Favorable Outcomes After Arthroscopic Posterior Bankart Repair for Traumatic Posterior Shoulder Instability in Collision Athletes

Purpose

To investigate the postoperative results of arthroscopic posterior Bankart repair for traumatic posterior shoulder instability in collision sports athletes and their clinical characteristics, including injury mechanism, symptoms, physical examination findings, and imaging features.

Methods

Between January 2011 and April 2022, a retrospective review was conducted of collision-sport athletes who underwent arthroscopic posterior Bankart repair for traumatic posterior shoulder instability at a single institution. The inclusion criteria were posterior instability caused by trauma, absence of generalized joint laxity, and arthroscopic posterior Bankart repair. All the patients had a minimum follow-up of 24 months. Patient demographics, injury mechanisms, imaging findings (evaluated using radiographs, computed tomography, and magnetic resonance imaging, including posterior labral tears, posterior glenoid bone loss, glenoid retroversion, and reverse Hill-Sachs lesions), return-to-play rates, recurrence, and postoperative shoulder pain and Rowe score were evaluated. Pre- and postoperative Rowe scores were compared using the Wilcoxon signed-rank test. A P value of <.05 was considered statistically significant.

Results

Of 517 shoulders operated on for instability, 21 (4.1%) had posterior instability. After excluding 8 cases, 17 shoulders from collision sports athletes were analyzed. The mean age of the athletes was 21.1 years. All patients had a positive posterior apprehension test, and 62% had a positive anterior apprehension test. Imaging revealed posterior glenoid bony defects in 85% of cases. The mean glenoid retroversion angle was 1.6° ± 3.6° as measured on axial computed tomography images. The mean follow-up period was 40.5 ± 22.9 months, and all patients returned to their preinjury level of sports activity at a mean of 6.5 ± 1.0 months postoperatively. At the time of injury, 6 patients experienced a posterior dislocation, one experienced a subluxation, and 6 reported only posterior shoulder pain. Before surgery, 7 patients had no history of complete dislocation, while 6 patients had recurrent dislocations (≥2 times). The mean interval from the first dislocation or symptom onset to surgery was 21.8 ± 20.8 months. The Rowe score improved significantly from 55 (range: 25-75) preoperatively to 95 (range: 50-100) postoperatively ( P <.01). The minimum clinically important difference for the Rowe score was 13.4 points, and 92% of patients exceeded this threshold. However, one patient (8%) experienced redislocations, and 4 patients (31%) reported residual pain, with 3 of these cases involving glenoid cartilage lesions at the time of surgery. Although postoperative pain was more common in patients with cartilage lesions (60%) than in those without (13%), this difference did not reach statistical significance ( P =.217).

Conclusions

Arthroscopic posterior Bankart repair for traumatic posterior shoulder instability in collision sports athletes resulted in a low recurrence rate, high return-to-play rate, and clinically meaningful improvement. Although not statistically significant, residual postoperative pain tended to be more common in patients with glenoid cartilage lesions observed at the time of surgery.

Level of Evidence

Level IV, retrospective therapeutic case series.

Posterior shoulder instability is estimated to occur in approximately 5% of all cases of shoulder instability. ^, Reports from the United States indicate an incidence rate of 0.046 per 1000 person-year, compared with a greater incidence in studies within the U.S. military. ^, It also has been reported that posterior shoulder instability and anterior shoulder instability differ in traumatic injury mechanism and that posterior instability is more common in women compared with anterior instability.

Furthermore, the exact mechanism of posterior shoulder instability is often unclear, and in cases of high general laxity, minor triggers can lead to voluntary subluxation. ^, Therefore, especially in the acute phase, the diagnosis and treatment strategy depend on the injury mechanism and the assessment of general laxity. Thus, the pathophysiology of posterior instability differs from that of anterior instability.

Reports to date on surgical outcomes for posterior shoulder instability are not limited to athletes but include patients at a wide range of activity levels, and joint laxity often contributes to posterior instability, even in the absence of a clear traumatic event. ^,^, Similar to anterior instability, posterior shoulder instability caused by definite trauma is also common in collision sports athletes. Kim et al. reported on the surgical outcomes of arthroscopic posterior Bankart repair in patients with posterior shoulder instability, including 5 rugby players, one of whom retired after surgery. Nonetheless, clinical characteristics and treatment outcomes specific to collision athletes with traumatic posterior shoulder instability have not been thoroughly documented.

The purpose of this study was to investigate the postoperative results of arthroscopic posterior Bankart repair for traumatic posterior shoulder instability in collision sports athletes and their clinical characteristics, including injury mechanism, symptoms, physical examination findings, and imaging features. We hypothesized that arthroscopic posterior Bankart repair for traumatic posterior shoulder instability in collision sports athletes would result in favorable clinical outcomes with a low recurrence rate and high rate of return to sport.

Methods

A retrospective study was conducted on collision sport athletes with traumatic posterior shoulder instability who underwent arthroscopic posterior Bankart repair at a single institution between January 2011 and April 2022, with a follow-up period of at least 2 years. Collision sports are defined as contact sports involving high-energy, intentional physical impact between players, such as rugby, American football, and wrestling. The study was conducted in accordance with the Declaration of Helsinki, and ethical review board approval and informed consent were obtained from all patients and relevant individuals before the study began (Research Ethics Committee of St. Luke’s International Hospital, Ref: 17-R036).

The inclusion criteria were as follows: (1) age 16 years or older, (2) no history of injury to the same shoulder, (3) injury caused by a clear traumatic event, (4) findings of posterior instability on physical examination (positive posterior apprehension test), ^, and (5) cases with bony Bankart lesions or posterior glenoid rim erosion identified on preoperative computed tomography (CT), and clear posterior Bankart lesions confirmed on magnetic resonance imaging (MRI) or during arthroscopy, were included. Even in the absence of a clear traumatic dislocation as the initial event, patients with posterior shoulder pain and no clear history of dislocation or subluxation, but who showed posterior instability on physical examination and imaging findings (e.g., posterior apprehension test and posterior labral tear), were diagnosed with symptomatic posterior instability and underwent surgical repair. The exclusion criteria were as follows: (1) postoperative follow-up less than 2 years, (2) incomplete medical information, (3) high general laxity (Beighton score >4), (4) cases in which concurrent anterior Bankart repair was performed, (5) patients presenting with anterior instability on physical examination and exhibiting a definitive anterior Bankart lesion requiring repair during surgery, as these findings were indicative of multidirectional instability, and (6) lack of interest to participate in the study.

Evaluation Items

Age at surgery, sex, side of injury, sports, competition level, and injury mechanism were considered evaluation items. One experienced shoulder surgeon (A.T.) examined the preoperative physical findings, and posterior apprehension test, anterior apprehension test, and muscle strength (manual muscle) test were conducted. In the anterior and posterior apprehension tests, a positive finding was defined as either a reproducible sensation of apprehension (dislocation fear) or pain provoked by applied stress. We also investigated radiographs (anterior-posterior and axial views), 3-dimensional CT images, and noncontrast 3-Tesla MRIs. All the imaging evaluations were performed by a musculoskeletal radiologist (T.N.). On radiographs, the presence or absence of a bony Bankart lesion was assessed ( Fig 1 A). CT scans were used to evaluate glenoid bone defects, and the percentage of bone loss was measured as discussed by Sugaya ( Fig 1 B). The presence of Hill-Sachs lesions or reverse Hill-Sachs lesions also was assessed, and glenoid retroversion was measured. Glenoid version was measured using the glenoid vault method as discussed by Friedman. MRI T2∗-weighted axial images were used to evaluate labral pathology, including complete anterior-to-posterior detachment and incomplete tears such as Kim’s lesion. ^,

Image, AltText currently not available — Fig 1

Surgical Procedure

All surgeries were performed by one experienced shoulder surgeon (A.T.) ( Fig 2 ). The patient was placed in the lateral decubitus position with lateral traction of the affected arm at 40° abduction and 10° anterior flexion. Anterior, anterosuperior, and posterior portals were used; we observed from the posterior and anterosuperior portals. The damaged posterior capsulolabrum complex and bony Bankart lesion were repaired using a suture anchor. The suture method was simple suture or double-row repair. ^, Rotator interval closure was performed in cases in which the surgeon identified significant attenuation or widening of the rotator interval. If a cartilage injury (glenoid labrum articular disruption [GLAD]) was present, a microfracture was performed depending on the defect size.

Surgical findings were evaluated the number of suture anchors and repair method (simple suture or double-row repair) and the clinical outcome (Rowe score) before and after surgery. The primary outcome variable was the change in the Rowe score from preoperative to postoperative evaluation.

Rehabilitation comprised a shoulder brace at 30° abduction and neutral rotation for the first 3 weeks postoperatively, followed by passive range of motion exercise. Patients began running at 8 weeks postoperatively, upper extremity strength training on the affected side at 10 weeks, skill exercises and weight training at 12 weeks, return to noncontact practice at 3 months postoperatively, and return to contact practice at 5 months postoperatively.

Statistical Analysis

The minimum clinically important difference (MCID) for the Rowe score was calculated as 13.4 points, on the basis of one-half standard deviation of the change from preoperative to postoperative scores. Pre- and postoperative Rowe scores were compared using the Wilcoxon signed-rank sum test. The Fisher exact test was used to evaluate the association between intraoperative glenoid cartilage lesions and postoperative residual shoulder pain due to the small sample size. The significance level was set at P <.05. Statistical analysis was performed using the R software (Foundation for Statistical Computing).

Results

Between January 2011 and April 2022, 517 shoulders were operated on for shoulder instability. Twenty-one shoulders (4.1%) of 20 collision sports athletes had traumatic posterior instability. One patient had bilateral injuries. After we excluded 1 patient who underwent concurrent rotator cuff repair, 1 with generalized joint laxity, 2 who were lost to follow-up due to relocation, and 4 patients who underwent concurrent anterior Bankart repair, 13 shoulders were included in the final analysis.

Patient Demographics

All patients were male. The mean age at the time of surgery was 21.2 ± 4.0 years. The mean follow-up period was 40.5 ± 22.9 months, and the mean interval from injury to surgery was 21.8 ± 20.8 months ( Table 1 ). The patient sample included 7 American football players, 5 rugby players and 1 martial arts athlete. All patients had a clear history of trauma during competition. Seven patients were injured by falling and hitting the ground with the shoulder, 3 patients were injured by forced horizontal extension, 2 patients were tackled and pushed backward, and 1 patient by forced horizontal flexion. At the time of injury, 6 patients experienced a posterior dislocation, 1 experienced a posterior subluxation, and 6 reported only posterior shoulder pain without overt instability. Three patients (23%) had experienced 2 to 5 dislocations, whereas another 3 patients (23%) had more than 5 dislocations. On physical examination, the posterior apprehension test was positive in all cases, with 7 cases of apprehension of dislocation and 6 cases of pain. The anterior apprehension test was positive in 8 patients (62%).

Table 1

Patient Demographics

Age, yr, mean ± SD (range)	21.2 ± 4.0 (18-32)
Mean follow-up, mo, mean ± SD (range)	40.5 ± 22.9 (24-83)
Duration from onset to surgery, mo, mean ± SD	21.8 ± 20.8
Time to return to sport, mo, mean ± SD	6.5 ± 1.0
Side, n (%)
Right	5 (38%)
Left	8 (62%)
Sex, n (%)
Male	13 (100%)
Female	0 (0%)
Competition level, n (%)
Professional	2 (15%)
College	10 (77%)
High school	1 (8%)
Sport, n (%)
American football	7 (54%)
Rugby	5 (38%)
Material arts	1 (8%)
Injury mechanism, n (%)
Anterior shoulder impact	7 (54%)
Horizontal extension	3 (23%)
Posterior axial load during flexion	2 (15%)
Horizontal flexion	1 (8%)
Symptoms at onset, n (%)
Dislocation	6 (46%)
Subluxation	1 (8%)
Pain	6 (46%)
Dislocation times, n (%)
0 times	7 (54%)
1 times	0 (0%)
2-5 times	3 (23%)
>5 times	3 (23%)
Posterior apprehension test, n (%)
Apprehension	7 (54%)
Pain	6 (46%)
Anterior apprehension test, n (%)
Negative	5 (38%)
Positive	8 (62%)

SD, standard deviation.

Radiography showed bony Bankart lesions in 8 patients ( Table 2 ), and CT showed bony defects in 11 patients (85%), with a median defect rate of 11.5% ± 7.5%. Bony Bankart lesions were seen in 11 patients (8585%), and Hill-Sachs lesions and reverse Hill-Sachs lesions were seen in 2 and 9 patients, respectively. The mean glenoid retroversion angle was 1.6° ± 3.6° as measured on axial CT images. Posterior complete labral detachment was identified in 9 patients (69%) on the basis of MRI.

Table 2

Imaging Findings

Radiography
Posterior bony Bankart lesion, n (%)	8 (62%)
CT
Glenoid bone loss
Positive, n (%)	11 (85%)
Defect rate (%): mean ± SD	11.5 ± 7.5
Reverse Hill-Sachs lesion
Positive, n (%)	9 (69%)
Hill-Sachs lesion
Positive, n (%)	2 (15%)
Glenoid retroversion (°): mean ± SD	1.6 ± 3.6
MRI, n (%)
Posterior labral complete detachment	9 (69%)
Posterior incomplete labral tear	4 (31%)
Partial detachment or irregular labral contour	3 (23%)
Reduced labral height	3 (23%)
Continuity between the labrum and glenoid is maintained	2 (15%)