Background: Although antihyperglycemic pharmacotherapy in frail older adults with type 2 diabetes mellitus (T2DM) is challenging, recommendations from international guidelines are mainly based on indirect evidence from trials not including frail participants.

Objective: This systematic review investigated the effectiveness and safety of pharmacotherapy in frail older adults with T2DM.

Material and methods: Randomized (RCT) and non-randomized prospective clinical trials (non-RCT) were searched in three electronic databases (Medline, Embase, Central) up to October 2018. Trials in older adults with T2DM who were assessed as significantly or severely impaired by defined cut-off scores of assessment instruments on frailty, activities of daily living or physical functional impairment were included.

Results: Two reviewers independently screened 17,391 references for inclusion and assessed risk of bias with ROBINS‑I. Five non-RCTs and no RCT were identified. Treatment of T2DM without insulin compared to insulin could be associated with increased improvement in cardiac functions in patients with cardiac resynchronization therapy and with decreased falls in frail older women. While better glycemic control with low variability and low HbA1c (hemoglobin A1c) values (<7%) was associated with better maintenance of physical function in community-dwelling older persons, higher HbA1c values (8-8.9%) were associated with a reduction in the composite outcome of death or functional decline in community-dwelling diabetic older adults with need for skilled assistance. Due to serious risk of bias in all studies, results should be considered with caution.

Conclusion: Well-designed, large-scale RCTs including this important group of patients are required to assess the effectiveness and safety of pharmacotherapy and HbA1c targets.