Early post-operative delirium: are surgery and anaesthesia themselves independent risk factors?

Background and Goal of Study: Aims of the study are to assess the incidence of early post-operative delirium in patients without predisposing factors and the eventual role played by subjects' age and gender, the kind of surgery they underwent and the anaesthetic treatment they received in increasing their risk. Materials and Methods: Following ethics committee approval, an epidemiological, prospective study was conducted on a consecutive cohort of patients admitted within 24 hours after major surgery to our ICU. Exclusion criteria were known predisposing factors for delirium, according to literature. Patients have been assessed daily, using the CAM-ICU, for 72 hours after surgery or until discharge or death. Data are given as means ± standard deviation and relative risk (RR); comparisons between groups of values were performed by unpaired Student's t-test for numeric variables and by χ2-test for univariate variables; a p value <0.05 (95% confidence interval) was considered statistically significant. Results and Discussion: 100 patients were examined, 52 of which were recruited (mean age 64±11.4 years). Incidence of post-operative delirium during the first 3 days was 28%; 86% of cases occurred in the first 24 hours; 87% of patients with delirium manifested the hypoactive form. Risk in patients >64 years old was not significantly different from younger subjects (RR=1.273, χ2=0.284, p>0.5); gender was not a risk factor (RR=1.406, χ2=0.466, 0.5<p<0.10). Cardiac surgery was not associated to an increased risk compared to other surgeries (RR=1.413, χ2=1.587, 0.5<p<0.10). Concerning the induction agent, thiopental was associated to an eight-fold higher risk compared to propofol (RR=8.0, χ2=4.256, 0.05<p<0.02); this increase was not dose-related (p=0.463). Neither the use of muscle relaxants nor the drug administered (pancuronium vs rocuronium) seem to determine a risk difference (χ2=2.33, 0.5<p<0.10), as well as the doses of midazolam (p=0.467)and fentanyl (p=0.484) and the halogenated compound used (isoflurane vs sevoflurane, χ2=3.201, 0.10<p<0.05). Concluslon(s): Early post-operative delirium occurs in more than 1 patient out of 4 without predisposing factors. The incidence peak is in the first 24 hours and the most common form is the hypoactive one. Age and gender do not seem to play a role in the development of this dysfunction. Cardiac surgery itself does not represent a risk factor. With regard to anaesthesia, the only factor statistically associated to a strong increase in the RR is the use of thiopental, regardless of its dose.