A systematic review on the prevalence and management of subclinical hypothyroidism in patients with Type 2 Diabetes Mellitus

Abstract

Background: Subclinical hypothyroidism (SCH) remains one of the most common biochemical manifestations of thyroid dysfunction.1 Similarly, type 2 diabetes mellitus (T2DM) is considered the most common metabolic disorder in clinical practice.2Methods: This is a systematic review to ascertain the prevalence and optimum management approach for thyroid dysfunction in patients with T2DM. We conducted a search on PubMed and Google scholar (Figure 1) for articles published between 2010-2020 using the following keywords: subclinical hypothyroidism, type 2 diabetes mellitus, thyroid diseases, diabetic retinopathy, diabetic nephropathy, and diabetic complications. Results: The prevalence of SCH in T2DM patients ranges from 7.8% to 23 % (average around 13.4%). In comparison, the prevalence of SCH in the general population ranges from 6% to 10%. SCH has a higher prevalence in females, older age >60 years old, long duration of T2DM, positive thyroid autoantibodies, glycated hemoglobin (Hba1c) 8%, and obese patients (risk factors).3 The prevalence of SCH in patients with diabetic retinopathy (DR) ranges from 17.3% to 43.3%. Also, the prevalence of SCH among patients with diabetic nephropathy (DN) ranges from 18.1% to 36%. Screening for thyroid dysfunction in T2DM at diagnosis is recommended and justified (Table 1). If patients with SCH have a risk of cardiovascular disease (CVD), DR, DN, symptomatic, presence of goiter, pregnant women, and continuous sustained increases of thyroid-stimulating hormone (TSH) on follow-up, it is recommended to start low dose levothyroxine which improves morbidity. Conclusion: The prevalence of SCH is relatively high in T2DM patients. This supports routine screening of such patients for thyroid dysfunction especially in patients with risk factors and diabetic complications (DR, DN), and consideration of thyroxine replacement wherever warranted although more evidence from randomized controlled trials is needed to explore the possible causal relationships.