Risk assessment for postoperative delirium: Cognitive impairment as major factor

Abstract

AbstractBackgroundElderly patients show an increased risk of postoperative delirium. Cognitive impairment constitutes a risk factor for delirium, whereas delirium can also accelerate cognitive decline. To improve prevention and therapy of delirium, planning of surgical and anaesthetic measures could integrate a range of psychiatric, psychosocial and somatic risk factors. Here, the predictive value of risk factors was investigated.MethodIn a longitudinal design, 120 patients aged 65 years and older were included. Risk factors were assessed before surgical interventions with a questionnaire for risk factors, the TE4D as a screening tool for cognitive impairment and depression and Reisberg’s GDS staging of severity of cognitive impairment. Postoperative prevalence of delirium was assessed with the Confusion Assessment Method for the Intensive Care Unit (CAM‐ICU). Pearson correlation was calculated for metric variables. If significant (p<.05), metric variables were transformed into dummy variables by calculating the highest possible odds ratio for delirium. For nominal variables, odds ratio for delirium was calculated. A stepwise binary logistic regression was conducted.ResultPrevalence of postoperative delirium was 13.3%. Results of the Pearson correlation indicated neither significant association between postoperative delirium and age (r(118) = .089, p > .05) nor education (r(118) = ‐.002, p > .05). Postoperative delirium and duration of general anaesthesia were found to be correlated (r(118) = .33, p < .01). Logistic regression of 20 risk factors (Table 1) resulted in a 6‐factor‐model including preoperative cognitive impairment (OR 33.67), severe disease (OR 23.05), premedication with narcotics or benzodiazepine (OR 9), fracture (OR 5.77), duration of general anaesthesia above 160 minutes (OR 5.56) and male gender (OR 2.29). For the resulting model predictive sensitivity was 1, specificity was .93, Youden index J=.93, accuracy .94.ConclusionFor the risk of delirium in elderly surgical patients, preoperative cognitive impairment demonstrated the highest odds ratio followed by somatic predisposition. To decrease the risk of delirium, the result of calculating the risk factors will give the chance to adapt anaesthetic and surgical procedures. A screening tool, working title Computation of Risk for Delirium (COR‐D), including a short version based on the presented data will be provided and further validated.