It has been fairly well established that Grade I injuries probably do not need surgical repair and will likely resolve with conservative management where progression to rupture is highly unlikely.^7,8 Conversely, patients with Grade IV or free rupture frequently do not make it to the hospital alive, and if they do, they are usually in extremis and require emergent treatment. Patients with Grade II to III fall into a “grey zone”. Some groups have shown that patients with Grade II injuries can be safely watched to allow appropriate stent graft sizing and stabilization of the patient before repair and, in fact, may not need repair at all.^6,9 There is a paucity of data available on patients with Grade III injuries. The Society for Vascular Surgery consensus guidelines on TEVAR for blunt aortic injury recommends “urgent” (<24 hours from injury) repair of all thoracic aortic injuries immediately after other injuries have been repaired.⁶ We hypothesized that delayed repair of thoracic aortic injuries in patients with Grade III injuries could be performed safely with no increase in aortic-related morbidity and mortality, thus optimizing patient stabilization and allowing other more urgent operative management to take precedence.

PATIENTS AND METHODS

A retrospective review of all patients undergoing TEVAR for blunt aortic injury from March 2004 to March 2012 was performed. All procedures were performed at a single institution by one of five full-time academic surgeons. Our center is the only Level I trauma center in the state, and during this period, there

all at the discretion of the operative surgeon. Heparin was used during placement of large sheaths; however, it was reversed as soon as the sheaths were removed. Intravascular ultrasound was used during deployment for sizing at the discretion of the surgeon and was not performed routinely. We did not revascularize any left subclavian arteries prophylactically.

Grading on Imaging

All emergent patients with suspected chest trauma underwent CTA initially. The initial diagnosis was based on CTA, and no patient underwent conventional arteriogram for diagnostic purpose. Based on preoperative CTA imaging, patients were classified into one of four categories, dependent on severity of injury. Grade I injuries consisted of aortic intimal tear; Grade II had intramural hematoma; Grade III had aortic pseudoaneurysm, and Grade IV had free aortic rupture. Size of the pseudoaneurysm and presence of mediastinal hematoma was noted. Grade I injuries were routinely managed medically with repeat imaging to assess progression of injury and in the absence of progression were not repaired. Grade II and higher injuries were offered TEVAR if anatomically and clinically suitable. For the purposes of this study, we limited evaluation to Grade III injuries only.

Statistical Analysis

Demographics and outcomes were compared between the immediate repair group and the delayed repair group. Categorical data were analyzed using a contingency table with Fisher's exact test and two-tailed p values. Because of small sample size, nonparametric tests were used to analyze variables with skewed distribution. Mean with range was used instead of SD. A p < 0.05 was considered statistically significant. Statistical analyses were performed using GraphPad (La Jolla, CA).

RESULTS

During this period, 43 patients were admitted with blunt aortic injuries. Of these, nine had Grade I injuries and were

Table IV.Survival and Length of Stay in the Immediate and Delayed Repair Groups

*Excludes two patients who were discharged and returned for aortic surgery

patients had scheduled CTA at the time of discharge and then at 30-day intervals. One of these patients underwent repair at Day 91 because of a slight increase in the size of the pseudoaneurysm. The other patient underwent repair electively with no change in the size. We had recommended TEVAR to both patients earlier, but they had initially declined surgery.

DISCUSSION

Patients with blunt aortic injury rarely have isolated vascular trauma. In our series, patients with Grade III injuries had on average 9.5 trauma diagnoses per patient, with the patients in the delayed repair group having a higher number of trauma diagnoses compared with the immediate repair group (11.3 vs. 7.1), despite similar ISS and GCS score on arrival. These patients are commonly in shock and have multiple injuries that need to be stabilized and addressed simultaneously. There is frequently a question of which injury should be addressed first, and often significant time in the operating room is required to handle all the surgical problems. Several earlier studies looking at blunt aortic injuries found that delayed repair of aortic injuries may be associated with improved survival regardless of associated injuries.^11–13 However, these studies did not have the patients divided by aortic injury grade, and thus, generalization to higher-grade injuries could not be made. These suggest that patients with Grade III blunt aortic injury can be safely watched and stabilized before definitive repair. An “emergent” repair of the aorta is certainly necessary with a Grade IV injury of the aorta or a change in the initial grade of the aortic injury, but we found no difference in Grade III patients who had their aortic injuries repaired immediately (within 24 hours) or in a delayed fashion. In addition, there are perceived benefits to waiting for aortic repair. It allows not only repair of other injuries in a timely fashion but also stabilization of the patient and adequate resuscitation, which may aid in graft sizing.8 In our small series, there was no significant difference in aortic neck diameter between the CTA obtained on arrival and any repeat CTA performed before definitive repair. In addition, a recent article by Rabin et al.13 suggests that early aortic repair may worsen concurrent traumatic brain injury, and thus, a delayed strategy for blunt aortic injury may be warranted.

Of 16 patients, 3 died of their other injuries before definitive aortic repair in the delayed repair group. In addition, one patient in the delayed group had undergone a successful TEVAR and died of other injuries. We had no aortic-related deaths in either group, and only one patient in the delayed group had a change in aortic pathology before definitive aortic repair. In this patient, respiratory issues developed, and