Metformin treatment and acute ischemic stroke outcomes in patients with type 2 diabetes: a retrospective cohort study

Background and purpose Preclinical studies have shown that metformin has neuroprotective actions in stroke. However, the optimal treatment timing and duration remain unknown. Herein, we examined the efficacy of metformin treatment on prognosis in acute ischemic stroke (AIS) patients, and assessed the optimal treatment timing and duration. Methods AIS patients with type 2 diabetes mellitus were retrospectively enrolled. Patients were grouped into those who never received metformin (MET − group), those who received metformin continuously before stroke and after admission (pre-stroke + /post-stroke + group), those who only received metformin before stroke onset (pre-stroke + /post-stroke − group), and those who only received metformin after admission (pre-stroke − /post-stroke + group). The all MET + group represents the sum of the three metformin treatment groups. The efficacy outcome was the 90-day modified Rankin Scale (mRS) score. Results In total, 309 eligible patients were included (MET − [ N = 130], pre-stroke + /post-stroke + [ N = 94], pre-stroke + /post-stroke − [ N = 30], pre-stroke − /post-stroke + [ N = 55]; all MET + [ N = 179]). The all MET + group had a trend toward a lower 90-day mRS score compared with that in the MET − group (1 [0–2] vs 1 [0–3], unadjusted odds ratio [OR] = 0.652, P = 0.041; adjusted OR = 0.752, P = 0.218). In the three metformin treatment groups, only the pre-stroke + /post-stroke + group had a significantly lower 90-day mRS score (1 [0–1] vs 1 [0–3], adjusted OR = 0.497, 95%CI = 0.289–0.854; P = 0.011) and higher proportion of mRS score 0–1 (78.7% vs. 61.5%, adjusted OR = 2.278, 95%CI = 1.108–4.680; P = 0.025) than the MET − group. Conclusion AIS patients with type 2 diabetes mellitus who receive continuous metformin treatment before stroke onset and after admission have improved functional outcome at 90 days.