Yongsung Kim

Background: Although radiographs are the first-line imaging modality, differentiating between neoplastic pathologic fractures and nonpathologic fractures on radiographs can sometimes be challenging. In this study, we aimed to develop and evaluate a deep learning model capable of distinguishing neoplastic pathologic fractures from nonpathologic fractures on hip radiographs to enhance diagnostic accuracy. Methods: This retrospective, multicenter study analyzed anteroposterior hip radiographs from patients who visited the emergency department at 4 different institutions. The deep learning model was trained on, and tested using, data from 338 patients at a single institution and externally validated on data from 488 patients across 3 additional institutions. Results: The model achieved an overall accuracy of 0.880, with a sensitivity of 0.882 and a specificity of 0.879, on the internal test set. It was then externally validated using the data of 488 patients (67 with neoplastic pathologic fracture and 421 with nonpathologic fracture) from institutions separate from where the model was developed. The model achieved an overall accuracy of 0.848, sensitivity of 0.910, and specificity of 0.786. Its performance was comparable with that of general orthopaedic surgeons. Conclusions: The developed deep learning model is a reliable and valid tool for distinguishing neoplastic pathologic fractures from nonpathologic fractures on hip radiographs. It has the potential to assist orthopaedic surgeons in resource-limited settings, where optimizing the interpretation of radiographs is critical for patient care. The model is publicly available at https://pathfxdx.org. Level of Evidence: Diagnostic Level III . See Instructions for Authors for a complete description of levels of evidence.

Background/Objectives: Denosumab has been considered effective for downstaging giant cell tumor of bone (GCTB), implying that it lowers the Campanacci grade rather than reducing tumor size. Preoperative tumor shrinkage holds therapeutic value by decreasing surgical complications. While previous studies have observed lesion shrinkage in some patients, no study has identified the types of patients likely to show treatment response. Thus, we sought to identify the clinical factors associated with post-denosumab GCTB size reduction. Methods: The data for 45 GCTB patients (29 females [64%], 16 males [36%], median age 32 years [range: 17–65]) who received denosumab were analyzed. Lesions were in the extremities (n = 25, 56%) or spinopelvic region (n = 20, 44%). Ten (22%) were recurrent. Lesion size reduction was assessed using two criteria: a ≥5% and a ≥5 mm decrease in the longest diameter. Univariate analyses were conducted for all variables, and those found to be significant were subjected to multivariate analyses. In addition, multicollinearity was evaluated. Subgroup analyses were performed based on lesion location and recurrence status. Results: Campanacci grade III predicted proportional shrinkage (≥5%) in all patients (OR 4.819, 95% CI 1.121–20.714) and in extremity (OR 11.171, 95% CI 1.023–122.014) and primary lesions (OR 5.781, 95% CI 1.181–28.297), and aneurysmal bone cyst (ABC)-like change was associated with absolute shrinkage (≥5 mm) in all patients (OR 8.734, 95% CI 1.159–65.845) and primary lesions (OR 11.936, 95% CI 1.074–132.69). The longest tumor diameter of ≥7 cm predicted absolute shrinkage in all patients (OR 12.380, 95% CI 1.038–147.694) and for spinopelvic lesions (OR 20, 95% CI 1.676–238.63). Conclusions: ABC-like change, Campanacci grade III, and the longest tumor diameter might predict post-denosumab GCTB shrinkage in all patients, though shrinkage varies with lesion location and recurrence status. These factors could help clinicians tailor treatment strategies in different settings. Further research is needed to explore how clinical factors pharmacologically influence denosumab-induced GCTB shrinkage.

Kim S, et al. Clin Orthop Surg. 2025 May;17:e35. https://doi.org/10.4055/cios24336

Cho S, et al. Clin Orthop Surg. 2023 Oct;15(5):853-863. https://doi.org/10.4055/cios22254

The research on surgical outcomes of hemiarthroplasty and reverse total shoulder arthroplasty using allograft-prosthesis composites for the proximal humeral oncologic condition is still scarce. Therefore, this study aimed to compare surgical outcomes of shoulder joint reconstruction with hemiarthroplasty and reverse total shoulder arthroplasty using allograft-prosthesis composites for tumors of the proximal humerus. Eleven patients underwent hemiarthroplasty or reverse total shoulder arthroplasty using allograft-prosthesis composites for tumors of the proximal humerus between July 2011 and April 2018 were reviewed. Radiographic analysis for bone union of allograft-host bone junction, implant loosening, stress shielding and shoulder dislocation or subluxation was performed. Functional outcomes were evaluated using visual analog scales for pain, range of motion, Simple Shoulder Test score and Musculoskeletal Tumor Society score. Furthermore, oncologic outcome and complications were also assessed, respectively. There were five patients with hemiarthroplasty (mean age, 23.2 years) and six patients with reverse total shoulder arthroplasty (mean age, 46.8 years, P = 0.05). Radiographically, there were no events associated with implant loosening, stress shielding and shoulder dislocation or subluxation in the two groups. There were no differences in functional outcomes between the two groups. There was no local recurrence in entire cohort. In the hemiarthroplasty group, one patient was required revision surgery to reverse total shoulder arthroplasty at postoperative 6 years due to rotator cuff dysfunction. In the reverse total shoulder arthroplasty group, one patient showed the fracture occurred at allograft-host bone junction at postoperative 6 months. Surgical outcomes of hemiarthroplasty with allograft-prosthesis composites were not inferior to reverse total shoulder arthroplasty when applied in properly selected patients. The authors recommended that hemiarthroplasty with allograft-prosthesis composites could be used for young age patients without glenoid metastasis involvement, and reverse total shoulder arthroplasty with allograft-prosthesis composites could be used for patients with old age or metastatic bone tumors.

Introduction: Hemicortical resection is challenging when a huge fungating tumor is covering the osteotomy site. We report the clinical outcome of hemicortical resection and reconstruction for primary bone tumors, especially with high‐grade histology and extensive circumferential involvement.Materials and Methods: We retrospectively reviewed 44 patients (males, n = 18; females, n = 26) who underwent hemicortical resection from 2005 to 2014.Results: The median follow‐up period was 46.0 (23–178) months. Disease‐specific, local recurrence‐free, and metastasis‐free survival rates of patients in the malignant group at 5 years were 96.6%, 84.5%, and 93.6%, respectively. Among 42 patients, there were local recurrences (n = 6), metastasis (n = 2), and death (n = 1). Surgical margin was an independent prognostic factor for local recurrence (hazard ratio = 5.7; p = 0.038). The recycled autograft and strut allograft groups did not show statistical difference in bone union. Failure rate was 31.8% and local recurrence was the most frequent, followed by infection.Conclusion: Hemicortical resection can be a feasible option for extremity malignant bone tumors. Regarding reconstruction, there were no difference between autograft and allograft in bone union rate. Surgical margin was an independent prognostic factor for local recurrence.

김용성, et al. J Korean Orthop Assoc. 2021 Feb;56(1):42-50. https://doi.org/10.4055/jkoa.2021.56.1.42

김용성, et al. J Korean Orthop Assoc. 2020 Dec;55(6):511-519. https://doi.org/10.4055/jkoa.2020.55.6.511

김용성, et al. J Korean Orthop Assoc. 2020 Oct;55(5):397-404. https://doi.org/10.4055/jkoa.2020.55.5.397