Increased Risk of Complications Associated with Magnetic Resonance Imaging Prior to External Fixation in Unstable Knee Dislocations.

Abstract

Unstable knee dislocations (KDs) may be assessed with magnetic resonance imaging (MRI) and treated with knee spanning external fixation (ex-fix) prior to definitive multi-ligament reconstruction. The order in which ex-fix and MRI are performed is debated due to the risk of complications. The following study evaluates the risks associated with MRI completion before versus after ex-fix placement, including risk of redislocation compared to risk of MRI related injury. KDs treated with knee spanning ex-fix were retrospectively reviewed at a Level-I academic center over a 10-year period. Exclusion criteria included no MRI completion or ex-fix removal prior to MRI. Study groups were defined as KDs that received MRI prior to ex-fix (MRI-first) and ex-fix prior to MRI (ex-fix-first). Demographic data were collected including age, sex, and body mass index (BMI). Primary outcomes were adverse events during MRI including redislocation/subluxation events, pin pullout, or thermal injury. Secondary outcomes included radiographic malreduction, early MRI termination, and repeat MRI within 3 months. Data analysis was conducted with Mann-Whitney U, Chi-squared, and Fisher's exact tests, with a p-value < 0.05 indicating significance. Twenty-four knees met inclusion criteria: 8 were MRI-first versus 16 which were ex-fix-first. No statistically significant differences in demographic variables were found between study groups. In the MRI-first group, 4/8 (50%) had a redislocation or severe subluxation event in the scanner versus 0/16 (0%) in the ex-fix-first group, (p=0.007). In the ex-fix first group, 0/16 (0%) suffered thermal injuries or pin pullout. MRI-first knees were significantly more radiographically malreduced in the sagittal and coronal plane (p=0.017 and 0.047 respectively after excluding MRIs completed with redislocation and KD Vs with marked tibial plateau displacement). Repeat MRI occurred in 3/8 (37.5%) of MRI-first knees and 1/16 (6.3%) of ex-fix-first knees (p=0.091). In the setting of grossly unstable knee dislocations, MRI prior to ex-fix placement put the patient at a high risk of redislocation and more often required repeat MRI. There were no cases of thermal injury or pin pullout with ex-fix placement before MRI. We advocate for a multidisciplinary protocol in order to assure prompt and appropriate treatment. Level III: Retrospective comparative study.