(069) Safety of concomitant use of antidiabetic drugs and oral anticoagulants: a systematic review of observational studies

Background: Several observational studies have recently assessed the safety of drug-drug interactions between antidiabetic drugs and oral anticoagulants (OACs).

Objectives: To systematically assess and synthesize current observational evidence on the risk of adverse clinical outcomes associated with concomitant use of antidiabetic drugs and OACs.

Methods: We systematically searched MEDLINE and EMBASE up to January 2024 for observational studies (cohort, [nested] case-control, case-only) assessing the risk of adverse clinical outcomes (hypoglycemia, bleeding) associated with concomitant use of any class of antidiabetic drugs and any class of OACs. The risk of bias was assessed using the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool.

Results: We identified 4 cohort studies and 2 self-controlled case series with an overall population of 1,044,507 patients; 5 studies focused on severe hypoglycemia and 1 on major bleeding. Regarding severe hypoglycemia, concomitant use of sulfonylureas (SU) and warfarin was associated with an up to 47% increased risk compared to SU use alone (n=4); results were inconsistent when compared to concomitant use of SU and direct oral anticoagulants (DOACs) (n=2). Concomitant use of SU and DOACs was associated with a 46% decreased risk of severe hypoglycemia compared to SU use alone (n=1). Concomitant use of metformin and warfarin was associated with a 73% increased risk of severe hypoglycemia compared to metformin use alone (n=1) and a 37% increased risk when compared to concomitant use of metformin and DOACs (n=1). Concomitant use of dipeptidyl peptidase-4 inhibitors (DPP-4i) and warfarin was associated with a 37% increased risk of severe hypoglycemia compared to concomitant use of DPP-4i and DOACs (n=1). Regarding major bleeding, concomitant use of SU and warfarin was not associated with a change in the risk compared to warfarin use alone (n=1), but concomitant use of metformin and warfarin was associated with a 15% decreased risk compared to warfarin use alone (n=1). According to ROBINS-I, 3 studies were at moderate, 1 at serious, and 2 at critical risk of bias. Biases included selections bias due to prevalent drug use, outcome-dependent censoring, and application of exclusion criteria after cohort entry, and information bias due to exposure and outcome misclassification.

Conclusions: Most studies have focused on the hypoglycemic risk of concomitant use of OACs and older antidiabetic drugs suggesting potential benefits for DOACs over warfarin. However, results should be interpreted cautiously given the high degree of methodological heterogeneity and several possible biases. Evidence on the hypoglycemic risk of concomitant use of OACs and newer antidiabetic drugs or on the bleeding risk of any combination is scarce.