Aim: This umbrella review summarises and compares synthesised evidence on the impact of subclinical hypothyroidism and its management on long-term clinical outcomes.
Methods: We conducted comprehensive searches on MEDLINE, EMBASE, Scopus, Web of Science, Cochrane Database of Systematic Reviews, JBI Evidence Synthesis, the PROSPERO register, Epistemonikos Database and PDQ Evidence from inception to February and July 2021 using keywords on subclinical hypothyroidism, treatment with levothyroxine, monitoring and primary outcomes (all-cause mortality, cardiovascular events, stroke, frailty fractures and quality of life). Only systematic reviews and meta-analyses on adult patient populations were considered. Study selection, data extraction and quality appraisal using AMSTAR-2 were done independently by two reviewers and discrepancies were resolved through discussion. Overlap across the selected reviews was also assessed, followed by a narrative synthesis of findings.
Results: A total of 763 studies were identified from literature searches; 20 reviews met inclusion criteria. Methodological quality ratings were high (n = 8), moderate (n = 7), and low (n = 5), but no reviews were excluded on this basis. Though there was slight overlap across all reviews, some pairwise comparisons had high corrected covered area scores. Compared to euthyroidism, untreated subclinical hypothyroidism was associated with a higher risk of cardiovascular events or death if Thyroid Stimulating Hormone was above 10mIU/L at baseline. Treatment was associated with a lower risk of death from all causes for patients younger than 70 years and possibly better cognitive and quality of life scores than untreated individuals. Evidence on the risk of strokes and fractures was inconclusive.
Conclusion: In the long term, treatment of subclinical hypothyroidism may be beneficial for some patient groups. However, the findings of this review are negatively impacted by the relative sparseness and poor quality of available evidence. Additional large and adequately powered studies are needed to investigate this topic further.
Systematic review registration: PROSPERO (CRD42021235172).