Accuracy of detecting burst of the lateral wall in intertrochanteric hip fractures with plain radiographs: Is postoperative CT necessary?

Background: Postoperative burst of the lateral femoral wall is thought to be the main predictor of reoperation for intertrochanteric fractures, which is routinely evaluated using plain radiographs. We retrospectively compared computed tomography (CT) scans and radiographs regarding the ability to detect burst of the lateral wall. We also investigated whether intramedullary nails may cause iatrogenic burst of the lateral wall. Methods: From January 2010 to December 2021, patients aged 65 years and older who undergone intertrochanteric fractures treated with the proximal femoral nail antirotation 2 (PFNA-II) were included. The incidence of burst of the lateral wall was evaluated with two different imaging modalities by two observers. Two rounds of evaluation were performed: (1) with plain radiographs alone; and (2) with CT scans combined with radiographs. Interobserver and intraobserver agreement (kappa value) for evaluation of the lateral wall burst was assessed. Results: A total of 1507 patients were included (362 males and 1145 females). Compared with radiographs alone (12.0 %, 181/1507 patients), a higher rate of lateral wall burst was found by CT scans combined with radiographs (72.9 %, 1098/1507 patients) for observer 1 at first reading (P < 0.001). Similar results were seen in other evaluations. Interobserver and intraobserver agreement was substantial for radiographs alone (kappa, 0.659-0.727) and almost perfect for CT scans combined with radiographs (kappa, 0.847-0.926). Conclusions: Computed tomography combined with radiographs is superior to radiographs alone for detecting burst of the lateral wall after intertrochanteric fracture fixation. Additionally, PFNAII could cause iatrogenic burst of the lateral wall for intertrochanteric fractures in the elderly.