982 D.L. HOLDEN E’r AL.

29. THORNHILL, T. S.: Unicompartmental Knee Arthroplasty. Clin. Orthop., 20!i: 121-131, 1986.30. TJ~RNSTRAND, B. A. E.; EGUND, NIELS; and HAGSTEDT, B. V.: High Tibial Osteotomy. A Seven-Year Clinical and Radiographic Follow-up.

Clin. Orthop., 160: 124-136, 1981.31. UEMATSU, ARINIWA, and KIM, E. E.: Role of Radionuclide Joint Imaging in High Tibial Osteotomy. Clin. Orthop., 144: 220-225, 1979.32. VAINIONP.~.~, SEPPO; L.Z, IKE, ERKKI; KIRVES, PEKKA; and TIUSANEN, PENTTI: Tibial Osteotomy for Osteoarthritis of the Knee. A Five to Ten-Year

Follow-up Study. J. Bone and Joint Surg., 63-A: 938-946, July 1981.

Copyright 1988 by The Journal of Bone and Joint Surgery, Incorporated

Treatment of Ununited Fractures of the Scaphoid byIliac Bone Grafts and Kirschner-Wire Fixation*

BY HERBERT H. STARK, M.D.’~, THOMAS A. RICKARD, M.D.S’, NORMAN P. ZEMEL, M.D.S’,

AND CHARLES R. ASHWORTH, M.D.’~, LOS ANGELES, CALIFORNIA

From Orthopaedic Hospital, California Medical Center, and Hospital of the Good Samaritan, Los Angeles

ABSTRACT: Of 151 ununited fractures of the scaph-old that were treated with lilac bone grafts and Kirsch-her-wire fixation through a volar approach, all but four(97 per cent) healed in an average of seventeen weeks.Three of the four failures resulted from obvious technicalerrors. Neither the preoperative existence of necrosis ofthe proximal fragment nor the location of the fractureaffected the results. When there was mild radioearpalarthritis preoperatively, it did not progress postopera-tively; if there was moderate radiocarpal arthritis pre-operatively, progression seldom was seen if a radialstyloidectomy was done. Displaced and unstable unun-ited fractures healed even if the deformity was not cor-rected completely. The principal benefit of the procedurewas relief of pain rather than an increase either in motionof the wrist or in strength of grip.

At least 5 .per cent of acute fractures of the scaphoidfail to unite after conservative treatment11’13"29m’35’53’61. Thefailures have been attributed to delay in beginning treatment,inadequate immobilization, displacement of the fragments,instability due to ligamentous injury, or inadequate bloodsupply of the proximal fragment4.lLI2ATA8,33,57,rl’~’69.

The regimens of treatment that have been advocatedfor ununited fractures include prolonged immobilization ofthe wrist in a plaster cast, open reduction and internal fix-ation of the fracture fragments, intercarpal fusion, arthro-desis of the wrist, replacement of the scaphoid with aprosthesis, and radial styloidectomy3’2°’22’43’4~’49’56’62"6.. Someauthors have advocated bone-grafting with or without in-

* No benefits in any form have been received or will be received froma commercial party related directly or indirectly to the subject of this article.No funds were received in support of this study.

"~ 2300 South Hope Street, Suite 400, Los Angeles, California 90007.

temal fixation of the fracture, or multiple drilling of thefragments, or fascial arthroplasty, or proximal row carpec-tomy, or electrical stimulation, or partial or total excisionof the scaphoid 1’2"4"6’8"9’27’28’40’57. Bone-grafting has been themc.st popular surgical treatment6A3’25’26’46’69.

Herbert and Fisher24 recommended discontinuing im-mc,bilization of the wrist in a plaster cast six weeks afteran acute fracture, regardless of whether healing was evidenton radiographs. If, after an additional two or three weeks,the fracture had not healed, he advocated open reductionusing bone grafts and internal fixation with a special screw.For undisplaced and stable fractures, many surgeons_p__referto continue immobilization of the wrist in a plaster cast forfour to six months before recommending an operativeprocedure~’3~’5~. If, however, the fragments are grossly dis-placed or unstable because of ligamentous or osseous dis-ruption, they recommend open ~eduction and internalfixation as soon as possible7"11"12"~’*A9’37’~8’42’53.

Materials and Methods

Since 1966, we have treated 151 ununited fractures ofthe scaphoid with iliac bone-grafting and internal fixationusing Kirschner wires. All of the patients had had symptomsfor at least six months before the operation. The location,displacement, and instability of each fracture; the presenceor absence of avascular necrosis; and the degree of radi-ocarpal arthritis were recorded. Radiographs that were madepreoperatively and during treatment, as well as those thatwere made at the most ~ecent follow-up examination (whichincluded at least six radiographs of the wrist), were re-viewed. Subjective complaints, range of motion of the wrist,artd strength of grip were recorded. The average length offollow-up was forty-two months (range, one year to morethan ten years). Twenty-three patients were examined within

e:-

n

te

TREATMENT OF UNUNITED FRACTURES OF THE SCAPHOID 983

one to two years after the index operation; ninety-five pa-tients, within two to three years; eighteen patients, withinthree to five years; four patients, within five to ten years;and eleven patients, more than ten years after the operation.

There were 138 male and thirteen female patients. Thedominant wrist was involved in more than half of the pa-tients. The average age at Operation was twenty-five years(range, twelve to fifty-nine years).

Eleven patients were less than seventeen years old. Thisshows that ununited fractures of the scaphoid in childrenand teen-agers are more common than has been generallyrealized~1’16’3°’~°’5s. All of the pediatric patients had an es-tablished non-union, and none had radiocarpal arthritis.Often the fracture had occurred at the age of ten or twelveyears, but symptoms did not arise until later. All elevenfractures in these patients united after the index procedure.

Seventy-eight of the 151 fractures (52 per cent) wereattributed to an athletic injury; thirty-two (21 per cent), a fall; and twenty-two (14 per cent), to a motor-vehicleaccident.. Twenty patients, including four who were lessthan seventeen years old, could not recall or were unsurewhen the fracture had occurred. All twenty had complainedof pain in the wrist for more than six months, and all hadcystic changes indicating that the fracture had been presentfor many months. For the 131 patients who had a definiteinjury, the time-interval between the injury and bone-graft-ing ranged from six to 108 months (average, twenty-threemonths). Thirty-two patients had the index operation morethan two years after the initial injury and fifty,three, withintwelve months after the injury. In twenty-nine of the 151patients (approximately 20 per cent), the injury was work-related.

Thirty-two fractures were in the proximal third of thebone; 106, in the middle third; and thirteen, in the distalthird (Table I). In forty-two patients the fracture fragmentswere displaced at least one millimeter, and in twenty-fivepatients the fracture was considered unstable because of adorsal intercalary-segment .instability configuration of thecarpal bones and a scapholuna~e angle of more than 60degrees.

The treatment before the index operation included theuse of a sling, taping of the wrist, the use of an elastic wrapor some other splint, immobilization with a short or longthumb-spica plaster cast, intra2articular injections of ste-roids, electrical stimulation, and internal fixation with orwithout bone-grafting. Eight patients had had a previousbone.graft, and three others had had at least three monthsof electrical stimulation supplemented with immobilizationof the wrist in a plaster cast.

The main indications for grafting were an ununitedfracture with disabling pain in 107 patients; pain and stiff-ness in the wrist in twenty-two; stiffness and extreme weak-ness of grip in five; and pain, weakness, and stiffness inthe wrist in seventeen. There was radiographic evidence ofmild radiocarpal arthritis (increased prominence or loss ofthe rounded contour of the radial stytoid process) in fortywrists; this was not a contraindication to grafting, and nei-

TABLE I

DATA ON ONE HUNDRED AND FIFTY-ONE FRACTURES OF THE SCAPHOID

UnitedWrists Fractures

No. Per Cent No. Per CentAverage Time

to Union(Wks.)

Total 151 147 97 17.0Site of fracture

Distal third 13 9 13 100 15.3Middle third 106 70 103 97 I7.4Proximal third 32 21 31 97 16.5

Preoperative 25 16 21 84 17.4avascular necrosis

Stabh; fractures 126 83 123 98 16.9~Unstable fractures 25 17 24 96 17.2Displaced fractures 42 28 41 98 17.0

ther was the presence of moderate radiocarpal arthritis (intwe’~tve wrists), as shown by narrowing or irregularity of theradiocarpal articulation, limited to the space between thestyloid process and the distal pole of the scaphoid (Fig. 1).However, when moderate arthritis was present, a styloi-dec~:omy was performed either at the time of grafting or asa subsequent procedure after the fracture had united (Figs.2-A through 2-D). The presence of severe radiocarpal ar-thritis was a contraindication for bone-grafting of an un-united fracture of the scaphoid.

Avascular necrosis of the proximal fragment, whichwas present in twenty-five wrists, was not a contraindicationto grafting; however, when the proximal fragment was foundto be sclerotic, fragmented, badly deformed, or devoid ofcav:ilage, bone-grafting was not performed. The latter com-plex of abnormalities was encountered only twenty-eighttimes in nineteen years; in each patient, the proximal fracturefragment was replaced with a hand-carved silicone-ru_.bberspacer7~. These patients were not included in the presentstudy. Except for these exclusions, the series of 151 patientsis consecutive.

Operative Technique ~

The Scaphoid is exposed through a straight or zigzagvol:ar incision. After the wrist capsule is incised longitu-dinally and the wrist is dorsiflexed, both parts of the scaph-oid as well as the articular surface of the radius can be seenreadily. A small, rectangular window of bone is removedfrem the volar aspect of the distal fragment immediatelyadjacent to the fracture. Through this opening, both frag-ments are cleared of fibrous tissue and dead bone, using alow-speed power burr or curet.

As emphasized by Matti4~’"~, a large cavity is fashionedin both the proximal and distal parts of the scaphoid. AChandler retractor is uied to protect the articular cartilageof the radioscaphoid joint. It also helps to correct angulation,malrotation, and displacement of the fragments. The volarpart of the cortex of the scaphoid is often deficient, and thisdeficiency permits an exaggerated volar tilt of the distalfragment. Realignment and reduction of the fracture, and

984

normal

H. H. STARK ET AL.

moderate

ARTHRITIS

mildFro. I

Illustration of mild and moderate radiocarpdl arthritis.

restoration of the bone to the proper length, is a difficultpart of the procedure7"~3’~4. Occasionally, intraoperative ra-diographs may have to be made. In the present series, cor-rection of the displacement ordinarily was assessed visuallyand not radiographically. The scaphoid is transfixed withtwo 0.035-inch (0.9-millimeter) Kirschner wires, which areinserted through the distal fragment into the proximal one,while the articular cartilages of the scaphoid and the radiusare protected with the retractor (Fig. 3). Correct placementof the wires is ensured by observing them through the volarwindow.

Cancellous bone from the ilium is packed into the cav-ity. ;The wires can be inserted after packing the cavity withbone, but it is easier to verify their location before insertingthe graft. Often a cortical bone graft can be fashioned to fitsnugly int~ the volar window, and then it should be im-mobilized with one additional 0.28-inch (0.7-millimeter)Kirschnei: wire. Theoretically, this technique helps to sta-bilize and support the volar aspect of the cortex, but in thisstudy there was no statistical evidence that it improved therate of union or the final result 7’~3’~4. The Kirschner wiresare cut off beneath the skin. The capsule is approximatedwith absorbable sutures, the skin is closed, and the extremityis immobilized in a long thumb-spica plaster splint, whilethe forearm is held in supination with the wrist in neutraland the thumb in opposition. Two weeks postoperatively,the sutures in the skin are removed. A long thumb-spicaplaster dressing is worn for six additional weeks, followedby a short thumb-spica cast that is worn until the fractureunites or the procedure is considered a failure. The Kirschnerwires are removed, usually under local anest[aesia, after thefracture has united. Once immobilization is discontinued,patients are permitted to use the wrist and hand for lightactivities, but strenuous and forceful activity is discouragedfor an additional two months. Formal physiotherapy is con-sidered to be unnecessary and a needless expense.

Results

Union was judged to be present when, on radiographicexamination, there were definite trabeculae across the siteof the fracture and the fracture gap had disappeared59. Weroutinely made a posteroanterior radiograph of the wrist inmaximum radial and ulnar deviation, a lateral radiograph,and both oblique radiographs, as well as a so-called fistradiograph of the wrist 34’3~’7°. In recent years, if healing wasquestionable, a tomogram or computerized tomography scanalso was made. We were conservative in deciding when afracture had healed, believing it better to immobilize thewrist longer than to risk an unsatisfactory result7.

There were no infections or postoperative hematomasin either operative site, nor did we encounter proble.rn_s withhealing of the wound, tenderness of the scar, or damage tosensory nerves.

All but four of the 151 fractures healed, in an averageof seventeen weeks postoperatively (Table I). The shortesttime to union was eight weeks ar{d the longest, thirty-threeweeks. Thirty-one of thirty-two fractures through the prox-imal third of the scaphoid healed in an average of 16.5weeks (range, 10.4 to twenty-four weeks); 103 of 106 frac-tures through the middle third, in an average of 17.4 weeks(range, eight to thirty-three weeks); and all thirteen fracturesthrough the distal third, in an average of 15.3 weeks (range,10.3 to seventeen weeks). Several authors have reported lower rate of union or a longer time to union, or both, afterbone-grafting of fractures of the proximal third of thescaphoid7"~°’5~’5~’~’~9. This was not found in the presentstudy, although the fractured wrists that had avascular ne-crosis as determined preoperatively on radiographs had aslightly lower rate of union than did those that did not haveavascular necrosis: twenty-one (84 per cent) of the twenty-five fractured wrists for which there was radiographic evi-dence of preoperative avascular necrosis of the proximalfragment united within an average of 17.4 weeks (range,

TREATMENT OF UNUNITED FRACTURES OF THE SCAPHOID 985

FIG. 2-A FIG. 2-B

Figs, 2-A through 2-D: A thirty-two-year-old man had injured the wrist while playing football.Fig. 2-A: Preoperative radiograph, made in November 1985, showing an ununited fracture of the scaphoid fifteen years after the injury. Synovial

pseudarthrosis and a large exostosis of the scaphoid are evident.Fig. 2-B: .Three weeks after the operation.

twelve to 28.1 weeks), while 100 per cent of the 126 frac-tured wrists that did not have avascular necrosis united (Ta-ble I). The necrosis eventually cleared in all twenty-onefractures that healed after grafting.

Sixteen patients who did not have avascular necrosisbefore the operation had radiographic evidence of necrosisin the proximal fragment afterward. Nevertheless, in thesepatients union occurred in an, average of 17.9 weeks (range,nine to 27.9 weeks), and the necrosis cleared afterward.The fracture that required 27.9 weeks to unite also hadprogression of preoperative radiocarpal arthritis, necessi-tating a fusion of the wrist because of pain. Only one otherpatient, in this grouphad radiographic evidence of mildradioscaphoid arthritis when examined eight years after the

operation, and that patient did not have pain in the wrist.All of the fractures that had failed to unite after previous

bone-grafting (eight fractures) or electrical stimulation(three fractures) united after our procedure.

Of the four fractures that failed to heal, three werethrough the middle third of the bone. In one, the displace-ment measured more than one millimeter and the scapho-lunate angle, more than 60 degrees. Three fractures wereundisplaced and stable, and one of them was through theproxim.al third, but all four had preoperative avascular ne-crosis of the proximal fragment. Technical error rather thanavascular necrosis was considered to be the most likely causeof three of these four failures. In two, the Kirschner wireswere not placed correctly and did not immobilize the frag-

FIG. 2-C FIG. 2-DFig. 2-C: Eight months after the operation. The fracture had healed, exostosis was still present, and there was moderate radiocarpal arthritis.Fig. 2-D: Radiograph made in May 1987, six months after styloidectomy of the radius and excision of the scaphoid exostosis.

ments. In another, one of the wi~es, although it had beenplaced correctly, backed out of the bone within eighl weeksafter the operation, an6 all of the wires had ~ b~ ~emove~before the bone united (Figs. 4-A t~ougg

Of the 147 patients whose fract~e united, ninety-ninedenied having pain on follow-up. Thi~y-four had slight paina~ strenuous use of the w~st, b~z ~e pai~ did notwith work or recreational pursuits, and they did not useanalgesics. Fou~een denied having ~ain, Opt 1bey were oc-casionNty annoyed by residuaI stiffness of the wrist. Onaverage, the grip strength as measured with the Jam~ dy-namometer (Fred Sammons, Buw Ridge, Illinois) increasedfrom 3~.4 kilo~ams ~f force b~fo~e ope, at{o~ t~ 39.6 kil-ogres of fore orchard. The average g~p strength of theoppositE (uninjured) hand was 46.6 Mlograms of force. Onlyseventeen patients ~ad no,at grip strength after the surgicalprocedure. On average; dorsiflexion of the wris~ increased2 de~ees, from an average of 55 degrees preoperatively.On average, radial deviation oft~e ~st increased 3 dogiesfrom 16 de~ees preoperatively, wRRe palma flexion andutn~ ~eviation ,emained unchange6 (a~erage, 60 and 31degrees).

rudiqc~N ~tis. The operation was successM in all of¯em, e float N1 uNted after b~ne-graffing and intemat fix-orion. None of the patients who hM mild ~hfitis ~ad pro-~ession of the ~ds after the fracture ~ea~ed, Nve of thet~Nve patients who had moderate ~dtis had a styloidec-tomy ~f the radius at t~e time of graftJ~g. N these fivepatients, at the most recent examinatioo (twenty-five,twenty-seven, fogy-eight, sixty-four, and sixty-six monthspostoperatively), ~e a~dtis had not pmgessed; three

aRer heavy ~erk, Five others wRo ~ad moderatepreope~fively had a ~dial styloide~I~my as a seconda~

procedure for pain after the scaphoid had united, Four hadc~mplcte relief of the pain. The fifth failed to obtain reliefeven afle~ an intercarpaI fusion, and ~ater the w~ist a~so wasfused. At the time of w~itiag, two patients had not yet hada sZyloidegtomy although it hag been recommended.

It has been mentioned already that severe radiocarpalarthritis was a contraindicatioo ~o our grafting procedure.l~tecause of two theoretical advantages of performing a sty-loidec~omy as a secondary procedure after the scaphoid hashealed ~ because the sty~oid process provides some support~o the scaphoid, and because if the fracture fails to unRe,fusion of the wrist is easier when the styloid process is stillpresen~ ~ we sometimes (.in fi’4e of te~ ,,vrists) did thestytoidectomy as a secondary procedure. At ~l~e-time ofw:gih~g, ~one of the forty patients who had mild arthritishad ~aa progression of the arthritis or had had a styloidec~foray.

Four patients, t~ree ~f whom had moderate radiocarpalarthritis preoperatively, continued to complain of pain inthe wrist after the fracture/~ad healed. At the time of writing,two had been advised to have a radiai styloidectomy as asecondary procedure. The other two had progressive ra-diosvaphoid arthritis e-~e~’~ though the fracture had healed.kt one we fused the wrist. The patient wh~ did not havearthritis preoperatively had the proximal portion of thescaphoid replaced with a hand-carved silicone-rubber spacertw~ years after bone-grafting; when last seen, nine yearslater, ~here was no more pain in the wrist.

Ununited fractures ~f the scaphoid have been termeddisplaced and unsta/ale w~en the scap~oJ~nate angle is morethan 45 degreesv, but in this study we considered the f, actureto be unstable when the scapholunate angle exceeded 60degrees. According t~ ou~ ~titeria, ~26 og the fractures wereg~ab~e and the other- twenty-five were unstable. Forty-twof, avtures, of which fifteen were classified as unstable ac~

TREATMENT OF UNUNITED FRACTURES OF THE SCAPHOID 987

FIG. 4-A FIG. 4-B

Figs. 4-A through 4-D: A fifty-nine-year-old woman did not know when she had injured the wrist. There had been pain in the wrist for three yearsbefore the operation.

Fig. 4-A: Preoperative radiograph, made in June 1977.Fig. 4-B: Three weeks after the operation.

cording to our criteria, had a displacement of one millimeteror more as measured on the posteroanterior radiographs.Since all but one or two ,of the unstable, or stable anddisplaced, or undisplaced fractures united, it was evident inour patients that neither displacement nor instability was animportant factor in achieving union (Table I).

Preoperatively, thirteen of the twenty-five patients whohad an unstable fracture had mild or moderate radiocarpalarthritis, whereas only thirty-nine of the 126 patients who

had a stable fracture had mild or moderate radiocarpal ar-thritis. This demonstrated that radiocarpal arthritis is morelikely to be present when the fracture is unstable. Afterunion of the fracture, the scapholunate angle was diminished(by an average of 11 degrees) in fifteen of the twenty-fourunstable fractures that healed. The scapholunate angle wasrestored to normal in six of the fifteen wrists, and in ninethe scapholunate angle was larger than it had been preop-eratively. The increase in the scapholunate angle was-con-

FtG. 4-C FIG. 4-D

Fig. 4-C: Eight weeks after the operation. One Kirschner wire had lost purchase in the distal fragment.Fig. 4-D: Five months after the operation. All Kirschner wires were removed nine weeks after the operation because of irritation of the skin. The

n’acture was ununited and the operation was a failure.

988 H. H. STARK ET AL.

sidered to be major in only two patients (10 and 15 degrees).Of the other seven patients, the angle was increased by 1degree in one patient, by 2 degrees in two, by 3 degrees inone, and by 6 degrees in three. Once an unstable fracturehad healed, radiocarpal arthritis seldom progressed even ifthe scapholunate angle and the abnormal configuration (dor-sal intercalary-segment instability) still were present. Wetried to correct displacement, malrotation, and malalignmentas well as to restore the length of the scaphoid and a normalrelationship between the lunate and scaphoid, but, evenwhen we did not accomplish that objective, most of ourpatients had satisfactory function of the wrist and were freeof pain once the fracture of the scaphoid had healed. How-ever, the length of follow-up for most of our patients wastoo short to be sure that the results will not deteriorate overtime51,59.

Five patients, who had had avascular necrosis but notradiocarpal arthritis preoperatively, had radiocarpal arthritisafter the fracture of the scaphoid had healed. In four of thesepatients, the arthritis was mild and did not cause pain. Thewrist Of the fifth patient became painful and stiff, and serialradiographs showed disintegration of the proximal pole ofthe scaphoid.

Discussion

From a technical standpoint, the volar approach thatwas popularized by Russe affords a clear view of any un-united fracture of the scaphoid and eliminates the need fora radial styloidectomy53. However, this approach may haveno advantage over a dorsal or lateral approach in terms ofpreserving the supply of blood to the bone. It does provideexcellent access to the volar part of the cortex of the scaph-oid, which, if deficient, can be reconstituted easily with acortical bone graft7’14’6s’66.

:. Our experience as well as that of others has shown thatradiographic evidence of avascutarity of the proximal frag-ment is .not a contraindication to a grafting proce-dure7,23,36,45,46,5L~3,69,7~. Fractures that are associated withavascutar necrosis of the proximal fragment usually uniteafter grafting and internal fixation, but we found that de-generative arthritis sometimes develops later in the wristsof such patients. If it did not, the result was satisfactory.

While in the operating theater, we were unable to de-termine whether there was sufficient vascularity of the prox-imal fragment to ensure union after grafting 7"2~. If theproximal fragment was large enough to accept a graft anda Kirschner wire, we did the procedure as described. Whenthe proximal fragment was fragmented or badly deformed,or the articular cartilage was damaged, or the proximalfragment was too small for Kirschner-wire fixation, we be-lieved, as have others 23’33"37’42’45’46’53, that bone-graftingwould fail, and such patients were excluded .from the seriesand treated in some other manner. This decision must bemade by the surgeon in the operating theater7~.

Advanced radiocarpal arthritis, a small avascular prox-imal fragment, or a lack of symptoms has been consideredto be a contraindication for bone-grafting of an ununited

fracture of the scaphoid. However, other studies have shownthat there is a risk that painful arthritis will develop in thewrist of an adult who has an ununited fracture of the scaph-oid, even if asymptomatic, and often the arthritis develops

4only a few years after the occurrence of the fracture4’3°’3~’ ~’5".We believe that most patients who have an ununited fractureof the scaphoid, even if asymptomatic, should be advisedof this risk and encouraged to have bone-grafting and in-ternal fixation of the fracture. Between 1982 and 1985, wesuccessfully grafted five ununited fractures of the scaphoidin patients who had very mild complaints and had minimum.stiffness of the wrist. We are convinced that this approachis justified in otherwise healthy patients39.

A successful result (union of the fracture) after treat-.ment of ununited fractures of the scaphoid with non-invasiveelectrical stimulation, as occurred in two series reported byothers~5’47, has not been nearly as frequent as after bone-grafting. Electrical treatment probably is contraindicatedwhen there is a major collapse of the fragments, or angulardisplacement, or a synovial pseudarthrosis~’~7, but these arenot contraindications to bone-grafting and internal fixation.We have used non-invasive electrical stimulation in rareinstances when a patient insisted on it, but have achieveda rate of union of only 70 per cent by this method. Ad-mittedly, our experience with it has been very limited.

Internal fixation of an unstable, ununited fracture ofthe scaphoid, with or without bone-grafting, has been ad-w~cated repeatedly, and various screws, as well as specialir~stmments for insertion of the screws, have been designedfor this purpose17’24’32"37’42"55. For example, insertion of theHerbert compression screw requires special equipment. Ita].s9 demands technical skill of a high order. One disad-wantage of this method is that it violates the scaphotrapezialjoint"-3’2~, and if the screw is misdirected, it can damage thea~icular cartilage. This method does have an advantage inthat it provides more rigid fixation than do Kirschneewires.

We are not the first to recommend Kirschner-wire fix-ation in conjunction with bone-grafting for an unstablefracture7"~3"5’*’6°’65, but, as far as we know, we are the firstto recommend the use of Kirschner wires in all graftedfi:actures. Because it may be difficult to judge the stabilitythat will be achieved with a bone graft, and because fixationwith Kirschner wires is easy to accomplish and adds littleto the operating time, we prefer to fix all grafted fractureswith them. We use at least two wires to immobilize thet~racture after grafting. To supplement the wire fixation, weuse a cast, as mentioned, for about eighteen weeks. Patientsreadily accept postoperative immobilization in a cast for thatprolonged period, because of the rate of success of theprocedure. Rasmussen et al. ~’ reported no correlation be-tween the duration of postoperative immobilization and theultimate range of motion of the wrist after grafting of anununited fracture of the scaphoid.

Gross displacement of three millimeters or more or ascapholunate angle of more than 60 degrees, or both, in anacute fracture indicates instability. Such a fracture has beenthought to be less likely to unite, when treated conserva-

TREATMENT OF UNUNITED FRACTURES OF THE SCAPHOID

TABLE II

RESULTS OF BONE-GRAFTING FOR UNUNITED FRACTURES OF THE SCAPHOID AS REPORTED IN THE ENGLISH-LANGUAGE LITERATURE

989

YearNo. of Rate of

Series Wrists Union(Per cent)

No. ofFractures

Fixed withKirschner

Wires Technique

19601962196819681968

1969

197419751980

1982

19841984

198519851986

1987

Russe53 22 90 0Pennsylvania Orthop. Soc.~ 21 76 0Verdan and Narakas69 45 98 0Dooley~° 23 86 0Mulder.5 100 97 0

Unger and StrykeW 42 79 0

T6mgren and Sandqvist67 45 71 0McDonald and Petrie36 48 80 0Cooney et al.7 44 86 20

22 91 2218 50 4

Schneider and Aulicino~ 31 87 9Boeckstyns and Busch6 28 86 0Fisk~3 41 73 *

Rasmussen et al. 5~ 28 71 0Green2t 45 75 0Steichen and Schreiber~ 25 92 ~

Stark et al.5927 81 0

Russe (volar inlay graft)Peg graft, some with styloidectomyRusse, some grafted if unhealed after three monthsRusseRusse, grafted if unhealed after four months;

failures thought to be due to inadequateimmobilization

Russe, grafted if unhealed after four months;failures thought to be due to inadequateimmobilization

Peg graft inserted through lateral approachRusseRusseDorsal inlay graft, some with radial styloidectomyPeg graft (Murray technique), some grafted

unhealed after four monthsRusseRusseWedge bone graft through lateral incision, styloid

of radius used as graftRusseRusseRusse, using graft from radius; non-union present

for at least five months in all wristsRusse; non-union present for at least seven months

in all wrists

* Kirschner wires were used occasionally.~" Kirschner wires were used if the fracture was thought to be unstable.

tively, than an undisplaced or stable fracture 7A2A4,as. How-ever, neither displacement nor instability in a long-standing,unhealed fracture deters healing if the fracture is reducedand grafted4s,sL54.

A summary of the results that have been reported byseveral authors after bone-grafting of ununited fractures of

the scaphoid (Table II) shows rates of success of 50 to__98per cem. We achieved union in 147 (97 per cent) of 151long-standing ununited fractures of the scaphoid, and oursis a much larger series than those reported previously. Webelieve: that our success was due, atfleast in part, to thestabilization of the fracture and graft with Kirschner wires.

References1. ADAMS, J. D., and LEONARD, R. D.: Fracture of the Carpal Scaphoid. A New Method of Treatment with a Report of One Case. New England J.

Med., 198: 401-404, 1928.2. AGNER, OLAF: Treatment of Ununited Fractures of the Carpal Scaphoid by Bentzon’s Operation. Acta Orthop. Scandinavica, 33: 56-65, 1963.3. BARNARD, LEONARD, and STUBBINS, S. G.: Styloidectomy of the Radius in the Surgical Treatment of Non-Union of the Carpal Na~icular. A

Preliminary Report. J. Bone and Joint Surg., 30-A: 98-102, Jan. 1948.4. BARR, J. S.; ELLISTON, W.; MUSNICK, HENRY; DELORME, T. L.; HANELIN, JOSEPH; and THIBODEAU, A. A.; Fracture of the Carpal Navicular

(Scaphoid) Bone. An End-Result Study in Military Personnel J. Bone and Joint Surg., 35-A: 609-625, July 1953.5. BENTZON, P. G. K., and RANDLOV-MADSEN, AAGE: On Fracture of the Carpal Sczphoid. A Method for Operative Treatment of Inveterate Fractures.

Acta Orthop. Scandinavica, 16: 30-39, 1946." 6. BOECKSTYNS, M. E; H., and BUSCH, PETER; Surgical Treatment of Scaphoid Pseudarthrosis. Evaluation of the Results after Soft Tissue Arthroplasty

and Inlay Bone Grafting. J. Hand Surg., 9A: 378-382, 1984.7. COONEY, W. P., III; DOBYNS, J. H.; and LINSCHEID, R. L.’ Nonunion of the Scaphoid. Analysis of the Results from Bone Grafting. J. Hand

Surg., 5: 343-354, 1980.8. CRABBE, W. A.: Excision of the Proximal Row of the Carpus. J. Bone and Joim Surg.., 46-B(4): 708-711, 1964.9. DAVIDSON, A. J.) and HORWITZ, i. T.: An Evaluation of Excision in the Treatment of Ununited Fractures of the Carpal Scaphoid (Navicular)

Bone. Ann. Surg., 108: 291-295, 1938.10. DOOLEY, B. J.: Inlay Bone Grafting for Non-Union of the Scaphoid Bone by the Anterior Approach. J. Bone and Joint Surg., 50-B(1): 102-I09,

1968.] I. EDDELAND, ALLAN; EIKEN, ODDVAR; HELLGREN, ERIK; and OHLSSON, N.-M.: Fractures of the Scaphoid. Scandinavian J. Plast. and Reconstr.

Surg., 9: 234-239, 1975.;2. FIsK, G. R.: Carpal Instability and the Fractured Scaphoid. Hunterian Lecture Delivered at the Royal College of Surgeons of England on 7th May

1968. Ann. Roy. Coll. Surg., 46: 63-76, 1970.

990 H.H. STARK ET AL.

13. FISK, G. R.: Non-Union of the Carpal Scaphoid Treated by Wedge Grafting. In Proceedings of the British Orthopaedic Association. J. Bone andJoint Surg., 66-B(2): 277, 1984.

14. FIsK, G. R.: The Wrist. J. Bone and Joint Surg., 66-B(3): 396-407, 1984.15. FRYKMAN, G. K.; TALEISNIK, JULIO; PETERS, GREGORY; KAUFMAN, RICnAR~; HELAL, BASIL; WOOD, V. E.; and UNSELL, R. S.: Treatment of

Nonunited Scaphoid Fractures by Pulsed Electromagnetic Field and Cast. J. Hand Surg., llA: 344-349, 1986.16. GAMBLE, J. G., and SIMMONS, S. C., HI: Bilateral Scaphoid Fractures in a Child. Clin. Orthop., 162: 125-128, 1982.17. GASSER, HERNAN: Delayed Union and Pseudarthrosis of the Carpal Navicular: Treatment by Compression-Screw Osteosynthesis. A Preliminary

Report on Twenty Fractures. J. Bone and Joint Surg., 47-A: 249-266, March 1965.18. GELBERMAN, R. H., and MENON, JAVASANKER: The Vascularity of the Scaphoid Bone. J. Hand Surg., 5: 508-513, 1980.19. GOLDMAN, SHERW~N; LIPSCOMB, P. R.; and TAYLOR, W. F.: Immobilization for Acute Carpal Scaphoid Fractures. Surg., Gynec. and Obstet.,

129: 281-284, 1969.20. GRANER, ORLANDO; LOPES, E. I.; CARVALHO, B. C.; and ATLAS, SAMUEL: Arthrodesis of the Carpal Bones in the Treatment of Kienbrck’s

Disease, Painful Ununited Fractures of the Navicular and Lunate Bones with Avascular Necrosis and Old Fracture-Dislocations of Carpal Bones.J. Bone and Joint Surg., 48-A: 767-774, June 1966.

21. GREEN, D. P.: The Effect of Avascular Necrosis on Russe Bone Grafting for Scaphoid Nonunion. J. Hand Surg., 10A: 597-605, 1985.22. HELOT, A. J.: A New Operation for Ununited Fracture of the Scaphoid. Jtn Proceedings of the South African Orthopaedic Association. J. Bone

and Joint Surg., 34-B(2): 329, 1952. 23. HERBERT, T. J.: Use of the Herbert Bone Screw in Surgery of the Wrist. Clin. Orthop., 202: 79-92, 1986.24. I-I~RBERT, T. J., and FISHER, W. E.: Management of the Fractured Scaphoid Using a New Bone Screw. J. Bone and Joint Surg., 66-B(1): 114-

123, 1984.25. HERNESS, D., and POSNER, M. A.: Some Aspects of Bone Grafting for Non-Union of the Carpal Navicular. Analysis of 41 Cases. Acta Orthop.

Scandinavica, 48: 373-378, 1977.26. HULL, W. J.; HOUSE, J. H.; GUSTILO, R. B.; KLEVEN, LOWELL; and THOMPSON, WAYNE: The Surgical Approach and Source of Bone Graft for

a Symptomatic Non-Union of the Scaphoid. Clin. Orthop., 115: 241-247, 1976.27. INGLIS, A. E., and JONES, E. C.: Proximal-Row Carpectomy for Diseases of the Proximal Row. J. Bone and Joint Surg., 59-A: 460-463, June

1977.28. JORGENSEN, E. C.: Proximal-Row Carpectomy. An End-Result Study of Twenty-two Cases. J. Bone and Joint Surg., 51-A: 1104-111 i, Sept.

1969.29. KLEINERT, J. M., and ZENN1, E. J., JR.: Nonunion of the Scaphoid, Review of Literature and Current Treatment. Orthop. Rev., 13: 19-35, March

1984.30. LARSON, BR~D; L|GHT, T. R.; and OGDEN, J. A.: Fracture and Ischemic Necrosis of the Immature Scaphoid. J. Hand Surg., 12A: 122-127, 1987.31. LESLIE, I. J., and DICKSON, R. A.: The Fractured Carpal Scaphoid. Natural History and Factors Influencing Outcome. J. Bone and Joint Surg.,

63-B(2): 225-230, 1981.32. LE~CSHON, ANTHONY; IRELAND, JOHN; and TR~CKEV, E. L.: The Treatmertt of Delayed Union and Non-Union of the Carpal Scaphoid by Screw

Fixation. J. Bone and Joint Surg., 66-B(1): 124-127, 1984.33. LICHTMAN, D. M., and ALEXANDER, C. E.: Decision Making in Scaphoid Non-Union. Orthop. Rev., 11: 55-67, March 1982.34. L~t.~DGREN, E.: Some Radiological Aspects on the Carpal Scaphoid and Its Fractures. Acta Chit. Scandinavica, 98: 538-548, 1949.35. LONDON, P. S.: The Broken Scaphoid Bone. The Case against Pessimism. J. Bone and Joint Surg., 43oB(2): 237-244, 1961.36. MCDONALD, GLEN, and PETRIE, DAVID: Un-United Fracture of the Scaphoid. Clin. Orthop., 108:110-114, 1975.37. MCLAUGHL~N, H. L.: Fracture of the Carpal Navicular (Scaphoid) Bone. Some Observations Based on Treatment by Open Reduction and Internal

Fixation. J. Bone and Joint Surg., 36-A: 765-774, July 1954.38. MCLAUGHLIN, H. L., and PARKES, J. C., II: Fracture of the Carpal Navicu:lar (Scaphoid) Bone. Gradations in Therapy Based upon Pathology.

Trauma, 9: 311-319, 1969.39. MACK, G. R.; BOSSE, M. J.; GELBERMAN, R. H.; and Ytr, E~IC: The Natural History of Scaphoid Non-Union. J. Bone and Joint Surg., 66-A:

504-509, April 1984.40. MATTI, HERMANN: Technik und Resultate meiner Pseudarthrosenoperation.. Zentralbl. Chir., 63: 1442-1453, 1936.41. MATTI, HERMANN: 0her die Behandlung der Navicularefraktur und der Refractura patellae dutch Plombierung mit Spongiosa. Zentralbl. Chir.,

64: 2353-2359, 1937. .42. MAUDSLEV, R. H., and CHEN, S. C.: Screw Fixation in the Managementof the Fractured Carpal Scaphoid. J. Bone and Joint Surg., 54-B(3):

432-441, 1972.43. MAZ~T, ROBERT, JR., and HOHL, MASON: Radial Styloidectomy and Styloidectomy Plus Bone Graft in the Treatment of Old Ununited Carpal

Scaphoid Fractures. Ann. Surg., 152: 296-302, 1960.44. :METCALFE, J. W.: The Vitallium Sphere Prosthesis for Nonunion of the Navicular Bone. J. Intemat. Coll. Surg., 22: 459-462, 1954.45. MULDER, J. D.: The Results of 100 Cases of Pseudarthrosis in the Scaphoid Bone Treated by the Matti-Russe Operation. J. Bone and Joint Surg.,

50-B(1): 110-115, 1968.46. MtrRR~’x’, GORDON: End Results of Bone-Grafting for Non-Union of the Carpal Navicular. J. Bone and Joint Surg., 28: 749-756, Oct. 1946.47. OSTER~AN, A. L., and BORa, F. W., JR.: Electrical Stimulation Applied to Bone and Nerve Injuries in the Upper Extremity. Orthop. Clin. North

America, 17: 353-364, 1986.48. PENNSYLVANIA ORTHOPEDIC SOCIETY SCIENTIFIC RESEARCH COMMITTEE: Evaluation of Treatment for Non-Union,.of the Carpal Navicular. J. Bone

and Joint Surg., 44-A: 169-174, Jan. 1962.49. PETEgSON, H. A., and LIPSCOM~, P. R.: Intercarpal Arthrodesis. Arch. Surg., 95: 127-134, 1967.50. PICK, R. Y., and SEPAL, DAWD: Carpal Scaphoid Fracture and Non-Union in an Eight-Year-Old Child. Report of a Case. J. Bone and Joint Surg.,

65-A: 1188-1189, Oct. 1983.51. RASMUSSEN, PAUL; SCh’WAn, J. P.; and JOHNSON, R. P.: Symptomatic Sc;tphoid Nonunions Treated by Russe Bone Grafting. Orthop. Rev., 14:

41-47, Jan. 1985.52. RUBY, L. K.; ST~SON, JOHN; and BELSI~, M. R.: The Natural History of Scaphoid Non-Union. A Review of Fifty-five Cases. J. Bone and Joint

Surg., 67-A: 428-432, March 1985.53. RUSSE, OT’ro: Fracture of the Carpal Navicular. Diagnosis, Non-Operative Treatment, and Operative Treatment. J. Bone and Joint Surg., 42-A:

759-768, July 1960.54. SCHNEIDER, L. H., and AUL~C~O, PAT: Nonunion of the Carpal Scaphoid. The Russe Procedure. J. Trauma, 22: 315-319, 1982.55. Sr~_~w, J. S.: A Biomechanical Comparison of Scaphoid Screws. J. Hand Surg., 12A: 347-353, 1987.56. SMITH, L’~MAN, and FR~ED~tAY, BARRY: Treatment of Ununited Fracture of the Carpal Navicular by Styloidectomy of the Radius. J. Bone and

Joint Surg., 38-A: 368-375, April 1956.57. SOTO-HALL, RALI’Xa, and HALDEMAN, K. O.: The Conservative and Operative Treatment of Fractures of the Carpal Scaphoid (Navicutar). J. Bone

and Joint Surg., 23: 841-850, Oct. 1941.58. Sotrr~COTT, R., and ROSMAN, M. A.: Non-Union of Carpal Scaphoid Fractures in Children. J. Bone and Joint Surg., 59-B(1): 20-23, 1977.59. STARK, ANDRe; BROTSTROM, L.-A.; and SVARTENGRnN, GUNNAR: Scapl~toid Nonunion Treated with the Matti-Russe Technique. Long-Term

Results. Clin. Orthop., 214: 175-180, 1987.60. STEICHEN, J. B., and SCnRnmER, D. R.: Radial Bone Graft with Prolonged Imn~obilization for Scaphoid Nonunions. Contemp. Orthop., 12: 19-

24, June 1986.61. STEWART, M. J.: Fractures of the Carpal Navicular (Scaphoid). A Report of 436 Cases. J. Bone and Joint Surg., 36-A: 998-1006, Oct. 1954.62. SUTRO, C. J.: Treatment of Nonunion of the Carpal Navicular Bone. Surgery, 20: 536-540, 1946.63. SWANSON, A. B.: Silicone Rubber Implants for the Replacement of the C~?al Scaphoid and Lunate Bones. Orthop. Clin. North America, 1: 299-

309, 1970.64. SWANSON, A. B.: Implant Arthroplasty in the Hand and Upper Extremity and Its Future. Surg. Clin. North America, 61: 369-382, 1981.

hit.,

~(3):

~rpal

e and

Bone

’7.Term

!: 19-

54.

: 299-

TREATMENT OF UNUNITED FRACTURES OF THE SCAPHOID 991

65. TALEISNIK, JULIO: The Wrist, pp. 105-148. New York, Churchill Livingstc,ne, 1985.66. TALEISNIK, JULIO, and KELLY, P. J.: The Extraosseous and Intraosseous Blood Supply of the Scaphoid Bone. J. Bone and Joint Surg., 48-A:

1125-1137, Sept. 1966.67. TORNGREN, STAFFAN, and SANDQVIST, STURE: Pseudarthrosis in the Scaphoid Bone Treated by Grafting with Autogenous Bone-Peg. A Follow-

up Study. Acta Orthop. Scandinavica, 45: 82-88, 1974.68. U~GER, H. S., and STRYKER, W. C.: Nonunion of the Carpal Navicular. Analysis of 42 Cases Treated by the Russe Procedure. Southern Med.

J., 62: 620-622, 1969.69. VERDAN, CLAUDE, and NARAKAS, ALGIMANTAS: Fractures and Pseudarthrosis of the Scaphoid. Surg. Clin. North America, 48: 1083-1095, 1968.70. ZEMEL, N. P., and STARK, H. H.: Fractures and Dislocations of the Carpal Bones. Clin. Sports Med., 5: 709-724, 1986.71. ZEMEL, N. P.; STARK, H. H.; ASHWORTH, C. R.; RICKARD, T. A.; and .ANDERSON, D. R.: Treatment of Selected Patients with an Ununited

Fracture of the Proximal Part of the Scaphoid by Excision of the Fragment and Insertion of a Carved Silicone-Rubber Spacer. J. Bone and JointSurg., 66-A: 510-517, April 1984.