Open Latarjet reduces residual apprehension, redislocation and possibility of dislocation arthropathy compared to arthroscopic Bankart repair despite greater bipolar bone loss in anterior glenohumeral instability

Abstract

Background

Arthroscopic Bankart repair (ABR) and the open Latarjet (OL) procedure are the most frequently preferred methods in the treatment of anterior glenohumeral instability. The aim of this study was to compare patients who underwent ABR or OL due to anterior glenohumeral instability in terms of functional capacity, glenohumeral bone loss, residual apprehension, redislocation, and dislocation arthropathy.

Methods

A total of 56 patients who underwent ABR or OL due to anterior glenohumeral instability between January 2018 and December 2021 were evaluated retrospectively. There were 32 patients in the ABR group and 24 patients in the OL group. Patients’ demographic characteristics, number of preoperative dislocations, and return-to-work, and follow-up periods were recorded. Glenoid bone loss, Hill–Sachs interval, and Hill–Sachs depth were measured from preoperative computed tomography sections. The American Shoulder and Elbow Surgeons score, the Western Ontario Shoulder Instability Index score, Rowe score, and joint range of motion were calculated. Patients were also asked about residual apprehension, postoperative redislocations, dislocation arthropathy, and surgery satisfaction.

Results

The mean age of the ABR group was 22.5 ± 3.9 years (28 men, 4 women), while that of the OL group was 25.0 ± 4.8 years (22 men, 2 women). The mean number of dislocations was 2.7 ± 1.3 in the ABR and 10.9 ± 5.5 in the OL (P = .001). Higher values of glenoid bone loss (%) (ABR: 6 ± 2; OL: 20 ± 4), Hill–Sachs interval (mm) (ABR: 8 ± 3; OL: 21 ± 3), and Hill–Sachs depth (mm) (ABR: 5 ± 2; OL: 8 ± 2) were measured in the OL (P = .001 for all), reflecting significantly more bone loss. Residual apprehension was detected in 10 patients in the ABR and 2 patients in the OL (P = .007). While 4 patients in the ABR had a history of redislocation, no redislocation occurred in the OL (P = .012). Dislocation arthropathy development was observed in 9 patients in the ABR and 4 patients in the OL (P = .038), according to the modified Samilson and Prieto classification. External rotation in adduction and external rotation in 90° abduction were approximately 5° higher in the OL (P = .011 and P = .016, respectively).

Conclusion

The ABR and OL methods both provide satisfactory outcomes in the treatment of anterior glenohumeral instability with appropriate indications. The OL procedure is preferred for patients with more dislocations and greater bipolar bone loss. Despite greater bipolar bone loss, the OL procedure provides lower rates of residual apprehension, redislocation, and dislocation arthropathy. Additionally, due to the stability it provides, there is less loss in external rotation.

The glenohumeral joint is the most frequently dislocated joint in the human body with an incidence rate of approximately 0.2/1000 person-years.^23,25 Traumatic anterior glenohumeral instability is the most common subtype and accounts for 90% of all glenohumeral instability.²⁶ Traumatic anterior shoulder instability is frequently associated with soft tissue and osseous injuries, such as anteroinferior glenoid labrum tears (Bankart lesions) and posterosuperior humeral head impaction (Hill–Sachs lesions).^7,15,40

After the first anterior shoulder dislocation, conservative treatment largely fails because the recurrence rate is approximately 90%, especially in young male patients.^20,31 Arthroscopic Bankart repair (ABR) and open Latarjet (OL) procedures are the most frequently preferred treatment methods for anterior glenohumeral instability.^41,42 The Instability Severity Index score (ISI score), defined by Balg and Boileau, considers the patient’s clinical and radiological risk factors simultaneously and provides guidance on the optimal surgical treatment to be selected.³ Although much research has been conducted over the years regarding the clinical outcomes of both of these techniques, there is some controversy regarding the ideal treatment choice. In a study conducted by Loppini et al with 670 patients to confirm the validation of the ISI score, while ABR gave successful results for patients with ISI score values of ≤3 and the OL procedure was found to be preferable for patients with ISI score values of >6, no definitive conclusion was drawn regarding ideal treatment for patients with ISI score values of 4-6.¹⁸

ABR is generally applied for patients who have minimal or no glenoid bone loss, and significant improvements are provided in terms of pain and function.^6,34 On the other hand, OL is preferred in revision cases where glenoid bone loss is greater than 13.5% and ABR failure has occurred.^27,36 The fact that OL provides lower rates of recurrent instability, apprehension, and revision surgery has led some surgeons to adopt OL as the primary surgical procedure.^4,13,42

The aim of this study was to compare patients who underwent ABR and OL procedures due to anterior glenohumeral instability in terms of functional capacity, residual apprehension, redislocation, and dislocation arthropathy. Moreover, more attention should be paid to the surgery method (ABR or OL) that has a higher risk of residual apprehension, redislocation, and dislocation arthropathy that may occur in the early follow-up period.

Materials and methods

The study protocol was approved by the relevant institutional review board (date: 29/11/2023; number: E1-23-4349). This is a retrospective cohort study conducted on patients who underwent ABR or OL procedures due to anterior glenohumeral instability. Patients who underwent ABR or OL due to anterior glenohumeral instability between January 2018 and December 2021 were evaluated retrospectively. The inclusion criteria were patients having undergone primary ABR or OL due to anterior glenohumeral instability with regular follow-up for at least two years. The exclusion criteria were multidirectional glenohumeral instability, remplissage procedures, revision surgeries, and lack of regular follow-up (Fig. 1). A total of 56 patients who fulfilled these criteria were included in the study. The demographic characteristics of the patients, number of preoperative dislocations, and follow-up and return-to-work times were recorded. Return-to-work times were evaluated in two different groups as high labor and low labor, depending on the effort the patients made in their work.

Figure 1.

Flowchart of the study. ABR, arthroscopic Bankart repair; OL, open Latarjet.

Surgical techniques

When deciding on the surgical technique, glenoid bone loss was primarily taken into consideration. If the glenoid bone loss was less than 13.5%, the ABR procedure was preferred, and if the glenoid bone loss was more than 13.5%, the OL procedure was preferred.³⁵ In almost all patients who underwent the OL procedure, the Hill–Sachs lesion was larger in parallel with the glenoid bone loss. Although bipolar bone loss was found, all of these were on-track lesions and therefore no remplissage procedure was performed in any case.⁵

Arthroscopic Bankart repair

All surgeries were performed by the same senior surgeon. ABR was performed with patients under general anesthesia and in the beach-chair position. A posterior portal was used for visualization, and anterosuperior and anteroinferior (5 o’clock) portals were used as the working portals. After portal placement and diagnostic arthroscopy, the glenoid rim and anterior capsulolabral complex were renewed with a shaver and rasp. A traction suture was temporarily placed to visualize the capsulolabral complex more clearly. Three suture anchors were then placed on the glenoid rim at the 5, 4, and 2 o’clock positions for the right shoulder. The capsulolabral complex was elevated with the help of the traction suture and the sutures were passed through it using a bird-beak suture passer. The sutures were then tied with the sliding suture technique, and the Bankart repair was completed.

OL procedure

All surgeries were performed by the same senior surgeon. The OL procedure was performed according to the technique developed by Walch and Boileau³⁸ with patients under general anesthesia and in the beach-chair position. An incision of approximately 5 cm was made from the coracoid process to the axillary fold. The coracoacromial ligament and pectoralis minor insertion were then released from the coracoid. The coracoid process was harvested to a length of at least 20 mm with the help of an osteotome, perpendicular to the coracoid body. The lower surface of the coracoid was decorticated with the help of a high-speed burr. The subscapularis was split horizontally at the distal 1/3 level. After the capsule was horizontally split in a similar manner, the anteroinferior glenoid rim was decorticated with the help of a high-speed burr. The coracoid graft was placed at the 2-5 o’clock position relative to the right shoulder, flush to the glenoid surface, and fixed with 2 partially threaded cannulated screws of 4.5 mm. The subscapularis tendon was closed with 2 or 3 stitches.

Rehabilitation

The shoulder joint was kept in a sling for 6 weeks. Passive controlled pendulum movements were started on the 2nd week. At the 4th week, passive forward flexion, abduction, and internal rotation movements were performed. External rotation was not allowed for 6 weeks. In the 6th week, active joint range of motion (ROM) exercise was started. In the 12th week, patients were trained for strengthening.

Radiological assessment

Bone loss assessment was performed by a senior surgeon via preoperative computed tomography imaging. Glenoid width, estimated glenoid bone loss, Hill–Sachs depth, and Hill–Sachs interval (width) were measured, as previously described.^5,10 While the Hill–Sachs interval and Hill–Sachs depth were calculated from axial sections, glenoid bone loss measurements were made specifically as a percentage based on three-dimensional reconstructions.²⁹ Plain shoulder radiographs were used to evaluate the grade and progression of dislocation arthropathy according to the modified Samilson and Prieto classification.³² Additionally, coracoid graft position and union were assessed via computed tomography in the OL group at the final follow-up.³³

Clinical assessment

The patients were followed-up regularly every 3 weeks in the first 3 months after surgery, every 6 weeks in the next 3 months, and every 3 months after the postoperative 6th month. Patients were evaluated with the American Shoulder and Elbow Surgeons (ASES) score, the Western Ontario Shoulder Instability Index (WOSI) score, and the Rowe score at the final follow-up period. Degrees of forward flexion, abduction, internal rotation, external rotation in adduction, and external rotation in 90° abduction were measured to evaluate the joint ROM at the final follow-up. ROM was measured using a goniometer by two orthopedic surgeons independently of each other. Then, the joint ROM degree was calculated by taking the average of the two measurements. The presence of residual apprehension was determined via apprehension test by the senior surgeon.¹⁶ Patients were asked about postoperative redislocation at the final follow-up visit. Finally, patients rated their satisfaction with the surgery as 1 (not satisfied), 2 (moderate), 3 (good), or 4 (excellent) at the final follow-up period.

Statistical analysis

Statistical power analysis was performed to calculate the sample size. Statistical analysis was performed using IBM SPSS Statistics 25.0 (IBM Corp., Armonk, NY, USA). If continuous variables were normally distributed, they were described as mean ± standard deviation values (P > .05 in the Kolmogorov–Smirnov or Shapiro–Wilk test (n < 30)), and if continuous variables were not normally distributed, they were described as median values. Comparisons between groups were conducted using the Student t-test for normally distributed data and the Mann-Whitney U test for data not normally distributed. Categorical variables were analyzed between groups using the chi-square test. Bonferroni correction was used to correct for multiple P value comparisons. Values of P < .05 were considered statistically significant.

Results

The demographic characteristics of the patients, mean number of preoperative dislocations, and follow-up and return-to-work periods (high labor and low labor) are presented in Table I. The patients were predominantly male in both groups and were similar in terms of the involved side. When compared in terms of the mean number of dislocations before surgery, the OL group had a rate approximately 4 times higher than that of the ABR group and this difference was significant (P = .001). The groups were similar in terms of mean age, time until returning to work, and follow-up period, with no significant differences observed in this regard.

Table I.

Demographic characteristics of the patients, preoperative dislocations, return-to- work, and follow-up times.

	ABR	OL	P
	n (%)	n (%)
Gender
Female	4 (12.5)	2 (8.3)	.628
Male	28 (87.5)	22 (91.7)
Side
Right	18 (56.2)	15 (62.5)	1.000
Left	14 (43.8)	9 (37.5)
Dominant side
Right	26 (81.3)	20 (83.3)	1.000
Left	6 (18.7)	4 (16.7)
Age (year)
Mean ± SD	22.5 ± 3.9	25.0 ± 4.8	.070
Median (Min-Max)	22 (17-31)	24 (17-33)
Dislocations
Mean ± SD	2.7 ± 1.3	10.9 ± 5.5	.001∗
Median (Min-Max)	2 (1-5)	11 (3-21)
Return to work (week)
High labor
Mean ± SD	14.3 ± 3.4	15.1 ± 4.1	.435
Median (Min-Max)	14 (8-18)	15 (9-22)
Low labor
Mean ± SD	9.2 ± 2.7	10.3 ± 2.6	.237
Median (Min-Max)	9 (7-15)	12 (7-16)
Follow-up time (mo)
Mean ± SD	50.5 ± 12.7	48.9 ± 11.4	.620
Median (Min-Max)	48 (24-66)	46 (24-62)

ABR, arthroscopic Bankart repair; OL, open Latarjet; n, number; SD, standard deviation.

^∗Statistically Significant.

The mean values obtained as a result of radiological measurements are shown as boxplots in Figure 2. Higher values of glenoid bone loss (%) (ABR: 6 ± 2; OL: 20 ± 4), Hill–Sachs interval (mm) (ABR: 8 ± 3; OL: 21 ± 3), and Hill–Sachs depth (mm) (ABR: 5 ± 2; OL: 8 ± 2) were measured in the OL group (P = .001 for all), reflecting significantly more bone loss.

Figure 2.

Mean glenoid bone loss, Hill–Sachs interval, and Hill–Sachs depth measurements in boxplots. ABR, arthroscopic Bankart repair; OL, open Latarjet.

The joint ROM measurements are presented in Table II. External rotation in adduction and external rotation in 90° abduction were found to be approximately 5° higher in the OL group, and these differences were significant (P = .011 and P = .016). The values of forward flexion, abduction, and internal rotation were also slightly higher in the OL group, although the differences were not statistically significant.

Table II.

Shoulder joint range of motion measurements at the postoperative first year.

	ABR Mean ± SD Median (Min-Max)	OL Mean ± SD Median (Min-Max)	P
Forward flexion (°)	169 ± 8	170 ± 9	.616
	170 (150-180)	170 (150-180)
Abduction (°)	162 ± 10	164 ± 10	.624
	160 (140-180)	165 (140-180)
Internal rotation (°)	52 ± 9	55 ± 8	.511
	50 (40-70)	55 (45-65)
External rotation in adduction (°)	46 ± 4	51 ± 12	.011∗
	45 (35-50)	50 (30-70)
External rotation in 90° abduction (°)	101 ± 5	106 ± 8	.016∗
	100 (90-110)	105 (90-120)

ROM, range of motion; ABR, arthroscopic Bankart repair; OL, open Latarjet.

^∗Statistically Significant.

The patients were compared functionally with the ASES, WOSI, and Rowe scores, and the mean values are shown as boxplots in Figure 3. No statistical differences were found between the groups for ASES scores (ABR: 88.6 ± 9.4; OL: 94.3 ± 5.9; P = .131), WOSI scores (ABR: 9.5 ± 3.6; OL: 6.6 ± 4.3; P = .221), or Rowe scores (ABR: 91.8 ± 8.9; OL: 96.4 ± 4.8; P = .343).

Figure 3.

Mean ASES score, WOSI score, and Rower score values in boxplots. ASES, American Shoulder and Elbow Surgeons; WOSI, Western Ontario Shoulder Instability Index; ABR, arthroscopic Bankart repair; OL, open Latarjet.

The distributions of residual apprehension, redislocation, and patient surgery satisfaction in the follow-up are given in Table III. While residual apprehension was detected in almost one-third of the ABR group, it was present in only 2 patients of the OL group (P = .007). The redislocation rate was 12.5% in the ABR group for a minimum follow-up period of 2 years, while no redislocation was observed in the OL group (P = .012). Good or excellent patient satisfaction was determined among 90.6% of the ABR group and 95.8% of the OL group, and these rates were statistically similar.

Table III.

Distributions of residual apprehension, postoperative redislocation, and patient satisfaction.

	ABR	OL	P
	n (%)	n (%)
Residual apprehension
Yes	10 (31.3)	2 (8.3)	.007∗
No	22 (68.7)	22 (91.7)
Redislocation
Yes	4 (12.5)	0	.012∗
No	28 (87.5)	24 (100.0)
Patient surgery satisfaction
Not Satisfied	0	0	.621
Moderate	3 (9.4)	1 (4.2)
Good	9 (28.1)	5 (20.8)
Excellent	20 (62.5)	18 (75.0)

ABR, arthroscopic Bankart repair; OL, open Latarjet; n, number.

^∗Statistically Significant.

Data regarding grade and progression of the dislocation arthropathy in preoperative and postoperative follow-up periods are obtained in Table IV. No arthropathy was detected in the preoperative period in either group. During the postoperative follow-up, dislocation arthropathy was detected in 9 patients in the ABR group (6 patients grade I, 2 patients grade II, and 1 patient grade III) and in 4 patients in the OL group (3 patients grade I and 1 patient grade II) (P = .038). The radiograph of a patient who developed grade 3 dislocation arthropathy after ABR is presented in Figure 4. Moreover, in patients with grade II and III dislocation arthropathy, forward flexion and external rotation were more restricted than in patients without arthropathy. When the coracoid graft position was evaluated in the OL group, it was seen that the graft was transferred flush to the joint in 18 patients, medial position in 3 patients, and lateral position in 3 patients. In addition, in 3 of the 4 patients with dislocation arthropathy in the OL group, graft placement was lateral to the joint, and in 1 patient, graft placement was flush to the joint.

Table IV.

Grade and progression of dislocation arthropathy in preoperative and postoperative follow-up period.

	ABR	OL	P
	n (%)	n (%)
Dislocation arthropathy
Preoperative	0	0	1.000
Postoperative	9 (28.1)	4 (16.7)	.038∗
Severe (≥3 points)	1 (3.1)	0	.635
Progression (≥2 points)	3 (9.4)	1 (4.1)	.455

ABR, arthroscopic Bankart repair; OL, open Latarjet; n, number.

Dislocation arthropathy was graded from 1 (mild) to 4 (severe) according to the modified Samilson and Prieto classification.

^∗Statistically Significant.

Figure 4.

Osteophyte is shown with a red arrow in the radiograph of the patient who developed grade III dislocation arthropathy according to the modified Samilson and Prieto classification at the third postoperative year after ABR. ABR, arthroscopic Bankart repair.

Finally, the relationships between the number of preoperative dislocations and functional scores were evaluated, as presented in Table V, and no correlation was detected between the number of dislocations and those scores.

Table V.

Evaluation of the correlation between the number of dislocations and functional scores.

	ASES score	WOSI score	Rowe score
Number of dislocations
r	0.11	−0.27	0.14
P	.477	.079	.383

ASES, American Shoulder and Elbow Surgeons; WOSI, Western Ontario Shoulder Instability Index.

In the ABR group, 4 patients were reoperated after redislocation. Revision ABR was performed on 1 patient, and revision OL was performed on 3 patients. No redislocation was observed during follow-up. Graft resorption was detected in 2 patients in the OL group. However, both of these patients had no complaints, and there was no residual apprehension. Therefore, no additional surgery was performed. In 1 patient, the screws were removed due to implant irritation. These were all complications in both groups.

Discussion

The most important finding of this study is that residual apprehension, redislocation, and possibility of dislocation arthropathy were significantly reduced despite greater bipolar bone loss in patients who underwent primary OL due to anterior glenohumeral instability.

ABR has been the most preferred treatment method for anterior glenohumeral instability in the last few decades.²⁹ However, despite reports of good medium-term clinical results, the effectiveness of ABR began being questioned over the years. Zimmermann et al⁴² reported that the success of the treatment decreased and patient dissatisfaction increased over time in patients who underwent ABR. According to a systematic review, approximately one-third of athletes who underwent ABR failed to return to preinjury levels.²¹ On the other hand, according to a study conducted in the United States, while approximately 500 OL procedures were performed in the treatment of anterior glenohumeral instability in 2009, >1300 OL procedures were performed in 2018, and the number of OL procedures will increase further toward 2030.¹ Pugliese et al²⁸ found that the OL procedure was more cost-effective in the long term than ABR in the treatment of anterior shoulder instability for patients with ISI score between 4 and 6. In light of this information, it is predicted that the OL procedure will become increasingly globally popular in the treatment of anterior glenohumeral instability in the coming years.

The success of surgery for anterior glenohumeral instability can be evaluated by the patient’s feeling of a stable shoulder, continuity of joint stability, and general patient satisfaction. Ernstbrunner et al reported 53% excellent satisfaction and a redislocation rate of 25% after ABR and 92% excellent satisfaction and a redislocation rate of 8% after OL for anterior shoulder instability patients over the age of 40.⁹ In a study conducted with collision sport athletes, a sense of apprehension and redislocations were respectively observed in 36.3% and 6.3% of the ABR group, while these rates were 27.5% and 0% in the OL group during a minimum postsurgery follow-up duration of 2 years.¹² Waltenspül et al found a residual apprehension rate of 33% and good or excellent results for 74% in the ABR group and a residual apprehension rate of 10% and good or excellent results for 96% in the OL group.³⁹ In a retrospective study evaluating the long-term results of these treatments, the residual apprehension rate was significantly 3 times higher after ABR compared to OL.⁴² In the same study, while redislocations had occurred in 13% of the ABR group at 10 years of follow-up, this rate was only 1% in the OL group.⁴² According to a recent systematic review and meta-analysis conducted on 3275 shoulders in the treatment of recurrent anterior shoulder instability, the Latarjet procedure reduces the risk of recurrence and redislocation.¹⁴ In a study conducted by Elamo et al,⁸ out of 48 patients who developed recurrence of instability after ABR, 30 underwent revision ABR and 18 underwent revision OL. While 13 patients (43%) in the revision ABR group had recurrence of instability symptoms, no patient in the revision OL group had symptoms. In the current study, results similar to the literature were obtained, and the residual apprehension rate was found to be significantly lower in the OL group. Additionally, no redislocations were observed in the OL group during a minimum follow-up period of 2 years. Good or excellent patient satisfaction was observed at a rate of 90.6% in the ABR group and 95.8% in the OL group. Although both treatments are extremely successful, stronger stability is achieved thanks to the triple blocking effect provided by the OL procedure, with the sling effect of the conjoint tendon, the bone effect of the graft, and the ligament effect of the coracoacromial ligament, thus preventing residual apprehension and redislocation.²² Moreover, the OL procedure provides extremely satisfactory results in revision surgery as well as its success in primary anterior glenohumeral instability surgery. The high rate of recurrence of instability after ABR should be taken into consideration when this procedure is preferred.

When defects in the glenoid as seen in anterior glenohumeral instability and Hill–Sachs lesions in the humerus occur together, they are categorized as bipolar bone loss, and this is an important risk factor for redislocation, especially after ABR.²⁴ According to a biomechanical cadaveric study, bipolar bone defects have a negative effect on glenohumeral stability.² It is also known that there is a relationship between bipolar bone loss and the number of preoperative dislocations; as the number of dislocations increases, bone defects grow.¹⁹ In a study conducted after the first dislocations experienced by athletes, the rate of glenoid bone loss was found to be significantly 8 times higher and the rate of off-track Hill–Sachs lesions was 9 times higher after OL compared to ABR.¹² Waltenspül et al³⁹) similarly applied primary OL for higher rates of glenoid bone loss. In our study, similar to the literature, the numbers of preoperative dislocations, glenoid bone defects, and Hill–Sachs lesions were significantly higher in the OL group. The fact that lower rates of residual apprehension and redislocation are seen after the OL procedure, even with greater bipolar bone losses, reveals the success of this procedure in treating glenohumeral stability.

In addition to obtaining a stable joint after surgery for shoulder instability, the absence of any limitation in shoulder ROM indicates the success of the surgery. Waltenspül et al³⁹ reported no significant difference in joint ROM between the groups of patients who underwent ABR and OL. In another study, although approximately 3° more external rotation was obtained in the OL group, the difference was not statistically significant.³⁰ On the other hand, Horinek et al¹¹ observed a 10° loss in forward flexion in the second postoperative year compared to the preoperative period among patients who underwent OL. Ernstbrunner et al⁹ observed a 22° decrease in the final follow-up of the ABR group compared to the preoperative period. In our study, approximately 5° more shoulder ROM was found in the OL group for both external rotation in adduction and external rotation in 90° abduction, and these differences were statistically significant. Residual apprehension, which is more common after ABR, causes patients to avoid external rotation movements and, as a result, to have more limited external rotation ROM values.

A wide variety of scales and questionnaires are available to functionally evaluate patients after shoulder surgery. Horinek et al¹¹ compared patients who underwent ABR or OL due to anterior shoulder instability with the single assessment numeric evaluation, a visual analog scale for pain, and the WOSI, and no significant differences were seen between the scores of the two groups. In another study in which patients over the age of 40 were followed for >10 years, patients who underwent ABR or OL were evaluated with the Walch–Duplay score and Constant score, and no significant difference was reported between the two groups.⁹ Rai et al³⁰ reported no significant differences in terms of ASES scores or Rowe scores between patients who underwent ABR or OL due to recurrent shoulder instability. Waltenspül et al³⁹ also evaluated these patient groups using WOSI and ASES scores and reported similar results in the two groups. In the current study, patients were evaluated functionally with the ASES score, WOSI score, and Rowe score. Similar to the literature, although the rates of satisfactory scores were not identical in the two groups, no significant difference was observed.

Dislocation arthropathy is an important complication that can occur after anterior glenohumeral instability surgery. Waltenspül et al³⁹ reported mild dislocation arthropathy in 29% of patients in the ABR group and 11% of patients in the OL group. In another study, more frequent, more severe and progressive dislocation arthropathy was detected in the ABR according to OL.⁹ According to a review by Vezeridis et al,³⁷ dislocation arthropathy develops due to excessive tightness after instability surgery. In the current study, a significantly higher rate of dislocation arthropathy was observed in the ABR group. Over-tight suturing of the capsulolabral complex during ABR increases the risk of dislocation arthropathy. Obtaining a more stable joint with the OL procedure reduces the risk of arthropathy due to instability. However, the most important rule to avoid dislocation arthropathy in the OL procedure is not to transfer the graft lateral to the joint. Lädermann et al,¹⁷ when they followed-up 117 patients to whom they underwent the OL procedure for a mean of 16.2 years, found that lateral coracoid transfer according to the glenoid rim predisposed to dislocation arthropathy. In our current study, the lateral position of the coracoid graft was detected in 75% of the patients who developed dislocation arthropathy after the OL procedure. Transfer of the coracoid graft to the lateral aspect of the joint is a risk factor for dislocation arthropathy.

The current study has some limitations, one of which is its retrospective design. There was no predetermined algorithm and randomization in the process of treatment selection; rather, the surgeon’s preference was generally at the forefront. The sample size was small and the mean follow-up period was relatively short. The presence of hyperlaxity was not evaluated in the patients, and this is a factor that may affect the presence of postoperative residual apprehension and the success of the surgery. Although the presence of residual apprehension was recorded during the last follow-up visit of the patients, it is not known when it started or whether a trauma caused it to start. Therefore, residual apprehension should be among the routine tests applied in the postoperative evaluation of patients who have undergone surgery for anterior glenohumeral instability. The method used for bone loss measurements is not standardized and different results may be obtained in measurements made with different methods. The scores used in functional evaluations and the number of preoperative dislocations are data based primarily on patient statements and may therefore be subjective.

Conclusion

ABR and OL procedures provide satisfactory outcomes for patients with anterior glenohumeral instability and appropriate indications. The OL procedure is preferred for patients with more dislocations and greater bipolar bone loss before surgery. Despite greater bipolar bone loss, the OL procedure provides lower rates of residual apprehension, redislocation, and the possibility of dislocation arthropathy. Additionally, due to the stability it provides, there is less loss of external rotation.

Disclaimers:

Funding: This study received no funding source.

Conflicts of interest: The authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.