Clinical and radiological factors associated with recurrent shoulder dislocation after conservative treatment of traumatic primary anterior shoulder dislocation

Abstract

Objective

The aim of this study is to evaluate the clinical and radiological factors affecting the development of recurrent shoulder dislocation in patients who have been followed up with conservative treatment after the traumatic primary anterior shoulder dislocation (TPASD), and to identify patient subgroups at high risk of recurrence.

Methods

Fifty-four patients (40 males and 14 females) aged 18–65 years who presented with first-time traumatic anterior shoulder dislocation and were treated conservatively were retrospectively evaluated. Patients with and without recurrent instability during a minimum follow-up of 12 months were compared regarding age, sex, affected side, Bankart lesion, bony Bankart lesion, greater tuberosity fracture, rotator cuff tear, Hill-Sachs lesion, Hill-Sachs lesion depth, glenoid version, and rotator interval width. Multivariable logistic regression analysis was performed to identify independent predictors of recurrent instability.

Results

The recurrence rate was 46.3% after a minimum follow-up of 12 months. Univariate analysis demonstrated significant associations between recurrent instability and younger age, male sex, the presence of bony Bankart lesions, the presence and depth of Hill-Sachs lesions, and the absence of rotator cuff tears. No significant differences were observed regarding affected side, Bankart lesions, greater tuberosity fractures, glenoid version, or rotator interval width. In multivariable logistic regression analysis, the presence of a Hill-Sachs lesion remained an independent predictor of recurrent instability (OR 7.88, 95% CI 1.77–35.08, p = 0.007).

Conclusion

Recurrent instability following conservative treatment of first-time traumatic anterior shoulder dislocation is strongly associated with structural bone defects. Although younger age and male sex were associated with recurrence in univariate analyses, multivariable analysis identified the presence of a Hill-Sachs lesion as the strongest independent predictor of recurrent instability. Detailed radiological assessment may therefore facilitate risk stratification and treatment decision-making following TPASD.

Introduction

Traumatic anterior shoulder dislocation(TASD) represents the most prevalent major joint dislocation, particularly among young and active individuals. The treatment approach following the initial attack remains the subject of controversy in clinical practice [1, 2]. Although conservative monitoring is the preferred initial treatment in many patients, recurrent shoulder dislocation remains a significant clinical problem [3]. The recurrence rates following the traumatic primary anterior shoulder dislocation (TPASD) have been documented in a broad range in the extant literature, and this variability is predominantly ascribed to the patient profile and heterogeneity in accompanying lesions [4]. As demonstrated in preceding studies, patient-based factors such as young age, male gender, and high activity level have been shown to increase the risk of recurrence [5]. However, in recent years, emphasis has been placed on the evaluation of instability not only through clinical parameters but also through imaging-based bone defect patterns. In particular, the concepts of Hill-Sachs lesions, Bankart lesions and bipolar bone loss have contributed to a more profound understanding of the pathophysiology of recurrent instability [5–8]. However, the extant literature does not yet present a definitive consensus on the clinical and radiological factors that determine the development of recurrent shoulder dislocation in patients treated conservatively. Moreover, a substantial proportion of extant studies have concentrated on the surgical population, and data specific to patients treated conservatively are limited [9]. This complicates the formulation of treatment decisions based on patient-specific risk stratification in clinical practice.

The aim of this study is to evaluate the clinical and radiological factors affecting the development of recurrent shoulder dislocation in patients who have been followed up with conservative treatment after the TPASD, and to identify patient subgroups at high risk of recurrence.

Materials and methods

Approval was obtained from the local ethics committee, and the entire process was conducted in accordance with the principles of the Declaration of Helsinki. The study comprised 54 patients (40 men and 14 women) aged between 18 and 65 years who presented with TPASD. The patients were treated with conservative methods and were subsequently followed up. None of the patients were competitive athletes and no patient had documented generalized ligamentous laxity. The mean follow-up period was 36.2 months. Patients with radiological imaging (X-ray and MRI) after the TPASD were retrospectively evaluated for recurrent shoulder dislocation during a minimum of 12 months of follow-up. Consequently, the cohort exhibiting recurrent dislocation during a minimum of 12 months of follow-up, as well as the cohort devoid of recurrent dislocation, underwent statistical analysis in terms of age, gender, side, the presence or absence of a Bankart lesion, a bony Bankart lesion, a tuberculum majus fracture, an MRI-detected rotator cuff tear following the initial dislocation, the presence of a Hill-Sachs (HS) lesion, the depth of the HS lesion, the glenoid version, and the rotator interval width.

Statistical analysis

Statistical analyses were performed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA). The normality of continuous variables was assessed using the Shapiro–Wilk test. Continuous variables with normal distribution were compared using the independent samples t-test, whereas non-normally distributed variables were analyzed using the Mann–Whitney U test (Wilcoxon rank-sum test). Categorical variables were compared using Pearson’s chi-square test or Fisher’s exact test, as appropriate. Variables demonstrating statistical significance in univariate analyses and considered clinically relevant were subsequently entered into a multivariable logistic regression model to identify independent predictors of recurrent instability. Statistical significance was defined as p < 0.05.

Results

In patients with TPASD who were subjected to conservative treatment, the recurrence rate of shoulder dislocation was found to be 46.29% following a minimum of 12 months of follow-up. Statistically significant differences were identified between the two groups with respect to various parameters, including age, male gender, the presence or absence of bone bankart lesions, the presence or absence of HS lesions, the depth of HS lesions, and the presence or absence of rotator cuff tears. No statistically significant differences were identified in relation to other parameters. (Table 1).

Table 1.

Comparison of clinical and radiologic characteristics between patients with and without recurrent shoulder dislocation

Variables	Non-recurrent group (n = 29)	Recurrent dislocation group (n = 25)	p value
Age (years)	44.37 ± 16.69	30.72 ± 11.30	0.008*
Gender	Male: 62.1% (n = 18) Female: 37.9% (n = 11)	Male: 88.0% (n = 22) Female: 12.0% (n = 3)	0.030*
Side	Right: 62.1% (n = 18) Left: 37.9% (n = 11)	Right: 64.0% (n = 16) Left: 36.0% (n = 9)	0.884
Bankart lesion	Present: 20.7% (n = 6) Absent: 79.3% (n = 23)	Present: 44.0% (n = 11) Absent: 56.0% (n = 14)	0.066
Bony Bankart lesion	Present: 0% (n = 0) Absent: 100% (n = 29)	Present: 20.0% (n = 5) Absent: 80.0% (n = 20)	0.011*
Greater tuberosity fracture	Present: 10.3% (n = 3) Absent: 89.7% (n = 26)	Present: 0% (n = 0) Absent: 100% (n = 25)	0.098
Rotator cuff tear	Present: 31.0% (n = 9) Absent: 69.0% (n = 20)	Present: 4.0% (n = 1) Absent: 96.0% (n = 24)	0.038*
Hill-Sachs lesion	Present: 37.9% (n = 11) Absent: 62.1% (n = 18)	Present: 88.0% (n = 22) Absent: 12.0% (n = 3)	< 0.001*
Hill-Sachs lesion depth	1.26 ± 1.71	3.91 ± 2.61	< 0.001*
Glenoid version (°)	−2.79 ± 5.45	−1.47 ± 4.39	0.395
Rotator interval width (mm)	20.25 ± 3.68	19.78 ± 2.02	0.722

Values are presented as mean ± standard deviation or percentage (n)

*Statistically significant (p < 0.05)

Multivariable logistic regression analysis

A multivariable logistic regression analysis was performed to determine independent predictors of recurrent instability. Age, sex, and the presence of a Hill-Sachs lesion were entered into the model based on their clinical relevance and significance in univariate analyses. Bony Bankart lesions were not included because all patients with a bony Bankart lesion experienced recurrent instability, resulting in complete separation and preventing reliable estimation within the logistic regression model. In the multivariable model, the presence of a Hill-Sachs lesion remained an independent predictor of recurrent instability (OR 7.88, 95% CI 1.77–35.08, p = 0.007). Neither age (OR 0.96, 95% CI 0.91–1.01, p = 0.108) nor male sex (OR 2.28, 95% CI 0.40–12.83, p = 0.351) retained statistical significance after adjustment. The results of the multivariable analysis are summarized in Table 2.

Table 2.

Multivariable logistic regression analysis for predictors of recurrent instability

Variable	Odds Ratio (OR)	95% CI	p value
Age	0.96	0.91–1.01	0.108
Male sex	2.28	0.40–12.83	0.351
Hill-Sachs lesion	7.88	1.77–35.08	0.007*

Discussion

Recent prognostic studies and systematic reviews over the past decade suggest that recurrence rates following conservative management of the TPASD cannot be expressed as a single rate, as they are highly dependent on patient characteristics. The most consistent predictors of recurrent instability are young age and male gender. In addition, participation in sports (contact/overhead) and high activity levels constitute an additional risk factor that significantly increases the likelihood of recurrence. Consequently, contemporary literature is deviating from the supposition that “conservative treatment = low risk”, instead advocating for risk stratification at the initial stage, even in patients undergoing conservative follow-up [4]. In the present study, the mean age of patients who developed recurrent shoulder dislocation was found to be significantly lower in comparison to the group which did not experience recurrence. Conversely, male gender was found to be significantly associated with recurrent dislocation. These findings are consistent with the results reported in large cohort studies and systematic reviews over the past decade. As indicated by the extant literature, higher connective tissue elasticity, particularly in younger age groups, high activity levels, and an increased tendency to return to sports, are among the main mechanisms explaining the risk of recurrence. The male gender has been demonstrated to be associated with increased participation in contact sports and elevated levels of mechanical loading. These results lend support to the notion that age and gender should be considered fundamental risk determinants when determining conservative follow-up.

In recent years, there has been a growing emphasis on clinical tools designed to predict the failure of conservative management. One such tool is the Nonoperative Instability Severity Index Score (NISIS), which is discussed as a practical model for predicting the risk of nonoperative failure, particularly in the athletic population. However, the extant literature also reports that instability scores may not work with the same accuracy in every population; in some cohorts, certain scores have been found to lack the expected predictive power. This approach endorses a multifaceted risk assessment methodology, encompassing the interrelated components of “age, sport, demand level, imaging (bone defect), and hyperlaxity”. In contrast to a unidimensional evaluation, this multifaceted approach is regarded as a more comprehensive and nuanced method of evaluating risk [10]. The present study revealed no significant difference between the groups in terms of the side on which the dislocation occurred. In a similar vein, glenoid version and rotator interval width were found to be non-associable with recurrent dislocation. The extant literature suggests that glenoid version has a stronger association with posterior instability, while its predictive value remains limited in cases of anterior instability. The findings concerning rotator interval width imply that this parameter alone might be inadequate for predicting the risk of recurrence. These findings are consistent with current approaches that argue that bone defect patterns and trauma severity are more decisive factors in anterior instability. The presence of soft tissue Bankart lesions was observed at a higher rate in the recurrent dislocation group in our study; however, this difference did not reach statistical significance. In contrast, bone Bankart lesions were only seen in the recurrent dislocation group and showed a significant association with recurrence. The literature from the last 10 years defines bone Bankart lesions as an important component of the “bipolar bone loss” spectrum and associates them more frequently with failure of conservative management. This suggests that not only labral injury but also disruption of glenoid rim integrity permanently affects the mechanics of instability. Although the association between greater tuberosity fractures and lower recurrence did not reach statistical significance, it is noteworthy that tuberculum majus fractures were observed only in the non-recurrent group. The extant literature suggests that tuberculum majus fractures may cause more limited damage to the capsulolabral structures through energy absorption by the bone, thereby reducing the risk of recurrence. This finding should be interpreted with caution due to the limited number of patients in our study. However, further studies examining this relationship in more detail may be beneficial.

A salient finding of our study is the strong correlation between the presence and severity of Hill-Sachs lesions and recurrent shoulder dislocation. In the recurrent dislocation group, Hill-Sachs lesions were observed with greater frequency and were found to have a significantly greater depth. As reported in the current literature, off-track Hill-Sachs lesions, as defined by the glenoid-track approach, have been identified as one of the strongest radiological predictors of recurrence in non-operative follow-up. Increased lesion depth has the potential to compromise stability by facilitating mechanical locking in the humeral head-glenoid contact area. Consequently, the findings of our study support the necessity to evaluate not only the presence of Hill-Sachs lesions but also their quantitative characteristics.

It is noteworthy that rotator cuff tears were observed to be significantly more prevalent in the group that did not experience recurrence. This finding is consistent with reports in the literature that “first shoulder dislocations in older age are often accompanied by rotator cuff tears and show lower recurrence rates.” The increased rigidity of the capsuloligamentous structure with advancing age and the decrease in activity level can be considered factors limiting the risk of recurrence in this patient group. Consequently, the presence of a rotator cuff tear may signify a distinct clinical phenotype, characterised more by pain and functional limitations than instability in younger populations. To further account for potential confounding, a multivariable logistic regression analysis was performed. Although several variables were associated with recurrence in univariate analyses, the adjusted model allowed assessment of their independent contribution to recurrent instability. Interestingly, after adjustment for age and sex, the presence of a Hill-Sachs lesion remained the only independent predictor of recurrent instability. This finding supports the growing body of evidence suggesting that structural bone defects may play a more decisive role in recurrent instability than demographic characteristics alone. The nearly eightfold increase in recurrence risk associated with Hill-Sachs lesions observed in our cohort further emphasizes the importance of detailed radiological assessment following primary traumatic anterior shoulder dislocation.

The results indicate that early surgical intervention may yield superior outcomes in terms of recurrence, particularly in young patients and those with high demand. However, it should be noted that functional outcomes and patient expectations following conservative follow-up are heterogeneous. Consequently, the prevailing approach tends to position conservative treatment not as the “default” option, but as a more rational choice in patients with a low risk of recurrence or in subgroups where surgery offers relatively fewer advantages [11].

This study has several limitations. First, its retrospective design may have introduced selection bias. Second, the sample size was relatively small, limiting statistical power and restricting the number of variables that could be included in the multivariable model. Third, all patients included in this cohort were non-athletes, and no patient had a documented history of generalized ligamentous laxity. Therefore, the effects of athletic participation and hyperlaxity, both recognized risk factors for recurrent instability, could not be evaluated. Furthermore, because all patients with a bony Bankart lesion experienced recurrent instability, this variable could not be reliably incorporated into the multivariable model. Finally, this was a single-center study, which may limit the generalizability of the findings.

This study demonstrates that recurrent shoulder dislocation following conservative treatment after the TPASD is not random; specifically, young age, male gender, and bone defect patterns (particularly the presence and depth of Hill-Sachs lesions and Bankart lesions) constitute significant risk factors for recurrence. Conversely, it was observed that the predictive value of certain anatomical parameters, including glenoid version and rotator interval width, is constrained. The findings emphasise the necessity of patient-specific risk stratification based on detailed radiological analysis, rather than solely clinical evaluation, when deciding on conservative follow-up. Multivariable analysis further demonstrated that the presence of a Hill-Sachs lesion was an independent predictor of recurrent instability. These findings emphasize the importance of detailed radiological assessment and individualized risk stratification when determining treatment strategies after TPASD.

Authors’ contributions

All of authors(Mert Kahraman Marasli, Kaan Berk Karasungur, Omer Furkan Oz, Ali Cavit, Birkan Kibar, Hakan Serhat Yanik) meet conditions 1, 2, 3 and 4. (1) made substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data; or the creation of new software used in the work; (2) drafted the work or revised it critically for important intellectual content; (3) approved the version to be published; and (4) agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

No funding.

Data availability

Data is available and can be sent upon request.

Declarations

Ethics approval and consent to participate

Approved by local ethics committee (Haydarpasa Numune Education and Research Hospital Editorial Board) with HNHEAH/GOAEK 2025/134. Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.