Postoperative delirium. A review of 80 primary data-collection studies.

We conducted an on-line search and manual searches for 1966 through 1992 to determine the incidence, diagnosis, risk factors, and treatment of postoperative delirium. Of the 374 citations found, 277 articles were excluded after criteria of relevance were applied. After methodologic criteria for validity were applied to the remaining 80 articles, 26 studies were retained for the final information synthesis. The incidence of postoperative delirium was 36.8% (range, 0% to 73.5%). Primary reasons for this disparity were insufficient sample size and inconsistent application of numerous diagnostic tools. One study provided statistically significant data that demonstrated that postoperative delirium is underdiagnosed by physicians and nurses. Four of the articles that met the established criteria provided risk factor data. Although age, preoperative cognitive impairment, and the use of anticholinergic drugs were significantly associated with postoperative delirium, gender, type and route of anesthesia, and sleep deprivation were not. Two studies demonstrated a decreased incidence of postoperative delirium when patients underwent preoperative psychiatric counseling or participated in a structured perioperative program. These findings indicate a need for (1) accurate incidence data with further definition of risk factors and (2) studies that address the diagnosis and treatment of this common postoperative problem.