• E R Marcantonio, G Juarez, L Goldman, C M Mangione, L E Ludwig, L Lind, N Katz, E F Cook, E J Orav, and T H Lee.
• Section for Clinical Epidemiology, Brigham and Women's Hospital, Boston, MA 02115.
• JAMA. 1994 Nov 16;272(19):1518-22.

ObjectiveTo examine the role of medications with known psychoactive properties in the development of postoperative delirium.DesignNested case-control study within a prospective cohort study.SettingGeneral surgery, orthopedic surgery, and gynecology services at Brigham and Women's Hospital, Boston, Mass.PatientsCases (n = 91) were patients enrolled in a prospective cohort study who developed delirium during postoperative days 2 through 5. One or two controls (n = 154) were matched to each case by the calculated preoperative risk for delirium using a predictive model developed and validated in the prospective cohort study.Main Outcome MeasuresMedication exposures were ascertained from the medical record by a reviewer blinded to the study hypothesis. Exposures to narcotics, benzodiazepines, and anticholinergics were recorded for the 24-hour period before delirium developed in the 91 cases and for the same 24-hour postoperative period for the 154 matched controls.ResultsDelirium was significantly associated with postoperative exposure to meperidine (odds ratio [OR], 2.7; 95% confidence interval [CI], 1.3 to 5.5) and to benzodiazepines (OR, 3.0; 95% CI, 1.3 to 6.8). Meperidine had similar associations with delirium whether administered via epidural or patient-controlled routes, although only the epidural route reached significance (OR, 2.4; 95% CI, 1.3 to 4.4; OR, 2.1; 95% CI, 0.4 to 10.7, respectively). For benzodiazepines, long-acting agents had a trend toward stronger association with delirium than did short-acting agents (OR, 5.4; 95% CI, 1.0 to 29.2; vs 2.6; 1.1 to 6.5), and high-dose exposures had a trend toward slightly stronger association than low-dose exposures (OR, 3.3; 95% CI, 1.0 to 11.0; vs 2.6; 0.8 to 9.1). Neither narcotics (OR, 1.4; 95% CI, 0.5 to 4.3) nor anticholinergic drugs (OR, 1.5; 95% CI, 0.6 to 3.4) were significantly associated with delirium as a class, although statistical power was limited because of the high use of narcotics and the low use of anticholinergics in the study population.ConclusionsClinicians caring for patients at risk for delirium should carefully evaluate the need for meperidine and benzodiazepines in the postoperative period and consider alternative therapies whenever possible.

26 keywords...

hide…