This study evaluated the relationship between the preoperative alkaline phosphatase (ALP) level and major postoperative neurological complications in patients undergoing cerebral bypass surgery. This was a retrospective analysis of a prospective database of all patients undergoing cerebral bypass surgery after a diagnosis of cerebrovascular stenosis or occlusion between May 2003 and August 2017. The patients were divided into tertiles based on serum alkaline phosphatase (ALP) levels (low: <63, intermediate: 63~79, and high: ALP > 79 IU/mL). The incidence of neurological events according to ALP level was analyzed. The study analyzed 211 cases. The incidence of acute infarction was highest in the third serum ALP tertile (5.7% vs. 2.9% vs. 16.9% in the first, second, and third tertile, respectively, p = 0.007). Logistic regression analysis showed that the third tertile of serum ALP was an independent predictor of acute cerebral infarction (odds ratio 3.346, 95% confidence interval 1.026−10.984, p = 0.045). On Kaplan−Meier time-to-event curves, the incidence of acute infarction increased significantly with ALP (log rank = 0.048). Preoperative serum ALP level can be used as a biomarker to predict acute cerebral infarction in patients undergoing cerebral bypass surgery for vascular stenosis or occlusion.