Diabetes mellitus inhibits complete recanalization in patients with middle cerebral artery occlusion

Methods: This was a retrospective cohort study. Data from 165 patients, with middle cerebral occlusion before t-PA therapy (from the YAMATO study databank), were retrospectively evaluated. Patients were classified into diabetic (D) or non-diabetic (ND) groups based on the history of diabetes mellitus (DM) or hemoglobin A1c levels of >= 6.5%. Arterial recanalization was assessed using magnetic resonance angiography or digital subtraction angiography at 2 points: 1) early recanalization, within 2 h; 2) delayed complete recanalization at 24 h. Good clinical outcome was defined as modified Rankin Scale score 0-2 at 3 months. Results: A total of 33 (21%) were classified into the D and 127 (79%) in the ND groups. Early recanalization was similarly in the D and ND groups (61% vs. 52%, p = 0.434). However, complete recanalization at 24 h was infrequent in the D group (13% vs. 43%, p = 0.002). Among patients with early recanalization, 4 (22%) of 18 patients in the D group and 32 (56%) of 57 patients in the ND group had complete recanalization at 24 h (p = 0.015); while among those without early recanalization, 17 (30%) in the ND and none in the D groups had complete recanalization at 24 h (p = 0.028). Multivariate regression analysis showed DM was one of the independent negative factors for complete recanalization at 24 h (odds ratio 0.113, 95%CI: 0.027-0.472, p = 0.003). At 3 months, group with complete recanalization at 24 h achieved higher frequency of good outcome (67% vs. 49%, p = 0.046). Conclusion: Diabetes might be a risk factor of incomplete recanalization at 24 h regardless of early recanalization.