Adverse events in cardiac surgery, a mixed methods retrospective study in an Italian teaching hospital

Rationale, aims, and objectives Adverse events (AEs) are a major concern in surgery, but the evidence in cardiac surgery is limited, especially on the contributory factors. According to the data of the National Outcomes Program, a unit was selected to conduct a mixed methods investigation into the incidence, type, and cause of AE, given its mortality rate that was double the national average on coronary artery bypass grafting, valve reparation, and replacement. Methods A retrospective investigation on the performance of a cardiac surgery, combining the routinely collected data on process and outcome measures with a 2-stage structured review of 280 medical records performed by 3 expert clinicians, with the support of a methodologist. Results At least one risk had been verified in 137 of 280 cases (48.9%, 95% CI, 43.1-54.8). The total number of AE was 42, with an incidence of 15% (95% CI, 10.8-20.2) and a preventability of 80.9% (95% CI, 69.1-92.8). In 11.9% of AE, the consequence is death, disability in 40.5%, and extended hospital stay in 69% of the cases. Adverse events are associated with problems in care management at the ward (89/137, 64.9%, 95% CI, 56.9-72.9), followed by surgical complications (46/137, 33.6%, 95% CI, 25.7-41.5) and infection/sepsis (32/137, 23.4%, 95% CI, 16.3-30.4). An active error was made by the health care workers in 31 of 42 cases with AE, either during the decision making or during the execution of an action. A total of 36 AEs were due to deficiencies attributed to organizational factors and 31 were linked to poor teamwork. Conclusions The mixed methods approach demonstrated how a deep understanding of AE and poor performance may emerge thanks to the combination of routinely available data and experts' evaluations. The main limitation of this study is its focus on the cardiac surgery rather than on the entire process of care. The evaluation could have been integrated with on-site observations and the analysis of reported incidents.