Background: An increased prevalence of thyroid dysfunction is reported in both type 1 and type 2 diabetes mellitus compared to the normal population. However, the cost-effectiveness or otherwise of screening for thyroid dysfunction in this group of patients is an unresolved question.
Aims: The aim of this study was to measure the effectiveness of screening for hypo- and hyperthyroidism, both sub-clinical and overt, in an ethnically diverse population of mainly type 2 diabetic patients referred to the hospital for the first time.
Methods: All new referrals to the Diabetes Clinic at our hospital are screened for thyroid dysfunction. We looked retrospectively at a total of 959 patients screened over a four-year period. Serum TSH level was measured in each patient and free T4 levels measured only if the TSH was abnormal. For each patient with an abnormal TSH, the past medical history of the patient was reviewed to determine whether or not the thyroid dysfunction was newly diagnosed by screening.
Results: patients (6.8%) had a suppressed TSH with a median free T4 of 19 pmol/L. Screening 959 patients picked up 3 undiagnosed cases of overt hyperthyroidism and 32 (3.3%) cases of sub-clinical hyperthyroidism. 25 patients (2.6%) had an elevated TSH, with a median free T4 of 13 pmol/L. Screening 959 patients picked up 5 undiagnosed cases of overt hypothyroidism and 13 (1.3%) cases of sub-clinical hypothyroidism.
Conclusion: Our results confirm a higher prevalence of thyroid dysfunction (especially sub-clinical hyperthyroidism) in our diabetic population (40% South Asians & Afro-Caribbean), compared to that reported in the general population. Also, a few cases of frank hyper- and hypothyroidism have been picked up routinely. Screening with TSH measurement is not expensive. However, the cost-effectiveness or otherwise of screening diabetic patients for thyroid dysfunction is likely to depend on a number of factors like local prevalence, ethnicity, as well as on the question of whether sub-clinical hypo- and hyperthyroidism should be treated, which is itself is not fully resolved. Higher than expected prevalence of sub-clinical hyperthyroidism in our ethnic patch of diabetic patients will need further evaluation.