Additional Scapular Spine Bone Grafting Combined with Bankart and Remplissage for Off‐Track Hill–Sachs Lesions with Subcritical Glenoid Bone Loss: Provides Better Stability

Abstract

Purpose

Arthroscopic Bankart repair combined with remplissage and autologous scapular spine bone grafting have been described as a treatment for off‐track Hill–Sachs lesions with subcritical glenoid bone defects in the anterior shoulder instability. However, whether these two techniques can achieve satisfactory postoperative outcomes is unclear, and there are few comparative studies between them. Therefore, this study compared the postoperative efficacy of the two techniques for off‐track Hill–Sachs lesions with subcritical glenoid bone loss.

Method

Between June 2017 and December 2020, 62 patients with shoulder instability due to Off‐Track Hill–Sachs lesions with subcritical glenoid bone loss underwent surgical treatment and were included in this regression study. Thirty‐two patients underwent arthroscopic Bankart repair combined with remplissage (B + R group), and 30 patients underwent additional autologous scapular glenoid bone grafting (additional bone grafting group). The general information of the patients was recorded. The patient's activity before and after surgery was recorded. The DASH score and Constant–Murley (CM) score were used to assess the patient's functional status; the Rowe score was used to evaluate the patient's shoulder stability. The shoulder function and stability before and after surgery were analyzed and compared between the two groups.

Results

The final DASH scores of the B + R group and the additional bone grafting group were significantly lower than those before surgery, with a statistically significant difference (9.76 ± 4.32 vs. 27.89 ± 6.63, 8.50 ± 3.32 vs. 28.0 ± 4.27, p = 0.000); the final CM scores of the two groups were significantly higher than those before surgery (88.71 ± 3.74 vs. 73.68 ± 3.74, 87.16 ± 2.29 vs. 71.37 ± 2.68, p = 0.000). There was no statistical difference in the final DASH score and final CM score between the two groups (p > 0.05). In terms of postoperative stability, the final Rowe scores of the two groups were significantly higher than those before surgery, with a statistically significant difference (89.06 ± 9.19 vs. 41.71 ± 4.13; 93.16 ± 4.99 vs. 42.33 ± 2.53, p = 0.000). Compared with the control group, the additional bone graft group achieved higher final Rowe scores (93.16 ± 4.99 vs. 89.06 ± 9.19, p = 0.032).

Conclusion

For patients with anterior shoulder instability due to off‐track Hill–Sachs lesions with subcritical glenoid bone loss, although Bankart and remplissage can achieve satisfactory clinical results, additional autogenous scapular spine bone grafting can provide better stability of the shoulder, especially for patients with high sports demands.

Technical diagram of additional scalar spine bone grading.

Anterior shoulder instability is a common disease in sports medicine. Over 90% of patients who suffer from recurrent anterior dislocation of the shoulder possess bone defects, including the humeral head and glenoid, which represent significant contributions to the recurrent dislocation. ¹ , ² , ³ , ⁴ As research continues to deepen, newly introduced concepts include “engaging/nonengaging Hill–Sachs lesion,” ⁵ “glenoid track,” ⁶ and “on/off track,” ⁷ which are used to describe and evaluate the connection between Hill–Sachs lesion and glenoid bone defect, and furthermore, the stability of the shoulder joint. Currently, the primary treatment strategies for anterior instability of the shoulder joint resulting from bipolar bone defects involve soft tissue surgery (such as Bankart repair and remplissage) and bony surgery (Latarjet and iliac bone grafting). In 2014, Di Giacomo et al. ⁷ classified bipolar bone defects into four categories, recommending arthroscopic Bankart and remplissage repair for patients with shoulder glenoid bone defects less than 25% and Hill–Sachs lesions that are off‐track. When the glenoid bone defect is ≥25%, bone surgery is recommended. However, recent studies have challenged the concept of a critical threshold for glenoid bone loss. ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² The current systematic review indicates that if the loss of glenoid bone exceeds 10%, the recurrence rate of dislocation will significantly increase; for subcritical glenoid bone loss, the recurrence rate of dislocation after soft tissue repair under arthroscopy is higher than that of bony surgery, and the complication rate of bony surgery is much higher than that of Bankart repair. ¹³

Clinically, it is relatively common to find patients with 10%–15% glenoid bone defects accompanied by off‐track Hill–Sachs lesions. Bankart and remplissage surgery is currently the most frequently utilized surgical approach. We recently implemented an autologous scapular spine bone grafting procedure as a treatment for 10%–15% subcritical glenoid bone defects associated with anterior instability, ¹⁴ which effectively improved shoulder joint stability and reduced the recurrence rate of shoulder dislocation. However, there has been no comparative study of these two techniques. The aim of this study is: (1) to compare the postoperative outcomes of patients with subcritical (10%–15%) glenoid bone loss and off‐track Hill–Sachs lesion who underwent treatment with these two techniques for shoulder instability and (2) to clarify whether additional autologous scapular glenoid bone grafting can provide better stability.

Methods

Patient Selection

The inclusion criteria for this study were: (1) recurrent anterior glenohumeral instability; (2) a follow‐up of at least 1 year after surgery; (3) preoperative three‐dimensional (3D) computed tomography (CT) or magnetic resonance imaging (MRI); (4) 10%–15% glenoid bone loss; and (5) off‐track Hill–Sachs lesion in glenoid track measurement. The exclusion criteria for this study included: (1) patients with multidirectional instability; (2) patients with incomplete clinical records, such as surgical and clinical records, or patients who were unable to attend follow‐up; and (3) patients with rotator cuff tears, biceps tendon injuries, or humeral avulsion of glenohumeral ligaments (HAGL) injuries.

General Information

From June 2017 to December 2020, a total of 217 patients with recurrent anterior shoulder dislocation were diagnosed and treated in our hospital, and 62 patients who met the above criteria were included for analysis. Thirty‐two patients underwent arthroscopic Bankart repair combined with remplissage, including 21 males and 11 females; the age ranged from 18 to 54 years, with an average of 31.41 years; the number of dislocations ranged from two to 40, with an average of 12.34; the follow‐up duration ranged from 13 to 63 months, with an average of 24.93 months; the average glenoid bone defect was 12.62%. Of the 30 patients who received additional treatment with autologous scapular spine bone grafting, 25 were male and five were female. The age range was 19 to 53 years, with an average of 29.17 years. The range of dislocation numbers is six to 50, and the average is 14.5. The follow‐up duration varied from 12 to 60 months, and the average duration was 28.6 months. The average glenoid bone defect is 13.24%. The baseline of general information in both groups was consistent (Table 1). This study was approved by the Ethics Committee of Sichuan Orthopaedic Hospital (KY2024‐029‐01).

Table 1.

General information comparison.

Baseline information	B + R group (n = 32)	Additional bone grafting group (n = 30)	T‐value	p‐value
Age (years)	31.41 ± 12.07	29.17 ± 10.48	0.777	0.440
Gender				0.096
Male	21 (65.6)	25 (83.3)
Female	11 (34.4)	5 (16.7)
Number of dislocations (times)	12.34 ± 9.39	14.5 ± 8.43	−0.949	0.346
Glenoid bone defect (%)	12.62 ± 1.61	13.24 ± 1.63	−1.512	0.136
Follow‐up (months)	24.93 ± 11.41	28.66 ± 13.48	−1.178	0.243

Data Collection

We collected data on patient mobility (including forward flexion, external rotation, and internal rotation) before and after surgery, using the Disability of Arm Shoulder and Hand (DASH) score and Constant–Murley (CM) score to evaluate patient function; using the Rowe score to assess shoulder joint stability; and collected preoperative patient 3D CT and/or MRI data. We recorded the general information of the patient, the number of dislocations, and any complications after surgery. Postoperative instability is defined as any dislocation or subluxation of the shoulder at any time after surgery; dislocation refers to the patient experiencing a “popping” or “dislocation” of the shoulder joint, requiring manual or self‐reduction; subluxation refers to the sensation of sliding or displacement of the shoulder joint.

Measurement of the Glenoid Bone Defect

The method of Sugaya et al. ¹⁵ was used to measure the defect of the glenoid. 3D CT reconstruction of the glenoid, assuming that the lower part of the pear‐shaped glenoid is approximately circular in shape, constructs a best‐fit circle in the en‐face view. The diameter of the glenoid is determined by taking the overall diameter of the circle, and the width of the glenoid bone defect is expressed as a percentage of this diameter, representing the percentage of glenoid bone defect (Figure 1).

FIGURE 1.

Measurement of percentage of glenoid defect. Three‐dimensional CT was used to reconstruct the glenoid, and the best fit circle was constructed in the en‐face view. The width (b) of glenoid defect was measured as the percentage of the total circular diameter (A).

Measurement of Hill–Sachs Interval (HSI)

In the posterior image of the humeral head 3D‐CT, assess the size of the Hill–Sachs lesion. The HSI refers to the distance between the insertion point of the rotator cuff and the innermost edge of the Hill–Sachs lesion. In the 3D‐CT image, draw a straight line along the medial edge of the rotator cuff insertion point, and a parallel line along the innermost edge of the Hill–Sachs lesion. The HSI is the distance between the two lines (Figure 2).

FIGURE 2.

Measurement of Hill–Sachs interval. Select the posterior image of humeral head 3D‐CT, make a straight line along the medial edge of rotator cuff insertion, make a parallel line along the innermost edge of Hill–Sachs lesions. The distance between the two lines is the Hill‐Sachs interval.

Evaluation of On‐Track/Off‐Track Hill‐Sachs Lesions

Referring to the calculation method of Di Giacomo et al., ⁷ we first measured the diameter of the glenoid bone, bone defect (Figure 1), and HSI (Figure 2), and then calculated the glenoid track (GT). GT calculation formula: 83% of the diameter of the glenoid bone minus the width of the glenoid bone defect, expressed as GT = 0.83A‐b. Compare the size of GT and HSI. If HSI > GT, it is an off‐track Hill–Sachs lesion; if HSI < GT, it is on‐track Hill–Sachs lesion.

Surgical Techniques

After general anesthesia and brachial plexus block anesthesia, take a lateral lying position, with the affected shoulder joint abduction of 40° and moderate forward flexion traction, and perform routine disinfection and drape. Routinely establish posterior, anterior, and anterolateral portals, re‐evaluate the extent of the glenoid bone defect and Hill–Sachs defect under arthroscopy, and confirm that there are no rotator cuff and biceps tendon injuries and no HAGL (humeral avulsion of glenohumeral ligament) injury. After exposing the bone defect on the upper posterior part of the humeral head, two anchors with wires were placed on the freshened bone surface, with the wires evenly distributed through the subscapularis, preparing for remplissage.

Turn to the anterosuperior portal, use a liberator knife to strip and release the retracted and adherent inferior glenohumeral ligament complex (IGHLC) from the glenoid neck until the 6 o'clock position of the glenoid, and use the shaver to remove the cortical bone of the glenoid rim. Insert three anchors at approximately 5 o'clock, 4:30, and 3 o'clock on the shoulder glenoid. Suture the IGHLC, perform Bankart repair. Finally turn to the subacromial space, and the double pulley technique is used to tie off the sutures, securing the subscapularis tendon to the Hill–Sachs bone defect, thereby completing the remplissage procedure.

For patients who need additional bone graft, we also place three anchors at the 5 o'clock, 4:30, and 3 o'clock positions of the glenoid. The anchor at the 5 o'clock position is called a “labral anchor,” while the anchors at the 4:30 and 3 o'clock positions are called “graft‐anchors.” At the midpoint of the medial and lateral edges of the spine of the scapula, make a 4 cm incision parallel to the spine of the scapula, and carefully separate the subcutaneous tissue and fascia layer by layer to expose the spine of the scapula. Depending on the size of the defect, take a bone graft of about 20 mm × 10 mm × 8 mm, drill two 1.5 mm tunnels in the center of the bone graft, and insert two “graft‐anchors” suture line through the tunnel. The bone graft is transported through the cannula. Use the suture lines that pass through the bone graft and the remaining suture lines to fix the IGHLC with a suture hook, to make the IGHLC cover the bone graft. Adjust the position of the bone graft so that it is flush with the glenoid neck; insert a liberator knife from the posterior portal to hold the bone graft in place to prevent displacement and twisting during knotting. Finally, tie the knot to secure the bone graft and perform the Bankart repair ¹⁶ (Figures 3 and 4).

FIGURE 3.

(A) The “labral anchor” was inserted at 5 o'clock and two sutures were passed through the inferior glenohumeral ligament complex (IGHLC) in preparation for subsequent Bankart repair. (B) The “graft‐anchors” were inserted at 4:30 and 3 o'clock. And the bone graft was obtained at the scapular spine. (C) One suture of each of the two “graft‐anchors” was correspondingly shuttled through the bone tunnels separately. And a cannula was used for transporting the bone graft. (D) The bone graft was covered by the IGHLC and integrated with the anterior glenoid rim. Two “graft‐anchors” sutures which were shuttled through the bone tunnels were then passed through the IGHLC. (E) After the IGHLC was pretensioned, sutures of the two “graft‐anchors” were tied up to fix the bone graft and repair the IGHLC simultaneously. Additional sutures were then tied up to accomplish the Bankart repair and reinforce the fixation of the bone graft. Reproduced with permission from Dai F et al. ¹⁶

FIGURE 4.

(A) Scapular spine bone graft harvest. (B‐C) Release the IGHLC and insert one “labral anchor” (blue arrows), two “graft‐anchors” (red and black arrow), and two sutures (yellow arrows) were passed through the IGHLC in preparation for subsequent Bankart repair. (D‐F) After implanting bone grafts, two “graft‐anchors” sutures (white arrows) which were shuttled through the bone tunnels were then passed through the IGHLC. (G‐I) Tying up the two “graft‐anchors” to fix the bone graft and repair the IGHLC simultaneously. Then complete Bankart repair and reinforce the fixation of the bone graft. Reproduced with permission from Dai F et al. ¹⁶

Postoperative Rehabilitation

Following surgery, the shoulder was stabilized using an abduction brace for a period of 6 weeks, followed by passive movement of the shoulder joint commencing after 6 weeks. Strength training began 10–12 weeks after surgery, while physical activity was introduced 6 months later.

Statistical Analysis

Statistical analysis was performed using SPSS 22.0 (IBM, USA). Quantitative data (age, number of dislocations, shoulder glenoid defects, follow‐up time, shoulder range of motion, DASH score, CM score, Rowe score) were expressed as x ± s. When the data distribution was normal, independent t‐tests were used for intergroup comparisons, and count data were expressed as frequency (%) and chi‐square tests were used for intergroup comparisons. Paired t‐test was used to compare the preoperative and postoperative differences. Significance was assumed for p values <0.05.

Results

Functional Outcome

The final DASH score in the Bankart repair combined with remplissage group (B + R group) was significantly lower than that before surgery, and the difference was statistically significant (9.76 ± 4.32 vs. 27.89 ± 6.63, p = 0.000). The final DASH score in the additional bone grafting group was significantly lower than that before surgery, and the difference was statistically significant (8.50 ± 3.32 vs. 28.0 ± 4.27, p < 0.000). Although the final DASH score in the B + R group was higher than that in the additional bone grafting group, the difference was not statistically significant (9.76 ± 4.32 vs. 8.50 ± 3.32, p = 0.2) (Table 2).

Table 2.

Comparison of preoperative and final follow‐up functional and stability scores in the B + R group.

Outcomes score	Preoperative	Final follow‐up	p‐value
DASH score	27.89 ± 6.63	9.76 ± 4.32	0.000*
CM score	73.68 ± 3.74	88.71 ± 3.74	0.000*
Rowe score	41.71 ± 4.13	89.06 ± 9.19	0.000*

^*Significant statistical difference.

The final CM score in the B + R group was significantly higher than that before surgery (88.71 ± 3.74 vs. 73.68 ± 3.74, p = 0.000). The final CM score in the additional bone graft group was significantly higher than that before surgery, with a statistically significant difference (87.16 ± 2.29 vs. 71.37 ± 2.68, p = 0.000). The final CM score in the B + R group was higher than that in the additional bone graft group, but the difference was not statistically significant (88.71 ± 3.74 vs. 87.16 ± 2.29, p = 0.053) (Table 3).

Table 3.

Comparison of preoperative and final follow‐up functional and stability scores in the additional bone grafting group.

Outcomes score	Preoperative	Final follow‐up	p‐value
DASH score	28.0 ± 4.27	8.50 ± 3.32	0.000*
CM score	71.37 ± 2.68	87.16 ± 2.29	0.000*
Rowe score	42.33 ± 2.53	93.16 ± 4.99	0.000*

^*Significant statistical difference.

Stability Outcome

In terms of postoperative stability, the final Rowe scores in the B + R group and the additional bone graft group were significantly higher than those before surgery, with a statistically significant difference (89.06 ± 9.19 vs. 41.71 ± 4.13; 93.16 ± 4.99 vs. 42.33 ± 2.53, p = 0.000). Compared to the B + R group, the additional bone graft group achieved a higher final Rowe score (93.16 ± 4.99 vs. 89.06 ± 9.19), with a statistically significant difference (p = 0.032) (Table 4).

Table 4.

Comparison of preoperative and final follow‐up functional and stability scores between the two groups.

Outcomes score	B + R group	Additional bone grafting group	T‐value	p‐value
Forward flexion (°)	175.31 ± 5.67	176.50 ± 6.03	−0.799	0.428
External rotation (°)	53.28 ± 6.42	54.16 ± 9.29	−0.434	0.666
DASH score	9.76 ± 4.32	8.50 ± 3.32	1.297	0.200
CM score	88.71 ± 3.74	87.16 ± 2.29	1.983	0.053
Rowe score	89.06 ± 9.19	93.16 ± 4.99	−2.202	0.032*

^*Significant statistical difference.

Healing and Complications

Three months after surgery, CT scans of the additional bone graft group showed that the bone grafts were all well healed, with no cases of bone graft displacement or non‐union.

Although neither group experienced re‐dislocation after surgery, two patients in the B + R group did experience subluxation, and six patients still had positive Crank tests at the last follow‐up. All these patients underwent proprioceptive function training and their symptoms improved. These patients were able to complete daily tasks, but physical demands were high. The group that underwent additional bone graft had no positive Crank test residuals, and they were all able to complete their daily tasks and participate in recreational sports (Figure 5).

FIGURE 5.

Typical case. (A) and (B) indicate that the patient had a severe glenoid bone defect and Hill–Sachs lesion, which was assessed as off‐track. (C) and (D) are CT images after the surgery of additional scapular spine bone grafting combined with Bankart and remplissage. (E) and (F) represent the CT and MRI images at 14 months postoperatively, showing that the bone graft has healed and is in excellent shape. (G)‐(I) showed the function of the patient at 14 months after operation, with excellent function and stability, and the Crank test (−).

Discussion

Comparison of Functional and Stability Scores

This is the first study to compare the postoperative outcomes of arthroscopic Bankart repair combined with remplissage and additional autogenous scapular spine bone grafting for the treatment of off‐track Hill–Sachs lesion with subcritical glenoid bone loss in shoulder instability. Our research indicates that, in the general population, both the routine Bankart with remplissage and the additional autogenous scapular spine grafting can be utilized to treat off‐track Hill–Sachs lesions associated with subcritical glenoid bone defects, resulting in satisfactory outcomes. Regardless of conventional soft tissue repair or additional bone grafting, the functional scores at the last follow‐up after surgery were better than those before surgery. Although there was no significant difference in functional scores between the two groups, additional bone grafting resulted in higher stability scores, especially in patients with high physical activity requirements.

Subcritical Glenoid Bone Loss

It has been determined that 25% of the glenoid bone defect is considered the critical point for both soft tissue repair and bony repair. However, recent studies have challenged this concept as research has deepened. Shin et al. ⁹ first challenged this concept from the perspective of biomechanics through a cadaveric model study, which showed that 15% or more of the glenoid defect should be considered as the critical bone loss that cannot be restored by soft tissue repair to restore glenohumeral stability. Later, in their case–control study with 169 participants, it was discovered that a loss of 17.3% of the articular glenoid bone was the critical value for deciding between soft tissue surgery and bony surgery. When the bone defect exceeded 17.3%, the failure rate of Bankart surgery alone was as high as 42.9%. However, when the bone defect was less than 17.3%, the failure rate was only 3.7%. ¹⁰ However, the study by Shaha et al. ¹⁷ set 13.5% as the critical value for glenoid bone defects, and patients with bone loss >13.5% had significantly lower SANE and WOSI scores than patients with bone defects <13.5%. A recent matched cohort analysis by Dekker et al. ¹¹ found that in individuals with high physical demands, when the GBL ≥15%, the risk of re‐dislocation after arthroscopic Bankart repair alone significantly increases.

There has been controversy regarding the treatment of shoulder instability with subcritical bone defects in the glenoid. The Latarjet procedure is a classic orthopaedic surgery, and although there are a large number of literature reports indicating its reliable efficacy and low recurrence rate, ¹⁸ , ¹⁹ more and more recent literature reports have shown that the Latarjet procedure has surgical risks and complications such as neurovascular injury, shoulder stiffness, coronoid process fractures, and postoperative osteoarthritis. ²⁰ , ²¹ Yang et al. ²² compared the postoperative outcomes of Bankart repair combined with remplissage and modified Latarjet in patients with subcritical glenoid bone loss and shoulder instability, and found that both surgical procedures can achieve satisfactory results in the initial surgery for the general population, but the complication rate is higher in the Latarjet group. However, for revision surgery, collision and contact athletes, and patients with glenoid bone loss >10%, the re‐dislocation rate of Latarjet is lower. Min et al. ²³ compared the postoperative outcomes of Bankart and Latarjet in active‐duty soldiers with sub‐critical glenoid defects (13.5% to 24%), and found that patients treated with open Latarjet had significantly higher SANE and WOSI scores than those treated with arthroscopic Bankart repair, but there was no significant difference in the rate of re‐dislocation between the two groups. Autologous or allogeneic iliac bone grafting is also a common orthopaedic surgical procedure used to treat shoulder instability with glenoid bone defects, but autologous iliac bone has potential complications such as persistent postoperative pain, iliac fracture, hematoma at the bone harvesting site, nerve injury, and gait abnormalities, ²⁴ , ²⁵ while allogeneic iliac bone has the disadvantages of high graft absorption and rejection rates. For patients with shoulder instability due to 10%–15% glenoid defect, our team proposed a method of autologous scapular spine bone grafting combined with Bankart repair, which has the advantages of convenient bone harvesting, safe operation, and few complications, and has achieved good clinical efficacy. ¹⁴

Bipolar Bone Defects

Bipolar bone defects in the humeral head and glenoid are significant contributors to recurrent anterior dislocation of the shoulder. In most cases of shoulder instability, the initial symptom is a Hill–Sachs lesion, which is subsequently followed by a glenoid defect. Repeated dislocation may result in an off‐track Hill–Sachs lesion. ⁴ Remplissage is the most frequently employed method for treating Hill‐Sachs lesions. In fact, when there is an off‐track Hill‐Sachs lesion, remplissage is the preferred surgical approach regardless of the size of the glenoid bone defect. ²⁶ , ²⁷ , ²⁸ Biomechanical evidence indicates that remplissage can attain stable results in bipolar lesions, with up to 30% Hill–Sachs bone defects and up to 17% glenoid defects. ²⁹ Multiple studies have demonstrated that additional remplissage can also yield more satisfactory clinical outcomes in clinical practice. ³⁰ , ³¹ , ³² MacDonald et al. ³² compared the clinical outcomes of Bankart repair alone and Bankart repair combined with remplissage in the treatment of off‐track Hill–Sach lesions and shoulder instability with less than 15% glenoid bone loss. They discovered that the recurrence rate was 18% in the Bankart group and 4% in the Bankart with remplissage group at 24 months. Our study is similar to the study in terms of inclusion criteria, both involving patients with off‐track Hill–Sach lesions and shoulder instability with less than 15% glenoid bone loss. However, our surgical protocol involves the addition of autogenous scapular spine bone grafting on top of Bankart with remplissage, and the results of the study also indicate that although soft tissue repair with Bankart with remplissage can achieve satisfactory postoperative outcomes, additional bone grafting theoretically increases the bone contact area of the glenoid, resulting in better shoulder stability.

Strengths and Limitations

To our knowledge, this is the first study to compare the treatment of shoulder instability between arthroscopic Bankart repair combined with remplissage and additional autogenous scapular spine bone grafting. The research results provide more options and references for personalized treatment of shoulder instability. And regarding limitations, first and foremost, this study is a retrospective study with a small sample size, so inevitable statistical bias may affect the final results. Second, the follow‐up time is relatively short, and some potential complications may be overlooked, requiring further demonstration with long‐term results. Third, the shoulder glenoid bone defect is limited to 10%–15%, and there is a lack of comparison with patients with defects exceeding 15% or even more. We hope to clarify the scope of glenoid bone defects that can be treated with additional scapular spine bone grafting through further comparative studies in the future.

Conclusion

Both arthroscopic Bankart repair combined with remplissage and additional autogenous scapular spine bone grafting treatments for patients with anterior shoulder instability due to off‐track Hill–Sachs lesions with subcritical glenoid bone loss can achieve good functional outcomes, but additional autogenous scapular spine bone grafting can provide better stability, especially for patients with a high demand for exercise. Additional autogenous scapular spine bone grafting is a better treatment option for this type of patient.

Conflict of Interest Statement

The authors declare no conflict of interest.

Author Contributions

Methodology, Ming Xiang and Fei Dai; Statistic analysis, Fei Dai and Jinsong Yang; Data collection, Qing Zhang and Yiping Li; Writing–Original Draft Preparation, Fei Dai; Writing–Review and Editing, Ming Xiang; All authors have read and agreed to the published version of the manuscript.