Predicting Role of Prestroke Glycemic Variability Estimated by Glycated Albumin for Reperfusion and Prognosis after Endovascular Treatment

Introduction: Glycated hemoglobin is widely used for the diagnosis of diabetes, but it is not accurately correlated with blood glucose fluctuations. We evaluated the impact of prestroke glycemic variability, measured by glycated albumin (GA) on reperfusion rate and stroke outcomes after endovascular treatment (EVT). Methods: We consecutively collected 310 EVT-treated patients for 60 months using a multicenter registry database. Primary outcome was unsuccessful reperfusion defined by modified Thrombolysis in Cerebral Infarction grade 0 to 2a. Secondary outcomes were occurrence of early neurologic deterioration (END), symptomatic hemorrhagic transformation (SHT) and a 3-month poor outcome (modified Rankin Scale >2). GA was measured in the morning after hospital admission with overnight fasting and determined to reflect high prestroke glycemic variability (GA >= 16.0%). Results: Over the median follow-up of 60 months of 310 patients, there were 64 (20.6%) events of unsuccessful reperfusion, 66 (21.3%) of END, 21 (6.8%) of SHT, and 180 (58.1%) of 3-month poor outcome. In the higher GA group (130, 41.9%), proportion of unsuccessful reperfusion, END, SHT, and poor outcome were higher than lower GA group. The multivariate analysis showed that higher GA was associated with unsuccessful reperfusion after EVT (adjusted odds ratio 4.13; 95% confidence interval [CI], 1.93-8.85). The area under the receiver operating characteristic of GA (0.644; 95% CI: 0.634-0.740) for predicting poor outcome was better than that of glycated hemoglobin (0.586; 95% CI: 0.529-0.642, p for DeLong's pairwise comparison = 0.005). Conclusion: GA, reflecting prestroke glycemic variability, could be a reliable parameter for predicting reperfusion rate and acute ischemic stroke outcome in this study.