Breaking protocol: Should destructive colon injuries be managed differently in a balanced blood product resuscitation era?

INTRODUCTION

Traumatic colon injuries are a significant source of morbidity. For several decades, an institutional colon injury management protocol has been followed, advising diverting ostomy for transfusion >6 U of packed red blood cells (pRBCs) or comorbidities. This protocol was based on resuscitation practices prior to balanced transfusion strategies and results in high stoma rates. With advancements in transfusion strategies and interventional radiology, this protocol warrants reevaluation.

METHODS

We reviewed 5 years of traumatic colon injuries at an urban Level 1 trauma center, analyzing destructive injuries treated with resection and anastomosis without diversion. Injury characteristics, management, and outcomes were collected. Factors associated with anastomotic leak, abscess, reoperation, and mortality were assessed. Variables included 24-hour pRBC transfusion, base deficit, body mass index (BMI), comorbidities, pancreatic injury, shock index, and protocol adherence.

RESULTS

Among 559 patients with operative colon injuries, 213 underwent resection and anastomosis without diversion. Anastomotic leak occurred in 10 patients (4.7%). Thirteen patients violated protocol, including nine exceeding the transfusion threshold without diversion. Protocol adherence based on pRBC transfusion resulted in lower abscess rates (24% vs. 67%, p = 0.010) and drain placement (17% vs. 56%, p = 0.012), with no difference in reoperation (8.3% vs. 11.1%, p = 0.555) or mortality (2.9% vs. 0%, p = 0.999). Other variables associated with abscess formation included BMI >35 kg/m 2 ( p = 0.039) and pancreatic injury ( p = 0.001). Multivariable logistic regression identified pRBCs (adjusted odds ratio [AOR], 1.30; p = 0.003) and concomitant pancreatic injury (AOR, 6.03; p = 0.006) as abscess predictors, and pRBCs were significant for abscess even at a >2 pRBC cutoff ( p = 0.049). Shock index (AOR, 21.2; p = 0.033) and BMI (AOR, 1.12; p = 0.044) were associated with anastomotic leak, as well as reoperation with ostomy (shock index AOR, 8.13 [ p = 0.052]; BMI AOR, 1.10 [ p = 0.041]).

CONCLUSION

In colon injury patients undergoing resection and anastomosis without diversion, transfusion volume correlates with abscess but not anastomotic leak, reoperation, ostomy, or mortality. Shock index and BMI may better predict anastomotic leak and reoperation. The colon injury protocol should be reevaluated.

LEVEL OF EVIDENCE: Therapeutic/Care Management; Level IV.