Cognitive and functional predictors and sequelae of postoperative delirium in elderly patients undergoing elective joint arthroplasty

BACKGROUND: Postoperative delirium (POD) is common in the elderly and associated with adverse outcomes. The cognitive and functional sequelae of POD in elective surgical patients are not known. We sought to determine whether (1) lower scores on sensitive neurocognitive tests are an independent risk factor for POD in elderly surgical patients, and (2) POD predicts cognitive and functional decline 3 months postoperatively. METHODS: We conducted a prospective, cohort study on patients ≥65 years old undergoing total hip or knee arthroplasty. Participants underwent preoperative neurocognitive and functional testing. POD was diagnosed using the Confusion Assessment Method. Patients who developed POD and matched controls underwent repeat neurocognitive and functional testing 3 months after surgery. RESULTS: Four hundred eighteen patients met entry criteria, and 42 (10%) developed POD. There were no differences in baseline Mini-Mental State Examination scores, alcohol abuse, depression, and verbal intelligence between groups. Independent predictors of POD included age, history of psychiatric illness, decreased functional status, and decreased verbal memory. For all tests, changes from before to 3 months after surgery were similar between those patients with POD and matched controls. CONCLUSIONS: Subtly reduced preoperative neurocognitive and functional status predict POD. However, in the small group that developed POD, there was no evidence of cognitive and functional decline 3 months after surgery. POD is associated with decreased preoperative cognitive reserve but, in elderly elective surgical patients, may be without adverse cognitive or functional sequelae 3 months postoperatively.