Failure rate and complications of angiography and embolization for abdominal and pelvic trauma

Angiography and embolization have become the treatment of choice after abdominal trauma or pelvic injury in hemodynamically stable patients with a suspicion of internal hemorrhage (contrast extravasation, pseudo-aneurysm, or a vessel cutoff diagnosed on computed tomographic scanning). Some studies, however, report a high incidence of rebleeding (failure) or complications. The aim of this study was to evaluate the failure rate and the complications in trauma patients undergoing such procedures.

All consecutive patients (n = 97) admitted to our Level I trauma center between January 2002 and December 2008 in whom angiography with or without embolization was performed were analyzed. Complications were classified as organ specific, puncture site related, and systemic. Additional interventions, required to treat complications, were documented.

The overall failure rate was 12%. Overall, 48 complications were documented in 28 patients. Organ-specific complications were observed in 18 patients (19%), especially abscess formation and infarction of the liver. Puncture site-related complications occurred in three patients. The incidence of contrast-induced nephropathy was 24%. Three patients developed renal failure. Nine of the 15 patients with rebleeding could be managed with reembolization or operative packing, resulting in an organ salvage rate of 93%. Most (83%) of the organ-specific complications and all of the puncture site-related complications could be managed conservatively or with percutaneous treatment.

In the present study, the failure rate and incidence of organ-specific and procedure-related complications were low and often could be managed with nonoperative minimally invasive interventions. Trauma patients undergoing angiography have a high chance (24%) of developing contrast-induced nephropathy and should therefore receive optimal prophylactic measures to avoid this complication.

Therapeutic study, level IV; prognostic/epidemiologic study, level III.