OBJECTIVETo explore the value of three-dimensional digital reconstruction techniques in the diagnosis and treatment of proximal humeral fractures.

METHODSFrom January, 2008 to May, 2010, 25 patients with proximal humeral fractures underwent preoperative examinations with conventional X-ray, plain CT scan and 3D-CT scan. Based on the two-dimensional gray-scale CT scan data in DICOM format, personalized virtual fracture models were reconstructed using Amira4.1 software and compared with the conventional imaging examination methods for fracture typing to identify the fracture classification. Personalized surgical treatment was implemented according to the typing of the fractures, and the patients were followed up postoperatively for 8 to 35 months (mean 21 months) for functional evaluation using the Constant-Murley score.

RESULTSIn the 25 cases of proximal humeral fractures, plain X-ray examination obtained a clear diagnosis and classification of the fractures in 15 cases, and CT scan, three-dimensional CT, and personalized three-dimensional model reconstruction confirmed the diagnosis and classification in 20, 22, and 25 cases, respectively. The humeral fractures healed generally 6 months after the operation, and the patients showed a Constant-Murley score ranging from 70 to 100 (mean 88).

CONCLUSIONSPreoperative personalized three-dimensional reconstruction of proximal humeral fracture can help obtain a more accurate diagnosis of the fracture type to facilitate the decision on the optimal surgical plan, surgical approach and internal fixation methods.