Does the quality of preoperative closed reduction of displaced ankle fractures affect wound complications after surgical fixation?

Category: Trauma Introduction/Purpose: Prompt reduction and stabilization of displaced ankle fractures is important to protect soft tissues, restore potential neurovascular deficits and prevent cartilage injury. Many of these injuries do eventually require surgical fixation. The purpose of this study is to determine whether the initial quality of ankle closed reduction based on radiographic criteria would affect outcomes such as ankle osteoarthritis and complications after surgery. Furthermore, we sought to develop a classification system for the quality of closed reduction that would be easy to use and provide interrater reliability. Methods: A retrospective analysis of patients who sustained isolated, closed ankle fractures with at least 3 months follow up postoperatively at two level 1 trauma centers was performed. Patient demographics and history, ankle fracture characteristics and reduction information as well as surgical outcomes and complications were collected. A grading classification for the quality of the initial closed reduction before surgery was developed based on standard AP or mortise and lateral ankle x-rays. The factors considered for rating the reduction included the degree of talar shift on the AP/mortise view, malleoli displacement, as well the relationship of a central plumb line to the center of the talar dome on the lateral x-ray. For ankle osteoarthritis, the Takakura classification was utilized. Three reviewers (1 resident, 2 attendings) independently reviewed and rated all imaging. Results: 161 patients were analyzed. 65% female, average age 50, average 4 days between injury and surgery, mean follow up of 12 months (3-58 months), and 17% wound complications. Psychiatric history was the single comorbidity significantly associated with complications (p=0.009). There was no difference in wound or infection complication rates based on initial closed reduction quality (p=0.17). Neither number nor quality of reductions correlated with increased osteoarthritis (p=0.19, 0.39 respectively). Worst graded reductions had shorter time to surgery, mean 1.4 vs 4.7 days for best reductions (p=0.03), suggesting a protective factor that may account for no association between reduction quality and wound complications. Interclass correlation coefficients for multiple observers showed very high consistency for grading of reduction quality based on the classification system (ICC >0.85, p