stto

U

r

S

Current Concepts With Video Illustration

The Rotator Interval: Pathology and Management

Trevor R. Gaskill, M.D., Sepp Braun, M.D., and Peter J. Millett, M.D., M.Sc.

Abstract: The rotator interval describes the anatomic space bounded by the subscapularis, supraspi-natus, and coracoid. This space contains the coracohumeral and superior glenohumeral ligament, thebiceps tendon, and anterior joint capsule. Although a definitive role of the rotator interval structureshas not been established, it is apparent that they contribute to shoulder dysfunction. Contracture orscarring of rotator interval structures can manifest as adhesive capsulitis. It is typically managednonsurgically with local injections and gentle shoulder therapy. Recalcitrant cases have beensuccessfully managed with an arthroscopic interval release and manipulation. Conversely, laxity ofrotator interval structures may contribute to glenohumeral instability. In some cases this can bemanaged with one of a number of arthroscopic interval closure techniques. Instability of the bicepstendon is often a direct result of damage to the rotator interval. Damage to the biceps pulley structurescan lead to biceps tendon subluxation or dislocation depending on the structures injured. Althoughsome authors describe reconstruction of this tissue sling, most recommend tenodesis or tenotomy ifit is significantly damaged. Impingement between the coracoid and lesser humeral tuberosity is arelatively well-established, yet less common cause of anterior shoulder pain. It may also contributeto injury of the anterosuperior rotator cuff and rotator interval structures. Although radiographicindices are described, it appears intraoperative dynamic testing may be more helpful in substantiatingthe diagnosis. A high index of suspicion should be used in association with biceps pulley damage oranterosuperior rotator cuff tears. Coracoid impingement can be treated with either open or arthros-copic techniques. We review the anatomy and function of the rotator interval. The presentation, physicalexamination, imaging characteristics, and management strategies are discussed for various diagnosesattributable to the rotator interval. Our preferred methods for treatment of each lesion are also discussed.

tihiv

.S.A.;any.me Re

M.D., Mic.net

Neer is widely credited with describing the antero-superior space between the subscapularis and

upraspinatus tendons as the rotator interval. The ro-ator interval has since become recognized as a dis-inct anatomic space. The early, classic contributionsf Neer and Foster,1 Rowe and Zarins,2 and Nobuhara

From The Steadman Clinic (T.R.G., P.J.M.), Vail, Colorado, University Munich, Klinikum rechts der Isar (S.B.), Munich, GermP.J.M. has received (or agreed to receive) from Arthrex and Ga

esearch. The other authors report no conflict of interest.Received April 28, 2010; accepted October 6, 2010.Address correspondence and reprint requests to Peter J. Millett,

te 1000, Vail, CO 81657, U.S.A. E-mail: drmillett@steadmanclin© 2011 by the Arthroscopy Association of North America0749-8063/10277/$36.00doi:10.1016/j.arthro.2010.10.004

Note: To access the video accompanying this report, visit the [Month] iss

Arthroscopy: The Journal of Arthroscopic and Related

and Ikeda3 proposed biomechanical roles for struc-ures within the rotator interval and fostered curiosityn its dysfunction. Subsequently, considerable effortas been devoted to better understanding the rotatornterval, yet its pathoanatomical role remains contro-ersial. As our knowledge of the biomechanical sig-

and the Department of Orthopaedic Sports Medicine, Technical

ady something of value exceeding US $500 to this manuscript or

.Sc., Shoulder Service, The Steadman Clinic, 181 W Meadow Dr,

ue of Arthroscopy at www.arthroscopyjournal.org.

1Surgery, Vol xx, No x (Month), 2011: pp xxx

.

tisfsght

sarabsacatmtC

lMtS

ft

hc

ccm

2 T. R. GASKILL ET AL.

nificance of the rotator interval structures evolves, sodoes our understanding of the disability it can create.This provides an opportunity to advance our ability totherapeutically intervene (Video 1, available at wwwarthroscopyjournal.org).

ANATOMY

The rotator interval is a triangular space located inhe anterosuperior portion of the glenohumeral joint. Its bounded by the supraspinatus superiorly and theubscapularis inferiorly, and the coracoid processorms its medial base. Contained within this triangularpace are the coracohumeral ligament (CHL), middlelenohumeral ligament (MGHL), and superior gleno-umeral ligament (SGHL), long head of the bicepsendon, and anterior joint capsule.

The configuration of these structures has been theubject of several anatomic descriptions. The presencend prominence of structures and foramen within theotator interval exhibit considerable variability. Jost etl.4 offered 1 of the more comprehensive depictionsased on 22 cadaveric shoulder specimens. They de-cribe the medial and lateral rotator interval as uniquenatomic structures (Fig 1). Laterally, the first layeronsists of the CHL fibers following the subscapularisnd supraspinatus to their respective insertions intohe humeral tuberosities. Layer 2 is composed ofeshing fibers of the bounding rotator cuff muscula-

ure and CHL. Layer 3 is primarily deep fibers of theHL inserting on the greater tuberosity. The deep

ayer is composed of the joint capsule and SGHL.edially, 2 layers are described. The CHL makes up

he superficial layer, and the deep layer comprises theGHL and joint capsule.The CHL is a broad, thin structure originating

rom the lateral coracoid base. It has an irregularrapezoidal shape and a variable insertion.5 Recent

reports suggest that the CHL is histologically moresimilar to capsule than ligamentous tissue.5 Others

ave concluded that it is an essential biomechanicalomponent of the rotator interval.6-10 Thus contro-

versy remains regarding the physiologic role of theCHL.4,7,10,11

The SGHL originates at the superior glenoid tuber-le and inserts near the lesser tuberosity at the foveaapitis.10 It crosses the floor of the rotator interval anday merge with the CHL.5 These ligaments, with

contributions from the bounding rotator cuff tendons,form the biceps reflection pulley. Traditionally, it wasthought that the SGHL or CHL (or both) was theprimary stabilizer of the long head of the biceps ten-don.12-14 More recent anatomic reports challenge thisbelief and suggest that subscapularis tendon integrityis responsible for biceps tendon stability.11

FUNCTION

The structures within the rotator interval are thesubject of multiple in vivo and in vitro biomechanicalstudies, and many of these are contrasting.3,4,6,8-11,15-19

FIGURE 1. Depiction of rota-tor interval anatomy. (A) Grossrepresentation of described ro-tator interval structures. (B)Cross-sectional representationof lateral rotator interval layersas described by Jost et al.4 Thesupraspinatus (SSP), CHL,long head of the biceps tendon(LHB), SGHL, and subscapu-laris (SSC) are depicted.

ctstislctmtNttgcpt

sClgt

mpottlib

oti

fodotmtt

cMocK

ps

ic

3ROTATOR INTERVAL DISEASE

Harryman et al.10 classically reported 1 of the firstomprehensive cadaveric studies examining the func-ion of rotator interval structures. They showed thatectioning of the rotator interval capsule and ligamen-ous structures increased passive glenohumeral flex-on, extension, external rotation, and adduction in 80houlders. Conversely, they showed that medial-to-ateral interval imbrication caused a reciprocal de-rease in these motions. They concluded that the in-erval functions as a “check-rein” against excessiveotion and limits posterior-inferior glenohumeral

ranslation. This study substantiated a report fromobuhara and Ikeda,3 who showed clinically that

ightening the rotator interval, by externally rotatinghe shoulder, resulted in decreased posterior-inferiorlenohumeral instability. Numerous clinical studiesiting altered motion with interval closure support thisroposed function, although the magnitude and direc-ion of limitations are debated.20-22 Therefore it is

thought that structures within the rotator interval pro-vide some degree of stability to the glenohumeral joint(Table 1).

Rotator interval structures also contribute to thetability of the long head of the biceps tendon. TheHL, SGHL, and subscapularis form a reflection pul-

ey, supporting the bicipital tendon as it exits thelenohumeral joint. Evidence supports each as criticalo the integrity of this structure.11,12,23 The transverse

humeral ligament is an inconsistent anatomic struc-ture.24 When present, it appears to provide little bio-

echanical support within the groove despite itsrominent location.25 A third and, perhaps, less obvi-us role of the rotator interval is its function in main-aining negative intra-articular pressure. Venting ofhe glenohumeral capsule contributes to greater trans-ation.26,27 Similarly, lesions of the rotator intervalnvolving the capsule could also contribute to insta-ility.

CONTRACTURES OF ROTATOR INTERVAL

Rotator interval contractures constitute a spectrumf disease ranging from mild rotator cuff impingemento debilitating adhesive capsulitis. In its most incapac-tating form, it is characterized by painful motion and

TABLE 1. Proposed Rotator Interval Function

● Contributes to glenohumeral stability● Increases stability of long head of biceps tendon

r● Limits excessive glenohumeral motion

requently causes rest and night pain. It is thought toccur more frequently and respond less readily iniabetic populations.28,29 It can also occur in the post-perative setting. Active and passive motion limita-ions are characteristic, and abnormal scapulothoracicotion may exist. Discomfort is frequently localized

o the deltoid insertion, and tenderness in the area ofhe coracoid is common.15

Radiographs are typically normal, and decreasedapsular volume may be apparent with arthrography.agnetic resonance imaging (MRI) studies report

bliteration of the subcoracoid fat plane and anatomichanges in the rotator interval30,31 (Fig 2). Studies byim et al.31,32 showed significant decreases in rotator

interval height, base, and area compared with normalshoulders. Others have identified thickening of theCHL and joint capsule within the rotator intervalcompared with normal age- and sex-matched con-trols30,33 (Fig 3).

The etiology of adhesive capsulitis is not com-letely understood. Injury, causing inflammation andcarring, has been proposed.3,17 Histologically, fibro-

blasts and myofibroblasts are prevalent, depositingdense matrices of collagen within the capsule.34 Works being conducted to identify abnormal expressions ofytokines, proteases, or growth factors that may be

FIGURE 2. Sagittal oblique T1-weighted MRI scan showing nor-mal appearance of CHL (arrow) and subcoracoid fat plane (aster-isk). The normal laxity of the CHL should be noted.

esponsible for the abnormal regulation of these

ur

ncna

cpt

fMas

4 T. R. GASKILL ET AL.

cells.34,35 Regardless of its etiology, some believe that,ltimately, a contracted CHL is the “essential” lesionesponsible for this entity.9

Symptomatic management is frequently all that isecessary to successfully treat adhesive capsulitis, be-ause it is generally a self-limiting disorder. Variousonsurgical modalities are used including nonsteroidalgents, local injections, and physical therapy.28 Satis-

factory outcomes are reported in 90% of patients withidiopathic capsulitis after local steroid injection,whereas oral steroid administration appears less effi-cacious.28,36,37

Recalcitrant cases may require manipulation or surgi-al soft-tissue release. Promising results have been re-orted for manipulation alone and both open and ar-hroscopic release techniques.38-42 Post-manipulation

gains in shoulder motion and function appear to besustained long-term as noted by Farrell et al.39 in theirollow-up of 26 shoulders 15 years after manipulation.

ore recently, reports of combined manipulation andrthroscopic release have also yielded good re-ults.40-43 In addition, hydrodilation has been shown to

be effective in properly selected patients.44

Preferred Technique

The initial management is nonsurgical. Gentle mo-

FIGURE 3. Sagittal oblique T1-weighted image showing thick,contracted CHL (arrow) and obliteration of normal subcoracoid fatplane (asterisk) in a patient with adhesive capsulitis.

tion exercises are prescribed, typically supervised by a

physical therapist. Aggressive therapy may worsensymptoms, especially in the early, painful stages ofadhesive capsulitis. Corticosteroid injections are usedto provide local symptomatic control and attempt tolimit any inflammatory process.

If these measures fail, an arthroscopic interval re-lease followed by a gentle manipulation under anes-thesia is recommended. Introduction of the arthros-cope must be done carefully in the contractedshoulder to avoid iatrogenic cartilage damage. Lessforce is required during manipulation, however, ifthe arthroscopic release is performed first. Thebeach-chair position and interscalene regional an-esthesia are used. Indwelling catheters can be usefulto facilitate early passive motion and decrease nar-cotic requirements.

The joint is insufflated before the surgeon carefullyintroduces the arthroscope. Visualization of the ante-rior-superior portion of the joint is usually possible,and an anterior working portal is established. A4.5-mm oscillating shaver and electrocautery are usedto debride synovitis and exuberant scar tissue (Fig 4).Once adequate visualization is established, a thick,robust capsule is usually confirmed. A hooked elec-trocautery device is used to divide the capsule justlateral to the glenoid labrum extending from the bicepstendon to the subscapularis. This release increases gle-nohumeral volume, allows the humeral head to lateralize,and improves visualization of the anterior-inferior gle-

FIGURE 4. Rotator interval before arthroscopic release with sy-novitis (asterisk) in left shoulder viewed from posterior with pa-tient in modified beach-chair position. (Subscap, subscapularis;

HH, humeral head.)

iodcwd

ltlp

Tteisp

rqfaaptiw

rri

dpcc

5ROTATOR INTERVAL DISEASE

nohumeral joint. Additional release is performed in thisarea as necessary to ensure mobility of the subscapularis(Fig 5). A gentle manipulation is then performed toprovide appropriate motion.

Postoperatively, adequate analgesia is critical tomaintain motion established intraoperatively. Pas-sive motion is begun shortly after surgery and con-tinued at least daily for the first 2 weeks. Patientsare then transitioned to a home exercise programwith intermittent supervision individualized to thetheir progress.

ROTATOR INTERVAL LAXITY

The opposite end of the spectrum encompasses re-dundancy of rotator interval structures. Laxity maycontribute to instability and resultant pain.4,7,10 Roweand Zarins2 reported an association between rotatornterval lesions and patients with recurrent instabilityf the shoulder. They therefore recommended ad-ressing these lesions during surgery. Despite thisorrelation, the precise contribution of structuresithin the rotator interval to glenohumeral stability isebated.Patients presenting with laxity of rotator interval

igamentous or capsular structures often report a his-ory of acute trauma or overuse injuries that eventuallyead to tissue insufficiency and laxity. Many of these

FIGURE 5. Rotator interval after arthroscopic release in left shoul-er viewed from posterior with patient in modified beach-chairosition. The following structures can be seen: coracoid (Cor),oracoacromial ligament (CA lig) with lines indicating its course,onjoined tendon (CJTen), and subscapularis (Subscap).

atients will complain of instability or early fatigue.

he sulcus sign may be present if inferior humeralranslation is excessive and normally disappears inxternal rotation because of tightening of the rotatornterval. The presence of a persistent sulcus sign de-pite glenohumeral external rotation may suggestathologic laxity when symptoms are elicited.3,10,16,17

Subtle inferior subluxation may be appreciated onadiographs, and although arthrograms are infre-uently used, they may show contrast extravasationrom interval lesions or abnormal contrast filling inbduction and external rotation.3 More recently, MRInd magnetic resonance arthrography have becomeopular imaging modalities because of the added soft-issue detail these techniques afford. Increased rotatornterval dimensions have been described in patientsith chronic instability compared with controls.32

Other authors suggest that these measurements are notsignificantly different between various instability pat-terns and controls.45

Symptomatic patients may require imbrication ofotator interval structures to decrease laxity. Isolatedotator interval foramen closure for recurrent instabil-ty has been reported with excellent results.16 It is

important to recognize, however, that rotator intervallesions are often not solely responsible for glenohu-meral instability. Failure to recognize other causes ofinstability will yield unsatisfactory results.46 Levine etal.46 reported on 50 patients undergoing revisionshoulder stabilization. External rotation was asymmet-rically limited by tight anterior structures in more than20% of patients in this series. Yet, these patientsremained unstable in abduction and external rotationbecause a patulous inferior glenohumeral ligamentcapsular complex was not addressed at the primarysurgery. Therefore isolated interval closure is indi-cated for only select patients.

Arthroscopic and open procedures have been de-scribed to imbricate rotator interval tissues. The re-sults of open coracohumeral imbrication of Harrymanet al.10 are often quoted to justify closure of the rotatorinterval in certain situations. They showed decreasedposterior and inferior translation after open medial-to-lateral imbrication of the CHL. These findings arefrequently extrapolated to arthroscopic closure despitefundamental differences in the interval closure tech-niques. This has resulted in an interesting controversyregarding the true function of, and indication for, arotator interval closure. In contrast to the CHL imbri-cation described by Harryman et al., the typical ar-throscopic SGHL-to-MGHL closure more accuratelydepicts closure of the subscapularis foramen. Recent

analysis of this technique by Provencher and col-

twtsbcoatissoitm

c3ol

6 T. R. GASKILL ET AL.

leagues20,47 and other authors48 has failed to replicatefindings reported by Harryman et al. By contrast,Savoie et al.49,50 report excellent clinical results usingan oblique suture orientation, propose that it moreaccurately reproduces the results of Harryman et al.,and suggest that it provides an increased measure ofsuccess in many cases. To our knowledge, no biome-chanical data currently exist to support this conclu-sion. To this end, Farber et al.51 compared a medial-o-lateral arthroscopic technique using suture anchorsith a superior-to-inferior technique in a cadaveric mul-

idirectional instability model. The medial-to-lateral clo-ure technique more closely restored normal motion,ut the findings of Harryman et al. could not beonfirmed despite use of a similar interval closurerientation. Mologne et al.20 reported improvement innterior translation after an arthroscopic SGHL-o-MGHL rotator interval closure. No improvementn posterior or inferior glenohumeral stability washown. Therefore arthroscopic SGHL-to-MGHL clo-ure may have a role in instability repair, but clinicalutcome studies are lacking and the long-term efficacys unclear. If interval closure is performed, tensioninghe closure in 30° of external rotation may avoidotion limitations.52

Preferred Technique

Satisfactory results have been reported both withand without rotator interval closure in posterior and

TABLE 2. Rotator Interval Closure Pearls

● External rotation during closure minimizes joint contracture● Begin just lateral to coracoid and proceed laterally as

necessary● Tie closure stitches extra-articularly to capture CHL for more

robust closure

multidirectional instability procedures. Although thebiomechanical role of an arthroscopic closure remainscontroversial, such a closure may be indicated in casesof a persistent sulcus sign despite external rotation orwhen subluxation persists despite adequate plication.In these cases an arthroscopic SGHL-to-MGHL clo-sure is used as an extension of the capsular shift.

A rotator interval closure may be performed withthe patient in either the beach-chair or lateral decub-itus position (Table 2). The rotator interval is visual-ized from the posterior portal. The arm is externallyrotated to minimize motion loss. An 8.25-mm cannulais positioned just outside the joint capsule. A curvedshuttling device is then passed through the cannula, pen-etrating the capsular tissue (typically the MGHL) justsuperior to the subscapularis, 3 to 4 mm lateral to theglenoid. A suture is advanced into the joint. The cannulais redirected superiorly, and a penetrating instrument ispassed anterior to the supraspinatus tendon through theSGHL base. The previously passed suture is retrievedthrough the same cannula in the anterior portal to avoidcapturing the deltoid. Sutures are tied extra-articularly,capturing as much capsular tissue as possible. Additionalsutures are placed laterally as necessary (Fig 6). Typi-ally, 2 sutures are used, although in more severe cases,can be placed. Postoperative rehabilitation is dependentn leading pathology, but external rotation is usuallyimited to 30° for the first 3 to 4 weeks.

BICEPS REFLECTION PULLEY TEARS

Early work by Petersson53 illustrated medial bicepstendon dislocation in 3.3% of cadaveric specimensand suggested the relation between subscapularis in-tegrity and stability of the biceps tendon. Subse-quently, the biceps reflection pulley was described asbeing formed by contributions from the CHL, SGHL

FIGURE 6. Right shoulderviewed from posterior before(A) and after (B) rotator inter-val closure with patient in mod-ified beach-chair position. (A)Passage of first rotator intervalsuture from the SGHL to theMGHL. (B) Closed rotator in-terval after placement of 2 su-tures. The humeral head (HH)and glenoid (GLEN) can alsobe seen.

r

rts(isdts

ssbmaai4

atmb

d

toh(saamCs

7ROTATOR INTERVAL DISEASE

and upper subscapularis.12 In these patients the super-ficial CHL was always intact, and until the rotatorinterval was opened, the lesion was “hidden” andeasily overlooked during both open and arthroscopicinspections.12,54

It is established that injury to this pulley complexesults in instability of the proximal biceps tendon.55

Bennett55 reported a classification system of bicepseflection pulley lesions based on the anatomic struc-ure injured (Fig 7). If either the intra-articular sub-capularis (type 1) or the medial head of the CHLtype 2) is incompetent, the biceps tendon will displayncreased intra-sheath mobility. When both the sub-capularis and medial CHL are disrupted, the bicepsislocates intra-articularly (type 3). Less commonly,he biceps tendon may dislocate anterior to the sub-capularis if the lateral CHL and leading edge of the

FIGURE 7. Classification system: normal (A);ype 1, tear of subscapularis without involvementf medial head of CHL (B); type 2, tear of medialead of CHL without subscapularis involvementC); type 3, tear of both medial head of CHL andubscapularis (D); type 4, tear of supraspinatusnd lateral head of CHL (E); and type 5, tears ofll stabilizing structures (F). (s, subscapularis; m,edial head of CHL; B, biceps; L, lateral head ofHL.) Arrows represent direction of biceps in-

tability. (Reprinted with permission.55)

ubscapularis are injured (type 4). Finally, when eachtructure is disrupted, complete loss of integrity of theicipital sheath occurs (type 5). Although instabilityay occur in isolation, it is frequently associated withsubscapularis and leading-edge supraspinatus tear asn “anterosuperior” rotator cuff tear.6,55 In 1 series,nternal anterosuperior impingement was found in4% of patients with pulley tears.14 In addition to

injury to the biceps reflection pulley, anterior-superiorimpingement is also thought to be responsible forinjury to the biceps tendon. Boileau et al.56 described

hypertrophic, hourglass-shaped biceps tendon en-rapped within the bicipital groove. This impedes nor-al gliding during elevation and leads to mechanical

lock and subsequent pain.Lesions of the biceps reflection pulley system are

ifficult to identify radiologically; however, several

tcsttshl

do

iprc

8 T. R. GASKILL ET AL.

subtle findings may be useful. Walch et al.12 describedhe “pulley sign” as an extra-articular collection ofontrast material anterior to the superior extent of theubscapularis. Its presence on magnetic resonance ar-hrography studies suggests injury to the biceps reflec-ion pulley complex. Other authors suggest that exten-ion of contrast to the cortex of the coracoid may beelpful in the preoperative diagnosis of rotator intervalesions.57

Lesions of the biceps reflection pulley can be ad-ressed surgically, but results are mixed. Reports ofpen repair by Walch et al.12 included subscapularis

reattachment and medial biceps sheath reconstruction.Of 22 patients in this series, 12 required scar removalor groove deepening to stabilize an enlarged tendon.This technique resulted in tendon rupture in 3 patientsand modest pain improvement, thus leading the au-thors to conclude that biceps tenodesis is a morereliable treatment. Other authors, however, advocatean arthroscopic repair technique and report statisti-cally improved pain, American Shoulder and ElbowSurgeons, and Constant scores at 2 years’ follow-up.58

Although repair of the biceps reflection pulley remainscontroversial, most authors appear to support tenode-sis of the biceps tendon to protect associated rotatorcuff repairs. Biceps tenodesis can be accomplishedthrough open or arthroscopic approaches by use ofsoft-tissue or bony repairs. The optimal techniquehas not been established, and many open and ar-throscopic techniques have resulted in good re-ported outcomes.59,60

Preferred technique

Our preference is to perform a subpectoral bicepstenodesis in patients with anterior shoulder pain andMRI or arthroscopic confirmation of damage to thebiceps tendon or its pulley system. The beach-chairposition with standard posterior viewing and anteriorworking portals are used for the arthroscopic portionof this procedure. The biceps tendon is advancedintra-articularly to visualize as much of the tendon aspossible, and when indicated, it is cut at its base. Theinferior border of the pectoralis major is palpated, anda longitudinal incision, centered over the inferior bor-der of the pectoralis major tendon, is made. The re-flected falx of the pectoralis major fascia is incised inline with the pectoralis tendon fibers at the inferioraspect of the pectoralis tendon. The horizontal fibersof the pectoralis tendon should be visualized to ensureappropriate location. Blunt dissection proximally,

deep to the pectoralis major tendon and along the

anteromedial humerus, will reveal the long head of thebiceps tendon in the bicipital groove. Retraction in-struments are placed, and the biceps tendon is deliv-ered into the surgical wound. It is trimmed approxi-mately 2 cm proximal to its musculotendinoustransition, and a No. 2 permanent suture is placedalong its proximal 1.5 cm in a whipstitch fashion. Adrill is used to fashion a tunnel approximately 15 mmin depth at the distal aspect of the bicipital groove.Care must be taken to ensure that the tunnel is drilledin the center of the humerus. An interference screw isused to fix the tendon within the bone tunnel andadvanced until it is flush with the humeral cortex. The2 ends of the suture are tied over the interferencescrew, enhancing screw-tendon security.

Postoperatively, a sling is used for no longer than 4weeks, and the rehabilitation plan should follow thatprescribed because of concomitant procedures. Re-sisted elbow flexion is not permitted for 6 weeks. If inisolation, passive glenohumeral and elbow motionshould begin immediately and advance to active mo-tion as the patient tolerates. Strengthening or heavylifting should be withheld for the first 6 postoperativeweeks.

CORACOID IMPINGEMENT

Coracoid impingement is a relatively well-established,yet less common cause of anterior shoulder pain(Table 3). Pain is presumed to occur because of im-pingement of the subscapularis between the coracoidprocess and lesser humeral tuberosity. It may be idio-pathic in nature or result from trauma, instability, oriatrogenic causes.61 Although the coracoacromialnterval involves the acromion, CHL, and coracoidrocess, it is the latter that is considered mostesponsible for altering the volume of the cora-oacromial arch.62

Russo and Togo63 suggested that this syndromeoccurs most commonly because of chronic overusewith the shoulder in a forward-flexed, adducted, andinternally rotated position. These patients typicallypresent with dull anterior shoulder pain exacerbatedby bringing the lesser tuberosity into contact with the

TABLE 3. Coracoid Impingement

● Tenderness in subcoracoid region● Pain in forward flexed, adducted, and internally rotated

position● Coracohumeral interval �5 mm

● Dynamic impingement at arthroscopy

ipbp

ee

vetttattalctwMs

i

assapnsp

n

vppbcitaj

9ROTATOR INTERVAL DISEASE

coracoid process, and it appears to be more commonin the midrange of flexion.64 Tenderness can be elic-ted in the soft tissues surrounding the coracoidrocess and lesser tuberosity. Others have notedurning pain during aggravating exercise such asush-ups.65 Injection of the subcoracoid region has

been advocated as a diagnostic and therapeutic in-tervention.66,67 Validated diagnostic criteria do notxist, and therefore it remains largely a diagnosis ofxclusion (Table 4).Standard radiographs may detect some anatomic

ariations that can contribute to this syndrome; how-ver, MRI or computed tomography is more useful inhis respect. The coracoid index is measured by de-ermining the distance the coracoid projects lateral tohe glenoid articular surface on axial images. Dines etl.67 reported the mean value in 67 normal shoulderso be 8.2 mm. A study by Friedman et al.,68 using MRIo measure the coracohumeral interval, found thatsymptomatic patients averaged 11 mm, with noneess than 4 mm, in maximal internal rotation. Byontrast, the mean coracohumeral interval in symp-omatic patients was 5.5 mm, and it appears to narrowith internal rotation.68,69 Subsequently, this index onRI was found to be only 5.3% sensitive but 97%

pecific for coracoid impingement.18 Other authorsreport no correlation between the coracohumeral in-terval and subscapularis injury in cadaveric speci-

FIGURE 8. Right shoulderiewing from posterior withatient in modified beach-chairosition. (A) Coracoid (Cor)efore coracoplasty. (B) Cora-oid after coracoplasty. Arrowsndicate the increased distanceo the subscapularis (Subscap)fter coracoplasty. (CT, con-oined tendon.)

TABLE 4. Key Points Regarding Coracoid Impingement

● Should be evaluated in any patient with other anterosuperiorrotator cuff or biceps injury

● Subtle anterior instability should be assessed● Diagnostic injection can be useful to localize symptoms● Narrowed coracohumeral interval can be evaluated by

computed tomography or MRI● Dynamic examination confirms adequate resection of coracoid

mens.70 They propose that functional anterior instabil-ity narrows the coracohumeral interval and may bemore important than static impingement in the settingof full-thickness rotator cuff tears.70 Therefore imag-ng may support but cannot establish this diagnosis.

The first line of treatment should include a trial ofctivity modification and rotator cuff and scapulartrengthening activities. If nonsurgical modalities fail,urgical decompression may be necessary. Options toccomplish this include open or arthroscopic coraco-lasty. It is also intuitive that anterior instability couldarrow the coracohumeral interval and, if indicated,hould be appropriately stabilized before coraco-lasty.71 Open coracoplasty is accomplished through a

deltopectoral approach allowing removal of the lateral1.5 cm of the coracoid. This requires protection of themusculocutaneous nerve and repair of the conjoinedtendon more medially after its dissection.65,67 Otherauthors believe that isolated coracoid impingement israre and advocate resection of the coracoacromialligament and acromioplasty in addition to partial cor-acoid resection.61 More recently, arthroscopic tech-iques have been popularized.72,73 The described ad-

vantage is avoiding detachment of the conjoinedtendon and less tissue dissection. Outcome studiesdetailing this relatively rare diagnosis are limited, butauthors typically report reasonable outcomes for bothopen and arthroscopic procedures.67,73

Preferred Technique

Arthroscopic coracoplasty is our preferred tech-nique to address coracoid impingement. It avoids de-tachment of the conjoined tendon and facilitates treat-ment of other concomitant intra-articular pathology.The coracoid and subcoracoid space should be exam-ined in patients with long head biceps and bicepsreflection pulley tears, subscapularis tears, and ante-

ttwapp

icspricdserisnt

i

ioirtmlttmts

f

10 T. R. GASKILL ET AL.

rior supraspinatus tears because of the high associa-tion with coracohumeral impingement as suggested byBraun et al.74,75 We perform this procedure with pa-ients in the beach-chair position under regional anes-hesia using standard posterior viewing and anteriororking portals. A coracoplasty can be performed bytransarticular approach or an extra-articular ap-

roach. Most commonly, we use a transarticular ap-roach.76 The capsule between the SGHL and MGHL

is opened with a shaver or radiofrequency device. Thecoracoacromial ligament fibers serve as an excellentlandmark leading to the lateral side of the coracoidsafely. Radiofrequency ablation is used to clear theposterior and lateral aspects of the coracoid. The con-joined tendon and coracoacromial ligament are pre-served, with removal of the lateral and posterior cor-acoid with an arthroscopic bur. The anterolateral 1 to1.5 cm of the coracoid is typically resected, and dy-namic examination should confirm sufficient resec-tion. Excessive resection risks coracoid fracture, andneurologic injury is possible inferior to the coracoid(Fig 8).

Postoperatively, rehabilitation is tailored based onother attendant procedures. If performed in isolation,rehabilitation should be similar to that of an isolatedsubacromial bursectomy. Motion begins immediately,and active range of motion is encouraged as early asdiscomfort will allow. We recommend that positionsof impingement be avoided for the first 2 postopera-tive weeks.

CONTROVERSIES AND FUTUREDIRECTIONS

The rotator interval continues to be an area ofntense interest for shoulder surgeons. Although pre-ise roles for the named ligamentous and capsulartructures are debated, our ability to satisfactorily treatathology of the rotator interval has become moreeliable. Despite these advances, controversy remainsn several areas. It is apparent that traditional arthros-opic rotator interval closure techniques have not toate successfully reproduced the biomechanical re-ults of open imbrication. This, in combination withquivocal clinical results, has resulted in uncertaintyegarding its utility. Other authors hold that rotatornterval closure is a critical aspect of shoulder recon-truction for all patients. More detailed analysis isecessary to determine the indication and optimalechnique for rotator interval closure.

Similarly, the etiology and prevalence of coracoid

mpingement remain contentious. Although coracoid

mpingement appears to be highly associated withther anterior shoulder damage, it is unknown whethert contributes to or results from these associated inju-ies.70,74 Biomechanical and clinical validation of in-ervention techniques will be necessary to establish aore precise role for surgical treatment. We will also

ikely see increased utilization of biologic interven-ion, through growth factor and enzymatic regula-ion, to augment or avoid surgical management inany scenarios. These efforts will ultimately con-

inue to improve outcomes for many complexhoulder patients.

Acknowledgment: The authors thank Heinrich Heueror his help preparing images for this article.

REFERENCES

1. Neer CS II, Foster CR. Inferior capsular shift for involuntaryinferior and multidirectional instability of the shoulder. Apreliminary report. J Bone Joint Surg Am 1980;62:897-908.

2. Rowe CR, Zarins B. Recurrent transient subluxation of theshoulder. J Bone Joint Surg Am 1981;63:863-872.

3. Nobuhara K, Ikeda H. Rotator interval lesion. Clin OrthopRelat Res 1987:44-50.

4. Jost B, Koch PP, Gerber C. Anatomy and functional aspects ofthe rotator interval. J Shoulder Elbow Surg 2000;9:336-341.

5. Yang HF, Tang KL, Chen W, et al. An anatomic and histologicstudy of the coracohumeral ligament. J Shoulder Elbow Surg2009;18:305-310.

6. Bennett WF. Subscapularis, medial, and lateral head coraco-humeral ligament insertion anatomy. Arthroscopic appearanceand incidence of “hidden” rotator interval lesions. Arthroscopy2001;17:173-180.

7. Boardman ND, Debski RE, Warner JJ, et al. Tensile propertiesof the superior glenohumeral and coracohumeral ligaments. JShoulder Elbow Surg 1996;5:249-254.

8. Itoi E, Berglund LJ, Grabowski JJ, Naggar L, Morrey BF, AnKN. Superior-inferior stability of the shoulder: Role of thecoracohumeral ligament and the rotator interval capsule. MayoClin Proc 1998;73:508-515.

9. Neer CS II, Satterlee CC, Dalsey RM, Flatow EL. The anat-omy and potential effects of contracture of the coracohumeralligament. Clin Orthop Relat Res 1992:182-185.

10. Harryman DT II, Sidles JA, Harris SL, Matsen FA III. The roleof the rotator interval capsule in passive motion and stability ofthe shoulder. J Bone Joint Surg Am 1992;74:53-66.

11. Arai R, Mochizuki T, Yamaguchi K, et al. Functional anatomyof the superior glenohumeral and coracohumeral ligamentsand the subscapularis tendon in view of stabilization of thelong head of the biceps tendon. J Shoulder Elbow Surg 2010;19:58-64.

12. Walch G, Nove-Josserand L, Levigne C, Renaud E. Tears ofthe supraspinatus tendon associated with “hidden” lesions ofthe rotator interval. J Shoulder Elbow Surg 1994;3:353-360.

13. Werner A, Mueller T, Boehm D, Gohlke F. The stabilizingsling for the long head of the biceps tendon in the rotatorcuff interval. A histoanatomic study. Am J Sports Med

2000;28:28-31.

5

5

5

5

5

11ROTATOR INTERVAL DISEASE

14. Habermeyer P, Magosch P, Pritsch M, Scheibel MT, Lichten-berg S. Anterosuperior impingement of the shoulder as a resultof pulley lesions: A prospective arthroscopic study. J ShoulderElbow Surg 2004;13:5-12.

15. Carbone S, Gumina S, Vestri AR, Postacchini R. Coracoidpain test: A new clinical sign of shoulder adhesive capsulitis.Int Orthop 2009;34:385-388.

16. Field LD, Warren RF, O’Brien SJ, Altchek DW, WickiewiczTL. Isolated closure of rotator interval defects for shoulderinstability. Am J Sports Med 1995;23:557-563.

17. Fitzpatrick MJ, Powell SE, Tibone JE, Warren RF. Theanatomy, pathology, and definitive treatment of rotator in-terval lesions: Current concepts. Arthroscopy 2003;19:70-79 (Suppl 1).

18. Giaroli EL, Major NM, Lemley DE, Lee J. Coracohumeralinterval imaging in subcoracoid impingement syndrome onMRI. AJR Am J Roentgenol 2006;186:242-246.

19. Paulson MM, Watnik NF, Dines DM. Coracoid impingementsyndrome, rotator interval reconstruction, and biceps tenodesisin the overhead athlete. Orthop Clin North Am 2001;32:485-493, ix.

20. Mologne TS, Zhao K, Hongo M, Romeo AA, An KN,Provencher MT. The addition of rotator interval closure afterarthroscopic repair of either anterior or posterior shoulderinstability: Effect on glenohumeral translation and range ofmotion. Am J Sports Med 2008;36:1123-1131.

21. Shafer BL, Mihata T, McGarry MH, Tibone JE, Lee TQ.Effects of capsular plication and rotator interval closure insimulated multidirectional shoulder instability. J Bone JointSurg Am 2008;90:136-144.

22. Yamamoto N, Itoi E, Tuoheti Y, et al. Effect of rotator intervalclosure on glenohumeral stability and motion: A cadavericstudy. J Shoulder Elbow Surg 2006;15:750-758.

23. Gambill ML, Timothy SM, Matthew TP. Dislocation of thelong head of the biceps tendon with intact subscapularis andsupraspinatus tendons. J Shoulder Elbow Surg 2006;15:e20-e22.

24. MacDonald K, Bridger J, Cash C, Parkin I. Transverse hu-meral ligament: Does it exist? Clin Anat 2007;20:663-667.

25. Braun S, Millett P, Yongpravat C, et al. Biomechanical eval-uation of shear force vectors leading to injury of the bicepsreflection pulley: A biplane fluoroscopy study on cadavericshoulders. Am J Sports Med 2010;38:1015-1024.

26. Warner J, Deng X, Warren R, Torzilli P, O’Brien S. Supero-inferior translation in the intact and vented glenohumeral joint.J Shoulder Elbow Surg 1993;2:99-105.

27. Gibb TD, Sidles JA, Harryman DT II, McQuade KJ, MatsenFA III. The effect of capsular venting on glenohumeral laxity.Clin Orthop Relat Res 1991:120-127.

28. Griggs SM, Ahn A, Green A. Idiopathic adhesive capsulitis. Aprospective functional outcome study of nonoperative treat-ment. J Bone Joint Surg Am 2000;82:1398-1407.

29. Massoud SN, Pearse EO, Levy O, Copeland SA. Operativemanagement of the frozen shoulder in patients with diabetes. JShoulder Elbow Surg 2002;11:609-613.

30. Mengiardi B, Pfirrmann CW, Gerber C, Hodler J, Zanetti M.Frozen shoulder: MR arthrographic findings. Radiology 2004;233:486-492.

31. Kim KC, Rhee KJ, Shin HD. Adhesive capsulitis of the shoul-der: Dimensions of the rotator interval measured with mag-netic resonance arthrography. J Shoulder Elbow Surg 2009;18:437-442.

32. Kim KC, Rhee KJ, Shin HD, Kim YM. Estimating the dimen-sions of the rotator interval with use of magnetic resonancearthrography. J Bone Joint Surg Am 2007;89:2450-2455.

33. Homsi C, Bordalo-Rodrigues M, da Silva JJ, Stump XM.Ultrasound in adhesive capsulitis of the shoulder: Is assess-

ment of the coracohumeral ligament a valuable diagnostictool? Skeletal Radiol 2006;35:673-678.

34. Bunker TD, Reilly J, Baird KS, Hamblen DL. Expression ofgrowth factors, cytokines and matrix metalloproteinases infrozen shoulder. J Bone Joint Surg Br 2000;82:768-773.

35. Hand GC, Athanasou NA, Matthews T, Carr AJ. The pathologyof frozen shoulder. J Bone Joint Surg Br 2007;89:928-932.

36. Levine WN, Kashyap CP, Bak SF, Ahmad CS, Blaine TA,Bigliani LU. Nonoperative management of idiopathic adhesivecapsulitis. J Shoulder Elbow Surg 2007;16:569-573.

37. Lorbach O, Anagnostakos K, Scherf C, Seil R, Kohn D, PapeD. Nonoperative management of adhesive capsulitis of theshoulder: Oral cortisone application versus intra-articular cor-tisone injections. J Shoulder Elbow Surg 2009;19:172-179.

38. Ng CY, Amin AK, Narborough S, McMullan L, Cook R,Brenkel IJ. Manipulation under anaesthesia and early physio-therapy facilitate recovery of patients with frozen shouldersyndrome. Scott Med J 2009;54:29-31.

39. Farrell CM, Sperling JW, Cofield RH. Manipulation for frozenshoulder: Long-term results. J Shoulder Elbow Surg 2005;14:480-484.

40. Harryman DT II, Matsen FA III, Sidles JA. Arthroscopicmanagement of refractory shoulder stiffness. Arthroscopy1997;13:133-147.

41. Tetro AM, Bauer G, Hollstien SB, Yamaguchi K. Ar-throscopic release of the rotator interval and coracohumeralligament: An anatomic study in cadavers. Arthroscopy 2002;18:145-150.

42. Pollock RG, Duralde XA, Flatow EL, Bigliani LU. The use ofarthroscopy in the treatment of resistant frozen shoulder. ClinOrthop Relat Res 1994:30-36.

43. Warner JJ, Allen A, Marks PH, Wong P. Arthroscopic releasefor chronic, refractory adhesive capsulitis of the shoulder.J Bone Joint Surg Am 1996;78:1808-1816.

44. Bell S, Coghlan J, Richardson M. Hydrodilatation in the man-agement of shoulder capsulitis. Australas Radiol 2003;47:247-251.

45. Provencher MT, Dewing CB, Bell SJ, et al. An analysis of therotator interval in patients with anterior, posterior, and multi-directional shoulder instability. Arthroscopy 2008;24:921-929.

46. Levine WN, Arroyo JS, Pollock RG, Flatow EL, Bigliani LU.Open revision stabilization surgery for recurrent anterior gle-nohumeral instability. Am J Sports Med 2000;28:156-160.

47. Provencher MT, Mologne TS, Hongo M, Zhao K, Tasto JP, AnKN. Arthroscopic versus open rotator interval closure: Biome-chanical evaluation of stability and motion. Arthroscopy 2007;23:583-592.

48. Van der Reis W, Wolf EM. Arthroscopic rotator cuff intervalcapsular closure. Orthopedics 2001;24:657-661.

49. Savoie FH, Holt MS, Field LD, Ramsey JR. Arthroscopicmanagement of posterior instability: Evolution of techniqueand results. Arthroscopy 2008;24:389-396.

50. Savoie FH. Authors’ reply. Arthroscopy 2010;25:110-111.1. Farber AJ, ElAttrache NS, Tibone JE, McGarry MH, Lee TQ.

Biomechanical analysis comparing a traditional superior-infe-rior arthroscopic rotator interval closure with a novel medial-lateral technique in a cadaveric multidirectional instabilitymodel. Am J Sports Med 2009;37:1178-1185.

2. Gartsman GM, Taverna E, Hammerman SM. Arthroscopicrotator interval repair in glenohumeral instability: Descriptionof an operative technique. Arthroscopy 1999;15:330-332.

3. Petersson CJ. Spontaneous medial dislocation of the tendon ofthe long biceps brachii. An anatomic study of prevalence andpathomechanics. Clin Orthop Relat Res 1986:224-227.

4. Braun S, Horan M, Millett P. Lesions of the biceps pulley: Arethere other common associated pathologies. Am J Sports Med.In press.

5. Bennett WF. Arthroscopic repair of anterosuperior (supraspi-natus/subscapularis) rotator cuff tears: A prospective cohort

with 2- to 4-year follow-up. Classification of biceps sublux-ation/instability. Arthroscopy 2003;19:21-33.

5

6

6

6

6

6

6

6

6

6

6

7

7

7

7

7

7

7

12 T. R. GASKILL ET AL.

56. Boileau P, Ahrens PM, Hatzidakis AM. Entrapment of thelong head of the biceps tendon: The hourglass biceps—Acause of pain and locking of the shoulder. J Shoulder ElbowSurg 2004;13:249-257.

57. Vinson EN, Major NM, Higgins LD. Magnetic resonanceimaging findings associated with surgically proven rotatorinterval lesions. Skeletal Radiol 2007;36:405-410.

58. Bennett WF. Arthroscopic bicipital sheath repair: Two-yearfollow-up with pulley lesions. Arthroscopy 2004;20:964-973.

9. Mazzocca AD, Rios CG, Romeo AA, Arciero RA. Subpectoralbiceps tenodesis with interference screw fixation. Arthroscopy2005;21:896e.1-896.e7.

0. Maier D, Jaeger M, Suedkamp NP, Koestler W. Stabilizationof the long head of the biceps tendon in the context of earlyrepair of traumatic subscapularis tendon tears. J Bone JointSurg Am 2007;89:1763-1769.

1. Gerber C, Terrier F, Ganz R. The role of the coracoid processin the chronic impingement syndrome. J Bone Joint Surg Br1985;67:703-708.

2. Renoux S, Monet J, Pupin P, et al. Preliminary note onbiometric data relating to the human coraco-acromial arch.Surg Radiol Anat 1986;8:189-195.

3. Russo R, Togo F. The subcoracoid impingement syndrome:Clinical, semeiologic and therapeutic considerations. Ital J Or-thop Traumatol 1991;17:351-358.

4. Gerber C, Terrier F, Zehnder R, Ganz R. The subcoracoidspace. An anatomic study. Clin Orthop Relat Res 1987:132-138.

5. Kragh JF Jr, Doukas WC, Basamania CJ. Primary coracoidimpingement syndrome. Am J Orthop (Belle Mead NJ) 2004;33:229-232; discussion 232.

6. McFarland EG, Selhi HS, Keyurapan E. Clinical evaluation ofimpingement: What to do and what works. J Bone Joint Surg

Am 2006;88:432-441.

7. Dines DM, Warren RF, Inglis AE, Pavlov H. The coracoidimpingement syndrome. J Bone Joint Surg Br 1990;72:314-316.

8. Friedman RJ, Bonutti PM, Genez B. Cine magnetic resonanceimaging of the subcoracoid region. Orthopedics 1998;21:545-548.

9. Millett P, Elser F, Anstett T, Brunkhorst J, Torry M, GiphartJ. In vivo analysis of coracohumeral distances in differentangles of rotation: A 3D biplane fluoroscopy study. Presentedat the Orthopaedic Research Society meeting, New Orleans,LA, March 2010.

0. Radas CB, Pieper HG. The coracoid impingement of thesubscapularis tendon: A cadaver study. J Shoulder Elbow Surg2004;13:154-159.

1. Patte D. The subcoracoid impingement. Clin Orthop Relat Res1990:55-59.

2. Lo IK, Burkhart SS. Arthroscopic coracoplasty through therotator interval. Arthroscopy 2003;19:667-671.

3. Karnaugh RD, Sperling JW, Warren RF. Arthroscopic treat-ment of coracoid impingement. Arthroscopy 2001;17:784-787.

4. Braun S, Horan M, Tello T, Millett P. Coracoid impingement:A prospective cohort study on the association between cora-coid interval narrowing and anterior shoulder pathology. Pre-sented at the Arthroscopy Association of North America An-nual Meeting, San Diego, CA, August 2009.

5. Greene T, Horan M, Millett P. Subcoracoid impingement:Factors associated with the size and location of the coracohu-meral interval. Presented at the Arthoscopy Association ofNorth America meeting, Hollywood, FL, May 2010.

6. Millett PJ, Sanders B, Gobezie R, Braun S, Warner JJ. Inter-ference screw vs. suture anchor fixation for open subpectoralbiceps tenodesis: Does it matter? BMC Musculoskelet Disord

2008;9:121.e1-121.e6.