Wei et al.

The Journal ofHAND SURGERY

metacarpophalangeal joint and reconstruction in greattoe-to-hand free tissue transfers. J HAND SURe [Am]9:645-9, 1984

13. Lister GD, Kalismen M, Tsai TM: Reconstruction of the

hand with free microneurovascular toe-to-hand transfer:experience with 54 toe transfers. Plast Reconstr Surg71:372-84, 1983

Treatment of nonunited scaphoid fractures bypulsed electromagnetic field and cast

Thirty-five of 44 nonunited scaphoid fractures that were at least 6 months old healed in a meantime of 4.3 months during pulsed electromagnetic field (PEMF) treatment using external coilsand a thumb spica cast. The mean time from the onset of the fracture to treatment was 40months. No concurrent operation was performed. Follow-up time averaged 8.4 months. Eightof nine fractures with avascular necrosis healed. Five of eight fractures in the proximal third

healed. Twelve (75%) of 16 patients treated in short-arm thumb spica casts and PEMF healedversus 22 (92%) of 24 patients treated initially in long-arm thumb spica casts and PEMF. have found PEMF to be a reliable alternative ~aethod of treating nonunited scaphoid fractures.

Because of the low risk, simplicity of use, and reliability, we recommend its consideration in thetreatment of undisplaced, nonunited fractures without carpal instability less than 5 years afterthe injury. Treatment should initially begin with a long-arm cast. (J HAND SURG 11A:344-9,

1986.)

Gary K. Frykman, M.D., Julio Taleisnik, M.D., Gregory Peters, M.D.,

Richard Kaufman, M.D., Basil Helal, M.B., M.Ch., Virchel E. Wood, M.D., and

Robert S. Unsell, M.D., Loma Linda and Irvine, Calif., Grand Rapids, Mich.,Clevdand Heights, Ohio, and London, England

Nonunion of the scaphoid continues to

represent a challenge to the surgeon who treats handinjuries. There are many treatment modalities advo-cated, including benign neglect,~ casting,~" 2 wrist de-nervation, 3 bone grafting, 4" 5 internal fixation withcompression screw,6" v radial styloidectomy,s excisionof proximal pole,2’ 9. ~0 proximal row carpectomy,9" nscaphoid replacement arthroplasty,~2 wrist arthrodesis, ~a

or combined scaphoid replacement arthroplasty and

From the Hand Surgery Service, Department of Orthopedic Surgery,Loma Linda University School of Medicine, Loma Linda, Calif.,University of California, Irvine, Irvine College of Medicine, Ir-vine, Calif., and London Hospital, Whitechapel, London, En-gland. "

Received for publication Nov. 16, 1982; accepted in revised formJuly 4, 1985.

Reprint requests: Gary K. Frykman, M.D., Loma Linda UniversitySchool of Medicine, Loma Linda, CA 92350.

limited carpal arthrodesis. ~ No one method is suitablefor treatment of all nonunited scaphoid fracture~, there-fore, a variety of methods must be included in thesurgeon’s armamentarium.15" ~ Better results are a-

chieved if union can be obtained before the developmentof carpal collapse and posttraumatic arthrosis. 16

Although the mechanisms of action on tissues arenot fully understood,~7 electrical treatment of nonunitedfractures of long bones has been successful in about80% of patients. ~8-2~ The rate for successful union oftibias has been as high as 87%.z~ -

There are two widely used but different concepts forelectrical treatment of nonunions. One concept has a

constant, direct 20 microampere current delivered toone or more cathodes, which are implanted directly at

the fracture site, from either an "invasive" (the cath-ode, anode, and battery are implanted through a surgicalprocedure)ZO, 22 or "semi-invasive" system (cathodes

344 THE JOURNAL OF HAND SURGERY

Vol. 11A, No. 3May 1986 Nonunited scaphoid fractures treated by PEMF and cast 345

Fig. 1. This method places inductive electromagnetic coilson opposite sides of the fracture site. These coils supply a15 Hz pulsed quasirectangular electrical signal.

are drilled percutaneously across the fracture site, andthe anode is a pad on the skin surface).18-~ The secondconcept uses a totally noninvasive system that involvesplacement of two electromagnetic coils outside the castwith an alignment block incorporated into the cast tocenter the coils over the fracture site ~9’ 23 (Figs.1 and 2).

Although an equivalent success rate has been re-ported for treatment of nonunited fractures by each ofthe three methods, a distinct advantage of the pulsedelectromagnetic field (PEMF) is that it requires no sur-gical procedure, it can be performed in the office, andthere is no danger of introducing infection. ~9. 2~

The first report of treatment of scaphoid nonunionswith an electrical field used an AO compression screwand a parallel Kirschner wire inserted into the scaphoid;both were connected to a transformer, which generatedan electrical alternating current field. -~ A 79% unionrate was reported with this method, ~nd a 60% unionrate With the screw-only method. Bora et al. 26 reportedhealing of. 12 (71%) of 17 using tl~e semi-invasivemethod on fionunited scaphoid fractures. He stated thatavascular necrosis was a contraindication, but.that aprevious bone grafting procedure was not a contrain-dication.

This article presents a retrospective study designedto evaluate the efficacy of noninvasive PEMF in thetreatment of nonunited fractures of the scaphoid.

Materials and methods

We reviewed all of the cases of nonunited scaphoidfractures that we treated with the PEMF system* from1979 through 1984. We examined patients’ records andinformation was obtained by means of a standardizedprotocol. To be included in this study, the scaphoid

*Bi-osteogen, System Electro-biology Inc., Fairfield, N.J.

Fig. 2. The electromagnetic coils are attached by a Velcrostrap over a thumb splca cast and applied to a patient with ascaphoid nonunion. The box transforms 110 volt AC linecurre~at to the proper pulsed signal. The patient wears theapparatus 8 to I0 hours a day.

fracVare had to be nonunited and at least 6 months old,regardless of previous treatment methods, with no sur-gical procedures having been performed during or justbefo:ce initiation of PEMF treatment. None of our pa-tients had collapse of the scaphoid or degeneration ofthe periscaphoid joints. Three patients showed loss ofcarpal alignment. This series represents consecutive--cases treated by us following a standardized protocol.~7

PEMF coils were centered over the scaphoid (Fig. 2)and attached to either a long-arm thumb spica cast ora short-arm thumb spica cast. "

¯ There were 50 patients initially treated. Three pa-tients failed to return for follow-up examination afterimmobilization was stopped and the electrical treatmentcoils were removed. Although their fractures had healedthese patients were excluded from our final study. Therewere three additional patients who received inadequatetreatment of 2 months or less (this did not conform toour protocol), and they were also excluded from thestudy. All of the remaining 44 patients were followeduntil their fractures healed, or until this method of treat-ment failed. There were 41 male and 3 female patientsin this group. Their ages ranged from 14 to 46 years,

346 Frykman et al.The Journal of

HAND SURGERY

Fig. 3. Uncomplicated healing of nonunited scaphoid with PEMF. A, Unhealed proximal thirdfracture 6 months after injury. B, Eight months after fracture when PEMF began. C, Appearance3 months after beginning PEMF. D, Eight months later.

with a mean age of 25 years. In 8 patients the fracturewas localized to the proximal third of the scaphoid, in33 patients to the middle third, and in 3 patients to thedistal third. Nine patients had had one previous bonegraft and two patients had two bone-grafting proceduresbefore PEMF treatment was initiated. Nine patients hadx-ray evidence of avascular changes of the proximalfragment. The length of time from injury to PEMFtreatment ranged from 6 to 241 months with a mean of40 months. All of the patients who were included inour final study were examined from 1 to 33 months,with a mean of 8.4 months, after immobilization andelectrical treatment was stopped.

Results

The fractures healed in 35(80%) of the 44 patients,as shown in Fig. 3. Five (62.5%) of the eight fracturesin the proximal third of the scaphoid healed and 30(83%) of the 36 fractures in the middle and distal third,,;healed with electrical treatment. Fractures of eight(89%) of the nine patients who had avascular necrosisbefore treatment healed. Eight (73%) of the eleven pa-tients who had undergone previous bone-grafting pro-cedures healed. The mean duration of electrical treat-ment and casting was 4.3 months, with a range of 2.5to 9 months. Only four patients had PEMF treatmentfor longer than 6 months.

Vol. 11A, No. 3May t986 Nonunited scaphoid fractures treated by PEMF and cast 347

Fig. 4. A, Scaphoid nonunion of 96 months’ duration in distal third. B, Synovial pseudarthrosis,proved by arthrograms, after fracture did not heal by PEMF. Synovial pseudarthrosis is a contrain-dication to PEMF treatment.

In four patients, records failed to indicate whetherabove-elbow or below-elbow immobilization was used.Of the remaining 40 patients, healing occurred in 12(75%) of the 16 who were treated in a short-arm thumbspica cast and in 22 (92%) of the 24 patients who weretreated initially in a long-arm thumb spica cast. Theduration of long-arm cast immobilization ranged from4 to 20 weeks, with a median of 6 weeks and wasfollowed by below-elbow immobilization for the bal-ance of the PEMF treatment.

Of the 35 patients with healed fractures, 33 returnedto their pre-injury work, 11 to heavy labor, 13 to mod-erately strenuous work, and 9 to light work. One patientwho performed heavy labor was not able to resume hispreviousoccupation, and in one patient this informationwas unknown.

At follow=up on the patients with healed fractures,wrist extension had a mean of 85.3%; flexion, 93.1%;radial deviation, 85.0%; and ulnar deviation, 91.1% ofthe opposite normal wrist. Grip strength at follow-upaveraged 83% of the opposite hand.

Analysis of failures

Of the nine patients whose fractures failed to heal,one had a synovial pseudarthrosis of the distal third ofthe scaphoid (that was proven by an arthrogram), 132months after the fracture (Fig. 4). Three patients haddorsal intercalated segment instability (DISI) pattern(shown on x-ray films) and should not have been con-sidered for this type of treatment. Two patients showed

HEALING OF SCAPHOID FRACTURES WITH PEMFVS. TIME SINCE FRACTURE

Healed

O0 ¯O0 0o0 Oo ¯

¯ 0000 ¯ 0o0 ¯ ¯ ¯0 0 0 0 0 00 0

Not Healed

__ l ,~4 I , g6 12 2 48 60 6 192 240

Time in Months

Fig. 5..The effect of pretreatment duration of scaphoid frac-ture on healing with PEMF.

poor cooperation and removed their own casts morethan once during treatment. There were only three fail---ures among the patients who conformed to the criteriafor inclusion in this study.

The fractures that failed to unite with PEMF had amean duration of fracture of 81 month~ (range 6 to 241months) versus a mean of 27 months for the fracturesthat healed. Fig. 5 shows that in fractures that weremore than 60 months old, only 2 (29%) of 7 healed,but 27 (90%) of 30 of fractures that were less than months old healed. This difference is statistically sig-nificant (p < 0.1301).

Discussion

Although there is some controversy in the literatureregarding the need to treat asymptomatic nonunions ofthe scaphoid,4" 16 recent reports indicate that the longer

348 Frykman et al.The Journal of

HAND SURGERY

the duration of the nonunion, the greater the degener-ative changes about the scaphoid.28’ 29 Mack et al.2s

suggest that "asymptomatic patients with an undis-placed stable nonunion should be advised of the pos-sibility of late degenerative changes." This study ofPEMF treatment was designed for this type of patient.

We believe that the main problem in objectively eval-uating the efficacy of PEMF treatment is the lack of areliable double-blind study between like series of pa-tients treated with immobilization alone, and thosetreated with immobilization and PEMF. However, thereare too many other variables for such a study to beaccurate. Instead, in our opinion, the patients in thesestudies themselves provided the comparison for the ef-ficacy of PEMF since in many cases, immobilizationalone was used unsuccessfully. In our series, withPEMF and casting, healing occurred in an average of4.3 months. In addition, a meaningful comparison maybe made between our patients and a reported series ofpatients with nonunited scaphoid fractures that weretreated with immobilization alone. Stewart2 reportedsuch a series. Ninety scaphoid fractures, which weremore than 1 month old, were treated with casting alone;they had a healing rate of 57% and a "probable" heal-ing rate (i.e., lost to follow-up) of 73%. The averagelength of immobilization was 20 weeks. Some patientswere treated with immobilization for as long as 59weeks. Immobilization of the fracture began an averageof 5 months after the injury. Although Stewart’s seriesand our series were not strictly comparable, the unionrate of our patients’ fractures was considerably greater(80% versus 57%). This becomes significant becauseof the longer time from injury to treatment in our series,40 months versus 5 months. Finally, union occurredslightly faster in our series, a mean of 4.3 months versus4.7 months.

Our results with PEMF are comparable to bone graft-ing where union rates of 50% to 100% have been re-ported,:4, 5, ~5, 30 However, PEMF may not be quite aseffective as the Russe bone graft technique; a recentreview of the literature reported it to lead to union onan average of 90% of patients.3x Our results are betterthan those reported following open reduction and in-ternal fixation with compression screw osteosynthesis;the latter technique has a reported success rate of 17 to73%.~. 15

The above-elbow immobilization clearly seems toenhance the success rate of PEMF treatment. This is

¯ confirmed by Beckenbaugh,32 who reported that healingoccurred in only 7 of 10 patients who were treated in

short-arm casts versus 12 (87%) of 14 of patients treatedin long-arm immobilization with the addition of PEMF.

Final grip strength in our patients (83% of normal)is comparable to that obtained following direct electricalcurrent treatment (76%)26 or bone grafting (85%)3x. Theoverall mean range of motion of the wrist in our patientswas 89% of normal. This was much greater than thatre]ported using the semiinvasive method (50% of wristdc,rsiflexion and 42% of wrist palmar flexion).26 Thus,preservation of motion with noninvasive techniquesseems to provide a distinct advantage over the semiin-vasive method.

From our experience, PEMF treatment is contrain-dicated for displaced fractures, for synovial pseudarth-rosis,a3 in the presence of periscaphoid osteoarthritis,artd for wrists that exhibit a loss of normal radiocarpalalignment. Adherence to these contraindications wouldhave eliminated four out of the nine patients in ourfailure group.

Our results also suggest that nonunited scaphoid frac-tures that are older than 5 years are much less likely tounite with PEMF. Other techniques are recommendedfc,r these fractures. Finally, patient selection is impor-tant, because this method requires the patient’s coop-eration in order to be successful. A newer portable,battery-operated unit is now available* and should al-low for easier patient compliance.

The advantages of PEMF treatment are evident: nopain, no anesthetic, no surgical risk, and no knowncomplications. The cost of this treatment is less thantl~e combined cost of hospitalization, anesthetic, andsurgical fees. The length of treatment is comparable totk~at following bone grafting procedures. We believe thatPEMF treatment should be specifically considered asan alternative method for treatment when previous sur-gery has failed, for problem fractures in the proxiNalthird of the scaphoid, and whenever surgery is not safeor is rejected by the patient.

;"*EBI Bone Healing Syste,m Model 420,~EBI Medical Systems, Fair-

field, N.J.

REFERENCES

1. London PS: The broken scaphoid bone. The case againstpessimism: J Bone Joint Surg [Am] 43:237-4~-, 1961

2. Stewart M J: Fractures of the carpal navicular (scaphoid).A report of 436 cases. J Bone Joint Surg [Am] 36:998-1006, 1954

3. Buck-Gramcko D: Denervation of the wrist. J HAlOSofia 2:54-61, 1977

Vol. llA, No. 3May 1986 Nonunited scaphoid fractures treated by PEMF and cast 349

4. Russo O: Fracture of the carpal navicular. Diagnosis,non-operative treatment, and operative treatment. J BoneJoint Surg [Am] 42:759-68, 1960

5. Herness D, Posner MA: Some aspects of bone graftingfor nonunion of the carpal navicular. Acta Orthop Scand48:373-8, 1977

6. Maudsley RH, Chen SC: Screw fixation in the manage-ment of the fractured carpal scaphoid. J Bone Joint Surg[Br] 54:432-41, 1972

7. Herbert TJ, Fisher WE: Management of the fracturedscaphoid using a new bone screw. J Bone Joint Surg66:114-23, 1984

8. Mazet R, Hohl M: Fractures of the carpal navicular. JBone Joint Surg [Am] 45:82-112, 1963

9. Taleisnik J: Fractures of the carpal bones. In Green DP,editor: Operative hand surgery, New York, 1982, Chur-chill Livingstone, pp 669-702

10. Zemel NP, Stark HH, Ashworth CR, Rickard TA, An-derson DR: Treatment of selected patients with an un-united fracture of the proximal part of the scaphoid byexcision of the fragment and insertion of a carved silicone- rubber spacer. J Bone Joint Surg [Am] 66:510-7, 1984

11. Inglis AE, Jones EC: Proximal-row carpectomy for dis-eases of the proximal row. J Bone Joint Surg [Am]59:460-3, 1977

12. Swanson AB: Silicone rubber implants for the replace-ment of the carpal scaphoid and lunate bones. OrthopClin North Am 1:299, 1970

13. Gordon LH, King D: Partial wrist arthrodesis for oldununited fractures of the carpal navicular. Am J Surg102:460-4, 1961

14. Watson HK, Goodman ML, Johnson TR: Limited wristarthrodesis. Part II: Intercarpal and radiocarpal combi-nations. J HAtqo StJRG 6:223-33, 1981

15. Cooney WP, Dobyns JH, Linscheid RL: Nonunion of thescaphoid: Analysis of the results from bone grafting. JHArqo StJg6 5:343-53, 1980

16. Kleinert JM, Zenni EJ: Nonunion of the scaphoid. Re-’,Jew of the literature and current treatment. OrthopaedicReview 13:125-40, 1984

17. Black J: Tissue response to exogenous electromagneticsignals. Orthop Clin North Am 15:15-31, 1984

18. Brighton CT, Black J, Friederiberg ZB, Esterhai JL, DayL J, Connolly JF: A multicenter study of the treatment ofnon-umon with constant direct current. J Bone Joint Surg[Am] 63:2-13, 1981

19. Ba:~sett CAL, Mitchell SN, Gaston SR: Pulsing electro-magnetic field treatment in ununited fractures and failedartlarodeses. JAMA 247:623-8, 1982

20. Pa’terson D, Lewis GN, Cass CA: Clinical experience inAustralia with an implanted bone growth stimulator(1976-1978). Orthop Trans 3:288, 1979

2 I. Brighton CT: The semi-invasive method of treating non-union with direct current. Orthop Clin North Am 15:33-45, 1984

22. Pzterson D: Treatment of nonunion with a constant directcu.rrent: A totally implantable system. Orthop Clin NorthAm 15:47-59, 1984

23. Bassett CAL: The development and application of pulsedelectromagnetic fields (PEMFs) for ununited fracturesand arthrodeses. Orthop Clin North Am 15:61-87, 1984

24. Bassett CAL, Mitchell SN, Gaston SR: Treatment ofttaunited tibial fractures with pulsing electromagneticfields. J Bone Joint Surg [Am] 63:511-23, 1981

25. Wilhelm K: Die ergebnisse der behandlund der navicu-lare-pseudarthrose mit magnetischen und elektrischen po-tentalien. Medizinische Orthopadische Technik 2:67-70,1978

26. Bora FW, Osterman AL, Woodbury DF, Brighton CT:Treatment of nonunion of the scaphoid by direct current.Orthop Clin North Am 15:107-12, 1984

27. Electro-Biology, Inc., Fairfield, N.J.: Procedure manualfor the prescribing, application, and management of theEBI Bi-Osteogen System 204, May, 1979.

28. Mack GR, Bosse MJ, Gelberman RH, Yu E: The naturalaistory of scaphoid nonunions. J Bone Joint Surg [Am]66:504-9, 1984

29. Ruby LK, Stinson J, Belsky MR: The natural history ofscaphoid nonunion. A review of fifty-five cases. J BoneJoint Surg [Am] 67:428-32, 1985

30. Schneider LH, Aulicino P: Nonunion of the carpal scaph-oid: The Russe procedure. J Trauma 22:315-9, 1982

31. Rasmussen P, Schwab JP, Johnson PR: Symptomaticscaphoid nonunions treated by Russe bone grafting. Or-thopaedic Review 14:41-7, 1985

32. Beckenbaugh RD: Noninvasive pulsed electromagnetic_stimulation in the treatment of scaphoid nonunions. Pre-sented at the American Society for Surgery of the HandAnnual Meeting, Las Vegas, Nev. Jan. 21, 1985

)