Prevalence of and risk factors for postoperative delirium among children after cardiac surgery in a Single-Centre retrospective study

Abstract

Due to the increasing focus on neurodevelopment in children with congenital heart disease (CHD), early predictive markers are crucial for implementing interventions and improving neurodevelopmental outcomes. As postoperative delirium (PD) is known to have a long-term impact on neurocognitive function in adults, studies on the prevalence of and modifiable risk factors for PD offer new perspectives. We conducted a retrospective, single-centre study screening for PD using the Cornell Assessment of Pediatric Delirium (CAPD). We distinguished PD from iatrogenic withdrawal syndrome (IWS) by using the Withdrawal Assessment Tool 1 (WAT-1). A confirmatory, multivariate regression analysis was performed and included various pre-, intra-, and postoperative variables. The screening compliance rate was 95% among the 311 patients. The prevalence of PD was 40.2%, and 46.4% of the patients developed IWS. Infants were at the highest risk for PD (OR 2.9, p = 0.05). Prolonged mechanical ventilation > 100 h (OR 7.4, p = 0.003), infusion therapy with ketamine (OR 3.3, p = 0.009), IWS (mild: OR 7.7, p = < 0.001, severe: OR 17.0, p = < 0.001) and low cardiac output syndrome (LCOS) (OR 3.9, p = 0.02) were significant predictive risk factors for PD. Overall, PD and IWS are highly prevalent in paediatric cardiac intensive care units (pCICUs), especially in infants and children with prolonged ventilation durations who require multiple sedatives. This is one of the most extensive single-centre studies in the pCICU population, and the results revealed that IWS and lactatemia in the context of LCOS are novel predictors of PD.

Keywords: Congenital heart disease (CHD); Cornell assessment of pediatric delirium (CAPD); Iatrogenic withdrawal syndrome (IWS); Neurodevelopment; Paediatric cardiac intensive care unit (pCICU); Paediatric delirium.

Fig. 1

Duration of delirium in all delirious patients (n = 118). Patients were screened positive for delirium when the CPAD score was ≥ 9 points for the first time (time of delirium hour 0). CAPD was performed as soon as the patient was awake (RASS ≥-2). CAPD: Cornell Assessment of Pediatric Delirium, RASS: Richmond Agitation and Sedation Scale.

Fig. 2

Multivariate logistic regression model for predicting postoperative delirium in children after cardiac surgery with or without bypass (n = 293). OR: Odds Ratio, CI: Confidence interval, WAT-1: Withdrawal Assessment Tool 1, IWS: Iatrogenic withdrawal syndrome, OTE: On-table extubation, RACHS-1: Risk Adjustment for Congenital Heart Surgery, CPBT: Cardiopulmonary bypass time.

Fig. 3

COMFORT-B scores for ventilated patients who were transferred to the pCICU (n = 154). The reference COMFORT-B score of 11–22 points indicates an adequate level of sedation while the patient is intubated. The patient was considered to be oversedated if the score was less than 11 points, with more than 22 points undersedated. OR: odds ratio, CI: Confidence interval.

Fig. 4

Different continuous infusion therapies predict delirium in children who remain ventilated after cardiac surgery (n = 146). Dex: Dexmedetomidine, OR: Odds ratio, CI: Confidence interval.

Fig. 5

Subgroup analysis of patients with different causalities for postoperative lactatemia and delirium (n = 83) OR: Odds ratio, CI: Confidence interval.