Objective : Postoperative delirium (POD) is characterized by an acute change in cognitive function and can result in longer hospital stays, higher morbidity rates, and more frequent discharges to long-term care facilities. In this study, we investigated the incidence and risk factors of POD in 224 patients older than 70 years of age, who had undergone a neurosurgical operation in the last two years. Methods : Data related to preoperative factors (male gender, >70 years, previous dementia or delirium, alcohol abuse, serum levels of sodium, potassium and glucose, and co-morbidities), perioperative factors (type of surgery and anesthesia, and duration of surgery) and postoperative data (length of stay in recovery room, severity of pain and use of opioid analgesics) were retrospectively collected and statistically analyzed. Results : POD appeared in 48 patients (21.4%) by postoperative day 3. When we excluded 26 patients with previous dementia or delirium, 17 spontaneously recovered by postoperative day 14, while 5 patients recovered by postoperative 2 months with medication, among 22 patients with newly developed POD. The univariate risk factors for POD included previously dementic or delirious patients, abnormal preoperative serum glucose level, pre-existent diabetes, the use of local anesthesia for the operation, longer operation time (>3.2 hr) or recovery room stay (>90 min), and severe pain (VAS>6.8) requiring opioid treatment (p<0.05). Backward regression analysis revealed that previously dementic patients with diabetes, the operation being performed under local anesthesia, and severe postoperative pain treated with opioids were independent risk factors for POD. Conclusion : Our study shows that control of blood glucose levels and management of pain during local anesthesia and in the immediate postoperative period can reduce unexpected POD and help preventing unexpected medicolegal problems and economic burdens.