Benefits and Harms of Deprescribing Antihyperglycemics for Adults with Type 2 Diabetes: A Systematic Review.

OBJECTIVES:Contemporary guidelines suggest relaxed glycemic targets in populations with type 2 diabetes mellitus (T2DM) at risk of hypoglycemia, including people with multimorbidity, limited life expectancy or frailty. However, overtreatment remains commonplace. To inform safe deprescribing, a previous systematic review investigated the benefits and harms of deprescribing antihyperglycemics, but identified only limited, very low-quality evidence. We sought to update that review and identify and describe newly published literature on the effects of deprescribing antihyperglycemics in older adults with T2DM.METHODS:We searched MEDLINE, EMBASE and the Cochrane Library (July 2015 to January 2021) for controlled studies published in English addressing the effects of deprescribing vs continuing antihyperglycemics in adults with T2DM. Two independent reviewers performed title, abstract and full-text screening, data extraction and risk-of-bias assessment. Cochrane's risk-of-bias tools, RoB 2 and ROBINS-I, were used. The findings were summarized narratively. GRADE (Grading of Recommendations, Assessment, Development and Evaluations) was used to evaluate the evidence.RESULTS:We identified 4 additional investigations-2 randomized controlled trials and 2 retrospective cohort studies. After deprescribing, 3 studies reported no clinically significant changes in glucose management and 2 studies reported reductions in adverse events (e.g. hypoglycemia, all-cause mortality and nonspine fractures). However, based on GRADE assessment, we found very low certainty of the evidence due to concerns of risk of bias (e.g. unmeasured confounding), imprecision, and indirectness.CONCLUSIONS:Deprescribing antihyperglycemic medications in older adults with T2DM is likely feasible and safe, and benefits may outweigh the harms. However, the evidence indicates very low certainty. Additional deprescribing studies are needed with rigorous methodologies and reporting.