Introduction
Hypothyroidism is the most common pregnancy-related thyroid condition, associated with an increased risk of miscarriage, preterm delivery, and offspring intellectual impairment. Such risks may be reduced through timely diagnosis and thyroxine initiation. To streamline access to Endocrinology services for women with thyroid dysfunction in pregnancy, endocrinologists at Sutherland Hospital implemented a dedicated referral and management algorithm in February 2021.
Aims:
To compare the proportion of referred patients to the Sutherland Hospital Gestational Thyroid Clinic who achieved target thyroid stimulating hormone (TSH) during pregnancy, before and after algorithm implementation. Secondary outcomes included changes in (a)number of patients commenced on levothyroxine prior to initial appointment, (b)median gestational age of initial specialist consultation, (c)rate of guideline-appropriate investigations and management, and (d)perinatal outcomes.
Methods:
A retrospective clinical audit of electronic medical records was performed. The first fifty consecutive patients referred to the Sutherland Hospital Gestational Thyroid Clinic with hypothyroidism (defined as TSH>2.5mIU/L or TSH values outside the reference range for the pregnancy trimester) between 1.04.2020 to 1.09.2020 (pre-intervention) and 1.04.2021 to 1.09.2021 (post-intervention) formed the two sample populations for comparison. Demographic, clinical, biochemical, and pregnancy outcome-related data were compared.
Results:
All examined women achieved target TSH during pregnancy with a median final TSH of 1.6mIU/L (IQR: 1.2-2.3mIU/L). A similar proportion of women who had an initial raised TSH were prescribed thyroxine prior to their first clinic appointment across both groups (p=0.83). After introducing the intervention tool, the first TSH measurement occurred earlier in pregnancy (median 5.5 vs 6.5 weeks, p=0.011) with a subsequent earlier specialist review (median 22 weeks in 2020, 19 weeks in 2021, p=0.032). Significantly more women with an elevated TSH underwent thyroid autoantibody testing post-intervention (55.5% vs 78%, p=0.035). There was no significant difference in perinatal outcomes. Interestingly, gestational diabetes was more common in the pre-intervention group (24% vs 4%, p=0.011). However, our study did not aim to examine the casual inference between thyroid dysfunction and gestational diabetes, noting hypothyroxinaemia is considered a relative risk factor for metabolic syndrome.
Conclusions:
While measures of service effectiveness (proportion of referred patients achieving target TSH in pregnancy) remained unchanged following the implementation of a dedicated referral and management tool, some measures of service efficiency increased. In particular, initial TSH measurement occurred earlier in pregnancy, followed by earlier endocrinologist specialist review, and increased detection of thyroid autoantibodies. Further studies are needed to establish recommendations for screening for thyroid dysfunction in patients with gestational diabetes.