This study was performed with approval from Dankook university hospital institutional research ethics committee and informed consent from all the patients (DKUH-2020-04-035). The procedures in this study were performed under the Declaration of Helsinki’s ethical principles for medical research involving human participants. Written informed consent was obtained from all individual participants included in the study. We enrolled 45 consecutive patients who had undergone an open-reduction and internal fixation procedure for a proximal humeral fracture at our institution during the period of January 2013 and May 2018.
Patients with 2-, 3-, or 4-part fractures were included in this study on the basis of the Neer classification. (14) Patients with surgical neck fractures deemed to be unstable by a surgeon were included. Exclusion from the study required that patients meet at least one of the following criteria: treatment with the minimally invasive plate osteosynthesis technique or fibular allograft, combined large to massive rotator cuff tears, pathological fractures, an irreparable head and/or tuberosity fragments, stable fractures with intact medial support, and an immature skeleton or loss to follow-up before 12 months of enrollment. The BCL screw fixation technique was used in the first 27 cases, and the FCL screw fixation technique was used in the final 18 consecutive cases. Patient information was obtained from the patients’ medical records, including age, sex, dominant hand, height, weight, body mass index, American Society of Anesthesiologists (ASA) classification, mechanism of injury, time to surgery, and smoking history.
Description of the BCL screw fixation procedure
Under general anesthesia, all the patients underwent surgery in the supine position using the standard deltopectoral approach. The fracture fragments are first temporarily reduced with Kirschner wires, and then sutures are passed through the rotator cuff tendon and fixed with a proximal humeral internal locking system (PHILOS) plate (DePuy Synthes, Zuchwil, Switzerland) with one 3.5-mm cortical screw and two or three BCL screws (Fig. 1). After the PHILOS plate fixation, the 3.5-mm cortical screw is loosened from the longitudinal combi-hole of the plate shaft. The rotator cuff tendons are then attached with four to five No. 5 non-absorbable braided sutures (Ethibond, Somerville, NJ, USA). The sutures are then passed through the 3.5-mm cortical screw and tied off. Finally, the cortical screw is fastened fully to complete the procedure.
Description of the FCL screw fixation procedure
The same position, approach, reduction, and fixation are used in the FCL and BCL screw fixation procedures. After placement, the periarticular proximal humeral locking plate (Zimmer Biomet, Warsaw, Indiana, USA) is secured with three FCL screws, and the rotator cuff tendon is augmented with multiple sutures tied to the plate through the suture holes (Fig. 2).
Postoperative rehabilitation and implant removal
After surgery, the affected arm was first kept in a splint for 1 week and then in a sling for the following 6 weeks. Pendulum, self-assisted circumduction, and gradual passive range of motion (ROM) exercises were implemented 1 week postsurgery if tolerated by the patient. Furthermore, 6 weeks postsurgery, active ROM exercises were implemented. Implant removal was conducted 3 months postsurgery, provided that bone union was accomplished. If the bone union was not complete at 3 months, implant removal was delayed until full union was observed. If the patient presented with stiffness of the shoulder joint, brisement was performed concomitantly at the time of implant removal.
Clinical and radiological evaluations
Clinical assessments included the constant score, the American Shoulder and Elbow Surgeons (ASES) score, and range of motion. These assessments were recorded by the physician’s assistant at the final follow-up. Full range of motion was measured and documented, including active forward flexion, abduction, external rotation, and internal rotation at the back. An independent examiner blinded to all patient data evaluated the values at each postoperative follow-up visit.
The radiological evaluation included both clavicle and anteroposterior views taken at regular intervals after surgery (3, 6, and 12 weeks; 6 months; and 1 year post surgery). The radiographic outcomes included union rate, union at 12 weeks, and structural alignment. Bony union was defined by the existence of a bridging callus in not less than three cortices in two planes. Delayed union was defined as an incomplete radiographical consolidation after ≥6 months. Alignment was evaluated immediately after the procedure and 1 year after surgery, using the Paavolainen method(14), which measures the humeral neck-shaft angle (NSA). The NSA was determined as a line perpendicular to another line, between the superior and inferior borders of the articular surface on anteroposterior radiography. NSA is formed by the intersection of this perpendicular line and another line, bisecting the humeral shaft (Fig. 3).(14)The NSA was analyzed by two independent examiners (J.K.K. and G.Y.K.). Individual values were measured, and the mean values were calculated. All complications were recorded, including fracture collapse, screw penetration, avascular necrosis of the humeral head, peri-implant fracture, postoperative infection, stiffness of the shoulder joint, and the necessity for additional surgery.
Statistical analyses
To establish if the continuous data were normally distributed, the Kolmogorov-Smirnov test was used. An independent t test was performed to analyze continuous variables; and a Pearson chi-square test, to analyze non-continuous variables. All statistical analyses were performed through the Statistical Package for Social Sciences version 25.0 (SPSS, Inc., an IBM Co., IL, USA).A p value of <0.05 denoted statistically significant differences in all the analyses.