- Email: [email protected]
Corrective osteotomy for symptomatic scaphoid malunion
Injury, Int. J. Care Injured (2005) 36, 1440—1448
www.elsevier.com/locate/injury
Corrective osteotomy for symptomatic scaphoid malunion Essam Awad El-Karef * Orthopaedic Surgery and Traumatology, Faculty of Medicine, Alexandria University, 203 Thiebah Street (26B), Sporting, Alexandria, Egypt Accepted 5 September 2005
KEYWORDS Scaphoid; Malunion; Corrective osteotomy; Intrascaphoid angle; Humpback
Summary A prospective study aimed at assessment of the outcome of management of symptomatic scaphoid malunion (hump-back deformity). The work included 13 scaphoid malunions in 13 patients. All patients complained of weak painful hand grip and limitation of wrist function. All cases were subjected to a corrective opening wedge scaphoid osteotomy with insertion of a trapezoid-shaped tricortical iliac bone graft. At the final assessment, after a mean follow-up period of 42 months, the achieved results were rated excellent in seven cases, good in four and fair in two according to the scoring system used. Objectively, the mean range of wrist motion and hand grip strength improved from 48% and 47% pre-operatively to 82% and 79% at the final assessment. Radiological parameters including height to length ratio, lateral intrascaphoid angle and dorsal cortical angle were also effectively improved. The intraoperative corrected carpal alignment has almost been maintained at the final followup. The procedure did not have serious drawbacks such as non-union or avascular necrosis and perhaps might delay the development of degenerative arthritis of the wrist. # 2005 Published by Elsevier Ltd.
Introduction Unstable scaphoid fractures, virtually always, occur with a degree of perilunate ligamentous injury.15 This allows the proximal scaphoid fragment to follow the lunate into hyperextension, while the distal fragment behaves as a part of the distal carpal row and moves with the trapezieum into flexion. Thus, the scaphoid develops a hump-back deformity.16 * Tel.: +20 3 5455374/12 2702514; fax: +20 3 5410240. E-mail address: [email protected]. 0020–1383/$ — see front matter # 2005 Published by Elsevier Ltd. doi:10.1016/j.injury.2005.09.003
Therefore, such unstable fractures are prone to malunion and non-union.1,10,12,19 Malunion of the scaphoid may also complicate conventional inlay grafting procedures in which union is achieved without deformity correction.11 Malunited scaphoid fractures should be considered as an intra-articular incongruity and are expected to produce abnormal carpal mechanics that may lead to several clinical problems such as pain, loss of mobility, weak hand grip and may herald the progression of degenerative arthritis.1,5,6 Despite the guarded natural history of scaphoid
Scaphoid malunion
1441
malunion, it has received relatively little attention in the literature. In the early 1970s, Segmuller pointed out that shortening of the scaphoid was detrimental for normal carpal kinematics and proposed a technique of interpositional or ‘‘sandwich’’ grafting.8 Fisk (1970) recognized the association of dorsal carpal instability patterns with the humpback deformity of the scaphoid.9 He reconstructed the collapsed scaphoid with a wedge of bone cut from the resected radial styloid. Fernandez in 1984, modified the Fisk’s procedure and used an iliac bone graft and internal fixation, and Lynch and Linscheid (1997) preferred to fix the reduced dorsally rotated lunate to the radius before grafting and fixing the scaphoid with a Herbert screw.13 The aim of this work was to study the outcome of corrective open wedge osteotomy and iliac bone grafting for treatment of symptomatic scaphoid hump-back deformity.
Patients and methods A longitudinal prospective study included 13 symptomatic malunited scaphoid fractures in 13 patients. The mean duration of follow-up was 42 months (ranging from 20 months to 10 years). All fractures were unilateral with no history of injury to the other wrist. All of them were post-unstable (displaced) waist fractures of the scaphoid (Herbert B2). All had been treated conservatively in a cast that was applied within the first two days after injury, and was continued for a mean period of 10 weeks (range: 8—14 weeks). All had healed in a malposition that was not recognised, at that time by the treating surgeon. All patients were men and the mean age at the time of surgery was 26 (range: 21—36) years. Twelve patients were heavy manual labourers and one had an office job. The dominant hand was involved only in five patients. The mechanism of injury was a simple fall on out stretched hand
Table 1 Pre-operatively and final subjective clinical assessment Symptoms
Preoperative No.
Moderate wrist pain (during ADL) Severe wrist pain (at rest) Subjective hand grip weakness Sense of stiffness Limitations of ADL performance
Final assessment No.
%
10
% 77
4
31 a
3
23
—
—
13
100
2
15 a
13 13
100 100
5 4
38 a 31 a
ADL: activities of daily living. a Significant differences between the pre-operative and last visit values.
in 11 patients and a road traffic accident in two. The mean time lapse between injury and reconstructive osteotomy was 22 (range: 18—30) months. None of the patients suffered from any form of specific arthritis. Subjective weakness of the hand grip, moderate or severe wrist pain (during activities of daily living or at rest), limitation of wrist movements and inability to perform activities of daily living were the most frequent symptoms, occuring in all cases (Table 1). Objectively, tenderness on the dorsal aspect of the wrist, on the snuff box and/or on pressing the scaphoid tubercle was elicited in all cases. A swollen wrist with signs of extensor tenosynovitis was reported in six cases. A prominent capitate was also a common finding being reported in seven cases. With the use of the other side as a standard, there was a pre-operative degree of loss in wrist movements, in certain directions, in all cases. The mean loss of wrist flexion—extension, forearm pronation— supination and wrist radio-ulnar deviation arcs of movement were 598 (range: 50—758), 258 (range: 15—758) and 108 (range: 5—108), respectively.
Table 2 Pre-operative and latest visit means (range) of wrist movements and grip strength and its percentage of the uninvolved side values Pre-operative
Last visit
Mean (range)
Percent of uninvolved
Mean (range)
Active movements Dorsiflexion Palmarflexion Pronation Supination Ulnar deviation Radial deviation
418 758 808 748 188 128
47 89 91 70 64 48
578 798 878 818 228 168
Grip strength (kg)
19 (13: 30)
47
32 (26: 36) a
a
(25: (65: (70: (40: (20: (10:
488) 808) 858) 808) 258) 158)
Significant differences between the pre-operative and last visit means.
(55: (60: (70: (65: (18: (14:
858) a 858) 908) 908) a 258) 228)
Percent of uninvolved 73 a 94 a 96 89 a 84 82 79 a
1442 Reduced dorsiflexion, supination and radial deviation more obvious than that in palmar flexion, pronation and ulnar deviation, in all cases (Table 2). Grip strength was measured bilaterally using the three grip positions of the JAYMAR dynamometer (Asimow Engineering Co., Los Angeles) according to the protocol described by Mathiowetz et al. (1984),14 and was expressed as a percentage of the strength of the uninvolved side. Relative weakness of grip strength was reported in all cases. The pre-operative mean grip strength of the affected hand was 19.2 kg-force compared with 41 kg-force in the uninjured hands: 47% of the unaffected side (range: 32—73%) (Table 2). Scaphoid series radiographs and CT scans taken along the axis of the scaphoid were performed preoperatively performed, for both wrists in all patients. All cases had a rather deformed scaphoid with shortening and flexion in the sagittal plane (Fig. 1), and ulnar deviation of the distal fragment in the frontal plane; this usually overlapped the radial aspect of the capitate in the PA views, particularly when the wrist was in a midprone and radial deviation position. In addition, an element of axial rotation in the horizontal plane with pronation of the distal fragment was always perceived. Threedimensional CT reconstructions, obtained for some cases, were also helpful in demonstrating all elements of the deformity, particularly the rotational malalignment (Fig. 2).
E.A. El-Karef
Figure 2 Three-dimensional CT scan of the wrist showing a malunited scaphoid with ulnar deviation and pronation of the distal fragment, scapholunate dissociation, dorsal rotation of the lunate, and mild prominence of the capitate head.
Early mild radiographic signs of radioscaphoid arthritis were demonstrated pre-operatively in three cases as fine subchondral sclerosis at the radial styloid. These three cases had the longest time lapse before surgery. The hump-back deformity was evaluated using the (articular) lateral intrascaphoid angle that was made by the intersection of two lines perpendicular to each of the proximal and
Figure 1 (A) A malunited foreshortened scaphoid with scapholunate dissociation and ulnar deviation of the distal scaphoid fragment. (B) A semi-supinated projection showing how much the scaphoid is angulated and shortened with a hump-back deformity. (C and D) Immediate post-operative AP and lateral views. (E and F) After union and before removal of the cast and the Kirschner wires. (G) Immediately following removal of the cast and the wires. (H) Four years follow-up picture.
Scaphoid malunion
1443
Figure 3 (A) A pre-operative CTscanning (middle cut) taken along the axis of the scaphoid. (B) The lateral intrascaphoid angle above 558. (C) The dorsal cortical angle less than 1408. (D) Height to length ratio is more than 5.5.
Table 3 Radiological assessment Pre-operative Scapholunate angle (8) Radiolunate angle (8) Capitolunate angle (8) Lateral intrascaphoid angle (8) Dorsal cortical angle (8) Height to length ratio a
Last visit
Mean
Range
Mean
Range
60 16 20 59 139 0.62
50—85 0—20 0—30 53—65 135—143 0.58—0.66
55 7 9 34 147 0.53
40—70 0—12 a 0—15 a 29—39 a 139—156 0.51—0.53
Significant differences between the pre-operative and last visit means.
distal articular surfaces1,18 in the middle scaphoid cut of the axial scaphoid CT scans with reference to the normal values of the contralateral side (Fig. 3A). The mean pre-operative lateral intrascaphoid angle was 598 (SD, 6) in comparison with 378 (SD, 2.4) of the unaffected side (Fig. 3B) (Table 3). The dorsal scaphoid angle was also used, in retrospect, for deformity quantification. It was made by two lines drawn tangential to the flattest portion of the dorsal cortex of the proximal and distal halves of the scaphoid in the lateral radiographs2,18 (Fig. 3C). The mean pre-operative dorsal cortical angle was 1398 (SD, 1.6) in comparison with 1548 (SD, 2.3) of the unaffected side (Table 3). All radiological parameters were measured thrice at different times by the two authors to avoid observer errors.
Methods All of the 13 patients were subjected to a corrective open wedge osteotomy and insertion of a tightfitting trapezoid shaped tricortical iliac crest bone graft with multiple Kirschner wires fixation.
Pre-operative planning Using the sound wrist as a standard, radiographic cut outs on tracing paper were made to determine the size and shape of the graft needed to restore the anatomic length and form of the scaphoid. The degree of carpal malalignment of the affected wrist and the normal and pathologic scapholunate and lunocapitate angles for each individual patient were also identified. Dimensions of the trapezoid shaped bone graft, theoretically, depend on the angle of malunion and the existing scaphoid shortening. Nevertheless, accurate calculations were not quiet simple because of the rotational component of the deformity. However, PA view in maximal ulnar deviation was helpful for anticipating the heights of the interpositional bone graft in the frontal plane (height of the radial and ulnar borders of the trapezoid-shaped graft) as the scaphoid should rotate to an extended position in maximal ulnar deviation and appears in its maximal length in this view. While measuring scaphoid length in the lateral images, as the distance between two lines touching its most proximal and
1444 distal articular surfaces, which are perpendicular to the palmar tangent of the scaphoid,2 should determine the palmar height of the graft (Fig. 3D).
Osteotomy technique The palmar approach was preferred as it provides better exposure of the whole scaphoid and the scapholunate joint, so, allowing lunate rotation correction and offers a direct and easy access to correct scaphoid palmar flexion that requires a palmar opening wedge grafting. Furthermore, it also reduces the danger of iatrogenic damage of the vascular supply of the scaphoid and accidental injuries of the superficial branches of the radial nerve. The capsular flaps that contain the strong radioscaphocapitate ligament were held on both sides with stay sutures to facilitate anatomic closure. A transverse osteotomy at the site of malunion was made with a fine osteotome, taking care not to damage the point of entry of the dorsal ridge vessels. The ‘‘hump-back’’ deformity and shortening were corrected by distracting the osteotomy site with a spreader clamp. This was simply helped by hyperextending the wrist. Dorsal rotation of the proximal fragment should simultaneously be corrected by manipulating the lunate with a fine bone spike. In the mean time, ulnar deviation of the distal pole was corrected by opening the osteotomy more on the ulnar side. Combined extension and supination of the distal fragment was sometimes necessary for restoration of full wrist extension. Achieving complete passive extension of the wrist, intraoperatively, while the two fragments were held open, before graft
E.A. El-Karef insertion, was used as an intra-operative clinical guide for axial rotation correction. A trapezoidal tricortical iliac crest bone graft was shaped to snugly fit the gap and tightly inserted into the osteotomy site. The graft was oriented so that a good cortical part lies palmarly and ulnarward guarding against collapse (Fig. 4). The concave inner cortical surface of the iliac bone was directed ulnar ward to fit with the capitate contour. The graft dimensions obtained pre-operatively were always less than that needed to be tightly fitting between the scaphoid fragments. The graft was then secured with two or three 1.2 mm Kirschner wires. Intraoperative fluoroscopic assessment of scaphoid length and alignment was a routine. It also helped in controlling correction of lunate rotation as guided by scapholunate and lunocapitate angles. The wires were then introduced further up to the radius transfixing the bone and the graft. Protruding edges of the graft were then trimmed to match with the proximal and distal fragments’ contour. Spontaneous stable derotation of the lunate usually took place after insertion of the trapezoidal graft and correction of the hump-back deformity, this may be due to restoration of normal scaphoid length that restored the strut function of the radial column of the carpus, which, in turn, controlled normal carpal relationships. However, if needed, intercarpal transfixing wire(s) were inserted to maintain carpal alignment and the corrected lunate rotation particularly in long-standing cases. After careful closure of the palmar capsule, including the radioscaphocapitate ligaments, the wires were cut short under the skin and a thumb and wrist plaster splint was applied.
Post-operative management Sutures were removed after 2 weeks and an above elbow scaphoid cast was applied until radiographic indicators of union were recognised with disappearance of the osteotomy sites and presence of crossing trabaculae between scaphoid fragments and the bone graft (range: 10—16 weeks). Kirschner wires were removed, at that time, through a small incision under local anesthesia and a removable splint or wrist brace was used afterward to allow progressing active exercises of the wrist and the patient was encouraged to use the hand for activities of daily living. Heavy manual work and strenuous sport activities were prohibited until supple movements and considerably good grip power were gradually restored.
Methods of assessment Figure 4 (A and B) The trapezoidal shaped graft. (C) After the graft has been tightly inserted.
After a mean follow-up of 42 months (range: 20 months to 10 years) all patients were invited for a
Scaphoid malunion late subjective, objective and radiographic evaluation. Wrist motion, pain, grip strength, working capacity and patient’s satisfaction) were graded according to the Mayo wrist scoring system. Both pre- and post-operative data were statistically analysed using the SPSS (Version 6) for windows using Student’s t and Chi square tests and the level of significance was p < 0.05.
Results According to the Mayo scoring system, the outcome was rated excellent in seven cases, good in four and fair in two.
Subjective evaluation
1445 Radiographic signs of union was detected (obliteration of any gaps and crossing trabeculae), in 12 cases, in a mean time of 12 weeks (range: 10—16 weeks) (Figs. 1 and 5). In the remaining case the graft has united to the proximal scaphoid fragment while its union to the distal one went into delayed union. It needed a lag screw fixation as an ordinary delayed union of a scaphoid fracture 6 months after osteotomy. None of the cases developed any signal of avascular necrosis. The height to length ratio2 was calculated for all patients post-operatively and in retrograde preoperatively with a mean ratio of 0.62 (SD, 0.02) and 0.53 (SD, 0.01) for the pre- and post-operative measurements, respectively. The mean (articular) lateral intrascaphoid angle at the last visit was 348 (SD, 1.9) in comparison with 598 (SD, 2.2) pre-
Seven patients were pleased with their outcome; apart from occasional mild aches on demanding activities, all of them were no longer symptomatic. Five were moderately satisfied and had persistent mild stiffness, occasional mild aches on strenuous activities or subjective sense of a weak hand. Only one of 13 patients was dissatisfied due to persistence of moderate wrist pain (during daily life activities) with sense of stiff weak hand grip (Table 1). Eleven patients reported improvement in work and sporting competency. Ten patients gradually returned to their previous jobs and sports within 4—10 months post-operatively, while three labourers chose to return to modified duties.
Objective clinical evaluation Although all patients suffered from post-plaster stiffness, substantially improved wrist movement was eventually obtained by all of them. The mean objective range of wrist movements, at the latest visit, was 88% of that of the normal side (Table 2). Grip strength did improve in all cases but never reached that of the opposite side. The mean last assessment grip strength was 79% of the uninjured side compared with 47% pre-operatively (Table 2).
Radiographic evaluation Comparative radiographs of both wrists and final assessment CT scans of all operated cases were assessed thrice by both authors at different times to avoid observer errors, to evaluate union, restoration of scaphoid length and shape and correction of carpal alignment.
Figure 5 (A) A malunited scaphoid with massively angulated distal fragment that appeared end on. (B) PA view with lateral deviation demonstrating scapholunate disruption, ulnar deviation of the distal fragment and a dorsolateral exostosis described previously by Fisk (1984). (C) Concomitant dorsally rotated lunate. (D—E) Immediate post-operative picture after correction of the deformity and alignment of the scaphoid. (F—G) Three years after reconstruction.
1446 operatively. While, the mean dorsal cortical angle at the final assessment was 1478 (SD, 2.8) in comparison to pre-operative value of 1398 (SD, 2.1). The difference between pre- and post-operative values of the three measurements (height to length ratio, lateral interascaphoid and dorsal cortical angles) had no statistical relationship with the clinical outcome. The intra-operatively corrected carpal alignment has almost been kept up, in all cases, till late followup as indicated by the maintenance of the corrected scapholunate and capitolunate angles to within normal ranges. The mean scapholunate angle on the operated side, at the last visit, was 558 (range: 40—708) in comparison with a mean of 608 (range: 50—858) and 488 (range: 40—658) for the pre-operative and sound side, respectively. In eight patients, no radiographic signs of degenerative changes could be detected at the most recent follow-up. While, the remaining five patients showed mild degenerative changes in the form of a mild sharpening of the radial styloid and mild subchondral sclerosis of the distal radius, three of them already had almost the same degree of changes existed pre-operatively. None of the patients had developed severe arthritis at late follow-up.
Complications Transient skin irritation with mild superficial pin track infection occurred in two cases.
Discussion The early and late consequences of scaphoid malunion, when it exceeds certain limits, should be expected to have a deteriorating natural history.4,6,12 Scaphoid malunion alters the carpal kinematics that may correlate with clinical symptoms such as weak grip strength and painful decrease in wrist motion. Such a situation is a pre-arthritic condition and may also predispose to premature wrist arthrosis1,4,16,19 that may spread to show evidence of scapholunate advanced collapse (SLAC) as described by Watson and Ballet (1984).20 The more severe the malunion the worse the prognosis is likely to be. Surgical correction, therefore, should be considered, at least in symptomatic patients. Although, scaphoid fracture malunion is now a well-recognised entity and evidence of an associated poor prognosis are mounting, the reported number of patients treated with scaphoid osteotomy is surprisingly small.8 The risk of creating iatrogenic avascular necrosis of the proximal fragment or converting a healed scaphoid into a non-united one
E.A. El-Karef may be the reasons. Non-symptomatic malunions with acceptable painless limitation of wrist function that do not need further treatment may be another cause as stated by Jiranek et al. (1992).11 Little is written about the indications for, and precise methods of treating scaphoid malunions. However, a corrective osteotomy seems to be a good reconstructive option as it indeed is directed at correction of the pathology and restoring normality, and its targets are to improve functions of the wrist and to reduce or delay the incidence of arthritis. Herbert in 1990 expected that corrective osteotomy of the scaphoid will become an increasingly common operation in the future years.8 Barton (1992)3 stated that it is certainly bold to divide a bone notoriously subject to non-union. Fernandez (1998)7 reviewed the English speaking literature of scaphoid osteotomy and found that neither iatrogenic avascular necrosis nor non-union had complicated any of the cases. Delaying correction until degenerative changes develop and treating the condition with limited or total wrist arthrodesis will lead to the undesirable loss of motion. While, wrist arthroplasty may be an unreliable long-term solution for young patients. Corrective scaphoid osteotomy, in the present work, was indicated only in relentlessly symptomatic malunion particularly in the presence of radiological indicators of bad prognosis such as lateral intrascaphoid angle greater than 558.1 Nevertheless, the degree of pain, stiffness and weakness should be weighed against the possible risks of post-osteotomy complications, non-union or avascular necrosis. The presence of long-established periscaphoid arthritis were considered a formal contraindication and should be carefully ruled out pre-operatively. Standardised pre-operative radiographic assessment of both wrists allowed sensible measurements of normal scaphoid shape and length and good evaluation of the normal and the pathologic intercarpal relationships, which consequently, facilitates better planning through a fair calculation of the size of the graft needed for each patient to restore anatomic length and realignment of the scaphoid. However, despite the extensive pre-operative planning, intra-operative assessment was always essential particularly for the rotational element of the deformity. Several methods have been described for determining the degree of deformation of a malunited scaphoid.1,2,4,8,17,18 The apparent shortening seen on PA radiographs and even that confirmed by CT, may be substantially less than the true shortening. Therefore, interposed graft height always exceeded the pre-operative measurements. Both ulnar and
Scaphoid malunion palmar heights of the graft were almost approaching twice the pre-operatively measured shortening in most of the cases. This observation provided a fairly accurate pre-operative estimation in the later cases of this series. Because of scaphoid irregularity and variability in patient positioning during CT and/or radiography18 and also because of the difficulty in determining where the articular surface ends and the cortex begins2 there is often inability to reproduce scaphoid axes. This leads to poor intra and inter-observer reliability of the lateral interascaphoid angle and adds to the intricacy of pre-operative planning. Moreover, axial CT scanning of the scaphoid is not a straightforward job. It required experience in hand positioning.2,12,17,18 Only few cuts were obtained for the scaphoid bone and the middle cut between the capitate and the radial border of the scaphoid was the one used for the measurements. The inserted graft must be tightly fitted between the scaphoid fragments to provide stability with good apposition of the osteotomised surfaces and to help graft incorporation. A combination of K. wire and screw fixation has been tried in the early two cases (Fig. 5) but was not a favorite due to possible damage of the core of the graft during screw insertion and the screw may hold the fragments apart and prevent collapse if bone resorption occur during healing. However, there was no statistical correlation between method of fixation and any of the clinical parameters at the final assessment. Although none of the patients included developed severe degenerative changes at the wrist, five of the 13 patients had radiographic evidence of mild degenerative changes at the radioscaphoid joint on latest follow-up (mean, 42 months), most of them were asymptomatic. These late findings may suggest that the procedure at least slows the process of degenerative arthritis. However, this depends on the individual functional demands. In the present study, good response to re-establishment of normal scaphoid anatomy following the procedure was documented in the majority of cases with cessation of their symptoms. However, there was persistent pain and positive clinical tests for ligament instability in four cases, despite radiographic evidence of scaphoid realignment. This may be attributed to deep scarring or an initially associated severe perilunate injury.8,15 The encouraging results of this work and those reported in the literature would suggest that corrective scaphoid osteotomy could be indicated in an effort to correct intra-articular incongruity and to restore the anatomic and functional integrity of the wrist joint. It should be performed at an early stage
1447 before development of degenerative changes. The procedure did not create serious complications such as avascular necrosis or nonunion and would also appear to be a measure to prevent late intercarpal collapse and, in theory, should provide better function than late salvage procedures.
Acknowledgements The author is grateful to Dr. Hesham Kotb, MD. Assistant professor in Diagnostic Radiology, Alexandria University, Egypt, for his tremendous help through out this work.
References 1. Amadio PC, Bequist TH, Smith DK, et al. Scaphoid malunion. J Hand Surg 1989;14A:679. 2. Bain GI, Bennett JD, Dermid M, et al. Measurement of the scaphoid deformity using longitudinal computed tomography: intra- and interobserver variability using various measurement technique. J Hand Surg 1998;17A:76. 3. Barton NJ. Twenty questions about scaphoid fractures. J Hand Surg 1992;17B:289. 4. Birchard D, Pichora D. Experimental corrective scaphoid osteotomy for scaphoid malunion with abnormal wrist mechanics. J Hand Surg 1990;15A:863. 5. Burgess RC. The effect of a simulated scaphoid malunion on wrist motion. J Hand Surg 1987;12A:774. 6. Femandez DL. A technique for anterior wedge-shaped grafts for scaphoid nonunions with carpal instability. J Hand Surg 1984;9A:733. 7. Fernandez DL, Martin CL, Gonzalez del Pino J. Scaphoid malunion: the significance of rotational malalignment. J Hand Surg 1998;23B:771. 8. Femandez DL, Eggli S. Scaphoid nonunion and malunion. How to correct deformity. Hand Clin 2001;17:631. 9. Fisk GR. Carpal instability and the fractured scaphoid. Ann R Collage Surg Engl 1970;46:63. 10. Fisk GR. The wrist: review article. J Bone Joint Surg 1984;66B:396. 11. Jiranek WA, Ruby LK, Millender LB, et al. Long-term results after Russe bone-grafting: the effect of malunion of the scaphoid. J Bone Joint Surg 1992;74A:1217. 12. Kozin SH. Incidence, mechanism and natural history of scaphoid fractures. Hand Clin 2001;17:515. 13. Lynch NM, Linscheid RL. Corrective osteotomy for scaphoid malunion: technique and long-term follow-up evaluation. J Hand Surg 1997;22 A:35. 14. Mathiowetz V, Weber K, Volland G, Kashman N. Reliability and validity of hand strength evaluation. J Hand Surg 1984;9A: 222. 15. Mayfield JK. Mechanism of carpal injuries. Clin Orthopaed Relat Res 1980;149:45. 16. Nakamura R, Imaeda T, Miura T. Scaphoid malunion. J Bone Joint Surg 1991;73B:134. 17. Sanders WE. Evaluation of the humpback scaphoid by computed tomography in the longitudinal axial plane of the scaphoid. J Hand Surg 1988;13A:182.
1448 18. Smith DK, Linscheid RL, Amadio PC, Berquist TH, Cooney WP. Scaphoid anatomy: evaluation with complex motion tomography. Radiology 1989;173:177. 19. Tomaino MM, King J, Pizillo M. Correction of lunate malalignment when bone grafting scaphoid nonunion with humpback
E.A. El-Karef deformity: rationale and results of a technique revisited. J Hand Surg 2000;25A:322. 20. Watson HK, Ballet FL. The SLAC wrist: Scapholunate advanced collapse pattern of degenerative arthritis. J Hand Surg 1984;9A:358.
www.elsevier.com/locate/injury
Corrective osteotomy for symptomatic scaphoid malunion Essam Awad El-Karef * Orthopaedic Surgery and Traumatology, Faculty of Medicine, Alexandria University, 203 Thiebah Street (26B), Sporting, Alexandria, Egypt Accepted 5 September 2005
KEYWORDS Scaphoid; Malunion; Corrective osteotomy; Intrascaphoid angle; Humpback
Summary A prospective study aimed at assessment of the outcome of management of symptomatic scaphoid malunion (hump-back deformity). The work included 13 scaphoid malunions in 13 patients. All patients complained of weak painful hand grip and limitation of wrist function. All cases were subjected to a corrective opening wedge scaphoid osteotomy with insertion of a trapezoid-shaped tricortical iliac bone graft. At the final assessment, after a mean follow-up period of 42 months, the achieved results were rated excellent in seven cases, good in four and fair in two according to the scoring system used. Objectively, the mean range of wrist motion and hand grip strength improved from 48% and 47% pre-operatively to 82% and 79% at the final assessment. Radiological parameters including height to length ratio, lateral intrascaphoid angle and dorsal cortical angle were also effectively improved. The intraoperative corrected carpal alignment has almost been maintained at the final followup. The procedure did not have serious drawbacks such as non-union or avascular necrosis and perhaps might delay the development of degenerative arthritis of the wrist. # 2005 Published by Elsevier Ltd.
Introduction Unstable scaphoid fractures, virtually always, occur with a degree of perilunate ligamentous injury.15 This allows the proximal scaphoid fragment to follow the lunate into hyperextension, while the distal fragment behaves as a part of the distal carpal row and moves with the trapezieum into flexion. Thus, the scaphoid develops a hump-back deformity.16 * Tel.: +20 3 5455374/12 2702514; fax: +20 3 5410240. E-mail address: [email protected]. 0020–1383/$ — see front matter # 2005 Published by Elsevier Ltd. doi:10.1016/j.injury.2005.09.003
Therefore, such unstable fractures are prone to malunion and non-union.1,10,12,19 Malunion of the scaphoid may also complicate conventional inlay grafting procedures in which union is achieved without deformity correction.11 Malunited scaphoid fractures should be considered as an intra-articular incongruity and are expected to produce abnormal carpal mechanics that may lead to several clinical problems such as pain, loss of mobility, weak hand grip and may herald the progression of degenerative arthritis.1,5,6 Despite the guarded natural history of scaphoid
Scaphoid malunion
1441
malunion, it has received relatively little attention in the literature. In the early 1970s, Segmuller pointed out that shortening of the scaphoid was detrimental for normal carpal kinematics and proposed a technique of interpositional or ‘‘sandwich’’ grafting.8 Fisk (1970) recognized the association of dorsal carpal instability patterns with the humpback deformity of the scaphoid.9 He reconstructed the collapsed scaphoid with a wedge of bone cut from the resected radial styloid. Fernandez in 1984, modified the Fisk’s procedure and used an iliac bone graft and internal fixation, and Lynch and Linscheid (1997) preferred to fix the reduced dorsally rotated lunate to the radius before grafting and fixing the scaphoid with a Herbert screw.13 The aim of this work was to study the outcome of corrective open wedge osteotomy and iliac bone grafting for treatment of symptomatic scaphoid hump-back deformity.
Patients and methods A longitudinal prospective study included 13 symptomatic malunited scaphoid fractures in 13 patients. The mean duration of follow-up was 42 months (ranging from 20 months to 10 years). All fractures were unilateral with no history of injury to the other wrist. All of them were post-unstable (displaced) waist fractures of the scaphoid (Herbert B2). All had been treated conservatively in a cast that was applied within the first two days after injury, and was continued for a mean period of 10 weeks (range: 8—14 weeks). All had healed in a malposition that was not recognised, at that time by the treating surgeon. All patients were men and the mean age at the time of surgery was 26 (range: 21—36) years. Twelve patients were heavy manual labourers and one had an office job. The dominant hand was involved only in five patients. The mechanism of injury was a simple fall on out stretched hand
Table 1 Pre-operatively and final subjective clinical assessment Symptoms
Preoperative No.
Moderate wrist pain (during ADL) Severe wrist pain (at rest) Subjective hand grip weakness Sense of stiffness Limitations of ADL performance
Final assessment No.
%
10
% 77
4
31 a
3
23
—
—
13
100
2
15 a
13 13
100 100
5 4
38 a 31 a
ADL: activities of daily living. a Significant differences between the pre-operative and last visit values.
in 11 patients and a road traffic accident in two. The mean time lapse between injury and reconstructive osteotomy was 22 (range: 18—30) months. None of the patients suffered from any form of specific arthritis. Subjective weakness of the hand grip, moderate or severe wrist pain (during activities of daily living or at rest), limitation of wrist movements and inability to perform activities of daily living were the most frequent symptoms, occuring in all cases (Table 1). Objectively, tenderness on the dorsal aspect of the wrist, on the snuff box and/or on pressing the scaphoid tubercle was elicited in all cases. A swollen wrist with signs of extensor tenosynovitis was reported in six cases. A prominent capitate was also a common finding being reported in seven cases. With the use of the other side as a standard, there was a pre-operative degree of loss in wrist movements, in certain directions, in all cases. The mean loss of wrist flexion—extension, forearm pronation— supination and wrist radio-ulnar deviation arcs of movement were 598 (range: 50—758), 258 (range: 15—758) and 108 (range: 5—108), respectively.
Table 2 Pre-operative and latest visit means (range) of wrist movements and grip strength and its percentage of the uninvolved side values Pre-operative
Last visit
Mean (range)
Percent of uninvolved
Mean (range)
Active movements Dorsiflexion Palmarflexion Pronation Supination Ulnar deviation Radial deviation
418 758 808 748 188 128
47 89 91 70 64 48
578 798 878 818 228 168
Grip strength (kg)
19 (13: 30)
47
32 (26: 36) a
a
(25: (65: (70: (40: (20: (10:
488) 808) 858) 808) 258) 158)
Significant differences between the pre-operative and last visit means.
(55: (60: (70: (65: (18: (14:
858) a 858) 908) 908) a 258) 228)
Percent of uninvolved 73 a 94 a 96 89 a 84 82 79 a
1442 Reduced dorsiflexion, supination and radial deviation more obvious than that in palmar flexion, pronation and ulnar deviation, in all cases (Table 2). Grip strength was measured bilaterally using the three grip positions of the JAYMAR dynamometer (Asimow Engineering Co., Los Angeles) according to the protocol described by Mathiowetz et al. (1984),14 and was expressed as a percentage of the strength of the uninvolved side. Relative weakness of grip strength was reported in all cases. The pre-operative mean grip strength of the affected hand was 19.2 kg-force compared with 41 kg-force in the uninjured hands: 47% of the unaffected side (range: 32—73%) (Table 2). Scaphoid series radiographs and CT scans taken along the axis of the scaphoid were performed preoperatively performed, for both wrists in all patients. All cases had a rather deformed scaphoid with shortening and flexion in the sagittal plane (Fig. 1), and ulnar deviation of the distal fragment in the frontal plane; this usually overlapped the radial aspect of the capitate in the PA views, particularly when the wrist was in a midprone and radial deviation position. In addition, an element of axial rotation in the horizontal plane with pronation of the distal fragment was always perceived. Threedimensional CT reconstructions, obtained for some cases, were also helpful in demonstrating all elements of the deformity, particularly the rotational malalignment (Fig. 2).
E.A. El-Karef
Figure 2 Three-dimensional CT scan of the wrist showing a malunited scaphoid with ulnar deviation and pronation of the distal fragment, scapholunate dissociation, dorsal rotation of the lunate, and mild prominence of the capitate head.
Early mild radiographic signs of radioscaphoid arthritis were demonstrated pre-operatively in three cases as fine subchondral sclerosis at the radial styloid. These three cases had the longest time lapse before surgery. The hump-back deformity was evaluated using the (articular) lateral intrascaphoid angle that was made by the intersection of two lines perpendicular to each of the proximal and
Figure 1 (A) A malunited foreshortened scaphoid with scapholunate dissociation and ulnar deviation of the distal scaphoid fragment. (B) A semi-supinated projection showing how much the scaphoid is angulated and shortened with a hump-back deformity. (C and D) Immediate post-operative AP and lateral views. (E and F) After union and before removal of the cast and the Kirschner wires. (G) Immediately following removal of the cast and the wires. (H) Four years follow-up picture.
Scaphoid malunion
1443
Figure 3 (A) A pre-operative CTscanning (middle cut) taken along the axis of the scaphoid. (B) The lateral intrascaphoid angle above 558. (C) The dorsal cortical angle less than 1408. (D) Height to length ratio is more than 5.5.
Table 3 Radiological assessment Pre-operative Scapholunate angle (8) Radiolunate angle (8) Capitolunate angle (8) Lateral intrascaphoid angle (8) Dorsal cortical angle (8) Height to length ratio a
Last visit
Mean
Range
Mean
Range
60 16 20 59 139 0.62
50—85 0—20 0—30 53—65 135—143 0.58—0.66
55 7 9 34 147 0.53
40—70 0—12 a 0—15 a 29—39 a 139—156 0.51—0.53
Significant differences between the pre-operative and last visit means.
distal articular surfaces1,18 in the middle scaphoid cut of the axial scaphoid CT scans with reference to the normal values of the contralateral side (Fig. 3A). The mean pre-operative lateral intrascaphoid angle was 598 (SD, 6) in comparison with 378 (SD, 2.4) of the unaffected side (Fig. 3B) (Table 3). The dorsal scaphoid angle was also used, in retrospect, for deformity quantification. It was made by two lines drawn tangential to the flattest portion of the dorsal cortex of the proximal and distal halves of the scaphoid in the lateral radiographs2,18 (Fig. 3C). The mean pre-operative dorsal cortical angle was 1398 (SD, 1.6) in comparison with 1548 (SD, 2.3) of the unaffected side (Table 3). All radiological parameters were measured thrice at different times by the two authors to avoid observer errors.
Methods All of the 13 patients were subjected to a corrective open wedge osteotomy and insertion of a tightfitting trapezoid shaped tricortical iliac crest bone graft with multiple Kirschner wires fixation.
Pre-operative planning Using the sound wrist as a standard, radiographic cut outs on tracing paper were made to determine the size and shape of the graft needed to restore the anatomic length and form of the scaphoid. The degree of carpal malalignment of the affected wrist and the normal and pathologic scapholunate and lunocapitate angles for each individual patient were also identified. Dimensions of the trapezoid shaped bone graft, theoretically, depend on the angle of malunion and the existing scaphoid shortening. Nevertheless, accurate calculations were not quiet simple because of the rotational component of the deformity. However, PA view in maximal ulnar deviation was helpful for anticipating the heights of the interpositional bone graft in the frontal plane (height of the radial and ulnar borders of the trapezoid-shaped graft) as the scaphoid should rotate to an extended position in maximal ulnar deviation and appears in its maximal length in this view. While measuring scaphoid length in the lateral images, as the distance between two lines touching its most proximal and
1444 distal articular surfaces, which are perpendicular to the palmar tangent of the scaphoid,2 should determine the palmar height of the graft (Fig. 3D).
Osteotomy technique The palmar approach was preferred as it provides better exposure of the whole scaphoid and the scapholunate joint, so, allowing lunate rotation correction and offers a direct and easy access to correct scaphoid palmar flexion that requires a palmar opening wedge grafting. Furthermore, it also reduces the danger of iatrogenic damage of the vascular supply of the scaphoid and accidental injuries of the superficial branches of the radial nerve. The capsular flaps that contain the strong radioscaphocapitate ligament were held on both sides with stay sutures to facilitate anatomic closure. A transverse osteotomy at the site of malunion was made with a fine osteotome, taking care not to damage the point of entry of the dorsal ridge vessels. The ‘‘hump-back’’ deformity and shortening were corrected by distracting the osteotomy site with a spreader clamp. This was simply helped by hyperextending the wrist. Dorsal rotation of the proximal fragment should simultaneously be corrected by manipulating the lunate with a fine bone spike. In the mean time, ulnar deviation of the distal pole was corrected by opening the osteotomy more on the ulnar side. Combined extension and supination of the distal fragment was sometimes necessary for restoration of full wrist extension. Achieving complete passive extension of the wrist, intraoperatively, while the two fragments were held open, before graft
E.A. El-Karef insertion, was used as an intra-operative clinical guide for axial rotation correction. A trapezoidal tricortical iliac crest bone graft was shaped to snugly fit the gap and tightly inserted into the osteotomy site. The graft was oriented so that a good cortical part lies palmarly and ulnarward guarding against collapse (Fig. 4). The concave inner cortical surface of the iliac bone was directed ulnar ward to fit with the capitate contour. The graft dimensions obtained pre-operatively were always less than that needed to be tightly fitting between the scaphoid fragments. The graft was then secured with two or three 1.2 mm Kirschner wires. Intraoperative fluoroscopic assessment of scaphoid length and alignment was a routine. It also helped in controlling correction of lunate rotation as guided by scapholunate and lunocapitate angles. The wires were then introduced further up to the radius transfixing the bone and the graft. Protruding edges of the graft were then trimmed to match with the proximal and distal fragments’ contour. Spontaneous stable derotation of the lunate usually took place after insertion of the trapezoidal graft and correction of the hump-back deformity, this may be due to restoration of normal scaphoid length that restored the strut function of the radial column of the carpus, which, in turn, controlled normal carpal relationships. However, if needed, intercarpal transfixing wire(s) were inserted to maintain carpal alignment and the corrected lunate rotation particularly in long-standing cases. After careful closure of the palmar capsule, including the radioscaphocapitate ligaments, the wires were cut short under the skin and a thumb and wrist plaster splint was applied.
Post-operative management Sutures were removed after 2 weeks and an above elbow scaphoid cast was applied until radiographic indicators of union were recognised with disappearance of the osteotomy sites and presence of crossing trabaculae between scaphoid fragments and the bone graft (range: 10—16 weeks). Kirschner wires were removed, at that time, through a small incision under local anesthesia and a removable splint or wrist brace was used afterward to allow progressing active exercises of the wrist and the patient was encouraged to use the hand for activities of daily living. Heavy manual work and strenuous sport activities were prohibited until supple movements and considerably good grip power were gradually restored.
Methods of assessment Figure 4 (A and B) The trapezoidal shaped graft. (C) After the graft has been tightly inserted.
After a mean follow-up of 42 months (range: 20 months to 10 years) all patients were invited for a
Scaphoid malunion late subjective, objective and radiographic evaluation. Wrist motion, pain, grip strength, working capacity and patient’s satisfaction) were graded according to the Mayo wrist scoring system. Both pre- and post-operative data were statistically analysed using the SPSS (Version 6) for windows using Student’s t and Chi square tests and the level of significance was p < 0.05.
Results According to the Mayo scoring system, the outcome was rated excellent in seven cases, good in four and fair in two.
Subjective evaluation
1445 Radiographic signs of union was detected (obliteration of any gaps and crossing trabeculae), in 12 cases, in a mean time of 12 weeks (range: 10—16 weeks) (Figs. 1 and 5). In the remaining case the graft has united to the proximal scaphoid fragment while its union to the distal one went into delayed union. It needed a lag screw fixation as an ordinary delayed union of a scaphoid fracture 6 months after osteotomy. None of the cases developed any signal of avascular necrosis. The height to length ratio2 was calculated for all patients post-operatively and in retrograde preoperatively with a mean ratio of 0.62 (SD, 0.02) and 0.53 (SD, 0.01) for the pre- and post-operative measurements, respectively. The mean (articular) lateral intrascaphoid angle at the last visit was 348 (SD, 1.9) in comparison with 598 (SD, 2.2) pre-
Seven patients were pleased with their outcome; apart from occasional mild aches on demanding activities, all of them were no longer symptomatic. Five were moderately satisfied and had persistent mild stiffness, occasional mild aches on strenuous activities or subjective sense of a weak hand. Only one of 13 patients was dissatisfied due to persistence of moderate wrist pain (during daily life activities) with sense of stiff weak hand grip (Table 1). Eleven patients reported improvement in work and sporting competency. Ten patients gradually returned to their previous jobs and sports within 4—10 months post-operatively, while three labourers chose to return to modified duties.
Objective clinical evaluation Although all patients suffered from post-plaster stiffness, substantially improved wrist movement was eventually obtained by all of them. The mean objective range of wrist movements, at the latest visit, was 88% of that of the normal side (Table 2). Grip strength did improve in all cases but never reached that of the opposite side. The mean last assessment grip strength was 79% of the uninjured side compared with 47% pre-operatively (Table 2).
Radiographic evaluation Comparative radiographs of both wrists and final assessment CT scans of all operated cases were assessed thrice by both authors at different times to avoid observer errors, to evaluate union, restoration of scaphoid length and shape and correction of carpal alignment.
Figure 5 (A) A malunited scaphoid with massively angulated distal fragment that appeared end on. (B) PA view with lateral deviation demonstrating scapholunate disruption, ulnar deviation of the distal fragment and a dorsolateral exostosis described previously by Fisk (1984). (C) Concomitant dorsally rotated lunate. (D—E) Immediate post-operative picture after correction of the deformity and alignment of the scaphoid. (F—G) Three years after reconstruction.
1446 operatively. While, the mean dorsal cortical angle at the final assessment was 1478 (SD, 2.8) in comparison to pre-operative value of 1398 (SD, 2.1). The difference between pre- and post-operative values of the three measurements (height to length ratio, lateral interascaphoid and dorsal cortical angles) had no statistical relationship with the clinical outcome. The intra-operatively corrected carpal alignment has almost been kept up, in all cases, till late followup as indicated by the maintenance of the corrected scapholunate and capitolunate angles to within normal ranges. The mean scapholunate angle on the operated side, at the last visit, was 558 (range: 40—708) in comparison with a mean of 608 (range: 50—858) and 488 (range: 40—658) for the pre-operative and sound side, respectively. In eight patients, no radiographic signs of degenerative changes could be detected at the most recent follow-up. While, the remaining five patients showed mild degenerative changes in the form of a mild sharpening of the radial styloid and mild subchondral sclerosis of the distal radius, three of them already had almost the same degree of changes existed pre-operatively. None of the patients had developed severe arthritis at late follow-up.
Complications Transient skin irritation with mild superficial pin track infection occurred in two cases.
Discussion The early and late consequences of scaphoid malunion, when it exceeds certain limits, should be expected to have a deteriorating natural history.4,6,12 Scaphoid malunion alters the carpal kinematics that may correlate with clinical symptoms such as weak grip strength and painful decrease in wrist motion. Such a situation is a pre-arthritic condition and may also predispose to premature wrist arthrosis1,4,16,19 that may spread to show evidence of scapholunate advanced collapse (SLAC) as described by Watson and Ballet (1984).20 The more severe the malunion the worse the prognosis is likely to be. Surgical correction, therefore, should be considered, at least in symptomatic patients. Although, scaphoid fracture malunion is now a well-recognised entity and evidence of an associated poor prognosis are mounting, the reported number of patients treated with scaphoid osteotomy is surprisingly small.8 The risk of creating iatrogenic avascular necrosis of the proximal fragment or converting a healed scaphoid into a non-united one
E.A. El-Karef may be the reasons. Non-symptomatic malunions with acceptable painless limitation of wrist function that do not need further treatment may be another cause as stated by Jiranek et al. (1992).11 Little is written about the indications for, and precise methods of treating scaphoid malunions. However, a corrective osteotomy seems to be a good reconstructive option as it indeed is directed at correction of the pathology and restoring normality, and its targets are to improve functions of the wrist and to reduce or delay the incidence of arthritis. Herbert in 1990 expected that corrective osteotomy of the scaphoid will become an increasingly common operation in the future years.8 Barton (1992)3 stated that it is certainly bold to divide a bone notoriously subject to non-union. Fernandez (1998)7 reviewed the English speaking literature of scaphoid osteotomy and found that neither iatrogenic avascular necrosis nor non-union had complicated any of the cases. Delaying correction until degenerative changes develop and treating the condition with limited or total wrist arthrodesis will lead to the undesirable loss of motion. While, wrist arthroplasty may be an unreliable long-term solution for young patients. Corrective scaphoid osteotomy, in the present work, was indicated only in relentlessly symptomatic malunion particularly in the presence of radiological indicators of bad prognosis such as lateral intrascaphoid angle greater than 558.1 Nevertheless, the degree of pain, stiffness and weakness should be weighed against the possible risks of post-osteotomy complications, non-union or avascular necrosis. The presence of long-established periscaphoid arthritis were considered a formal contraindication and should be carefully ruled out pre-operatively. Standardised pre-operative radiographic assessment of both wrists allowed sensible measurements of normal scaphoid shape and length and good evaluation of the normal and the pathologic intercarpal relationships, which consequently, facilitates better planning through a fair calculation of the size of the graft needed for each patient to restore anatomic length and realignment of the scaphoid. However, despite the extensive pre-operative planning, intra-operative assessment was always essential particularly for the rotational element of the deformity. Several methods have been described for determining the degree of deformation of a malunited scaphoid.1,2,4,8,17,18 The apparent shortening seen on PA radiographs and even that confirmed by CT, may be substantially less than the true shortening. Therefore, interposed graft height always exceeded the pre-operative measurements. Both ulnar and
Scaphoid malunion palmar heights of the graft were almost approaching twice the pre-operatively measured shortening in most of the cases. This observation provided a fairly accurate pre-operative estimation in the later cases of this series. Because of scaphoid irregularity and variability in patient positioning during CT and/or radiography18 and also because of the difficulty in determining where the articular surface ends and the cortex begins2 there is often inability to reproduce scaphoid axes. This leads to poor intra and inter-observer reliability of the lateral interascaphoid angle and adds to the intricacy of pre-operative planning. Moreover, axial CT scanning of the scaphoid is not a straightforward job. It required experience in hand positioning.2,12,17,18 Only few cuts were obtained for the scaphoid bone and the middle cut between the capitate and the radial border of the scaphoid was the one used for the measurements. The inserted graft must be tightly fitted between the scaphoid fragments to provide stability with good apposition of the osteotomised surfaces and to help graft incorporation. A combination of K. wire and screw fixation has been tried in the early two cases (Fig. 5) but was not a favorite due to possible damage of the core of the graft during screw insertion and the screw may hold the fragments apart and prevent collapse if bone resorption occur during healing. However, there was no statistical correlation between method of fixation and any of the clinical parameters at the final assessment. Although none of the patients included developed severe degenerative changes at the wrist, five of the 13 patients had radiographic evidence of mild degenerative changes at the radioscaphoid joint on latest follow-up (mean, 42 months), most of them were asymptomatic. These late findings may suggest that the procedure at least slows the process of degenerative arthritis. However, this depends on the individual functional demands. In the present study, good response to re-establishment of normal scaphoid anatomy following the procedure was documented in the majority of cases with cessation of their symptoms. However, there was persistent pain and positive clinical tests for ligament instability in four cases, despite radiographic evidence of scaphoid realignment. This may be attributed to deep scarring or an initially associated severe perilunate injury.8,15 The encouraging results of this work and those reported in the literature would suggest that corrective scaphoid osteotomy could be indicated in an effort to correct intra-articular incongruity and to restore the anatomic and functional integrity of the wrist joint. It should be performed at an early stage
1447 before development of degenerative changes. The procedure did not create serious complications such as avascular necrosis or nonunion and would also appear to be a measure to prevent late intercarpal collapse and, in theory, should provide better function than late salvage procedures.
Acknowledgements The author is grateful to Dr. Hesham Kotb, MD. Assistant professor in Diagnostic Radiology, Alexandria University, Egypt, for his tremendous help through out this work.
References 1. Amadio PC, Bequist TH, Smith DK, et al. Scaphoid malunion. J Hand Surg 1989;14A:679. 2. Bain GI, Bennett JD, Dermid M, et al. Measurement of the scaphoid deformity using longitudinal computed tomography: intra- and interobserver variability using various measurement technique. J Hand Surg 1998;17A:76. 3. Barton NJ. Twenty questions about scaphoid fractures. J Hand Surg 1992;17B:289. 4. Birchard D, Pichora D. Experimental corrective scaphoid osteotomy for scaphoid malunion with abnormal wrist mechanics. J Hand Surg 1990;15A:863. 5. Burgess RC. The effect of a simulated scaphoid malunion on wrist motion. J Hand Surg 1987;12A:774. 6. Femandez DL. A technique for anterior wedge-shaped grafts for scaphoid nonunions with carpal instability. J Hand Surg 1984;9A:733. 7. Fernandez DL, Martin CL, Gonzalez del Pino J. Scaphoid malunion: the significance of rotational malalignment. J Hand Surg 1998;23B:771. 8. Femandez DL, Eggli S. Scaphoid nonunion and malunion. How to correct deformity. Hand Clin 2001;17:631. 9. Fisk GR. Carpal instability and the fractured scaphoid. Ann R Collage Surg Engl 1970;46:63. 10. Fisk GR. The wrist: review article. J Bone Joint Surg 1984;66B:396. 11. Jiranek WA, Ruby LK, Millender LB, et al. Long-term results after Russe bone-grafting: the effect of malunion of the scaphoid. J Bone Joint Surg 1992;74A:1217. 12. Kozin SH. Incidence, mechanism and natural history of scaphoid fractures. Hand Clin 2001;17:515. 13. Lynch NM, Linscheid RL. Corrective osteotomy for scaphoid malunion: technique and long-term follow-up evaluation. J Hand Surg 1997;22 A:35. 14. Mathiowetz V, Weber K, Volland G, Kashman N. Reliability and validity of hand strength evaluation. J Hand Surg 1984;9A: 222. 15. Mayfield JK. Mechanism of carpal injuries. Clin Orthopaed Relat Res 1980;149:45. 16. Nakamura R, Imaeda T, Miura T. Scaphoid malunion. J Bone Joint Surg 1991;73B:134. 17. Sanders WE. Evaluation of the humpback scaphoid by computed tomography in the longitudinal axial plane of the scaphoid. J Hand Surg 1988;13A:182.
1448 18. Smith DK, Linscheid RL, Amadio PC, Berquist TH, Cooney WP. Scaphoid anatomy: evaluation with complex motion tomography. Radiology 1989;173:177. 19. Tomaino MM, King J, Pizillo M. Correction of lunate malalignment when bone grafting scaphoid nonunion with humpback
E.A. El-Karef deformity: rationale and results of a technique revisited. J Hand Surg 2000;25A:322. 20. Watson HK, Ballet FL. The SLAC wrist: Scapholunate advanced collapse pattern of degenerative arthritis. J Hand Surg 1984;9A:358.