preterm delivery and intrauterine growth restriction.