Should Emergent and Emergent Salvage Operations be Included in Hospital STS Outcomes?
William Z Chancellor1, James H Mehaffey1, Jared P Beller1, Elizabeth D Krebs1, Kenan Yount1, Robert B Hawkins1, Clifford Fonner2, Alan M Speir3, Mohammed A Quader4, Jeffrey B Rich5, Leora T Yarboro1, Nicolas Teman1, Gorav Ailawadi1
1University of Virginia, Charlottesville, VA;2Virginia Cardiac Services Quality Initiative, Virginia Beach, VA;3INOVA Heart and Vascular Institute, Falls Church, VA;4Virginia Commonwealth University, Richmond, VA;5Heart and Vascular Institute, Cleveland Clinic, Cleveland, OH
Outcomes in cardiac surgery are benchmarked against national Society of Thoracic Surgeons (STS) data and include patients undergoing elective, urgent, emergent, and salvage operations. In contrast, the National Cardiovascular Data Registry (NCDR) excludes very high-risk percutaneous coronary intervention (PCI) data to allow interventional cardiologist to perform extreme cases without affecting their reported outcomes. We hypothesize the STS risk calculator does not adequately characterize the risk of emergent and salvage operations and these cases should be excluded from quality reporting.
All patients who underwent emergent or salvage surgery with an STS predicted risk score in a regional consortium of 19 cardiac surgery centers (2001-2017) were evaluated. Outcomes were compared to STS calculated preoperative risk scores, which were available from a regional STS database. Observed to expected (O:E) ratios were calculated for mortality, permanent stroke, renal failure, prolonged ventilation, reoperation, deep sternal wound infection, and composite major morbidity or mortality at each center.
Among 85,819 cardiac surgery patients with calculated STS predicted risk scores, 2,647 patients had emergent surgery and 131 patients underwent salvage operations. Predicted Risk of Mortality (PROM) for all patients was 2.3% but increased to 9.9% for emergent cases and 32.3% for salvage operations. The STS PROM adequately predicted mortality for emergent cases (O:E= 0.96, p=0.56), but did not predict mortality for salvage operations (O:E= 1.35, p=0.005). In contrast, STS risk scores failed to adequately predict composite morbidity and mortality for both emergent and salvage operations (emergent: O:E= 1.15, p=0.0001; salvage: O:E= 1.23, p=0.0005). Importantly, there was substantial variability in the distribution of emergent and salvage cases (N=1.17-33.35 emergent cases per year and N=0-2.65 salvage cases per year) and O:E ratios among centers (Figure).
The current STS risk models do not adequately predict outcomes for emergent and salvage cardiac operations due to low volumes and substantial variation across centers. Therefore, we recommend for the exclusion of emergent and salvage operations from quality reporting to avoid risk aversion in these potentially life-saving operations.
Figure: Observed to expected mortality ratios for emergent (Range: 0.5-1.7) and salvage (Range: 0-4.9) cases at each center.
Back to 2018 Program