Prevalence of ICU Delirium in Postoperative Pediatric Cardiac Surgery Patients.

OBJECTIVES: The objective of this study was to determine the prevalence of ICU delirium in children less than 18 years old that underwent cardiac surgery within the last 30 days. The secondary aim of the study was to identify risk factors associated with ICU delirium in postoperative pediatric cardiac surgical patients. DESIGN: A 1-day, multicenter point-prevalence study of delirium in pediatric postoperative cardiac surgery patients. SETTING: Twenty-seven pediatric cardiac and general critical care units caring for postoperative pediatric cardiac surgery patients in North America. PATIENTS: All children less than 18 years old hospitalized in the cardiac critical care units at 06:00 on a randomly selected, study day. INTERVENTIONS: Eligible children were screened for delirium using the Cornell Assessment of Pediatric Delirium by the study team in collaboration with the bedside nurse. MEASUREMENT AND MAIN RESULTS: Overall, 181 patients were enrolled and 40% (n = 73) screened positive for delirium. There were no statistically significant differences in patient demographic information, severity of defect or surgical procedure, past medical history, or postoperative day between patients screening positive or negative for delirium. Our bivariate analysis found those patients screening positive had a longer duration of mechanical ventilation (12.8 vs 5.1 d; p = 0.02); required more vasoactive support (55% vs 26%; p = 0.0009); and had a higher number of invasive catheters (4 vs 3 catheters; p = 0.001). Delirium-positive patients received more total opioid exposure (1.80 vs 0.36 mg/kg/d of morphine equivalents; p < 0.001), did not have an ambulation or physical therapy schedule (p = 0.02), had not been out of bed in the previous 24 hours (p < 0.0002), and parents were not at the bedside at time of data collection (p = 0.008). In the mixed-effects logistic regression analysis of modifiable risk factors, the following variables were associated with a positive delirium screen: 1) pain score, per point increase (odds ratio, 1.3; 1.06-1.60); 2) total opioid exposure, per mg/kg/d increase (odds ratio, 1.35; 1.06-1.73); 3) SBS less than 0 (odds ratio, 4.01; 1.21-13.27); 4) pain medication or sedative administered in the previous 4 hours (odds ratio, 3.49; 1.32-9.28); 5) no progressive physical therapy or ambulation schedule in their medical record (odds ratio, 4.40; 1.41-13.68); and 6) parents not at bedside at time of data collection (odds ratio, 2.31; 1.01-5.31). CONCLUSIONS: We found delirium to be a common problem after cardiac surgery with several important modifiable risk factors.

Assuring Sustainable Gains in Interdisciplinary Performance Improvement: Creating a Shared Mental Model During Operating Room to Cardiac ICU Handoff.

OBJECTIVE: To understand sustainability and assure long-term gains in multidisciplinary performance improvement using an operating room to cardiac ICU handoff process focused on creation of a shared mental model. DESIGN: Performance improvement cohort project with pre- and postintervention assessments spanning a 4-year period. SETTING: Twenty-six bed pediatric cardiac ICU in a tertiary care children's hospital. PATIENTS: Cardiac surgery patients admitted to cardiac ICU from the operating room following cardiac surgery. INTERVENTIONS: An interdisciplinary workgroup overhauled our handoff process in 2010. The new algorithm emphasized role delineation, standardized communication, and creation of a shared mental model. Our "I-5" mnemonic allowed validation and verification of a shared mental model between multidisciplinary teams. Staff orientation and practice guidelines were revised to incorporate the new process, visual aids were distributed and posted at each patient's bedside, and lapses/audit data were discussed in multidisciplinary forum. MEASUREMENTS AND MAIN RESULTS: Audits assessing equipment and information transfer during handoff were performed 8 weeks following implementation (n = 29), repeated at 1 year (n = 37), 3 years (n = 15), and 4 years (n = 50). Staff surveys prior to implementation, at 8 weeks, and 4 years postintervention assessed satisfaction. Comprehensiveness of information transfer improved in the 4 years following implementation, and staff satisfaction was maintained. At 4 years, discussion of all elements of information transfer was 94%, increased from 85% 8 weeks following implementation and discussion of four or more information elements was 100% increased from 93%. Of the 73% of staff who completed the survey at 4 years, 91% agreed that they received all necessary information, and 87% agreed that the handoff resulted in a shared mental model. CONCLUSIONS: Our methods were effective in creating and sustaining high levels of staff communication and adherence to the new process, thus achieving sustainable gains. Performance improvement initiatives require proactive interdisciplinary maintenance to be successful long term.