Background: In-hospital mortality and morbidity are - if well defined - readily measurable and objective parameters for monitoring standard of care within a single institution and for comparisons between centres. The Estimation of Physiologic Ability and Surgical Stress (E-PASS) score was initially developed to predict adverse postoperative effects for patients requiring elective gastrointestinal surgery ranging from laparoscopic cholecystectomy through to transthoracic esophagectomy. Our aim was to review whether the E-PASS scoring system could be used without restrictions in hepatic surgery as a means of correctly predicting morbidity and mortality.
Methods: E-PASS predictor equations were prospectively collected and analyzed retrospectively for 243 patients requiring hepatic resections between 2002-2006. The Comprehensive Risk Score (CRS) was calculated using the E-PASS equations as previously stated, which includes calculation of the Pre-Operative Risk Score (PRS) and the Surgical Stress Score (SSS). Patients were divided into 5 severity groups, also as previously stated, for whom expected adverse outcomes increase with increasing CRS. Observed morbidity and mortality rates were compared with rates predicted by E-PASS using either the Fisher’s Exact Test, or for larger sample sizes the chi2 Test. The Wilcoxon rank-sum Test and the t-Test were applied for comparison of PRS and SSS between patients with and without morbidity or mortality.
Results: The observed and predicted overall mortality rates were 3.3 and 3.7 per cent respectively, morbidity rates were 31 and 28 per cent. The E-PASS model showed no significant difference between expected and observed in-hospital mortality (p= 0.641), indicating that it predicted outcome effectively. E-PASS under-predicted morbidity and showed significant lack of fit (chi2= 11.1, 3d.f. p= 0.011). Although comparison of PRS and SSS between patients with and without complications revealed no overall significant difference (t= -0.37, 241d.f. p= 0.714 and t= -1.69, 241d.f. p= 0.093), group specific comparisons showed lack of fit for groups 1, 2 and 4. Equally, patients who died postoperatively did not have a significantly higher PRS or SSS (p= 0.157 and p=0.305).
Conclusions: These data suggest that E-PASS does up to a certain extent accurately predict outcome in patients undergoing hepatic resections. This was especially true for predicting mortality. Morbidity was however under-predicted in the E-PASS model. A modified, new logistic equation might be required for liver-specific resections in order to correctly foresee postoperative complications and mortality after hepatic surgery.