BACKGROUND: There are many benefits of repair over ligation of major venous injuries (MVIs) following penetrating trauma, but the risk of pulmonary embolism (PE) is not well defined. We hypothesized that rates of PE are comparable between repair and ligation of MVI. METHODS: All penetrating trauma patients with MVI requiring an operation from 2003 to 2012 (n = 158) were retrospectively reviewed. Propensity scores were based on a logistic regression model using patient and injury characteristics. A 1:1 fixed ratio nearest neighbor matching was performed to compare outcomes of the repair and ligation cohorts. Data are reported as mean ± SD if parametric, or median (interquartile range) if not, and compared using a t test, Mann-Whitney U-test, χ2, or Fisher's exact test, as appropriate. RESULTS: The population was 89% male, age 32 ± 12 years, 74% gunshot wound, Injury Severity Score of 19 ± 13, length of stay of 9 (18) days, 3.8% PE, and a mortality of 21.5%. Repair was performed in 37% (n = 59), ligation was performed in 60% (n = 94), and 3% required both. With ligation versus repair, ligation patients were generally more critically injured; 48-hour survival was 78% versus 93% (p = 0.0083), initial Glasgow Coma Scale (GCS) score was 12 ± 5 versus 14 ± 3 (p = 0.003), initial base excess was -9 ± 8 versus -5 ± 5 mEq/L (p = 0.003), more packed red blood cells were transfused (12 (14) U vs. 9 (12) U; p = 0.032), and major arterial injury was more likely (86% vs. 42%, p < 0.001), but the PE rate was identical (5.9%) in propensity-matched cohorts. In those who developed a PE, all were receiving standard thromboprophylaxis. CONCLUSION: Following penetrating trauma, the risk of PE between repair and ligation of MVI is comparable. LEVEL OF EVIDENCE: Epidemiologic study, level III.
BACKGROUND: There are many benefits of repair over ligation of major venous injuries (MVIs) following penetrating trauma, but the risk of pulmonary embolism (PE) is not well defined. We hypothesized that rates of PE are comparable between repair and ligation of MVI. METHODS: All penetrating traumapatients with MVI requiring an operation from 2003 to 2012 (n = 158) were retrospectively reviewed. Propensity scores were based on a logistic regression model using patient and injury characteristics. A 1:1 fixed ratio nearest neighbor matching was performed to compare outcomes of the repair and ligation cohorts. Data are reported as mean ± SD if parametric, or median (interquartile range) if not, and compared using a t test, Mann-Whitney U-test, χ2, or Fisher's exact test, as appropriate. RESULTS: The population was 89% male, age 32 ± 12 years, 74% gunshot wound, Injury Severity Score of 19 ± 13, length of stay of 9 (18) days, 3.8% PE, and a mortality of 21.5%. Repair was performed in 37% (n = 59), ligation was performed in 60% (n = 94), and 3% required both. With ligation versus repair, ligation patients were generally more critically injured; 48-hour survival was 78% versus 93% (p = 0.0083), initial Glasgow Coma Scale (GCS) score was 12 ± 5 versus 14 ± 3 (p = 0.003), initial base excess was -9 ± 8 versus -5 ± 5 mEq/L (p = 0.003), more packed red blood cells were transfused (12 (14) U vs. 9 (12) U; p = 0.032), and major arterial injury was more likely (86% vs. 42%, p < 0.001), but the PE rate was identical (5.9%) in propensity-matched cohorts. In those who developed a PE, all were receiving standard thromboprophylaxis. CONCLUSION: Following penetrating trauma, the risk of PE between repair and ligation of MVI is comparable. LEVEL OF EVIDENCE: Epidemiologic study, level III.