Background
Existing scoring systems to predict mortality in acute pancreatitis may not be directly applicable to the emergency department (ED). The objective of this study was to derive and validate the ED-SAS, a simple scoring score using variables readily available in the ED to predict mortality in patients with acute pancreatitis.
Methods
This retrospective observational study was performed based on patient data collected from electronic health records across 2 independent health systems; 1 was used for the derivation cohort and the other for the validation cohort. Adult patients who were eligible presented to the ED, required hospital admission, and had a confirmed diagnosis of acute pancreatitis. Patients with chronic or recurrent episodes of pancreatitis were excluded. The primary outcome was 30-day mortality. Analyses tested and derived candidate variables to establish a prediction score, which was subsequently applied to the validation cohort to assess odds ratios for the primary and secondary outcomes.
Results
The derivation cohort included 599 patients, and the validation cohort 2011 patients. Thirty-day mortality was 4.2 and 3.9%, respectively. From the derivation cohort, 3 variables were established for use in the predictive scoring score: ≥2 systemic inflammatory response syndrome (SIRS) criteria, age > 60 years, and SpO2 < 96%. Summing the presence or absence of each variable yielded an ED-SAS score ranging from 0 to 3. In the validation cohort, the odds of 30-day mortality increased with each subsequent ED-SAS point: 4.4 (95% CI 1.8–10.8) for 1 point, 12.0 (95% CI 4.9–29.4) for 2 points, and 41.7 (95% CI 15.8–110.1) for 3 points (c-statistic = 0.77).
Conclusion
An ED-SAS score that incorporates SpO2, age, and SIRS measurements, all of which are available in the ED, provides a rapid method for predicting 30-day mortality in acute pancreatitis.