PSRC Main Site
Annual Meeting Home
Final Program
Past & Future Meetings


Back to 2016 Joint Meeting Posters


Assessing the Predictive Accuracy of the ACS NSQIP Surgical Risk Calculator in Open Ventral Hernia Repair
Marten N. Basta, MD1, Andrew R. Bauder, BS2, Stephen J. Kovach, MD2, John P. Fischer, MD, MPH2.
1Brown University Rhode Island Hospital, Providence, RI, USA, 2University of Pennsylvania Health System, Philadelphia, PA, USA.

PURPOSE:
Preoperative surgical risk assessment is critical for clinical decision-making. The American College of Surgeon’s (ACS) Risk Calculator estimates risk for outcomes based upon individual risk profiles. Recent studies have reported inaccuracies amongst certain populations. This study assesses the predictive accuracy of the ACS Risk Calculator in patients undergoing open ventral hernia repair (VHR).
METHODS:
A review of patients undergoing open VHR between 7/1/2007-7/1/2014 by a single surgeon was performed. Risk factors and outcomes were collected as defined by NSQIP. Thirty-day outcomes included serious complication, venous thromboembolism (VTE), medical morbidity, surgical site infection (SSI), unplanned reoperation, mortality, length of stay. Patient profiles were entered into the ACS Risk Calculator and outcome-specific risk predictions recorded. Prediction accuracy was assessed using Brier scores and ROC under the curve (AUC).
RESULTS:
142 patients were included. ACS predictions were accurate for cardiac complications (Brier=0.02), VTE (Brier=0.08), reoperation (Brier=0.10), and mortality (Brier=0.01). Significantly underestimated outcomes included SSI (Brier=0.14), serious complication (Brier=0.30), and any complication (Brier=0.34). Discrimination ranged from highly accurate (Mortality, AUC=0.99) to indiscriminate (SSI, AUC=0.57). Predicted length of stay was 3-fold shorter than observed (2.4 vs. 7.4 days, p<0.001).
CONCLUSION:
The ACS Risk Calculator accurately predicted medical complications, reoperation, and 30-day mortality. Surgical site infections, serious complications, and length of stay were significantly underestimated. These findings suggest additional considerations are needed to better estimate complications after open VHR.


Back to 2016 Joint Meeting Posters