Influence of intraoperative and postoperative factors on the predictive capacity of the delirium risk model for cardiac surgery patients (DELIPRECAS): An observational multicentre study

Adverse intraoperative events such as hypotensive phases [4], acute haemorrhage with anaemia and the need and number of blood transfusions have been associated with an increased risk of postoperative delirium development [1], as well as the postoperative determination of acid-base imbalance [5]. [...]the objective of this study is to evaluate the influence of factors occurring during the intraoperative and postoperative periods on the risk of developing delirium in patients undergoing cardiac surgery, as well as to evaluate whether its inclusion in the model would improve its predictive power. The result of the multivariate backward step analysis was six independent risk factors for the development of delirium: a SOFA score in the 24 first hours ≥8 (odds ratio [OR] 3.06, 95% confidence interval [CI] 1.24–7.56, p = 0.015), a lactic acid value >2.2 mmol/l in the first 24 h (OR 3.54, 95% CI 1.73–7.25, p = 0.001) and the preoperative ones that already comprised the DELIPRECAS model: an MMSE score of 25–26 (OR 7.07, 95% CI 3.01–17.24, p < 0.001) or < 25 (OR 4.68, 95% CI 1.79–12.22, p = 0.002), low physical activity (OR 2.23, 95% CI 1.02–4.87, p = 0.044) and age > 65 years (OR 3.56, 95% CI 1.16–10.9, p = 0.026). According to the Hosmer–Lemeshow test, the fit of the two models is good. [...]this study shows that: (1) when predicting the risk of developing delirium, the consideration of intraoperative and postoperative factors does not substantially improve the predictive capacity of the model and (2) given that its application 24 h after surgery delays the identification of patients with a higher risk of developing delirium and, consequently, the application of preventive measures to avoid it, the efficacy offered by the DELIPRECAS early prediction model is greater.