Operative Treatment of Isolated Split Greater Tuberosity Fracture (mutch type II) of Proximal Humerus with Modified PHILOS

Background: The greater tuberosity (GT) of the proximal humerus is the attachment point of the rotator cuff, which plays a role in the movement of the shoulder joint and is the core of the entire shoulder joint. Material and Methods: In our current study, 40 patients with isolated split GT fracture (mutch type II) from july 2017 to January 2019, which was typed by J.Mutch professor, Canada in 2014, were employed in the study. They were divided into two groups: the Modified PHILOS plate group (group A, n=20) and the Hollow Screw group (group B, n=20). The functional scores Constant‑Murley Score (CMS), American Shoulder and Elbow Surgeons (ASES), The University of California at Los Angeles shoulder rating scale (UCLA) and Visual Analogue Score (VAS) were recorded in both pre-op and post-op last follow. Results: Compared with last-follow, all shoulder scores (CMS, ASES, and UCLA) of group A was significantly better than group B (P<0.05), but VAS (P>0.05). Moreover, in post-op complications, there were one GT malunion and one shoulder pain in group A, but in group B, there were seven GT disappear (35%), three GT malunion and two patients with repeated shoulder pain. Conclusion: The Modified PHILOS plate in treating isolated split GT fracture (mutch type II) was found to be effective than hollow screw in the short term follow. However, there are still some post-op complications in both groups.

to fix humeral shaft in PHFs. Isolated Mutch type II GT fracture has only one large fracture fragment, and we should use a small and shorter plate to fix that fragment. So we cut off the shaft PHILOS plate and leave a 48 millimeter plate for use, which has a shorter length, incision and avoids the lesion of the axillary nerve.
In the past decades, many types were suggested to guide the treatment of PHFs and the most widely accepted classical were Neer and AO typing. In 1970, Neer divided PHFs into four partial fractures, which displaced more than 1 cm is considered to be appropriate for surgical treatment [8]. In recent years, many people began to pay more attention to GT fractures [9][10]. However, the two classical typing did not differentiate the isolated GT fracture. Moreover, the size, shape and fragment movement of the GT fragment could reflect different injury mechanisms. On the other hand, the morphology fracture fragment also influenced the choice of specific surgical methods.
In 2014, the GT fractures were further divided into three types: avulsion, split and depressed by professor J.Mutch in Canada, in which split fracture of the GT (as specific Fig. 1.II) was the most common, accounting for the 41% of entire fracture of the GT [11]. With that background, we tried to use this new morphology type and modified traditional PHILOS plate fixation in our study.

Ethical statement
All procedures were approved by the Ethical Committee of Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University (SWMCTCM2017-0710) and performed by the 1964 Helsinki declaration and amendments or comparable ethical standards. The specific illustration and agreement were obtained from all individual participants included in our study.

Inclusion Criteria
Patients were employed to meet such inclusion criteria: i) Aged 18-80, patient's history, physical examination, X-Ray and CT (2D and 3D) should conform to the humeral fractures of GT. ii) Confirmed isolated Mutch II (split) GT fracture of the proximal humerus, (2 people assessed respectively, and if there were any divergences to the results of the classification, it would be given to the third person to take a comprehensive evaluation and got the final results included in this experiment. iii) The patient and family members were willing to cooperate with the postoperative treatment. Then the doctor illustrated the advantages and disadvantages of surgical and the study plan before an operation.
Finally, the patient and the family signed on the agreement.

Exclusion Criteria
Other Patients should be excluded with such criteria: i) Type Mutch I or III of humeral GT fracture. ii) Patient whoever was followed up less than one year and did not cooperate with the medical staff for postoperative follow-up evaluation. iii) Patients with complications such as severe neurological and vascular injury. iv) The patient who had a history of mental illness or epilepsy or who are medicine or narcotic abuser.

Pre-op assessment
On admission, asking disease history in detail to be familiar with the time and mechanism of injury.
Judged their ROM through the shoulder examination and X-ray, CT (3D), and MRI should be taken, especially for CT scan. According to the author's experience, the fracture fragment showed better on CT (3D), which was very important for specific plans and assessments before surgery. Therefore, we recommended that all patients should have CT (3D), and MRI could better understand whether the patient had combined rotator cuff lesion, to make a comprehensive strategy.

Surgical technique
Group A: Modified PHILOS plate; with satisfactory anesthesia, patients were placed in the beach chair position, and taken the split deltoid incision; sharply and bluntly separated subcutaneous tissue, deep fascia, muscle, and other tissues layer-by-layer, until the fracture was exposed. (Noted to protect the axillary nerve, marked and protected Posterior humeral circumflex artery). On the other hand, we prepared the modified PHILOS plate and cut off the three holes screw on the shaft. Moreover, the fragment was accurately fixed by K-wires, and the modified plate was at 5 mm under the top of the GT and the medial intertubercular sulcus. The correct position was checked with fluoroscopy. Finally, fragment reduction and screw length were ensured by fluoroscopy.
Group B: The hollow screw; the same preparation as group A. Then, 2.0 K-wires wires were used to have reduction and fixation under fluoroscopy in a 90°. (Note: The direction of the inserting needle should not be parallel to the biomechanical direction of the supraspinatus muscle pull as far as possible; the best position was to form an angle and increase the fixation biomechanics). Two K-wires were used to fix fracture fragments, and then, the C-arm was used to fluoroscopy reset effect. If the reset effect was not effective as expected, adjust its position until having a reduction. Screws have been fixed by the guidance K-wire.

Postoperative review and functional exercise
Functional exercises were started two days after the operation. The arm was supported by a sling for three weeks, which was protected at the abduction of 45 degrees. We tried to avoid making the fixed GT fragment displacement via the supraspinatus or shoulder activities. Passive motion exercises began in the third weeks of operation. Active motion exercises and strengthening exercises were started at six weeks after surgery. Further continual follow-ups were carried out at every 1, 3, 6 and The data of Shoulder function scores of CMS, ASES, UCLA, and VAS in the two groups were expressed as the mean ± standard deviation (± S) analyzed by independent samples t test. The test level was α=0.05.

The pre-op baseline
Compared with group A and group B, there were no significant statistical differences in ages, gender, injured sites (left and right) and injured time between them (P>0.05) ( Table 1). Note: ap 0.05,compared with group B.

X-Ray
As was shown in Fig 2, with post-op follow-up of 1, 6 and 12 months, we could learn that the mutch type II GT fractures with fixation of modified PHILOS might show better outcomes. Moreover, we could employ a shorter incision for the lower part of PHILOS had been cutting down. On the other hand, in

Shoulder Functional Score
The respectively pre-op and post-op last follow-up scores of A and B 50±2.28, which shows that modified PHILOS plate is better than hollow screw (Fig 4. A-D & Table   2) (P< 0.05).

Complications
In our study, although we have gotten an acceptable shoulder function, there was still one GT malunion and one shoulder pain in group A. In group B, there were seven GT disappeared (35%), three GT malunion and two patients with repeated shoulder pain. which mean the morphological type of the greater tuberosity is better than the Neer and AO types.
Conservative and surgical treatment in displaced fractures of the GT had been reported [12][13][14][15]. Park TS recommended that GT fractures with a displacement greater than 5 mm should be treated operatively [16]. On the other hand, Rath E had shown that conservative treatment with a displacement of less than 3 mm could achieve clinical results [17][18]. In our study, isolated GT fracture, J.Mutch type, was employed to be our diagnosis. Mover, treatment and investigation algorithm reviewed by professor Rouleau DM, was also adopted in this research. However, this classification also had some disadvantages. As is shown in Fig 3. Our results show that the Modified PHILOS provides a stable fixation and patients get a better functional outcome. Moreover, the modified PHILOS has some specific advantages as follows. It is a small locking plate shorter than normal PHILOS plate, especially good for elder patients with osteoporosis. We can employ a small incision and protect the soft tissue. However, the costeffectiveness is higher than hollow screw group. Furthermore, the modified PHILOS can't fix the superposterior part of GT fracture with multiple fragments, and this may be one of the reasons cased shoulder dysfunction. On the other hand, a hollow screw is cheaper than modified PHILOS.
Additionally, we fix the fragment with a percutaneous way, which has a smaller incision, and less soft tissue lesion [24]. But it is hard for us to perform it on osteoporosis patients.
Older patients with more than 60 years are likely to sustain PHFs [25][26]. In our study, we employ However, our study also has some limitations. i) The modified PHILOS plate does not fix the superposterior fragments of GT. ii) J.Mutch type was only assessed by X-ray without 3D CT and MRI, which can't provide a better assessment of whether the fracture combines with rotator cuff tears and the number of fragments. If the rotator cuff tear is included, it is better for us to use suture anchor when fixing the fracture. In later studies, MRI should be performed routinely. iii) We have no idea about vascular injury occurred when GT fracture happened, and this might affect results of GT disappeared and malunion. iv)The patients participating in this study were all chosen from the same hospital, and the number of cases was small. There were geographical limitations, and the operator was a professor who had studied in Germany with extensive experience in the shoulder joint.

Conclusion
It is necessary to classify and treat subtypes of greater tuberosity alone. Morphological types for GT fracture is simply and practically. However, the classification does not enroll the injury mechanism and biomechanics, and the mechanism of the injury need to be considered in the process of clinical surgery, which aims at restoring mechanism balance biomechanical of rotator cuff. Under the guidance of this classification, the Modified PHILOS plate in treating Mutch type II GT fractures was found to be effective than hollow screw in the short term follow. However, there are still some post-op complications in both groups.

Availability of data and materials
The datasets used and analysed during the current study available from the correspond-ing author on reasonable request. Figure 1 14 The Morphology Classification of Isolated Humerus Greater tuberosity fracture. Type I avulsion fracture involves small fragments of bone with a horizontal fracture line. Type II split fracture involves a large fragment with a vertical fracture line. Type III depressed fracture involves a fragment that is displaced inferiorly.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download. Data.xlsx