Al

1

SCIENTIFIC ARTICLE

A Cadaver Model That Investigates Irreducible

Metacarpophalangeal Joint Dislocation

Ahmed M. Afifi, MD, Amanda Medoro, BS, Christina Salas, MS, Mahmoud Reda Taha, PhD,Tahseen Cheema, MD

Purpose Controversy exists over the pathologic anatomy of irreducible dorsal metacarpopha-langeal (MCP) dislocation. The aim of this work is to develop a cadaveric model of MCPjoint dislocation that closely simulates the clinical situation and to study the structures aroundthe MCP joint and their contribution to irreducibility of the dislocation.

Methods Nine fresh-frozen cadaveric specimens were amputated at the radiocarpal joint andstabilized in a specially formulated fixture. The dislocation was created by an impact loaddelivered by a servohydraulic testing machine, at a displacement rate of 1000 mm/s and with amaximum displacement of 60 mm. An irreducible dislocation was successfully created in 6 indexfingers. An attempt at closed reduction was followed by a dissection of the dislocated joint.

Results In the 6 examined specimens, the flexor tendons were ulnar to the joint in all cases,the radial digital nerve was superficial (5 cases) or radial (5 cases) to the metacarpal head,and the lumbrical was usually radial (5 of 6 cases) to the joint. Division of the superficialtransverse metacarpal ligaments, natatory ligaments, flexor tendons, or lumbricals does notaid reduction of the dislocation. Division of the volar plate was necessary for reduction of thedislocation in all 6 cases, whereas division of the deep transverse metacarpal ligaments doesnot allow reduction of the dislocation.

Conclusions We present a model for creating an irreducible MCP joint dislocation using animpact load that simulates the clinical situation. The volar plate is the primary structurepreventing reduction of the dislocation. Division of the deep transverse metacarpal ligamentis not effective in reducing the dislocation. The flexor tendons, lumbricals, superficialtransverse metacarpal ligament and natatory ligaments do not contribute to irreducibility.The anatomy of the structures surrounding the MCP joint is variable, and careful dissectionto prevent iatrogenic injuries is mandatory. (J Hand Surg 2009;34A:1506–1511. Copyright© 2009 by the American Society for Surgery of the Hand. All rights reserved.)

Key words Metacarpophalangeal joint, dislocation, volar plate, deep transverse metacarpalligament.

1qts

MONG KAPLAN’S MANY contributions to hand sur-gery is his description of the pathologic anat-omy of complex (irreducible) metacarpopha-

angeal (MCP) joint dislocation in a classic article in

FromtheDepartmentofPlasticSurgery,ClevelandClinic,Cleveland,OH;DepartmentofCivilEngineering,University of New Mexico, Albuquerque, NM; Department of Orthopedics, University of New Mexico,Albuquerque, NM; Department of Plastic Surgery, Cairo University, Cairo, Egypt.

Received for publication December 30, 2008; accepted in revised form June 2, 2009.

No benefits in any form have been received or will be received related directly or indirectly to the

subject of this article.

506 � © ASSH � Published by Elsevier, Inc. All rights reserved.

957.1 This article, based on a single case, is still widelyuoted in the literature. Kaplan believed that a noosehat formed around the joint (by the natatory ligament,uperficial transverse metacarpal ligament [STML],

orresponding author: Ahmed M. Afifi, MD, Cleveland Clinic, Department of Plastic Surgery, A60,500 Euclid Avenue, Cleveland, OH 44195; e-mail: ahafifi@yahoo.com.

363-5023/09/34A08-0019$36.00/0oi:10.1016/j.jhsa.2009.06.001

C9

0d

lic te

METACARPOPHALANGEAL JOINT DISLOCATION 1507

flexor tendons, and lumbricals) contributed to the irre-ducibility of the dislocation. However, many surgeonsbelieve that Kaplan’s description is not entirely accu-rate.2–4 An alternative theory suggested that the deeptransverse metacarpal ligament (DTML) and the volarplates form a continuous band volar to the MCP joint,that this band remains intact but is displaced dorsal tothe metacarpal head at the site of the dislocated joint,and that interruption of this band by dividing either theDTML or the volar plate will allow reduction of thedislocation.2,3 Previous cadaveric models of MCP dis-location involved an open surgical dissection of thejoint, release of the volar plate proximally, and manu-ally dislocating the joint, thereby re-creating the anat-omy as usually observed during surgery as opposed tore-creating the usual traumatic force that caused thedislocation. The aim of this work is to (1) develop acadaveric model of MCP dislocation in the index fingerthat closely simulates the clinical situation; (2) definethe location of the STML, natatory ligaments, DTML,flexor tendons, lumbrical, index finger radial neurovas-

FIGURE 1: The hand stabilized in the servohydrau

cular bundle, and volar plate in an irreducible MCP

JHS �Vol A, O

dislocation; and (3) evaluate their contribution to theirreducibility of the dislocation.

MATERIALS AND METHODSNine fresh-frozen cadaveric specimens (6 male and 3female) were used in this study, ranging in age from 38to 59 years. The specimens were thawed at room tem-perature for approximately 12–24 hours before thestudy. The specimens were examined to exclude thepresence of previous pathology or scars in the hand orfingers.

The specimens were amputated at the level of thewrist (radiocarpal joint). The flexor tendons were su-tured to the extensor tendons at the amputation level tomaintain their tension. The hand was then stabilizedbetween 2 metal plates on a specially fabricated test jigand fixed in place using 2 Steinmann pins passed in avolar–dorsal direction at the level of the distal carpalrow and proximal metacarpals (Fig. 1). The index fingerwas extended at the MCP joint to approximately 90°. Itwas then fixed to the actuator test jig with 2 transverse

sting apparatus before creation of the dislocation.

1.5-mm (0.059-in) K-wires, first at the level of the head

ctober

1508 METACARPOPHALANGEAL JOINT DISLOCATION

of the proximal phalanx and the second more distallynear the distal interphalangeal joint. A servohydraulictesting machine (858 Bionix; MTS Systems, Minneap-olis, MN) equipped with a 15 kN maximum capacityload cell was used to deliver an impact load to the fingerand create the dislocation. This load was applied par-allel to the coronal plane of the hand, in a distal-to-proximal direction, at a displacement rate of 1000 mm/sand with a maximum displacement of 60 mm. Thepalmar surface of the hand was oriented away fromthe axis of loading, simulating a bent positiontoward the posterior forearm. The hand was re-leased from the jig and tested to confirm the pres-ence of dislocation, its irreducibility, and the pres-ence of any fractures. Two specimens were excludedbecause of a proximal phalanx fracture or a metacar-pal fracture. These were both from the same cadaver,and they were notably more muscular and bulkierthan the other specimens. A third specimen wasexcluded because of an error in loading with theservohydraulic testing machine. In the remaining 6specimens, an index finger MCP dislocation wasproduced, and an attempt at closed reduction wasperformed by 2 authors, using a force comparable tothat used clinically.5 Specimens with irreducibleMCP dislocation were surgically dissected to identifythe location of the STML, natatory ligaments,DTML, flexor tendons, lumbrical, index finger radialneurovascular bundle, and volar plate and their con-tribution to irreducibility. Surgical dissection pro-ceeded through a volar approach. Initially, the struc-tures forming the noose around the metacarpal head(natatory ligament, STML, flexor tendons, and lum-bricals, as described by Kaplan) were examined andthen divided. If release of these structures failed toallow reduction of the dislocation, the volar plate wasidentified and divided along its midline, and an at-tempt at reduction was performed. After reduction,the original irreducible state was reproduced bysuturing the volar plate behind the metacarpalhead, similar to its original position. The DTMLwas then divided halfway between the volar platesof the index and long fingers, and joint reductionwas attempted.

RESULTSIn the 6 index fingers in this study, an irreducible MCPdislocation was successfully created, and attempts atclosed reduction failed. Surgical dissection of the irre-ducible joints revealed that, in all the cases, the volarplate was avulsed from its proximal attachment to the

metacarpal head and interposed in the joint, preventing

JHS �Vol A, O

reduction (Figs. 2, 3). The flexor tendons were dis-placed ulnarly, with the A2 pulley intact and the tendontethered between the second and third metacarpal heads(Fig. 4). The radial neurovascular bundle of the indexfinger was found on the volar aspect of the metacarpalhead immediately deep to the volar skin in 3 cases (Fig.4) and was found radial to the metacarpal head in theother 3 cases. The lumbrical muscle was found radial tothe metacarpal head in 5 cases (Fig. 5) and ulnar in 1case (Fig. 6). The natatory ligament was distant fromthe dislocation and had no contribution to the irreduc-ibility. The STML was superficial to the neck of themetacarpal and did not contribute to the irreducibility ofthe dislocation. The DTML was at least partially incontinuity with the volar plate, although the exact in-tegrity of its whole length is difficult to assess in thepresence of the dislocation. In 3 specimens, a sesamoidbone was found in the volar plate (Fig. 3). Completerelease of the natatory ligament, STML, flexor tendons,

FIGURE 2: A dislocated index MCP joint with a freer is underthe DTML, which is still attached to the volar plate (VP) lyingdorsal to the metacarpal (M) head. The roof of the A2 pulley(FS, flexor sheath) and a section of the flexor tendons (FT)have been excised. Note the location of the flexor sheathmedial to the joint and that the joint is still dislocated evenafter the tendons have been excised.

and lumbricals did not allow reduction of the disloca-

ctober

METACARPOPHALANGEAL JOINT DISLOCATION 1509

tion. Release of the volar plate allowed easy reductionof the dislocation in all cases. However, with the volarplate severed, the joint could be easily redislocated.Repair of the incised volar plate dorsal to the metacar-pal head (in its original place during the initial dissec-tion) re-created the irreducibility. Releasing the DTMLat this stage did allow more mobility of the volar platebut not to the extent of allowing reduction. The repair inthe volar plate was then re-created with the volar platein its normal position, volar to MCP joint. This pre-vented redislocation of the joint in all cases.

DISCUSSIONDorsal MCP dislocations are divided into reducible andirreducible (also referred to as complex) dislocations.Kaplan described the anatomy of irreducible MCP dis-location in a classic article in 1957, based on observa-tions from a single clinical case.1 He described the volarplate as being entrapped in the joint dorsal to the meta-carpal head, with a noose around the metacarpal head

FIGURE 3: Dorsal view of the dislocated index MCP jointafter excision of the extensor tendons (ET). The freer is underthe volar plate (VP) lying dorsal to the metacarpal head. Notethe sesamoid bone (S) in the volar plate. (Prox Ph, proximalphalanx).

formed by the natatory ligament dorsally, the STML

JHS �Vol A, O

volarly, and the flexor tendons and the lumbricals oneither side. Kaplan’s original description is still often

FIGURE 5: The lumbrical muscle (L) is usually found lateral to themetacarpal head (M) with the flexor tendons (FT) lying medial.

FIGURE 4: The A2 pulley (FS, flexor sheath) has beenpartially excised to show the flexor tendons (FT) medial to themetacarpal head (M). The neurovascular bundle (NV) is lyingin a superficial position volar to the metacarpal head and isliable to iatrogenic injury during a volar skin incision.

quoted,6,7 although authors have varied in their descrip-

ctober

1510 METACARPOPHALANGEAL JOINT DISLOCATION

tion of the structures preventing reduction.2–4,8 In thisstudy, we attempted to create a model for MCP dislo-cation through application of an impact force to theextended index finger and to study the structures sur-rounding the MCP joint and their contribution to irre-ducibility of the dislocation. We opted to examine onlythe index finger because it is the joint most commonlydislocated.

Previous attempts at creating a dorsal MCP disloca-tion in cadavers were limited to manual dislocation ofthe joint by an open surgical dissection through a volarapproach, detaching the volar plate from its attachmentto the metacarpal proximally, and forcefully hyperex-tending the joint.3,4 These efforts focused on re-creatingthe pathologic anatomy as usually observed during sur-gery, as opposed to creating the typical traumatic forcethat causes the injury. We present a model for MCPdislocation that simulates the clinical scenario moreclosely, as it does not involve incisions or dissectionsaround the joint or surgical detachment of the volarplate before creating the dislocation. In addition, thecausative force is represented by a high-rate impactload, similar to the clinical situation.

The STML (also known as the transverse ligamentof the palmar aponeurosis) constitutes the transversedistal fibers of the palmar aponeurosis, running trans-versely between the pretendinous bands. The natatoryligament (interdigital ligament) consists of thin fiberslying in the distal-most extent of the web spaces justbeneath the skin. In this study, we found that neither theSTML nor the natatory ligament contributes to irreduc-ibility of the MCP dislocation, as they were distant fromthe dislocation, they were under only moderate tension,

FIGURE 6: A dislocated MCP joint with the first lumbrical (L)muscle lying medial to the metacarpal head. (FT, flexortendon; DN, digital nerve).

and their division did not alter the dislocation or the

JHS �Vol A, O

adjacent anatomy. Moreover, complete division of thelumbricals or the flexor tendons does not allow reduc-tion of the dislocation. Kaplan’s original theory of anoose forming around the metacarpal head and prevent-ing reduction was not borne out in any of the cadaverstested.

The DTML (also referred to as interglenoid liga-ment) is a sturdy ligament connecting the volar plates.The connection between the DTML and the volar plateis inseparable, and it receives the insertion of the sagittalband dorsally and the flexor sheath volarly (Zancollireferred to this point as the dorsal assemblage nucl-eus).9,10 The volar plate (glenoid plate) and theDTML (interglenoid ligament) seem to form a con-tinuous band volar to the MCP joint. In MCP dislo-cation, this band shifts dorsal to the affected MCPjoint.10 Several authors have argued that division ofeither the volar plate or the DTML will interrupt thisband and allow its relocation to its normal positionvolar to the MCP joint.2,3,11,12 In our cadaver dissec-tions, we found that division of the volar plate isnecessary to allow reduction of the dislocation. Di-vision of the DTML alone, aiming to interrupt thecontinuous band formed by the volar plate and theligaments, did not allow reduction. This could beexplained by the insertion of several other structuresto the junction of the DTML and the volar plate(specifically the flexor pulley, the saggital band, andpossibly connections to the tendons of the lumbricalsand the interossei) maintaining the irreducibility ofthe volar plate. In addition, identification of theDTML in the presence of the dislocation is techni-cally difficult. The difference between our findingsand other studies could be explained by the differentmethods of creating the dislocation, because in otherstudies the different structures were identified beforedislocating the joint, and it is possible that the otherligaments and tendons attaching to the junction of theDTML and volar plate were severed during the initialdissection.

In all the examined specimens, the flexor tendonswere intact and ulnar to the joint, with an intact pulleysystem. The radial neurovascular bundle was foundsuperficial to the MCP joint immediately under the skinin 50% of cases, placing it at a high risk of injury duringthe skin incision. The lumbrical tendon was found ulnarto the joint in 1 case, in contrast to previous reports inwhich it is usually radial to the metacarpal head.1,6,7

This study shows that reduction was achieved onlyafter release of the volar plate, which was done afterrelease of the structures forming the noose around the

MCP head. There exists the possibility that release of

ctober

METACARPOPHALANGEAL JOINT DISLOCATION 1511

both the volar plate and the structures forming the nooseis necessary to achieve reduction. Although this is un-likely, further studies are needed to test this by revers-ing the order of the surgical release.

Our findings do confirm that the volar plate is indeedthe structure usually preventing reduction and that di-vision of the volar plate should be the primary aim ofsurgery. Perhaps the biggest controversy in surgicalmanagement of the dislocation is the approach: volar ordorsal?13 Proponents of the volar approach argue itssuperiority for 2 reasons. The first advantage is accessto the flexor tendons, STML, natatory ligament, andintrinsic tendons, which are the structures suggested byKaplan to contribute to irreducibility. With the evidencefrom this study and the experience of other authors thatthe volar plate is the culprit behind irreducibility andthat the noose theory is inaccurate, this argument isrebutted. The second advantage of a volar approach isthe opportunity to repair the incised volar plate in itscorrect position after joint reduction. We found thatsuch repair does make the joint more stable and resis-tant to redislocation. However, there is no evidencefrom clinical studies that choosing not to repair thevolar plate will lead to long-term complications.14

We present a novel model for creation of an irreduc-ible MCP dislocation using an impact load simulatingthe clinical situation. Our findings confirm that the volarplate is the primary structure preventing reduction ofthe dislocation. Division of the DTML is not effectivein reducing the dislocation, and the noose theory doesnot seem to be accurate. The flexor tendons are dis-placed ulnarly, the lumbricals (usually) radially, and theradial neurovascular bundle is either radial or superfi-cial to the metacarpal head (just under the skin, makingit prone to iatrogenic injury during the skin incision).The anatomy of the structures surrounding the MCP is

variable, and careful dissection to prevent iatrogenic

JHS �Vol A, O

injuries is mandatory if a volar approach is used. Fur-ther studies are necessary to compare the volar anddorsal approaches, assess the stability of the MCP jointafter release of the volar plate, and examine a largernumber of cadavers.

REFERENCES1. Kaplan EB. Dorsal dislocation of the metacarpophalangeal joint of

the index finger. J Bone Joint Surg 1957;39A:1081–1086.2. Green DP, Terry GC. Complex dislocation of the metacarpophalangeal

joint. Correlative pathological anatomy. J Bone Joint Surg 1973;55A:1480–1486.

3. Barry K, McGee H, Curtin J. Complex dislocation of the metacarpo-phalangeal joint of the index finger: a comparison of the surgicalapproaches. J Hand Surg 1988;13B:466–468.

4. Decoster TA, McGrew D, Omer GE. Complex dorsal dislocation ofthe metacarpophalangeal joint, the deep transverse metacarpal liga-ment as a barrier to reduction. Iowa Ortho J 1988;8:9–12.

5. Mclaughlin HL. Complex “locked” dislocation of the metacarpopha-langeal joints. J Trauma 1965;5:683–688.

6. Slade JF, Oetgen ME. Phalangeal injuries. In: Trumble TE, BudoffJE, eds. Hand surgery update. 4th ed. Rosemont, IL: AmericanSociety of Surgery of the Hand, 2007:1–19.

7. Lattanza LL, Choi PD. Intraarticular injuries of the metacarpopha-langeal and carpometacarpal joints. In: Berger RA, Weiss AC, eds.Hand surgery. Philadelphia: Lippincott Williams & Wilkins, 2004:176–179.

8. Orozco JR, Rayan GM. Complex dorsal metacarpophalangeal jointdislocation caused by interosseous tendon entrapment: case report.J Hand Surg 2008;33A:555–557.

9. Yu Hl, Chase RA, Strauch B. Atlas of hand anatomy and clinicalimplications. St. Louis, MO: Mosby, 2004:228–229.

10. Zancolli E. Structural and dynamic basis of the hand surgery. Phil-adelphia: J.B. Lippincott, 1979:23.

11. Murphy AF, Stark HH. Closed dislocation of the metacarpophalan-geal joint of the index finger. J Bone Joint Surg 1967;49A:1579–1586.

12. Nussbaum R, Sadler AH. An isolated, closed, complex dislocation ofthe metacarpophalangeal joint of the long finger: a unique case.J Hand Surg 1986;11A:558–561.

13. Becton JL, Christian JD Jr, Goodwin HN, Jackson JG III. A simpli-fied technique for treating the complex dislocation of the indexmetacarpophalangeal joint. J Bone Joint Surg 1975;57A:698–700.

14. Sodha S, Breslow GD, Chang B. Percutaneous technique for reduc-tion of complex metacarpophalangeal dislocations. Ann Plast Surg

2004;52:562–565.

ctober