article osteoarthritis

REVIEW ARTICLE

Tnrmt(toozaa

toakwoBwussvtl

Osteoarthritis of the WristKrista E. Weiss, Craig M. Rodner, MD

From Harvard College, Cambridge, MA and Department of Orthopaedic Surgery, University of ConnecticutHealth Center, Farmington, CT.

Osteoarthritis of the wrist is one of the most common conditions encountered by handsurgeons. It may result from a nonunited or malunited fracture of the scaphoid or distal radius;disruption of the intercarpal, radiocarpal, radioulnar, or ulnocarpal ligaments; avascularnecrosis of the carpus; or a developmental abnormality. Whatever the cause, subsequentabnormal joint loading produces a spectrum of symptoms, from mild swelling to considerablepain and limitations of motion as the involved joints degenerate. A meticulous clinical andradiographic evaluation is required so that the pain-generating articulation(s) can be identi-fied and eliminated. This article reviews common causes of wrist osteoarthritis and theirsurgical treatment alternatives. (J Hand Surg 2007;32A:725–746. Copyright © 2007 by theAmerican Society for Surgery of the Hand.)Key words: Wrist, osteoarthritis, arthrodesis, carpectomy, SLAC.

ob(aum

rldht

DBpiwTtccjpabcs

here are several different causes, both idio-pathic and traumatic, of wrist osteoarthritis.Untreated cases of idiopathic carpal avascular

ecrosis, as in Kienböck’s1 or Preiser’s2 disease, mayesult in radiocarpal arthritis. Congenital wrist abnor-alities, such as Madelung’s deformity,3,4 can lead

o radiocarpal, ulnocarpal, and distal radioulnar jointDRUJ) incongruities, producing altered loading pat-erns and premature cartilage degeneration through-ut the wrist. Another nontraumatic form of wriststeoarthritis occurs at the scaphotrapezium-trape-oid (STT) joint; this is usually idiopathic in originnd is present when all of the other joints of the wristre normal.5

Traumatic causes of wrist arthritis include injurieso ligament and bone. Although injury to many of thether wrist ligaments can lead to progressive wristrthrosis, chronic scapholunate tears in particular arenown to produce intercarpal instability, alteredrist kinematics and joint loading, and degenerationf the radiocarpal joint. As described by Watson andallet,6 scapholunate advanced collapse (SLAC)rist arthritis progresses in a predictable fashion,ltimately leading to pancarpal arthritis. Fracture andubsequent nonunion of the scaphoid also leads to aeries of predictable degenerative changes, first in-olving the radial styloid tip and then progressing tohe more proximal radioscaphoid joint and eventually
eading to pancarpal arthritis (Fig. 1). This sequence I
f events is analogous to SLAC wrist and haseen termed scaphoid nonunion advanced collapseSNAC). Wrist osteoarthritis can also occur second-ry to an intra-articular fracture of the distal radius orlna or from an extra-articular fracture resulting inalunion and abnormal joint loading.The surgical treatment of the osteoarthritic wrist

ests on basic principles that take into account theocation of the arthritis and the most reliable proce-ure that might eliminate the patient’s pain, improveis/her function, and prevent further progression ofhe pain-generating degenerative process.

iagnosisefore deciding which reconstructive surgery is ap-ropriate for a particular patient, a meticulous phys-cal examination and a radiographic assessment ofhich wrist joints are pain generators are essential.he importance of palpating the joints of the wrist to

ry to ascertain the area of the most tenderness orrepitus cannot be overstated. Possible culprits in-lude any one or combination of the radiocarpaloint, ulnocarpal joint, DRUJ, or any of the intercar-al joints, including degeneration within a row (suchs the trapeziotrapezoid or pisotriquetral joints) oretween the proximal and distal rows (ie, the mid-arpal joint). A key question is whether a plannedurgery will cure a patient of his/her chief complaint.
n cases in which the pain generator may be less
The Journal of Hand Surgery 725

otoksed

fts(mdpfwjtot(rwaaw

TFhicc

avnptctipai

ij

F(aa ed wi

Fpeh

726 The Journal of Hand Surgery / Vol. 32A No. 5 May–June 2007

bvious, prognostic anesthetic and/or cortisone injec-ions can be exceedingly helpful. Carpometacarpalsteoarthritis, especially of the thumb, must also beept in mind in the differential diagnosis as a perva-ive cause of hand arthritis, but it is usually fairlyasy to distinguish from more proximal areas ofisease on clinical examination.Because the surgical procedures that are per-

ormed on the osteoarthritic wrist are directly relatedo which joints are preserved and which are de-troyed, a meticulous examination of posteroanteriorPA) and lateral radiographs is also critical. Oneust thoroughly examine the radioscaphoid and ra-

iolunate joints for evidence of narrowing, osteo-hyte formation, and/or subchondral sclerosis or cystormation. In addition, the head of the capitate,hich is the primary articulation for the midcarpal

oint, must be evaluated closely.7 As a further step,he surgeon should carefully evaluate each of thether individual joints throughout the wrist to ensurehat there is no evidence of focal arthritis presentFig. 2). Subtle triquetral or ulnar-sided lunate scle-osis or cyst formation may be signs of ulnocarpalear. Although more advanced imaging studies such

s magnetic resonance imaging or bone scintigraphyre rarely necessary, they may play a role in theorkup of more subtle cases of wrist osteoarthritis.

reatmentrequently, patients with wrist osteoarthritis can beelped by nonsurgical means. The use of splint or castmmobilization, nonsteroidal anti-inflammatory medi-ations, and selective intra-articular injections of corti-

igure 1. With both chronic scapholunate ligament injury aA) stage I, affecting the radial styloid–scaphoid junction, (B)ffecting the entire radioscaphoid joint and the capitolunatend intercarpal joints and may involve the DRUJ. (Reproduc

osteroids may provide patients with improved function t

nd decreased pain. In many patients with more ad-anced disease, however, the pain relief granted byonsurgical methods is limited. As can be the case inatients with progressive wrist osteoarthritis, such ashose with SLAC or SNAC wrists or evolving lunateollapse from Kienböck’s disease, nonsurgical modali-ies may be sufficient at first but often become increas-ngly ineffective at alleviating pain as time goes on. If aatient has debilitating wrist pain referable to an osteo-rthritic joint or joints on imaging studies, surgicalntervention should be considered.

The goal of performing surgery for such a patients to eliminate the pain associated with the arthriticoint(s) while trying to preserve as much motion as

phoid nonunions, progressive osteoarthritis can occur withII, affecting the entire radioscaphoid joint, and (C) stage III,ation, (D) Stage IV, osteoarthritis affects both the radiocarpalth permission.120)

igure 2. When examining a posteroanterior radiograph, therimary joints (solid rectangles) to carefully evaluate for degen-rative changes are the radioscaphoid, radiolunate, and capitateead. The secondary joints (dotted rectangles) affected include

nd scastage

articul

he STT, ulnolunate (with impingement), and radioulnar.

pgtiwTrthmrpSfIcicc

wptae

ciswanotwmsomftidlpsp

SAw

Fcfrs

Weiss and Rodner/Osteoarthritis of the Wrist 727

ossible in the adjacent joints by maintaining con-ruity between the undamaged articular surfaces ofhe wrist. In the most severe cases of arthritis, whichnvolve the entire carpus and distal radius, a totalrist arthrodesis has traditionally been required.8–11

his surgery would eliminate all movement at theadiocarpal and ulnocarpal joints. A total wrist ar-hroplasty (TWA) may also be possible, although thisas been more typically used in patients with rheu-atoid arthritis because of the permanent activity

estrictions required.12 If the radiolunate joint ap-ears to be intact, as in the classic description of aLAC (or SNAC) wrist, then a scaphoid excision andour-corner arthrodesis may be an appealing option.f the capitate head is preserved, a proximal rowarpectomy (PRC) also becomes attractive. If there issolated ulnocarpal joint or DRUJ pathology, surgi-ally addressing the ulna via shortening or partial oromplete resection may be appropriate.

Based on several biomechanic studies13,14 of therist, general postoperative motion parameters for aatient who has had a motion-preserving reconstruc-ive procedure are well delineated. Simulated wristrthrodesis and carpal excision studies13,14 in cadav-rs have shown that any fusion involving the mid-

igure 3. (A) When exposing the dorsal wrist during a totaompartment is opened, allowing a radial transposition of theor harvesting autogenous bone graft. (Reproduced with peetinaculum at the end of the procedure, which avoids adhes

Table 1. Postoperative Motion After Wrist Procedu

Type of Procedure

Arthrodeses involving the midcarpal jointArthrodeses involving the radiocarpal joint

Without excision of distal scaphoidWith excision of the distal scaphoid

PRC

ite.

arpal joint will allow a postoperative arc of motionn the range of 50% to 67% of that of the unaffectedide. Arthrodeses involving the radiocarpal joint,hich preserves motion at the midcarpal joint, will

llow a residual arc of motion of 33% to 40% of aormal motion arc unless the distal pole of the scaph-id is excised as well, which adds an additional 15%o 25%.15 Elimination of the proximal carpal rowill preserve approximately 50% to 75% of the nor-al motion arc (Table 1). These biomechanic data

hould be kept in mind when reviewing the clinicalutcome literature from the various surgical treat-ent options for wrist osteoarthritis. Other important

actors to keep in mind when determining whichreatment is most appropriate for a particular patients age (both absolute and physiologic) and functionalemands. Although most reconstructive proceduresikely have enough longevity to provide satisfactoryain relief for the lifetime of an elderly patient, theame cannot be said with certainty for a youngererson.

urgical Approachcommon surgical approach to the osteoarthritic

rist, as in scaphoid excision and four-corner arth-

t arthrodesis, a longitudinal incision is used and the thirdndon and exposing Lister’s tubercle, which can be removed

on.121) (B) The EPL tendon is left superficial to the closednd potential rupture from sharp bone edges at the bone graft

Usual Final Motion (Flexion/Extension Arc), %Relative to the Unaffected Side

50–67

33–4050–6750–75

l wrisEPL te

rmissiions a

res

rwapdtcgLattdoAammcist

olr(cblboopcn

mgtieaottta

ww

W

●

●

●

Ftdip

Fpp


odeses, proximal row carpectomies (PRC), totalrist arthrodeses, and total wrist arthroplasties, useslongitudinal incision in line with the third metacar-al to the distal radius.11,16 This incision is takenown through the skin and subcutaneous tissues tohe dorsal retinacular structures. The third dorsalompartment is opened and the extensor pollicis lon-us (EPL) tendon is transposed radially, exposingister’s tubercle, which is often removed to obtainutogenous bone graft (Fig. 3). In the floor of thehird compartment and extending distally betweenhe second and fourth compartments, a subcapsularissection exposing the dorsal carpus can easily bebtained and extended out to the third metacarpal.lternative capsular incisions include a transverse

pproach (good for limited arthrodeses of smalleragnitude) and a dorsal “T” incision (useful forore extensive limited arthrodeses or proximal row

arpectomies (Fig. 4). The dorsal capsule may bencised longitudinally or transversely in a ligament-paring fashion. The relatively recent description ofhe ligament-sparing exposure provides an incision

igure 4. (A) A longitudinal (A), transverse (B), or combina-ion “T” approach (C), can be used for the dorsal capsulotomyuring a PRC or limited wrist arthrodesis. (B) Alternatively, annverted “T” incision can be used. (Reproduced withermission.122)

igure 5. (A) A ligament-sparing exposure preserving portiorovides excellent exposure to the dorsal aspect of the wrist
ermission.123) (B) The capsular flap is elevated off the carpus (C)
rientation that may minimize scarring across theigaments of the dorsal wrist and improve end flexionange of motion after reconstructive proceduresFig. 5).17 A longitudinal or transverse capsular in-ision provides excellent exposure to the carpalones themselves and to Lister’s tubercle, an excel-ent source of local autogenous bone graft. It shoulde noted that the posterior interosseous nerve residesn the floor of the fourth dorsal compartment and isften resected, presumably to help with postoperativeain control. Complete exposure of the third meta-arpal in a total wrist arthrodesis is important but isot necessary during more limited wrist arthrodeses.Transverse dorsal skin incisions are useful duringore limited wrist arthrodeses when exposing a re-

ion of more localized intercarpal arthritis such ashe STT joint. The advantage of limited transversencisions is that their appearance after surgery isxcellent with little, if any, scar formation. The dis-dvantage of this incision is that it can be extendednly with difficulty, and exposure with protection ofhe sensory nerve branches can be problematic. Ifhere is any possibility of needing more exposure athe arthrodesis site, alternative incisions are moreppropriate.

During our surgical approach to the osteoarthriticrist, we generally follow several basic principles,hich are described as follows.

rist Approach: Basic Principles

Keep the capsular incisions precise and maintainthe integrity of the capsule as much as possible toallow a careful repair.Preserve the extrinsic and intrinsic ligaments of thewrist that are not affected by the procedure toprevent secondary instability from occurring.In motion-preserving procedures, try to use trans-verse (or ligament-sparing) capsular incisions

the dorsal intercarpal ligaments and radiocarpal ligamentsg both PRC and limited wrist arthrodesis. (Reproduced with

ns ofdurin

to the radial side exposing the carpus.

●

●

●

●

●

TFEbtvwGtaaptutftiSuopatmls

Ftpso

Fobsrt

Fb


when possible to improve postoperative range ofmotion.Be careful during exposure to protect fibers of thesensory branch of the radial and ulnar nerves,which can be vulnerable during these exposures.Excise the posterior interosseous nerve in the floorof the fourth compartment to assist in denervationof the carpus.In the extended longitudinal exposure of the wristduring total wrist arthrodesis, leave the EPL ten-don transposed dorsally after the procedure.When undertaking a total or limited wrist arthro-desis, use only good-quality autogenous cancel-

igure 6. A specialized, precontoured dorsal plate with au-ogenous bone graft is used for total wrist arthrodesis in mostatients, although fusion may be obtained with the use of aliding bone graft from the distal radius, multiple pin fixation,r smaller plates.

igure 7. During total wrist arthrodesis, a complete columnf bone graft should extend from the radius to the metacarpaleing packed into at any joint that underlies the plate. Alluch joints should have the hard subchondral cortical boneemoved, exposing cancellous bone before the placement of

phe autograft. (Reproduced with permission.11)

lous bone graft obtained from the distal radius,proximal ulna, or iliac crest. Carefully place thisgraft into the joints to be fused after all of thedegenerative cartilage and hard subchondral bonehas been fully removed to expose the native can-cellous carpus. Use of excised carpal bones as asource of graft is generally insufficient, yieldinglimited, hard, poor-quality material; this should beused only to augment good-quality graft place-ment.Use fluoroscopic examination during surgery toensure that hardware placement and osseous align-ment is appropriate before wound closure and thatany hardware used for fixation during reconstruc-tion does not impinge on the overall postoperativerange of motion.

otal Wrist Arthrodesisirst reported in the beginning of the 1900s byly,18,19 who was treating wrists diseased with tu-erculosis, total wrist arthrodesis has since becomehe standard surgical treatment for patients with se-ere pancarpal degenerative joint disease who stillish to undertake considerable heavy labor.11,16

enerally, good function occurs after surgery, al-hough patients report trouble with personal hygienend working in tight spaces.20,21 Grip is quite strongfter this procedure, provided that the arthrodesis islaced in an alignment that allows some extension ofhe wrist, generally in the range of 10° to 15°. These of autogenous bone graft with dorsal plate fixa-ion has been shown to produce predictably excellentusion rates from 93% to 100% (Fig. 6).11,16,22 Al-hough distal radius bone graft is usually sufficient,liac crest bone graft can also be used (Figs. 7, 8).pecialized precontoured dorsal plates have becomebiquitous and decrease the need for long-term post-perative immobilization (Fig. 9). During the initialostoperative period, the patient should be encour-ged to work aggressively on his/her digital mo-ion and pronation/supination of the wrist to opti-ize long-term functional recovery. Although the

iterature is rather consistent in showing an out-tanding fusion rate with current techniques, re-

igure 8. Careful packing of autogenous bone graft followedy plate fixation is performed. (Reproduced with permission.11)

orts of the ability of total wrist arthrodesis to

satc

TBcchtdhap

wLdrbavtTpio

Fsf

Ftc


uccessfully relieve pain are a bit more vari-ble.23–25 Not recognizing concomitant DRUJ ar-hritis or ulnocarpal impaction are potential compli-ations of this procedure.16,26,27

igure 9. Postoperative (A) PA (B) and lateral radiographshow a well-aligned total wrist arthrodesis using a dorsalusion plate.

igure 10. (A) A fourth-generation–design total wrist implanhritis. (B) Posteroanterior and (C) lateral radiographs show e
omponents.
otal Wrist Arthroplastyecause many of the activities of daily living (eg,ombing hair, fastening buttons, writing) are madeonsiderably easier with some wrist motion,12 TWAas become an increasingly attractive surgical al-ernative in lower-demand individuals who haveebilitating pancarpal osteoarthritis. Patients whoave had a total wrist arthrodesis on one side andTWA on the other prefer the motion-sparing

rocedure.28,29

Designed by Swanson, the first total wrist implantas made of silicone and provided good pain relief.ong-term follow-up data have shown a high inci-ence of implant failure,30–33 and issues regardingeactive synovitis and secondary osteolysis haveeen reported.34,35 Although a new generation ofrticulated wrists were designed to improve on pre-ious implants and provide an enhanced arc of mo-ion, they suffered from instability problems.12

hrough innovative research and by learning fromast mistakes, total wrist designs have since becomencreasingly sophisticated, with attempts made toptimize component fixation and their articulating

ed without cement into a 65-year-old patient with osteoar-t alignment with an interference fit of the radial and carpal

t placxcellen

gahmaattpaw1issv(

DIbjoutnloo(saro

Dapiodlmdsojotth

maoddetswc

Fdua

Fttflt


eometry.36–39 Although most data are understand-bly from patients with rheumatoid arthritis, TWAas been shown to predictably provide pain relief inost patients while maintaining approximately a 60°

rc of wrist motion split evenly between extensionnd flexion.12,36 Although many patients with post-raumatic osteoarthritis are young and active andherefore may shy away from TWA because of itsostoperative activity restrictions, which tradition-lly include avoidance of impact loading of therist and infrequent lifting of objects heavier than0 lb (4.5 kg), a lower demand patient may be andeal candidate for this procedure.12 Newer de-igns may also be implanted with or without re-ection of the distal ulnar head providing moreersatility for use in patients with osteoarthritisFig. 10).

istal Ulnan the ulnar neutral wrist, the ulnocarpal articulationears 20% of the load transferred across the wristoint.40 When ulnar variance increases, which mayccur in the setting of shortened distal radius mal-nions, Palmer and Werner40 showed that the loadransmission across the ulna increases and that ab-ormal mechanical loading of the ulnocarpal carti-ages ensues. Furthermore, when the relative positionf the distal radius or ulna is altered by fracturer injury to the triangular fibrocartilage complexTFCC), radioulnar or ulnocarpal ligaments, or otheroft tissues surrounding the distal ulna, the preciserticulation between the shallow sigmoid notch of theadius and the convex ulnar head is disrupted, and

igure 11. A PA radiograph shows ulnocarpal abutment syn-rome with secondary cystic formation in the lunate andlnar head at the site of impaction. This finding is nearlylways associated with a central TFCC tear.

steoarthritis may result. f

Arthritis involving the ulnocarpal joint and/orRUJ is often associated with rheumatoid arthritis

nd other inflammatory disorders41 because of theropensity of these conditions for destroying thentegrity of the ulnocarpal wrist ligaments.42 Devel-pmental abnormalities, such as Madelung’s growtheformity of the volar and ulnar distal radius,3,4 mayead to a marked loss of radial height and malalign-ent of the radiocarpal joint, with progressive DRUJ

isruption and ulnocarpal impaction. Traumatic in-ults to the distal radius, ulnar head, or the ligamentsf the ulnar wrist may also lead to altered ulnocarpaloint and DRUJ loading and serve as a precursor forsteoarthritis. A key to treating osteoarthritic condi-ions of the ulnar wrist is determining precisely whathe pain-generating articulation is, often with theelp of differential anesthetic or cortisone injections.If ulnocarpal arthrosis is thought to be the pri-ary pain generator, surgical treatment should be

imed at performing either a distal radius osteot-my to restore radial height (as in the setting of aistal radius shortened malunion) or shortening theistal ulna to effectively unload what is a diseased,burnated joint (Fig. 11). This constitutes the ra-ionale behind performing an ulnar diaphysealhortening osteotomy43,44 or a wafer resection,hich can be performed either open45 or arthroscopi-

ally (Fig. 12).46 A distal ulna wafer resection may

igure 12. (A) A arthroscopic wafer procedure is performedhrough a debrided TFCC tear, with care taken to fully rotatehe forearm during burring of the distal ulna to completelyatten out the entire distal ulnar prominence. (B) Alterna-ively, a formal ulnar shortening osteotomy can be per-
ormed. (Reproduced with permission.124)

npttawsrauhp

poWeupnbltbkwdtts(p

LLctbjtarAotlafvlb

Ftigr ssue e

Fhbv


ot be appropriate in patients with considerable ulnarositivity (ie, �3–4 mm).47 A complete excision ofhe ulnar head, or so-called Darrach procedure,48 is araditional surgery used to eliminate ulnocarpal jointnd/or DRUJ sources of pain that has been associatedith serious postoperative complications such as in-

tability of the distal ulna stump.49–53 Partly as aesponse to this problem of radioulnar convergencend impingement seen after resecting the distallna,50,54 endoprosthetic replacement of the ulnaread has attracted a great deal of interest and showsromise.55–58

If osteoarthritis of the DRUJ is thought to be aatient’s primary pain generator, a partial resectionf the distal ulna such as those described byatson59–60 or Bowers61 should be strongly consid-

red (Fig. 13). Although the Watson matched distallna resection and the Bowers hemiresection inter-osition arthroplasty differ with respect to whether orot interposing soft tissue in the area of the excisedone is advantageous, their basic concepts are simi-ar. To avoid the pitfalls of radioulnar convergencehat can occur after a traditional Darrach procedure,oth of these hemiresection techniques emphasizeeeping the ulnar styloid and attached TFCC intacthile resecting enough of the radial aspect of theistal ulna to avoid postoperative impingement. Al-ernatives to distal ulna hemiresection include fusinghe DRUJ and creating a more proximal pseudarthro-is through which pronation/supination may occurthe so-called Sauve-Kapandji procedure)62–64 or re-

igure 13. (A) Resection of the distal ulnar head (Darrach proo make the level of resection proximal to the sigmoid notchmpingement. (B) Hemiresection of the distal ulnar head, wiood ulnar soft-tissue-stabilizing structures. (C) Alternativelesecting a segment of ulnar proximal maintaining the soft-ti

lacing the ulnar head with an endoprosthesis.58 p

imited Wrist Arthrodesisimited wrist arthrodeses involve fusing a part of thearpus in a local fashion and maintaining motionhrough a portion of the wrist joint that is not affectedy the arthritis itself.65 After determining whichoints of the wrist are the arthritic pain generators,reatment can be undertaken to fuse the affectedrticulations while allowing motion to occur at theemaining wrist joints that are free from arthritis.lthough some motion is lost after any partial fusionf the wrist, by maintaining perfect congruity be-ween the remaining uninvolved joints, excellentong-term pain relief and wrist stability can bechieved.13,14 In most cases, local bone graft takenrom the distal radius is more than sufficient to pro-ide an adequate fusion.11 If there is any doubt thatocal bone graft is sufficient or of appropriate quality,one graft should be taken from the iliac crest to

e) may be performed for osteoarthritis. Care should be takenradius in order to minimize the possibility of post-operativeithout tendon interposition, is performed in a patient with

uve-Kapandji may be performed by fusing the DRUJ andnvelope. (Reproduced with permission.125)

igure 14. During a limited wrist arthrodesis, at least onealf to three quarters of the joint surfaces are denuded toleeding cancellous bone, with only a small portion of theolar joint kept to preserve carpal spacing. (Reproduced with

cedurof theth or wy, a Sa

ermission.126)

abaqwsaworapbser

tt

L

●

●

●

●

●

●

●

FFa ion an

Fd


ugment the fusion. Using a patient’s scaphoid forone graft during a scaphoid excision, four-cornerrthrodesis may be problematic because of the fre-uently sclerotic nature of the scaphoid in patientsith SLAC and SNAC wrists. This may explain

ome of the remarkably high nonunion rates in recentrticles reviewing four-corner arthrodeses performedith dorsal circular plates and predominantly scaph-id autograft66,67 compared with limited wrist arth-odeses using K-wires and distal radius cancellousutograft.68 Bone graft substitute materials, althougherhaps appropriate as an augment to autogenousone graft, are not at this time sufficient unto them-elves. Future bioengineering and research will likelyxpand the role of bone growth factors and syntheticeplacements during these procedures.

To maximize union rates and minimize complica-ions, limited wrist arthrodeses should try to adhereo the following technical principles.

igure 15. When undertaking limited wrist arthrodesis, compailure to do so will cause dorsal impingement during wrthrodesis, lack of impingement is confirmed with wrist flex

igure 16. K-wires are commonly used to provide fixation
uring intercarpal arthrodeses.
imited wrist arthrodesis: technical principles

Take down at least half of the joint surfaces beingfused while leaving a small portion of the volarjoint intact to preserve carpal spacing (Fig. 14).Denude osteoarthritic surfaces through the hardsubchondral bone to expose cancellous trabeculaeto provide more predictable fusion surfaces.Use high-quality autogenous cancellous graft, usu-ally from the distal radius, and pack it tightly intothe spaces exposed at the joints being fused.Use rigid fixation to allow earlier range-of-motionexercises and to decrease stiffness in the postop-erative period.Check hardware placement with fluoroscopy dur-ing surgery to ensure that the hardware does notpenetrate into the joints that are not being fusedand that the length of the hardware will not causesoft-tissue or bony impingement problems.Make certain that the carpal bones are in theirnormal positions before arthrodesis. This is specif-ically important with the lunate, which is com-monly tilted into a dorsal (DISI) or volar interca-lated segmental instability (VISI) pattern. Inchronic scapholunate ligament tears, for example,the lunate will often be in a DISI pattern and needsto be brought out of this extended position. Usinga 1.6-mm (0.062-in) K-wire as a joystick, the lu-nate should be placed in its normal alignmentbefore obtaining fixation (Fig. 15).During the postoperative period, take serial radio-graphs to ensure that a successful fusion has

duction of the lunate out of its abnormal position is required.tension. With appropriate reduction of the lunate befored extension. (Reproduced with permission.123)

lete rerist ex

occurred.

L

u

●

●

●

●

Fvswr

Faw

Fscbn

Fi


imited wrist arthrodesis: modes of fixation.Deciding which definitive method of fixation to

se is controversial and highly surgeon dependent.

K-wires have long been regarded as a mainstay inthe treatment of hand and wrist arthrodeses. Theyare inexpensive, easy to use, and provide reliablefixation (Fig. 16). Although there are several po-tential disadvantages to K-wire fixation, includingthe requirements for cast immobilization and even-tual pin removal, the risk of pin protrusion, sensorynerve irritation, and pin track infection, resultshave generally been quite good with this tech-nique.68–70

Standard staples either inserted with manual im-plantation or power implantation provide limitedcompression but provide good stability between

igure 17. Staples (both static and memory types) can pro-ide excellent fixation of the carpal bones being fused. Carehould be taken to avoid any possible dorsal impingementith wrist extension between the staple edge and the distal

adius.

igure 18. Headless screws can be used for limited wristrthrodesis and provide the advantage of no protruding hard-
are. (Reproduced with permission.127) s
bones.71 Because staples can protrude on the dor-sal aspect of the carpus, they may impinge duringwrist extension, and care must be taken to ensurethat this is minimized (Fig. 17). Newer staples thatprovide a memory compression force may providemore optimal outcomes.Screws can provide extremely stable compressionacross an arthrodesis site.72,73 Headless screwsthat are buried beneath the carpal bones offer min-imal impingement risk. Getting optimal placementof these types of screws can be problematic be-cause of orientation difficulties (Fig. 18).Dorsal circular plates have been designed to lie in arecessed fashion on the carpus to minimize impinge-

igure 19. Postoperative (A) PA and (B) lateral radiographshow excellent position of the carpal bones in this four-orner arthrodesis. Two screws affix each of the 4 boneseing fused, and the height of the 4 bones is restored to itsative position.

igure 20. The area of arthrodesis at the STT joint forms annverted “T”. C, capitate; Tm, trapezium; Td, trapezoid; S,
caphoid. (Reproduced with permission.128)

ioma

a

SSttj3prgettpmw(t

Fps(


ment during wrist extension while providing rigidfixation to minimize the duration of cast immobili-zation. Such plates theoretically promote circumfer-ential compression of the bones being fused whenscrew tightening occurs while allowing for the place-ment of autogenous bone graft through the center ofthe plate at the arthrodesis site (Fig. 19).74 The initialenthusiasm for dorsal circular plates during limitedwrist arthrodeses75 has been tempered by severalother investigators’ reports66,67,76,77 of high non-union and complication rates.

It is important to note that rigid hardware in and oftself will not achieve a successful arthrodesis with-ut appropriate technique and bone graft. There re-ains no consensus on the ideal method of fixation to

chieve a successful intercarpal arthrodesis.We will now discuss the common limited wrist

igure 21. (A) A preoperative PA radiograph shows moderatlacement into the STT interval. (C) The grafting technique. (Rhowing fixation with 3 K-wires. (E) Various methods of fixaReproduced with permission.122)

rthrodesis in turn. r

caphotrapezium-trapezoid arthrodesis. TheTT joint is a relatively common site of focal arthri-

is, especially in elderly women.5,78 A limited dorsalransverse incision is often used, because the entireoint can be visualized through a relatively small-cm incision followed by a transverse capsular ex-osure.5,78,79 After the STT joint is entered and theemaining cartilage denuded, local autogenous bonerafting from the distal radius is harvested. Watsont al68 and Wollstein and Watson79 argue that rela-ively more autograft is required for STT arthrodeseshan for SLAC wrist reconstructions, because it isarticularly important to maintain the external di-ensions of the STT joint (Fig. 20). Internal fixationith K-wires has high union rates and relief of pain

Fig. 21).68,79 Although arthrodesis is a standardreatment for STT osteoarthritis, its use in treating

arthrosis. (B) An intraoperative demonstration of bone graftuced with permission.122) (D) A postoperative PA radiographcluding K-wires, staples, and headless screws can be used.

e STTeprodtion in

otatory subluxation of the scaphoid has proved to be

mtpbtoa

Stiptaic

Rnrajreslt

smtt(

ScswwnpaiKcw

FaitgStclormpcrA

Fdopcw

Fd


uch more controversial.78,80–82 A radial styloidec-omy should always be performed along with thisrocedure to allow maximum motion after fusion haseen accomplished, with great care taken to not harmhe volar radiocarpal ligaments.83,84 One can expect,n average, approximately 65% of normal motionfter surgery.

caphocapitate arthrodesis. Scaphocapitate ar-hrodesis is generally used for ligament instabilitynvolving the scapholunate joint. For all intents andurposes, this procedure provides the same func-ional outcome as an STT arthrodesis by maintainingrelatively extended position of the scaphoid, allow-

ng reduction of the scapholunate interval. This pro-edure is rarely used for osteoarthritis.85

adioscapholunate arthrodesis. Radioscapholu-ate arthrodesis is commonly used after severe distaladius fractures that have resulted in degenerationlong the entire radiocarpal joint while the midcarpaloint remains well preserved.86,87 Fixation of theadius, scaphoid, and lunate is achieved with a vari-ty of fixation methods, including K-wires, poweredtaples, or a dorsal circular recessed plate. In general,ocal autogenous bone graft is sufficient, although

igure 22. (A) A postoperative PA radiograph shows a ra-ioscapholunate arthrodesis. (B) Excision of the distal scaph-id is now recommended, because this minor additionalrocedure unlocks the fused radiocarpal joint from the mid-arpal joint, improving eventual motion. (Part B reproducedith permission.130)

his may need to be augmented with synthetic f

ubstitutes. One can expect about 33% of normalotion to occur, but this can be increased substan-

ially to 50% to 60% of normal motion by excisinghe distal half of the scaphoid during the procedureFig. 22).15,88

capholunocapitate arthrodesis. Scapholuno-apitate arthrodesis can be used for patients withcapholunate instability and in those rare patientsith isolated midcarpal degenerative joint diseaseho have a good radiocarpal joint.89 The scapholu-ate joint needs to be reduced carefully and com-letely before arthrodesis. It is important to performradial styloidectomy with this procedure to prevent

mpingement during motion. Internal fixation with-wires or dorsal plates75 can be used (Fig. 23). One

an expect approximately 33% to 50% of normalrist motion after this procedure.

our-corner (capitate–lunate–hamate–triquetrum)rthrodesis. Scapholunate advanced collapse wrists the most common degenerative arthritis pattern inhe wrist, and it follows a predictable order of pro-ression after injury to the scapholunate ligament.6

caphoid nonunion advanced collapse wrist is essen-ially identical in its natural history, although lessommon. Degenerative changes begin at the articu-ation between the radial styloid and the distal scaph-id and progress predictably to the entirety of theadioscaphoid joint, followed by the development ofidcarpal arthritis. The radiolunate joint is typically

reserved. Degeneration of the radiolunate joint, as isharacteristic of advanced Kienböck’s disease, rep-esents a contraindication to four-corner arthrodesis.s developed and popularized by Watson,6,68 the

igure 23. A patient who had a scapholunocapitate arthro-esis 5 years earlier shows no secondary arthritic changes on
ollow-up examination radiographs (PA shown).

Sslca2s

dlltw(c

dPdspserpfipetc

Fdfl ows ap


LAC wrist reconstruction involves complete exci-ion of the scaphoid and arthrodesis of the capitate,unate, hamate, and triquetrum. In non-SLACases, in which there is isolated midcarpal osteo-rthritis and a pristine radioscaphoid joint (Fig.4), a four-corner arthrodesis in situ withoutcaphoid excision is a viable treatment.

To perform a SLAC wrist reconstruction, we use aorsal longitudinal skin incision followed by either aongitudinal or transverse (ligament-sparing) capsu-ar incision. During excision of the scaphoid, care isaken to preserve the long radiolunate ligament,hich prevents ulnar translation of the carpal mass

Fig. 25).90 Before setting the alignment of the four-

igure 24. Preoperative (A) PA and (B) lateral radiographs shemonstrates an in situ four-corner arthrodesis using a receuoroscopy of the wrist in PA (D) and lateral (E) projections sh

orner arthrodesis, one must be careful to fully re- t

uce the lunate out of its DISI position (Fig. 26).rovisional fixation with K-wires can be used, andefinitive fixation is obtained with either K-wires,taples, headless screws, or a recessed circular dorsallate. Scapholunate advanced collapse wrist recon-truction has been well studied and has producedxcellent results in the literature.7,65,68,91–93 Severalecent articles66,67,76,77 suggest that dorsal circularlates are inferior to more traditional methods ofxation. Vance et al,66 for example, reviewed 58atients who had four-corner arthrodesis fixed withither dorsal plates or traditional pin, staple, or screwechniques. They found a significantly higher rate ofomplications such as nonunion or impingement in

olated midcarpal arthrosis. (C) An intraoperative photographircular plate and distal radius bone graft. An intraoperativepropriate plate alignment without hardware impingement.

ow isssed c

he circular plate fixation group (48%) versus the

tpsgctsctdfpptpmtustWtctp

mtsfgcScoc

Se

PPpaepioci

Feposr

FbKprbKdda


raditional fixation group (6%). Proponents of dorsallate fixation note that the bone graft used in thistudy was not equally distributed between the 2 studyroups. Whereas morselized bone graft from the ex-ised scaphoid alone was used in 20 of the 27 pa-ients in the plate fixation group, presumably non-clerotic– quality cancellous distal radius or iliacrest autograft was used in 25 of the 31 patients inhe traditional fixation group. The report by Ken-all et al67 of 18 four-corner arthrodeses per-ormed by 4 hand surgeons with dorsal circularlate fixation showed a 63% nonunion rate. Oneossible explanation for this high nonunion rate ishat the plate may have obscured determination ofotential union. Another possible explanation is thatorselized scaphoid graft alone was again used as

he source of bone graft in 16 of the 18 wrists. It isnclear whether the high nonunion rates in thesetudies were due to the type of fixation used or to theype of bone graft that was predominantly harvested.

hat is clear from reviewing these articles is theechnical demands of the dorsal circular plate, in-luding the importance of recessing the plate belowhe level of the dorsal lunate cartilage to avoid im-ingement during wrist extension (Fig. 27).Based on the literature, one can expect approxi-ately 75% to 80% of normal grip strength and 40%

o 60% of normal motion after four-corner arthrode-is.68,91–95 Before reviewing the results obtainedrom PRC, it is important to note when a PRC isenerally contraindicated. When the cartilage of theapitate head is normal, as in stages I and II SLAC orNAC wrists, either scaphoid excision and four-orner arthrodesis or PRC represent viable treatmentptions (Fig. 28). When there is degeneration of the

igure 25. During a four-corner arthrodesis with scaphoidxcision, care must be taken when excising the scaphoid torevent injury to the long radiolunate ligament (arrow). If ansteotome is used to fragment the scaphoid for excision, ithould be oriented in line with the fiber direction of the longadiolunate ligament.

apitate head, however, as in stage III SLAC or r

NAC wrists, a PRC is generally contraindicatedxcept in very low-demand patients (Fig. 29).

roximal Row Carpectomyroximal row carpectomy is an excellent salvagerocedure for the wrist with considerable radiocarpalrthritis provided that the capitate has not had degen-rative changes.96,97 Proximal row carpectomy iserformed through either a transverse or longitudinalncision. The bones of the proximal row are removedn either side of the tendons of the fourth dorsalompartment (Fig. 30). During excision of the prox-mal row, one must be extremely careful to preserve

igure 26. (A) An intraoperative photograph of the lunateeing brought out of its malaligned DISI position by using a-wire as a joystick and then pinning the lunate in its reducedosition through the dorsal aspect of the radius. (B) A fluo-oscopic image is used to confirm the position of the lunateefore proceeding with the surgery. (C) The initial joystick-wire is then removed, allowing unfettered access to theorsal carpus while maintaining the lunate position via theistal radius K-wire, which is in the volar half of the lunatellowing free dorsal access to the joints being fused. (Part C
eproduced with permission.126)

twtnfotio

qt

lAtpPBwcaotenidat3f

Ffi

Fpct

Fpc


he integrity of the radioscaphocapitate ligament,hich prevents ulnar translation of the capitate off

he distal radius.90,98 This ligament is especially vul-erable when removing the scaphoid in a piecemealashion with an osteotome and mallet (Fig. 31). Thesteotome should be aligned with the orientation ofhe fibers of the radioscaphocapitate ligament so that,n the case of inadvertently deep penetration, thesteotome will split rather than transect the fibers.After removing the scaphoid, lunate, and tri-

uetrum, the distal carpal row migrates proximally sohat the head of the capitate articulates with the

igure 27. An intraoperative photograph of a dorsal plateully recessed below the dorsal lunate cartilage so as to avoidmpingement during wrist extension.

igure 28. (A) Posteroanterior and (B) lateral radiographs of aatient with stage II SLAC wrist in which the capitate headartilage is pristine and a PRC should be considered as a
reatment option.
unate fossa of the distal radius (Figs. 32, 33).99,100

lthough the arcs of curvature of the capitate andhe lunate are not entirely congruent and someoint loading occurs during motion, the results ofRC have generally been excellent.92,94,95,97,100–104

ased on 2 separate cohorts of patients,102,103 eachith a minimum follow-up period of 10 years, one

an conclude that patients having a PRC willchieve over 80% of normal grip strength and 60%f normal motion. Although secondary degenera-ive changes at the radiocapitate articulation mayventually be seen on plain radiographs, they doot usually preclude a successful clinical outcomen most patients, especially in functionally lower-emand patients.103 Younger, active patients are atgreater risk for developing secondary degenera-

ive changes at the proximal capitate head (Fig.4). DiDonna et al103 found that there were 4ailures in 22 wrists studied that required arthro-

igure 29. (A) Posteroanterior and (B) lateral radiographs of aatient with stage III SNAC wrist in which the capitate headartilage appears compromised and a PRC is contraindicated.

Figure 30. In stepwise fashion, the scaphoid, lunate, andtriquetrum are exposed and removed via either the radial orulnar side of the fourth dorsal compartment. ECU, extensorcarpi ulnaris; EDC, extensor digitorum communis; EDQP,extensor digiti quinti proprius; EIP, extensor indicis proprius.
( Reproduced with permission.122)

dP

tasfhpas

PETiw

S3cfteliecmoac

sf1saps

FPcp

Flsdtr

Fa

FnN


esis at an average time of 7 years after the initialRC.103

Comparing patients with post-traumatic wrist arthri-is who had a PRC with those who had a total wristrthrodesis, De Smet et al105 found no difference in griptrength but less disability in the PRC group. A morerequent comparison in the wrist osteoarthritis literatureas been between the 2 most common motion-sparingrocedures used today: PRC versus scaphoid excisionnd four-corner arthrodesis. Thus far, only relativelyhort-term outcome data are known.

roximal Row Carpectomy Versus Scaphoidxcision and Four-Corner Arthrodesishere are no prospective, randomized trials compar-

ng scaphoid excision and four-corner arthrodesisith PRC as alternative motion-sparing treatments of

igure 31. While removing the scaphoid when during aRC, care must be taken not to injure the radioscapho-apitate ligament in the volar depths. (Reproduced withermission.131)

igure 32. After PRC, reduction of the capitate head into theunate fossa of the distal radius is accomplished. A radialtyloidectomy may also be performed, usually at level “A”epicted in the figure. Care must again be taken to preserve

he radioscaphocapitate ligament attachment at the distal
adius. (Reproduced with permission.122) c
LAC or SNAC wrist arthritis. In fact, there are onlystudies92–94 to our knowledge that compare the

linical outcomes of these surgeries in a head-to-headashion. Before reviewing these articles, it is impor-ant to first understand the theoretic advantages ofach procedure. Advocates of PRC cite its relativelyower technical demands, decreased postoperativemmobilization, and—perhaps most of all—its inher-nt lack of nonunion risk.95,104 Proponents of four-orner arthrodesis cite its theoretic advantage ofaintaining wrist height; a more physiologic range

f motion through a preserved, congruent radiolunaterticulation; and lack of eventual degenerativehanges in the radiolunate joint.

Three studies attempt to compare the clinical re-ults of PRC with those of scaphoid excision andour-corner arthrodesis. The first, by Tomaino et al in994,92 retrospectively reviewed 24 wrists treated forymptomatic SLAC wrist at an average of 5.5 yearsfter surgery. At the time of follow-up evaluation, allatients in the study had satisfactory pain relief, griptrength, and function except for 3 patients in the

igure 33. A postoperative PA radiograph shows excellentlignment of the capitate head in the lunate fossa after PRC.

igure 34. A PA radiograph at the 3-year follow-up exami-ation of an active 35-year-old patient who had had a PRC.ote the considerable secondary degenerative changes at the

apitate head.

PsadtmpotddSfetd

swamgatalawoMPowsamc

tCtfeafmtKbaf(

trtsgapwpbpfetsa

ssPcws

PPfsfqpteTvpmtcrfawtrPtosfp


RC group. One of these patients developed progres-ive radiocapitate arthritis necessitating a total wristrthrodesis. There were no statistically significantifferences in subjective and objective results be-ween the 2 groups other than residual range ofotion: PRC preserved a greater arc of motion com-

ared with limited wrist arthrodesis, mostly becausef significantly greater wrist extension. In light ofhese comparable outcomes between the 2 proce-ures, Tomaino et al92 concluded that either proce-ure is a justifiable choice for reconstructing theLAC wrist. While they favor scaphoid excision andour-corner arthrodesis when capitolunate arthritisxists (stage III SLAC), they recommend the lessechnically demanding PRC for stage I or II SLACisease.One year later, Wyrick et al94 published a similar

tudy comparing 11 wrists that had a PRC and 17rists that had a scaphoid excision and four-corner

rthrodesis. The PRC group was followed up at aean of 37 months and these wrists had an average

rip strength of 94% of the opposite wrist and a totalrc of motion of 115°, which was 64% of the con-ralateral wrist motion. These data compare favor-bly with the limited wrist arthrodesis group, fol-owed up at a mean of 27 months, in which theverage grip strength was 74% of the oppositerist and the total arc of motion was 95°, or 47%f the opposite wrist. Using the grading criteria ofinami et al106 for intercarpal arthrodeses, the

RC group was found to have 11 of 11 successfulutcomes versus the four-corner arthrodesis group,hich had 5 of 17 failures. Although their data

howed more polarity than those of Tomaino etl92 a year earlier, Wyrick et al94 similarly recom-end PRC as their motion-sparing procedure of

hoice in wrists without capitolunate degeneration.The most recent article comparing the outcomes of

hese 2 reconstructive procedures was written byohen and Kozin in 2001.93 Unlike its predecessors,

his study examined 2 cohorts of patients who wererom separate institutions that performed exclusivelyither PRC or scaphoid excision and four-cornerrthrodesis. The length of follow-up study was againairly short: 19 months for the PRC group and 28onths for the four-corner arthrodesis group. As was

he case in the study of Tomaino et al,92 Cohen andozin93 found no statistically significant differencesetween the groups with respect to flexion–extensionrc of motion (81° in the PRC group, 80° in theour-corner arthrodesis group) and grip strength
71% of the opposite wrist in the PRC group, 79% in s
he four-corner arthrodesis group). Furthermore, painelief and patient satisfaction were similar betweenhe 2 groups based on a plethora of outcome mea-ures. Their data, however, showed a significantlyreater ulnar–radial deviation arc in the four-cornerrthrodesis group (70% of the opposite wrist) com-ared with the PRC group (51% of the oppositerist). In all, the study of Cohen and Kozin93 sup-orts the notion that there are minimal differencesetween the short-term clinical outcomes of these 2rocedures. It would certainly be interesting in theuture to determine the incidence of eventual degen-rative changes for each of these procedures to de-ermine if any factors influence the incidence ofecondary arthrosis (eg, age or gender of the patientt the time of the initial procedure).

After our discussion of the workhorses of theurgical treatment of wrist osteoarthritis—namely,caphoid excision, four-corner arthrodesis, andRC—it is important to recognize 2 other types ofarpectomies that may play a role in treating patientsith wrist osteoarthritis: pisiform excision and distal

caphoid excision.

isiformectomyisiform excision is the standard surgical treatmentor pisotriquetral osteoarthritis. Embedded within theubstance of the flexor carpi ulnaris tendon, the pisi-orm is a sesamoid bone that lies volar to the tri-uetrum. Pisotriquetral arthrosis is most commonlyost-traumatic in nature and has been delineatedhrough cadaveric dissections showing cartilageburnation on the periphery of the pisiform.107,108

enderness due to pisotriquetral arthritis can be pro-oked clinically by passive wrist extension or by theisiform tracking test, during which the pisiform isade to grind radially and ulnarly up against the

riquetrum while the wrist is flexed to relax the flexorarpi ulnaris tendon.109 Multiple semilateral plainadiographs (with the wrist in full extension and theorearm in 30° of supination, with the wrist in neutralnd the forearm in 30° of supination, and with therist fully flexed and the forearm in 45° of supina-

ion while the thumb is maximally abducted) areecommended to visualize the pisotriquetral joint.109

erhaps the best tool in helping the clinician reachhe correct diagnosis is pain relief with the injectionf a local anesthetic into the pisotriquetral joint. Witheveral studies110–115 supporting its efficacy, pisi-orm excision has become the treatment of choice foratients with pisotriquetral arthritis refractory to con-
ervative measures. In their evaluation of 20 wrists

t1ssbwsenfea

DCccsorssr

itaeqDawdcfatdcc

ttckuatwtisbtetoac

SOtietat

aotticasivfa

F(


hat had pisiformectomy, Lam et al114 reported that5 patients had complete relief of symptoms and 5till had residual mild discomfort. Furthermore, theyhowed that there were no significant differencesetween the strengths and ranges of motion of therists that had the procedure and the contralateral

ide. In describing his surgical technique, Rayan109

mphasizes identification and protection of the ulnarerve, careful subperiosteal dissection of the pisi-orm, and meticulous repair of the soft-tissue conflu-nce and flexor carpi ulnaris tendon overlying the sitefter the bone is excised.

istal Scaphoidectomyomplete excision of the scaphoid has been dis-ussed previously in the context of SLAC wrist re-onstruction. Excision of only the distal pole of thecaphoid was also briefly mentioned, in the contextf improving motion after a radioscapholunate arth-odesis. Indeed, distal scaphoidectomy has beenhown in both a cadaveric model15 and a clinicaltudy88 to increase flexion and radial deviation afteradioscapholunate arthrodeses.

Another possible role for distal scaphoid excisions in the treatment of isolated STT joint osteoarthri-is.116 Although arthrodesis remains the most widelyccepted treatment for STT arthritis, distal scaphoidxcision may represent an interesting alternative re-uiring less postoperative immobilization (Fig. 35).istal scaphoidectomy has also been proposed as an

lternative to more extensive procedures in patientsho have chronic scaphoid nonunions and resultantistal radioscaphoid or intercarpal osteoarthritis. Inases of symptomatic scaphoid nonunions that haveailed traditional fixation and bone grafting attempts,relatively simple technical procedure to eliminate

his focal area of osteoarthritis is an excision of theistal scaphoid pole. Although there are certainlyoncerns about promoting instability of the proximalarpal row, there are a few studies117–119 that suggest

igure 35. (A) Intraoperative photograph showing conside
B) Resection of the distal half of the scaphoid is performed (C) with
hat distal scaphoid excision may be efficacious inhe treatment of recalcitrant scaphoid nonunion withoncomitant arthritis. In 2006, Ruch and Papadoni-olakis119 reported on 13 patients with scaphoid non-nions who were treated with distal scaphoidectomyfter prior surgical treatment at a mean follow-upime of 5 years. They showed an increase of meanrist flexion and extension by 23° and 29°, respec-

ively; a dramatic decrease in pain; and an averagencrease in the capitolunate angle before and afterurgery of only 2°. This is in contrast to prior worky Garcia-Elias et al116 that showed an increase inhe capitolunate angle of 9° after distal scaphoidxcision. Ruch and Papadonikolakis119 concludedhat distal scaphoid excision is a valuable treatmentption for certain patients in whom multiple attemptst union have failed and who have no associatedomplete scapholunate ligament tears.

ummarysteoarthritis of the wrist is a commonly encoun-

ered condition and has a variety of causes, fromdiopathic to traumatic. With a meticulous clinicalxamination and analysis of radiographs, careful at-ention paid to the patient’s age and functional needs,nd adherence to basic surgical principles, successfulreatment of the osteoarthritic wrist can be expected.

Despite the successful outcomes that can bechieved in treating patients with wrist osteoarthritis,ne should be careful to counsel patients that any ofhe surgical treatment options presented have posi-ives and negatives, not all of which are evident in themmediate perioperative recovery period. Some pro-edures are fairly simple and have limited morbidityssociated with their use, such as PRC and distalcaphoidectomy; however, their long-term durabilitys not always ensured. Other more technically in-olved procedures, such as scaphoid excision andour-corner arthrodesis, may have a higher morbiditynd complexity initially, as well as a longer recovery

degenerative changes at the distal pole of the scaphoid.
rable the resection level confirmed by intraoperative fluoroscopy.

pypomp

RF

aa

SM

R


eriod, but may provide better durability in aounger patient. Longer term follow-up data com-aring the results of alternative surgeries for wriststeoarthritis will surely help the hand surgery com-unity to most effectively and accurately educate

atients on these matters in the future.

eceived for publication January 16, 2007; accepted in revised formebruary 7, 2007.No benefits in any form have been received or will be received fromcommercial party related directly or indirectly to the subject of this

rticle.Corresponding author: Craig Rodner, MD, Department of Orthopaedic

urgery, University of Connecticut Health Center, 263 Farmington Ave,ARB 4013, Farmington, CT 06034-4037; e-mail: [email protected] © 2007 by the American Society for Surgery of the Hand0363-5023/07/32A05-0020$32.00/0doi:10.1016/j.jhsa.2007.02.003

eferences1. Luo J, Diao E. Kienbock’s disease: an approach to treat-

ment. Hand Clin 2006;22:465–473.2. Lauder AJ, Trumble TE. Idiopathic avascular necrosis of

the scaphoid: Preiser’s disease. Hand Clin 2006;22:475–484.

3. Arora AS, Chung KC, Otto W. Madelung and the recog-nition of Madelung’s deformity. J Hand Surg 2006;31A:177–182.

4. Carter PR, Ezaki M. Madelung’s deformity. Surgical cor-rection through the anterior approach. Hand Clin 2000;16:713–721.

5. Watson HK, Hempton RF. Limited wrist arthrodeses. I.The triscaphoid joint. J Hand Surg 1980;5:320–327.

6. Watson HK, Ballet FL. The SLAC wrist: scapholunateadvanced collapse pattern of degenerative arthritis. J HandSurg 1984;9A:358–365.

7. Watson HK, Goodman ML, Johnson TR. Limited wristarthrodesis. Part II: intercarpal and radiocarpal combina-tions. J Hand Surg 1981;6:223–233.

8. Abbott LC, Saunders JBDM, Bost FC. Arthrodesis of thewrist with the use of grafts of cancellous bone. J Bone JointSurg 1942;24:883–898.

9. Clayton ML, Ferlic DC. Arthrodesis of the arthritic wrist.Clin Orthop 1984;187:89–93.

10. Haddad RJ Jr, Riordan DC. Arthrodesis of the wrist. Asurgical technique. J Bone Joint Surg 1967;49A:950–954.

11. Weiss APC, Hastings H II. Wrist arthrodesis for traumaticconditions: a study of plate and local bone graft application.J Hand Surg 1995;20A:50–56.

12. Anderson MC, Adams BD. Total wrist arthroplasty. HandClin 2005;21:621–630.

13. Douglas DP, Peimer CA, Koniuch MP. Motion of thewrist after simulated limited intercarpal arthrodeses. Anexperimental study. J Bone Joint Surg 1987;69A:1413–1418.

14. Palmer AK, Werner FW, Murphy D, Glisson R. Functionalwrist motion: a biomechanical study. J Hand Surg 1985;10A:39–46.

15. McCombe D, Ireland DC, McNab I. Distal scaphoid exci-sion after radioscaphoid arthrodesis. J Hand Surg 2001;
26A:877–882.
16. Hastings H II, Weiss APC, Quenzer D, Wiedeman GP,Hanington KR, Strickland JW. Arthrodesis of the wrist forpost-traumatic disorders. J Bone Joint Surg 1996;78A:897–902.

17. Berger RA, Bishop AT, Bettinger PC. New dorsal capsu-lotomy for the surgical exposure of the wrist. Ann PlastSurg 1995;35:54–59.

18. Ely LW. A study of joint tuberculosis. Surg Gyencol Obstet1910;10:561–572.

19. Ely LW. An operation for tuberculosis of the wrist. JAMA1920;75:1707–1709.

20. Weiss APC, Wiedeman G Jr, Quenzer D, Hanington KR,Hastings H II, Strickland JW. Upper extremity functionafter wrist arthrodesis. J Hand Surg 1995;20A:813–817.

21. Field J, Herbert TJ, Prosser R. Total wrist fusion. A func-tional assessment. J Hand Surg 1996;21B:429–433.

22. Wright CS, McMurtry RY. AO Arthrodesis in the hand.J Hand Surg 1983;8:932–935.

23. O’Bierne J, Boyer MI, Axelrod TS. Wrist arthrodesis usinga dynamic compression plate. J Bone Joint Surg 1995;77B:700–704.

24. Sagerman SD, Palmer AK. Wrist arthrodesis using a dy-namic compression plate. J Hand Surg 1996;21B:437–441.

25. De Smet L, Truyen J. Arthrodesis of the wrist forosteoarthritis: outcome with a minimum follow-up of 4years. J Hand Surg 2003;28A:575–577.

26. Zachary SV, Stern PJ. Complications following AO/ASIFwrist arthrodesis. J Hand Surg 1995;20A:339–344.

27. Jebson PJ, Adams BD. Wrist arthrodesis: review of currenttechnique. J Am Acad Orthop Surg 2001;9:53–60.

28. Vicar AJ, Burton RI. Surgical management of the rheuma-toid wrist—fusion or arthroplasty. J Hand Surg 1986;11A:790–797.

29. Meuli HC. Total wrist arthroplasty. Experience with a nonce-mented wrist prosthesis. Clin Orthop 1997;342:77–83.

30. Fatti JF, Palmer AK, Mosher JF. The long-term results ofSwanson silicone rubber interpositional wrist arthroplasty.J Hand Surg 1986;11A:166–175.

31. Fatti JF, Palmer AK, Greenky S, Mosher JF. Long-termresults of Swanson interpositional wrist arthroplasty: partII. J Hand Surg 1991;16A:432–437.

32. Jolly SL, Ferlic DC, Clayton ML, Dennis DA, Stringer EA.Swanson silicone arthroplasty of the wrist in rheumatoid arthritis:a long-term follow-up. J Hand Surg 1992;17A:142–149.

33. Stanley JK, Tolat AR. Long-term results of Swanson Si-lastic arthroplasty in the rheumatoid wrist. J Hand Surg1993;18B:381–388.

34. Peimer CA, Medige J, Eckert BS, Wright JR, Howard CS.Reactive synovitis after silicone arthroplasty. J Hand Surg1986;11A:624–638.

35. Kistler U, Weiss AP, Simmen BR, Herren DB. Long-termresults of silicone wrist arthroplasty in patients with rheu-matoid arthritis. J Hand Surg 2005;30A:1282–1287.

36. Menon J. Universal Total Wrist Implant: experience with acarpal component fixed with three screws. J Arthroplasty1998;13:515–523.

37. Divelbiss BJ, Sollerman C, Adams BD. Early results of theUniversal total wrist arthroplasty in rheumatoid arthritis.J Hand Surg 2002;27A:195–204.

38. Rizzo M, Beckenbaugh RD. Results of biaxial total wrist


arthroplasty with a modified long metacarpal stem. J HandSurg 2003;28A:577–584.

39. Grosland NM, Rogge RD, Adams BD. Influence of artic-ular geometry on prosthetic wrist stability. Clin Orthop2004;421:134–142.

40. Palmer AK, Werner FW. Biomechanics of the distal radio-ulnar joint. Clin Orthop 1984;187:26–35.

41. Blank JE, Cassidy C. The distal radioulnar joint in rheu-matoid arthritis. Hand Clin 1996;12:499–513.

42. Glowacki KA. Hemiresection arthroplasty of the distalradioulnar joint. Hand Clin 2005;21:591–601.

43. Friedman SL, Palmer AK. The ulnar impaction syndrome.Hand Clin 1991;7:295–310.

44. Chun S, Palmer AK. The ulnar impaction syndrome: fol-low-up of ulnar shortening osteotomy. J Hand Surg 1993;18A:46–53.

45. Feldon P, Terrono AL, Belsky MR. Wafer distal ulnaresection for triangular fibrocartilage tears and/or ulna im-paction syndrome. J Hand Surg 1992;17A:731–737.

46. Wnorowski DC, Palmer AK, Werner FW, Fortino MD.Anatomic and biomechanical analysis of the arthroscopicwafer procedure. Arthroscopy 1992;8:204–212.

47. Markolf KL, Tejwani SG, Benhaim P. Effects of waferresection and hemiresection from the distal ulna on load-sharing at the wrist: a cadaveric study. J Hand Surg 2005;30A:351–358.

48. Darrach W. Partial excision of the lower shaft of the ulna.Ann Surg 1912;56:802–803.

49. af Ekenstam F, Engkvist O, Wadin K. Results from resec-tion of the distal end of the ulna after fractures of the lowerend of the radius. Scand J Plast Reconstr Surg 1982;16:177–181.

50. Bell MJ, Hill RJ, McMurtry RY. Ulnar impingement syn-drome. J Bone Joint Surg 1985;67B:126–129.

51. Bieber EJ, Linscheid RL, Dobyns JH, Beckenbaugh RD.Failed distal ulna resections. J Hand Surg 1988;13A:193–200.

52. Stuart PR, Berger RA, Linscheid RL, An KN. The dorso-palmar stability of the distal radioulnar joint. J Hand Surg2000;25A:689–699.

53. Minami A, Iwasaki N, Ishikawa J, Suenaga N, Yasuda K,Kato H. Treatments of osteoarthritis of the distal radioulnarjoint: long-term results of three procedures. Hand Surg2005:10:243–248.

54. McKee MD, Richards RR. Dynamic radio-ulnar conver-gence after the Darrach procedure. J Bone Joint Surg 1996;78B:413–418.

55. Scheker LR, Babb BA, Killion PE. Distal ulna prostheticreplacement. Orthop Clin North Am 2001;32:365–376.

56. Sauerbier M, Hahn ME, Fujita M, Neale PG, Berglund LJ,Berger RA. Analysis of dynamic distal radioulnar conver-gence after ulnar head resection and endoprosthesis implan-tation. J Hand Surg 2002;27A:425–434.

57. Gordon KD, Dunning CE, Johnson JA, King GJ. Kinematicsof ulnar head arthroplasty. J Hand Surg 2003;28B:551–558.

58. Berger RA, Cooney WP III. Use of an ulnar head endo-prosthesis for treatment of an unstable distal ulnarresection: review of mechanics, indications, and surgicaltechnique. Hand Clin 2005;21:603–620.

59. Watson HK, Ryu JY, Burgess RC. Matched distal ulna
resection. J Hand Surg 1986;11A:812–817.
60. Watson HK, Gabuzda GM. Matched distal ulna resectionfor posttraumatic disorders of the distal radioulnar joint.J Hand Surg 1992;17A:724–730.

61. Bowers WH. Distal radioulnar joint arthroplasty: thehemiresection-interposition technique. J Hand Surg 1985;10A:169–178.

62. Kapandji IA. The Kapandji-Sauve operation. Its techniquesand indications in non rheumatoid diseases. Ann Chir Main1986;5:181–193.

63. Lamey DM, Fernandez DL. Results of the modified Sauve-Kapandji procedure in the treatment of chronic posttrau-matic derangement of the distal radioulnar joint. J BoneJoint Surg 1998;80A:1758–1769.

64. DeSmet LA, Van Ransbeeck H. The Sauve-Kapandji pro-cedure for posttraumatic wrist disorders: further experi-ence. Acta Orthop Belg 2000;66:251–254.

65. Siegel JM, Ruby LK. A critical look at intercarpalarthrodesis: review of the literature. J Hand Surg 1996;21A:717–723.

66. Vance MC, Hernandez JD, Didonna ML, Stern PJ. Com-plications and outcome of four-corner arthrodesis: circularplate fixation versus traditional techniques. J Hand Surg2005;30A:1122–1127.

67. Kendall CB, Brown TR, Millon SJ, Rudisill LE Jr, SandersJL, Tanner SL. Results of four-corner arthrodesis usingdorsal circular plate fixation. J Hand Surg 2005;30A:903–907.

68. Watson HK, Weinzweig J, Guidera PM, Zeppieri J, Ash-mead D. One thousand intercarpal arthrodeses. J Hand Surg1999;24A:307–315.

69. McAuliffe JA, Dell PC, Jaffe R. Complications of intercar-pal arthrodesis. J Hand Surg 1993;18A:1121–1128.

70. Larsen CF, Jacoby RA, McCabe SJ. Nonunion rates oflimited carpal arthrodesis: a meta-analysis of the literature.J Hand Surg 1997;22A:66–73.

71. Benkeddache Y, Gottesman H, Fourrier P. Multiple sta-pling for wrist arthrodesis in the nonrheumatoid patient.J Hand Surg 1984;9A:256–260.

72. Nelson DL, Manske PR, Pruitt DL, Gilula LA, Martin RA.Lunotriquetral arthrodesis. J Hand Surg 1993;18A:1113–1120.

73. Sennwald GR, Ufenast H. Scaphocapitate arthrodesis forthe treatment of Kienbock’s disease. J Hand Surg 1995;20A:506–510.

74. Enna M, Hoepfner P, Weiss AP. Scaphoid excision withfour-corner fusion. Hand Clin 2005;21:531–538.

75. Walsh EF, Manuel JLM, Weiss APC. Scaphoid-lunate-capitate fusion for scapholunate ligament injuries. AtlasHand Clin 2003;8:279–285.

76. Shindle MK, Burton KJ, Weiland AJ, Domb BG, WolfeSW. Complications of circular plate fixation for four-cornerarthrodesis. J Hand Surg 2007;32B:50–53.

77. Chung KC, Watt AJ, Kotsis SV. A prospective outcomesstudy of four-corner wrist arthrodesis using a circular lim-ited wrist fusion plate for stage II scapholunate advancedcollapse wrist deformity. Plast Reconstr Surg 2006;118:433–442.

78. Kleinman WB. Long-term study of chronic scapho-lunateinstability treated by scapho-trapezio-trapezoid arthrodesis.
J Hand Surg 1989;14A:429–445.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1


79. Wollstein R, Watson HK. Scaphotrapeziotrapezoid arthro-desis for arthritis. Hand Clin 2005;21:539–543.

80. Frykman EB, Af Ekenstam F, Wadin K. Triscaphoid arth-rodesis and its complications. J Hand Surg 1988;13A:844–849.

81. Kleinman WB, Carroll CT IV. Scapho-trapezio-trapezoidarthrodesis for treatment of chronic static and dynamicscapho-lunate instability: a 10-year perspective on pitfallsand complications. J Hand Surg 1990;15A:408–414.

82. Ishida O, Tsai TM. Complications and results of scapho-trapezio-trapezoid arthrodesis. Clin Orthop 1993;287:125–130.

83. Watson HK, Ryu J. Evolution of arthritis of the wrist. ClinOrthop 1986;202:57–67.

84. Siegel DB, Gelberman RH. Radial styloidectomy: an ana-tomical study with special reference to radiocarpal intra-capsular ligamentous morphology. J Hand Surg 1991;16A:40–44.

85. Moy JO, Peimer CA. Scaphocapitate fusion in the treat-ment of Kienböck’s disease. Hand Clin 1993;9:501–504.

86. Inoue G, Tamura Y. Radiolunate and radioscapholunatearthrodesis. Arch Orthop Trauma Surg 1992;111:333–335.

87. Nagy L, Büchler U. Long term results of radioscapholunatefusion following fractures of the distal radius. J Hand Surg1997;22B:705–710.

88. Garcia-Elias M, Lluch A, Ferreres A, Papini-Zorli I, Ra-himtoola ZO. Treatment of radiocarpal degenerative osteo-arthritis by radioscapholunate arthrodesis and distal scaph-oidectomy. J Hand Surg 2005;30A:8–15.

89. Rotman MB, Manske PR, Pruitt DL, Szerzinski J. Scapho-capitolunate arthrodesis. J Hand Surg 1993;18A:26–33.

90. Berger RA, Landsmeer JM. The palmar radiocarpalligaments: a study of adult and fetal human wrist joints.J Hand Surg 1990;15A:847–854.

91. Ashmead D IV, Watson HK, Damon C, Herber S, Paly W.Scapholunate advanced collapse wrist salvage. J Hand Surg1994;19A:741–750.

92. Tomaino MM, Miller RJ, Cole I, Burton RI. Scapholunateadvanced collapse wrist: proximal row carpectomy or lim-ited wrist arthrodesis with scaphoid excision? J Hand Surg1994;19A:134–142.

93. Cohen MS, Kozin SH. Degenerative arthritis of thewrist: proximal row carpectomy versus scaphoid exci-sion and four-corner arthrodesis. J Hand Surg 2001;26A:94 –104.

94. Wyrick JD, Stern PJ, Kiefhaber TR. Motion-preservingprocedures in the treatment of scapholunate advanced col-lapse wrist: proximal row carpectomy versus four-cornerarthrodesis. J Hand Surg 1995;20A:965–970.

95. Krakauer JD, Bishop AT, Cooney WP. Surgical treatmentof scapholunate advanced collapse. J Hand Surg 1994;19A:751–759.

96. Bach AW, Almquist EE, Newman DM. Proximal rowfusion as a solution for radiocarpal arthritis. J Hand Surg1991;16A:424–431.

97. Begley BW, Engber WD. Proximal row carpectomy inadvanced Kienböck’s disease. J Hand Surg 1994;19A:1016–1018.

98. Stern PJ, Agabegi SS, Kiefhaber TR, Didonna ML. Prox-

imal row carpectomy. J Bone Joint Surg 2005;87A(suppl 1pt 2):166–174.

99. Inglis AE, Jones EC. Proximal row carpectomy for diseasesof the proximal row. J Hand Surg 1977;59A:460–463.

00. Imbriglia JE, Broudy AS, Hagberg WC, McKernan D.Proximal row carpectomy: clinical evaluation. J Hand Surg1990;15A:426–430.

01. Tomaino MM, Delsignore J, Burton RI. Long-term resultsfollowing proximal row carpectomy. J Hand Surg 1994;19A:694–703.

02. Jebson PJ, Hayes EP, Engber WD. Proximal rowcarpectomy: a minimum 10-year follow-up study. J HandSurg 2003;28A:561–569.

03. DiDonna ML, Kiefhaber TR, Stern PJ. Proximal rowcarpectomy: study with a minimum of ten years of follow-up. J Bone Joint Surg 2004:86A:2359–2365.

04. Wyrick JD. Proximal row carpectomy and intercarpal ar-throdesis for the management of wrist arthritis. J Am AcadOrthop Surg 2003;11:277–281.

05. De Smet L, Degreef I, Truyen J, Robijns F. Outcome of twosalvage procedures for posttraumatic osteoarthritis of thewrist: arthrodesis or proximal row carpectomy. Acta ChirBelg 2005;105:626–630.

06. Minami A, Agino T, Minami M. Limited wrist fusions.J Hand Surg 1988;13A:660.

07. Yamaguchi S, Viegas SF, Patterson RM. Anatomic studyof the pisotriquetral joint: ligament anatomy and cartilagi-nous change. J Hand Surg 1998;23A:600–606.

08. Rayan GM, Jameson BH, Chung KW. The pisotriquetraljoint: anatomic, biomechanical, and radiographic analysis.J Hand Surg 2005;30A:596–602.

09. Rayan GM. Pisiform ligament complex syndrome andpisotriquetral arthrosis. Hand Clin 2005;21:507–517.

10. Carroll RE, Coyle MP Jr. Dysfunction of the pisotriquetraljoint: treatment by excision of the pisiform. J Hand Surg1985;10A:703–707.

11. Johnston GH, Tonkin MA. Excision of pisiform in pisotri-quetral arthritis. Clin Orthop 1986;210:137–142.

12. Belliappa PP, Burke FD. Excision of the pisiform in piso-triquetral osteoarthritis. J Hand Surg 1992;17B:133–136.

13. Trail IA, Linscheid RL. Pisiformectomy in young patients.J Hand Surg 1992;17B:346–348.

14. Lam KS, Woodbridge S, Burke FD. Wrist function afterexcision of the pisiform. J Hand Surg 2003;28B:69–72.

15. Arner M, Hagberg L. Wrist flexion strength after excisionof the pisiform bone. Scand J Plast Reconstr Surg 1984;18:241–245.

16. Garcia-Elias M, Lluch AL, Ferreres A, Castillo F, Saffar P.Resection of the distal scaphoid for scaphotrapeziotrap-ezoid osteoarthritis. J Hand Surg 1999;24B:448–452.

17. Malerich MM, Clifford J, Eaton B, Eaton R, Littler JW.Distal scaphoid resection arthroplasty for the treatment ofdegenerative arthritis secondary to scaphoid nonunion.J Hand Surg 1999;24A:1196–1205.

18. Soejima O, Iida H, Hanamura T, Naito M. Resection of thedistal pole of the scaphoid for scaphoid nonunion withradioscaphoid and intercarpal arthritis. J Hand Surg 2003;28A:591–596.

19. Ruch DS, Papadonikolakis A. Resection of the scaphoid
distal pole for symptomatic scaphoid nonunion after failed

1

1

1

1

1

1

1

1

1

1

1

1


previous surgical treatment. J Hand Surg 2006;31A:588–593.

20. Imbriglia JE. Proximal row carpectomy, In: Berger RA,Weiss APC, eds. Hand surgery. Philadelphia: Lippincott,Williams & Wilkins, 2004:1332–1334.

21. Weiss APC. Principles of limited wrist arthrodesis. In:Berger RA, Weiss APC, eds. Hand surgery. Philadelphia:Lippincott, Williams & Wilkins, 2004:1290.

22. Calandruccio JH. Proximal row carpectomy. J Am SocSurg Hand 2001;1:112–122.

23. Shin AY. Four-corner arthrodesis. J Am Soc Surg Hand2001;1:93–111.

24. Pomerance J. Arthroscopic debridement and/or ulnar short-ening osteotomy for TFCC tears. J Am Soc Surg Hand2002;2:95–101.

25. Katolik LI, Trumble T. Distal radioulnar joint dysfunction.
J Am Soc Surg Hand 2005;5:8–29.
26. Weiss APC. Principles of limited wrist arthrodesis. In:Berger RA, Weiss APC, eds. Hand surgery. Philadelphia:Lippincott, Williams & Wilkins, 2004:1292.

27. Cohen MS. Four-corner fusion. In: Berger RA, Weiss APC,eds. Hand surgery. Philadelphia: Lippincott, Williams &Wilkins, 2004:1315.

28. Burge PD. Scaphotrapeziotrapezoid and scaphocapitate fu-sions. In: Berger RA, Weiss APC eds. Hand surgery.Philadelphia: Lippincott, Williams & Wilkins, 2004:1300.

29. Burge PD. Scaphotrapeziotrapezoid and scaphocapitate fusions.In: Berger RA, Weiss APC, eds. Hand surgery. Philadelphia:Lippincott, Williams & Wilkins, 2004:1304–1305.

30. Garcia-Elias M, Lluch AL, Ferreres A. Partial arthrodesisfor the treatment of radiocarpal osteoarthritis. J Am SocSurg Hand 2005;5:100–108.

31. Wolfe SW. Scapholunate instability. J Am Soc Surg Hand
2001:1:45–60.

article osteoarthritis

Documents