Incidence, risk factors, prediction and long-term effects of cognitive impairment and/or confusional states after elective orthopedic surgery

Research project P 13624 Predicition of prostoperative confusional states Peter FISCHER 11.10.1999 Background: Delirium, or acute confusional state, is a common postoperative complication with higher mortality and complication rates, poor functional recovery, and longer length of stay. As improvements in surgical and anesthetic techniques allow older and sicker patients to undergo procedures, the incidence of postoperative delirium is likely to rise and knowledge on risk factors as a basis of prediction and prevention of delirium is urgently needed. Despite its clinical importance and implications for health policy, delirium has remained a relatively neglected subject of research. Objective: The purpose of this study is fourfold: 1) to determine the incidence of delirium after elective orthopedic surgery, 2) to contrast the clinical outcomes of patients with and without delirium, 3) to identify demographic, clinical, and psychological risk factors of postoperative delirium, and 4) to develop and validate a predictive model for postoperative delirium. Point 1 includes the description and incidence of incomplete manifestations of postoperative delirium. Point 2 will answer the important question of the influence of deliria on development or accelleration of development of age-related cognitive decline. Thus, the study wants to give first evidence to the question of whether delirium influences age-associated cognitive decline (Alzheimer`s disease). Point 3 includes the important questions which type of anesthesia (general versus epidural) is followed by smaller cognitive deficits, which perioperative variables (anoxia, hypovolaemia, medications, pain, etc.) raise the vulnerability for delirium and whether incipient (very mild) Alzheimer`s disease favours the occurrence of postoperative delirium as a baseline risk factor. Design: Prospective cohort analytic study. Using DSM-IV criteria, three standardized scales , and a battery of neuropsychological tests for the detection of incomplete and complete manifestations of delirium, we will follow up a cohort of elderly patients undergoing planned hip-arthroplasty or knee-arthroplasty due to coxarthrosis or gonarthrosis, respectively. We will collect risk factor data from baseline, and perioperative factors from orthopedic and anesthetic protocols and factors from the postoperative period. Setting: Orthopedic and Psychiatric Department (Memory Clinic) of the University Hospital of Vienna. Patients: 180 plus 60 (validation cohort) elderly patients undergoing elective orthopedic surgery (hip-replacement or knee-replacement). Statistical analysis: In bivariate analysis, rates of delirium will be calculated when each risk factor was present or absent. Crude relative risks will be determined as the number of events (delirium) per person in the group with the risk factor present relative to the number of such events per person in the group without risk factor. Variables with relative risk estimates of 1,5 or greater will be selected for evaluation in a logistic regression analysis. The predictive model for delirium will then be cross-validated in another cohort.