Comparison of the Outcomes With Suture Anchors Versus Anatomical Locking Plates for Humeral Greater Tuberosity Fractures: a Case Control Study.

Background: Humeral greater tuberosity fractures account for approximately one-fth of the fractures of the proximal humerus. There are a variety of internal xation devices used in the treatment of fractures. Suture anchors and anatomical locking plates are the two commonly used internal xation devices. However, there are no systematic studies on which device is more advantageous for the treatment of greater tuberosity fractures. Materials and methods: We retrospectively analyzed the medical records and radiological results of patients with humeral greater tuberosity fractures who underwent surgical treatment and were followed up for at least one year from June 2017 to March 2020. A total of 31 patients were included in the study, including 13 males and 18 females, with an average age of 51.4 years old (18 to 75 years old). According to the different xation methods, the patients were divided into the suture anchor treatment group (13 cases) and the anatomical plate treatment group (18 cases). The clinical and radiological results of these two internal xation devices were evaluated by various criteria. Results: All 31 patients were followed up successfully, with a mean follow-up time of 27.74 months. There were no differences in patient demographic and characteristic information between the two groups. The blood loss in the suture anchor treatment group was less than that in the anatomical plate treatment group. In contrast, the average operation time of the anatomical plate treatment group was shorter than that of the suture anchor treatment group, but the results were not signicantly different. The postoperative Visual analog scale score, Constant-Murley shoulder joint function score and range of motion of the two groups were not signicantly different. In addition, there was no signicant difference in the postoperative complication rate between the two groups. Conclusion: Both suture anchors and anatomical plates can achieve satisfactory clinical results in the treatment of humeral greater tuberosity fractures, and there is no signicant difference in the long-term functional results between the two groups. These two xation devices can be exibly selected according to the surgeon's experience and fracture characteristics.


Introduction
Proximal humeral fractures are one of the most common types of upper limb fractures. They are the third most frequent fractures in elderly patients after hip and distal radius fractures. They are mostly caused by low-energy trauma and are common in elderly women with osteoporosis3. Greater tuberosity fractures account for one-fth of proximal humeral fractures. They usually occur in young male patients with better bone quality, and the energy of the injury is usually higher. The main mechanisms of injury for greater tuberosity fractures are falls from a height over an outstretched hand with the elbow to the ground in external rotation associated with shoulder abduction or direct trauma on the lateral side of the shoulder10. Three classi cation systems are commonly used to describe greater tuberosity fractures: the AO, Neer, and morphological classi cations13. Approximately 5%-57% of greater tuberosity fractures of the humerus are combined with shoulder dislocation. Inappropriate treatment can lead to acromion impingement, shoulder joint pain, and limited mobility, which affect the daily life and work of patients17.
Fortunately, most of these fractures are stable and nondisplaced or minimally displaced and treated conservatively. A recent study indicated that 85%-95% of greater tuberosity fractures do not require surgical treatment, and only 5%-15% of displacements greater than 5 mm require surgical intervention2. A variety of surgical devices have been used in the past, including lag screws, trans-osseous suture xation, tension band wiring, proximal humeral internal locking system (PHILOS), suture anchors, and arthroscopic techniques. However, the optimal treatment of displaced fractures is still controversial18.
The choice of methods for treating isolated displaced fractures of the greater tuberosity mainly depends on the fracture type, the bone quality, the fracture fragment size, and the preference of the surgeon. Bony healing of the displaced greater tuberosity critically depends on anatomical reduction and rigid xation, and mechanical stability further allows early rehabilitation of the shoulder to achieve satisfactory clinical outcomes16. Anatomical locking plates and suture anchors are currently two commonly used xation methods for the treatment of these fractures. However, there is no systematic study on the long-term clinical results between the two devices. Some scholars believe that arthroscopic-assisted suture anchor xation can achieve satisfactory clinical results. However, arthroscopy has higher requirements for technology and equipment, and a long learning curve is required. In addition, a study found that there were few important differences between arthroscopic suture anchor xation and ORIF for isolated displaced greater tuberosity fractures7. This limits its clinical popularity. The anatomical locking plate is also widely used for fractures of the greater tuberosity, and most studies have con rmed that it can also achieve satisfactory long-term results. However, there is no report on the differences between suture anchors and anatomical locking plates. Therefore, we retrospectively analyzed the clinical results for these two xation devices in the treatment of greater tuberosity fractures. The hypothesis of this study was that the clinical results obtained with suture anchors and anatomical locking plates for humeral greater tuberosity fractures would reveal equal functional and radiographic outcomes.

General information
The study design was approved by the Guyuan People's Hospital Clinical Research Ethics Committee and performed in accordance with the Declaration of Helsinki. Informed consent was obtained from the parents/guardians of the patients included in the study. In this study, 31 patients with humeral greater tuberosity fractures from March 2015 to June 2019 were taken as the research objects, including 12 males and 19 females, with an average age of 51.4 years old (18 to 75 years). The causes of injury included falling from heights, tra c accidents, and falling while walking or standing. All patients received shoulder X-rays and CT before surgery, and 14 patients (45.16%) underwent shoulder MRI scans to better determine rotator cuff injury. According to the AO classi cation, these fractures were type A extra-articular unifocal fractures (unifocally displaced greater tuberosity fracture A1.2). Among them, 19 cases were comminuted fractures, 20 cases were combined with rotator cuff injury, and 9 cases were combined with shoulder joint dislocation (Table 1. Demographics of patients). Spontaneous relocation of the shoulder joint might have occurred prior to clinical assessment. Cases might have been falsely assigned to the group of patients without shoulder dislocation. Therefore, a detailed medical history should be requested after hospitalization to avoid omissions. According to different surgical methods, the patients were divided into a suture anchor treatment group (11 cases) and an anatomical locking plate treatment group (20 cases). The postoperative follow-up was at least one year. All operations were performed by the same team of surgeons. During the operation, we also carefully checked to con rm whether cases were combined with damage to the rotator cuff and other accessory structures and treated them as needed.

Inclusion and exclusion criteria
All patients were chosen by the following criteria.
The inclusion criteria were as follows: (1) patients with a closed fracture with normal shoulder joint function before injury; (2) clinical symptoms, signs and radiological imaging ndings diagnosed as humeral greater tuberosity fracture; (3) fracture fragment displacement > 5 mm; and (4) follow-up for one year or more after the operation.

Suture anchor group
The patient was under brachial plexus or general anesthesia. After anesthesia was satis ed, patients were placed in a beach chair position, and the surgical area was routinely disinfected. The incision extends from the anterolateral edge of the acromion to the distal end with a length of 5-6 cm. The gap between the anterior and middle deltoids is carefully separated, exposing the displaced greater tuberosity fracture fragment and the rotator cuff attached to it. When the distal end of the incision is separated, attention should be given to protecting the axillary nerve after clearance of hematoma and scar tissue.
The displaced fracture fragment was reduced and temporarily xed with Kirschner wires. The rotator cuff is passed, and two suture anchors are inserted (Smith & Nephew Endoscopy, Andover, MA, USA) obliquely into the humeral head on the edge of the articular cartilage. When combined with a rotator cuff injury, the suture on the suture anchor can be used for supplemental xation. Then, a small hole is drilled in the humeral shaft 1-2 cm below the distal fracture line, and the anchor thread is passed through the small hole and tied. Alternatively, the end of the suture is squeezed with the distal row of anchors. After C-arm uoroscopy to determine the reduction of the fracture and the position of the screw was satisfactory, the shoulder joint is passively moved to check the stability of the fracture, and the incision is sutured layer by layer ( Figure 1).

Anatomical locking plate group
The posture and incision were the same as those in the suture anchor treatment group. The displaced greater tuberosity fracture fragment and the attached rotator cuff are exposed. After the fracture is reduced, Kirschner wire is used to temporarily x it. Subsequently, the anatomical locking plate (Double Medical Technology Inc, Xiamen, China.) is placed below the tip of the greater tubercle and 0.8 cm outside the biceps groove, screwed in with the appropriate length screw, sutured to the surrounding rotator cuff and tied to the plate through the suture eyelets for supplemental xation. After C-arm uoroscopy to determine that the reduction of the fracture and the position of the screw is satisfactory, the shoulder joint is passively moved to check the stability of the fracture, and the incision is washed and sutured layer by layer ( Figure 2).

Postoperative Management
Symptomatic treatments such as oral analgesics were given after the operation. An intravenous drip of antibiotics was given 0.5-1 hour before the operation, and another intravenous drip of antibiotics was given 12 hours after the operation to prevent infection. The postoperative rehabilitation protocols were similar for patients in the two groups. The shoulder and elbow bands were externally xed for two weeks after the operation. After two weeks, the external xation was removed, and the shoulder joint was passively moved. Part of the exion and extension activities were started four weeks after the operation. Internal and external rotation and abduction were performed six weeks after the operation. The surrounding bones were reviewed after three months. After the scab formed, strength training gradually started, and some antagonistic training began approximately half a year after the operation (Figure 3). All of these activities and training methods were completed under the guidance of professional rehabilitation therapists.

Data Collection
Surgical times were recorded during the perioperative period. Surgical time was de ned as the time from initiation of the incision to the time when suture of the incision was nished. Radiographs were obtained immediately after surgery to evaluate the adequacy of fracture reduction, which was determined by measuring the degree of residual displacement. Our follow-up doctors followed up patients in the outpatient clinic or by telephone at two weeks, one month, two months, three months, six months, and twelve months after surgery. Observe or take pictures of the relevant angle, range of motion, local skin, and patient satisfaction. The carrying angle and the range of motion were measured using a full-circle goniometer and compared with that of the contralateral arm. X-ray examination was required at each follow-up to determine the healing of the fracture. A detailed clinical and radiological examination was performed on all patients and documented. The images were blindly reviewed by a senior resident surgeon. The charts were reviewed by another resident surgeon.

Statistical Analysis
The data were analyzed using the Statistical Package for Social Sciences (SPSS version 13.0). The t-test was used to compare the postoperative clinical results, including ROM and functional scores. Statistical signi cance was set at a probability less than 0.05. The results are expressed as the means ± standard deviation (SD), and p<0.05 indicates a statistically signi cant difference.

Results
A total of 31 eligible patients were included in this retrospective study. No signi cant differences were found between the two groups according to demographics (Table 1), suggesting the comparability of the two groups. All 31 patients were followed up after surgery. All patients had achieved primary incision healing in both groups at the last follow-up. The blood loss in the suture anchor treatment group was less than that in the anatomical plate treatment group (Table 2). In contrast, the average operation time of the anatomical plate treatment group was shorter than that of the suture anchor treatment group, but the results were not signi cantly different. The postoperative VAS score and Constant-Murley shoulder function score were better in the suture anchor treatment group than in the locking plate treatment group, but the difference was not statistically signi cant. There was no signi cant difference in hospitalization time between the two groups. In terms of complications, the complication rate was 7.69% in the suture anchor treatment group and 11.11% in the locking plate treatment group. One patient in the plate treatment group had acromion impingement. One patient in each group had fracture displacement after surgery, but the displacement was less than 3 mm. After conservative treatment, the joint function recovered satisfactorily. For ROM, all indices, including forward exion, abduction, external rotation and internal rotation, were also non-signi cant (Table 3).

Discussion
Proximal humerus fractures are the third most frequent fractures in elderly patients after hip and Colles fractures. Greater tuberosity fractures account for one-fth of proximal humeral fractures. They usually occur in young male patients with better bone quality, and the energy of the injury is usually higher1. In our series of cases, there were more female patients than male patients. Regarding the age of the patient, our results are consistent with those previously reported. Our study found that these fractures were more common in 58.06% of females vs 41.94% of males, and the more affected group had an age < 65 years (83.87%) vs > 65 years (16.13%). The causes of injury included falling from heights, tra c accidents, and falling while walking or standing. The greater tuberosity of the humerus is one of the important attachment points of the rotator cuff. The supraspinatus and infraspinatus can induce superior displacement of the greater tuberosity. The proximal displacement of this part after fracture can lead to subacromial impingement and limitation of abduction and external rotation5. In addition, simple fractures of the greater tuberosity of the humerus are not common and are often associated with shoulder dislocation and rotator cuff tears. A study found that secondary interventions are required more frequently after shoulder dislocation. Surgery is most likely required for associated soft tissue lesions rather than for secondary displacements14. In our cases, 20 were combined with rotator cuff injury, accounting for 64.52%, and 9 cases were combined with shoulder joint dislocation, accounting for 29.03%.
The diagnosis and classi cation of isolated greater tuberosity fractures are mainly based on standard plain radiographs. However, these fractures may be challenging to identify because of osseous overlap9. Surgeons feel slightly more con dent about their treatment recommendation when assessing CT images with radiographs compared with radiographs alone4. In our study, all patients underwent CT examination before surgery. There are still many controversies regarding detailed surgical indications. Most scholars believe that conservative treatment can obtain satisfactory clinical and imaging results for fractures with a displacement of less than 5 mm, while surgical intervention is required for a greater tubercle fracture of the humerus with a fracture displacement of more than 5 mm, even 3 mm for a population with high demands on shoulder function12. Malunion is one of the most common complications following surgical treatment of proximal humeral fractures. The deltoid force required for abduction is signi cantly higher when the greater tuberosity is displaced more than 5mm11. Due to the limited dimension of the subacromial space, even small residual superior displacement may cause clinical impairment. Therefore, we deem that the surgical indication for greater tuberosity should comprehensively consider the displacement of the fragment, the width of the subacromial space, and the patient's rotator cuff injury.
Patients who sustain an isolated greater tuberosity fracture are younger and more active than those who sustain other proximal humerus fractures. Therefore, in the treatment of GT fractures, there is increased emphasis on returning to high activity and function6. Mechanical stability is essential to allow early functional rehabilitation to achieve satisfying clinical results. There are multiple xation methods, including screws, locking plates, suture anchors, and tension bands, to treat this type of injury. However, no established technique is considered the gold standard for the treatment of displaced greater tuberosity fractures. When screw xation is selected, small or thin bones are likely to cause fracture fragments to split, resulting in xation failure. The anatomical locking plate can better x the fracture fragment and can repair the injured rotator cuff during the operation. Since its application, satisfactory clinical treatment results have been achieved. In one study, the authors used a self-adjusted calcaneus titanium plate (Litos) that was cut into a 6 or more holed small plate. Excellent postoperative outcomes were obtained with no complications and no secondary loss of reduction in ten patients15. The results of a recent study show that a novel plate (Kaisidis plate, Fa. Königsee) has good biological properties in the treatment of displaced greater tuberosity fractures. However, the most critical points of the Kaisidis system are the screws that are connected to the bone, which may be limited in patients with osteoporosis6. In our cases, the functional results of the plate we used in the treatment of greater tuberosity fractures were encouraging. However, it seems to be di cult to x fractures that are smaller and closer to the articular surface.
The suture anchor is the other tool generally described for xation of greater tuberosity fractures in both open and arthroscopic surgeries. A biomechanical study suggested that suture anchor constructs would be stronger than xation constructs using screws for humeral greater tuberosity fractures8. The suture anchoring technique assisted by arthroscopy has achieved satisfactory results since its application.
Despite the bene t of reduced soft tissue damage and skin incision, there are higher requirements for doctors' arthroscopic techniques and a longer learning curve, so the development in primary hospitals is partially restricted. In addition, some studies believe that the arthroscopic technique has no advantage over open reduction in the treatment of simple humeral greater tuberosity avulsion fractures due to the longer operation time7. Using a small incision through the deltoid approach with suture anchors to treat this type of injury can effectively avoid the above problems. In addition, a biomechanical study concluded that there is no signi cant difference between double-row suture anchor xation and the suture-bridge technique in the treatment of greater tuberosity fractures8. Therefore, we consider that a familiar xation method can be selected according to the experience of the surgeon.
It is not clear whether the suture anchors or anatomical locking plates have better functional results. For this reason, we compared the clinical treatment results of the small incision suture anchoring technique through the deltoid muscle and the locking plate for the treatment of humeral greater tuberosity fractures. In our study, the blood loss in the suture anchor treatment group was less than that in the anatomical plate treatment group. In contrast, the average operation time of the anatomical plate treatment group was shorter than that of the suture anchor treatment group. There was no signi cant difference in operation time, postoperative VAS score, hospital stay or postoperative function score. The complication rate was 7.69% in the suture anchor treatment group and 11.11% in the locking plate treatment group. Considering the small sample size and differences in the study, the results should be interpreted cautiously. One patient in each group had fracture displacement after surgery, but the displacement was less than 3 mm. After conservative treatment, the joint function recovered satisfactorily. Although the two have similar functional results, we found that suture anchors have advantages in cases where the fracture is smaller and closer to the articular surface, and plate xation could provide rigid xation and cover a larger fracture area. One of the bene ts of suture xation is that it depends on the strength of the rotator cuff tendons themselves and not only on the bone quality. In older patients with osteopenic bone suture anchors, healing and stability could be helpful. In addition, the suture anchors do not require a second operation to remove them. Therefore, we consider that satisfactory functional results can be achieved with both xing methods. However, suture anchors seem to be more advantageous in cases where the fracture is smaller and closer to the articular surface, while the anatomical locking plate is more secure in cases where the fracture is larger. Our study compared the functional results of the two xed methods, but the study also has certain limitations. First, this is a nonrandom retrospective study, and its results may be biased. Second, the small number of cases in the study may also have a certain impact on the results.

Conclusion
Both suture anchors and anatomical plates can achieve satisfactory clinical results in the treatment of humeral greater tubercle fractures, and there is no signi cant difference in the long-term functional results between the two groups. These two xation devices can be exibly selected according to the surgeon's experience and fracture characteristics.

Declarations
Ethics approval and consent to participate: This study was approved by the medical ethics review board of Guyuan People's Hospital. Written informed consent was obtained from the parents/guardians of the patients.
Consent for publication: Not applicable.
Availability of data and materials: All data generated or analysed during this study are included in this published article.    performed. (C) The X-ray examination showed that the fracture had healed six months after the operation.