A simple model for predicting postoperative delirium in older patients undergoing elective orthopedic surgery

OBJECTIVE: To determine the incidence, and severity of postoperative delirium (POD) in older patients undergoing elective orthopedic procedures and to identify potential pre-operative risk factors. DESIGN: A prospectively studied cohort of elective orthopedic surgery patients. SETTING: A University teaching hospital. PATIENTS: Eighty patients who attended a preadmission clinic and, subsequently, underwent elective orthopedic surgery. All patients were aged 60 years or older, and all spoke English. MEASUREMENTS: Patients underwent preoperative medical, cognitive, and activities of daily living assessment with standardized instruments and were followed postoperatively with daily visits, telephone interviews with attending nurses using a modified confusion assessment model (CAM), and repeated cognitive testing. Suspected delirium was verified by direct physician assessment. RESULTS: The elective group had 14 (17.5%) cases of POD, of which six (7.5%) were severe. These incidences are low compared with those of nonelective surgery groups reported elsewhere in the literature. Stepwise multiple logistic regression identified two POD risk factors: clock-drawing scores ≤ 6 (OR = 9.0, CI, 2.8 to 45.6) and male gender (OR = 5.6, CI 1.9 to 33.8). CONCLUSION: A simple model using clock-drawing scores and male gender for preoperative identification of elective patients at greatest risk for POD appears sensitive, predictive, and practical for the preadmission clinic setting, but it should be validated in a prospective trial.