- Email: [email protected]
Bipolar Bone Loss in Patients With Anterior Shoulder Dislocation: A Comparison of Adolescents Versus Adult Patients
Bipolar Bone Loss in Patients With Anterior Shoulder Dislocation: A Comparison of Adolescents Versus Adult Patients Brian C. Lau, M.D., Devin Conway, B.S., Patrick F. Curran, M.D., Brian T. Feeley, M.D., and Nirav K. Pandya, M.D.
Purpose: To compare bipolar bone loss by evaluating the degree of glenoid bone loss, Hill-Sachs lesion size, and glenoid track in adolescents and adults with shoulder dislocations. Methods: We performed a retrospective review between 2012 and 2016 of surgical and nonsurgical patients with a history of anterior shoulder dislocations (primary or recurrent) who underwent magnetic resonance imaging of the affected shoulder. The exclusion criteria included multidirectional instability, prior surgery, and posterior dislocation. Patients were grouped into 2 groups: adolescents (aged 10-19 years) and adults (aged 20 years). The groups were compared regarding measures of glenoid bone loss (best-fit circle technique) and Hill-Sachs lesion size (medial margin of rotator cuff footprint to medial margin of Hill-Sachs lesion). If the medial margin of a Hill-Sachs lesion was within the glenoid track, it was defined as on track; if it was more medial than the glenoid track, it was defined as off track. Results: We identified 45 adolescents (mean age, 16.1 years) and 30 adults (mean age, 28.9 years) with anterior shoulder dislocations. There was no significant difference in percentage of bone loss between adolescents (mean, 8.4%) and adults (mean, 9.9%; P ¼ .23). There was no significant difference in Hill-Sachs lesion size between adolescents (mean, 12.7 mm) and adults (mean, 9.9 mm; P ¼ .12). There were 12 patients with off-track lesions. Off-track lesions were present in 11 of 45 adolescents (24.4%) and 1 of 30 adults (3.3%). Adolescents had an increased risk of having an off-track lesion (odds ratio, 9.38; 95% confidence interval, 1.14-77.1). A subgroup analysis identified multiple dislocations as an independent risk factor for an off-track lesion (odds ratio, 4.15; 95% confidence interval, 0.85-20.23). Conclusions: This study shows that adolescence and a history of multiple dislocations are independent risk factors for a greater likelihood of glenoid off-track lesions. The findings support the use of bipolar assessment of shoulder dislocators, especially in adolescents and multiple dislocators. Level of Evidence: Level III, retrospective comparative study.
A
dolescence has been identified as the most significant risk factor for recurrent anterior shoulder dislocations despite comprehensive nonoperative treatment.1,2 Similar to the situation in adults,
From the Division of Shoulder and Sports Medicine Surgery, Department of Orthopaedic Surgery, University of California San Francisco Medical Center (B.C.L., D.C., P.F.C., B.T.F., N.K.P.), San Francisco; and University of California San Francisco Medical Center, Children’s Hospital of Oakland (N.K.P.), Oakland, California, U.S.A. The authors report the following potential conflict of interest or source of funding: N.K.P. receives support from Orthopediatrics. Unpaid consultant on devices, none of which pertain to the submitted article. There is no payment provided for services. Received December 2, 2016; accepted April 6, 2017. Address correspondence to Brian C. Lau, M.D., University of California San Francisco Medical Center, 500 Parnassus Ave, MU 320-W, San Francisco, CA 94143, U.S.A. E-mail: [email protected] Ó 2017 by the Arthroscopy Association of North America 0749-8063/161191/$36.00 http://dx.doi.org/10.1016/j.arthro.2017.04.004
shoulder dislocations account for 90% of shoulder instability cases and usually occur after a fall during a sporting activity.3 In adolescent patients, however, there is a 75% to 80% rate of recurrent dislocations with nonoperative treatment.4-6 Outcomes of arthroscopic treatment of shoulder instability in the adolescent population have shown a much higher rate of failure than in the adult population.7,8 Shymon et al.8 reported a failure rate of 21% in the adolescent population within 2 years and a 5-year survival rate of only 49% after shoulder stabilization. This is compared with a failure rate of 8% to 11% in adults after 11 years.7 The reasons for this are not completely known. It may be because of the engagement of these athletes in high-demand collision sports that stress the repair unlike their adult counterparts. Other risk factors for the increased rate of recurrent instability in this population may include the amount of glenoid bone loss after initial dislocation. A degree of
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2017: pp 1-7
1
2
B. C. LAU ET AL.
glenoid bone loss occurs in up to 40% of traumatic anterior shoulder dislocations.9 Another risk factor may be the degree of humeral head injury. The incidence of a humeral head injury (Hill-Sachs lesion) in shoulder instability ranges from 38% to 88%.10,11 In contrast, several studies did not find that larger Hill-Sachs lesion size was a potential cause of recurrence.10-13 These studies, however, did not assess the relation of the HillSachs lesion with glenoid bone loss. More recently, Yamamoto et al.14 identified a model to account for both glenoid and humeral head (bipolar) bone loss: the glenoid track. In this concept, 2 terms were defined to simultaneously characterize the relation of glenoid bone loss and its relation to Hill-Sachs lesions: on track and off track.14 Metzger et al.15 modified the glenoid track model to allow for clinical application and found, through a regression analysis, that younger age and multiple dislocations were predictive of off-track lesions. To date, however, there has not been a direct comparison of the glenoid track concept between adolescent and adult patients with shoulder instability. The purpose of this study was to compare bipolar bone loss by evaluating the degree of glenoid bone loss, Hill-Sachs lesion size, and glenoid track lesions in adolescents and adults with shoulder dislocations. We hypothesized that adolescent patients would have similar glenoid bone loss but greater likelihood of off-track lesions because of the relation with the Hill-Sachs lesion.
Methods After obtaining institutional review board approval (No. 16-19,515), we performed a retrospective review of surgical and nonsurgical patients with a history of primary and recurrent anterior shoulder dislocations over a 4-year period (2012-2016) who underwent magnetic resonance imaging (MRI) with arthrograms of the affected shoulder. Dislocation was defined by an episode requiring a reduction. The exclusion criteria included prior surgery on the affected shoulder, multidirectional instability, and posterior dislocation. Demographic data, sports played, and number of dislocations were collected. The patients were grouped into 2 groups: adolescents (aged 10-19 years) and adults (aged 20 years). We used 20 years of age as a cutoff on the basis of prior studies that have established age, in particular less than 20 years, as a risk factor for recurrent instability.2,16-19 The groups were compared regarding bipolar bone loss by measuring glenoid bone loss and Hill-Sachs injury size as determined on MRI. All patients were evaluated for any glenoid bone loss and Hill-Sachs defects. On the basis of the extent of the bipolar lesion, the glenoid track was defined as outside and engaging the
glenoid on the humeral head (off track) or as inside and non-engaging (on track). MRI Assessment As described by Yamamoto et al.,14 the size of the glenoid track was determined by the amount of glenoid bone loss. When there is no glenoid defect, the width of the glenoid track is 84% of the glenoid width. Yamamoto et al. showed in a cadaveric study that this is the contact point of the glenoid with the arm in 60 of abduction to the medial margin of the rotator cuff footprint on the humerus. When there is a bony defect at the anterior rim of the glenoid, the defect width should be subtracted from the value representing 84% of the length to obtain the true width of the glenoid track.14 If the medial margin of the Hill-Sachs lesion was more medial than the glenoid track, it was defined as off track and more likely to engage and have a higher risk of recurrent instability. Conversely, if the medial margin of the Hill-Sachs lesions was within the glenoid track, it was defined as on track and less likely to engage14 (Fig 1). Our study used the modified glenoid track method as defined by Metzger et al.15 on the basis of clinical magnetic resonance (MR) images performed by a single evaluator. By use of the sagittal-oblique image, the bare spot was identified and the glenoid width was measured using this as the central reference point as described by Huysmans et al.20,21 To determine the expected glenoid width, a best-fit circle was placed on the inferior third of the glenoid centered on the bare spot and the diameter of the circle was measured to calculate the expected width prior to bone loss (Fig 2).20 The glenoid bone loss percentage and glenoid track were both determined from these measurements. The glenoid track was calculated as 84% of the actual glenoid width. In the setting of bone loss, the amount of bone loss was subtracted from 84% of the actual glenoid width to calculate the glenoid track.14 The size of the Hill-Sachs lesion was determined by the method reported by Saito et al.22 It was calculated based on the distance from the medial margin of the rotator cuff footprint to the medial margin of the Hill-Sachs lesions, in millimeters, by use of the largest distance on a coronal image (Fig 2). If this value was greater than the previously determined glenoid track (84% of actual glenoid width minus any bone loss), then the humeral lesion was determined to be off track. If the value was less than the glenoid track (84% of actual glenoid width minus any bone loss), then the humeral lesion was determined to be on track. The described method of combined glenoid bone loss, Hill-Sachs assessment, and glenoid tracking has been shown to have good intraobserver and interobserver reliability and high correlation (>0.89).23
3
BIPOLAR BONE LOSS
Statistical Analysis Descriptive statistics were performed to determine the demographic characteristics of each study population. Student t tests for parametric data were performed to compare the 2 groups: adolescents (aged 10-19 years) and adults (aged 20 years). Adjusted odds ratios were used to determine adolescents’ risk of having off-track lesions. To test intraobserver reliability, the same measurements were performed at 3 months by the same evaluator (B.C.L.). To test interobserver reliability, a second evaluator (P.F.C.) performed imaging measurements using the aforementioned techniques. Pearson correlation coefficients were used to determine reliability. The k statistic was used to test the intraobserver and interobserver reliability of on- and off-track classification.
Results We identified 45 adolescents (aged 10-19 years; mean age, 16.1 years; 12 female adolescents) and 30 adults (aged 20-39 years; mean age, 28.9 years; 4 female adults) (Table 1). There was no significant difference in body mass index between the 2 groups (P ¼ .31). There was no significant difference in glenoid bone loss between adolescents (mean percentage of glenoid bone loss, 8.4%) and adults (mean percentage of glenoid bone loss, 9.9%; P ¼ .23) (Table 2). There was no significant difference in Hill-Sachs lesion size in adolescents (mean size, 12.7 mm) compared with adults (mean size, 9.9 mm; P ¼ .12) (Table 2). There were a total of 12 patients with off-track lesions. Off-track lesions were present in 11 of 45 adolescents (24.4%) and 1 of 30 adults (3.3%) (Table 2). Adolescents had an increased risk of having an off-track lesion (odds ratio, 9.38, 95% confidence interval, 1.14-77.1) (Table 3). A subgroup analysis comparing all patients (adolescents and adults) with 1 dislocation (single) versus those with 2 or more dislocations (multiple) was performed. There was no significant difference in bone loss between single dislocators (mean percentage of glenoid bone loss, 8.4%) and multiple dislocators (mean percentage of glenoid bone loss, 9.5%; P ¼ .47) (Table 4). To determine whether the size of glenoid bone loss correlated with the number of dislocations, a Pearson correlation was performed, with R ¼ 0.25 (P ¼ .83). The multiple-dislocations group had a significantly larger Hill-Sachs lesion size (mean, 14.3 mm) than the single-dislocation group (mean, 8.1 mm; P ¼ .0003) (Table 4). To determine whether the size of Hill-Sachs lesions correlated with the number of dislocations, a Pearson correlation was performed, with R ¼ 0.26 (P ¼ .82). The rate of off-track lesions was 2 of 34 (5.9%) in the single-dislocation group and 10 of 41 (24.4%) in the multiple-dislocations group (Table 3). The presence of multiple dislocations was an independent risk factor for an off-track lesion (odds ratio, 4.15; 95% confidence interval, 0.85-20.23) (Table 4).
Fig 1. The distances from the medial margin of the contact area to the edge of the articular surface of the humeral head (MA) and to the medial margin of the cuff attachment site on the greater tuberosity (MF) are measured. (A, lateral margin of articular surface; C, articular center of humeral head; F, footprint; M, most medial point.) Figure courtesy of Sonali Feeley.
The types of activities that patients participated in are included in Table 1. The most common mechanism of injury for adolescents was football, at 37.7%. The most common mechanism of injury for adults was gym exercises (weight lighting, yoga), at 43.3%. Injuries in adolescents were more associated with contact sports (football, wrestling, basketball), high-risk sports (snowboarding, climbing), or high-energy events (motor vehicle accidents), at 68.9% (31 of 45), compared with adults, at 23.3% (7 of 30). Intraobserver reliability for the percentage of glenoid bone loss and Hill-Sachs lesion size had Pearson correlation coefficients of 0.91 (P ¼ .02) and 0.87 (P ¼ .03), respectively. Interobserver reliability for the percentage of glenoid bone loss and Hill-Sachs lesion size had Pearson correlation coefficients of 0.86 (P ¼ .03) and 0.83 (P ¼ .04), respectively. The k statistics for intraobserver and interobserver classification into on track and off track were 0.86 and 0.81, respectively.
Discussion The principal finding of this study was that adolescent patients with shoulder instability have a greater
4
B. C. LAU ET AL.
Fig 2. The glenoid track is calculated as 84% of the actual glenoid width measured on the sagittal-oblique magnetic resonance image. (A) Sagittal-oblique fluid-sensitive image. A best-fit circle is placed on the glenoid to calculate the expected width prior to bone loss. The percentage of bone loss and glenoid track can be determined. (B) Coronal-oblique fluid-sensitive image. The distance from the medial edge of the rotator cuff footprint to the medial margin of the Hill-Sachs lesion is measured on the coronal magnetic resonance image. A Hill-Sachs lesion width that is greater than the glenoid track is considered outside the glenoid track and deemed off track.
likelihood of having “off-track” shoulder lesions than adult patients with anterior shoulder dislocations. Adolescents had a 9.4 increased risk of having an off-track lesion. A history of multiple dislocations (2) regardless of age was also an independent risk factor, with a 4.2 increased risk of an off-track lesion. These findings suggest that glenoid track assessment should be performed in the adolescent population with a history of shoulder dislocations and in multiple dislocators regardless of age as part of their diagnostic workup.
The key to the glenoid track concept is that the size and location of the Hill-Sachs lesion in relation to the glenoid bone loss are important factors for the risk of engagement and recurrent instability. In this study, glenoid bone loss and Hill-Sachs lesion size were similar between adolescents and adults. This finding highlights that although related to size, the location of the HillSachs lesion is also an important variable to consider. For example, the larger the size of the Hill-Sachs lesion, the more likely it will extend medially on the humeral head and engage the glenoid. Similarly, a smaller
Table 1. Demographic Characteristics n Mean age (range), yr Gender, n Mean BMI (range) Sport or mechanism of injury, n Football Gym exercises (weight lifting or yoga) Wrestling Basketball Baseball or softball Cheerleading Volleyball Soccer Snowboarding or skiing Swimming Surfing Climbing Motor vehicle accident Miscellaneous (lifting baby, moving mattress, towel drying hair, running from bee) BMI, body mass index; F, female; M, male.
Adolescents 45 16.1 (13-19) 33 M and 12 F 25.7 (20.5-33.2) 17 2 5 4 3 1 2 2 4
5
Adults 30 28.9 (20-39) 26 M and 4 F 23.45 (19.9-30.1)
13 1 2 1 2 2 2 1 1 1 4
P Value <.001 .31
5
BIPOLAR BONE LOSS Table 2. Comparison of Glenoid Bone Loss, Hill-Sachs Lesion Size, and Glenoid Tracking in Adolescents Versus Adults n Mean glenoid bone loss Mean expected glenoid diameter, mm Mean resultant glenoid diameter, mm Mean Hill-Sachs lesion size, mm Glenoid tracking, n Off track On track
Adolescents 45 8.4% 25.4 23.3 12.77 11 34
Adults P Value 30 9.9% .67 26.0 .32 23.5 .75 10.0 .12 1 29
NOTE. Adolescents were aged 13 to 19 years, and adults were aged 20 years or older.
Hill-Sachs lesion that is more medialized is also more likely to engage the glenoid (off track) when the arm is placed into abduction and external rotation. Adolescence is a known risk factor for recurrent dislocations during nonoperative management1,2 and after operative management,8 and the findings of this study suggest that off-track lesions may contribute to this increased risk. An off-track lesion was found in 24.4% of adolescent patients, whereas only 3.3% of adults had an off-track lesion. As such, these findings support previous studies’ recommendations on the assessment of bipolar bone lesions in shoulder instability and on planning surgical or conservative treatment, especially in younger patients.14,15,24 The differences between adolescents and adults may also be a result of the injury mechanism. In this study, adolescents were more likely to have an injury during a specific contact athletic event, with football being the most common mechanism, occurring in 37.7% of adolescents. Other high-risk activities that adolescents were more likely to engage in were wrestling, basketball, and snowboarding or skiing. The adult population underwent injury most commonly from gym exercises, such as weight lifting or yoga, at a rate of 43.3%. It is possible that the more traumatic nature of adolescent injuries may result in a higher rate of offtrack lesions. In addition, the findings from this study identified multiple dislocations as a risk factor for having an offtrack glenoid lesion. An interesting finding was that the presence of multiple dislocations did not affect the amount of glenoid bone loss but was associated with
Table 3. ORs of Adolescence and History of Multiple Dislocations Adolescence (age 13-19 yr) History of multiple dislocations
OR 9.38 4.15
CI, confidence interval; OR, odds ratio.
95% CI 1.14-77.1 0.85-20.23
P Value <.01 <.01
Table 4. Comparison of Glenoid Bone Loss, Hill-Sachs Lesion Size, and Glenoid Tracking for Single Versus Multiple Dislocations
n Adolescents/adults, n No. of patients with 2 dislocations No. of patients with 3 dislocations No. of patients with 4 dislocations No. of patients with 5 dislocations No. of patients with 7 dislocations Mean glenoid bone loss Mean Hill-Sachs lesion size, mm Glenoid tracking, n Off track On track
Single Multiple P Dislocation Dislocations Value 34 41 21/13 24/17 20 11 4 5 1 8.4% 9.5% .47 8.1 14.3 .0003 2 32
10 31
NOTE. The single-dislocation group had 1 dislocation before magnetic resonance imaging, whereas the multiple-dislocations group had 2 or more dislocations before magnetic resonance imaging.
larger Hill-Sachs lesion size. A history of multiple dislocations is a known risk factor for recurrent instability because of possible effects on the articular cartilage and soft tissues,25 and the findings from this study suggest that subsequent dislocations’ effect on humeral head and glenoid tracking may be another etiology for increased risk of recurrent instability. It is also important to note that in this study, the frequency of recurrent dislocations did not correlate with Hill-Sachs lesion size. A possible explanation is that the increased HillSachs defect occurs during the second dislocation and that subsequent dislocations may not increase the defect. However, it is also possible that a larger number of patients or a population with a larger range of dislocations would better bear out this difference. The natural history of glenohumeral dislocations is that adolescents have a greater likelihood, up to 75% to 80%, of recurrent dislocations5,6,26,27; however, it is important to note that in this study, the adolescent and adult patients were relatively equally distributed in the single-dislocation and multiple-dislocations groups. Among adolescents, 46.7% had a single dislocation and 53.3% had multiple dislocations. Among adults, 43.3% had a single dislocation and 56.7% had multiple dislocations. The findings from this study suggest that subsequent dislocation events have a greater effect on the humeral head than on the glenoid regardless of age. Although traditional data have shown that there is greater glenoid loss with recurrent dislocations,28 some recent data have suggested that recurrent dislocations may have a greater effect on Hill-Sachs lesions. Kim et al.29 compared patients with a single dislocation and patients with recurrent dislocations and found that compared with single dislocators, a larger percentage of recurrent dislocators had Hill-Sachs lesions (37% more)
6
B. C. LAU ET AL.
and glenoid erosion (13.5% more). A recent study compared recurrent dislocators and patients with recurrent subluxation without frank dislocation and also showed a similar presence of glenoid erosion between the groups but a significant increase in Hill-Sachs lesions in recurrent dislocators.30 It remains unclear whether both the glenoid and humeral head defects need to be treated in off-track lesions. Clinicians may choose to perform an isolated Bankart repair, glenoid bone grafting, or Hill-Sachs treatment with remplissage or bone grafting. A recent biomechanical study of off-track lesions found that Bankart repair prevented engagement of 67% and 0% of shoulders in 60 and 90 of external rotation, respectively.31 In contrast, adding Hill-Sachs remplissage to Bankart repair prevented engagement in all shoulders, 100%, at 60 and 90 of external rotation. The authors noted that Bankart repair with remplissage resulted in shoulder stiffness but was necessary to prevent engagement of off-track lesions. Di Giacomo et al.32 developed a treatment algorithm based on the percentage of glenoid loss and glenoid track classification. Glenoid treatment was determined independently of the Hill-Sachs lesion, with less than 25% bone loss being treated with arthroscopic Bankart repair and more than 25% bone loss being treated with the Latarjet procedure. The authors suggested that Hill-Sachs lesion treatment be based on glenoid track classification, with only off-track lesions warranting the remplissage procedure. They also suggested possible humeral bone grafting in off-track lesions with large glenoid defects (>25% bone loss). Limitations This study has several limitations. MRI was chosen in this study for a number of reasons. First, MR arthrography imaging is the most common advanced imaging modality for shoulder instability to evaluate labral and other soft-tissue injuries for surgical planning and therefore may be a more generalizable imaging modality. MR arthrograms have been shown to have sensitivity and specificity of between 86% and 98% in identifying labral and cartilage injury.33 Second, ionizing radiation exposure risk in the adolescent population has led to less use of computed tomography (CT) scans in this at-risk population. Last, MRI assessment of glenoid bone loss has been shown to be as accurate as CT scans in recent studies.15,21,34 We believe that MR arthrography imaging was the most practical and generalizable imaging modality to assess bone loss, but we recognize that 3-dimensional CT scanning is also highly accurate. Of note, MR arthrograms are more sensitive than non-arthrogram MRI for bone loss.33 Our study population mostly comprised male patients, which may not be representative of the general population. However, prior studies have shown
that young male patients are the population most at risk of shoulder dislocations after a traumatic sporting event.35 This study found that younger patients engaged in higher-risk activities including contact sports at a greater rate than adults, which may affect the rate of bipolar injury. The body mass indexes of the 2 groups were similar, which shows that the physical sizes of the patients were similar, but given the inherent differences in age-related activities, such as activity level, high-risk activities, or contact sports, it is difficult to control for these factors. In addition, our study found no significant difference in Hill-Sachs lesion size between adolescents and adults, although the average size was larger in the adolescent group. It is possible that there is a true difference that this study was underpowered to detect. Interobserver and intraobserver reliability for percentage of glenoid bone loss and Hill-Sachs lesion size ranged from 0.83 to 0.91, which is similar to interobserver and intraobserver reliabilities shown by prior studies using MRI with Pearson correlation coefficients.23 However, it is important to note that prior studies have also identified variability of glenoid track classification. A recent study showed that glenoid track classification had 72% agreement, with the Hill-Sachs measurement having a high level of variability, with a coefficient of variability of 19.2%.36 Our study had k statistics of 0.86 for intraobserver glenoid track classification and 0.81 for interobserver classification.
Conclusions This study shows that adolescence and a history of multiple dislocations are independent risk factors for a greater likelihood of glenoid off-track lesions. The findings support the use of bipolar assessment of shoulder dislocators, especially in adolescents and multiple dislocators.
Acknowledgment The authors acknowledge Sonali Feeley for making Figure 1.
References 1. Flinkkilä T, Hyvönen P, Ohtonen P, Leppilahti J. Arthroscopic Bankart repair: Results and risk factors of recurrence of instability. Knee Surg Sports Traumatol Arthrosc 2010;18:1752-1758. 2. te Slaa RL, Wijffels MP, Brand R, Marti RK. The prognosis following acute primary glenohumeral dislocation. J Bone Joint Surg Br 2004;86:58-64. 3. Mazzocca AD, Brown FM Jr, Carreira DS, Hayden J, Romeo AA. Arthroscopic anterior shoulder stabilization of collision and contact athletes. Am J Sports Med 2005;33:52-60. 4. Arciero RA, Wheeler JH, Ryan JB, McBride JT. Arthroscopic Bankart repair versus nonoperative treatment for
BIPOLAR BONE LOSS
5.
6.
7. 8.
9. 10.
11.
12.
13.
14.
15.
16. 17. 18.
19.
20.
21.
acute, initial anterior shoulder dislocations. Am J Sports Med 1994;22:589-594. Bottoni CR, Wilckens JH, DeBerardino TM, et al. A prospective, randomized evaluation of arthroscopic stabilization versus nonoperative treatment in patients with acute, traumatic, first-time shoulder dislocations. Am J Sports Med 2002;30:576-580. Deitch J, Mehlman CT, Foad SL, Obbehat A, Mallory M. Traumatic anterior shoulder dislocation in adolescents. Am J Sports Med 2003;31:758-763. Harris JD, Gupta AK, Mall NA, et al. Long-term outcomes after Bankart shoulder stabilization. Arthroscopy 2013;29:920-933. Shymon SJ, Roocroft J, Edmonds EW. Traumatic anterior instability of the pediatric shoulder: A comparison of arthroscopic and open Bankart repairs. J Pediatr Orthop 2015;35:1-6. Postacchini F, Gumina S, Cinotti G. Anterior shoulder dislocation in adolescents. J Shoulder Elbow Surg 2000;9:470-474. Rowe CR, Zarins B, Ciullo JV. Recurrent anterior dislocation of the shoulder after surgical repair. Apparent causes of failure and treatment. J Bone Joint Surg Am 1984;66:159-168. Yiannakopoulos CK, Mataragas E, Antonogiannakis E. A comparison of the spectrum of intra-articular lesions in acute and chronic anterior shoulder instability. Arthroscopy 2007;23:985-990. Rowe CR, Patel D, Southmayd WW. The Bankart procedure: A long term end-result study. J Bone Joint Surg Am 1978;60:1-16. Ungersbock A, Michel M, Hertel R. Factors influencing the results of a modified Bankart procedure. J Shoulder Elbow Surg 1995;4:365-369. Yamamoto N, Itoi E, Abe H, et al. Contact between glenoid and the humeral head in abduction, external rotation, and horizontal extension: A new concept of glenoid track. J Shoulder Elbow Surg 2007;16:649-656. Metzger PD, Barlow B, Leonardelli D, Peace W, Solomon DJ, Provencher MT. Clinical application of the “glenoid track” concept for defining humeral head engagement in anterior shoulder instability. Orthop J Sports Med 2013;1:2325967113496213. Henry JH, Genung JA. Natural history of glenohumeral dislocation revisited. Am J Sports Med 1982;10:135-137. McLaughlin H, Cavallaro WU. Primary anterior dislocation of the shoulder. Am J Surg 1950;80:615-621. passim. Rowe CR, Sakellarides HT. Factors related to recurrences of anterior dislocations of the shoulder. Clin Orthop Relat Res 1960;20:40-48. Wasserstein DN, Sheth U, Colbenson K, et al. The true recurrence rate and factors predicting recurrent instability after nonsurgical management of traumatic primary anterior shoulder dislocation: A systematic review. Arthroscopy 2016;32:2616-2625. Huysmans PE, Haen PS, Kidd M, Dhert WJ, Willems JW. The shape of the inferior part of the glenoid: A cadaveric study. J Shoulder Elbow Surg 2006;15:759-763. Huijsmans PE, Haen PS, Kidd M, Dhert WJ, van der hulst VP, Willems JW. Quantification of a glenoid defect
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
7
with three-dimensional computed tomography and magnetic resonance imaging: A cadaveric study. J Shoulder Elbow Surg 2007;16:803-809. Saito H, Itoi E, Mingawa H, Yamamoto N, Tuoheti Y, Seki N. Location of the Hill-Sachs lesion in shoulders with recurrent anterior dislocation. Arch Orthop Trauma Surg 2009;129:1327-1334. Shaha JS, Cook JB, Rowles DJ, Bottoni CR, Shaha SH, Tokish JM. Clinical validation of the glenoid track concept in anterior glenohumeral instability. J Bone Joint Surg Am 2016;98:1918-1923. Omori Y, Yamamoto N, Koishi H, et al. Measurement of the glenoid track in vivo as investigated by 3-dimensional motion analysis using open MRI. Am J Sports Med 2014;42: 1290-1295. Brophy RH, Marx RG. Osteoarthritis following shoulder instability. Clin Sports Med 2005;24:47-56. Roberts SB, Beattie N, McNiven ND, Robinson CM. The natural history of primary anterior dislocation of the glenohumeral joint in adolescence. J Bone Joint Br 2015;97:520-526. Eljabu W, Klinger HM, von Knoch M. The natural course of shoulder instability and treatment trends: A systematic review. J Orthop Traumatol 2017;18:1-8. Sugaya H, Moriishi J, Dohi M, Kon Y, Tsuchiya A. Glenoid rim morphology in recurrent anterior glenohumeral instability. J Bone Joint Surg Am 2003;85:878-884. Kim DS, Yoon YS, Yi CH. Prevalence comparison of accompanying lesions between primary and recurrent anterior dislocation in the shoulder. Am J Sports Med 2010;38:2071-2076. Shin SJ, Ko YW, Jeon YS, Lee J, Kim RG, Baek H. Comparison of intra-articular findings and clinical features between patients with symptomatic anterior instability after recurrent shoulder subluxation and dislocation. Arthroscopy 2017;33:527-533. Hartzler RU, Bui CN, Jeong WK, et al. Remplissage of an off-track Hill-Sachs lesion is necessary to restore biomechanical glenohumeral joint stability in a bipolar bone loss model. Arthroscopy 2016;32:2466-2476. Di Giacomo G, Itoi E, Burkhart SS, et al. Evolving concept of bipolar bone loss and the Hill-Sachs lesion: From “engaging/non-engaging” lesion to “on-track/off-track” lesion. Arthroscopy 2014;30:90-98. Steinbach LS. MRI of shoulder instability. Eur J Radiol 2008;68:57-71. Tian CY, Shang Y, Zheng ZZ. Glenoid bone lesions: Comparison between 3D VIBE images in MR arthrography and nonarthrographic MSCT. J Magn Reson Imaging 2012;36:231-236. Owens BD, Duffey ML, Nelson BJ, et al. The incidence and characteristics of shoulder instability at the United States Military Academy. Am J Sports Med 2007;35: 1168-1173. Schneider AK, Hoy G, Ek ET, et al. Interobserver and intraobserver variability of glenoid track measurements. J Shoulder Elbow Surg 2017;26:573-579.
Purpose: To compare bipolar bone loss by evaluating the degree of glenoid bone loss, Hill-Sachs lesion size, and glenoid track in adolescents and adults with shoulder dislocations. Methods: We performed a retrospective review between 2012 and 2016 of surgical and nonsurgical patients with a history of anterior shoulder dislocations (primary or recurrent) who underwent magnetic resonance imaging of the affected shoulder. The exclusion criteria included multidirectional instability, prior surgery, and posterior dislocation. Patients were grouped into 2 groups: adolescents (aged 10-19 years) and adults (aged 20 years). The groups were compared regarding measures of glenoid bone loss (best-fit circle technique) and Hill-Sachs lesion size (medial margin of rotator cuff footprint to medial margin of Hill-Sachs lesion). If the medial margin of a Hill-Sachs lesion was within the glenoid track, it was defined as on track; if it was more medial than the glenoid track, it was defined as off track. Results: We identified 45 adolescents (mean age, 16.1 years) and 30 adults (mean age, 28.9 years) with anterior shoulder dislocations. There was no significant difference in percentage of bone loss between adolescents (mean, 8.4%) and adults (mean, 9.9%; P ¼ .23). There was no significant difference in Hill-Sachs lesion size between adolescents (mean, 12.7 mm) and adults (mean, 9.9 mm; P ¼ .12). There were 12 patients with off-track lesions. Off-track lesions were present in 11 of 45 adolescents (24.4%) and 1 of 30 adults (3.3%). Adolescents had an increased risk of having an off-track lesion (odds ratio, 9.38; 95% confidence interval, 1.14-77.1). A subgroup analysis identified multiple dislocations as an independent risk factor for an off-track lesion (odds ratio, 4.15; 95% confidence interval, 0.85-20.23). Conclusions: This study shows that adolescence and a history of multiple dislocations are independent risk factors for a greater likelihood of glenoid off-track lesions. The findings support the use of bipolar assessment of shoulder dislocators, especially in adolescents and multiple dislocators. Level of Evidence: Level III, retrospective comparative study.
A
dolescence has been identified as the most significant risk factor for recurrent anterior shoulder dislocations despite comprehensive nonoperative treatment.1,2 Similar to the situation in adults,
From the Division of Shoulder and Sports Medicine Surgery, Department of Orthopaedic Surgery, University of California San Francisco Medical Center (B.C.L., D.C., P.F.C., B.T.F., N.K.P.), San Francisco; and University of California San Francisco Medical Center, Children’s Hospital of Oakland (N.K.P.), Oakland, California, U.S.A. The authors report the following potential conflict of interest or source of funding: N.K.P. receives support from Orthopediatrics. Unpaid consultant on devices, none of which pertain to the submitted article. There is no payment provided for services. Received December 2, 2016; accepted April 6, 2017. Address correspondence to Brian C. Lau, M.D., University of California San Francisco Medical Center, 500 Parnassus Ave, MU 320-W, San Francisco, CA 94143, U.S.A. E-mail: [email protected] Ó 2017 by the Arthroscopy Association of North America 0749-8063/161191/$36.00 http://dx.doi.org/10.1016/j.arthro.2017.04.004
shoulder dislocations account for 90% of shoulder instability cases and usually occur after a fall during a sporting activity.3 In adolescent patients, however, there is a 75% to 80% rate of recurrent dislocations with nonoperative treatment.4-6 Outcomes of arthroscopic treatment of shoulder instability in the adolescent population have shown a much higher rate of failure than in the adult population.7,8 Shymon et al.8 reported a failure rate of 21% in the adolescent population within 2 years and a 5-year survival rate of only 49% after shoulder stabilization. This is compared with a failure rate of 8% to 11% in adults after 11 years.7 The reasons for this are not completely known. It may be because of the engagement of these athletes in high-demand collision sports that stress the repair unlike their adult counterparts. Other risk factors for the increased rate of recurrent instability in this population may include the amount of glenoid bone loss after initial dislocation. A degree of
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2017: pp 1-7
1
2
B. C. LAU ET AL.
glenoid bone loss occurs in up to 40% of traumatic anterior shoulder dislocations.9 Another risk factor may be the degree of humeral head injury. The incidence of a humeral head injury (Hill-Sachs lesion) in shoulder instability ranges from 38% to 88%.10,11 In contrast, several studies did not find that larger Hill-Sachs lesion size was a potential cause of recurrence.10-13 These studies, however, did not assess the relation of the HillSachs lesion with glenoid bone loss. More recently, Yamamoto et al.14 identified a model to account for both glenoid and humeral head (bipolar) bone loss: the glenoid track. In this concept, 2 terms were defined to simultaneously characterize the relation of glenoid bone loss and its relation to Hill-Sachs lesions: on track and off track.14 Metzger et al.15 modified the glenoid track model to allow for clinical application and found, through a regression analysis, that younger age and multiple dislocations were predictive of off-track lesions. To date, however, there has not been a direct comparison of the glenoid track concept between adolescent and adult patients with shoulder instability. The purpose of this study was to compare bipolar bone loss by evaluating the degree of glenoid bone loss, Hill-Sachs lesion size, and glenoid track lesions in adolescents and adults with shoulder dislocations. We hypothesized that adolescent patients would have similar glenoid bone loss but greater likelihood of off-track lesions because of the relation with the Hill-Sachs lesion.
Methods After obtaining institutional review board approval (No. 16-19,515), we performed a retrospective review of surgical and nonsurgical patients with a history of primary and recurrent anterior shoulder dislocations over a 4-year period (2012-2016) who underwent magnetic resonance imaging (MRI) with arthrograms of the affected shoulder. Dislocation was defined by an episode requiring a reduction. The exclusion criteria included prior surgery on the affected shoulder, multidirectional instability, and posterior dislocation. Demographic data, sports played, and number of dislocations were collected. The patients were grouped into 2 groups: adolescents (aged 10-19 years) and adults (aged 20 years). We used 20 years of age as a cutoff on the basis of prior studies that have established age, in particular less than 20 years, as a risk factor for recurrent instability.2,16-19 The groups were compared regarding bipolar bone loss by measuring glenoid bone loss and Hill-Sachs injury size as determined on MRI. All patients were evaluated for any glenoid bone loss and Hill-Sachs defects. On the basis of the extent of the bipolar lesion, the glenoid track was defined as outside and engaging the
glenoid on the humeral head (off track) or as inside and non-engaging (on track). MRI Assessment As described by Yamamoto et al.,14 the size of the glenoid track was determined by the amount of glenoid bone loss. When there is no glenoid defect, the width of the glenoid track is 84% of the glenoid width. Yamamoto et al. showed in a cadaveric study that this is the contact point of the glenoid with the arm in 60 of abduction to the medial margin of the rotator cuff footprint on the humerus. When there is a bony defect at the anterior rim of the glenoid, the defect width should be subtracted from the value representing 84% of the length to obtain the true width of the glenoid track.14 If the medial margin of the Hill-Sachs lesion was more medial than the glenoid track, it was defined as off track and more likely to engage and have a higher risk of recurrent instability. Conversely, if the medial margin of the Hill-Sachs lesions was within the glenoid track, it was defined as on track and less likely to engage14 (Fig 1). Our study used the modified glenoid track method as defined by Metzger et al.15 on the basis of clinical magnetic resonance (MR) images performed by a single evaluator. By use of the sagittal-oblique image, the bare spot was identified and the glenoid width was measured using this as the central reference point as described by Huysmans et al.20,21 To determine the expected glenoid width, a best-fit circle was placed on the inferior third of the glenoid centered on the bare spot and the diameter of the circle was measured to calculate the expected width prior to bone loss (Fig 2).20 The glenoid bone loss percentage and glenoid track were both determined from these measurements. The glenoid track was calculated as 84% of the actual glenoid width. In the setting of bone loss, the amount of bone loss was subtracted from 84% of the actual glenoid width to calculate the glenoid track.14 The size of the Hill-Sachs lesion was determined by the method reported by Saito et al.22 It was calculated based on the distance from the medial margin of the rotator cuff footprint to the medial margin of the Hill-Sachs lesions, in millimeters, by use of the largest distance on a coronal image (Fig 2). If this value was greater than the previously determined glenoid track (84% of actual glenoid width minus any bone loss), then the humeral lesion was determined to be off track. If the value was less than the glenoid track (84% of actual glenoid width minus any bone loss), then the humeral lesion was determined to be on track. The described method of combined glenoid bone loss, Hill-Sachs assessment, and glenoid tracking has been shown to have good intraobserver and interobserver reliability and high correlation (>0.89).23
3
BIPOLAR BONE LOSS
Statistical Analysis Descriptive statistics were performed to determine the demographic characteristics of each study population. Student t tests for parametric data were performed to compare the 2 groups: adolescents (aged 10-19 years) and adults (aged 20 years). Adjusted odds ratios were used to determine adolescents’ risk of having off-track lesions. To test intraobserver reliability, the same measurements were performed at 3 months by the same evaluator (B.C.L.). To test interobserver reliability, a second evaluator (P.F.C.) performed imaging measurements using the aforementioned techniques. Pearson correlation coefficients were used to determine reliability. The k statistic was used to test the intraobserver and interobserver reliability of on- and off-track classification.
Results We identified 45 adolescents (aged 10-19 years; mean age, 16.1 years; 12 female adolescents) and 30 adults (aged 20-39 years; mean age, 28.9 years; 4 female adults) (Table 1). There was no significant difference in body mass index between the 2 groups (P ¼ .31). There was no significant difference in glenoid bone loss between adolescents (mean percentage of glenoid bone loss, 8.4%) and adults (mean percentage of glenoid bone loss, 9.9%; P ¼ .23) (Table 2). There was no significant difference in Hill-Sachs lesion size in adolescents (mean size, 12.7 mm) compared with adults (mean size, 9.9 mm; P ¼ .12) (Table 2). There were a total of 12 patients with off-track lesions. Off-track lesions were present in 11 of 45 adolescents (24.4%) and 1 of 30 adults (3.3%) (Table 2). Adolescents had an increased risk of having an off-track lesion (odds ratio, 9.38, 95% confidence interval, 1.14-77.1) (Table 3). A subgroup analysis comparing all patients (adolescents and adults) with 1 dislocation (single) versus those with 2 or more dislocations (multiple) was performed. There was no significant difference in bone loss between single dislocators (mean percentage of glenoid bone loss, 8.4%) and multiple dislocators (mean percentage of glenoid bone loss, 9.5%; P ¼ .47) (Table 4). To determine whether the size of glenoid bone loss correlated with the number of dislocations, a Pearson correlation was performed, with R ¼ 0.25 (P ¼ .83). The multiple-dislocations group had a significantly larger Hill-Sachs lesion size (mean, 14.3 mm) than the single-dislocation group (mean, 8.1 mm; P ¼ .0003) (Table 4). To determine whether the size of Hill-Sachs lesions correlated with the number of dislocations, a Pearson correlation was performed, with R ¼ 0.26 (P ¼ .82). The rate of off-track lesions was 2 of 34 (5.9%) in the single-dislocation group and 10 of 41 (24.4%) in the multiple-dislocations group (Table 3). The presence of multiple dislocations was an independent risk factor for an off-track lesion (odds ratio, 4.15; 95% confidence interval, 0.85-20.23) (Table 4).
Fig 1. The distances from the medial margin of the contact area to the edge of the articular surface of the humeral head (MA) and to the medial margin of the cuff attachment site on the greater tuberosity (MF) are measured. (A, lateral margin of articular surface; C, articular center of humeral head; F, footprint; M, most medial point.) Figure courtesy of Sonali Feeley.
The types of activities that patients participated in are included in Table 1. The most common mechanism of injury for adolescents was football, at 37.7%. The most common mechanism of injury for adults was gym exercises (weight lighting, yoga), at 43.3%. Injuries in adolescents were more associated with contact sports (football, wrestling, basketball), high-risk sports (snowboarding, climbing), or high-energy events (motor vehicle accidents), at 68.9% (31 of 45), compared with adults, at 23.3% (7 of 30). Intraobserver reliability for the percentage of glenoid bone loss and Hill-Sachs lesion size had Pearson correlation coefficients of 0.91 (P ¼ .02) and 0.87 (P ¼ .03), respectively. Interobserver reliability for the percentage of glenoid bone loss and Hill-Sachs lesion size had Pearson correlation coefficients of 0.86 (P ¼ .03) and 0.83 (P ¼ .04), respectively. The k statistics for intraobserver and interobserver classification into on track and off track were 0.86 and 0.81, respectively.
Discussion The principal finding of this study was that adolescent patients with shoulder instability have a greater
4
B. C. LAU ET AL.
Fig 2. The glenoid track is calculated as 84% of the actual glenoid width measured on the sagittal-oblique magnetic resonance image. (A) Sagittal-oblique fluid-sensitive image. A best-fit circle is placed on the glenoid to calculate the expected width prior to bone loss. The percentage of bone loss and glenoid track can be determined. (B) Coronal-oblique fluid-sensitive image. The distance from the medial edge of the rotator cuff footprint to the medial margin of the Hill-Sachs lesion is measured on the coronal magnetic resonance image. A Hill-Sachs lesion width that is greater than the glenoid track is considered outside the glenoid track and deemed off track.
likelihood of having “off-track” shoulder lesions than adult patients with anterior shoulder dislocations. Adolescents had a 9.4 increased risk of having an off-track lesion. A history of multiple dislocations (2) regardless of age was also an independent risk factor, with a 4.2 increased risk of an off-track lesion. These findings suggest that glenoid track assessment should be performed in the adolescent population with a history of shoulder dislocations and in multiple dislocators regardless of age as part of their diagnostic workup.
The key to the glenoid track concept is that the size and location of the Hill-Sachs lesion in relation to the glenoid bone loss are important factors for the risk of engagement and recurrent instability. In this study, glenoid bone loss and Hill-Sachs lesion size were similar between adolescents and adults. This finding highlights that although related to size, the location of the HillSachs lesion is also an important variable to consider. For example, the larger the size of the Hill-Sachs lesion, the more likely it will extend medially on the humeral head and engage the glenoid. Similarly, a smaller
Table 1. Demographic Characteristics n Mean age (range), yr Gender, n Mean BMI (range) Sport or mechanism of injury, n Football Gym exercises (weight lifting or yoga) Wrestling Basketball Baseball or softball Cheerleading Volleyball Soccer Snowboarding or skiing Swimming Surfing Climbing Motor vehicle accident Miscellaneous (lifting baby, moving mattress, towel drying hair, running from bee) BMI, body mass index; F, female; M, male.
Adolescents 45 16.1 (13-19) 33 M and 12 F 25.7 (20.5-33.2) 17 2 5 4 3 1 2 2 4
5
Adults 30 28.9 (20-39) 26 M and 4 F 23.45 (19.9-30.1)
13 1 2 1 2 2 2 1 1 1 4
P Value <.001 .31
5
BIPOLAR BONE LOSS Table 2. Comparison of Glenoid Bone Loss, Hill-Sachs Lesion Size, and Glenoid Tracking in Adolescents Versus Adults n Mean glenoid bone loss Mean expected glenoid diameter, mm Mean resultant glenoid diameter, mm Mean Hill-Sachs lesion size, mm Glenoid tracking, n Off track On track
Adolescents 45 8.4% 25.4 23.3 12.77 11 34
Adults P Value 30 9.9% .67 26.0 .32 23.5 .75 10.0 .12 1 29
NOTE. Adolescents were aged 13 to 19 years, and adults were aged 20 years or older.
Hill-Sachs lesion that is more medialized is also more likely to engage the glenoid (off track) when the arm is placed into abduction and external rotation. Adolescence is a known risk factor for recurrent dislocations during nonoperative management1,2 and after operative management,8 and the findings of this study suggest that off-track lesions may contribute to this increased risk. An off-track lesion was found in 24.4% of adolescent patients, whereas only 3.3% of adults had an off-track lesion. As such, these findings support previous studies’ recommendations on the assessment of bipolar bone lesions in shoulder instability and on planning surgical or conservative treatment, especially in younger patients.14,15,24 The differences between adolescents and adults may also be a result of the injury mechanism. In this study, adolescents were more likely to have an injury during a specific contact athletic event, with football being the most common mechanism, occurring in 37.7% of adolescents. Other high-risk activities that adolescents were more likely to engage in were wrestling, basketball, and snowboarding or skiing. The adult population underwent injury most commonly from gym exercises, such as weight lifting or yoga, at a rate of 43.3%. It is possible that the more traumatic nature of adolescent injuries may result in a higher rate of offtrack lesions. In addition, the findings from this study identified multiple dislocations as a risk factor for having an offtrack glenoid lesion. An interesting finding was that the presence of multiple dislocations did not affect the amount of glenoid bone loss but was associated with
Table 3. ORs of Adolescence and History of Multiple Dislocations Adolescence (age 13-19 yr) History of multiple dislocations
OR 9.38 4.15
CI, confidence interval; OR, odds ratio.
95% CI 1.14-77.1 0.85-20.23
P Value <.01 <.01
Table 4. Comparison of Glenoid Bone Loss, Hill-Sachs Lesion Size, and Glenoid Tracking for Single Versus Multiple Dislocations
n Adolescents/adults, n No. of patients with 2 dislocations No. of patients with 3 dislocations No. of patients with 4 dislocations No. of patients with 5 dislocations No. of patients with 7 dislocations Mean glenoid bone loss Mean Hill-Sachs lesion size, mm Glenoid tracking, n Off track On track
Single Multiple P Dislocation Dislocations Value 34 41 21/13 24/17 20 11 4 5 1 8.4% 9.5% .47 8.1 14.3 .0003 2 32
10 31
NOTE. The single-dislocation group had 1 dislocation before magnetic resonance imaging, whereas the multiple-dislocations group had 2 or more dislocations before magnetic resonance imaging.
larger Hill-Sachs lesion size. A history of multiple dislocations is a known risk factor for recurrent instability because of possible effects on the articular cartilage and soft tissues,25 and the findings from this study suggest that subsequent dislocations’ effect on humeral head and glenoid tracking may be another etiology for increased risk of recurrent instability. It is also important to note that in this study, the frequency of recurrent dislocations did not correlate with Hill-Sachs lesion size. A possible explanation is that the increased HillSachs defect occurs during the second dislocation and that subsequent dislocations may not increase the defect. However, it is also possible that a larger number of patients or a population with a larger range of dislocations would better bear out this difference. The natural history of glenohumeral dislocations is that adolescents have a greater likelihood, up to 75% to 80%, of recurrent dislocations5,6,26,27; however, it is important to note that in this study, the adolescent and adult patients were relatively equally distributed in the single-dislocation and multiple-dislocations groups. Among adolescents, 46.7% had a single dislocation and 53.3% had multiple dislocations. Among adults, 43.3% had a single dislocation and 56.7% had multiple dislocations. The findings from this study suggest that subsequent dislocation events have a greater effect on the humeral head than on the glenoid regardless of age. Although traditional data have shown that there is greater glenoid loss with recurrent dislocations,28 some recent data have suggested that recurrent dislocations may have a greater effect on Hill-Sachs lesions. Kim et al.29 compared patients with a single dislocation and patients with recurrent dislocations and found that compared with single dislocators, a larger percentage of recurrent dislocators had Hill-Sachs lesions (37% more)
6
B. C. LAU ET AL.
and glenoid erosion (13.5% more). A recent study compared recurrent dislocators and patients with recurrent subluxation without frank dislocation and also showed a similar presence of glenoid erosion between the groups but a significant increase in Hill-Sachs lesions in recurrent dislocators.30 It remains unclear whether both the glenoid and humeral head defects need to be treated in off-track lesions. Clinicians may choose to perform an isolated Bankart repair, glenoid bone grafting, or Hill-Sachs treatment with remplissage or bone grafting. A recent biomechanical study of off-track lesions found that Bankart repair prevented engagement of 67% and 0% of shoulders in 60 and 90 of external rotation, respectively.31 In contrast, adding Hill-Sachs remplissage to Bankart repair prevented engagement in all shoulders, 100%, at 60 and 90 of external rotation. The authors noted that Bankart repair with remplissage resulted in shoulder stiffness but was necessary to prevent engagement of off-track lesions. Di Giacomo et al.32 developed a treatment algorithm based on the percentage of glenoid loss and glenoid track classification. Glenoid treatment was determined independently of the Hill-Sachs lesion, with less than 25% bone loss being treated with arthroscopic Bankart repair and more than 25% bone loss being treated with the Latarjet procedure. The authors suggested that Hill-Sachs lesion treatment be based on glenoid track classification, with only off-track lesions warranting the remplissage procedure. They also suggested possible humeral bone grafting in off-track lesions with large glenoid defects (>25% bone loss). Limitations This study has several limitations. MRI was chosen in this study for a number of reasons. First, MR arthrography imaging is the most common advanced imaging modality for shoulder instability to evaluate labral and other soft-tissue injuries for surgical planning and therefore may be a more generalizable imaging modality. MR arthrograms have been shown to have sensitivity and specificity of between 86% and 98% in identifying labral and cartilage injury.33 Second, ionizing radiation exposure risk in the adolescent population has led to less use of computed tomography (CT) scans in this at-risk population. Last, MRI assessment of glenoid bone loss has been shown to be as accurate as CT scans in recent studies.15,21,34 We believe that MR arthrography imaging was the most practical and generalizable imaging modality to assess bone loss, but we recognize that 3-dimensional CT scanning is also highly accurate. Of note, MR arthrograms are more sensitive than non-arthrogram MRI for bone loss.33 Our study population mostly comprised male patients, which may not be representative of the general population. However, prior studies have shown
that young male patients are the population most at risk of shoulder dislocations after a traumatic sporting event.35 This study found that younger patients engaged in higher-risk activities including contact sports at a greater rate than adults, which may affect the rate of bipolar injury. The body mass indexes of the 2 groups were similar, which shows that the physical sizes of the patients were similar, but given the inherent differences in age-related activities, such as activity level, high-risk activities, or contact sports, it is difficult to control for these factors. In addition, our study found no significant difference in Hill-Sachs lesion size between adolescents and adults, although the average size was larger in the adolescent group. It is possible that there is a true difference that this study was underpowered to detect. Interobserver and intraobserver reliability for percentage of glenoid bone loss and Hill-Sachs lesion size ranged from 0.83 to 0.91, which is similar to interobserver and intraobserver reliabilities shown by prior studies using MRI with Pearson correlation coefficients.23 However, it is important to note that prior studies have also identified variability of glenoid track classification. A recent study showed that glenoid track classification had 72% agreement, with the Hill-Sachs measurement having a high level of variability, with a coefficient of variability of 19.2%.36 Our study had k statistics of 0.86 for intraobserver glenoid track classification and 0.81 for interobserver classification.
Conclusions This study shows that adolescence and a history of multiple dislocations are independent risk factors for a greater likelihood of glenoid off-track lesions. The findings support the use of bipolar assessment of shoulder dislocators, especially in adolescents and multiple dislocators.
Acknowledgment The authors acknowledge Sonali Feeley for making Figure 1.
References 1. Flinkkilä T, Hyvönen P, Ohtonen P, Leppilahti J. Arthroscopic Bankart repair: Results and risk factors of recurrence of instability. Knee Surg Sports Traumatol Arthrosc 2010;18:1752-1758. 2. te Slaa RL, Wijffels MP, Brand R, Marti RK. The prognosis following acute primary glenohumeral dislocation. J Bone Joint Surg Br 2004;86:58-64. 3. Mazzocca AD, Brown FM Jr, Carreira DS, Hayden J, Romeo AA. Arthroscopic anterior shoulder stabilization of collision and contact athletes. Am J Sports Med 2005;33:52-60. 4. Arciero RA, Wheeler JH, Ryan JB, McBride JT. Arthroscopic Bankart repair versus nonoperative treatment for
BIPOLAR BONE LOSS
5.
6.
7. 8.
9. 10.
11.
12.
13.
14.
15.
16. 17. 18.
19.
20.
21.
acute, initial anterior shoulder dislocations. Am J Sports Med 1994;22:589-594. Bottoni CR, Wilckens JH, DeBerardino TM, et al. A prospective, randomized evaluation of arthroscopic stabilization versus nonoperative treatment in patients with acute, traumatic, first-time shoulder dislocations. Am J Sports Med 2002;30:576-580. Deitch J, Mehlman CT, Foad SL, Obbehat A, Mallory M. Traumatic anterior shoulder dislocation in adolescents. Am J Sports Med 2003;31:758-763. Harris JD, Gupta AK, Mall NA, et al. Long-term outcomes after Bankart shoulder stabilization. Arthroscopy 2013;29:920-933. Shymon SJ, Roocroft J, Edmonds EW. Traumatic anterior instability of the pediatric shoulder: A comparison of arthroscopic and open Bankart repairs. J Pediatr Orthop 2015;35:1-6. Postacchini F, Gumina S, Cinotti G. Anterior shoulder dislocation in adolescents. J Shoulder Elbow Surg 2000;9:470-474. Rowe CR, Zarins B, Ciullo JV. Recurrent anterior dislocation of the shoulder after surgical repair. Apparent causes of failure and treatment. J Bone Joint Surg Am 1984;66:159-168. Yiannakopoulos CK, Mataragas E, Antonogiannakis E. A comparison of the spectrum of intra-articular lesions in acute and chronic anterior shoulder instability. Arthroscopy 2007;23:985-990. Rowe CR, Patel D, Southmayd WW. The Bankart procedure: A long term end-result study. J Bone Joint Surg Am 1978;60:1-16. Ungersbock A, Michel M, Hertel R. Factors influencing the results of a modified Bankart procedure. J Shoulder Elbow Surg 1995;4:365-369. Yamamoto N, Itoi E, Abe H, et al. Contact between glenoid and the humeral head in abduction, external rotation, and horizontal extension: A new concept of glenoid track. J Shoulder Elbow Surg 2007;16:649-656. Metzger PD, Barlow B, Leonardelli D, Peace W, Solomon DJ, Provencher MT. Clinical application of the “glenoid track” concept for defining humeral head engagement in anterior shoulder instability. Orthop J Sports Med 2013;1:2325967113496213. Henry JH, Genung JA. Natural history of glenohumeral dislocation revisited. Am J Sports Med 1982;10:135-137. McLaughlin H, Cavallaro WU. Primary anterior dislocation of the shoulder. Am J Surg 1950;80:615-621. passim. Rowe CR, Sakellarides HT. Factors related to recurrences of anterior dislocations of the shoulder. Clin Orthop Relat Res 1960;20:40-48. Wasserstein DN, Sheth U, Colbenson K, et al. The true recurrence rate and factors predicting recurrent instability after nonsurgical management of traumatic primary anterior shoulder dislocation: A systematic review. Arthroscopy 2016;32:2616-2625. Huysmans PE, Haen PS, Kidd M, Dhert WJ, Willems JW. The shape of the inferior part of the glenoid: A cadaveric study. J Shoulder Elbow Surg 2006;15:759-763. Huijsmans PE, Haen PS, Kidd M, Dhert WJ, van der hulst VP, Willems JW. Quantification of a glenoid defect
22.
23.
24.
25. 26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
7
with three-dimensional computed tomography and magnetic resonance imaging: A cadaveric study. J Shoulder Elbow Surg 2007;16:803-809. Saito H, Itoi E, Mingawa H, Yamamoto N, Tuoheti Y, Seki N. Location of the Hill-Sachs lesion in shoulders with recurrent anterior dislocation. Arch Orthop Trauma Surg 2009;129:1327-1334. Shaha JS, Cook JB, Rowles DJ, Bottoni CR, Shaha SH, Tokish JM. Clinical validation of the glenoid track concept in anterior glenohumeral instability. J Bone Joint Surg Am 2016;98:1918-1923. Omori Y, Yamamoto N, Koishi H, et al. Measurement of the glenoid track in vivo as investigated by 3-dimensional motion analysis using open MRI. Am J Sports Med 2014;42: 1290-1295. Brophy RH, Marx RG. Osteoarthritis following shoulder instability. Clin Sports Med 2005;24:47-56. Roberts SB, Beattie N, McNiven ND, Robinson CM. The natural history of primary anterior dislocation of the glenohumeral joint in adolescence. J Bone Joint Br 2015;97:520-526. Eljabu W, Klinger HM, von Knoch M. The natural course of shoulder instability and treatment trends: A systematic review. J Orthop Traumatol 2017;18:1-8. Sugaya H, Moriishi J, Dohi M, Kon Y, Tsuchiya A. Glenoid rim morphology in recurrent anterior glenohumeral instability. J Bone Joint Surg Am 2003;85:878-884. Kim DS, Yoon YS, Yi CH. Prevalence comparison of accompanying lesions between primary and recurrent anterior dislocation in the shoulder. Am J Sports Med 2010;38:2071-2076. Shin SJ, Ko YW, Jeon YS, Lee J, Kim RG, Baek H. Comparison of intra-articular findings and clinical features between patients with symptomatic anterior instability after recurrent shoulder subluxation and dislocation. Arthroscopy 2017;33:527-533. Hartzler RU, Bui CN, Jeong WK, et al. Remplissage of an off-track Hill-Sachs lesion is necessary to restore biomechanical glenohumeral joint stability in a bipolar bone loss model. Arthroscopy 2016;32:2466-2476. Di Giacomo G, Itoi E, Burkhart SS, et al. Evolving concept of bipolar bone loss and the Hill-Sachs lesion: From “engaging/non-engaging” lesion to “on-track/off-track” lesion. Arthroscopy 2014;30:90-98. Steinbach LS. MRI of shoulder instability. Eur J Radiol 2008;68:57-71. Tian CY, Shang Y, Zheng ZZ. Glenoid bone lesions: Comparison between 3D VIBE images in MR arthrography and nonarthrographic MSCT. J Magn Reson Imaging 2012;36:231-236. Owens BD, Duffey ML, Nelson BJ, et al. The incidence and characteristics of shoulder instability at the United States Military Academy. Am J Sports Med 2007;35: 1168-1173. Schneider AK, Hoy G, Ek ET, et al. Interobserver and intraobserver variability of glenoid track measurements. J Shoulder Elbow Surg 2017;26:573-579.