This study, featuring contributions from past and present Penn faculty, retrospectively compared outcomes of trauma patients with grade IV or V duodenal and/or pancreatic injuries who underwent pancreaticoduodenectomy (PD) versus non-PD surgical management. They found that patients who underwent the Whipple procedure experienced significantly higher complication rates and longer lengths of stay without a clear survival benefit.
Summary:
Severe pancreaticoduodenal trauma is rare, accounting for less than 5% of penetrating traumas, yet it carries high morbidity (up to 87%) and mortality (13%–50%) due to frequent concomitant injuries. Previous studies suggest that primary repair with drainage may reduce duodenal leak rates and shorten hospital stays compared to more complex surgical reconstructions. The authors hypothesized that PD does not improve outcomes compared to non-PD surgical approaches for grade IV–V pancreaticoduodenal injuries.
This study included trauma patients aged 15 years and older who were treated for grade IV–V duodenal injuries and/or grade IV–V pancreatic injuries requiring surgical intervention at 35 Level 1 trauma centers from 2010 to 2020. Patients were identified via trauma registries and retrospective chart review. The cohort was divided into two groups: those who underwent PD (either single- or multi-stage) and those who received non-PD surgical management, which included primary repair alone (PRA), wide extraluminal drainage without repair, or complex repair with adjunctive measures other than PD (CRAM). Logistic regression analysis was used to assess the study’s primary outcomes, which included GI complications, length of stay, readmission, and mortality.
A total of 95 patients were identified, with 32 undergoing PD and 63 receiving non-PD surgical interventions. The majority of patients were male (82%), with a mean age of 25.5 years and a mean BMI of 24.4 kg/m². Most patients presented with normal systolic blood pressure (SBP) (median 120 mmHg, IQR 102–140), a normal Glasgow Coma Scale (GCS) score (median 15, IQR 13–15), and a high Injury Severity Score (median 26, IQR 17–34). Patients in the PD group had significantly higher rates of grade V duodenal injuries (50% vs. 19%, p=0.004) and grade V pancreatic injuries (66% vs. 22%, p<0.001) compared to those in the non-PD group.
Although duodenal leak rates (23%) and anastomotic leak rates (10%) were similar between both groups (p>0.05), PD patients experienced significantly higher rates of GI-related complications (69% vs. 44%, p=0.031), including intra-abdominal abscess, GI bleed, ileus, and enterocutaneous fistula. PD was also associated with significantly longer ICU stays (16.5 vs. 6 days, p=0.012) and hospital stays (33.5 vs. 24.5 days, p=0.017) compared to non-PD. However, there were no significant differences in mortality (12.5% PD vs. 23.8% non-PD, p=0.279) or 30-day readmission rates (31.3% PD vs. 27% non-PD, p=0.810).
Subgroup analysis of patients without ampullary injuries (n=60) showed that PD was more frequently performed in those with grade V pancreatic injuries, pancreatic head injuries, and concomitant pancreatic and duodenal injuries. However, these patients had significantly higher anastomotic leak rates compared to those managed with non-PD interventions (30% vs. 4%, p=0.028).
While this study is limited by its retrospective nature and lack of granularity in the setting of multi-institutional chart review, it is a large cohort study for a rare event. The retrospective nature of the study also limits the ability to control for all confounding variables that may have influenced surgical decision-making. Variability in practice patterns across different trauma centers, along with differences in surgeon experience and subspecialty training, could have contributed to disparities in patient outcomes. Lastly, despite using standardized AAST grading criteria, intraoperative interpretations of injury severity—particularly regarding "massive disruption of the pancreatic head"—may have varied among surgeons, introducing further variability into the study findings.
Bottom Line:
Grade IV and V duodenal and pancreatic traumatic injuries can be managed with either PD or alternative surgical interventions, though trauma PD is associated with high morbidity, prolonged hospital stays, and increased GI-related complications without a clear survival benefit. These findings suggest that non-PD surgical approaches should be strongly considered in patients without ampullary involvement or with less severe injuries to minimize postoperative complications and improve recovery outcomes.