FRACTURE OF THE DISTAL RADIUS

A PROSPECTIVE COMPARISON BETWEEN TRANS-STYLOID AND KAPANDJI FIXATIONS

E. LENOBLE, C. DUMONTIER, D. GOUTALLIER, A. APOIL

From the Henri Mondor Hospital, Creteil and the Saint Antoine Hospital, Paris, France

We performed a prospective study on 96 patients with Fractures of the distal radius represent one-sixth of all
extra-articular or intra-articular fractures of the distal fractures in patients over 50 years of age (Owen et al 1982;
radius with a dorsally displaced posteromedial Jupiter 1991), but there is surprising disagreement on clas
fragment. After closed reduction, we compared sification (Gartland and Werley 1951; Lidstrom 1959; Cas
trans-styloid fixation and immobilisation with Kapandji taing 1964; Frykman 1967; Jenkins 1989), treatment
fixation and early mobilisation. Forty-two patients of (DePalma 1952; Scheck 1962; Pool 1973; Kapandji 1976;
mean age 57.1 years ± 18.1 (SD) were treated by Cooney, Linscheid and Dobyns 1979; Sarmiento, Zagorski
trans-styloid K-wire fixation and 45 days of short-arm and Sinclair 1980; Clancey 1984; Dias et al 1987), and
cast immobilisation. Fifty-four patients of mean age correlation between radiological and functional outcome
57.7 years ± 18.7 (SD) had Kapandji fixation and (Cassebaum 1950; Cooney, Dobyns and Linscheid 1980;
immediate mobiisation according to the originator. Clancey 1984; Jupiter 1991).
All the patients had clinical and radiological review at The commonest operative method of treatment is percu
about six weeks and at 3, 6, 12 and 24 months after the taneous Kirschner (K)-wire fixation except for comminuted
operation. Pain, range of movement and grip strength fractures which are managed by external fixation (Cooney
were tested clinically, and changes in dorsal tilt, radial et al 1979; Weber and Szabo 1986; Jenkins et al 1987). A
tilt, ulnar variance, and radial shortening were assessed common sequel of K-wire fixation and postoperative
radiologically. Statistical analysis was applied to immobilisation is stiffness but Py (personal communica
comparisons with the normal opposite wrist. tion, 1969) described a method which required no immob
Pain and reflex sympathetic dystrophy were more ilisation, and good clinical and radiological results were
frequent after Kapandji fixation and early mobilisation, reported by Desmanet (1989). Kapandji (1976) reintro
but the range of motion was better although this duced the concept of immediate mobilisation and presented
became statistically insignificant after six weeks. The his technique of intrafocal K-wire fixation, claiming that
radiological reduction was better soon after Kapandji this encouraged early use of the hand, and reduced the
fixation, but there was some loss of reduction and incidence of reflex sympathetic dystrophy (RSD). Epinette
increased radial shortening during the first three et al (1982) reported good results in the short term, and in
postoperative months. The clinical result at two years 1985 Kapandji introduced a number of technical modifica
was similar in both groups. tions but did not report his results (Kapandji 1987).
J Bone Joint Surg (Br) 1995;77-B:562-7.
Few prospective studies have compared the outcome of
Received 4 August 1994; Accepted after revision 2 November 1994 K-wire fixation with different durations of immobilisation
(Pool 1973; Dias et al 1987; McAuliffe et al 1987). We
aimed to compare trans-styloid and Kapandji K-wire fixa
tion and postoperative management. We must stress that
our purpose was not to compare the two methods sepa
rately, but to assess as a whole the results of the two most
commonly used methods in France for posteriorly dis
placed distal radial fractures.
E. Lenoble, MD, Chief Clinical Assistant
D. Goutallier, MD, Professor of Orthopaedic Surgery
Department of Orthopaedic Surgery and Trauma, Henri Mondor Hospital, PATIENTS AND METHODS
51 Avenue du Maréchalde Lattre de Tassigny, 94010 Creteil Cedex,
France. We initially recruited 120 patients excluding those with
C. Dumontier, MD, Consultant Orthopaedic Surgeon anteriorly displaced fractures, fracture-dislocations, open
A. Apoil, MD, Professor of Orthopaedic Surgery
Department of Orthopaedic Surgery and Trauma, Saint Antoine Hospital, fractures, multiple trauma, or previous fractures of either
Paris, France. wrist or elbow which would preclude comparison. We also
Correspondence should be sent to Dr E. Lenoble. excluded comminuted fractures which could not be stabi
©1995British Editorial Society of Bone and Joint Surgery lised by K-wire fixation and fractures with more than two
030l-620X195/41001 $2.00 intra-articular fragments. Of the 120 patients, 15 were lost

562 ThE JOURNALOF BONE AND JOINTSURGERY

 FRACTURE OF THE DISTAL RADIUS 563

Table I. Details of the fracture classifications ineach

groupClassificationFhafion‘frans-styloldKapandjiTotalCastaingExtra-articular,

displaced172542Intra-articular,
displaced,252954posteromedial
fragmentTotal425496FrykmanExtra-articular1/2131730Radiocarpal

involvement3/4172643Distal
joint
involvement5/64812Radiocarpal
radio-ulnar joint
radio-ulnar7/88311joint
and distal
involvementTotal425496AOExtra-articular,

comminutionA2111829Extra-articular,
dorsal
volarA36713comminutionIntra-articular,
dorsal and

comminutionCl131528Intra-articular,
no
comminutionC2121426Total425496

Fig. 2

Fig. 1 Kapandji fixation of an extra-articular fracture of the distal radius. Two

1.8 mm K-wires are introduced by hand through the fracture line, one
Trans-styloid fixation of an extra-articular fracture of the distal radius. lateral and one posterior. These are initially directed at 90°to the axis of
Two 1.8 nun K-wires are inserted, using a powered driver, through the the radius and then redirected at 45°and impacted into the opposite
radial styloid anteriorly and posteriorly to reach the opposite cortex. cortex.

to follow-up, three died before completion of the study, and
six were excluded because of missing data. The 96 remain
ing patients were all skeletally mature with no other age
limit, with closed posteriorly displaced, extra-articular or
intra-articular distal radial fractures with a posteromedial
fragment.
The fractures were classified by the methods of Castaing
(1964) and Frykman (1967), and by the AO system (Table
I) on the basis of prereduction posteroanterior and lateral
films and axial traction radiographs, taken under
thesia to determine the feasibility of reduction and the
extent of fracture more clearly. The choice of method of K-
wire fixation was determined by simple random
the operating theatre. Of the 42 patients in the trans-styloid
group, 13 were male (31%), 38 were right-handed (9 1%)
and in 15 (36%) the fracture was of the dominant limb. The
average age at injury was 57.1 years ±18.1 (SD) (18 to 88).
Of the 54 patients in the Kapandji group, 18 were male
(33%), 52 were right-handed (96%), and 28 had their injury
in the dominant limb (52%). The average age was 57.7 ±
18.7 (SD)(20 to 82).
The operation was performed under regional or general
anaesthesia, with radiological control of reduction and K-
wire positioning.
anaes Operative techniques. When trans-styloid fixation had
been selected, reduction was by traction in the axis of the
second and third digits with care taken to avoid flexion to
selection in protect the anterior cortex. 1\wo 1.8 mm K-wires were
inserted by a powered driver through the radial styloid
anteriorly and posteriorly, to reach the opposite cortex (Fig.

VOL 77-B, No. 4, JULY 1995

 564 E. LENOBLE, C. DUMONTIER, D. GOUTALLIER, A. APOIL

@ .—@.

Fig. 3

1@ A distal radial fracture treated by the Kapandji

method.

1). After operation, the wrist was immobilised in a short
arm plaster for 45 days when the wires were removed
local or regional anaesthesia.
When the method of Kapandji had been selected, the
fracture was reduced by manual traction and 1 cm skin
incisions were made. Vessels, nerves and tendons were retrac
ted, and one lateral and one posterior 1.8 mm K-wires were
introduced by hand at the fracture line (Kapandji 1976). They
were initially directed at 900 to the axis of the radius and
under then, at the middle of the metaphysis, redirected at 45°and
advanced to impact into the opposite cortex (Figs 2 and 3).
Any posteromedial fragment was secured by a third K-wire
inserted more medially in the same way. The K-wires were
not bent over, but all were buried under the skin. The wrist
was mobilised immediately postoperatively. K-wires were
removed at 45 days under local or regional anaesthesia.

Fig.4

Radiological indices measured on posteroanterior and lateral films (R = radius axis; S = scaphoid axis; L = lunate axis; M = third metacarpal
length; H = height of the campus; RS = radial shortening; RT = radial tilt; DT = dorsal tilt; SLA = scapholunate angle; RLA = radiolunate angle;
UV = ulnar variance).

ThE JOURNALOF BONE AND JOINTSURGERY

@ @;

FRAC'rURE OF THE DISTAL RADIUS 565

40 and Cassebaum 1965; Rubinovich and Renme 1983). We

@ .@ . I KAPANDJI 0 @s-sm.oc also measured the scapholunate space, ulnar and radial
@.3m ,@
@ .@ :@• 20.5 translation and the radiolunate and scapholunate angles
(Fig. 4). We distinguished between failures of reduction
@.. zl i-.,. — .‘.-‘ 194 18.7
and secondary displacement, and at the final follow-up
@ @)18 13.4
looked for any signs of osteoarthritis.
We used Student's t-test to compare averages or percen
tages in two samples, and analysis of variance where
@ @1; !± 8mth1@th24mthL appropriate.
Follow-up
RESULTS
Fig.5
After trans-styloid fixation, patients were immobilised for
Levels of pain reported by patients on a 0 to 100 point scale. an average of 42.9 ±8 days (37 to 54). The K-wires were
removed at an average of 44.4 ±3.4 days (37 to 54).
Following Kapandji fixation, K-wires were removed at 46.7
±4.5 days (40 to 60).
Both methods were conducted under fluoroscopic con More patients treated by the method of Kapandji had
trol, and the stability of the reduction and fixation was pain but the difference was statistically insignificant (Fig.
tested passively. AU patients had a stable fixation. The 5). All movements had recovered a better range at 45 days
criteria for satisfactory reduction were determined by com in the Kapandji group (Table II). This difference persisted
parison with the normal wrist: there should be no articular after 45 days but was no longer statistically significant.
step, congruency of the distal radio-ulnar joint, and no Grip and pinch were also slighfly better in the Kapandji
more than 2°of tilt in either plane. group up to 12 months but the difference was not statis
Review. Pain was recorded by the patient on a scale of 0 tically significant.
(no pain) to 100 (unbearable pain) and range of movement There was better reduction in both radial (Fig. 6) and
in all planes was measured by goniometer. Grip and pinch palmar (Fig. 7) tilt in the Kapandji group, immediate
strength was measured by Jamar dynamometers. Both postoperative volar tilt being greater than on the non
range of movement and strength were recorded as percen injured side. Some loss of reduction occurred in both
tages of those in the non-injured limb without allowance groups during the first three months and was worse in the
for dominance. RSD was diagnosed clinically and radio Kapandji group. Ulnar variance (Fig. 8) was initially
logically and confirmed by isotope studies. restored in both groups, but at 45 days the variance was
Radiographs were taken immediately after operation and positive and it increased in value up to the third month.
at 45 days and 3, 6, 12 and 24 months, with posteroanterior This increase was statistically more significant in the
and lateral films (Meyrueis 1984; Palmer 1987). The mdi Kapandji group.
ces described by Youm et al (1978) were calculated, with In the Kapandji group six fractures had been over
frontal and sagittal inclination, the radio-ulnar index and reduced by 15°in an anterior direction. In one the anterior
radial shortening (Gartland and Werley 195 1; Older, Stabler tilt became worse, but in the other five, two converted to a

259.@.8

24
8'
@0
23

0
.@
22

0
8' 21

p0*- 45daya 3months 6monthel2mo,th. 24month. pod 45day. 3months 6month. l2month. 24months
operativeop.ra@ve

Follow-up Follow-up

Fig. 6 Fig.7

Changes in the average radial tilt in degrees. The opposite normal side Changes in the average palmar tilt in degrees. The opposite normal side
averaged 23.10 in the Kapandji series and 23.8° in the trans-styloid averaged 8.2° in the Kapandji series and 9.10 in the trans-styloid
series. patients.

VOL 77-B, No. 4, JULY 1995

566 E. LENOBLE, C. DUMONTIER, D. GOUTALLIER, A. APOIL

E aftertreatment
Table II. Wrist and hand function at intervals
E thepercentage
by trans-styloid and Kapandji methods as
wristFixationFunction
of that in the uninjured

C Trans-styloidMovement Kapandji
. TRAP4S-STYLOID weeksFlexion
at 6 to 8
46Extension 59
0
44Radial 58
45day. 3month. Smooth. l2months 24months
38Ulnardeviation 60
Follow-up 52Pronation
deviation 61
-1
62Supination 80
64Gripat6weeks 76

Fig. 8
Change in the average ulnar variance in nun. The opposite normal side
averaged —0.8mm in the Kapandji series and —0.9mm in the trans-styloid
series.
posterior tilt. There were two over-reductions in the trans
styloid group, but neither became worse. One case in each
group had a step in the articular surface of less than 2 mm.
There were four cases of DISI and seven of VlSI dis
tributed evenly through the two groups. Four wrists showed
early osteoarthritis which predated the injury. There was no
statistical difference in any other radiological index. The
presence of a posteromedial intra-articular fragment did not
influence either the clinical or the radiological results.
There were more complications after Kapandji fixation,
with eight patients having symptoms related to the sensory
branch of the radial nerve in this group, as against three in
the trans-styloid group. These symptoms always became
apparent after removal of the pins and persisted at 24
months. The same ratio of 8:3 also applied to bone-scan
confirmed RSD. Two of the patients with problems with the
radial nerve developed RSD. One fracture displaced after
trans-styloid fixation, requiring the application of an extern
al fixator.
There were no tendon or vascular complications and no
dysfunctions of the median nerve. There was superficial
infection of pin tracks in three Kapandji and one trans
styloid patient. This was always around K-wires which had
not been buried enough or which emerged through the skin
later and all resolved spontaneously after the planned
removal of the K-wires.
DISCUSSION
Range of movement was statistically better after Kapandji
fixation until a short period after the K-wires had been
removed and occurred at the cost of increased pain; this
corresponded to the time of immobilisation of the trans
styloid group. Forty-five days after cast removal, this
advantage became statistically insignificant and did not
influence the final range of movement. The increased pain
after the Kapandji procedure appeared to be due in part to
the early mobilisation, and it seems that postoperative
immobilisation for 10 to 15 days could be helpful, as
proposed by McAuliffe et al (1987).
37at 42
56atlyear
3 months 60
83Pinchat6weeks 84
56at 60
70atlyear
3 months 72
89 87
Long-term radiological results were identical, except for
ulnar variance, which was positive in both groups, but
greater on average after the Kapandji procedure. Although
there was more excessive anterior reduction with Kapandji
fixation, loss of reduction in the sagittal plane was also
more common in this group and these combined to give
similar average long-term radiological results. The intra
focal location of the Kapandji K-wires does not appear able
to resist the forces induced by early mobilisation or to
avoid secondary displacement.
Both radial-nerve lesions and RSD were more common
after the Kapandji procedure. The radial-nerve complica
tions were related to K-wire removal rather than to early
motion or scar formation. At the time of removal of the K-
wires, care may be needed to free branches of the radial
nerve which are trapped in scar tissues. In trans-styloid
fixation the K-wires are more distal and away from bran
ches of the radial nerve. RSD seemed to be related to pain
during early mobilisation and to the presence of radial
nerve lesions.
Some believe that distal radial fractures have a good
outcome regardless of the treatment (Peltier 1984), but in
common with others we believe that the outcome varies
with the type of fracture and the treatment (Cassebaum
1950; Green 1975; Fisk 1980). Most surgeons advocate
reduction of the fracture, but we question whether this is
always necessary, particularly in elderly patients with a
restricted functional requirement (Palmer 1988).
The results of closed reduction and cast immobilisation
of distal radial fractures are poor in 13% to 37% of cases
(Gartland and Werley 195 1; Castaing 1964; Frykman 1967;
Altissimi et al 1986), and we prefer to fix the fracture with
percutaneous K-wires after closed reduction. DePalma
(1952) defended ulnoradial percutaneous K-wire fixation
and Castaing (1964) advised passing the wire into the ulna,
but this blocks forearm rotation and could cause cross

THE JOURNAL OF BONE AND JOINT SURGERY

 FRACTURE OF THE DISTAL RADiUS
union. Kapandji (1976) proposed the intrafocal placement
of K-wires to allow early mobilisation with little risk of
secondary displacement. Over 80% of good results were
reported with this technique by Epinette et al (1982). We
found more radial shortening after Kapandji fixation, but
were unable to demonstrate any early clinical or radio
logical disadvantages, although radial shortening and dorsal
tilt both increase the pressure on the triangular fibro
cartilage and painful ulnocarpal impingement may develop
later (Palmer 1987; Short et al 1987).
Kapandji fixation improves the immediate clinical result,
but this advantage disappears later, and the early post
operative mobilisation appears to increase pain. The final
results of the two methods which we studied are identical in
terms of dorsal, volar, and radial tilt, but radial shortening
was better controlled by trans-styloid fixation. Lesions of
the radial nerve and RSD were more frequently seen after
Kapandji fixation and removal of the wires must be per
formed with great care.
No benefits in any form have been received or will be received from a
commercial party related directly or indirectly to the subject of this
article.
REFERENCES
Altissimi M, Antenucci R, Places C, Mancini GB. Long-term results of
conservative treatment of fractures of the distal radius. Clin Orthop
1986;206:202-10.
Cassebaum WH. Colles' fracture: a study of end results. JAm Med Assn
1950;l43:963-5.
Castaing J. Lea fractures récentes
de l'extrémité
inférieure
du radius chez
l'adulte. Rev Chir Orthop 1964;50:58l-696.
Clancey GJ. Percutaneous Kirschner-wire fixation of Colles' fractures: a
prospective study of thirty cases. J Bone Joint Surg (Am)
1984;66-A:1008-14.
Cooney WP ifi, Dobyns JH, Llnscheid RL. Complications of Colles'
fractures. J Bone Joint Surg (Am] 1980;62-A:613- 9.
Cooney WP ifi, Llnscheid RL, Dobyns JR. External pin fixation for
unstable Colles' fractures. J Bone Joint Surg (Am] l979;61-A:840-5.
DePalma AF. Comminuted fractures of the distal end of the radius treated
by ulnar pinning. J Bone Joint Surg (Am] l952;34-A:65l-62.
Dias JJ, Wray CC, Jones JM, Gregg PJ. The value of early mobilisation
in the treatment of Colles' fractures. J Bone Joint Surg (Br]
1987;69-B:463-7.
Desmanet E. L'ostéosynthèse
par double embrochage souple du radius:
traitement functionnel du fractures de l'éxtr@mité
inférieuredu radius:
a proposdun série
de 130cas.AnnChirMain 1989;8:193-206.
Epinette JA, Lehut JM, Cavenaile M, Bouretz JC, Decoulx J. Fracwre
de Pouteau-Colles: double embrochage intra-focal en berceau selon
Kapandji: a propos d'une sériehomogène de soixante-douze cas. Ann
ChirMain 1982;l:71-83.
Fisk GR. An overview of injuries of the WriSL Clin Orthop
1980;149:137-44.
VOL 77-B, No. 4, JULY 1995
567
Frykman G. Fracture of the distal radius including sequelae-shoulder
hand-finger syndrome, disturbance in the distal radio-ulnar joint and
impairment of nerve function: a clinical and experimental study. Acta
Orthop Scand 1967;Suppl 108.
Gartland JJ Jr, Werley CW. Evaluation of healed Colles' fractures. J
Bone Joint Surg (Am) 195 l;33-A:895-907.
Green DP. Pins and plaster treatment of comminuted fractures of the
distal end of the radius. J Bone Joint Surg (Am] 1975;57-A:304-10.
Jenkins NH. The unstable Colles' fracture. J Hand Surg (Br)
1989;14:149-54.
Jenkins Nil, Jones DG, Johnson SR, Mlntowt-Czyz Wi. External
fixation of Colles' fractures: an anatomical study. J Bone Joint Surg
(Br] 1987;69-B:207-1 1.
Jupiter JB. Current concepts review. Fractures of the distal end of the
radius. J Bone Joint Surg [Am) 199l;73-A:461-9.
Kapandji A. L'osteosynthèse par double embrochage intra-focal: traite
ment fonctionneldes fracturesnon-articulairesde l'extrémité
infér
ieuredu radius.Ann Chir 1976;30:903-8.
Kapandji A. L'embrochage intra-focal des fractures de l'extrémité
infer
ieure du radius dix ans aprés.
Ann Chir Main 1987;6:57-63.
Lidstrom A. Fractures of the distal end of the radius: a clinical and
statistical study of end results. Acta Orthop Scand l959:Suppl 41.
McAullffe TB, Hilliar KM, CoMes CJ, Grange WJ. Early mobilisation
of Colles' fractures: a prospective trial. J Bone Joint Surg (Br)
1987;69-B:727-9.
Meyruels ,JP. Instabilité
du carpe: circonstancesde diagnosticet étude
clinique. Ann Chir Main 1984;3:313-6.
Older TM, Stabler EV,CassebaumWH. Colles'fracture:evaluationand
selection of therapy. J Trauma 1965;5:469-76.
Owen RA, Melton LI ifi, Johnson KA, Ustrup DM, Riggs BL
Incidence of Colles' fracture in a North American Community. Am J
Public Health 1982;72:605-7.
Palmer AK. The distal radioulnar joint: anatomy biomechanics and
triangular fibrocartilage complex abnormalities. Hand Clin
l987;3:31-40.
Palmer AK. Fractures of the distal radius. In: Green DP, ed. Operative
hand surgery. New York, etc: Churchill Livingstone, 1988:991-1026.
Peltier LF. Fractures of the distal end of the radius: an historical account.
Clin On/mop 1984;187:l8-22.
Pool C. Colles fracture: a prospective study of treatment. J Bone Joint
Surg (Br] 1973;55-B:540-4.
Rubinovich RM, Rennie WR. Colles' fracture: end results in relation to
radiologic parameters. Can J Surg 1983;26:361-3.
Sarmiento A, Zagorski JB, Sinclalr WF. Functional bracing of Colles'
fractures: a prospective study of immobilisation in supination vs
pronation. CliiiOrthop 1980;146:175-83.
Scheck M. Long-term follow-up of treatment of commninuted fractures of
the distal end of the radius by transfixation with Kirschner wires and
cast.J BoneJoint Surg[Am] l962;44-A:337-5l.
Short WH, Palmer AK, Werner FW, Murphy DJ. A biomechanical
study of distal radial fractures. J Hand Surg (Am] l987;12-A:529-
34.
Weber SC, Szabo RM. Severely comminuted distal radial fracture as an
unsolved problem: complications associated with external fixation and
pins and plaster techniques. J Hand Surg (Am] 1986;! l-A:l57-65.
Youm Y, McMurtry RY, Flatt AE, Gifiepsie TB. Kinematics of the
wrist. I: an experimental study of radial-ulnar deviation and fiexion
extension. J Bone Joint Surg (Am) l978;60-A:423-31.