Thyroid autoimmunity in Type 1 diabetes: systematic review and meta‐analysis

Abstract

To review the incidence and evidence for screening for thyroid autoimmunity and thyroid dysfunction in Type 1 diabetes. Systematic review and meta-analysis. Inclusion criteria were prospective cohort studies screening for thyroid autoimmunity and/or dysfunction (defined as an abnormal thyroid-stimulating hormone level) in Type 1 diabetes. Exclusion criteria included pregnancy and thyroid dysfunction before diabetes onset. Outcomes examined were: incidence of thyroid autoimmunity and/or dysfunction; association between thyroid autoimmunity and dysfunction; and cost-effectiveness. Data sources were MEDLINE, EMBASE, the Cochrane Library, manual searching and contact with authors, with limitations to English language and human studies. Meta-analysis was performed using random effects models. We identified 14 eligible studies, involving 2972 young people and 789 adults with Type 1 diabetes. Follow-up ranged from 1-18 years. None of the studies were of good methodological quality (Newcastle Ottowa Scale score > 7). The incidence of thyroid dysfunction (11 studies) ranged from 27 (95% CI 15-45) to 246 (95% CI 118-453) per 10 000 patient-years and thyroid autoimmunity (four studies) from 13 (95% CI 0.3-71) to 326 (95% CI 194-510). The risk of thyroid dysfunction was higher in those with thyroid autoimmunity: summary risk ratio 25 (95% CI 9-71) and was higher in children (49, 95% CI 16-150) compared with adults (7, 95% CI 3-13). No studies examined cost-effectiveness of screening. There is a markedly increased risk of thyroid dysfunction in people with Type 1 diabetes and thyroid autoimmunity. The optimal method or frequency of screening could not be determined from available data. Future studies should examine whether screening improves clinical outcomes in this population.