Active extravasation of the abdomen and pelvis in trauma using 64MDCT

The objective of this study was to determine the clinical and management implications of the finding of active extravasation in blunt or penetrating trauma patients evaluated with abdomino-pelvic computed tomography (CT) using 64MDCT technology. This HIPAA compliant, retrospective study was IRB-approved, and the need for consent was waived. All adult patients scanned with 64MDCT who sustained blunt or penetrating abdomino-pelvic trauma and had findings of active extravasation at our Level I trauma center during a 30-month period were included. Two radiologists reviewed all abdomino-pelvic CT scans and characterized the active hemorrhage by location, extent, and attenuation on all available phases of imaging. Subsequent therapy and disposition were determined by reviewing the patients' medical records. The relationship between the location of a source of extravasation and subsequent clinical outcome was evaluated using Fischer's exact test. The relationship between the size and attenuation of the active hemorrhage and patient outcome were compared using the Wilcoxon rank sum test. One hundred and twenty-five patients with active extravasation were included. Patients with solid organ or pelvic injuries that were managed conservatively or had a negative digital subtraction angiogram had statistically significant smaller areas of active extravasation when compared to those that required intervention or died. When the attenuation values of extravasation are normalized to the intravascular attenuation achieved after intravenous contrast injection, no significant differences were seen based on subsequent clinical outcome. Based on location, those patients with solid organ, gastrointestinal/mesenteric, and pelvic sources of bleeding showed statistically significant higher likelihood of requiring subsequent intervention or dying, compared with those patients with subcutaneous, intramuscular, or retroperitoneal sources of active extravasation who were more likely to be managed conservatively (p < 0.0001, p = 0.005, p = 0.006, respectively). In blunt and penetrating trauma patients evaluated using 64MDCT technology, the location and size of the region of active extravasation are predictive of the type of subsequent clinical management. Normalized attenuation values of the active extravasation, however, are not predictive of subsequent management.