A propensity-matched analysis of contemporary outcomes of blunt popliteal artery injury

Objective: Traumatic popliteal artery injury is associated with an increased propensity for limb loss, morbidity, and mortality above an already elevated baseline risk to life and limb. Previous studies of outcomes in this patient group have been limited by selection bias. This study analyzed outcomes after blunt popliteal artery injury using propensity matching to reduce confounding variables associated with multiple mechanisms of traumatic vascular injury and to identify factors associated with amputation. Methods: A retrospective review was conducted of prospectively collected data from the National Trauma Data Bank. Patients were identified using International Classification of Diseases, Ninth Revision codes related to patterns of blunt injury associated with popliteal arterial injury or intervention. Using Trauma Quality Improvement Program variables as a reference, specific characteristics were collected. Variables found significant on univariate analysis were used to generate propensity-matched amputation and nonamputation cohorts. Multivariate logistic regression was used to assess for risk factors associated with amputation and inpatient mortality. Results: In total, 3029 patients with blunt popliteal artery injury were identified; 628 (20.7%) underwent amputation. Patients who underwent amputation presented with more frequent hypotension (systolic blood pressure of 0-99mmHg, 22.7% vs 12.8%; P < .001) and tachycardia (heart rate >120 beats/min, 28.5% vs 14.5%; P <.001). Limb loss was also associated with concurrent popliteal vein injury (18.3% vs 8.7%; P <.001) and tibial nerve injury (5.3% vs 1.3%; P <.001) as well as with elevated Injury Severity Score (median, 13 vs 9; P <.001) and lower extremity Abbreviated Injury Scale score (3 vs 2; P < .001). Subsequently, 794 patients were divided into equal number propensity-matched amputation and nonamputation cohorts. Regression analysis revealed that patients with diabetes mellitus (odds ratio [OR], 1.763; P = .049), popliteal vein injury (OR, 1.657; P = .012), or tibial nerve injury (OR, 3.537; P = .007) were more likely to undergo amputation. Further regression analysis with patients matched for Injury Severity Score revealed that age >= 86 years (OR, 38.092; P = .009), patellar fracture (OR, 3.445; P = .036), and elevated Abbreviated Injury Scale score (OR, 1.101; P <.001) were associated with higher risk of inpatient death. Conclusions: Trauma patients who sustain blunt popliteal artery injury are at an increased risk of amputation. Propensitymatched analysis revealed that concurrent popliteal vein and tibial nerve injury but not severity of tissue injury predicted limb loss.