Advertisement

Anteroinferior Glenoid Rim Fractures Are a Relatively Common Finding in Shoulder Instability Patients Aged 50 Years or Older but May Not Portend a Worse Prognosis

Open AccessPublished:September 08, 2022DOI:https://doi.org/10.1016/j.asmr.2022.07.012

      Purpose

      To investigate the incidence of anteroinferior glenoid rim fractures (AGRFs) after anterior shoulder instability (ASI) in patients aged 50 years or older, identify risk factors for surgical intervention for AGRFs, compare initial treatment strategies, and compare clinical outcomes of patients with and without associated AGRFs.

      Methods

      An established geographic medical record system was used to identify patients aged 50 years or older with ASI between 1994 and 2016. Patients with radiographic evidence of AGRFs were identified and matched 1:1 to patients without AGRFs. Outcome measures included recurrent instability, recurrent pain events, conversion to arthroplasty, and osteoarthritis graded with the Samilson-Prieto classification for post-instability arthritis.

      Results

      Overall, 177 patients were identified, with a mean follow-up period of 10.8 years. Of these patients, 41 (23.2%) had AGRFs and were matched to 41 control patients without AGRFs. The average age was 58.6 and 58.2 years for the AGRF and control groups, respectively. Rates of surgical intervention (27% vs 49%), recurrent instability (12% vs 20%), progression of osteoarthritis (34% vs 39%), and conversion to arthroplasty (2% vs 5%) were similar between AGRF patients and controls. For patients with AGRFs, increased bone fragment size (odds ratio, 1.1) and increased body mass index (odds ratio, 1.2) correlated with an increased risk of surgery. The cutoff value for an increased risk of surgery in patients with AGRFs was a fragment size 33% of the glenoid width or greater.

      Conclusions

      Of patients aged 50 years or older at presentation of ASI, 23.2% presented with an associated AGRF. A fragment size 33% of the glenoid width or greater and a higher patient body mass index were significant factors for surgical intervention; however, most patients did not require surgery and still showed acceptable clinical outcomes, and the most common reason for surgical intervention was a rotator cuff tear. Overall, the presence of an AGRF did not portend a worse prognosis as treatment strategies and long-term outcomes including recurrent instability, progression of osteoarthritis, and conversion to arthroplasty were similar to those in patients without AGRFs.

      Level of Evidence

      Level III, retrospective comparative study.
      The shoulder is the most commonly dislocated joint in the body owing to its dynamic range of motion and reliance on capsular and soft-tissue supporting structures.
      • Chan A.G.
      • Kilcoyne K.G.
      • Chan S.
      • Dickens J.F.
      • Waterman B.R.
      Evaluation of the Instability Severity Index score in predicting failure following arthroscopic Bankart surgery in an active military population.
      Shoulder dislocations in older patients are relatively frequent, with some studies reporting that up to 20% of all first-time anterior dislocations occur in patients older than 60 years.
      • Murthi A.M.
      • Ramirez M.A.
      Shoulder dislocation in the older patient.
      ,
      • Gumina S.
      • Postacchini F.
      Anterior dislocation of the shoulder in elderly patients.
      Although older patients are less likely to experience recurrent shoulder instability events, persistent morbidity due to rotator cuff injury, fracture, or nerve injury has been well documented.
      • Murthi A.M.
      • Ramirez M.A.
      Shoulder dislocation in the older patient.
      ,
      • Hawkins R.J.
      • Bell R.H.
      • Hawkins R.H.
      • Koppert G.J.
      Anterior dislocation of the shoulder in the older patient.
      Anteroinferior glenoid rim fractures (AGRFs), also known as “bony Bankart fractures,” are typically associated with traumatic anterior glenohumeral instability events.
      • Piasecki D.P.
      • Verma N.N.
      • Romeo A.A.
      • Levine W.N.
      • Bach Jr., B.R.
      • Provencher M.T.
      Glenoid bone deficiency in recurrent anterior shoulder instability: Diagnosis and management.
      These occur as the dislocated humeral head impacts the anterior glenoid, labrum, and capsule during a dislocation event. Particularly in the older population, glenoid fractures may occur owing to weakened and osteoporotic bone.
      • Sonnabend D.H.
      Treatment of primary anterior shoulder dislocation in patients older than 40 years of age. Conservative versus operative.
      Structurally, these lesions play a role in shoulder stability, but their impact on recurrent instability in the older patient is not as well studied as capsulolabral complex injuries in younger patients.
      • Randelli P.
      • Ragone V.
      • Carminati S.
      • Cabitza P.
      Risk factors for recurrence after Bankart repair a systematic review.
      • Shin S.J.
      • Yun Y.H.
      • Kim D.J.
      • Yoo J.D.
      Treatment of traumatic anterior shoulder dislocation in patients older than 60 years.
      • Duethman N.C.
      • Bernard C.D.
      • Leland D.
      • et al.
      Multiple instability events at initial presentation are the major predictor of failure of nonoperative treatment for anterior shoulder instability.
      • Leland D.P.
      • Bernard C.D.
      • Keyt L.K.
      • et al.
      An age-based approach to anterior shoulder instability in patients under 40 years old: Analysis of a US population.
      The current literature regarding the risk of recurrent instability in patients with acute anterior glenoid rim fractures offers inconclusive findings. Robinson et al.
      • Robinson C.M.
      • Kelly M.
      • Wakefield A.E.
      Redislocation of the shoulder during the first six weeks after a primary anterior dislocation: Risk factors and results of treatment.
      reported that the presence of an acute glenoid rim fracture after a first-time traumatic anterior dislocation was a significant risk factor for early redislocation after nonoperative management. Conversely, other investigations have suggested that the presence of a glenoid rim fracture after a primary shoulder dislocation does not increase the risk of recurrent instability for some patient groups.
      • Olds M.
      • Ellis R.
      • Donaldson K.
      • Parmar P.
      • Kersten P.
      Risk factors which predispose first-time traumatic anterior shoulder dislocations to recurrent instability in adults: A systematic review and meta-analysis.
      ,
      • Salomonsson B.
      • von Heine A.
      • Dahlborn M.
      • et al.
      Bony Bankart is a positive predictive factor after primary shoulder dislocation.
      It is clear that untreated chronic glenoid rim fractures may play a role in recurrent instability and arthritic progression owing to erosion, osteolysis, and continued bone loss.
      • Spiegl U.J.
      • Smith S.D.
      • Todd J.N.
      • Coatney G.A.
      • Wijdicks C.A.
      • Millett P.J.
      Biomechanical comparison of arthroscopic single- and double-row repair techniques for acute bony Bankart lesions.
      A distinction must be made between acute rim fractures with an associated bone fragment and cases of attritional glenoid bone loss without a remaining bone fragment because treatment options and clinical outcomes differ.
      • Dickens J.F.
      • Slaven S.E.
      • Cameron K.L.
      • et al.
      Prospective evaluation of glenoid bone loss after first-time and recurrent anterior glenohumeral instability events.
      For the purposes of this investigation, AGRF patients will encompass patients who experience acute fractures of the anteroinferior glenoid rim without attritional bone loss in the setting of anterior shoulder instability (ASI).
      Much of the published literature on traumatic ASI has focused on younger patient populations. There remains a paucity of available data regarding traumatic shoulder dislocations associated with anterior glenoid rim fractures in older patients. The purposes of this study were to investigate the incidence of AGRFs after ASI in patients aged 50 years or older, identify risk factors for surgical intervention for AGRFs, compare initial treatment strategies, and compare clinical outcomes of patients with and without associated AGRFs. We hypothesized that AGRFs would be prevalent in this patient population and would correlate with an increased likelihood of surgical intervention.

      Methods

      Study Population and Design

      After institutional review board approval was obtained from both Mayo Clinic (16-007084) and Olmsted Medical Center (042-OMC-16), the Rochester Epidemiology Project (REP) was used to identify patients who experienced ASI between January 1994 and July 2016. The REP is an electronic collection system of complete medical records involving a United States–based geographic cohort of more than 500,000 patients, all of whom were residents of Olmsted County, Minnesota, or neighboring counties in southeast Minnesota and western Wisconsin. The methodology and generalizability of the REP have previously been described in detail.
      • St Sauver J.L.
      • Grossardt B.R.
      • Yawn B.P.
      • et al.
      Data resource profile: The Rochester Epidemiology Project (REP) medical records-linkage system.
      ,
      • St Sauver J.L.
      • Grossardt B.R.
      • Yawn B.P.
      • Melton III, L.J.
      • Rocca W.A.
      Use of a medical records linkage system to enumerate a dynamic population over time: The Rochester Epidemiology Project.
      Patient records were identified by searching for International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes for ASI. Patient charts were then individually reviewed in detail to confirm the diagnosis of ASI and obtain the necessary study details. All patients were assessed by a primary care, emergency medicine, or orthopaedic physician. Patients were included if they had 1 or more anterior shoulder dislocation at age 50 years or later and were aged 50 years or older when ASI was first diagnosed. The exclusion criteria consisted of patients with multidirectional instability, posterior shoulder instability, previous surgery on the affected shoulder, or an initial ASI event occurring before the age of 50 years.
      Patients with AGRFs were identified by reviewing radiographs, computed tomography (CT) scans, or magnetic resonance imaging (MRI) scans of the affected shoulder. Images were initially reviewed by a musculoskeletal radiologist and were confirmed by the senior author (C.L.C.). The size of the AGRF bone fragment was measured on CT or MRI at the time of the instability diagnosis. The inferior portion of the glenoid rim was approximated to a true circle on sagittal views, and the size of the bone fragment was calculated as a percentage of the glenoid rim diameter.
      • Nakagawa S.
      • Mae T.
      • Yoneda K.
      • Kinugasa K.
      • Nakamura H.
      Influence of glenoid defect size and bone fragment size on the clinical outcome after arthroscopic Bankart repair in male collision/contact athletes.
      Patients with radiographic evidence of an acute anterior glenoid rim fracture were matched 1:1 with patients who had ASI without a glenoid fracture. For the matched-cohort analysis, patients were only included if they had at least 2 years of follow-up. The groups were matched based on age (±5 years), sex, occupation (laborer vs non-laborer), and dominant-sided versus non–dominant-sided dislocation. The surgical criteria for patients with AGRFs were dependent on patient factors including degree of instability, activity level, size of the glenoid fragment, and chronicity of the injury.
      Patient medical records were reviewed to collect demographic data (age at diagnosis, sex, body mass index [BMI], and occupation), smoking history, presence of diabetes, previous surgery on the affected shoulder, injury characteristics, and radiographic characteristics at presentation. Information regarding treatment strategies (operative vs nonoperative) and surgical details (open vs arthroscopic stabilization, method for addressing the AGRF, labral repair, rotator cuff repair, Latarjet procedure, and so on) was recorded. Patients were considered to have trialed an initial course of nonoperative management if they did not undergo surgery within 6 months of initial presentation. Outcome measures included recurrent instability, recurrent pain events, conversion to arthroplasty, and osteoarthritis graded with the Samilson-Prieto classification for post-instability arthritis.
      • Samilson R.L.
      • Prieto V.
      Dislocation arthropathy of the shoulder.

      Statistical Analysis

      Data were collected and stored in Microsoft Excel 2010 (Microsoft) and analyzed with JMP Pro (version 14.1.0; SAS Institute). Patient characteristics were presented with descriptive statistics using means, percentages, and 95% confidence intervals of the mean when appropriate. Univariate analysis of clinical and radiographic characteristics was performed to identify independent risk factors for surgery, and logistic regression analysis was performed to determine odds ratios (ORs). To determine cutoff values for independent risk factors, receiver operating characteristic (ROC) analysis with calculation of the area under the ROC curve (AUC) was performed. After analysis of data for parametric or nonparametric assumptions, continuous variables were compared between age groups using the Student t test or Wilcoxon rank sum test; categorical variables were similarly compared by use of χ2 analysis or the Fisher exact test. P < .05 was considered significant. Post hoc power analysis of the comparative analysis using rates of surgical intervention for each group yielded 21.4% power.

      Results

      Patient Demographic Data and Incidence

      Our initial database query of patients aged 50 years or older with a history of an ASI event resulted in 422 patients. Subsequently, 177 patients with a clinically confirmed diagnosis of first-time ASI occurring at age 50 years or later were identified, with a mean age of 58.9 (range, 50-70 years). Of these patients, 41 (23.2%) had radiographic evidence of an AGRF at initial presentation for the shoulder instability event.
      For the matched-cohort analysis, the 41 patients with AGRFs were matched in a 1:1 fashion with the control group without radiographic evidence of an AGRF. There were no differences in baseline demographic and clinical data between the 2 groups (Table 1). The mean total follow-up time for the overall cohort was 10.8 ± 5.9 years (range, 2.0-25.4 years).
      Table 1Patient Baseline Demographic and Clinical Data
      AGRF Group (n = 41)Control Group (n = 41)P Value
      Age, yr58.6 ± 4.958.2 ± 4.8.741
      Sex: male/female22/1922/19>.999
      BMI31.4 ± 7.932.4 ± 7.3.547
      Dominant dislocation22 (54)20 (49).842
      Occupation.356
       Laborer10 (24)9 (22)
       Non-laborer31 (76)28 (68)
       Unknown04 (10)
      Current or former smoker17 (41)25 (61).421
      Presence of diabetes7 (18)4 (10).393
      Previous subluxations at age ≥ 50 yr2 (5)2 (5)>.999
      Previous dislocations at age ≥ 50 yr4 (10)3 (7)>.693
      Clinical follow-up, yr10.1 ± 5.511.4 ± 6.3.420
      NOTE. Data are expressed as mean ± standard deviation or count (percentage).
      AGRF, anteroinferior glenoid rim fracture; BMI, body mass index.

      Radiographic Findings

      All patients underwent radiography at initial presentation. MRI scans were available for 48 patients (59%) overall and were performed at a median of 39 days (interquartile range [IQR], 14-117 days) after the initial instability event, and CT scans were available for 17 patients (21%) overall (14 AGRF patients and 3 controls) and were performed at a median of 7 days (IQR, 0-13.75 days) after presentation. Findings from radiographic evaluation of the 2 cohorts are presented in Table 2. An anteroinferior glenoid fracture was visible on radiographs in 35 patients in the AGRF cohort, whereas in 6, it was only visible on advanced imaging. Twenty patients had advanced imaging available for measurements, with a mean fragment size encompassing 25.6% (range, 11%-46%) of the transverse width of the articulating glenoid surface.
      Table 2Radiographic Characteristics
      AGRF GroupControl GroupP Value
      Radiographyn = 42n = 42
       Hill-Sachs lesion28 (68)13 (32).002
      Statistically significant (P < .05).
       Humeral head fracture5 (12)4 (10).724
       Samilson-Prieto graden = 27n = 20.975
      018 (66)13 (65)
      18 (30)6 (30)
      21 (4)1 (5)
      MRIn = 18n = 30
       Labral tear15 (83)16 (54).035
      Statistically significant (P < .05).
       Rotator cuff tear.010
      Statistically significant (P < .05).
      Partial9 (50)6 (20)
      Complete6 (33)23 (77)
       Hill-Sachs lesion14 (74)14 (42).133
      NOTE. Data are expressed as count (percentage).
      AGRF, anteroinferior glenoid rim fracture; MRI, magnetic resonance imaging.
      Statistically significant (P < .05).

      Management Decisions

      There were no differences in treatment strategies between the 2 cohorts (Table 3). Overall, 11 patients with AGRFs (27%) and 16 controls (39%) received operative intervention at a median of 93 days (IQR, 24-135 days) after the anterior shoulder dislocation. In 8 of the 11 patients with AGRFs managed operatively, surgical intervention for the glenoid fracture was performed because of instability (median time to surgery, 20.5 days), consisting of suture anchor fixation alone in 4, screw fixation in 1, both suture anchor and screw fixation in 1, reverse total shoulder arthroplasty (rTSA) with glenoid bone grafting in 1, and the Latarjet procedure in 1. The patients with AGRFs who required more robust fixation beyond suture anchors alone (screw fixation or Latarjet procedure) had larger glenoid fragment widths (35%-45%). The individual patient who was treated with rTSA with glenoid bone grafting was treated 279 days after initial presentation. The remaining 3 patients underwent surgical intervention for a rotator cuff pathology exclusively. For patients in the control group, rotator cuff injury was the most common reason for surgical intervention (81%). Multiple dislocation events at age 50 years or later were not associated with an increased rate of surgical intervention (P > .999).
      Table 3Treatment Strategies
      AGRF GroupControl GroupP Value
      Overall surgery1116.172
       Nonoperative treatment initially1 (9)5 (32).350
       Surgical treatment initially10 (91)11 (68).350
      Type of surgery>.254
       Open4 (36)4 (25)
       Arthroscopic6 (55)6 (37.5)
       Both open and arthroscopic1 (9)6 (37.5)
      Surgical pathology addressed
       Rotator cuff injury6 (55)13 (81).280
       Labral injury6 (55)5 (31).264
       Biceps injury1 (9)7 (47).105
       Humeral fracture1 (9)0.423
      NOTE. Data are expressed as count (percentage).
      AGRF, anteroinferior glenoid rim fracture.

      Risk Factors for Surgery in Patients With AGRFs

      Patients with AGRFs who underwent surgery to address glenoid fractures had a greater BMI (P = .035) and an increased AGRF size (P = .030) (Table 4). On logistic regression analysis, unit increases in preoperative BMI (OR, 1.1; P = .026) and AGRF size (OR, 1.2; P = .015) were found to be significant factors associated with surgery. All patients with a fragment size smaller than 20% were treated nonoperatively. Patients who required surgical stabilization of the glenoid fracture had fragment sizes ranging from 21% to 45% of the glenoid width. ROC curve cutoff values for surgical intervention were a fragment size of 33% (AUC, 0.79; sensitivity, 44%; specificity, 100%) and BMI of 33 (AUC, 0.72; sensitivity, 64%; specificity, 77%).
      Table 4Risk Factors for Surgery in AGRF Patients
      SurgeryNonoperativeP Value
      BMI35.8 ± 9.429.5 ± 6.5.035
      Statistically significant (P < .05).
      AGRF size, %30.1 ± 9.122 ± 5.7.035
      Statistically significant (P < .05).
      NOTE. Data are expressed as mean ± standard deviation.
      AGRF, anteroinferior glenoid rim fracture; BMI, body mass index.
      Statistically significant (P < .05).

      Outcomes and Complications

      There were no differences in recurrent instability events after initial presentation (Table 5). Documented recurrent instability was observed in 5 AGRF patients (12%) and 8 control patients (20%) (P = .547). The median time to recurrent instability was 134 days (IQR, 36-1,266 days) and 954.5 days (IQR, 50.75-4,275.25 days) for the AGRF and control patients, respectively (P = .306). There were no differences in recurrent pain episodes after initial presentation between the 2 groups (P = .078). In patients who underwent operative intervention, there were no recurrent instability events postoperatively. No differences in physical examination findings for forward flexion (P = .570) and external rotation (P = .601) at 2 years’ follow-up were observed between the 2 groups. Overall, 11 AGRF patients (27%) and 18 control patients (47%) reported returning to their occupations (P = .067). Arthritis progression based on an increased Samilson-Prieto grade was present in 9 AGRF and 9 control patients (P = .780) at final follow-up. Revision surgery was performed in 1 AGRF patient (9%) and 4 control patients (25%) (P = .619).
      Table 5Outcomes
      AGRF GroupControl GroupP Value
      Pain25 (63)33 (83).078
      Recurrent instability5 (12)8 (20).547
      Time to recurrent instability, d134 (36-1,266)954.5 (50.75-4,275.25).310
      Return to work, d49 (17-66)45 (14-91).893
      Return to work after surgery, d54.5 (43.75-288.75)90 (52.5-101.75)
      Progression of OA
      Data are reported for patients with radiographs available at final follow-up.
      9 (34)9 (39).780
      Revision surgery1 (9)4 (25).619
      Physical examination findings at final follow-up
       Forward flexion, °155 (102.5-170)150 (100-170).570
       External rotation, °50 (30-70)50 (30-70).601
      NOTE. Data are expressed as count (percentage) or median (interquartile range).
      AGRF, anteroinferior glenoid rim fracture; OA, osteoarthritis.
      Data are reported for patients with radiographs available at final follow-up.
      Overall, 1 AGRF patient (2%) and 2 control patients (5%) received rTSA. The AGRF patient underwent rTSA after initial nonoperative management, whereas the 2 control patients underwent eventual rTSA at 4.1 years and 11.4 months after the initial rotator cuff repair.
      Complications including adhesive capsulitis, infection, and nerve palsy were evaluated. Overall, adhesive capsulitis developed in 9 patients (3 AGRF patients vs 6 controls, P = .481), postsurgical infection developed in 1 control patient, and nerve palsy developed in 7 patients (2 AGRF patients vs 5 controls, P = .431).

      Discussion

      The primary finding of this study is that an AGRF is a relatively common finding in patients who experience an anterior shoulder dislocation at age 50 years or older. Patients with a glenoid fragment size 33% of the glenoid width or greater and a BMI of 33 of higher were more likely to undergo surgical intervention. In our overall cohort of ASI patients aged 50 years or older, 23.2% had radiographic evidence of an AGRF. The presence of an AGRF at the initial instability event did not significantly necessitate operative intervention, and most patients in the study (73% of AGRF patients and 61% of controls) did not require surgery after traumatic ASI. Concomitant rotator cuff pathology was the most common reason for surgery in both the AGRF cohort (55%) and control cohort (81%). Long-term outcomes including recurrent instability, revision surgery, progression of osteoarthritis, and conversion to arthroplasty were similar between the 2 groups.
      Although concomitant rotator cuff injuries with anterior shoulder dislocation are well documented in elderly patients, the osseous pathoanatomy of anterior instability events in this population is not as clear.
      • Murthi A.M.
      • Ramirez M.A.
      Shoulder dislocation in the older patient.
      In younger patients, injury to the anterior glenoid labrum is reported to be present in up to 97% of cases of primary anterior shoulder dislocation whereas the presence of associated glenoid rim fractures ranges from 8.6% to 16%.
      • Dickens J.F.
      • Slaven S.E.
      • Cameron K.L.
      • et al.
      Prospective evaluation of glenoid bone loss after first-time and recurrent anterior glenohumeral instability events.
      ,
      • Griffith J.F.
      • Antonio G.E.
      • Yung P.S.
      • et al.
      Prevalence, pattern, and spectrum of glenoid bone loss in anterior shoulder dislocation: CT analysis of 218 patients.
      In older patients with ASI, the prevalence of glenoid rim fractures is not as well studied. Abballe et al.
      • Abballe V.D.
      • Walter W.R.
      • Lin D.J.
      • Alaia M.J.
      • Alaia E.F.
      Anterior shoulder instability in the aging population: MRI injury pattern and management.
      conducted an MRI investigation of patients aged 40 years or older and reported that 37.5% had visible glenoid bone defects at initial presentation; however, the location and type of glenoid bone defect were not clearly defined. Radiographic evidence of Hill-Sachs lesions and labral tears was more likely to be present in patients with AGRFs. These findings may be expected owing to the overlapping pathophysiology of an AGRF and a Hill-Sachs lesion. As the humeral head translates anteriorly during the shoulder instability event, a compression fracture may occur along the posterosuperior aspect of the humeral head as it collides with the anterior glenoid and labrum.
      • Provencher M.T.
      • Frank R.M.
      • Leclere L.E.
      • et al.
      The Hill-Sachs lesion: Diagnosis, classification, and management.
      Hill-Sachs lesions rarely occur in isolation; accordingly, we would expect to observe coexisting injuries including anterior glenohumeral ligamentous or osseous pathologies.
      • Provencher M.T.
      • Frank R.M.
      • Leclere L.E.
      • et al.
      The Hill-Sachs lesion: Diagnosis, classification, and management.
      It is interesting to note that although AGRF patients were more likely to experience coexisting bony injuries including Hill-Sachs lesions, patients without AGRFs more commonly experienced concomitant rotator cuff injuries. This discrepancy may result from subtle differences in the mechanism of injury. During the initial traumatic instability event, energy transfer directly to osseous structures may result in AGRFs and associated bony lesions while sparing the rotator cuff. Conversely, if the force of impact is absorbed by the rotator cuff and supporting soft-tissue structures, this may reduce the likelihood of fracture, particularly in the older patient in whom asymptomatic age-related attritional tears are likely to be present.
      • Tempelhof S.
      • Rupp S.
      • Seil R.
      Age-related prevalence of rotator cuff tears in asymptomatic shoulders.
      In this study, patients with AGRFs were surgically managed at a similar rate to the matched controls (27% vs 49%), and 73% of patients with AGRFs were treated conservatively. This finding suggests that the presence of an AGRF does not necessitate operative intervention and patients in this age group may be successfully managed nonoperatively. Spiegl et al.
      • Spiegl U.J.
      • Ryf C.
      • Hepp P.
      • Rillmann P.
      Evaluation of a treatment algorithm for acute traumatic osseous Bankart lesions resulting from first time dislocation of the shoulder with a two year follow-up.
      developed a treatment algorithm for osseous Bankart lesions and reported encouraging clinical outcomes after using a conservative strategy for patients with small lesions (<5%), whereas patients with medium-sized (12.5%-25%) or large-sized (≥25%) Bankart fractures may benefit from a surgical treatment strategy.
      • Kim Y.K.
      • Cho S.H.
      • Son W.S.
      • Moon S.H.
      Arthroscopic repair of small and medium-sized bony Bankart lesions.
      ,
      • Scheibel M.
      • Magosch P.
      • Lichtenberg S.
      • Habermeyer P.
      Open reconstruction of anterior glenoid rim fractures.
      Our findings follow a similar trend as a greater fragment size, particularly 33% or greater, was associated with a greater incidence of surgical intervention. Particularly for patients with a fragment size of 35% or greater, a more robust stabilization with screw fixation or the Latarjet procedure may be necessary to achieve adequate surgical stabilization, whereas those with smaller fragment sizes may be optimally treated with suture anchor fixation. Conversely, in younger patients, surgical stabilization after a first-time instability event is associated with a significantly decreased risk of recurrence and improved outcomes compared with nonoperative treatment.
      • De Carli A.
      • Vadala A.P.
      • Lanzetti R.
      • et al.
      Early surgical treatment of first-time anterior glenohumeral dislocation in a young, active population is superior to conservative management at long-term follow-up.
      Thus, in the younger age group, a lower threshold for surgical management is recommended after primary anterior shoulder dislocation. However, in the middle-aged to older population, repairing the AGRF may not be necessary for smaller fragments and nonoperative treatment may be sufficient to stabilize the shoulder.
      Our results suggest that AGRFs may not significantly contribute to recurrent instability events in older patients. The current literature reports an overall rate of recurrent instability between 4% and 20% in elderly patients after primary traumatic anterior dislocation.
      • Gumina S.
      • Postacchini F.
      Anterior dislocation of the shoulder in elderly patients.
      ,
      • Pevny T.
      • Hunter R.E.
      • Freeman J.R.
      Primary traumatic anterior shoulder dislocation in patients 40 years of age and older.
      In our cohort, 14% of AGRF patients and 17% of control patients experienced recurrent instability. Regarding the time to recurrent instability between the 2 groups, although there were no statistically significant differences, the median time to recurrent instability was shorter in patients with AGRFs. Given the prevalence of concomitant rotator cuff tears in both groups, further investigation may be worthwhile to evaluate whether the presence of both an AGRF and a rotator cuff tear may contribute to an earlier recurrence of instability. Overall, the recurrence rate was low compared with younger patients. Duethman et al.
      • Duethman N.C.
      • Bernard C.D.
      • Leland D.
      • et al.
      Multiple instability events at initial presentation are the major predictor of failure of nonoperative treatment for anterior shoulder instability.
      reported a 60.7% rate of recurrent instability after nonoperative treatment in a group of patients younger than 40 years. The difference in the overall recurrence rate between younger and older patients may be explained in part by the resultant pathologies after traumatic anterior instability: Whereas older patients tend to have concomitant rotator cuff injuries, younger patients are likely to have resultant damage to the anterior stabilizing structures and glenohumeral ligaments.
      • Murthi A.M.
      • Ramirez M.A.
      Shoulder dislocation in the older patient.
      Additionally, younger patients may be involved in high-risk physical activities that contribute to an increased likelihood of redislocation.
      • Leland D.P.
      • Bernard C.D.
      • Keyt L.K.
      • et al.
      An age-based approach to anterior shoulder instability in patients under 40 years old: Analysis of a US population.
      To our knowledge, only 1 study has previously investigated AGRFs in older patients with respect to recurrent instability: Robinson et al.
      • Robinson C.M.
      • Kelly M.
      • Wakefield A.E.
      Redislocation of the shoulder during the first six weeks after a primary anterior dislocation: Risk factors and results of treatment.
      performed an investigation of patients with first-time ASI and identified a subset of 27 patients with associated glenoid rim fractures aged between 35 and 80 years. They reported a 37% recurrence rate and determined that the presence of an isolated glenoid rim fracture and the presence of a bifocal fracture (glenoid rim and greater tuberosity) were both significant risk factors for redislocation.
      • Robinson C.M.
      • Kelly M.
      • Wakefield A.E.
      Redislocation of the shoulder during the first six weeks after a primary anterior dislocation: Risk factors and results of treatment.
      However, all patients in their study were treated operatively owing to gross instability on initial assessment. Therefore, these findings may represent more severe cases and may not be generalizable to the entire spectrum of AGRFs.

      Limitations

      There are several important limitations to consider when interpreting the findings of our investigation. Given the retrospective nature of the study, the results and conclusions may be susceptible to the inherent biases of the retrospective process, including dependence on the accuracy and completeness of documentation within the patient records. Accordingly, information regarding comparisons of preoperative and postoperative range of motion after ASI was limited and unable to be analyzed in this study. Owing to the nature of the REP and retrospective design, standardized radiographic protocols and advanced imaging (MRI and/or CT) were not universally available for all patients. Because of the heterogeneity of available advanced imaging, comparisons between the 2 groups based on imaging characteristics must be carefully considered when interpreting these findings. It is possible that the number of AGRFs within our study may be under-reported given that CT scans are superior to radiographs and MRI scans for assessing glenoid fractures. Additionally, the ROC analysis to determine cutoff values for surgery based on glenoid fragment size was limited by available imaging and, as a result, must be interpreted with caution. Treatment strategies were based on physician preference because a standardized management protocol was not used for the included patients. Subsequently, certain valuable data points such as range of motion could not be accurately assessed in this study.

      Conclusions

      Of patients aged 50 years or older at presentation of ASI, 23.2% presented with an associated AGRF. A fragment size 33% of the glenoid width or greater and a higher patient BMI were significant factors for surgical intervention; however, most patients did not require surgery and still showed acceptable clinical outcomes, and the most common reason for surgical intervention was a rotator cuff tear. Overall, the presence of an AGRF did not portend a worse prognosis as treatment strategies and long-term outcomes including recurrent instability, progression of osteoarthritis, and conversion to arthroplasty were similar to those in patients without AGRFs.

      Acknowledgment

      The authors acknowledge support from the Foderaro-Quattrone Musculoskeletal-Orthopaedic Surgery Research Innovation Fund.

      Supplementary Data

      References

        • Chan A.G.
        • Kilcoyne K.G.
        • Chan S.
        • Dickens J.F.
        • Waterman B.R.
        Evaluation of the Instability Severity Index score in predicting failure following arthroscopic Bankart surgery in an active military population.
        J Shoulder Elbow Surg. 2019; 28: e156-e163
        • Murthi A.M.
        • Ramirez M.A.
        Shoulder dislocation in the older patient.
        J Am Acad Orthop Surg. 2012; 20: 615-622
        • Gumina S.
        • Postacchini F.
        Anterior dislocation of the shoulder in elderly patients.
        J Bone Joint Surg Br. 1997; 79: 540-543
        • Hawkins R.J.
        • Bell R.H.
        • Hawkins R.H.
        • Koppert G.J.
        Anterior dislocation of the shoulder in the older patient.
        Clin Orthop Relat Res. 1986; : 192-195
        • Piasecki D.P.
        • Verma N.N.
        • Romeo A.A.
        • Levine W.N.
        • Bach Jr., B.R.
        • Provencher M.T.
        Glenoid bone deficiency in recurrent anterior shoulder instability: Diagnosis and management.
        J Am Acad Orthop Surg. 2009; 17: 482-493
        • Sonnabend D.H.
        Treatment of primary anterior shoulder dislocation in patients older than 40 years of age. Conservative versus operative.
        Clin Orthop Relat Res. 1994; : 74-77
        • Randelli P.
        • Ragone V.
        • Carminati S.
        • Cabitza P.
        Risk factors for recurrence after Bankart repair a systematic review.
        Knee Surg Sports Traumatol Arthrosc. 2012; 20: 2129-2138
        • Shin S.J.
        • Yun Y.H.
        • Kim D.J.
        • Yoo J.D.
        Treatment of traumatic anterior shoulder dislocation in patients older than 60 years.
        Am J Sports Med. 2012; 40: 822-827
        • Duethman N.C.
        • Bernard C.D.
        • Leland D.
        • et al.
        Multiple instability events at initial presentation are the major predictor of failure of nonoperative treatment for anterior shoulder instability.
        Arthroscopy. 2021; 37: 2432-2439
        • Leland D.P.
        • Bernard C.D.
        • Keyt L.K.
        • et al.
        An age-based approach to anterior shoulder instability in patients under 40 years old: Analysis of a US population.
        Am J Sports Med. 2020; 48: 56-62
        • Robinson C.M.
        • Kelly M.
        • Wakefield A.E.
        Redislocation of the shoulder during the first six weeks after a primary anterior dislocation: Risk factors and results of treatment.
        J Bone Joint Surg Am. 2002; 84: 1552-1559
        • Olds M.
        • Ellis R.
        • Donaldson K.
        • Parmar P.
        • Kersten P.
        Risk factors which predispose first-time traumatic anterior shoulder dislocations to recurrent instability in adults: A systematic review and meta-analysis.
        Br J Sports Med. 2015; 49: 913-922
        • Salomonsson B.
        • von Heine A.
        • Dahlborn M.
        • et al.
        Bony Bankart is a positive predictive factor after primary shoulder dislocation.
        Knee Surg Sports Traumatol Arthrosc. 2010; 18: 1425-1431
        • Spiegl U.J.
        • Smith S.D.
        • Todd J.N.
        • Coatney G.A.
        • Wijdicks C.A.
        • Millett P.J.
        Biomechanical comparison of arthroscopic single- and double-row repair techniques for acute bony Bankart lesions.
        Am J Sports Med. 2014; 42: 1939-1946
        • Dickens J.F.
        • Slaven S.E.
        • Cameron K.L.
        • et al.
        Prospective evaluation of glenoid bone loss after first-time and recurrent anterior glenohumeral instability events.
        Am J Sports Med. 2019; 47: 1082-1089
        • St Sauver J.L.
        • Grossardt B.R.
        • Yawn B.P.
        • et al.
        Data resource profile: The Rochester Epidemiology Project (REP) medical records-linkage system.
        Int J Epidemiol. 2012; 41: 1614-1624
        • St Sauver J.L.
        • Grossardt B.R.
        • Yawn B.P.
        • Melton III, L.J.
        • Rocca W.A.
        Use of a medical records linkage system to enumerate a dynamic population over time: The Rochester Epidemiology Project.
        Am J Epidemiol. 2011; 173: 1059-1068
        • Nakagawa S.
        • Mae T.
        • Yoneda K.
        • Kinugasa K.
        • Nakamura H.
        Influence of glenoid defect size and bone fragment size on the clinical outcome after arthroscopic Bankart repair in male collision/contact athletes.
        Am J Sports Med. 2017; 45: 1967-1974
        • Samilson R.L.
        • Prieto V.
        Dislocation arthropathy of the shoulder.
        J Bone Joint Surg Am. 1983; 65: 456-460
        • Griffith J.F.
        • Antonio G.E.
        • Yung P.S.
        • et al.
        Prevalence, pattern, and spectrum of glenoid bone loss in anterior shoulder dislocation: CT analysis of 218 patients.
        AJR Am J Roentgenol. 2008; 190: 1247-1254
        • Abballe V.D.
        • Walter W.R.
        • Lin D.J.
        • Alaia M.J.
        • Alaia E.F.
        Anterior shoulder instability in the aging population: MRI injury pattern and management.
        AJR Am J Roentgenol. 2021; 216: 1300-1307
        • Provencher M.T.
        • Frank R.M.
        • Leclere L.E.
        • et al.
        The Hill-Sachs lesion: Diagnosis, classification, and management.
        J Am Acad Orthop Surg. 2012; 20: 242-252
        • Tempelhof S.
        • Rupp S.
        • Seil R.
        Age-related prevalence of rotator cuff tears in asymptomatic shoulders.
        J Shoulder Elbow Surg. 1999; 8: 296-299
        • Spiegl U.J.
        • Ryf C.
        • Hepp P.
        • Rillmann P.
        Evaluation of a treatment algorithm for acute traumatic osseous Bankart lesions resulting from first time dislocation of the shoulder with a two year follow-up.
        BMC Musculoskelet Disord. 2013; 14: 305
        • Kim Y.K.
        • Cho S.H.
        • Son W.S.
        • Moon S.H.
        Arthroscopic repair of small and medium-sized bony Bankart lesions.
        Am J Sports Med. 2014; 42: 86-94
        • Scheibel M.
        • Magosch P.
        • Lichtenberg S.
        • Habermeyer P.
        Open reconstruction of anterior glenoid rim fractures.
        Knee Surg Sports Traumatol Arthrosc. 2004; 12: 568-573
        • De Carli A.
        • Vadala A.P.
        • Lanzetti R.
        • et al.
        Early surgical treatment of first-time anterior glenohumeral dislocation in a young, active population is superior to conservative management at long-term follow-up.
        Int Orthop. 2019; 43: 2799-2805
        • Pevny T.
        • Hunter R.E.
        • Freeman J.R.
        Primary traumatic anterior shoulder dislocation in patients 40 years of age and older.
        Arthroscopy. 1998; 14: 289-294