Intramedullary nailing of diaphyseal femur fractures secondary to gunshot wounds: predictors of postoperative malrotation.

OBJECTIVES

The purpose of this study was to determine significant factors that may impact the postoperative differences in femoral version (DFV) and differences in femoral length (DFL) between the fixed and uninjured sides after intramedullary nailing (IMN) secondary to gunshot wounds.

DESIGN

Retrospective data registry study.

SETTING

Academic level I trauma center.

PATIENTS: Over a 10-year period, 417 patients underwent IMN of a diaphyseal femur fracture (OTA/AO 32A-C). Of these, 57 patients sustained fractures caused by gunshots and had a postoperative computed tomographic scanogram.

MAIN OUTCOME MEASURES: DFV and DFL. The effect of the following variables on DFV and DFL were determined through univariate and stepwise multivariate regression analyses: age, sex, body mass index, trauma fellowship-trained versus nontrauma surgeon, daytime versus nighttime surgery, antegrade versus retrograde nail insertion, use of traction, type of operating table, and AO and Winquist classifications.

RESULTS

The mean postoperative DFV for all patients was 8.62 degrees (±6.67 degrees). Postoperative DFV greater than 15 degrees was found in 12.3% of all patients. After IMN, no significant differences in DFV were found with increasing complexity of AO/OTA or Winquist fracture classification. None of the aforementioned independent variables were significantly predictive of postoperative DFV in univariate or multivariate analyses. The mean postoperative DFL for all patients was 5.25 mm (±4.36 mm). In a multivariate model, classification as Winquist type 3 or 4 was weakly (adjusted R = 0.075) but significantly predictive of less DFL than categorization as type 1 or 2 (P = 0.027).

CONCLUSIONS

Although gunshot-associated femur fractures may present surgical challenges for treatment through IMN, acceptable femoral rotation and length are obtainable regardless of the fracture complexity or a variety of demographic and surgically-related variables.

LEVEL OF EVIDENCE: Prognostic level II.