Selection of complications to define failure to rescue as an optimal quality improvement metric

Failure to rescue (FTR), defined as death after a surgical complication, is strongly impacted by systems-level care processes. The purpose of this study was to optimize the definition of FTR by developing the methodology for, and evaluating the subsequent impact of, adding complications to the Society of Thoracic Surgeons (STS) definition.

Patients undergoing coronary artery bypass grafting and/or valve operations from 2011 to 2024 in Michigan were included. Complications were considered for the FTR definition on the basis of the complication's association with mortality, event rate, FTR rate, interhospital variability, and percent of operative deaths accounted for by the FTR definition. Risk-adjusted FTR rates were calculated for 34 hospitals.

Of 92,860 cases, 37,162 (40%) patients developed any of 17 complications and 2066 (2.2%) died. In addition to the STS FTR complications (stroke, renal failure, reoperation, prolonged ventilation), 5 additional complications demonstrated high FTR and interhospital variation (cardiac arrest, sepsis, pneumonia, gastrointestinal events, and anticoagulation bleeding events, ie, "STS+5"). The current STS FTR definition accounted for 70% of mortalities, whereas STS+5 accounted for 82%. After risk adjustment, the STS+5 compared with the STS FTR definition changed hospital FTR rates between -19.2% and 19.1%, yet interhospital variability was similar (range, 3.5-50.7% vs 3.7-47.1%).

Adding 5 complications to the STS FTR definition captures more mortalities while retaining similar interhospital variation. A more comprehensive FTR definition will better account for variation in complication-specific FTR by hospital. Leveraging FTR for quality improvement within cardiac surgery will require further work to identify the optimal FTR definition.