Locking plate fixation of three- and four-part proximal humeral fractures.

OBJECTIVES We evaluated the functional results of open reduction and internal fixation with a locking plate in patients with three- or four-part fractures of the proximal humerus. METHODS We reviewed 32 consecutive patients (22 women, 10 men; mean age 63 years; range 29 to 82 years) who were treated with open reduction and internal fixation using the PHILOS locking plate for comminuted proximal humeral fractures. According to the Neer classification, 12 patients (mean age 56 years) had three-part fractures, 19 patients (mean age 67 years) had four-part fractures, and one patient had a four-part fracture dislocation. Ten patients were in the age group of <60 years, 22 patients were in the age group of 60=or>years. All the patients were evaluated with plain radiographs preoperatively; in addition, computed tomography was used in 14 patients in whom articular surface and tuberculum displacement could not be assessed adequately. The operation was performed through a standard deltopectoral approach, and minimal soft tissue dissection was used aiming not to impair vascularization of the fracture fragments. A cerclage wire was used to help reduction in 12 patients. An oblique screw was inserted to stabilize the medial colon in cases in which medial cortical contact was insufficient. Bone grafting was not used in any of the patients. Active-assisted and passive exercises of the shoulder were initiated on the second postoperative day. Active abduction to 90 degrees was allowed two weeks after surgery. During follow-up, implant failure, loss of reduction, malunion, and bone healing were assessed on plain radiographs. Bone scintigraphy was performed after 12 postoperative months for the detection of avascular necrosis. The results were assessed using the Constant shoulder score. The mean follow-up period was 25 months (range 18 to 36 months). RESULTS An anatomic or near-anatomic reduction was obtained in 29 patients (90.6%). In two patients, the fractures were fixed in a varus position, and in one patient, the greater tubercle was displaced proximally. All fractures united in a mean of three months (range 2 to 5 months). The mean Constant score of the patients was 79.5 (range 50 to 100). The results were excellent in 13 patients (40.6%), good in nine patients (28.1%), fair in eight patients (25%), and poor in two patients (6.3%). The mean Constant scores were 88.3 (range 69 to 100) and 74.2 (range 50 to 100) in three-part and four-part fractures, and 88.3 (range 71 to 100) and 75.5 (range 50 to 100) in the age groups of <60 years and =or>60 years, respectively. Constant scores showed significant differences with respect to the number of comminution and age groups (p=0.03). Avascular necrosis was observed in two patients. None of the patients had reduction loss, implant failure, deep infection, or neurovascular injury, and none required implant removal. CONCLUSION Preservation of humeral head vascularity through minimal soft tissue dissection, fixation with a locking plate, and early postoperative motion were effective in decreasing potential complications following surgical treatment of three- and four-part proximal humeral fractures. The degree of fracture comminution and age of the patients affect functional results significantly.