9th Annual Meeting of the European Delirium Association location:Cremona, Italy date:6-7 November 2014
BACKGROUND:State anxiety, a temporary and acute reaction with feelings of tension and apprehension, and its relation to postoperative delirium has only been investigated in a few studies. However, inadequate statistical power and less reliable delirium assessment have hampered any conclusion. The aim of this study is therefore to determine if preoperative state anxiety is a risk factor for postoperative delirium.
METHODS:A secondary analysis comprising data from two studies was performed. Patients (≥65 years) admitted for elective cardiac surgery or presenting with a hip fracture at the emergency department were included. State anxiety was measured preoperatively using the State-Trait Anxiety Inventory (STAI). Delirium was measured pre- and postoperatively by trained research nurses using the Confusion Assessment Method (CAM). The severity of delirium was measured using the Delirium Index.
RESULTS:Eighty six patients with a hip fracture and 83 patients undergoing cardiac surgery were included (n=169). Forty two patients (24.9%) developed delirium postoperatively (24 (27.9%) hip fracture patients and 18 (21.7%) cardiac surgery patients). State anxiety was not associated with postoperative delirium (rb=0.027, p=0.267), its duration (rho=-0.19, p=0.906) or its severity (rho=0.183, p=0.278). State anxiety was not an independent predictor of postoperative delirium (OR 1.06 (0.89-1.25)) using multivariate logistic regression analysis. Independent predictors for the total sample were MMSE, lowest intraoperative diastolic blood pressure and valve or osteosynthesis surgery. A subgroup analysis identified MMSE, osteosynthesis surgery and lowest intraoperative diastolic blood pressure as independent predictors among hip fracture patients and valve surgery, intraoperative systolic blood pressure and number of home medications among cardiac surgery patients.
CONCLUSIONS:No significant relationship between state anxiety and delirium was found. However, methodological problems, i.e. considerable missing data for state anxiety, multicollinearity and low statistical power was observed and should be taken into consideration for further research.