Arthroscopic repair is sufficient for treating recurrent shoulder instability in patients with bipolar bone defects and minor glenoid bone loss

1 Introduction

Bipolar bone loss is defined as a combined bony defect of both the humeral head and glenoid and is thought to occur in up to 84% of patients with recurrent anterior shoulder instability. However, in view of limited literature examining bipolar bone loss, the true incidence of such lesions is not known and bipolar bone loss is not routinely quantified on pre-operative imaging or arthroscopy. First line treatment of anterior shoulder instability is often an arthroscopic Bankart repair, which only addresses the labrum and does not address the glenohumeral bone loss.^1–6

However, a growing body of evidence emphasizes the importance of early detection and management of such glenohumeral bone defects in the management of recurrent shoulder disability.^1,7–12 Anteroinferior osseous glenoid lesions (bony Bankart lesions) and posterolateral humeral head impression fractures (Hill-Sachs lesions) are recognized risk factors for recurrence of shoulder instability and failure of soft tissue repair, with up to 67% of patients experiencing recurrence of symptoms following soft tissue repair in the presence of either a bony Bankart or Hill-Sachs lesion.^{1,3,7,8,11–13}

If bone loss of either the humerus or the glenoid increase risk of recurrence, then both articular surfaces should be considered when predicting the risk of recurrent dislocation.^3,4,10 To better illustrate the arc of the glenohumeral joint, the glenoid concept track was developed to predict which patients with bipolar bone loss would have engaging lesions.^3,10,14 As defined by Yamamoto et al. the glenoid track is defined as the contact zone of the glenoid with the humeral head. Hill-Sachs lesions that remain in contact with the glenoid do not engage and are “on-track” lesions, while lesions that are out of the glenoid track have a high risk of dislocation and are “off-track” lesions.^2,10,12

While classically, the threshold for addressing glenoid bone loss (GBL) with surgery was 20–25%, newer clinical studies indicate that GBL as low as 13.5% is correlated with worse clinical outcomes for patients with bipolar bone loss after Bankart repair.^{1,3,4,8–10,15–17} Although treatment algorithms for bipolar bone loss have been proposed with GBL thresholds for recommending bony reconstructive procedures ranging from 15 to 25%, none have been validated with clinical studies.^1,3,7–9 Further research into the specific relationship between the specific amount of bone loss, the presence of an “on-track” or “off-track” Hill-Sachs lesion, and post-operative clinical outcomes is therefore required to better guide clinical decision making.

We aimed to compare the post-operative clinical outcomes of patients with recurrent shoulder instability and minor glenoid bone loss, defined as GBL <15%, treated with arthroscopic soft tissue repair between patients with bipolar bone loss, monopolar bone loss, and no bone loss. We hypothesize that shoulder instability patients with bipolar bone lesions and minor glenoid bone loss can be treated adequately with arthroscopic soft tissue repair only and without the use of bony augmentation procedures.

2 Materials and methods

2.3

2.3 Radiological parameters

All pre-operative MRIs were evaluated by an orthopaedic surgeon for the presence of a Bankart lesion, Hill-Sachs lesion, tears of the superior labrum from anterior to posterior (SLAP), chondrolabral lesions, biceps tendinosis, and bone loss. Humeral and glenoid bone loss were quantified as per the methods described by Di Giacomo et al. which are further outlined below .¹⁰ For the humerus, the width, length, and depth of the Hill-Sachs lesion was recorded, while for the glenoid, the height and width of the glenoid was measure and a best-fit circle was fit on onto the glenoid, with loss estimated as a percentage of the best-fit circle, as shown in Fig. 1.

Close

Fig. 1 Estimating glenoid bone loss. Using the sagittal cut of the MRI shoulder, a best-fit circle was measured to fit the glenoid shape in order to estimate glenoid bone loss. The glenoid track was measured as 83% of the diameter of the glenoid minus d, the width of the glenoid bone loss, as described by Di Giacomo et al.

Export to PPT

3 Results

A total of 98 patients fulfilled the selection criteria and were included for analysis in this study. Baseline demographic information is outlined in Table 2. As described above, the radiological parameters were evaluated for all pre-operative MRIs and are described in full in Table 3.

All patients who met the inclusion criteria were confirmed to have only undergone an arthroscopic Bankart repair with or without capsular shift or plication. No patients were found to have undergone bony reconstructive surgeries such as a Latarjet-Bristow procedure. Patients with no bone loss were found to be statistically younger than patients with monopolar or bipolar bone loss (p = 0.01). Of these patients, 28 (31%) had bipolar bone loss, with only 3 of these patients identified as having off-track Hill-Sachs lesions. For patients with bipolar bone loss, mean GBL was found to be 6.66% and ranged from 1 to 15%.

No statistically significant differences were observed in the post-operative outcomes, expectations about surgery, and satisfaction with surgery between patients with no bone loss, monopolar bone loss, and bipolar bone loss at all time points (p < 0.05). Clinical outcomes assessed pre-operatively and post-operatively up to 2-years are shown in full in Table 4.

Post-operatively, 5 patients (5.1%) in the study population were found to have complications. Two patients, one with no bone loss and one with bipolar bone loss, were found to have post-operative shoulder dislocations, while three patients, two with no bone loss and one with bipolar bone loss, were found to have post-operative pain. No post-operative wound infections or weakness were noted. No statistically significant differences were observed in complication rates between patients with no bone loss, monopolar bone loss, and bipolar bone loss at 2-years post-operatively (p > 0.05). Complication rates assessed post-operatively at 2-years are shown in full in Table 4.

4 Discussion

The most important findings of the present study were that, while patients with no bone loss were found to be significantly younger than other patients, a comparison of post-operative outcomes, expectations about surgery, satisfaction with surgery, and complications between patients with no bone loss, monopolar bone loss, and bipolar bone loss showed no statistically significant differences. A total of 98 patients met the selection criteria, with approximately 28.5% of the study population identified to have bipolar bone loss. Fewer than 11% of the patients with bipolar bone loss and less than 5% of the entire study population predicted to have an “off-track” Hill-Sachs lesion at risk of engagement.

While previous studies suggested patients with bipolar bone defects and subcritical GBL as low as 13.5% were correlated with worse outcomes after Bankart repair, our results suggest that patients with bipolar bone lesions and minor GBL up to 15% undergoing arthroscopic soft tissue repair only had similar clinical outcomes and complications rates up to 2-years post-operatively compared to those with monopolar or no bone loss.^12,15,16 Although three of the patients in the bipolar bone loss group had “off-track” lesions at risk of engagement and multiple patients had a GBL of 15%, they were not noted to have any complications at 2 years post-operatively and had similar clinical outcomes scores to those with “on-track” lesions. Risk of recurrence post-operatively was low amongst all three study populations and not found to be statistically different between the groups. Our results thus are supportive of the bipolar bone loss treatment algorithm and GBL threshold suggested by Gowd et al. which recommended that, in the absence of “off-track” Hill-Sachs lesions as predicted using the glenoid track concept, open or arthroscopic soft tissue stabilization procedures provide sufficient treatment for patients with bipolar bone lesions and GBL <15%.^1,4,12

Although the literature suggests that bipolar bone loss can occur in up to 70% of first time shoulder dislocations and up to 84% of chronic shoulder instability, our data show only 28.6% of patients with recurrent shoulder instability who underwent arthroscopic soft tissue repair had bipolar bone loss.^5,19 However, there are numerous factors that might contribute to the lower proportion of bipolar bone loss observed in our study. Firstly, our study chose to focus only on patients with minor GBL who underwent arthroscopic soft tissue stabilization procedures only. Therefore, patients with greater degrees of GBL, those who underwent bony reconstructive procedures, or those who opted for conservative management without surgery would have been excluded from analysis. Furthermore, our study predominantly uses MRI images for quantifying glenoid and humeral bone loss. Although MRI images have been used to glenohumeral bone loss, this may also contribute to the difference in bipolar bone loss detected compared to studies using CT scanning.^19–21 Therefore, although the proportion of bipolar bone loss observed in our study is lower than what's reported in the literature, the study population may not reflect the general population of patients with recurrent shoulder instability.

Our study is not without limitations. While bone loss of the glenoid and humerus are best visualized via 3D CT scanning of bilateral shoulders, this can be costly and exposes patients to more radiation. Although recent research indicates that glenoid bone loss and Hill-Sachs lesions can be effectively quantified by 3D MRI as an alternative, the gold standard is still a CT and our recommendations may differ from studies based on 3D CT imaging.^12,20–22 Furthermore, the calculations for determining on-track and off-track Hill-Sachs lesions as described by Itoi and Di Giacomo required CT measurements, which were not available in our study. As a retrospective analysis of prospectively collected data, our study is also limited by sample size and a minimum 2-years follow-up period. Furthermore, our study was limited by a low number of bipolar bone loss in patients and even fewer engaging Hill-Sachs lesions compared to the literature. It would also be helpful to include additional objective outcome metrics, such as range of motion and muscle strength, in future studies. Furthermore, we focused on the Constant-Murley and UCLA Shoulder Scores, which have been used, albeit uncommonly, to assess outcomes after shoulder stabilization surgery, as opposed to other scores such as the Western Ontario Shoulder Instability Index (WOSI).²³

Our results suggest that arthroscopic soft tissue repair is sufficient for the treatment of recurrent shoulder instability in patients with bipolar bone defects and minor glenoid bone loss, which may be useful in providing evidence for treatment algorithms and recommendations.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Andrew Chia Chen Chou: Data curation, Formal analysis, Writing - original draft, Writing - review & editing. Benjamin Joseph Kang: Data curation, Formal analysis, Writing - original draft, Writing - review & editing. Aaron Junjie Tan: Data curation, Formal analysis, Writing - original draft, Writing - review & editing. Denny Tijauw Tjoen Lie: Writing - original draft, Writing - review & editing.