SCIENTIFIC ARTICLE Volar Plate Fixation to Treat Scaphoid Nonunion: A Case Series With Minimum 3 Years of Follow-Up

Ignacio Esteban-Feliu, MD,*† Sergi Barrera-Ochoa, PhD,†‡§ Nuria Vidal-Tarrason, MD,* Bernat Mir-Simon, PhD,k{ Alex Lluch, MD,*# Xavier Mir-Bullo, PhD*†

Purpose Several options exist for treating scaphoid nonunion. For selected cases, some authors recommend using a volar buttress plate. The aim of the study was to report the clinical and radiological outcomes achieved at a minimum of 3 years’ follow-up of treating scaphoid nonunion with a scaphoid volar plate. Methods We retrospectively reviewed 15 patients with symptomatic scaphoid nonunion treated with scaphoid plate osteosynthesis and a graft from the volar aspect of the distal between January 2011 and November 2013. The patients’ average age was 32 years (range, 21e62 years). No patient had undergone previous scaphoid surgery. Clinical assessments included range of motion (ROM), grip strength, and a 10-point visual analog scale rating of pain. Further subjective assessment was performed using validated measurement tools. All patients underwent plain radiography with 6 different views and computed tomography before and after surgery. The scapholunate angle was measured to evaluate carpal alignment. Results The mean follow-up period was 42 months (range, 36e51 months). Bone union was obtained in 13 of 15 patients (87%) at an average of 5 months (range, 3e8 months) after surgery. Significant improvements in the scapholunate angle and visual analog scale pain score were observed. However, both wrist ROM and grip strength remained significantly decreased relative to the opposite side (62% and 55% of normal, respectively). Four patients experienced hardware complications: plate breakage (1 scaphoid) and screw back-out (3 scaphoids). Impingement between the radial styloid and the scaphoid plate was identified in 6 patients. Five patients required additional surgery to remove the plate. Conclusions Our study indicates unsatisfactory results treating scaphoid nonunion with a volar scaphoid plate. Although the rate of union was comparable with that of other series, we noted more complications. We are concerned about both the high proportion of hardware complications and required secondary surgical procedures. (J Surg Am. 2017;-(-):1.e1-e8. Copyright Ó 2017 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic IV. Key words Internal fixation, scaphoid nonunion, volar plate.

CAPHOID FRACTURE IS THE MOST common fracture places it at high risk for nonunion, delayed consoli- affecting the carpal , accounting for dation, or osteonecrosis.2,3 In some scaphoids in S roughly 60% of the total.1 It is well known that which nonunion occurs, collapse into a humpback the inherent poor vascular supply of the scaphoid deformity may lead to a pattern of dorsal intercalated

From the *Hand Surgery Unit, Orthopaedic Surgery, Hospital Universitari Vall d’Hebron; No benefits in any form have been received or will be received related directly or indirectly the †Hand Surgery and Microsurgery Unit, Hospital Universitari Quiron-Dexeus, ICATME; the to the subject of this article. ‡Department of Pediatric Hand Surgery and Microsurgery, Universitat de Barcelona; the Corresponding author: Ignacio Esteban-Feliu, MD, Hand Surgery Unit, Orthopaedic §Department of Pediatric Orthopedic Surgery, Hospital Sant Joan de Deu, Universitat de Surgery, Hospital Universitari Vall d’Hebron, Passeig de la Vall d’Hebron, 119-129, 08035 k ’ Barcelona; the Department of Surgery, UD-Vall d Hebron School of Medicine, Universitat Barcelona, Spain. Autònoma de Barcelona; the {Medcom Advance SA, Viladecans; and #Hand Surgery, Institut Kaplan, Barcelona, Spain. 0363-5023/17/---0001$36.00/0 https://doi.org/10.1016/j.jhsa.2017.12.004 Received for publication February 14, 2017; accepted in revised form December 4, 2017.

Ó 2017 ASSH r Published by Elsevier, Inc. All rights reserved. r 1.e1 1.e2 VOLAR PLATE FIXATION FOR SCAPHOID NONUNION segment instability.4,5 Failure of a findings of punctate bleeding. Three patients were to heal may result in a predictable pattern of wrist diagnosed and treated nonsurgically at first presenta- arthritis, scaphoid nonunion advanced collapse.6 To tion and then developed nonunion. The remaining minimize the risk of arthritis, the goal of treatment patients had misdiagnosed fractures that had never should be to consolidate the fracture with the been treated. None of the patients had previously un- scaphoid in anatomical alignment.7,8 dergone scaphoid surgery. Several options exist for treating scaphoid nonunion.9 However, despite numerous reports, no Surgical technique consensus has been reached as to the best form of The same volar approach between the management.10 Using a single compression screw and the flexor carpi radialis tendon was used to with open reduction and length restoration by volar expose the scaphoid nonunion and the bone graft wedge grafting is considered by many to be the donor site from the distal radius. The wrist capsule treatment of choice, yielding a union rate of as high was opened and the scaphoid viewed. We aggres- e as 94%.11 14 Nonetheless, for very unstable fractures sively debrided the nonunion site until punctate or some nonunions of the scaphoid, some authors bleeding was visualized at both the proximal and the recommend the use of 2 compression screws15 or a distal poles. This debridement was performed with buttress plate.16 a spherical cutter at a low rotation speed under The 1.5-mm scaphoid volar plate (Medartis AG, adequate irrigation to minimize the risk of thermal Basel, Switzerland), used in our study, was designed necrosis. After the proximal and distal ends were to be placed on the volar surface of the scaphoid. Our debrided, a wedge-shaped bone graft was harvested aim was to examine both clinical and radiological from the anterolateral corner of the volar distal radius outcomes after a minimum of 3 years of follow-up in based on the technique of Aguilella and Garcia- patients for whom a scaphoid volar plate was used to Elias.17 Additional cancellous bone was harvested treat scaphoid nonunion. Complication rates and the from the distal radius to further fill defects within the frequency and reasons for secondary salvage pro- proximal and distal poles of the scaphoid. We then cedures were examined. stabilized the scaphoid with a 1.5-mm plate placed on the volar surface in buttress mode (Fig. 1). We aimed to have at least 3 screws on either side of the fracture. PATIENTS AND METHODS All the screws that we placed were nonlocking. Once Clinical and radiographic data were obtained from fixation was complete, the wrist was examined clin- medical records of patients who had undergone open ically and under fluoroscopy. reduction internal fixation with scaphoid plate After surgery, the wrist and were immobi- osteosynthesis for scaphoid nonunion at our institu- lized in a dorsal wrist plaster cast for 3 weeks. A tion between January 2011 and November 2013 removable orthosis was then used for the next 8 to 9 (Table 1). One independent evaluator (I.E.-F.) carried weeks, during which the patient was instructed to out all the assessments. This included 15 patients perform active range of motion (ROM) exercises. The who had symptomatic nonunion of the scaphoid period of immobilization was based on the results of that had persisted for more than 6months. All of clinical and radiographic assessments to assess bone these cases had been diagnosed both clinically and union throughout the follow-up period. radiographically. Prior to data collection, local ethics committee approval was obtained. All patients were Assessments made aware of the treatment objectives, understood Clinical assessments included measuring wrist ROM the risks and benefits of the procedure, and had in flexion, extension, pronation, supination, and ulnar provided signed informed consent. No external and radial deviation; measuring grip strength as a mean funding was received for the study. of 3 attempts, in kilograms, with correction for hand The average age of the patients was 32 years (range, dominance, using a hydraulic hand dynamometer; 21e62 years), and 14 of 15 were men. The dominant asking about return to prefracture levels of activity; and hand had been injured in 12 patients. Patients with asking patients to rate their current level of pain on a capitolunate or pancarpal arthritis were excluded 10-point visual analog scale (0 ¼ no pain; 10 ¼ severe (scaphoid nonunion advanced collapse IIIeIV). Four pain). Further reported assessment included a Quick patients had a humpback deformity, but none had Disabilities of the , Shoulder, and Hand (Quick- , based on the appearance of bone on DASH) score (0 ¼ no disability; 100 ¼ total disability), computed tomography (CT) scan and intraoperative and Mayo Wrist Score (0 ¼ maximal limitations;

J Hand Surg Am. r Vol. -, - 2017 VOLAR PLATE FIXATION FOR SCAPHOID NONUNION 1.e3

TABLE 1. Patients and Results

Treatment Stage Mechanism Before (Herbert) Follow-Up Case Age Sex Side of injury Job Fracture (see Table 3) DISI (mo)

1 38 M Right Fall Manual worker Misdiagnosed D2, stage 2 No 51 2 48 M Right Fall Transporter Misdiagnosed D2, stage 2 No 50 3 27 M Right Fall Manual worker Misdiagnosed D2, stage 2 No 48 4 30 M Right Sport Manual worker Misdiagnosed D2, stage 2 No 41 5 24 M Right Fall Manual worker Misdiagnosed D2, stage3 Yes 39 6 62 W Left Fall Housewife Misdiagnosed D2, stage 3 Yes 46 7 24 M Right Sport Student Misdiagnosed D2, stage 2 No 37 8 24 M Right Sport Student Misdiagnosed D2, stage 2 Yes 40 9 21 M Right Fall Waitress Misdiagnosed D2, stage 2 No 43 10 29 M Right Fall Office worker Misdiagnosed D2, stage 3 Yes 37 11 31 M Right Sport Office worker Cast D2, stage 2 No 41 12 37 M Left Fall Businessman Cast D2, stage 2 No 43 13 25 M Right Fall Unemployed Cast D2, stage 3 No 48 14 30 M Right Sport Manual worker Misdiagnosed D2, stage 2 No 36 15 26 M Right Fall Student Misdiagnosed D1, stage 1 No 36

DISI, dorsal intercalated segment instability.

FIGURE 1: Intraoperative images. A Curettage of sclerotic margins at a nonunion site. B, C A wedge-shaped corticocancellous graft from the volar distal radius is inserted into the volar defect. D Final position of the plate on the volar surface of the scaphoid.

100 ¼ no limitations). All measurements were per- minimum of 3 years. We measured the scapholunate formed by an independent evaluator (I.E.-F.). angle before and after surgery in each patient Before surgery, all patients had plain x-rays and a to evaluate carpal alignment. The x-ray CT scan, with magnetic resonance imaging added for protocol always included 6 different views for all 5 patients over the course of follow-up. Patients were patients—anteroposterior; lateral; lateral with wrist assessed before surgery and after surgery at 10 days, 3 flexion; lateral with wrist extension; and and 6 weeks, 3 and 6 months, and thereafter annually, anteroposterior view with both radial and ulnar including radiological reviews that extended over a deviation—and a CT scan after surgery (Figs. 2, 3).

J Hand Surg Am. r Vol. -, - 2017 1.e4 VOLAR PLATE FIXATION FOR SCAPHOID NONUNION

FIGURE 2: A, B Preoperative x-rays and C CT scan.

FIGURE 3: Postoperative x-rays. A Radial deviation. B Ulnar deviation. C Flexion. D Extension.

Statistical analysis after surgery. The proportion healed at each ascer- Data that were normally distributed were described tainment point was 7% at 3 months, 67% at 6 months, using means, minimum and maximum values, and and 13% at 8 months. We ordered CT scans 3 months SDs, where appropriate. Nonparametric data were after surgery to assess for interval healing. If healing described using medians, minimum and maximum was not confirmed on this CT scan, patients continued values, and interquartile ranges, as appropriate. to have restricted use of their wrist, and repeat scans Differences in pre- and postoperative data were were ordered subsequently. Two patients (cases 3 and identified using paired Student t tests for normally 11) failed to achieve union at the end of follow-up. distributed data and Wilcoxon test for nonparametric Carpal alignment, as indicated by the scapholunate data. To compare the wrist operated on against the angle, was improved after surgery; on average, the unaffected, contralateral wrist, dependent Student mean scapholunate angle improved from 54 ( SD t tests were used. All statistical tests were 2-sided, 9 ) before surgery to 46 ( SD 6 ) after surgery and statistical significance was accepted at the level (P < .05). The patients’ mean visual analog scale pain of P of .05 or less. rating also improved from preoperative (6) to post- operative (4) assessment (P < .05). The postoperative RESULTS DASH score ranged from 43 to 84 (mean, 63), and the The average interval between the initial injury and the postoperative Mayo Wrist Score ranged from 30 to 70 surgical treatment was 19 months (range, 9e34 (mean, 54). months). The mean period of follow-up was 42 The mean ROM (flexion-extension arc) was 94 months (range, 36e51 months) (Tables 2, 3). after surgery. This ROM corresponded to 61% that Bone union was obtained in 13 of 15 patients of the opposite side (154)(P < .05). Grip strength (87%), at an average of 5 months (range, 3e8 months) averaged 30 kg after surgery, which was 55% that of

J Hand Surg Am. r Vol. -, - 2017 TABLE 2. Functional Results

ROM ROM Grip Strength Grip Strength Time to DASH Case Affected Unaffected Affected Unaffected Union (mo) Mayo Wrist Score Score Function Outcome

1 80 200 22 48 8 50 84 Reduced Delayed union OA LT IAINFRSAHI NONUNION SCAPHOID FOR FIXATION PLATE VOLAR Impingement

adSr Am. Surg Hand J 2 110 190 34 68 4 55 52 Full Impingement 3 70 130 38 70 - 35 66 Reduced Nonunion Plate rupture 4 110 110 40 50 5 70 43 Full Good ROM and function 5 110 170 36 52 6 65 51 Full Good ROM and function r 6 120 170 34 54 5 60 62 Full Impingement Vol. 7 80 135 42 52 6 30 64 Full Occasional mild pain - 8 110 145 36 68 3 60 78 Reduced Screw loosening , - 9 70 115 33 54 6 70 54 Full Impingement 2017 10 110 140 42 58 5 35 62 Full Good ROM and function 11 135 130 40 50 - 65 53 Reduced Nonunion Screw loosening 12 70 170 20 38 5 50 81 Reduced Impingement 13 70 190 32 54 6 50 57 Reduced Screw loosening 14 60 170 32 52 8 45 76 Full Delayed union 15 100 140 340 54 4 65 56 Reduced Impingement 1.e5 1.e6 VOLAR PLATE FIXATION FOR SCAPHOID NONUNION

TABLE 3. Scaphoid Fracture: Herbert Classification

Type Stage Description

A Stable acute fractures B Unstable acute fracture (B2 complete fracture of the waist) C Delayed nonunion D1 Fibrous nonunion 1 Stable, no degenerative changes D2 Pseudarthrosis 2 Collapseþ, osteoarthritis 3 Collapseþþ, osteoarthritisþþ 4 Collapseþþþ, osteoarthritisþþþ 5 Avascular necrosis

FIGURE 4: Persistent nonunion with broken plate and screws.

FIGURE 5: Computes tomography demonstrates a large erosion in the distal radius from radiocarpal impingement. A Coronal view. B Sagittal view. the contralateral, unaffected side (55 kg); this was surgery. In a second patient, we performed implant also a statistically significant difference (P < .05). removal and a free vascularized bone graft from the Four patients had hardware complications. One medial femoral condyle using a compression screw experienced breakage of the plate during the fourth and successful bone union was obtained. month of follow-up (Fig. 4). There was screw back- Plain x-rays (with 6 different views) and a CT scan out in 3 scaphoids. were obtained for all our patients after surgery. In 1 patient with persistent nonunion, we removed Impingement between the radial styloid and the only the implant because the patient refused further scaphoid plate, especially during wrist flexion and

J Hand Surg Am. r Vol. -, - 2017 VOLAR PLATE FIXATION FOR SCAPHOID NONUNION 1.e7 radial deviation, was observed in six patients, both on was greater than that reported for other series.23 We plain x-rays and on CT (Fig. 5). Three of these 6 want to emphasize the high mean DASH score in our required additional surgery to remove their scaphoid series because it implies a very substantial degree of plate. patient-reported disability, which is much greater than what is usually reported. Overall, in our series of 15 patients, we observed DISCUSSION unsatisfactory results using a volar scaphoid plate to Treatment of scaphoid nonunion with plates has been treat scaphoid nonunion. Although the rate of union described previously. The use of AO plates for was comparable with that achieved in other series, we scaphoid nonunion was described almost a quarter identified more complications than previously re- century ago by Braun et al.18 Previously, Huene and ported. Relative to using a compression screw, a volar Huene19 had suggested using an Ender compression plate appears to be associated with poorer functional hooked blade plate system to treat difficult cases that outcomes and a greater risk of complications. were not ideally suited to compression screw fixation. Interpretation of our results warrants caution, given The buttress plate was designed for unstable the retrospective nature of this series and its inherent fractures or nonunions with humpback deformities limitations—like the small number of patients studied to counteract the compression forces and the axial and the lack of any control group against which to bending that occurs when the compression screw compare outcomes. As such, multicenter, long-term cannot provide stability.20,21 An additional advantage studies in which patients are randomized to either a of this plate is that it prevents extrusion of the cor- treatment or a control arm remain necessary. However, ticocancellous wedge graft.12 owing to the results of our study, we have moved away To date, the results achieved with volar plating of from the scaphoid volar plate in favor of the headless scaphoid nonunion with a bone graft have been screw treatment. encouraging. Ghoneim22 reported successful bone union in 93% of 14 cases of scaphoid nonunion, REFERENCES achieving more than 70% of the ROM and grip strength observed in the contralateral (unaffected) 1. Alshryda S, Shah A, Odak S, Al-Shryda J, Ilango B, Murali SR. hand. Meanwhile, Leixnering et al23 treated 11 pa- Acute fractures of the scaphoid bone: systematic review and meta- analysis. Surgeon. 2012;10(4):218e229. tients with a scaphoid plate for established nonunion 2. Geissler WB, Slade JF. Fractures of the . In: and achieved a bone union rate of 100% and very few Wolfe SW, Hotchkiss RN, Pederson WC, Kozin SH, eds. Green’s Operative Hand Surgery. 6th ed. Philadelphia: Elsevier; 2011: sequelae, limited to minor symptoms or mildly e 24 639 707. reduced function. Dodds et al used volar buttress 3. Steinmann SP, Adams JE. Scaphoid fractures and nonunions: diag- plates in conjunction with a vascularized bone graft nosis and treatment. J Orthop Sci. 2006;11(4):424e431. in 6 patients, and reported initial outcomes that were 4. Nakamura R, Horii E, Watanabe K, Tsunoda K, Miura T. Scaphoid non-union: factors affecting the functional outcome of open reduction excellent. and wedge grafting with Herbert screw fixation. J Hand Surg Br. In the present study, bony union was achieved in 1993;18(2):219e224. 13 of 15 (87%), which were followed for a 5. Tomaino M, King J, Pizillo M. Correction of lunate malalignment minimum of 3 years. In the literature, reported rates when bone graft scaphoid non-union with humpback deformity: rationale and results of a technique revisited. J Hand Surg Am. of union after headless compression screw and bone 2000;25(2):322e329. e grafting have varied from 71% to 95%.7,12,13,25 28 6. Kawamura K, Chung KC. Treatment of scaphoid fractures and and the mean period for bone union has been 5 nonunions. J Hand Surg Am. 2008;33(6):988e997. 7. Trumble TE, Salas P, Barthel T, Robert KQ. Management of months. The mean period to union reported for other scaphoid nonunions. J Am Acad Orthop Surg. 2003;11(6):380e391. 28 series has been 3.4 months. 8. Brooks S, Wluka AE, Stuckey S, Cicuttini F. The management of In the present study, certain mechanical compli- scaphoid fractures. J Sci Med Sport. 2005;8(2):181e189. 9. Waitayawinyu T, Pfaeffle HJ, McCallister WV, Nemechek NM, cations occurred. One of the reasons could be that Trumble TE. Management of scaphoid nonunions. Orthop Clin North there is only 1 plate design, which we find to be a Am. 2007;38(2):237e249. problem because only 1 plate design cannot accom- 10. Pinder RM, Brkljac M, Rix L, Muir L, Brewster M. Treatment of modate the different sizes and shape of the scaphoid. scaphoid nonunion: a systematic review of the existing evidence. J Hand Surg Am. 2015;40(9):1797e1805. In only 1 other series, involving scaphoid nonunion 11. Merrell GA, Wolfe SW, Slade JF. Treatment of scaphoid nonunions: patients treated with a volar buttress plate,25 minimal quantitative meta-analysis of the literature. J Hand Surg Am. screw back-out was reported in a single case. We also 2002;27(4):685e691. fi fi 12. Cooney WP, Linscheid RL, Dobyns JH, Wood MB. Scaphoid identi ed a signi cant decrease in ROM (61%) in the nonunion: role of anterior interpositional bone grafts. J Hand Surg affected versus the unaffected hand, a decrease that Am. 1988;13(5):635e650.

J Hand Surg Am. r Vol. -, - 2017 1.e8 VOLAR PLATE FIXATION FOR SCAPHOID NONUNION

13. Daly K, Gill P, Magnussen PA, Simonis RB. Established nonunion scaphoid fractures. Arch Orthop Trauma Surg. 2016;136(11): of the scaphoid treated by volar wedge grafting and Herbert screw 1623e1628. fixation. J Bone Joint Surg Br. 1996;78(4):530e534. 22. Ghoneim A. The unstable nonunited scaphoid waist fracture: 14. Radford PJ, Matthewson MH, Meggitt BF. The Herbert screw for results of treatment by open reduction, anterior wedge grafting, and delayed and non-union of scaphoid fractures: a review of fifty cases. internal fixation by volar buttress plate. J Hand Surg Am. 2011;36(1): J Hand Surg Br. 1990;15B:455e459. 17e24. 15. Garcia RM, Leversedge FJ, Aldridge JM, Richard MJ, Ruch DS. 23. Leixnering M, Pezzei C, Weninger P, et al. First experiences with a Scaphoid nonunions treated with 2 headless compression screws and new adjustable plate for osteosynthesis of scaphoid nonunions. bone grafting. J Hand Surg Am. 2014;39(7):1301e1307. J Trauma. 2011;71(4):933e938. 16. Bain GI, Turow A, Phadnis J. Dorsal plating of unstable scaphoid frac- 24. Dodds SD, Patterson JT, Halim A. Volar plate fixation of recalcitrant tures and nonunions. Tech Hand Up Extrem Surg. 2015;19(3):95e100. scaphoid nonunions with volar carpal artery vascularized bone graft. 17. Aguilella L, Garcia-Elias M. The anterolareal corne of the radial Tech Hand Up Extrem Surg. 2014;18(1):2e7. metaphysis as a source of bone graft for the treatment of scaphoid 25. Eggli S, Fernandez DL, Beck T. Unstable scaphoid fracture nonunion. J Hand Surg Am. 2012;37(6):1258e1262. nonunion: a medium-term study of anterior wedge grafting proced- 18. Braun C, Gross G, Bühren V. Osteosynthesis using a buttress plate— ures. J Hand Surg Br. 2002;27(1):36e41. a new principle for stabilizing scaphoid pseudarthroses [in German]. 26. Inoue G, Shionoya K. Herbert screw fixation by limited access for Unfallchirurg. 1993;96(1):9e11. acute fractures of the scaphoid. J Bone Joint Surg. 1997;79(3): 19. Huene DR, Huene DS. Treatment of nonunions of the scaphoid with 418e421. the Ender compression blade plate system. J Hand Surg Am. 27. Shah J, Jones WA. Factors affecting the outcome in 50 cases of 1991;16(5):913e922. scaphoid nonunion treated with Herbert screw fixation. J Hand Surg 20. Muller ME, Allgower M, Schneider R, Willenegger H. Buttress plate. Br. 1998;23(5):680e685. In: Allgower M, ed. Manual of Internal Fixation. Berlin: Springer- 28. Tsuyuguchi Y, Murase T, Hidaka N, Ohno H, Kawai H. Anterior Verlag; 1991:208e210. wedge-shaped bone graft for old scaphoid fractures or non-unions. 21. Jurkowitsch J, Dall’Ara E, Leixnering M. Rotational stability An analysis of relevant carpal alignment. J Hand Surg Br. in screw-fixed scaphoid fractures compared to plate-fixed 1995;20(2):194e200.

J Hand Surg Am. r Vol. -, - 2017