Traumatic Vascular Disruption in the Thoracocervical Region: Surgical Outcome and Predictors of Survival

Abstract

Traumatic vascular injury in the thoracocervical region is associated with significant morbidity and mortality. The goal of this study was to examine surgical outcome of these injuries and to establish predictors of survival. The medical records of all patients sustaining traumatic injury to the carotid, vertebral, subclavian, and proximal axillary vessels and undergoing operative repair were retrospectively reviewed. Sixteen patients (11 men, 5 women, mean age 31.1 ±16.6) during a four-year period underwent emergent surgical repair for traumatic vascular disruption. There were 6 lesions of the carotid, 7 of the subclavian, and 2 of the proximal axillary vessels and 1 vertebral artery injury. Sixty- two percent of the patients were hemodynamically unstable on admission. In these patients the diagnosis was established by clinical examination. Hemodynamically stable patients with a high index of suspicion for vascular injury underwent urgent arteriog raphy to confirm the diagnosis and assist in the operative approach. Surgical procedures included lateral arteriorrhaphy (2), end-to-end anastomoses (2), synthetic interposition grafts (3), and autogenous vein graft (2). Seven patients underwent emergent ligation of the subclavian, carotid, or vertebral artery. Despite 3 of 16 patients presenting in either unresponsive shock or with severe head trauma, overall mortality was 25%. Nonsurvivors had significantly higher injury severity scores than survivors (36.3 ±13.1 vs 19.8 ±7.9, P < 0.05). Length of time from trauma scene to hospital was a significant predictor of survival (33.4 ± 13.3 vs 50.8 ±6.5 minutes) as was the presence of associated organ system injury (2.8 ±0.9 organ systems vs 4.3 ±0.9, P < 0.05) in survivors and nonsur vivors, respectively. No neurologic compromise or limb ischemia was noted in the group undergoing arterial ligation. Among similarly injured vessels, the type of surgical repair did not influence patient survival. The outcome of surgical repair of a traumatic lesion of the thoracocervical vessels depends upon the rapid recognition of the injury, time delay to the trauma unit, and the magnitude of associated injury.

The impact of time to hemostatic intervention and delayed care for patients with traumatic hemorrhage: A systematic review

“…Title and abstract screening excluded 1,889 records, and 66 full texts were evaluated. Twenty-four studies including 10,168 patients met the full eligibility criteria 9,16–38 . Reasons for exclusion included different patient population (n = 19), wrong exposure (n = 18), wrong study design (n = 4), and abstract without necessary data (n = 1).…”

“…Table 1 summarizes the characteristics of the included studies 9,16–38 . The two largest studies contributed 56.2% (5,716 of 10,168) of the included patients, 18,25 and 15 of 24 studies contained more than 100 patients.…”

“…Twenty-three of 24 studies had mortality or survival as outcomes of interest 9,16,18–38 . Most studies (15 of 23 [65.2%]) reported in-hospital mortality, while four reported 30-day mortality, three examined 24-hour mortality, two assessed 28-day mortality, and one reported time to mortality (Table 1).…”

“…Studies evaluating patients requiring only operative intervention for abdominal trauma were most common (n = 9), followed by those evaluating patients requiring thoracic and/or neck surgery (n = 4), angioembolization for pelvic fractures (n = 4), surgery or angioembolization for any bleeding injuries (n = 4), surgery for abdominal or chest injuries (n = 2), and angioembolization for blunt abdominal solid organ injury (n = 1). Twelve studies included patients with severe head injuries, 9,[17][18][19]23,24,[26][27][28]30,31,38 while seven specifically excluded either all head injuries or those with head Abbreviated Injury Scale scores of greater than 2. The remainder of studies did not comment on inclusion of patients with traumatic brain injuries.…”

“…2). Nine studies (40.9%) were at very high risk of bias because of lack of control of confounding, 16,19,22,28,32,[34][35][36][37] and, for Rosengart et al, 32 very high risk of bias due to missing data. An additional nine studies (40.9%) were at high risk of bias, as important confounders were not controlled for despite control of other confounders.…”