Radiographic Accuracy of Identifying Anterolateral Tibial Plafond Involvement in Pronation External Rotation Ankle Fractures

OBJECTIVES: To evaluate the rate of anterolateral tibial plafond involvement in pronation-external rotation (PER) ankle fractures and analyze the accuracy of radiographs, using CT as the reference standard among fractures with paired imaging, in detecting anterolateral tibial plafond involvement, impaction, and predicting surgeons' stated intent to pursue direct visualization and articular reduction, as determined by surgeon assessment of the imaging findings. METHODS: Design: Multi-institutional retrospective chart review. SETTING: Two level 1 trauma centers in the United States. PATIENT SELECTION CRITERIA: Adult patients operative treated PER ankle fractures (AO/OTA 44C) from 2020-2022 and adequate preoperative radiographs, defined as an AP, mortise and lateral view of the injured ankle were included. A subset of these patients also had preoperative CT imaging of the injured ankle and were therefore evaluated with both imaging modalities. OUTCOME MEASUREMENTS AND COMPARISONS: All images were reviewed by 4 fellowship-trained orthopaedic trauma surgeons. For this study, anterolateral tibial plafond involvement was defined as the presence of any fracture line or impaction involving the anterolateral plafond, and impaction (a subset of involvement) as where the anterolateral plafond was crushed and depressed. A finding was considered present if at least 3 of the 4 reviewers indicated its presents on first-round responses. The rates of involvement and impaction were determined among all fractures based on radiographs alone, and separately among the subset of patients with paired radiographs and CT. Using CT as the gold standard, we determined the accuracy of radiographs and compared surgeons' stated intent to pursue direct visualization and articular reduction of the anterolateral plafond fracture before versus after review of CT imaging. RESULTS: Eighty-three PER ankle fractures were included (mean age 43.3 years [range, 17-89]; 63.9% male), of which 37 also had CT scans (mean age 39.2 years [range, 17-83]; 73.0% male). Among all 83 fractures, anterolateral tibial plafond involvement and impaction were identified on radiographs in 14.5% and 4.8% of cases, respectively. Among the 37 PER ankle fractures with both plain radiographs and CT scans, anterolateral plafond involvement was identified in 13.5% of radiographs versus 37.8% of CT scans (p=0.346) and impaction in 5.4% of radiographs versus 8.1% of CT scans (p=0.005). The diagnosis of impaction using plain radiographs was correct in 66.7% of fractures compared with CT, resulting in a sensitivity of 67%, specificity of 100%, positive predictive value (PPV) of 100%, and negative predictive value (NPV) of 97%. The diagnosis of anterolateral plafond involvement using plain radiographs had a sensitivity of 21%, specificity of 91%, PPV of 60%, and NPV of 66% compared with CT. Plain radiographs predicted surgeons' stated intent to pursue direct visualization and articular reduction in 66.7% of cases and had a PPV of 100% and NPV of 97%. CONCLUSIONS: In PER ankle fractures, the rates of anterolateral tibial plafond involvement and impaction were 37.8% and 8.1% on CT, respectively. Plain radiographs demonstrated high specificity but limited sensitivity for identifying both impaction (67%) and involvement (21%). Because anterolateral plafond impaction was relatively uncommon in PER fractures, a CT obtained for this purpose alone would yield a positive finding in only a minority of cases; nonetheless, CT should be considered for PER ankle fractures more broadly, given that plain radiographs miss a meaningful proportion of both involvement and impaction and interobserver reliability of radiograph-based assessment was poor. LEVEL OF EVIDENCE: Prognostic level III.