- Email: [email protected]
Classic Versus Congruent Coracoid Positioning During the Latarjet Procedure: An In Vitro Biomechanical Comparison
Classic Versus Congruent Coracoid Positioning During the Latarjet Procedure: An In Vitro Biomechanical Comparison Harm W. Boons, M.D., Joshua W. Giles, B.E.Sc., Ilia Elkinson, M.B.B.S., James A. Johnson, Ph.D., and George S. Athwal, M.D., F.R.C.S.C.
Purpose: The purpose of this biomechanical study was to compare the classic Latarjet technique and congruent-arc modification with respect to glenohumeral stability, joint stiffness, translation, and range of motion. Methods: Eight cadaveric forequarters were tested on a shoulder simulator that applied loads independently to the conjoint tendon, long head of biceps, rotator cuff, and deltoid. The test conditions included: intact, 30% glenoid defect, and reconstruction of the defect with the classic and congruent Latarjets. The Latarjet techniques were randomly ordered, with the outcome variables being anterior dislocation, glenohumeral translation, rotational range of motion, and joint stiffness. Results: All 8 specimens dislocated after creation of a 30% glenoid defect. The classic Latarjet stabilized 7 of 8 specimens, whereas the congruent-arc modification stabilized all specimens (8/8). In abduction neutral rotation, there was no difference in joint translation between techniques (P ¼ .613). In abduction external rotation, there was significantly greater anterior humeral head translation after the congruent technique than after the classic (9.9 and 6.5 mm, respectively, P ¼ .013). Rotational range of motion was significantly reduced after classic (25.8 ) and congruent (22.2 ) transfers as compared with the 30% defect (P .041). Joint stiffness in the abducted, externally rotated position was significantly reduced in the 30% defect as compared with intact (P ¼ .012), congruent (P ¼ .015), and classic (P < .001) conditions. In all abduction positions, the intact was not significantly different from the Latarjet techniques, and the techniques did not significantly differ from each other (P .102). Conclusions: The classic and congruent-arc Latarjet techniques restore shoulder stability and motion in cases of considerable bone loss. The techniques do not substantially differ in rotational range of motion or joint stiffness. The congruent-arc technique, however, does result in significantly greater anterior humeral head translation, as compared with the classic technique, before reaching a stable non-dislocated endpoint. Clinical Relevance: On the basis of this biomechanical model, both the classic and congruent-arc Latarjet techniques can be used to stabilize a shoulder with substantial glenoid bone loss. Further clinical and biomechanical studies are required to determine if particular clinical circumstances exist where 1 technique has an advantage over the other.
T
raditional soft tissue capsulolabral repairs, in the setting of glenoid bone defects, have been associated with higher failure rates.1,2 Several surgical strategies exist for the management of shoulder instability associated with substantial glenoid bone loss. These options include glenoid bone reconstitution with allograft or autograft, transfer of the coracoid tip, the
From the HULC Bioengineering Research Laboratory, Hand and Upper Limb Centre, St. Joseph’s Health Care, University of Western Ontario, London, Ontario, Canada. The authors report the following source of funding in relation to this article: Arthrex donated the screws used for graft fixation. Received April 8, 2012; accepted September 25, 2012. Address correspondence to George S. Athwal, M.D., F.R.C.S.C., HULC, St. Joseph’s Health Care, University of Western Ontario, 268 Grosvenor Street, London, ON N6A 4L6, Canada. E-mail: [email protected] Ó 2013 by the Arthroscopy Association of North America 0749-8063/12225/$36.00 http://dx.doi.org/10.1016/j.arthro.2012.09.007
so-called Bristow procedure, and transfer of the coracoid body termed the Latarjet procedure.3,4 No particular surgical technique has shown any clinical superiority over another. The Latarjet procedure involves transfer of the coracoid process to the anterior glenoid. In the original description,5 which we have termed the classic Latarjet procedure, the coracoid is secured to the glenoid with the lateral side of the coracoid sitting flush with the glenoid articular surface. Recently, a modification to the technique, termed the congruent-arc procedure, has been described by De Beer and Roberts6; in this procedure, the coracoid is rotated 90 so that the inferior surface is flush with the glenoid face. The purported advantage of the congruent-arc procedure is that the radius of curvature of the inferior coracoid surface matches the radius of curvature of the glenoid rim,7 so that contouring of the graft is not required.6 It has also been reported that orienting the graft in the congruent-arc
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 29, No 2 (February), 2013: pp 309-316
309
310
H. W. BOONS ET AL.
method allows reconstitution of a significantly greater glenoid bone deficiency, as the coracoid is wider than it is thick.7,8 Additionally, Ghodadra et al.8 found better normalization of glenoid articular surface contact pressures and area when the coracoid graft was placed using the congruent-arc method as compared with the classic orientation; however, they did not specifically assess joint stability. Controversy exists on the ideal orientation of the coracoid graft. We are unaware of any biomechanical studies comparing the 2 Latarjet techniques in their ability to stabilize the unstable glenohumeral joint. Therefore, the purpose of this in vitro study was to compare glenohumeral joint stability, translation, and range of motion between the classic and congruent-arc Latarjet techniques. We hypothesized that both techniques would enhance joint stability without altering shoulder range of motion and that there would be no significant differences between them with respect to these biomechanical parameters.
Methods Specimens and Shoulder Simulator Eight fresh-frozen cadaveric forequarters with a mean age of 74 8 years were tested on an in vitro shoulder simulator that independently applies loads to the periarticular muscles9,10 (Fig 1). Specimens underwent
Fig 1. An in vitro shoulder simulator was used to conduct comparative testing of the classic and congruent-arc Latarjet procedures. A cadaveric shoulder girdle is mounted in the simulator (soft tissues have been removed for clarity) by cementing the inferior aspect of the scapula. To simulate physiologic muscle loads, the rotator cuff muscles (supraspinatus, infraspinatus, subscapularis, and teres minor), the deltoid (anterior, middle and posterior heads), the conjoined group, and the long head of biceps were sutured, attached to cables, and linked to pneumatic actuators. (A) Marker for optical motion capturing system. (B) Mini-actuators attached to humerus used to load conjoined group and long head of biceps. (C) The scapula mount on the simulator, which secures the specimen and allows for scapular abduction. (D) A 6-df load cell secured to the humerus via a cemented intramedullary rod. (E) Actuators used to load the rotator cuff and deltoid.
computed tomography before testing to exclude specimens with fractures, arthritis, or deformity. Prior to transection of the shoulder specimen at the humerus midshaft, the elbow transepicondylar axis was registered and saved using an Optotrack Certus optical motion capturing system (Northern Digital, Waterloo, ON, Canada). Specimen preparation involved cementing into the humeral shaft an intramedullary rod that contained a 6 degree-of-freedom (df) load cell (Mini45, ATI Industrial Automation, Apex, NC), a humeral optical marker to track 6 of motion, and 2 mini-actuators (Bimba, University Park, IL) that would be used to apply physiologic loads to the conjoined tendon group and long head of biceps (Fig 1). Skin, subcutaneous tissues, and muscle from the inferior aspect of the scapular body were removed. Specimens were rigidly attached to the simulator by cementing the inferior aspect of the scapular body in 12 of declination. To simulate physiologic muscle loads, the musculotendinous junctions of the rotator cuff muscles (supraspinatus, infraspinatus, subscapularis, and teres minor); the deltoid (anterior, middle, and posterior heads); the conjoined group; and the long head of biceps were sutured (No. 5 Ethibond, Ethicon, Somerville, NJ) and attached to cables linked to pneumatic actuators. Computer-controlled pneumatic actuators were used to independently apply loads to the muscles. Muscles were loaded as follows9-11: anterior deltoid, 5 N; middle deltoid, 5 N; posterior deltoid, 5 N; supraspinatus, 7.5 N; subscapularis, 7.5 N; infraspinatus and teres minor together, 7.5 N; conjoined tendon, 10 N; and the long head of biceps, 5 N. The simulator allows unconstrained glenohumeral motion and allows scapular abduction. Composite shoulder abduction is accomplished by using a glenohumeral-to-scapulothoracic rhythm of 2:112; for example, 90 of composite abduction consists of 60 glenohumeral and 30 scapulothoracic abduction. A clinically relevant coordinate system was created to assess instantaneous glenohumeral rotations and translations. Testing Protocol To repeatedly access the glenohumeral joint to create the defects and perform the Latarjet procedures, an extended lesser tuberosity osteotomy was conducted. This osteotomy was conducted with a microsagittal saw and included the entire lesser tuberosity and approximately 1 cm of proximal anterior humeral cortex to allow secure fixation with 2 3.5-mm bicortical bolts. The lesser tuberosity osteotomy has been previously validated with this experimental model and found to be similar to the intact condition, with no significant differences in range of motion, stiffness, or joint translation.9,10 The following test conditions were evaluated: intact specimen, Bankart lesion with an anteroinferior
CORACOID POSITIONING DURING LATARJET PROCEDURE
capsulotomy, 30% anterior glenoid defect, a Latarjet procedure with the coracoid oriented in the classic manner, and a congruent-arc Latarjet. The Bankart lesion and the anteroinferior capsulotomy were created from the 3-o’clock to 6-o’clock position (right shoulder) on the glenoid. Once the Bankart lesion and the capsulotomy were created, the humerus was placed into abduction and external rotation and forcibly dislocated in the anteroinferior direction to propagate the surgical lesions to ensure anterior instability. The 30% anterior glenoid bone defect was created using the clinical description of bone loss described by Sugaya et al.13 The diameter of the inferior circle of the glenoid was measured and a 30% defect was created with a microsagittal saw oriented toward the 3 o’clock position (right shoulder). The size of the defect was confirmed with digitizations using the Optotrack Certus optical motion capturing system (Northern Digital), which has an accuracy of 0.1 mm and a resolution of 0.01 mm. We selected a 30% glenoid defect as this was the size tested by Ghodadra et al.8 and Wellmann et al.14 Additionally, we endeavored to create a defect
Fig 2. The Latarjet procedure involves transfer of the coracoid and the attached conjoined tendon to the anterior glenoid. (A, B) The original description of the procedure, termed the classic Latarjet, involves securing the coracoid to the glenoid with its inferior surface compressed to the glenoid vault and its lateral edge sitting flush with the glenoid articular surface. (C, D) The congruent-arc modification of the Latarjet involves rotating the coracoid graft 90 so that the inferior surface is flush with the glenoid face and the medial edge is secured to the anterior glenoid.
311
where there would be minimal controversy in selecting the Latarjet procedure as a reconstructive option. The classic Latarjet procedure5 was performed by osteotomizing the coracoid body at its angle and transferring it to the anterior glenoid to reconstitute the bone defect (Fig 2A and B). The graft was secured with 2 bicortical 3.75-mm titanium screws (Arthrex, Naples, FL). The congruent-arc modification was conducted as outlined by De Beer et al.6 The osteotomized coracoid was rotated 90 and the inferior surface of the coracoid was positioned flush with the glenoid articular surface and secured with two 3.75-mm titanium screws (Fig 2C and D). Both techniques were performed on each specimen, with the order randomized. The coracoid graft was placed through a subscapularis split between the upper two-thirds and the lower one-third. The split in the subscapularis was not repaired after the transfer. The conjoint tendon was passed through the subscapularis split and attached to an independent miniactuator, which was attached to the midhumeral shaft. This actuator provided a constant force on the conjoint group in an effort to recreate the normal physiologic function of
312
H. W. BOONS ET AL.
the muscle group in the native and also in the transferred position in various shoulder positions. The position of the corocoid graft was critical; therefore, we endeavored to place the graft in an identical manner from specimen to specimen. The inferior tip of the coracoid, for both the classic and congruent techniques, was placed at the inferiormost aspect of the glenoid osteotomy. This position was mapped and confirmed with the optical tracking system. Shoulder stability was quantified in terms of prevention of dislocation and glenohumeral joint stiffness (N/mm). The property of glenohumeral joint stiffness was obtained by applying an anteriorly directed 80-N force to the humeral head while measuring the magnitude of head displacement with the motion capture system until a soft tissue endpoint was reached (stability) or until dislocation. Dislocation was defined as a sudden and rapid medial displacement of the humeral head rotational center, as recorded by the motion capture system and confirmed by visual observation by 2 observers. Joint stiffness, therefore, was the extent to which the glenohumeral joint resisted anterior translation to an externally applied force. We chose the biomechanical assessment of stiffness as one of our stability assessment outcomes because it is robust enough to incorporate both the bony shoulder joint stabilizers and the contributions of the soft tissues. We believed this was an important parameter when assessing the Latarjet as it incorporates the bony effect of the transferred coracoid and also the dynamic soft tissue sling effect. The quasistatic 80-N force was applied through a uniaxial load cell (Model 34 Precision Miniature, Honeywell, Golden Valley, MN). Joint stiffness was tested in adduction and 90 of composite abduction with the humerus in both neutral rotation and 60 of external rotation. Internal and external rotational range of motion of the glenohumeral joint was recorded in adduction and 90 of composite abduction. Neutral rotation corresponded to the transepicondylar axis being parallel to the body plane, at both levels of elevation. Motion assessment started with the arm in neutral rotation, followed by maximal internal rotation to a predefined torque (0.8 Nm) and then maximal external rotation to the same torque. This torque was determined by taking the average value obtained from repeated trials by an orthopaedic surgeon (G.S.A.) via rotation of the humerus until a resistance consistent with a clinical evaluation was reached. Outcome Variables and Statistical Analyses The outcome variables for each condition in the testing protocol included glenohumeral joint translation, humeral head dislocation, joint stiffness, and rotational range of motion. Glenohumeral joint translation was assessed in shoulder adduction, 90 of composite
abduction, neutral rotation, and external rotation. We chose to test 8 specimens, as this was the number chosen by Wellmann et al.14 and Yamamoto et al.15 A repeated-measures study design was used to evaluate test conditions. To assess overall trends in the joint stiffness outcome variable, a 3-way analysis of variance across all configurations was performed using SPSS Version 17.0 (SPSS, Chicago, IL). Additionally, 1-way analyses of variance and pairwise comparison procedures were performed for all outcome variables at each tested shoulder configuration to assess specific joint configurations and conditions. Statistical significance was attained when P < .05.
Results Joint Stability and Translation All 8 specimens dislocated after creation of a 30% anterior glenoid bone defect in abduction neutral and abduction external rotation. The classic Latarjet technique resulted in 1 specimen dislocating in abduction neutral rotation and none in abduction external rotation. The congruent-arc technique effectively prevented dislocation in all specimens in both positions. In abduction neutral rotation, there was no difference in glenohumeral joint translation between the classic and congruent-arc reconstructions (15.7 and 17.2 mm, respectively; mean difference, 1.5 8.0 mm; P ¼ .613). In abduction external rotation, however, the congruentarc technique resulted in significantly greater anterior humeral head translation on the glenoid before reaching a stable soft tissue endpoint, as compared with the classic (6.5 and 9.9 mm, respectively; mean difference, 3.5 2.9 mm; P ¼ .013). Please see Table 1 for summary of conditions and outcome variables. Joint Stiffness According to the results of a 3-way ANOVA, joint stiffness was significantly affected by the level of internaleexternal rotation across all joint conditions for adduction and abduction (P ¼ .001). As well, there is a significant trend between joint conditions across both levels of abduction and degrees of internaleexternal rotation (P ¼ .008). One-way analysis of variance in adduction neutral rotation found that the unstable 30% glenoid defect had significantly lower stiffness than the intact state (mean difference, 12.6 8.6 N/mm; P ¼ .048) and trended toward lower stiffness than the congruent-arc technique (mean difference, 3.7 2.6 N/mm; P ¼ .052); however, it was not significantly different from the classic technique (mean difference, 5.6 5.2 N/mm, P ¼ .168). In adduction external rotation, both the classic (mean difference, 8.2 3.2 N/mm; P ¼ .003) and the congruentarc (mean difference, 6.1 3.2 N/mm; P ¼ .013) techniques significantly enhanced joint stiffness as compared with the 30% anterior glenoid defect condition. Stiffness
313
CORACOID POSITIONING DURING LATARJET PROCEDURE Table 1. Conditions and Outcome Variables Condition Parameter
Intact
Dislocation 0/8 Mean anterior humeral head translation (mm) Abduction and neutral rotation 8.1 Abduction and external rotation 3.9 Mean joint stiffness (N/mm) Adduction and neutral rotation 14.2 Adduction and external rotation 18.0 Abduction and neutral rotation 9.9 Abduction and external rotation 23.8 Mean rotational range of motion ( ) Adduction 47 Abduction 61
Bankart Lesion
30% Glenoid Defect
Classic Latarjet
Congruent-Arc Latarjet
ANOVA (P value)
8/8
8/8
1/8
0/8
d
* *
* *
15.7 6.5
17.2 9.9
.694 .021
4.75 6.75 3.9 8.1
1.6 1.8 1.8 1.9
7.2 10.0 4.2 9.2
5.3 7.9 4.2 8.9
.019 .033 .005 .004
46 77
52 75
42 49
43 53
.531 .001
*Translation not calculated as all specimens dislocated.
values of the intact specimen never significantly differed from those of either the classic or congruent-arc technique, and there was no significant difference between the 2 techniques (P .234) (Fig 3, Table 1). In abduction neutral rotation, the 30% anterior glenoid defect had significantly lower stiffness than the intact state (mean difference, 8.1 5.5N/mm; P ¼ .026) and the congruent-arc technique (mean difference, 2.4 1.7 N/mm; P ¼ .027); however, the stiffness did not significantly differ from that of the classic technique (mean difference, 2.4 2.7 N/mm, P ¼ .235). In abduction external rotation, the 30% glenoid defect state had significantly lower stiffness values than the intact state (mean difference, 21.9 13.0 N/mm, P ¼ .012), the classic Latarjet (7.3 1.4 N/mm, P < .001), and the congruent-arc technique (7.0 4.2 N/mm, P ¼ .015). In all abduction test positions, the intact state did not significantly differ from the 2 Latarjet techniques, and the techniques did not significantly differ from each other (P .102).
Fig 3. Glenohumeral joint stiffness values (mean and SD) for the 4 conditions tested (intact specimen, 30% anterior glenoid defect, classic Latarjet, and congruent-arc Latarjet) in the different joint configurations. Symbols denote statistically significant (P < .05) differences.
Range of Motion In adduction, no significant differences in internale external rotational range of motion were observed between any test conditions (P > .05) (Fig 4). In abduction, internaleexternal rotational range of motion was significantly reduced after the classic (mean difference, 25.8 18.0 , P ¼ .032) and congruent-arc (mean difference, 22.2 16.7 , P ¼ .041) Latarjet procedures as compared with the 30% anterior glenoid defect. There were no significant differences between the 2 techniques and the intact condition, nor did the techniques significantly differ from each other (P .282) (Table 1).
Discussion The results of this study indicate that both the classic and congruent-arc Latarjet orientations were able to eliminate dislocation in all specimens, except for 1 specimen treated with the classic repair when tested in the abducted neutral configuration. However,
314
H. W. BOONS ET AL.
Fig 4. Mean rotational range of motion values for the 4 conditions tested (intact specimen, 30% anterior glenoid defect, classic Latarjet, and congruent-arc Latarjet) in adduction and 90 of composite abduction. Symbols denote statistically significant (P < .05) differences.
a difference was found in the magnitude of anterior joint translation experienced by the humeral head when tested in the position of apprehension. Specifically, the congruent-arc orientation allowed significantly greater anterior humeral head translation before reaching a stable soft tissue endpoint. This is not unexpected, as orienting the coracoid in the congruent-arc manner results in greater bone reconstitution,7 which leads to a greater arc length of the glenoid articular surface. The kinematics of constructing a wider glenoid articular surface, however, are unknown. It is conceivable that when reconstructing a large glenoid bone deficit with the classic Latarjet, the associated decreased humeral head translations may constrain the joint, which may predispose to arthritis as the humeral head articulates with a narrower glenoid. With the congruent technique, the additional glenoid arc length may only come into play during instability-provoking maneuvers, which would then allow greater contained translation before dislocation. Unfortunately, the corollary of a greater surface area for articulation with the congruent-arc technique results in a smaller surface area for graft-toglenoid healing. This is a potential disadvantage of the congruent-arc method, as the contact area between the coracoid graft and the glenoid is decreased. This occurs because the coracoid is wider than it is thick, which results in a smaller surface area of coracoid contacting the glenoid for bony union.7 In our specimens, the mean width, thickness, and length of the coracoid grafts were 16 1, 10 2, and 28 2 mm, respectively. These corocoid dimensions are in keeping with those reported by Armitage et al.7 One specimen in the classic Latarjet group continued to dislocate. We have no explanation for why this specimen continued to remain unstable after the classic Latarjet procedure. The specimen, orientation of the graft, fixation of the graft, and size of the glenoid bone defect were critically examined to determine if any
deviations from the protocol had occurred or if there were any failures, and nothing was identified. This same specimen, however, was effectively stabilized with the congruent-arc Latarjet. Glenohumeral joint stiffness, along with joint dislocation and translation, was used as an objective outcome measure to describe the concept of joint stability. Stiffness was a measure of the humeral head’s ability to resist anterior translation and dislocation by an externally applied force. Our results indicate that there were no significant differences in joint stiffness between the intact condition, the classic Latarjet, and the congruent-arc modification. Universally, the stiffness values obtained for the 30% anterior glenoid defect were substantially lower than those for the intact condition and the 2 Latarjet techniques. Additionally, during evaluation of internaleexternal rotation range of-motion of the glenohumeral joint, it was confirmed that neither technique would cause significant changes in this outcome measure as compared with the intact condition. Ghodadra et al.8 biomechanically compared the glenohumeral contact pressures of the classic Latarjet, the congruent-arc modification, and iliac crest structural grafting. They reported that the optimum position for restoration of intact contact pressures was a graft placed flush to the glenoid surface. Therefore, we chose only to compare the different Latarjet techniques in the flush position. Additionally, they reported that the congruentarc modification yielded displayed better contact pressure restoration than the classic orientation. The findings of improved contact pressures is not unexpected, as Armitage et al.7 reported that a coracoid graft positioned in the congruent-arc manner has a radius of curvature nearly identical to that of the intact anterior glenoid rim. Wellmann et al.14 reported that the Latarjet procedure, with the coracoid oriented in the classic manner, is biomechanically superior to a contoured structural bone
CORACOID POSITIONING DURING LATARJET PROCEDURE
graft in reducing anterioinferior glenohumeral joint translation. These improved biomechanical properties may be attributed to the stabilizing “sling effect” of the conjoint tendon. Wellmann et al.14 chose to load the conjoined tendon with 10 N of force applied in a static manner. Giles et al.11 conducted a biomechanical simulator study assessing the role of the conjoined tendon in shoulder stability. They recommended applying 10 N of force to the conjoined tendon, as it was the minimum amount of force that had substantial effects on joint stability without influencing humeral head translation. Therefore, in the present study we applied 10 N of force to the conjoined tendon in a dynamic fashion through an actuator attached to the humeral shaft, mimicking the line of action of the native conjoined group. Shoulder instability with glenoid bone deficiency can present clinically as various-sized defects. Yamamoto et al.16 recently reported that a glenoid defect with a width greater than 19% of the glenoid height remains unstable after soft tissue Bankart repair. In the present study, only a 30% anterior glenoid bone deficiency was tested. We chose a 30% deficiency because it represents a larger-sized defect that would typically undergo bone grafting and because this size of defect was tested in previous biomechanical studies.8,14 Although congruent-arc positioning allows reconstitution of greater glenoid bone loss, our results reveal no significant differences in the outcomes assessed, other than the magnitude of anterior humeral head translation. Additionally, the strength of the congruent-arc reconstruction is unknown. It is conceivable that the congruent-arc orientation has less strength as it has less bony support because the graft is thinner. This lack of support may lead to cantilever bending moments, potentially increasing the risk of non-union and fixation failure. In contrast, the classic technique may have improved strength; however, it has inferior glenohumeral contact pressures8 and requires intraoperative graft contouring. Both techniques have surgeon advocates, advantages, and disadvantages.3,4,17-20 Limitations In this biomechanical simulator study we tried to simulate native human shoulder conditions. There are, however, limitations to the study, especially those that are inherent to all biomechanical simulations, such as the use of elderly donor specimens. Also, the results of this study represent time zero kinematics; the effects of soft tissue healing and adaptive changes are unknown. Another limitation is that we did not conduct our own power analysis; we used the power analysis of Yamamoto et al.,15 which concluded that 8 specimens were necessary. As such, because Yamamoto et al.15 and Wellmann et al.14 used 8 specimens, we also used 8 specimens. Additionally, we used the same shoulder specimens to perform both the classic and congruent-arc
315
techniques and, therefore, used the same coracoid process. This was accomplished by drilling 2 tunnels for the classic technique and 2 tunnels for the congruent-arc technique; therefore, the additional drill tunnels may have weakened the graft. This potential weakening of the coracoid graft, however, did not affect stability testing as all grafts were securely fixed and none fractured or failed during testing. The difference in humeral head translation between the classic and congruent techniques was statistically significant; however, a mean change of 3.5 mm may not be clinically relevant. Finally, the results of this study are applicable to the 30% glenoid bone loss model created. Therefore, the Latarjet procedure would likely have behaved differently if tested on smaller-sized glenoid defects.
Conclusions The classic and congruent-arc Latarjet techniques restore shoulder stability and motion in cases of considerable bone loss. The techniques do not substantially differ in rotational range of motion or joint stiffness. However, the congruent-arc technique does result in significantly greater anterior humeral head translation than with the classic technique, before reaching a stable non-dislocated endpoint.
Acknowledgment The authors acknowledge the assistance of Dr. Kenneth Faber, Dr. Robert Litchfield, and Dr. Louis Ferreira.
References 1. Boileau P, Villalba M, Hery JY, Balg F, Ahrens P, Neyton L. Risk factors for recurrence of shoulder instability after arthroscopic Bankart repair. J Bone Joint Surg Am 2006;88:1755-1763. 2. Burkhart SS, De Beer JF. Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: Significance of the inverted-pear glenoid and the humeral engaging HilleSachs lesion. Arthroscopy 2000;16:677-694. 3. Beran MC, Donaldson CT, Bishop JY. Treatment of chronic glenoid defects in the setting of recurrent anterior shoulder instability: A systematic review. J Shoulder Elbow Surg 2010;19:769-780. 4. Provencher MT, Bhatia S, Ghodadra NS, et al. Recurrent shoulder instability: Current concepts for evaluation and management of glenoid bone loss. J Bone Joint Surg Am 2010;92(Suppl 2):133-151. 5. Latarjet M. Treatment of recurrent dislocation of the shoulder. Lyon Chir 1954;49:994-997. 6. De Beer JF, Roberts C. Glenoid bone defects: Open Latarjet with congruent arc modification. Orthop Clin North Am 2010;41:407-415. 7. Armitage MS, Elkinson I, Giles JW, Athwal GS. An anatomic, computed tomographic assessment of the coracoid process with special reference to the congruentarc Latarjet procedure. Arthroscopy 2011;27:1485-1489.
316
H. W. BOONS ET AL.
8. Ghodadra N, Gupta A, Romeo AA, et al. Normalization of glenohumeral articular contact pressures after Latarjet or iliac crest bone-grafting. J Bone Joint Surg Am 2010;92: 1478-1489. 9. Elkinson I, Giles JW, Faber KJ, et al. The effect of the remplissage procedure on shoulder stability and range of motion: An in vitro biomechanical assessment. J Bone Joint Surg Am 2012;94:1003-1012. 10. Giles JW, Elkinson I, Ferreira LM, et al. Moderate to large engaging HilleSachs defects: An in vitro biomechanical comparison of the remplissage procedure, allograft humeral head reconstruction, and partial resurfacing arthroplasty. J Shoulder Elbow Surg 2012;21:1142-1151. Epub 2011 Oct 29. 11. Giles JW, Boons HW, Ferreira LM, Johnson JA, Athwal GS. The effect of the conjoined tendon of the short head of the biceps and coracobrachialis on shoulder stability and kinematics during in-vitro simulation. J Biomech 2011;44:1192-1195. 12. McQuade KJ, Smidt GL. Dynamic scapulohumeral rhythm: The effects of external resistance during elevation of the arm in the scapular plane. J Orthop Sports Phys Ther 1998;27:125-133. 13. 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.
14. Wellmann M, Petersen W, Zantop T, et al. Open shoulder repair of osseous glenoid defects: Biomechanical effectiveness of the Latarjet procedure versus a contoured structural bone graft. Am J Sports Med 2009;37:87-94. 15. Yamamoto N, Itoi E, Abe H, et al. Effect of an anterior glenoid defect on anterior shoulder stability: A cadaveric study. Am J Sports Med 2009;37:949-954. 16. Yamamoto N, Muraki T, Sperling JW, et al. Stabilizing mechanism of bone-graft of a large glenoid defect. J Bone Joint Surg Am 2012;92:2059-2066. 17. Burkhart SS, De Beer JF, Barth JR, Cresswell T, Roberts C, Richards DP. Results of modified Latarjet reconstruction in patients with anteroinferior instability and significant bone loss. Arthroscopy 2007;23:1033-1041. 18. Chen AL, Hunt SA, Hawkins RJ, Zuckerman JD. Management of bone loss associated with recurrent anterior glenohumeral instability. Am J Sports Med 2005;33:912-925. 19. Millett PJ, Clavert P, Warner JJ. Open operative treatment for anterior shoulder instability: When and why? J Bone Joint Surg Am 2005;87:419-432. 20. Piasecki DP, Verma NN, Romeo AA, Levine WN, Bach BR Jr, Provencher MT. Glenoid bone deficiency in recurrent anterior shoulder instability: Diagnosis and management. J Am Acad Orthop Surg 2009;17: 482-493.
Purpose: The purpose of this biomechanical study was to compare the classic Latarjet technique and congruent-arc modification with respect to glenohumeral stability, joint stiffness, translation, and range of motion. Methods: Eight cadaveric forequarters were tested on a shoulder simulator that applied loads independently to the conjoint tendon, long head of biceps, rotator cuff, and deltoid. The test conditions included: intact, 30% glenoid defect, and reconstruction of the defect with the classic and congruent Latarjets. The Latarjet techniques were randomly ordered, with the outcome variables being anterior dislocation, glenohumeral translation, rotational range of motion, and joint stiffness. Results: All 8 specimens dislocated after creation of a 30% glenoid defect. The classic Latarjet stabilized 7 of 8 specimens, whereas the congruent-arc modification stabilized all specimens (8/8). In abduction neutral rotation, there was no difference in joint translation between techniques (P ¼ .613). In abduction external rotation, there was significantly greater anterior humeral head translation after the congruent technique than after the classic (9.9 and 6.5 mm, respectively, P ¼ .013). Rotational range of motion was significantly reduced after classic (25.8 ) and congruent (22.2 ) transfers as compared with the 30% defect (P .041). Joint stiffness in the abducted, externally rotated position was significantly reduced in the 30% defect as compared with intact (P ¼ .012), congruent (P ¼ .015), and classic (P < .001) conditions. In all abduction positions, the intact was not significantly different from the Latarjet techniques, and the techniques did not significantly differ from each other (P .102). Conclusions: The classic and congruent-arc Latarjet techniques restore shoulder stability and motion in cases of considerable bone loss. The techniques do not substantially differ in rotational range of motion or joint stiffness. The congruent-arc technique, however, does result in significantly greater anterior humeral head translation, as compared with the classic technique, before reaching a stable non-dislocated endpoint. Clinical Relevance: On the basis of this biomechanical model, both the classic and congruent-arc Latarjet techniques can be used to stabilize a shoulder with substantial glenoid bone loss. Further clinical and biomechanical studies are required to determine if particular clinical circumstances exist where 1 technique has an advantage over the other.
T
raditional soft tissue capsulolabral repairs, in the setting of glenoid bone defects, have been associated with higher failure rates.1,2 Several surgical strategies exist for the management of shoulder instability associated with substantial glenoid bone loss. These options include glenoid bone reconstitution with allograft or autograft, transfer of the coracoid tip, the
From the HULC Bioengineering Research Laboratory, Hand and Upper Limb Centre, St. Joseph’s Health Care, University of Western Ontario, London, Ontario, Canada. The authors report the following source of funding in relation to this article: Arthrex donated the screws used for graft fixation. Received April 8, 2012; accepted September 25, 2012. Address correspondence to George S. Athwal, M.D., F.R.C.S.C., HULC, St. Joseph’s Health Care, University of Western Ontario, 268 Grosvenor Street, London, ON N6A 4L6, Canada. E-mail: [email protected] Ó 2013 by the Arthroscopy Association of North America 0749-8063/12225/$36.00 http://dx.doi.org/10.1016/j.arthro.2012.09.007
so-called Bristow procedure, and transfer of the coracoid body termed the Latarjet procedure.3,4 No particular surgical technique has shown any clinical superiority over another. The Latarjet procedure involves transfer of the coracoid process to the anterior glenoid. In the original description,5 which we have termed the classic Latarjet procedure, the coracoid is secured to the glenoid with the lateral side of the coracoid sitting flush with the glenoid articular surface. Recently, a modification to the technique, termed the congruent-arc procedure, has been described by De Beer and Roberts6; in this procedure, the coracoid is rotated 90 so that the inferior surface is flush with the glenoid face. The purported advantage of the congruent-arc procedure is that the radius of curvature of the inferior coracoid surface matches the radius of curvature of the glenoid rim,7 so that contouring of the graft is not required.6 It has also been reported that orienting the graft in the congruent-arc
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 29, No 2 (February), 2013: pp 309-316
309
310
H. W. BOONS ET AL.
method allows reconstitution of a significantly greater glenoid bone deficiency, as the coracoid is wider than it is thick.7,8 Additionally, Ghodadra et al.8 found better normalization of glenoid articular surface contact pressures and area when the coracoid graft was placed using the congruent-arc method as compared with the classic orientation; however, they did not specifically assess joint stability. Controversy exists on the ideal orientation of the coracoid graft. We are unaware of any biomechanical studies comparing the 2 Latarjet techniques in their ability to stabilize the unstable glenohumeral joint. Therefore, the purpose of this in vitro study was to compare glenohumeral joint stability, translation, and range of motion between the classic and congruent-arc Latarjet techniques. We hypothesized that both techniques would enhance joint stability without altering shoulder range of motion and that there would be no significant differences between them with respect to these biomechanical parameters.
Methods Specimens and Shoulder Simulator Eight fresh-frozen cadaveric forequarters with a mean age of 74 8 years were tested on an in vitro shoulder simulator that independently applies loads to the periarticular muscles9,10 (Fig 1). Specimens underwent
Fig 1. An in vitro shoulder simulator was used to conduct comparative testing of the classic and congruent-arc Latarjet procedures. A cadaveric shoulder girdle is mounted in the simulator (soft tissues have been removed for clarity) by cementing the inferior aspect of the scapula. To simulate physiologic muscle loads, the rotator cuff muscles (supraspinatus, infraspinatus, subscapularis, and teres minor), the deltoid (anterior, middle and posterior heads), the conjoined group, and the long head of biceps were sutured, attached to cables, and linked to pneumatic actuators. (A) Marker for optical motion capturing system. (B) Mini-actuators attached to humerus used to load conjoined group and long head of biceps. (C) The scapula mount on the simulator, which secures the specimen and allows for scapular abduction. (D) A 6-df load cell secured to the humerus via a cemented intramedullary rod. (E) Actuators used to load the rotator cuff and deltoid.
computed tomography before testing to exclude specimens with fractures, arthritis, or deformity. Prior to transection of the shoulder specimen at the humerus midshaft, the elbow transepicondylar axis was registered and saved using an Optotrack Certus optical motion capturing system (Northern Digital, Waterloo, ON, Canada). Specimen preparation involved cementing into the humeral shaft an intramedullary rod that contained a 6 degree-of-freedom (df) load cell (Mini45, ATI Industrial Automation, Apex, NC), a humeral optical marker to track 6 of motion, and 2 mini-actuators (Bimba, University Park, IL) that would be used to apply physiologic loads to the conjoined tendon group and long head of biceps (Fig 1). Skin, subcutaneous tissues, and muscle from the inferior aspect of the scapular body were removed. Specimens were rigidly attached to the simulator by cementing the inferior aspect of the scapular body in 12 of declination. To simulate physiologic muscle loads, the musculotendinous junctions of the rotator cuff muscles (supraspinatus, infraspinatus, subscapularis, and teres minor); the deltoid (anterior, middle, and posterior heads); the conjoined group; and the long head of biceps were sutured (No. 5 Ethibond, Ethicon, Somerville, NJ) and attached to cables linked to pneumatic actuators. Computer-controlled pneumatic actuators were used to independently apply loads to the muscles. Muscles were loaded as follows9-11: anterior deltoid, 5 N; middle deltoid, 5 N; posterior deltoid, 5 N; supraspinatus, 7.5 N; subscapularis, 7.5 N; infraspinatus and teres minor together, 7.5 N; conjoined tendon, 10 N; and the long head of biceps, 5 N. The simulator allows unconstrained glenohumeral motion and allows scapular abduction. Composite shoulder abduction is accomplished by using a glenohumeral-to-scapulothoracic rhythm of 2:112; for example, 90 of composite abduction consists of 60 glenohumeral and 30 scapulothoracic abduction. A clinically relevant coordinate system was created to assess instantaneous glenohumeral rotations and translations. Testing Protocol To repeatedly access the glenohumeral joint to create the defects and perform the Latarjet procedures, an extended lesser tuberosity osteotomy was conducted. This osteotomy was conducted with a microsagittal saw and included the entire lesser tuberosity and approximately 1 cm of proximal anterior humeral cortex to allow secure fixation with 2 3.5-mm bicortical bolts. The lesser tuberosity osteotomy has been previously validated with this experimental model and found to be similar to the intact condition, with no significant differences in range of motion, stiffness, or joint translation.9,10 The following test conditions were evaluated: intact specimen, Bankart lesion with an anteroinferior
CORACOID POSITIONING DURING LATARJET PROCEDURE
capsulotomy, 30% anterior glenoid defect, a Latarjet procedure with the coracoid oriented in the classic manner, and a congruent-arc Latarjet. The Bankart lesion and the anteroinferior capsulotomy were created from the 3-o’clock to 6-o’clock position (right shoulder) on the glenoid. Once the Bankart lesion and the capsulotomy were created, the humerus was placed into abduction and external rotation and forcibly dislocated in the anteroinferior direction to propagate the surgical lesions to ensure anterior instability. The 30% anterior glenoid bone defect was created using the clinical description of bone loss described by Sugaya et al.13 The diameter of the inferior circle of the glenoid was measured and a 30% defect was created with a microsagittal saw oriented toward the 3 o’clock position (right shoulder). The size of the defect was confirmed with digitizations using the Optotrack Certus optical motion capturing system (Northern Digital), which has an accuracy of 0.1 mm and a resolution of 0.01 mm. We selected a 30% glenoid defect as this was the size tested by Ghodadra et al.8 and Wellmann et al.14 Additionally, we endeavored to create a defect
Fig 2. The Latarjet procedure involves transfer of the coracoid and the attached conjoined tendon to the anterior glenoid. (A, B) The original description of the procedure, termed the classic Latarjet, involves securing the coracoid to the glenoid with its inferior surface compressed to the glenoid vault and its lateral edge sitting flush with the glenoid articular surface. (C, D) The congruent-arc modification of the Latarjet involves rotating the coracoid graft 90 so that the inferior surface is flush with the glenoid face and the medial edge is secured to the anterior glenoid.
311
where there would be minimal controversy in selecting the Latarjet procedure as a reconstructive option. The classic Latarjet procedure5 was performed by osteotomizing the coracoid body at its angle and transferring it to the anterior glenoid to reconstitute the bone defect (Fig 2A and B). The graft was secured with 2 bicortical 3.75-mm titanium screws (Arthrex, Naples, FL). The congruent-arc modification was conducted as outlined by De Beer et al.6 The osteotomized coracoid was rotated 90 and the inferior surface of the coracoid was positioned flush with the glenoid articular surface and secured with two 3.75-mm titanium screws (Fig 2C and D). Both techniques were performed on each specimen, with the order randomized. The coracoid graft was placed through a subscapularis split between the upper two-thirds and the lower one-third. The split in the subscapularis was not repaired after the transfer. The conjoint tendon was passed through the subscapularis split and attached to an independent miniactuator, which was attached to the midhumeral shaft. This actuator provided a constant force on the conjoint group in an effort to recreate the normal physiologic function of
312
H. W. BOONS ET AL.
the muscle group in the native and also in the transferred position in various shoulder positions. The position of the corocoid graft was critical; therefore, we endeavored to place the graft in an identical manner from specimen to specimen. The inferior tip of the coracoid, for both the classic and congruent techniques, was placed at the inferiormost aspect of the glenoid osteotomy. This position was mapped and confirmed with the optical tracking system. Shoulder stability was quantified in terms of prevention of dislocation and glenohumeral joint stiffness (N/mm). The property of glenohumeral joint stiffness was obtained by applying an anteriorly directed 80-N force to the humeral head while measuring the magnitude of head displacement with the motion capture system until a soft tissue endpoint was reached (stability) or until dislocation. Dislocation was defined as a sudden and rapid medial displacement of the humeral head rotational center, as recorded by the motion capture system and confirmed by visual observation by 2 observers. Joint stiffness, therefore, was the extent to which the glenohumeral joint resisted anterior translation to an externally applied force. We chose the biomechanical assessment of stiffness as one of our stability assessment outcomes because it is robust enough to incorporate both the bony shoulder joint stabilizers and the contributions of the soft tissues. We believed this was an important parameter when assessing the Latarjet as it incorporates the bony effect of the transferred coracoid and also the dynamic soft tissue sling effect. The quasistatic 80-N force was applied through a uniaxial load cell (Model 34 Precision Miniature, Honeywell, Golden Valley, MN). Joint stiffness was tested in adduction and 90 of composite abduction with the humerus in both neutral rotation and 60 of external rotation. Internal and external rotational range of motion of the glenohumeral joint was recorded in adduction and 90 of composite abduction. Neutral rotation corresponded to the transepicondylar axis being parallel to the body plane, at both levels of elevation. Motion assessment started with the arm in neutral rotation, followed by maximal internal rotation to a predefined torque (0.8 Nm) and then maximal external rotation to the same torque. This torque was determined by taking the average value obtained from repeated trials by an orthopaedic surgeon (G.S.A.) via rotation of the humerus until a resistance consistent with a clinical evaluation was reached. Outcome Variables and Statistical Analyses The outcome variables for each condition in the testing protocol included glenohumeral joint translation, humeral head dislocation, joint stiffness, and rotational range of motion. Glenohumeral joint translation was assessed in shoulder adduction, 90 of composite
abduction, neutral rotation, and external rotation. We chose to test 8 specimens, as this was the number chosen by Wellmann et al.14 and Yamamoto et al.15 A repeated-measures study design was used to evaluate test conditions. To assess overall trends in the joint stiffness outcome variable, a 3-way analysis of variance across all configurations was performed using SPSS Version 17.0 (SPSS, Chicago, IL). Additionally, 1-way analyses of variance and pairwise comparison procedures were performed for all outcome variables at each tested shoulder configuration to assess specific joint configurations and conditions. Statistical significance was attained when P < .05.
Results Joint Stability and Translation All 8 specimens dislocated after creation of a 30% anterior glenoid bone defect in abduction neutral and abduction external rotation. The classic Latarjet technique resulted in 1 specimen dislocating in abduction neutral rotation and none in abduction external rotation. The congruent-arc technique effectively prevented dislocation in all specimens in both positions. In abduction neutral rotation, there was no difference in glenohumeral joint translation between the classic and congruent-arc reconstructions (15.7 and 17.2 mm, respectively; mean difference, 1.5 8.0 mm; P ¼ .613). In abduction external rotation, however, the congruentarc technique resulted in significantly greater anterior humeral head translation on the glenoid before reaching a stable soft tissue endpoint, as compared with the classic (6.5 and 9.9 mm, respectively; mean difference, 3.5 2.9 mm; P ¼ .013). Please see Table 1 for summary of conditions and outcome variables. Joint Stiffness According to the results of a 3-way ANOVA, joint stiffness was significantly affected by the level of internaleexternal rotation across all joint conditions for adduction and abduction (P ¼ .001). As well, there is a significant trend between joint conditions across both levels of abduction and degrees of internaleexternal rotation (P ¼ .008). One-way analysis of variance in adduction neutral rotation found that the unstable 30% glenoid defect had significantly lower stiffness than the intact state (mean difference, 12.6 8.6 N/mm; P ¼ .048) and trended toward lower stiffness than the congruent-arc technique (mean difference, 3.7 2.6 N/mm; P ¼ .052); however, it was not significantly different from the classic technique (mean difference, 5.6 5.2 N/mm, P ¼ .168). In adduction external rotation, both the classic (mean difference, 8.2 3.2 N/mm; P ¼ .003) and the congruentarc (mean difference, 6.1 3.2 N/mm; P ¼ .013) techniques significantly enhanced joint stiffness as compared with the 30% anterior glenoid defect condition. Stiffness
313
CORACOID POSITIONING DURING LATARJET PROCEDURE Table 1. Conditions and Outcome Variables Condition Parameter
Intact
Dislocation 0/8 Mean anterior humeral head translation (mm) Abduction and neutral rotation 8.1 Abduction and external rotation 3.9 Mean joint stiffness (N/mm) Adduction and neutral rotation 14.2 Adduction and external rotation 18.0 Abduction and neutral rotation 9.9 Abduction and external rotation 23.8 Mean rotational range of motion ( ) Adduction 47 Abduction 61
Bankart Lesion
30% Glenoid Defect
Classic Latarjet
Congruent-Arc Latarjet
ANOVA (P value)
8/8
8/8
1/8
0/8
d
* *
* *
15.7 6.5
17.2 9.9
.694 .021
4.75 6.75 3.9 8.1
1.6 1.8 1.8 1.9
7.2 10.0 4.2 9.2
5.3 7.9 4.2 8.9
.019 .033 .005 .004
46 77
52 75
42 49
43 53
.531 .001
*Translation not calculated as all specimens dislocated.
values of the intact specimen never significantly differed from those of either the classic or congruent-arc technique, and there was no significant difference between the 2 techniques (P .234) (Fig 3, Table 1). In abduction neutral rotation, the 30% anterior glenoid defect had significantly lower stiffness than the intact state (mean difference, 8.1 5.5N/mm; P ¼ .026) and the congruent-arc technique (mean difference, 2.4 1.7 N/mm; P ¼ .027); however, the stiffness did not significantly differ from that of the classic technique (mean difference, 2.4 2.7 N/mm, P ¼ .235). In abduction external rotation, the 30% glenoid defect state had significantly lower stiffness values than the intact state (mean difference, 21.9 13.0 N/mm, P ¼ .012), the classic Latarjet (7.3 1.4 N/mm, P < .001), and the congruent-arc technique (7.0 4.2 N/mm, P ¼ .015). In all abduction test positions, the intact state did not significantly differ from the 2 Latarjet techniques, and the techniques did not significantly differ from each other (P .102).
Fig 3. Glenohumeral joint stiffness values (mean and SD) for the 4 conditions tested (intact specimen, 30% anterior glenoid defect, classic Latarjet, and congruent-arc Latarjet) in the different joint configurations. Symbols denote statistically significant (P < .05) differences.
Range of Motion In adduction, no significant differences in internale external rotational range of motion were observed between any test conditions (P > .05) (Fig 4). In abduction, internaleexternal rotational range of motion was significantly reduced after the classic (mean difference, 25.8 18.0 , P ¼ .032) and congruent-arc (mean difference, 22.2 16.7 , P ¼ .041) Latarjet procedures as compared with the 30% anterior glenoid defect. There were no significant differences between the 2 techniques and the intact condition, nor did the techniques significantly differ from each other (P .282) (Table 1).
Discussion The results of this study indicate that both the classic and congruent-arc Latarjet orientations were able to eliminate dislocation in all specimens, except for 1 specimen treated with the classic repair when tested in the abducted neutral configuration. However,
314
H. W. BOONS ET AL.
Fig 4. Mean rotational range of motion values for the 4 conditions tested (intact specimen, 30% anterior glenoid defect, classic Latarjet, and congruent-arc Latarjet) in adduction and 90 of composite abduction. Symbols denote statistically significant (P < .05) differences.
a difference was found in the magnitude of anterior joint translation experienced by the humeral head when tested in the position of apprehension. Specifically, the congruent-arc orientation allowed significantly greater anterior humeral head translation before reaching a stable soft tissue endpoint. This is not unexpected, as orienting the coracoid in the congruent-arc manner results in greater bone reconstitution,7 which leads to a greater arc length of the glenoid articular surface. The kinematics of constructing a wider glenoid articular surface, however, are unknown. It is conceivable that when reconstructing a large glenoid bone deficit with the classic Latarjet, the associated decreased humeral head translations may constrain the joint, which may predispose to arthritis as the humeral head articulates with a narrower glenoid. With the congruent technique, the additional glenoid arc length may only come into play during instability-provoking maneuvers, which would then allow greater contained translation before dislocation. Unfortunately, the corollary of a greater surface area for articulation with the congruent-arc technique results in a smaller surface area for graft-toglenoid healing. This is a potential disadvantage of the congruent-arc method, as the contact area between the coracoid graft and the glenoid is decreased. This occurs because the coracoid is wider than it is thick, which results in a smaller surface area of coracoid contacting the glenoid for bony union.7 In our specimens, the mean width, thickness, and length of the coracoid grafts were 16 1, 10 2, and 28 2 mm, respectively. These corocoid dimensions are in keeping with those reported by Armitage et al.7 One specimen in the classic Latarjet group continued to dislocate. We have no explanation for why this specimen continued to remain unstable after the classic Latarjet procedure. The specimen, orientation of the graft, fixation of the graft, and size of the glenoid bone defect were critically examined to determine if any
deviations from the protocol had occurred or if there were any failures, and nothing was identified. This same specimen, however, was effectively stabilized with the congruent-arc Latarjet. Glenohumeral joint stiffness, along with joint dislocation and translation, was used as an objective outcome measure to describe the concept of joint stability. Stiffness was a measure of the humeral head’s ability to resist anterior translation and dislocation by an externally applied force. Our results indicate that there were no significant differences in joint stiffness between the intact condition, the classic Latarjet, and the congruent-arc modification. Universally, the stiffness values obtained for the 30% anterior glenoid defect were substantially lower than those for the intact condition and the 2 Latarjet techniques. Additionally, during evaluation of internaleexternal rotation range of-motion of the glenohumeral joint, it was confirmed that neither technique would cause significant changes in this outcome measure as compared with the intact condition. Ghodadra et al.8 biomechanically compared the glenohumeral contact pressures of the classic Latarjet, the congruent-arc modification, and iliac crest structural grafting. They reported that the optimum position for restoration of intact contact pressures was a graft placed flush to the glenoid surface. Therefore, we chose only to compare the different Latarjet techniques in the flush position. Additionally, they reported that the congruentarc modification yielded displayed better contact pressure restoration than the classic orientation. The findings of improved contact pressures is not unexpected, as Armitage et al.7 reported that a coracoid graft positioned in the congruent-arc manner has a radius of curvature nearly identical to that of the intact anterior glenoid rim. Wellmann et al.14 reported that the Latarjet procedure, with the coracoid oriented in the classic manner, is biomechanically superior to a contoured structural bone
CORACOID POSITIONING DURING LATARJET PROCEDURE
graft in reducing anterioinferior glenohumeral joint translation. These improved biomechanical properties may be attributed to the stabilizing “sling effect” of the conjoint tendon. Wellmann et al.14 chose to load the conjoined tendon with 10 N of force applied in a static manner. Giles et al.11 conducted a biomechanical simulator study assessing the role of the conjoined tendon in shoulder stability. They recommended applying 10 N of force to the conjoined tendon, as it was the minimum amount of force that had substantial effects on joint stability without influencing humeral head translation. Therefore, in the present study we applied 10 N of force to the conjoined tendon in a dynamic fashion through an actuator attached to the humeral shaft, mimicking the line of action of the native conjoined group. Shoulder instability with glenoid bone deficiency can present clinically as various-sized defects. Yamamoto et al.16 recently reported that a glenoid defect with a width greater than 19% of the glenoid height remains unstable after soft tissue Bankart repair. In the present study, only a 30% anterior glenoid bone deficiency was tested. We chose a 30% deficiency because it represents a larger-sized defect that would typically undergo bone grafting and because this size of defect was tested in previous biomechanical studies.8,14 Although congruent-arc positioning allows reconstitution of greater glenoid bone loss, our results reveal no significant differences in the outcomes assessed, other than the magnitude of anterior humeral head translation. Additionally, the strength of the congruent-arc reconstruction is unknown. It is conceivable that the congruent-arc orientation has less strength as it has less bony support because the graft is thinner. This lack of support may lead to cantilever bending moments, potentially increasing the risk of non-union and fixation failure. In contrast, the classic technique may have improved strength; however, it has inferior glenohumeral contact pressures8 and requires intraoperative graft contouring. Both techniques have surgeon advocates, advantages, and disadvantages.3,4,17-20 Limitations In this biomechanical simulator study we tried to simulate native human shoulder conditions. There are, however, limitations to the study, especially those that are inherent to all biomechanical simulations, such as the use of elderly donor specimens. Also, the results of this study represent time zero kinematics; the effects of soft tissue healing and adaptive changes are unknown. Another limitation is that we did not conduct our own power analysis; we used the power analysis of Yamamoto et al.,15 which concluded that 8 specimens were necessary. As such, because Yamamoto et al.15 and Wellmann et al.14 used 8 specimens, we also used 8 specimens. Additionally, we used the same shoulder specimens to perform both the classic and congruent-arc
315
techniques and, therefore, used the same coracoid process. This was accomplished by drilling 2 tunnels for the classic technique and 2 tunnels for the congruent-arc technique; therefore, the additional drill tunnels may have weakened the graft. This potential weakening of the coracoid graft, however, did not affect stability testing as all grafts were securely fixed and none fractured or failed during testing. The difference in humeral head translation between the classic and congruent techniques was statistically significant; however, a mean change of 3.5 mm may not be clinically relevant. Finally, the results of this study are applicable to the 30% glenoid bone loss model created. Therefore, the Latarjet procedure would likely have behaved differently if tested on smaller-sized glenoid defects.
Conclusions The classic and congruent-arc Latarjet techniques restore shoulder stability and motion in cases of considerable bone loss. The techniques do not substantially differ in rotational range of motion or joint stiffness. However, the congruent-arc technique does result in significantly greater anterior humeral head translation than with the classic technique, before reaching a stable non-dislocated endpoint.
Acknowledgment The authors acknowledge the assistance of Dr. Kenneth Faber, Dr. Robert Litchfield, and Dr. Louis Ferreira.
References 1. Boileau P, Villalba M, Hery JY, Balg F, Ahrens P, Neyton L. Risk factors for recurrence of shoulder instability after arthroscopic Bankart repair. J Bone Joint Surg Am 2006;88:1755-1763. 2. Burkhart SS, De Beer JF. Traumatic glenohumeral bone defects and their relationship to failure of arthroscopic Bankart repairs: Significance of the inverted-pear glenoid and the humeral engaging HilleSachs lesion. Arthroscopy 2000;16:677-694. 3. Beran MC, Donaldson CT, Bishop JY. Treatment of chronic glenoid defects in the setting of recurrent anterior shoulder instability: A systematic review. J Shoulder Elbow Surg 2010;19:769-780. 4. Provencher MT, Bhatia S, Ghodadra NS, et al. Recurrent shoulder instability: Current concepts for evaluation and management of glenoid bone loss. J Bone Joint Surg Am 2010;92(Suppl 2):133-151. 5. Latarjet M. Treatment of recurrent dislocation of the shoulder. Lyon Chir 1954;49:994-997. 6. De Beer JF, Roberts C. Glenoid bone defects: Open Latarjet with congruent arc modification. Orthop Clin North Am 2010;41:407-415. 7. Armitage MS, Elkinson I, Giles JW, Athwal GS. An anatomic, computed tomographic assessment of the coracoid process with special reference to the congruentarc Latarjet procedure. Arthroscopy 2011;27:1485-1489.
316
H. W. BOONS ET AL.
8. Ghodadra N, Gupta A, Romeo AA, et al. Normalization of glenohumeral articular contact pressures after Latarjet or iliac crest bone-grafting. J Bone Joint Surg Am 2010;92: 1478-1489. 9. Elkinson I, Giles JW, Faber KJ, et al. The effect of the remplissage procedure on shoulder stability and range of motion: An in vitro biomechanical assessment. J Bone Joint Surg Am 2012;94:1003-1012. 10. Giles JW, Elkinson I, Ferreira LM, et al. Moderate to large engaging HilleSachs defects: An in vitro biomechanical comparison of the remplissage procedure, allograft humeral head reconstruction, and partial resurfacing arthroplasty. J Shoulder Elbow Surg 2012;21:1142-1151. Epub 2011 Oct 29. 11. Giles JW, Boons HW, Ferreira LM, Johnson JA, Athwal GS. The effect of the conjoined tendon of the short head of the biceps and coracobrachialis on shoulder stability and kinematics during in-vitro simulation. J Biomech 2011;44:1192-1195. 12. McQuade KJ, Smidt GL. Dynamic scapulohumeral rhythm: The effects of external resistance during elevation of the arm in the scapular plane. J Orthop Sports Phys Ther 1998;27:125-133. 13. 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.
14. Wellmann M, Petersen W, Zantop T, et al. Open shoulder repair of osseous glenoid defects: Biomechanical effectiveness of the Latarjet procedure versus a contoured structural bone graft. Am J Sports Med 2009;37:87-94. 15. Yamamoto N, Itoi E, Abe H, et al. Effect of an anterior glenoid defect on anterior shoulder stability: A cadaveric study. Am J Sports Med 2009;37:949-954. 16. Yamamoto N, Muraki T, Sperling JW, et al. Stabilizing mechanism of bone-graft of a large glenoid defect. J Bone Joint Surg Am 2012;92:2059-2066. 17. Burkhart SS, De Beer JF, Barth JR, Cresswell T, Roberts C, Richards DP. Results of modified Latarjet reconstruction in patients with anteroinferior instability and significant bone loss. Arthroscopy 2007;23:1033-1041. 18. Chen AL, Hunt SA, Hawkins RJ, Zuckerman JD. Management of bone loss associated with recurrent anterior glenohumeral instability. Am J Sports Med 2005;33:912-925. 19. Millett PJ, Clavert P, Warner JJ. Open operative treatment for anterior shoulder instability: When and why? J Bone Joint Surg Am 2005;87:419-432. 20. Piasecki DP, Verma NN, Romeo AA, Levine WN, Bach BR Jr, Provencher MT. Glenoid bone deficiency in recurrent anterior shoulder instability: Diagnosis and management. J Am Acad Orthop Surg 2009;17: 482-493.