Epidemiological state, predictive model for mortality, and optimal management strategy for pancreatic injury: A...

Data for establishing the optimal management modalities for pancreatic injury are lacking. Herein, we aimed to describe the epidemiology, identify mortality predictors, and determine the optimal management strategy for pancreatic injury.

We identified patients with pancreatic injury between 2004 and 2017 recorded in the Japan Trauma Data Bank. The primary outcome was mortality. Multivariable logistic regression analyses were used to identify factors significantly associated with mortality and to develop a predictive model. Patients were also classified according to the Organ Injury Scaling of the American Association for the Surgery of Trauma (AAST grade I/II or III/IV). Outcomes were compared based on significant confounder-adjusted treatment strategy.

Overall, 743 (0.25%) patients had pancreatic injury. Traffic accident was the most common aetiology. The overall mortality rate was 17.5%, while it was 4.7% for isolated pancreatic injury. AAST grade, Revised Trauma Scale score on arrival, age, and coexistence of severe abdominal injury aside from pancreatic injury were independently associated with mortality. A predictive model for mortality comprising these four variables showed excellent performance, with an area under the receiver operating characteristic curve of 0.89 (95% confidence interval [CI], 0.85-0.93). The in-hospital mortality was higher in patients who underwent celiotomy than in those who did not among those with AAST grade I/II (15.1% vs. 5.3%) and III/IV (13.8% vs. 12.3%). After adjusting for confounders, these differences were not significant with the adjusted odds ratios of 1.41 (95% CI, 0.55-3.60) and 0.54 (95% CI, 0.17-1.67) for AAST grade I/II and III/IV, respectively.

AAST grade, Revised Trauma Scale score on arrival, age, and coexistence of severe abdominal injury aside from pancreatic injury were prognostic factors of mortality after pancreatic injury. Confounder-adjusted analysis did not show that operative management was superior to non-operative management for survival. Non-operative management may be a reasonable strategy for select pancreatic injury patients, especially in institutions where expertise in interventional endoscopy is available.