Study finds CT angiography for pelvic trauma identifies arterial bleeding

When critically injured patients are brought to a trauma center, the trauma team needs to identify sources of blood loss quickly and efficiently. Among the existing protocols for patients with pelvic trauma, computed tomography (CT) scanning has been widely used to detect the presence of a hematoma that can indicate arterial bleeding. At the 2007 annual meeting of the Orthopaedic Trauma Association, Jodi Siegel, MD, presented the results of a study indicating that CT angiography could also be helpful.

“In this study, we added CT angiography to the initial trauma scan—and found a 100% negative predictive value for arterial bleeding that required therapeutic angiography,” lead author Paul Tornetta III, MD, explained.

“By quickly ruling out arterial bleeding, we were able to manage these patients more efficiently

than we have in the past—keeping them in the proper location for resuscitation.”

The authors of this study had two primary objectives. “We wanted to ascertain the predictive value of CT angiography in determining the need for therapeutic angiography in patients who sustained pelvic trauma,” said Dr. Siegel. “We also wanted to determine if definable arterial bleeding correlated with anatomic injury.”

New protocols established
The study was conducted at a level-one trauma center over 2 years with 58 nonconsecutive patients who had sustained pelvic trauma.

The protocol for patients who had sustained a pelvic or acetabular fracture that had been identified on the initial anteroposterior pelvic radiograph included a CT angiography as part of the patient’s initial trauma CT scan. The CT angiography was performed on a 25-second delay scan, after contrast introduction.

According to the study, patients who exhibited “clinical evidence of hemodynamic instability based on continued blood product requirements and/or hypotension (systolic blood pressure less than 100 mmHg),” were sent for therapeutic angiography despite standard resuscitation and regardless of the results of their CT angiography.

Pelvic and acetabular fractures were evaluated using either the Young and Burgess or the Letournel classification system. Lateral compression (LC) type 1 and anteroposterior compression (APC) type 1 pelvic fractures were considered stable. LC types 2 and 3, APC types 2 and 3, and vertical shear injuries were considered unstable.

The results of the therapeutic angiogram were compared to the results from the earlier CT angiographic study. “We also compared the location of the bleeding seen on the CT angiography to the findings of the interventional angiography and the anatomic location of the pelvic injury,” explained Dr. Siegel.

100 percent negative predictive value
The 36 male and 22 female patients (average age, 43 years) had an average injury severity score (ISS) of 18.6. Eighteen patients had positive results on the CT angiography and were referred for additional treatment. Two patients from this group were referred for therapeutic angiography but died before it could be performed.

Of the remaining 16 patients, 11 had interventional angiography with 8 positive findings. Seven patients had major vessel injuries treated with coils. The remaining patient had a cut-off vessel that was observed.

Only one of the 40 patients with negative CT angiography went on to interventional angiography. Although this patient continued to require blood, there was no evidence of arterial bleeding.

“The negative predictive value of CT angiography for pelvic arterial bleeding that required therapeutic angiography was 100 percent,” said Dr. Tornetta. “The positive predictive value of CT angiography for angiographically treatable bleeding was 70 percent—if one includes the two early deaths as these patients presumably had arterial injuries.”

The accuracy of the CT angiography was quite impressive. The location of the vessel identified on CT angiography was consistent with the therapeutic angiography in all patients. The location of the arterial bleeding was consistent with a fracture or dislocation in 9 of 11 patients.

In one patient, pelvic arterial bleeding that did not correlate with his pelvic injury was documented—a superior gluteal artery with a sacral fracture. In this particular patient, the fracture location was not useful in predicting the location of the arterial bleeding. Fracture-dislocations of the sacroiliac joint had the highest rate of positive CT angiographic findings.

The patient group with a positive CT angiogram had a higher percentage of unstable pelvic fractures than the negative CT angiography group (67 percent vs 40 percent).

Accuracy was encouraging
“I was encouraged by the accuracy in identifying the location of pelvic arterial bleeding using CT angiography,” Dr. Tornetta said. “In patients with an arterial bleed who went on to get formal interventional angiography as a treatment modality, the CT angiography was effective in helping identify the location of the problem. The CT angiographic findings made the interventional angiography and treatment more efficient.

“In summary, the inclusion of a CT angiogram holds the promise of reducing the time spent on resuscitation efforts by ruling out pelvic arterial bleeding in some, and making the therapeutic angiogram more efficient in the others.”

Disclosure information for Drs. Siegel and Tornetta is available at www.aaos.org/disclosure. Annie Hayashi is the senior science writer for AAOS Now. She can be reached at hayashi@aaos.org