Introduction

Recurrent anterior shoulder dislocation was seen commonly in younger patients. Although less frequent, we could also see some older patients with recurrent anterior shoulder dislocation. The majority of published studies have shown that the most common pathology of older patients with recurrent anterior shoulder dislocation was rotator cuff tears. The prevalence of full-thickness rotator cuff tears of the patient with recurrent anterior shoulder dislocation aged over 40 years was from 30 to 90 % [1–4]. Most studies tended to focus on rotator cuff tears as a significant pathology of older patients with recurrent anterior shoulder dislocation. But we should remember that an isolated simple rotator cuff tear cannot cause anterior dislocation and the essential pathology of shoulder dislocations is functional deficiency of the anterior capsular mechanism [5] which provides anterior shoulder stability.

There have been some reports on pathologies of the anterior capsular mechanism in older patients. Some authors reported that a Bankart lesion was observed at the same rate as in younger patients [6, 7], while others reported that midsubstance complete tear of the inferior glenohumeral ligament (capsular tear) or humeral avulsion of the glenohumeral ligament (HAGL lesion) were observed with rotator cuff tears [8, 9]. However, most reports included patients with rotator cuff tears and the details of the pathologies of the anterior capsular mechanism without rotator cuff tears have not been elucidated.

Therefore, in this study, in order to eliminate influences of rotator cuff tears, we excluded those patients with full-thickness rotator cuff tears and evaluated the pathologies focusing on the anterior capsular mechanisms in older patients with anterior shoulder instability. The purpose of this study was to assess the characteristics of the pathology of the anterior capsular mechanism in older patients with recurrent anterior shoulder dislocation in the absence of rotator cuff tears.

Materials and methods

Between 2002 and 2007, 400 shoulders with recurrent anterior shoulder dislocation underwent arthroscopic treatment at our institution. Excluding five shoulders with full-thickness rotator cuff tears, 395 shoulders (283 men and 112 women) were the subjects of this study. In order to assess differences of pathologies between older patients with late-onset, older patients with early-onset and younger patients, the patients were divided into three groups by the age at the time of operation and the age at the time of the initial dislocation. A cut-off line of the age for separation was set at 40 years as previously applied in the same kind of studies [2–4, 6, 8]. Group A consisted of patients who experienced the first dislocation at 40 years of age or older and underwent surgical treatment. Group B consisted of patients who experienced the first dislocation under 40 years old and underwent surgical treatment over 40 years old. Group C consisted of patients who experienced the first dislocation under 40 years old and underwent surgical treatment under 40 years old. Groups A, B and C involved nine, 31 and 355 shoulders, respectively (Table 1). The average age at the time of operation was 58.1 (40–72) years in group A, 49.7 (40–64) years in group B and 23.3 (11–39) years in group C. The average age at the time of initial dislocation was 53.6 (40–70) years in group A, 23.1 (9–39) years in group B and 18.6 (9–36) years in group C.

Table 1 Patient profiles
Full size table

An arthroscopic examination was performed for all patients. We evaluated the pathology of the anterior capsular mechanism and assessed the existence of Bankart lesions, midsubstance complete capsular tears [10] and HAGL lesions [11]. We defined as a midsubstance complete capsular tear when an interruption of the fibres of the inferior glenohumeral ligament was observed and the muscle belly of the subscapularis was recognised through the interrupted area (Fig. 1) [10]. A HAGL lesion was defined as when the anterior capsular ligamentous complex avulsed from the humeral head and the muscle belly of the subscapularis was recognised through the avulsed area (Fig. 2).

Fig. 1
figure 1

Midsubstance complete capsular tear. Interruption of the capsular fibres was observed and the muscle belly of the subscapularis was recognised through the interrupted area. IGHL inferior glenohumeral ligament

Full size image
Fig. 2
figure 2

HAGL lesion. The anterior capsular ligamentous complex avulsed from the humeral head and the muscle belly of the subscapularis was recognised through the avulsed area. IGHL inferior glenohumeral ligament

Full size image

The prevalence of the pathologies of the anterior capsular mechanism was compared between the three groups. We used Fisher’s test for statistical analysis and P < 0.05 was set as the significance level.

Results

A Bankart lesion was found in five shoulders of group A (55.6 %), in 27 shoulders of group B (87.1 %) and in 337 shoulders of group C (94.9 %). A midsubstance complete capsular tear was found in five shoulders of group A (55.6 %), in nine shoulders of group B (29 %) and in 54 shoulders of group C (15.2 %). A HAGL lesion was found in one shoulder of group A (11.1 %), in 2 shoulders of group B (6.5 %) and in 11 shoulders of group C (3.1 %).

According to the detailed assessment of all pathologies in group A (Fig. 3), an isolated Bankart lesion was found in three shoulders (33.3 %), a Bankart lesion associated with a capsular tear was found in two shoulders (22.2 %), a Bankart lesion associated with a HAGL lesion was not found (0 %), an isolated capsular tear was found in three shoulders (33.3 %) and an isolated HAGL lesion was found in one shoulder (11.1 %). In group B, an isolated Bankart lesion was found in 20 shoulders (64.5 %), a Bankart lesion associated with a capsular tear was found in seven shoulders (22.6 %), a Bankart lesion associated with a HAGL lesion was found in one shoulder (3.2 %), an isolated capsular tear was found in two shoulders (6.5 %) and an isolated HAGL lesion was found in one shoulder (3.2 %). In group C, an isolated Bankart lesion was found in 290 shoulders (81.7 %), a Bankart lesion associated with a capsular tear was found in 43 shoulders (12.1 %), a Bankart lesion associated with a HAGL lesion was found in four shoulders (1.1 %), an isolated capsular tear was found in 11 shoulders (3.1 %) and an isolated HAGL lesion was found in seven shoulders (2.0 %).

Fig. 3
figure 3

Total pathologies of each group. The prevalence of isolated Bankart lesions in groups A and B was lower and that of capsular tears in group A was higher than in group C

Full size image

We compared the prevalence of the pathologies between the three groups. An isolated Bankart lesion was seen in 81.7 % of group C (younger group), while it was seen in 64.5 % of group A and in 33.3 % of group B; both groups A and B were significantly lower (P = 0.0023 and P = 0.0317, respectively) than group C. The prevalence of an isolated capsular tear in the younger group was 3.1 %, while in group B it was 33.3 %, which was significantly higher (P = 0.0033) (Fig. 4). Other pathologies did not have significant differences between each of the groups.

Fig. 4
figure 4

Comparison of prevalence of pathologies between the three groups. The prevalence of the shoulders with an isolated Bankart lesion in the younger group was 81.7 %, while in group A it was 64.5 % and in group B it was 33.3 %, which were significantly lower (P = 0.0023 and P = 0.0317. respectively). The prevalence of an isolated capsular tear in the younger group was 3.1 %, while in group B it was 33.3 %, which was significantly higher (P = 0.0033). Other pathologies did not have significant differences between each of the groups

Full size image

Discussion

There were some reports on pathologies of the anterior capsular mechanism of recurrent anterior shoulder dislocation in older patients. Araghi et al. investigated 11 shoulders with recurrent anterior shoulder dislocation with onset over 40 years of age. They reported that anterior capsular lesions were observed in all shoulders, and nine of 11 shoulders had anterior capsulolabral detachment, four shoulders had a rotator interval defect, two shoulders had anteroinferior capsular redundancy and one shoulder had a HAGL lesion associated with a small rotator cuff tear [6]. Ikegami et al. reported ten shoulders with recurrent anterior shoulder dislocation in patients aged 50 years or older who underwent surgical treatment. They stated that the pathology with onset under 50 years of age was Bankart lesions and that with onset over 50 years of age was massive rotator cuff tears associated with Bankart lesions [7]. They concluded that Bankart lesions were mainly observed in young patients.

On the other hand, Neviaser and Neviaser reported on 11 shoulders with recurrent anterior shoulder dislocation with onset over 40 years of age and observed subscapularis tendon tears with avulsion of the anterior capsule from the lesser tuberosity but no Bankart lesion in any shoulders [8]. Okamura et al. reported on nine shoulders with recurrent anterior shoulder dislocation with onset over 50 years of age. They stated that severe functional deficit of the anterior stabilising structures such as a tear of both subscapularis tendon and anterior band of the inferior glenohumeral ligament was a characteristic of pathologies in older patients [9]. They concluded that the existence of capsular injury should be remembered as a pathological condition of older patients. All of them also noted that surgical treatment for each lesion of the anterior capsular mechanism was successful. But their studies consisted of a small number of patients including those with rotator cuff tears and the details of the pathologies of the anterior capsular mechanism had not been identified. Therefore, only those patients with recurrent anterior shoulder dislocation without rotator cuff tears were included and examined focusing on the pathologies of the anterior capsular mechanism in this study.

Among younger patients (group C) in this study, the main pathology of recurrent shoulder dislocation was a Bankart lesion, whereas a capsular tear and HAGL lesion rarely occurred. Comparing groups A and C, the prevalence of an isolated Bankart lesion in group A was significantly lower, and that of an isolated capsular tear in group A was significantly higher than in group C. Comparing groups B and C, the prevalence of an isolated Bankart lesion in group B was significantly lower than in group C. The prevalence of a Bankart lesion associated with a capsular tear in group B tended to be higher than in group C, but there was no significant difference.

From these results, we could speculate that in older patients with recurrent anterior shoulder dislocation with onset under 40 years of age such as group B, a Bankart lesion occurred at a young age, and a capsular tear occurred in the period of repetitive dislocations. On the other hand, with regard to recurrent anterior shoulder dislocations developed at over 40 years of age such as group A, we should consider the existence of a capsular tear as a main pathology.

Against this background, the vulnerability of the anterior capsular ligamentous complex due to aging was considered to be the cause of capsular tears. Reeves performed experiments on the tensile strength of the anterior capsular mechanism of the shoulder. He reported that the weakest point was the attachment of the glenoid labrum in the younger subjects, whereas in older people calcification of tissues made the midsubstance of the capsule and subscapular tendon weaker. He supposed that both capsular rupture and subscapularis tendon damage occurred at the time of an acute dislocation in older people [12]. Lee et al. also performed the tensile strength test of the inferior glenohumeral ligament targeting younger and older shoulders. They reported that the inferior glenohumeral ligament tore at the attachment of the glenoid labrum in the younger shoulders (the average age was 38 years) and tore on midsubstance in the older shoulders (the average age was 83 years), as seen in the report by Reeves. They also noted that the impaired strength of the inferior glenohumeral ligament of the older shoulders was only 46 % of that of the younger shoulders [13].

This study showed that we could speculate there were two types of characteristics in older patients with recurrent shoulder dislocation in the absence of rotator cuff tears. One of them was that those patients with onset under 40 years of age had a Bankart lesion in the early stage and then associated with a capsular tear due to repetitive dislocations during a long period. The other was that those patients with onset over 40 years of age were injured with a capsular tear in the acute stage and then changed over to recurrent shoulder dislocation. From this study, if we treat older patients with recurrent anterior shoulder dislocation in the absence of rotator cuff tears, we should keep in mind the existence of a capsular tear and evaluate the whole anterior capsular mechanism meticulously.

There are several limitations to this study. This was retrospective review and we did not evaluate the pathologies of non-operatively treated patients. The small number of patients and differences of male to female ratios made statistical analysis of data difficult. Furthermore, we did not evaluate the outcome of treatment of these patients [14]. We should do further studies in order to determine if our findings have true clinical relevance.

Conclusions

This study showed that as characteristics of recurrent anterior shoulder dislocation in older patients in the absence of rotator cuff tears, the prevalence of an isolated Bankart lesion was lower and that of isolated and associated capsular tears was higher than those of younger patients. So we should keep in mind the existence of a capsular tear and evaluate the whole anterior capsular mechanism meticulously.