Shoulder Latarjet Surgery Shows Wide Variation in Reported Indications, Techniques, Perioperative Treatment, and Definition of Outcomes, Complications, and Failure: A Systematic Review
To systematically review and compare the surgical indications, technique, perioperative treatment, outcomes measures, and how recurrence of instability was reported and defined after coracoid transfer procedures.
Methods
A systematic review of the literature examining open coracoid transfer outcomes was conducted according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines using the Cochrane registry, MEDLINE, and EMBASE databases from 2010 to 2020. Inclusion criteria included open coracoid transfer techniques, including the Bristow or Latarjet technique, full text availability, human studies, and English language.
Results
A screen of 1,096 coracoid transfer studies yielded 72 studies, which met inclusion criteria with a total of 4,312 shoulders. One study was a randomized controlled trial, but the majority of them were retrospective. Of those, 65 studies reported on postoperative outcome scores, complication rates, revision rate, and recurrence rates. Forty-three reported on range of motion results. Thirty studies reported on primary coracoid transfer only, 7 on revision only, and 30 on both primary and revision, with 5 not reporting. Average follow-up was 26.9 months (range: 1-316.8 months). Indications for coracoid transfer, technique, perioperative care, complications, and how failure was reported varied greatly among studies.
Conclusions
Latarjet and coracoid transfer surgery varies greatly in its indications, technique, and postoperative care. Further, there is great variation in reporting of complications, as well as recurrence and failure and how it is defined. Although coracoid transfer is a successful treatment with a long history, greater consistency regarding these factors is essential for appropriate patient education and surgeon knowledge.
Level of Evidence
Level IV, systematic review of Level I-IV studies.
Introduction
Despite the commonality of anterior shoulder instability, no agreed upon algorithm exists for its treatment.
Arthroscopic instability repair is successful, less invasive, and offers lower complication rates when compared with open bone block procedures. However, recurrence has been reported to be higher than open procedures, especially in circumstances with significant bone loss, Hill Sachs lesions, or both.
Open Bankart repair remains a valuable treatment option and is typically more commonly used in patients who play sports that are at higher risk of recurrent instability or those who failed arthroscopic repair and have recurrent instability without bone loss. Some surgeons advocate for earlier treatment with coracoid transfer, while others reserve the procedure for significant glenoid bone loss. With the recently improved understanding of Hill Sachs lesions and the glenoid track, as well as the development of newer bony augmentation procedures, such as distal tibial allograft, the treatment algorithm for anterior shoulder instability treatment is currently even more disputed.
Coracoid transfer, such as the Latarjet or Bristow procedure, extends the glenoid articular arc, while the sling effect of the conjoint tendon is believed to add dynamic stability in abduction and external rotation.
Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: Significance of the inverted-pear glenoid and the humeral engaging Hill-Sachs lesion.
Conversely, the Latarjet procedure is more invasive then arthroscopic and other open techniques, can be technically challenging, and complications rates have been reported as high as 30%.
Coracoid graft osteolysis after the Latarjet procedure for anteroinferior shoulder instability: A computed tomography scan study of twenty-six patients.
Treatment decisions and techniques also appear to be quite regional. The Latarjet or coracoid transfer procedure is commonly considered for primary shoulder instability in some areas of the world due to its reliable outcome; conversely, it is reserved in the setting of bone loss and multiple failed previous surgeries in other regions.
Latarjet techniques also vary greatly with many possible variations. The size of harvested coracoid graft, handling of the subscapularis muscle, graft positioning and fixation alternatives, method for capsular repair, and adjuncts such as remplissage are only some areas of variability. Beyond indications and surgical techniques, postoperative care and return to activities also differ.
Therefore, the purpose of this study was to systematically review and compare the surgical indications, technique, perioperative treatment, outcomes measures, and how recurrence of instability was reported and defined after coracoid transfer procedures. It was hypothesized that there will be great diversity associated with coracoid transfer procedures in regard to indication for surgery, surgical technique, postoperative care, as well as how outcomes and recurrence rate are reported, and therefore, will pose a challenge in comparing studies.
Methods
Search Strategy and Study Selection
This study followed the Preferred Reporting Items for Systematic Review and Meta-Analysis statement (PRSIMA) guidelines.
As this study was a systematic review of published studies, institutional review board (IRB) approval was not required. A systematic literature review was conducted using the PubMed and Scopus/Embase/Cochrane libraries databases and included dates of publication from 2010 through 2020. This computerized search was performed in October of 2020. The terms used included: (Latarjet OR Open Latarjet OR Latarjet procedure OR Bristow OR open Bristow OR Bristow procedure OR Bristow-Latarjet OR coracoid transfer) AND (anterior shoulder instability OR shoulder instability). Two investigators (K.K.T. and L.A.P., medical students), performed a separate, manual study selection from the resulting list to exclude duplicates and to select those specifically related to the relevant items. Studies returned by the literature search were assessed for relevance first based on title, followed by abstract, and lastly, the full manuscript text prior to final study inclusion and data extraction. In case of any discrepancies in article selection between the two investigators, a third investigator was involved (J.W.A.). Previous articles from the literature review and applicable references from the included articles were used if they met the inclusion criteria.
Eligibility Criteria
Clinical outcome studies that included the following criteria were considered eligible: published in the English language, reported clinical and/or functional outcomes, and reported rates of intraoperative and/or postoperative complications following coracoid transfer for the correction of anterior shoulder instability. Studies were excluded when they included any of the following: arthroscopic procedure, human cadaveric studies, animal studies, clinical studies without measures of postoperative outcomes, review articles, case reports, presentations, abstracts, editorial articles, or surveys. There was no minimum follow-up requirement.
Data Extraction and Quality Appraisal
Each study was categorized by the country, number of patients and shoulders, specifics of surgical technique, postoperative imaging and rehabilitation, range of motion (ROM) measures, postoperative outcome scores, length of follow-up, complications, and recurrence/revision rates, and how those were defined. An assessment of methodological quality was conducted by two reviewers using the Cochrane Collaboration tool.
Two authors (K.K.T. and L.A.P., medical students) independently assessed the potential assessed risk of bias of the studies included using the methodological index for non-randomized studies (MINORS), a methodological index for nonrandomized studies
CONSORT GROUP (Consolidated Standards of Reporting Trials). The revised CONSORT statement for reporting randomized trials: explanation and elaboration.
A third tiebreaker (J.W.A.) was designated in the case of discrepancy to resolve the dispute. The items of the questionnaire were scored 0 if not reported, 1 when reported but inadequate, and 2 when reported and adequate. The ideal score was 16 for noncomparative studies and 24 for comparative studies. Studies with a MINORS score of 13 to 16 for noncomparative studies or 21 to 23 for comparative studies were considered at low risk of bias and those ≤12 for noncomparative studies or ≤20 for comparative studies at high risk of bias.
Results
Literature Selection and Risk of Bias
An online database search using the above databases produced a total of 1,096 publications. After nonrelevant and identical articles were excluded, 163 abstracts were evaluated for inclusion. A total of 72 studies were identified for inclusion. A PRISMA flow diagram can be seen in Fig 1 detailing study identification and screening. One study was a randomized controlled trial, one was a Level II study, 23 were Level III studies, and 47 were Level IV studies. No Level V studies were included.
Fig 1Transparent reporting of systematic reviews and meta-analysis flow diagram outlining the search strategy results from an initial search to those studies that were included.
Risk of bias was assessed using MINORS criteria and ranged from 9 to 21 (Table 1). The average MINORS score for noncomparative studies was 11 and 18 for comparative studies. Five noncomparative studies were, therefore, considered low risk of bias, while 4 comparative studies were low risk. Forty (88.9%) noncomparative studies were, therefore, considered high risk of bias, while 24 (85.7%) comparative studies were high risk.
Table 1Quality Scoring of Included Studies Based on the MINORS Criteria and Coleman Score
Study
MINORS Score
Coleman Score
Abdelhady et al.
11
61
Ali et al.
20
53
Atalar et al.
12
51
Aydin et al.
15
51
Bah et al.
18
75
Balestro et al.
9
63
Baverel et al.
20
75
Beranger et al.
9
67
Bessiere et al. (2013)
21
65
Bessiere et al. (2014)
21
80
Bhatia et al.
9
61
Bokshan et al.
20
63
Bonnevialle et al.
12
64
Bouju et al.
11
68
Cautiero et al.
10
54
Chaudhary et al.
11
71
Cho et al.
15
53
Colegate-Stone et al.
12
54
Cunningham et al.
19
66
Dauzere et al.
10
59
De Carli et al.
16
57
De l’Escalopier et al.
10
56
Di Giacomo et al.
12
68
Domos et al.
13
75
du Plessis et al.
9
56
Ebrahimzadeh et al.
11
62
Edouard et al.
11
71
Emami et al.
12
64
Ernstbrunner et al.
14
68
Ersen Bayram et al.
16
58
Ersen Birisik et al.
15
69
Flinkkilä and Sirniö
12
64
Flinkkila et al.
21
71
Frank et al.
18
78
Frank et al.
12
69
Gordins et al.
12
63
Gough et al.
11
75
Holzer et al.
9
61
Hovelius et al. (2011)
18
66
Hovelius et al. (2012)
15
73
Ikemoto et al.
11
63
Jeon et al.
19
61
Kawasaki et al.
12
74
Kee et al.
12
68
Kordasiewicz et al.
18
72
Lateur et al.
11
59
Marion et al.
21
71
Metais et al.
15
54
Minkus et al.
9
71
Mizuno et al.
10
60
Mook et al.
14
78
Moon et al.
11
69
Moroder et al. (2017)
13
68
Moroder et al. (2019)
19
72
Moura et al.
10
71
Neyton et al.
12
70
Privitera et al.
11
67
Raiss et al.
12
66
Ranalletta et al.
12
72
Ropars et al.
20
72
Roulet et al.
13
60
Ruci et al.
10
72
Schmid et al.
12
65
Shah et al.
12
63
Tasaki et al.
12
61
Vadala et al.
9
64
Waterman et al.
11
52
Xu et al.
17
68
Yang et al. (2018)
19
64
Yang et al. (2016)
16
68
Zhu et al.
19
78
Zimmermann et al.
15
71
MINORS, Methodological Items for Non-Randomized Studies (score).
A total of 4,312 shoulders were included. Sample sizes ranged from 11 to 262 and were predominantly male (range: 16-100%) with a mean age ranging from 18.9 to 62 years. The average follow-up was 26.9 months (range: 1-316.8 months). Thirty-nine studies were from Europe (20 from France), 15 from Asia, and 9 from the United States (Table 2).
Table 2Demographics and Surgical Specifics Reported with Latarjet
Study
Year
Country
Level of Evidence
Sample Size (Shoulders/Patients)
Mean Age
Male/Female
Type of Surgery
Subscapularis Handling
Graft Position
No. of Screws
Capsule Repaired to Glenoid
Capsule Repaired to CAL
Capsule Closed Side to Side
Abdelhady et al.
2014
Egypt
IV
14/13
24.4
9/4
Primary
Split
Congruent Arch
2
Yes
Ali et al.
2020
-
III
-/15
29.5
12/3
Both
Split
Traditional
2
Yes
Atalar et al.
2013
Turkey
IV
35/35
35
33/2
Both
Split
Congruent Arc
2
Yes
Aydin et al.
2012
Turkey
III
13/13
29.6
NA
Primary
-
Bristow
1
-
-
-
Bah et al.
2018
France
III
43/43
24.25
7/36
Primary
Split
Traditional
2
Balestro et al.
2015
France
IV
12/11
28.6
7/4
Primary
Split
Traditional
2
Yes
Baverel et al.
2018
France
III
110/106
-
88/18
Primary
Split
Traditional
2
Yes
Group A Competitive Athlete
-
-
-
57/61
21.7
51/6
-
-
-
-
-
-
-
Group B Recreational Athlete
-
-
-
49/49
22.5
37/12
-
-
-
-
-
-
-
Beranger et al.
2016
France
IV
-/47
27.9
46/1
-
Split
Traditional
2
Yes
Bessiere et al.
2013
France
III
51/51
25
49/2
Primary
-
Traditional
2
-
-
-
Bessiere et al.
2014
France
III
93/90
26
89/4
Primary
Split
Traditional
2
Yes
Bhatia et al.
2013
India
IV
-/7
25
NA
-
Split
-
2
-
-
-
Bokshan et al.
2017
USA
III
163/163
29.8
134/29
-
-
-
-
Bonnevialle et al.
2013
France
IV
6/6
33.3
NA
Revision
Takedown
Bristow
1
Yes
Bouju et al.
2014
France
IV
70/68
26.7
48/20
Primary
Split
Traditional
2
Yes
Cautiero et al.
2017
Italy
IV
26/25
25.9
23/25
Primary
Split
Traditional
1
Yes
Chaudhary et al.
2015
India
IV
24/24
31.8
23/1
Revision
Split
Traditional
2
Yes
Cho et al.
2015
Korea
III
35/35
27.8
32/3
Both
Split
Traditional
2
Yes
Colegate-Stone et al.
2015
South Africa
IV
-/56
24
50/6
Both
Split
Congruent Arch
2
Yes
Cunningham et al.
2015
Switzerland
III
-/36
26
34/2
Primary
Split
Traditional
2
Yes
Dauzere et al.
2016
France
IV
55/54
25.5
67/1
Primary
Split
Traditional
2
Yes
De Carli et al.
2018
Italy
III
40/40
28
24/16
Primary
Takedown
Traditional
2
Yes
De l’Escalopier et al.
2018
France
IV
20/20
26.5
N/A
Primary
Split
Traditional
2
Yes
Di Giacomo et al.
2011
Italy
IV
26/26
28.6
19/7
Primary
Split
Traditional
2
-
-
-
Domos et al.
2019
France
IV
-/99
46
60/39
Primary
Split
Traditional
2
Yes
du Plessis et al.
2017
South Africa
IV
-/29
37.2
0/29
Both
Split
Traditional
2
-
-
-
Ebrahimzadeh et al.
2015
Iran
IV
36/36
24.6
35/1
Primary
Split
Traditional
2
-
-
-
Edouard et al.
2010
France
IV
-/20
27
20/0
Primary
Split
Traditional
2
Yes
Emami et al.
2011
Iran
IV
30/30
30.56
30/0
Primary
Split
Traditional
-
Yes
Ernstbrunner et al.
2019
Switzerland
IV
40/39
48
30/9
Both
Split
Traditional
2
Yes
Ersen Bayram et al.
2017
Turkey
III
65/62
-
42/20
-
Split
Traditional
2
-
-
-
Group A Epileptic
-
-
-
11/9
31.3
6/3
-
-
-
-
-
-
-
Group B Non-Epileptic
-
-
-
54/53
31.2
36/17
-
-
-
-
-
-
-
Ersen Birisik et al.
2017
Turkey
III
-/48
-
-
Both
-
Traditional
2
-
-
-
Group A Tenotomy
-
-
-
-/20
33.2
18/2
-
Takedown
-
-
-
-
-
Group B Split
-
-
-
-/28
33.9
24/4
-
Split
-
-
-
-
-
Flinkkilä and Sirniö
2015
Finland
IV
-/49
28.4
45/7
Revision
Split
Traditional
2
Yes
Flinkkila et al.
2019
Finland
IV
-
-
-
-
-
-
-
Primary
-
-
47/47
32
36/11
Primary
Split
Traditional
2
Revision
-
-
52/52
33
42/10
Revision
Split
Traditional
2
Frank et al.
2018
USA
III
-/50
25.4
48/2
Both
Split
Traditional
2
Frank et al.
2018
USA
IV
133/133
28.5
100/33
Primary
Split
Traditional
2
Yes
Gordins et al.
2014
Sweden
IV
31/31
26.7
23/8
Both
Split
Bristow
1
-
-
-
Gough et al.
2017
England
IV
50/48
27
46/2
Both
Split
Traditional
1
-
-
-
Holzer et al.
2013
Switzerland
IV
-/148
28.4
106/42
-
Takedown
Traditional
1
Yes
Hovelius et al.
2011
Sweden
III
97/96
27.8
82/15
Both
Split
Congruent Arch
1
Yes
Hovelius et al.
2012
Sweden
III
-
-
-
-
-
-
-
Series 1
-
-
-
118/-
27
95/23
Both
Split
Congruent Arch
1
Yes
Series 2
-
-
-
167/-
28
142/25
Both
Split
Congruent Arch
1
Yes
Series 3
-
-
-
34/-
26
30/4
Both
Split
Congruent Arch
1
Yes
Ikemoto et al.
2011
Brazil
IV
-/26
28
26/0
-
Takedown + Split
Traditional
2
Yes
Jeon et al.
2018
Korea
III
31/31
27.4
26/5
Primary
Split
Traditional
2
Yes
Kawasaki et al.
2018
Japan
IV
176/152
18.9
-
Primary
Split
Bristow
1
Yes
Kee et al.
2017
Korea
IV
110/110
23.8
100/10
Both
Split
Traditional
2
Yes
Kordasiewicz et al.
2016
Poland
III
47/-
28
45/2
Primary
Split
Traditional
1 or 2
Yes
Lateur et al.
2018
France
IV
40/38
34.5
32/6
-
Split
Traditional
1
-
-
-
Marion et al.
2016
France
II
22/22
27.3
16/6
Both
Split
Traditional
2
Yes
Metais et al.
2016
France
III
-/38
27.8
N/A
Both
Split
Traditional
2
Yes
Minkus et al.
2019
Germany
IV
29/29
28
25/4
Revision
Split
Traditional
2
Yes
Mizuno et al.
2014
France
IV
68/60
29.4
54/14
Primary
Split
Traditional
2
Yes
Mook et al.
2016
USA
IV
38/38
26
33/5
Both
Split
Traditional
2
Yes
Yes
Moon et al.
2015
Korea
IV
44/44
24.5
41/3
Both
Split
Traditional
2
Yes
Moroder et al.
2017
Austria
IV
25/25
62
13/12
Both
Split
Traditional + Bristow
1 or 2
Yes
Moroder et al.
2019
Germany
I
25/25
31
28/2
Both
Split
Traditional
2
Yes
Moura et al.
2018
Portugal
IV
102/102
26
89/13
Primary
Takedown
Traditional
1
Yes
Neyton et al.
2012
France
IV
37/34
23.4
34/0
Primary
Split
Traditional
2
Yes
Privitera et al.
2018
USA
IV
73/73
25.8
64/9
Both
Split
Traditional
2
Yes
Raiss et al.
2012
France
IV
14/12
31
10/2
Both
Split
Traditional
2
Yes
Ranalletta et al.
2018
Argentina
IV
65/65
26.8
63/2
Revision
Split
Congruent Arch
2
-
-
-
Ropars et al.
2015
France
IV
77/75
26.3
54/23
Primary
Split
Traditional
2
Yes
Group A ACR+
-
-
-
-/39
26.1
22/17
-
-
-
-
-
-
-
Group B ACR-
-
-
-
-/38
26.5
32/6
-
-
-
-
-
-
-
Roulet et al.
2019
France
IV
256/262
25.1
222/43
Primary
Split
Traditional
2
Yes
Ruci et al.
2015
Albania
IV
45/42
-
40/2
Primary
Split
Bristow
1
-
-
-
Schmid et al.
2012
Switzerland
IV
49/49
29
37/12
Revision
Split
Traditional
2
Yes
Shah et al.
2012
USA
IV
48/47
30
39/9
Both
Split
Traditional
2
Yes
Tasaki et al.
2015
Japan
IV
40/38
21
37/1
Both
Split
Bristow
1
Vadala et al.
2017
Italy
IV
24/24
27.2
22/2
Both
Takedown
-
-
Waterman et al.
2016
USA
IV
65/64
25.9
59/5
Both
-
-
-
-
-
-
Xu et al.
2019
China
III
26/26
31.23
20/6
-
Split
-
2
Yes
Yang et al.
2018
USA
III
91/91
30
86/5
Both
-
-
-
-
-
-
Yang et al.
2016
USA
III
-
-
-
Both
Split
Traditional
2
Yes
Yes
Group B with >25% Bone Loss
-
-
12/12
26.5
8/4
-
-
-
-
Group A with <25% Bone Loss
-
-
42/40
22.3
33/7
-
-
-
-
Zhu et al.
2017
China
III
44/44
34.8
32/12
Both
Split
Traditional
2
Yes
Zimmermann et al.
2016
Switzerland
III
93/-
30.8
82/11
Primary
Split
Traditional
2
Yes
Dashes denote no data available. CAT, coracoacromial ligament.
Thirty studies reported on primary coracoid transfer only, 7 on revision only, and 30 on both primary and revision. Five studies did not indicate whether primary or revision surgeries were included. Seven studies performed a subscapularis takedown, while the rest performed a subscapularis splitting technique. One study that performed a takedown technique included revision Latarjet surgery only, while two included both primary and revision Latarjet surgeries, and the others did not specify. Five studies did not comment on the handling of the subscapularis. Seven studies performed the congruent arc method, while the rest used the traditional method. Fifty-three studies used two screws in their patients, while 15 used a single screw, and 2 studies used a combination of 1 or 2 screws, based on patient specifics. In regard to capsular management, 9 studies repaired the capsule to the glenoid, 39 studies repaired the capsule to the coracohumeral ligament stump, and 7 studies repaired the capsule side to side only, while the remaining studies did not comment on capsular management (Table 2).
Preoperative and Postoperative Imaging
The reporting on the type of preoperative imaging was also limited. Thirty-seven studies noted that they obtained preoperative radiographs. Eight studies obtained MRIs. Thirty-two studies obtained CT scans, 15 3D CTs, while 7 studies obtained CTs or MRIs on all patients (Table 3). The type of postoperative imaging obtained was also sparsely reported. Forty-two studies obtained postoperative radiographs, 19 CTs, 8 3D CTs, and 7 studies obtained multiple postoperative CTs (Table 3).
Tables 3Perioperative Episode of Care: Imaging, and Postoperative Limitations and Rehabilitation
Imaging
Type
Preoperative Number of Studies
Postoperative Number of Studies
Radiograph
37
42
CT
32
19
3D CT
15
8
Multiple CTs
0
7
MRI
8
0
CT or MRI
7
0
Postoperative Rehabilitation
Number of Studies Reported
Average Length (Days)
Sling
46
21 (0-42)
Number of Studies Reported
Median Time to Begin (Days)
Formal PT
12
14 (1-90)
Passive ROM
43
1 (0-30)
Active ROM
32
28 (0-90)
External Rotation
22
42 (0-84)
Strengthening
25
56 (42-158)
Return to Sport
43
120 (90-360)
CT, computed tomography; MRI, magnetic resonance imaging; PT, physical therapy; ROM, range of motion; 3D, three-dimensional.
Average length of recommended postoperative sling use was 21 days with many studies allowing immediate sling removal and the longest recommending 42 days. The date of the start of formal physical therapy was only reported in 12 studies, which ranged from 1 to 90 days postoperatively with the median being 14 days. When to begin passive range of motion also varied greatly, with the average being 1 day postoperatively (range: 0-30 days). The date of starting active range of motion also varied, with the median being at 28 days (range: 3 days-3 months). External rotation was limited typically for 6 weeks (range: 0-12 weeks), and strengthening was typically started at 8 weeks (range: 6-22.5 weeks). Return to sport was normally allowed at 4 months postoperatively (range: 3-12 months) (Table 3).
Patient-Reported Outcome Measures
Sixty-one studies reported preoperative patient-reported outcome scores with the Rowe score being reported most commonly. Other reported scores included Walch-Duplay, American Shoulder and Elbow Surgeons Shoulder (ASES), organ injury scale (OIS), University of California, Los Angeles (UCLA), Western Ontario Shoulder Instability Index (WOSI), Single Assessment Numeric Evaluation (SANE), Constant, simple shoulder test (SST), Constant-Murley, visual analog scale (VAS), Disabilities of the Arm, Shoulder and Hand (DASH), and subjective shoulder value (SSV). Four studies reported neither preoperative nor postoperative outcome scores. The Rowe score was the most commonly reported postoperative score (47 studies). Tables 4 and 5 show the postoperative scores reported and outcomes from each study evaluated.
Table 4Patient-Reported Outcome Measures Reported After Latarjet
Study
Mean Postoperative Outcomes
Follow-Up Time (mo)
Rowe
Walch-Duplay
ASES
Oxford
UCLA
WOSI
SANE
Constant
VAS
SSV
Abdelhady et al.
91.07 ± 16.6
33.64
Ali et al.
78 ± 11
670 ± 372
30.5 ± 5.25
Atalar et al.
89.1 ± 9.2
91.3 ± 11
1.8 ± 0.6
24 ± 12.2
Aydin et al.
81.9
66 ± 18.3
Bah et al.
92.8 ± 11.3
93.6 ± 13.8
2.8 ± 1.1
89.5 ± 6.3
47.3 ± 9.07
Balestro et al.
90.8 ± 7.5
24
Baverel et al.
Group A Competitive Athlete
84.2 ± 16.4
196.4 ± 202.5
0.7 ± 0.8
91.5 ± 8.3
44 ± 14.5
Group B Recreational Athlete
69.5 ± 22
357.7 ± 479.5
1.1 ± 1.5
86.1 ± 15.7
49 ± 14.5
Beranger et al.
46.8 ± 9.7
Bessiere et al.
85 ± 13
1.62 ± 1.25
90.9 ± 17.5
66
Bessiere et al.
78 ± 22.5
90 ± 17.5
72 ± 17
Bhatia et al.
95 ± 3.7
12.5 ± 0.7
43.8 ± 31.1
20.6
Bokshan et al.
1
Bonnevialle et al.
82 ± 10.5
70 ± 10
73.3 ± 17.5
40 ± 10.3
Bouju et al.
82.6 ± 15.6
89.7 ± 13.75
91.9 ± 9
156 ± 24
Cautiero et al.
94.7
53 ± 9
Chaudhary et al.
95 ± 7.5
26 ± 1.5
Cho et al.
91 ± 8
32 ± 2
.1 ± 0.3
30.4 ± 11.2
Colegate-Stone et al.
3
Cunningham et al.
91
6.6 ± 5.9
Dauzere et al.
78 ± 12.5
89 ± 8
21 ± 12
De Carli et al.
95.6
32.3
111
De l’Escalopier et al.
91.8 ± 9.9
89.2 ± 9.7
195.6 ± 27
Di Giacomo et al.
17.5 ± 6.7
Domos et al.
87 ± 7
83 ± 8.5
75 ± 7.75
1.2 ± 2
87 ± 7
156 ± 60
du Plessis et al.
714 ± 456
69.8 ± 23.4
-
Ebrahimzadeh et al.
95.7 ± 3.6
32.1 ± 1.9
96.5 ± 4
37 ± 13.3
Edouard et al.
88.3 ± 17.1
90.8 ± 13.7
21
Emami et al.
77.66
60 ± 18
Ernstbrunner et al.
89 ± 12
81 ± 9
91 ± 13
132 ± 24
Ersen Bayram et al.
39.2 ± 27
Group A Epileptic
93.6
91.7
91
45.2
Group B Non- Epileptic
94.4
93.1
91.7
-
Ersen Birisik et al.
Group A Tenotomy
96.3 ± 4.1
96.1 ± 2.2
43.8 ± 16.3
Group B Split
93.2 ± 5
91.9 ± 4.9
23.5 ± 8.0
Flinkkilä and Sirniö
19.9 ± 7.5
83.9 ± 15.6
84.9 ± 14.8
38 ± 15.25
Flinkkila et al.
-
Primary
85 ± 15
88 ± 13
34.8 ± 16.8
Revision
76 ± 22
80 ± 18
55.2 ± 31.2
Frank et al.
91.06 ± 8.78
74.30 ± 21.84
80.68 ± 7.21
0.67 ± 0.97
45 ± 20
Frank et al.
3
Gordins et al.
85.85 ± 15.27
78.83 ± 18.88
396 ± 6
Gough et al.
88 ± 15
95 ± 1.5
89 ± 2.25
32 ± 17
Holzer et al.
175.2 ± 59.4
Hovelius et al.
87.5
87.8
84.2
204 ± 27
Hovelius et al.
-
Series 1
180 ±19.5
Series 2
85.5
80.1
204 ± 39
Series 3
82.2
75.2
72 ± 9.9
Ikemoto et al.
93
34
38 ± 17
Jeon et al.
91.1 ± 16.1
32.3 ± 3.4
0.2 ± 0.5
30.9 ± 12.25
Kawasaki et al.
87.7
534.3
51.5
Kee et al.
87.6
32.6
1.6
31 ± 21.75
Kordasiewicz et al.
87.8 ± 18.75
83.9 ± 22.5
0.77 ± 1.25
54.2 ± 12.8
Lateur et al.
95.87
42.4375
94.89
0
96.875
48 ± 18
Marion et al.
451 ± 158.7
29.8 ± 4.4
Metais et al.
83.9
85.9
22.7 ± 4.1
Minkus et al.
91 ± 10
89 ± 11.25
76 ± 14.75
90 ± 7.75
82 ± 15
27 ± 1.7
Mizuno et al.
89.6
90.4 ± 15
240 ± 12
Mook et al.
89.2 ± 10.9
87 ± 12.8
38.4 ± 17.7
Moon et al.
90 ± 13
33 ± 2
1.3 ± 0.7
25.9 ± 8.7
Moroder et al. (2017)
77 ± 20
75 ± 21
556 ± 298
65 ± 18
1.4 ± 2.3
70 ± 22
108 ± 39
Moroder et al. (2019)
90 ± 12
250 ± 300
90 ± 15
24
Moura et al.
91.23 ± 11.46
63.96 ± 32.9
Neyton et al.
93 ± 10
86 ± 16.3
1.6 ± 1.4
144 ± 42.3
Privitera et al.
87.9 ± 17.1
557 ± 504
1.3 ± 2.1
51.6 ± 24
Raiss et al.
76 ± 16.3
99.6 ± 57
Ranalletta et al.
90 ± 7
1.4 ± 1
44 ± 21
Ropars et al.
85.2 ± 17.2
81.7 ± 17.2
91 ± 9.5
55
Group A ACR+
83.9 ± 20.1
79.8 ± 18.6
91 ± 7.5
-
Group B ACR-
88.4 ± 15.8
82.3 ± 19.3
91 ± 11.4
-
Roulet et al.
3
Ruci et al.
90 ± 5.5
46 ± 18
Schmid et al.
84.6 ± 15
78.5 ± 25
38 ± 10
Shah et al.
86.3 ± 19.3
81.5 ± 20.4
9.4 ± 12.3
Tasaki e tal.
97.5 ± 1.16
30.5 ± 4.5
Vadala et al.
93.8 ± 5
33.5 ± 1
94 ± 5
95.6 ± 3
24
Waterman et al.
28.8
Xu et al.
96.23 ± 2.10
91.54 ± 2.38
31.83 ± 1.35
50 ± 17.5
67.6 ± 3
Yang et al.
352 ± 286
85.3 ± 9.6
1.55 ± 1.88
38.4 ± 30
Yang et al.
-
A: <25% Bone Loss
356 ± 361
85.5 ± 11.4
42 ± 24
B: >25% Bone Loss
475 ± 294
77.1 ± 7.22
42 ± 24
Zhu et al.
97.1 ± 2.5
93.3 ± 9.9
96.5 ± 3.8
37.4 ± 9.4
Zimmermann et al.
88.77 ± 14.63
119 ± 23.2
Data are expressed as means ± SD. Numbers with only a mean did not report standard deviation. Hyphens indicate no data available.
ASES, American Shoulder and Elbow Surgeons Shoulder; SANE, Single Assessment Numeric Evaluation; SSV, subjective shoulder value; UCLA, University of California, Los Angeles; VAS, visual analog scale; WOSI, Western Ontario Shoulder Instability Index.
Postoperative Recurrent Instability, Revision Surgery, and Complications
Sixty studies reported if postoperative dislocation occurred. Of those 12 studies that did not, 4 studies did report on subluxation. Overall, 42 studies reported on subluxation. Apprehension was evaluated in 25 studies. Six studies reported patients had postoperative instability but did not define how that was determined. Eight studies reported on subjective instability. The highest number of dislocations reported in a study was 9 (5.4%), while the highest number of subluxations was 20 (12.0%). The highest number of apprehension events reported was 17 (18.2%) (Table 6).
Sixty-four studies reported an overall complication rate of coracoid transfer surgery with the median being 16.7% (range: 2.3–78.6%). Of the studies that reported complications, graft fracture, nonunion, hematoma, infection, prominent hardware, prominent graft, screw breakage, osteoarthritis, recurrent instability, nerve injury/paresthesia, persistent pain and stiffness, thromboembolic events, and wound healing issues were described (Table 7).
Of the coracoid transfer cases evaluated, 5.1% underwent future revision surgery with the average number being 3 per study (range: 0-26). Of the revisions, 97 revisions were due to recurrent instability (2.2%), while many studies did not report the cause. Other reasons for revision surgery included lateral placement of the graft, screw breakage, screw removal, hematoma, infection, nonunion, and bone block fracture (Table 8).
Coracoid transfer surgery varies greatly in its indications, technique, and postoperative care. Further, there is great variation in reporting of complications, as well as recurrence and failure and how it is defined. Although coracoid transfer is a successful treatment with a long history, greater consistency regarding these factors is essential for appropriate patient education and surgeon knowledge. The number of studies that reported details regarding surgical technique, the perioperative episode of care, as well as complications and revision are not sufficient. More importantly, the way recurrence was defined varies greatly with few studies reporting subjective instability and apprehension. Further, few high-level studies exist, as most are Level III or IV, with only one Level I study meeting inclusion criteria. The risk of bias is also high. Future studies and discussion regarding coracoid transfer patient reported outcomes require a more stringent evaluation of continued instability, complications, and a discussion regarding the periepisode of care, as the literature varies greatly, which make comparisons difficult.
Among these studies, indications for coracoid transfer vary greatly. France has the highest number of studies of any country that met inclusion criteria with all 10 studies using the Latarjet in the primary setting. Great debate exists regarding this with many reserving coracoid transfer for off-track lesions or those with recurrent instability who participate in risky activities due to some reports of high complication rates. Most studies used a subscapularis split and the Latarjet technique with traditional graft positioning, rather than the congruent arch technique. Few studies reported whether the capsule was repaired to the glenoid, but the majority performed a coracoacromial ligament repair to the capsule. Further details regarding capsular management in studies are necessary as debate still exists on the ideal capsular management.
The perioperative episode of care also varies greatly in both its reporting and method. Few studies report what imaging was obtained preoperatively or postoperatively. Fifteen of 44 studies obtained CT scans preoperatively (9 obtained 3D CTs), while 11 obtained them on all postoperative patients (4 3D CTs). Five studies obtained multiple postoperative CT scans. Only 3 studies reported that they obtained preoperative MRIs, and none obtained an MRI postoperatively. It is unknown what imaging modalities are required to achieve the best outcomes, with some believing that postoperative CT scan is useful to evaluate graft healing, while others believe that information is not helpful and prefer to rely on patient symptoms and physical examination. Postoperative treatment and physical therapy also varies greatly, with some requiring no sling or formal therapy until 1 month after surgery and others using a sling for 6 weeks with immediate formal therapy. With the lack of consensus and discussion regarding these important aspects of the perioperative episode of care, evidence-based recommendations for surgeon and patient education are limited.
The outcome scores used also vary greatly among studies. The Rowe score was the most common postoperative outcome reported with high success rates. The shortest follow-up was 1 month, and the longest was 33 months, demonstrating the importance of long-term studies in the future. The incidence of arthritic change at long-term follow-up is, therefore, not well defined. Even with long-term studies evaluating patient and radiographic outcomes, making evidence-based recommendations would be difficult because of the multiple techniques regarding capsular repair. Studies comparing these variables would be helpful.
The way recurrence was reported also varied greatly. Only eight studies reported on subjective stability and 25 with apprehension of the 72 studies. Most studies reported on subluxation and dislocation. This outlines the need for inclusion of the patient’s own beliefs regarding the stability of their shoulder and if they have the confidence in it. Defining recurrence and failure by dislocation alone is not sufficient in expressing the patients true shoulder stability. Further studies and discussions should include these variables to provide a more complete picture of the patients’ surgical success.
Of the 2,831 shoulders in which complications were discussed, there were 473 complications reported (16%), with the most common being arthritis (3.3%), followed by graft osteolysis (1.3%) and fibrotic union (2.1%), as well as hardware complications (1.6%). Some studies have reported complication rates up to 30%; this is likely dependent on considerations of what is a complication.
Because of few studies obtaining postoperative CT scans, the true incidence of healing rates, graft osteolysis, or fibrotic union is also unknown. However, importance of healing of the graft for a successful outcome is also undetermined. Hardware complications can range from devastating prominent hardware causing complete joint erosion to pain from soft tissue irritation. Much discussion exists regarding the high incidence of hardware issues with coracoid transfer. Special care must be taken for appropriate placement and of the graft and hardware to minimize this important issue.
The majority of studies did report on the incidence of reoperation with an overall incidence of 5.1%. Recurrent instability (1.4%) and hardware issues (1.9%) were the main reasons for revision surgery. Although these numbers are not exorbitantly high, the incidence of repeat instability remains after the coracoid transfer. Many believe that no matter the bone loss or clinical scenario, the coracoid transfer will be sufficient; however, published reports regarding evaluating the on-track concept after the addition of the bone block are important.
Clinical and anatomic predictors of outcomes after the Latarjet procedure for the treatment of anterior glenohumeral instability with combined glenoid and humeral bone defects.
A high risk for bias was seen in nearly all the studies. This combined with the lack of standardized reporting on outcomes and revision, as well as how instability is defined, makes comparison of published data difficult. This likely stems from the varying indications, much of which is region dependent. Greater transparency with these factors is essential from indications, the episode of care, and postoperative outcomes and follow-up to determine treatment success is needed.
Limitations of this study include the many variations of coracoid transfers, including both Bristow and Latarjet techniques. Restricting this to Latarjet procedures was considered, but it was believed that many of the international studies would be excluded and, therefore, would bias this evaluation to mostly the European and the U.S. experience. Further, arthroscopic coracoid transfers were excluded. The authors also believe that because of the breadth of techniques, these all should be represented to give the most accurate evaluation of coracoid transfer procedures, in general. Indications for coracoid transfer also likely bias these results, as performing a Latarjet after multiple arthroscopic and open stabilization procedures may have greater risk of instability when compared with the Latarjet done in the primary setting. Inherent to most systematic reviews, many of the included studies had missing or incomplete data making proper evaluation difficult. The authors attempted to compare surgical indications among studies; however, the majority lacked these specific details, which made these assessments inaccurate and incomplete.
Conclusions
Latarjet and coracoid transfer surgery varies greatly in its indications, technique, and postoperative care. Further, there is great variation in reporting of complications, as well as recurrence and failure and how it is defined. Although coracoid transfer is a successful treatment with a long history, greater consistency regarding these factors is essential for appropriate patient education and surgeon knowledge.
Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: Significance of the inverted-pear glenoid and the humeral engaging Hill-Sachs lesion.
Coracoid graft osteolysis after the Latarjet procedure for anteroinferior shoulder instability: A computed tomography scan study of twenty-six patients.
CONSORT GROUP (Consolidated Standards of Reporting Trials). The revised CONSORT statement for reporting randomized trials: explanation and elaboration.
Clinical and anatomic predictors of outcomes after the Latarjet procedure for the treatment of anterior glenohumeral instability with combined glenoid and humeral bone defects.
The authors report the following potential conflicts of interest or sources of funding: M.T.P. has received royalties from Arthrex and Elsevier and is a paid consultant for Arthrex and Joint Research Foundation. Full ICMJE author disclosure forms are available for this article online, as supplementary material.
If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
00853-7&id=gr1.jpg){kind=link}