ORIGINAL ARTICLE
Dorsal Vascularized Grafting for Scaphoid Nonunion: A Comparison of Two Surgical Techniques Fábio A. Caporrino, MD, PhD, João Baptista G. Dos Santos, MD, PhD, Fernando T. Penteado, MD, MSc, Vinícius Y. de Moraes, MD, João Carlos Belloti, MD, PhD, and Flávio Faloppa, MD, PhD
Objectives: To compare the radiographic and functional outcomes of 2 surgical techniques for treating scaphoid nonunion.
Design: Randomized prospective study. Patients/Participants: Researchers assessed the outcomes every 2 weeks until bone healing and at discharge. Interventions: (1) Vascularized bone grafting (VBG) using the 1, 2 intercompartmental suprareticular artery and (2) a distal radius nonvascularized bone graft. Main Outcomes/Interventions: Time to union (primary), union rate, and functional outcomes.
Results: Seventy-five patients were followed for 29 months; 2 were lost to final follow-up. Both groups had similar baseline characteristics. The VBG group reached bone union earlier by 12 days (P = 0.002), but union rates were similar (P = 0.312). There was also less ulnar deviation in the VBG group (P = 0.03). There were no other differences between either intervention groups.
Conclusions: Although the VBG group attained earlier union, this may not be clinically meaningful, nor justify the greater technical difficulty and use of resources associated with this intervention. Key Words: scaphoid bone, fracture, nonunion, randomized controlled trial, bone transplantation, pseudarthrosis, prospective studies, consort
Level of Evidence: Therapeutic Level II. See instructions for authors for a complete description of levels of evidence. (J Orthop Trauma 2014;28:e44–e48)
BACKGROUND Scaphoid fractures are common1 and are usually treated by conservative measures with high healing rates.2,3 However, when displaced, this type of fracture is usually treated operatively. Likewise, proximal pole fractures are frequently Accepted for publication July 11, 2013. From the Disciplina de Cirurgia da Mão e Membro Superior, Escola Paulista de Medicina, Universidade Federal de São Paulo, São Paulo, Brazil. The authors report no conflict of interest. Reprints: Fábio A. Caporrino, MD, PhD, Disciplina de Cirurgia da Mão e Membro Superior, Escola Paulista de Medicina, Universidade Federal de São Paulo, Rua Borges Lagoa, 778 São Paulo, Brazil (e-mail: fa.caporrino@ gmail.com). Copyright © 2013 by Lippincott Williams & Wilkins
e44
| www.jorthotrauma.com
treated surgically. Scaphoid fractures have a 10% nonunion rate.4,5 The pain and disability after a neglected scaphoid nonunion are well documented in the literature.6,7 In a clinical setting, the extended period of casting after surgery until union is a concern for both the patient and orthopaedic surgeon.8,9 In a comprehensive systematic review, screws and wedge grafting were reported to have up to a 94% union rate.10 Nonvascularized bone grafting (NVBG) is considered technically easier and less time demanding.11 A vascularized bone graft (VBG) for situations involving small proximal fragments and proximal pole avascular necrosis may be more appropriate.12,13 Practical implications for surgical treatment are crucial for deciding which technique should be used.1,4 Published reports focus on trials that offer surgeons the best surgical technique.12,14,15 Vascularized-enhanced grafts hypothetically could improve union rates, but clinical reports demonstrate heterogeneous union rates ranging from 27% to 100%.1,4,5,16–18 In addition, these results are based on a small sampling and often do not include objective and validated outcome instruments. No robust evidence has been provided for treatment decisions regarding which technique is more effective.10 In addition, no controlled studies have assessed the effectiveness of VBG as an alternative for situations in which proximal pole avascular necrosis is not the major concern,10,19 such as waist and distal poles scaphoid nonunions, failure after screw fixation or NVBG.10,19 We hypothesized that VBG could enhance union rates and achieve union within a shorter time. The aim of this study was to evaluate 2 grafting options for treating scaphoid nonunion and to determine which one is more effective: VBG (as described by Zaidemberg20) or distal radius NVBG.21 Data from this study included radiographic and functional evaluation.
METHODS This study was approved by the local ethics committee, and the study complied with the CONSORT statement.22
Study Participants We included 75 consecutive patients with scaphoid nonunion who were referred to our institution, a university hand surgery section, responsible for the treatment of most hand and wrist injuries in the metropolitan area. Patients were treated on an outpatient surgery basis and were included after J Orthop Trauma Volume 28, Number 3, March 2014
J Orthop Trauma Volume 28, Number 3, March 2014
an initial clinical examination and plain radiograph (posteroanterior, lateral, and 2 oblique views) assessment that confirmed scaphoid nonunion. Nonunion was defined as absence of radiographic signs of union/healing after at least 3 months of conservative treatment. We also considered signs of nonunion as bone resorption, cysts, and sclerosis in line with the fracture. Exclusion criteria included: (1) patients age (,16 years), (2) wrist arthritis (defined as radiocarpal joint narrowing with the presence of arthritic bone cysts), such as that which is secondary to scaphoid nonunion advanced collapse, (3) previous self-reported high-energy wrist trauma, and (4) refusal to participate in this study. We collected demographic data, including age, sex, and work status before the surgical procedure.
Interventions Patients were randomized into 2 intervention groups. Group allocation was performed according to the instructions contained in 75 sealed opaque envelopes sequentially numbered in a random order. A research assistant not linked to the study performed this step. After a patient was admitted for clinical consultation, passed our inclusion criteria, and consented to the surgery, an opaque envelope was attached to the patient’s medical records and opened just before the surgery. Patients underwent an elective operation. A panel of 10 hand surgeons (with a minimum of 5 years surgical experience) performed all surgeries with the aid of hand surgery residents. All surgeries were performed with brachial plexus anesthesia. A tourniquet was used in all the cases and was deflated after pedicle dissection in cases of VBG to check the viability of the graft. Patients were admitted and discharged from the outpatient clinic on the same day. All patients received analgesics (acetaminophen) before surgery if needed. The VBG patients and other patients were maintained in a short-arm cast postoperatively. The NVBG patients were maintained in a long-arm thumb spica cast for a 6-week period. After surgery, patients were scheduled to return for periodic assessment. As union was one major concern, we assessed it by radiographs every 2 weeks, until bone healing was reached. Patients were discharged from clinical and functional assessment when the surgeon considered the follow-up to be satisfactory. This occurred between 24 and 30 months after surgery. At that time, functional assessments were performed.
Description of Surgical Interventions Vascularized Bone Graft A pedicled VBG using the dorsal supraretinacular artery of the distal radius was performed, as described by Zaidemberg.20 The wrist was approached by an oblique dorsoradial incision. The extensor retinaculum was divided, with the first compartment tendon retracted volarly and the second retracted ulnarly. After identification of the blood vessel on the periosteum, a bone graft was centered on the dorsal supraretinacular artery [1,2-intercompartimental supraretinacular vessel (1,2 ICSRA vessel)] that was perfusing the bone block. Ó 2013 Lippincott Williams & Wilkins
Dorsal Vascularized Grafting for Scaphoid Nonunion
Before transfer of the graft, the nonunion site was prepared to receive the graft. Then, the graft and its pedicle were elevated from distal radius to fit to the cavity made in the scaphoid. The graft was then stabilized using 2 Kirschner wires.
NVBG From the Distal Radius The wrist was approached through a longitudinal incision along the flexor carpi radialis.23,24 The flexor carpi radialis tendon was retracted ulnarly. The flexor pollicis longus and the pronator were retracted to expose the radius. A volar rectangular distal radius cortico-cancelous bone graft was removed. The graft was transferred to the scaphoid nonunion site and fixed with Kirschner wires.
Outcomes Radiologic Wrist radiographs were obtained at the time of the initial assessment and every 2 weeks until scaphoid union was reached. Radiographs were standardized as follows: posteroanterior, lateral, and 2 oblique views (semipronated and semisupinated). The nonunion was considered to be healed when there was evidence of bony trabeculae passing through the graft area (posteroanterior and the lateral projections), including the proximal and distal contact area between the graft the scaphoid poles. Assessments were performed by 2 of the authors, both of whom performed or participated in most of the surgical procedures. When consensus could not be reached, a CT scan was obtained to verify bridging trabeculae.
Functional At the end of the follow-up period, patients were evaluated by global criteria, as proposed by Cooney.25 These criteria consider pain, functional status, range of motion, and wrist strength. For each of the above criteria, a total of 25 points could be assigned. The sum of these criteria grades the intervention result as excellent (90–100 points), good (80–90 points), regular (65–80 points), and poor (,65 points). Wrist grip strength was assessed using the Jamar dynamometer (Jamar, Clifton, NJ). The results from these measurements were expressed as the difference in the values between the affected and unaffected sides (absolute index of limitation) after cast removal. Active wrist range of motion (wrist flexion and extension) was measured using a goniometer. We performed measurements for the affected and unaffected wrist, positioning the goniometer along the dorsum of the wrist. A therapist independent of the study performed these measurements. Time between the surgery and return to work activities was measured in days following the index procedure. Return to work was then categorized as: (1) return to work with no restrictions, (2) return to work with restrictions, and (3) no return to work. Pain was assessed subjectively: no/mild pain (pain with wrist loading only), moderate pain (symptoms present during daily activities), and intense pain (spontaneous, requiring narcotics). We did not assess surgical complications in this study. www.jorthotrauma.com |
e45
J Orthop Trauma Volume 28, Number 3, March 2014
Caporrino et al
Randomization Method Statistical Methods and Sample Size Determination An experienced statistician performed all analyses. At the beginning of the study, a power analysis was performed with the aim to detect a 7-day mean difference for the above interventions for scaphoid time to union during postoperative follow-up. We believe that a 1-week difference is the minimum acceptable difference between the groups in which a patient/surgeon may decide to change the implemented technique to reach this benefit. We performed a sample size calculation using a power of 0.90 with 0.95 confidence and a 12-day SD (as previously documented11), yielding a sample size of 35 patients per group. Forty patients were included per intervention group to account for any potential loss to follow-up. Descriptive statistics were performed by calculating the mean and SD for numerical data. For categorical data, numbers and percentages were used. To assure the normality of the data, 2 independent methods were used: (1) visual analysis of the distribution of histograms, and (2) the Kolmogorov–Smirnov test. If any of these indicated that the data were not normally distributed, a nonparametric statistical test was used. For comparing mean values, Student t test was used most commonly, except when data lacked normality of distribution, and the Mann–Whitney U test was performed instead. For categorical data, the x2 test was used most commonly, except when the data lacked normality of distribution, and the Fisher exact test was used instead. No statistical corrections were implemented for missing data (missing at random principle). We considered P , 0.05 as significant and provided 95% confidence interval for numerical data. All the analyses followed the intention-to-treat principle.
RESULTS
Of 75 patients, 73 patients reached the final follow-up (NVBG mean = 28.6 months, VBG mean = 29.4 months). Two patients did not complete the final follow-up assessment, both from the NVBG group. There were no significant differences between participants’ baseline data for demographical, radiographic, and functional outcomes (Tables 1 and 2). For the main outcomes, we found that union was achieved 12 days earlier in the VBG group (Fisher exact test, P = 0.002). No difference in the union rate was found between the groups (88.6% for VBG vs. 80.0% for NVBG; x2 test, P = 0.312). Ulnar deviation was decreased in the VBG group (Student t test, P = 0.003), and a trend showing less radial deviation was found in the VBG group as well (Student t test, P = 0.064). No significant differences were found for the remaining outcomes, as shown in Tables 3 and 4.
DISCUSSION Published reports document that union rates after treatment of a scaphoid nonunion are high for most of the proposed surgical techniques.1,13,26 The motivation and background biologic rationale for this study were based on our
e46
| www.jorthotrauma.com
TABLE 1. Baseline Data Mean (SD)
Variable
Technique
Age, y
NVBG VBG NVBG VBG NVBG VBG NVBG VBG
29.1 26.1 27.6 36.9 74.6 70.7 76.3 79.1
NVBG VBG NVBG VBG NVBG VBG NVBG VBG
15.7 14.3 31.3 29.7 60 57 55.5 53.8
Time to surgery, mo Wrist grip strength, kgf† Flexion + extension†, degrees Radial deviation, degrees Ulnar deviation, degrees Scapholunate angle, degrees Contralateral scapholunate angle, degrees
Range
N
P
(8.11) (7.1) (25.1) (49.1) (19.3) (23.6) (15.7) (16.3)
18–53 18–56 3–84 6–240 30–120 21–133 47–98 10–110
40 35 40 35 40 35 40 35
0.100
(6.7) (6) (6.7) (8.4) (10.3) (10.1) (6) (5.5)
0–30 0–30 18–46 14–44 46 38 48 42
40 35 40 35 40 35 40 35
0.440* 0.456 0.454
0.354 0.367 0.205 0.201
*Mann–Whitney U test. For the remaining overall analyses, Student t test was used. †Expressed as a percentage compared with the nonaffected side.
belief that a VBG could aid in increasing the local blood supply, leading to the promotion of earlier bone healing. The results of this study demonstrate that union was indeed observed earlier when the scaphoid nonunion was treated with a VBG. For functional outcomes however, including range of motion, wrist strength, and Cooney criteria, no significant differences were found, with the exception of ulnar deviation, which demonstrated a mean of 4 degree difference favoring NVBG. We believe that this ulnar deviation difference is not clinically relevant and the tolerance limits for accuracy of measurement of this range are likely to be greater than this difference. We feel that future studies should assess range of motion as a primary end point to demonstrate whether VBG is related to decreased range of motion.
TABLE 2. Patient Characteristics Characteristics* Gender (male) Heavy laboral activity Handness (right) Affected hand (right) Recognized previous trauma Previous treatment Preoperative pain Nonpain or mild Moderate or intense pain Localization (waist) Localization (proximal pole)
NVBG
VBG
P
38 20 36 20 38 15
(95) (50) (90) (50) (95) (37.5)
33 23 31 18 32 16
(94.3) (65.7) (88.6) (51.4) (91.4) (45.7)
.0.999† 0.170 .0.999† 0.902 0.659† 0.471
8 32 30 8
(20) (80) (75) (20)
9 26 23 2
(25.7) (74.3) (65.7) (5.7)
0.555 0.378 0.175†
*Data expressed as n (%). †Fisher exact test. For the overall analyses, x2 test was used.
Ó 2013 Lippincott Williams & Wilkins
J Orthop Trauma Volume 28, Number 3, March 2014
Dorsal Vascularized Grafting for Scaphoid Nonunion
TABLE 3. Outcomes at Patient Discharge Variable*
Technique
Follow-up, mo
NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG
Time to union, d Wrist strength, kgf‡ Range of motion‡, degrees Radial deviation, degrees Ulnar deviation, degrees Time to return to work, wk Scapholunate angle, degrees Cooney functional score
Mean (SD) 28.6 29.4 69.7 58 86.1 89.2 78.1 71.5 15.2 12.6 29.4 25.4 9.3 9.3 56.7 54.4 75.2 74.5
(8.3) (7.7) (15.1) (10.3) (12.6) (15) (12.3) (17.8) (5.3) (5.6) (5.8) (8.5) (2.9) (4.1) (6.5) (6.9) (5.6) (7.2)
Range
N
P
20–48 18–46 42–97 28–75 60–127 58–142 50–98 13–104 6–26 6–26 18–40 10–40 4–16 2–18 48–72 44–70 60–90 60–90
38 35 38 35 38 35 38 35 38 35 38 35 38 35 38 35 38 35
0.487† 0.002† 0.384 0.092 0.064 0.033 0.980 0.177 0.662†
*Data were missing for 2 patients due to follow-up loss (NVBG group). †Mann–Whitney U test. For the overall analyses, Student t test was used. ‡Expressed as a percentage compared with the nonaffected side. p , 0.05 considered as significant.
TABLE 4. Outcomes at Patient Discharge Increased density (Yes) Shortening (Yes) Resorption (Yes) Union (Yes) Postoperative pain score Absent or mild pain Moderate or intense pain Laboral status No return With restrictions No restrictions
NVBG n (%)* 9 21 38 32
(22.5) (52.5) (100) (80)
P
VBG n (%) 8 19 35 31
(22.9) (54.3) (100) (88.6)
0.971 0.558 † 0.312
32 (84.2) 6 (15.8)
34 1
0.109‡
9 (22.5) 6 (15) 25 (62.5)
4 (11.4) 7 (20) 24 (68.6)
0.429
*Data were missing for 2 patients because of follow-up loss (NVBG group). †Impossible to calculate. ‡Fisher exact test. For the overall analyses, x2 test was used.
Ó 2013 Lippincott Williams & Wilkins
ACKNOWLEDGMENTS The authors thank Ms. Katelyn Godin for the final revision of the text. REFERENCES
It is important to acknowledge the limitations of this study. We did not perform a final follow-up assessment based on a well-defined schedule and most occurred between the 24th and 30th week after surgery. We measured union by plain radiographs, which are known to incur in false-positive rates.27 CT was used when we had difficulties in assessing the union status. We did not perform the evaluations using blinded and independent assessors for both radiographs and functional outcomes, which may have introduced bias.19,28 Our findings are also limited because we did not consider complications or cost-effectiveness associated with earlier union. Future studies should examine these factors. Last, it would have been beneficial to perform a subgroup analysis on
Variable*
those patients receiving Workers’ Compensation for their fracture as this type of patient has been shown to have a worse prognosis based on their compensation status.29 Our sample was powered to determine differences for time to union; therefore, other differences might not have been encountered. Researchers should consider our methodological limitations when considering further prospective studies. The result of this study demonstrates the benefits of performing a VBG for treating scaphoid nonunion with no recognized avascular necrosis. The procedure yields similar functional results to NVGB. Earlier time to union could be a theoretical benefit in improving function, however, it did not seem to affect function in this study. Therefore, one should balance the real advantage of performing a more technically demanding procedure with the aim of promoting this earlier union.
1. Gray RR, Shin AY. Vascularized bone grafting of scaphoid nonunions. Oper Tech Sports Med. 2010;18:155–162. 2. Dawson JS, Martel AL, Davis TR. Scaphoid blood flow and acute fracture healing. A dynamic MRI study with enhancement with gadolinium. J Bone Joint Surg Br. 2001;83:809–814. 3. Bhat M, McCarthy M, Davis TR, et al. MRI and plain radiography in the assessment of displaced fractures of the waist of the carpal scaphoid. J Bone Joint Surg Br. 2004;86:705–713. 4. Kawamura K, Chung KC. Treatment of scaphoid fractures and nonunions. J Hand Surg Am. 2008;33:988–997. 5. Cooney WP, Dobyns JH, Linscheid RL. Fractures of the scaphoid: a rational approach to management. Clin Orthop Relat Res. 1980;149:90–97. 6. Sauerbier M, Bickert B, Trankle M, et al. Surgical treatment possibilities of advanced carpal collapse (SNAC/SLAC wrist). Unfallchirurg. 2000; 103:564–571. 7. Krimmer H, Krapohl B, Sauerbier M, et al. Post-traumatic carpal collapse (SLAC- and SNAC-wrist)-stage classification and therapeutic possibilities. Handchir Mikrochir Plast Chir. 1997;29:228–233. 8. Pao VS, Chang J. Scaphoid nonunion: diagnosis and treatment. Plast Reconstr Surg. 2003;112:1666–1676; quiz 1677; discussion 1678–1679. 9. Jaroma H, Suomalainen O, Hannikainen E, et al. Surgical treatment for nonunion of the carpal scaphoid. Ann Chir Gynaecol. 1985;74:239–243. 10. Merrell GA, Wolfe SW, Slade JF III. Treatment of scaphoid nonunions: quantitative meta-analysis of the literature. J Hand Surg Am. 2002;27: 685–691. 11. Caporrino FA, Faloppa F, Santos JBG, et al. Tratamento cirúrgico da pseudartrose do escafóide com enxerto ósseo vascularizado da extremidade dorsal e distal do rádio, baseado na artéia supra-retinacular intercompartimental 1,2. Rev Bras Ortop. 2003;38:522–533. 12. Waitayawinyu T, McCallister WV, Katolik LI, et al. Outcome after vascularized bone grafting of scaphoid nonunions with avascular necrosis. J Hand Surg Am. 2009;34:387–394. 13. Kakar S, Bishop AT, Shin AY. Role of vascularized bone grafts in the treatment of scaphoid nonunions associated with proximal pole avascular necrosis and carpal collapse. J Hand Surg Am. 2011;36:722–725; quiz 725. 14. Malizos KN, Zachos V, Dailiana ZH, et al. Scaphoid nonunions: management with vascularized bone grafts from the distal radius: a clinical and functional outcome study. Plast Reconstr Surg. 2007; 119:1513–1525. 15. Moraes VY, Belloti JC, Moraes FY, et al. Hierarchy of evidence relating to hand surgery in Brazilian orthopedic journals. Sao Paulo Med J. 2011; 129:94–98.
www.jorthotrauma.com |
e47
Caporrino et al
16. Alva D, Vijayaraghavan J, Kotecha A, et al. Treatment of scaphoid nonunion with combined vascularized distal radius graft and iliac crest bone graft. Tech Hand Up Extrem Surg. 2008;12:132–135. 17. Beck E. Russe surgical procedures in treatment of scaphoid pseudarthrosis (Russe I, Russe II) [in German]. Handchir Mikrochir Plast Chir. 1992;24:59–61. 18. Mathoulin C, Gras M, Roukos S. Vascularized bone grafting from the volar distal radius for carpal bones reconstruction [in French]. Chir Main. 2010;29(suppl 1):S65–S76. 19. Rizzo M, Moran SL. Vascularized bone grafts and their applications in the treatment of carpal pathology. Semin Plast Surg. 2008;22: 213–227. 20. Zaidemberg C, Siebert JW, Angrigiani C. A new vascularized bone graft for scaphoid nonunion. J Hand Surg Am. 1991;16:474–478. 21. Adams BD, Blair WF, Reagan DS, et al. Technical factors related to Herbert screw fixation. J Hand Surg Am. 1988;13:893–899. 22. Altman D, Schulz K, Moher D, et al. The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med. 2001;134:663–694.
e48
| www.jorthotrauma.com
J Orthop Trauma Volume 28, Number 3, March 2014
23. Russe O. Fracture of the carpal navicular. Diagnosis, non-operative treatment, and operative treatment. J Bone Joint Surg Am. 1960;42-A: 759–768. 24. Matti H. Technik und resultate meiner pseudarthrosenoperation. Zbl Fur Chir. 1936;63:1442–1453. 25. Cooney WP, Bussey R, Dobyns JH, et al. Difficult wrist fractures. Perilunate fracture-dislocations of the wrist. Clin Orthop Relat Res. 1987; 214:136–147. 26. Matsuki H, Ishikawa J, Iwasaki N, et al. Non-vascularized bone graft with Herbert-type screw fixation for proximal pole scaphoid nonunion. J Orthop Sci. 2011;16:749–755. 27. Steinmann SP, Adams JE. Scaphoid fractures and nonunions: diagnosis and treatment. J Orthop Sci. 2006;11:424–431. 28. Straw RG, Davis TR, Dias JJ. Scaphoid nonunion: treatment with a pedicled vascularized bone graft based on the 1,2 intercompartmental supraretinacular branch of the radial artery. J Hand Surg Br. 2002;27:413. 29. de Moraes VY, Godin K, Tamaoki MJS, et al. Workers’ compensation status: does it affect orthopaedic surgery outcomes? A meta analysis. PLoS One. 2012;7:e50251.
Ó 2013 Lippincott Williams & Wilkins
Dorsal Vascularized Grafting for Scaphoid Nonunion: A Comparison of Two Surgical Techniques Fábio A. Caporrino, MD, PhD, João Baptista G. Dos Santos, MD, PhD, Fernando T. Penteado, MD, MSc, Vinícius Y. de Moraes, MD, João Carlos Belloti, MD, PhD, and Flávio Faloppa, MD, PhD
Objectives: To compare the radiographic and functional outcomes of 2 surgical techniques for treating scaphoid nonunion.
Design: Randomized prospective study. Patients/Participants: Researchers assessed the outcomes every 2 weeks until bone healing and at discharge. Interventions: (1) Vascularized bone grafting (VBG) using the 1, 2 intercompartmental suprareticular artery and (2) a distal radius nonvascularized bone graft. Main Outcomes/Interventions: Time to union (primary), union rate, and functional outcomes.
Results: Seventy-five patients were followed for 29 months; 2 were lost to final follow-up. Both groups had similar baseline characteristics. The VBG group reached bone union earlier by 12 days (P = 0.002), but union rates were similar (P = 0.312). There was also less ulnar deviation in the VBG group (P = 0.03). There were no other differences between either intervention groups.
Conclusions: Although the VBG group attained earlier union, this may not be clinically meaningful, nor justify the greater technical difficulty and use of resources associated with this intervention. Key Words: scaphoid bone, fracture, nonunion, randomized controlled trial, bone transplantation, pseudarthrosis, prospective studies, consort
Level of Evidence: Therapeutic Level II. See instructions for authors for a complete description of levels of evidence. (J Orthop Trauma 2014;28:e44–e48)
BACKGROUND Scaphoid fractures are common1 and are usually treated by conservative measures with high healing rates.2,3 However, when displaced, this type of fracture is usually treated operatively. Likewise, proximal pole fractures are frequently Accepted for publication July 11, 2013. From the Disciplina de Cirurgia da Mão e Membro Superior, Escola Paulista de Medicina, Universidade Federal de São Paulo, São Paulo, Brazil. The authors report no conflict of interest. Reprints: Fábio A. Caporrino, MD, PhD, Disciplina de Cirurgia da Mão e Membro Superior, Escola Paulista de Medicina, Universidade Federal de São Paulo, Rua Borges Lagoa, 778 São Paulo, Brazil (e-mail: fa.caporrino@ gmail.com). Copyright © 2013 by Lippincott Williams & Wilkins
e44
| www.jorthotrauma.com
treated surgically. Scaphoid fractures have a 10% nonunion rate.4,5 The pain and disability after a neglected scaphoid nonunion are well documented in the literature.6,7 In a clinical setting, the extended period of casting after surgery until union is a concern for both the patient and orthopaedic surgeon.8,9 In a comprehensive systematic review, screws and wedge grafting were reported to have up to a 94% union rate.10 Nonvascularized bone grafting (NVBG) is considered technically easier and less time demanding.11 A vascularized bone graft (VBG) for situations involving small proximal fragments and proximal pole avascular necrosis may be more appropriate.12,13 Practical implications for surgical treatment are crucial for deciding which technique should be used.1,4 Published reports focus on trials that offer surgeons the best surgical technique.12,14,15 Vascularized-enhanced grafts hypothetically could improve union rates, but clinical reports demonstrate heterogeneous union rates ranging from 27% to 100%.1,4,5,16–18 In addition, these results are based on a small sampling and often do not include objective and validated outcome instruments. No robust evidence has been provided for treatment decisions regarding which technique is more effective.10 In addition, no controlled studies have assessed the effectiveness of VBG as an alternative for situations in which proximal pole avascular necrosis is not the major concern,10,19 such as waist and distal poles scaphoid nonunions, failure after screw fixation or NVBG.10,19 We hypothesized that VBG could enhance union rates and achieve union within a shorter time. The aim of this study was to evaluate 2 grafting options for treating scaphoid nonunion and to determine which one is more effective: VBG (as described by Zaidemberg20) or distal radius NVBG.21 Data from this study included radiographic and functional evaluation.
METHODS This study was approved by the local ethics committee, and the study complied with the CONSORT statement.22
Study Participants We included 75 consecutive patients with scaphoid nonunion who were referred to our institution, a university hand surgery section, responsible for the treatment of most hand and wrist injuries in the metropolitan area. Patients were treated on an outpatient surgery basis and were included after J Orthop Trauma Volume 28, Number 3, March 2014
J Orthop Trauma Volume 28, Number 3, March 2014
an initial clinical examination and plain radiograph (posteroanterior, lateral, and 2 oblique views) assessment that confirmed scaphoid nonunion. Nonunion was defined as absence of radiographic signs of union/healing after at least 3 months of conservative treatment. We also considered signs of nonunion as bone resorption, cysts, and sclerosis in line with the fracture. Exclusion criteria included: (1) patients age (,16 years), (2) wrist arthritis (defined as radiocarpal joint narrowing with the presence of arthritic bone cysts), such as that which is secondary to scaphoid nonunion advanced collapse, (3) previous self-reported high-energy wrist trauma, and (4) refusal to participate in this study. We collected demographic data, including age, sex, and work status before the surgical procedure.
Interventions Patients were randomized into 2 intervention groups. Group allocation was performed according to the instructions contained in 75 sealed opaque envelopes sequentially numbered in a random order. A research assistant not linked to the study performed this step. After a patient was admitted for clinical consultation, passed our inclusion criteria, and consented to the surgery, an opaque envelope was attached to the patient’s medical records and opened just before the surgery. Patients underwent an elective operation. A panel of 10 hand surgeons (with a minimum of 5 years surgical experience) performed all surgeries with the aid of hand surgery residents. All surgeries were performed with brachial plexus anesthesia. A tourniquet was used in all the cases and was deflated after pedicle dissection in cases of VBG to check the viability of the graft. Patients were admitted and discharged from the outpatient clinic on the same day. All patients received analgesics (acetaminophen) before surgery if needed. The VBG patients and other patients were maintained in a short-arm cast postoperatively. The NVBG patients were maintained in a long-arm thumb spica cast for a 6-week period. After surgery, patients were scheduled to return for periodic assessment. As union was one major concern, we assessed it by radiographs every 2 weeks, until bone healing was reached. Patients were discharged from clinical and functional assessment when the surgeon considered the follow-up to be satisfactory. This occurred between 24 and 30 months after surgery. At that time, functional assessments were performed.
Description of Surgical Interventions Vascularized Bone Graft A pedicled VBG using the dorsal supraretinacular artery of the distal radius was performed, as described by Zaidemberg.20 The wrist was approached by an oblique dorsoradial incision. The extensor retinaculum was divided, with the first compartment tendon retracted volarly and the second retracted ulnarly. After identification of the blood vessel on the periosteum, a bone graft was centered on the dorsal supraretinacular artery [1,2-intercompartimental supraretinacular vessel (1,2 ICSRA vessel)] that was perfusing the bone block. Ó 2013 Lippincott Williams & Wilkins
Dorsal Vascularized Grafting for Scaphoid Nonunion
Before transfer of the graft, the nonunion site was prepared to receive the graft. Then, the graft and its pedicle were elevated from distal radius to fit to the cavity made in the scaphoid. The graft was then stabilized using 2 Kirschner wires.
NVBG From the Distal Radius The wrist was approached through a longitudinal incision along the flexor carpi radialis.23,24 The flexor carpi radialis tendon was retracted ulnarly. The flexor pollicis longus and the pronator were retracted to expose the radius. A volar rectangular distal radius cortico-cancelous bone graft was removed. The graft was transferred to the scaphoid nonunion site and fixed with Kirschner wires.
Outcomes Radiologic Wrist radiographs were obtained at the time of the initial assessment and every 2 weeks until scaphoid union was reached. Radiographs were standardized as follows: posteroanterior, lateral, and 2 oblique views (semipronated and semisupinated). The nonunion was considered to be healed when there was evidence of bony trabeculae passing through the graft area (posteroanterior and the lateral projections), including the proximal and distal contact area between the graft the scaphoid poles. Assessments were performed by 2 of the authors, both of whom performed or participated in most of the surgical procedures. When consensus could not be reached, a CT scan was obtained to verify bridging trabeculae.
Functional At the end of the follow-up period, patients were evaluated by global criteria, as proposed by Cooney.25 These criteria consider pain, functional status, range of motion, and wrist strength. For each of the above criteria, a total of 25 points could be assigned. The sum of these criteria grades the intervention result as excellent (90–100 points), good (80–90 points), regular (65–80 points), and poor (,65 points). Wrist grip strength was assessed using the Jamar dynamometer (Jamar, Clifton, NJ). The results from these measurements were expressed as the difference in the values between the affected and unaffected sides (absolute index of limitation) after cast removal. Active wrist range of motion (wrist flexion and extension) was measured using a goniometer. We performed measurements for the affected and unaffected wrist, positioning the goniometer along the dorsum of the wrist. A therapist independent of the study performed these measurements. Time between the surgery and return to work activities was measured in days following the index procedure. Return to work was then categorized as: (1) return to work with no restrictions, (2) return to work with restrictions, and (3) no return to work. Pain was assessed subjectively: no/mild pain (pain with wrist loading only), moderate pain (symptoms present during daily activities), and intense pain (spontaneous, requiring narcotics). We did not assess surgical complications in this study. www.jorthotrauma.com |
e45
J Orthop Trauma Volume 28, Number 3, March 2014
Caporrino et al
Randomization Method Statistical Methods and Sample Size Determination An experienced statistician performed all analyses. At the beginning of the study, a power analysis was performed with the aim to detect a 7-day mean difference for the above interventions for scaphoid time to union during postoperative follow-up. We believe that a 1-week difference is the minimum acceptable difference between the groups in which a patient/surgeon may decide to change the implemented technique to reach this benefit. We performed a sample size calculation using a power of 0.90 with 0.95 confidence and a 12-day SD (as previously documented11), yielding a sample size of 35 patients per group. Forty patients were included per intervention group to account for any potential loss to follow-up. Descriptive statistics were performed by calculating the mean and SD for numerical data. For categorical data, numbers and percentages were used. To assure the normality of the data, 2 independent methods were used: (1) visual analysis of the distribution of histograms, and (2) the Kolmogorov–Smirnov test. If any of these indicated that the data were not normally distributed, a nonparametric statistical test was used. For comparing mean values, Student t test was used most commonly, except when data lacked normality of distribution, and the Mann–Whitney U test was performed instead. For categorical data, the x2 test was used most commonly, except when the data lacked normality of distribution, and the Fisher exact test was used instead. No statistical corrections were implemented for missing data (missing at random principle). We considered P , 0.05 as significant and provided 95% confidence interval for numerical data. All the analyses followed the intention-to-treat principle.
RESULTS
Of 75 patients, 73 patients reached the final follow-up (NVBG mean = 28.6 months, VBG mean = 29.4 months). Two patients did not complete the final follow-up assessment, both from the NVBG group. There were no significant differences between participants’ baseline data for demographical, radiographic, and functional outcomes (Tables 1 and 2). For the main outcomes, we found that union was achieved 12 days earlier in the VBG group (Fisher exact test, P = 0.002). No difference in the union rate was found between the groups (88.6% for VBG vs. 80.0% for NVBG; x2 test, P = 0.312). Ulnar deviation was decreased in the VBG group (Student t test, P = 0.003), and a trend showing less radial deviation was found in the VBG group as well (Student t test, P = 0.064). No significant differences were found for the remaining outcomes, as shown in Tables 3 and 4.
DISCUSSION Published reports document that union rates after treatment of a scaphoid nonunion are high for most of the proposed surgical techniques.1,13,26 The motivation and background biologic rationale for this study were based on our
e46
| www.jorthotrauma.com
TABLE 1. Baseline Data Mean (SD)
Variable
Technique
Age, y
NVBG VBG NVBG VBG NVBG VBG NVBG VBG
29.1 26.1 27.6 36.9 74.6 70.7 76.3 79.1
NVBG VBG NVBG VBG NVBG VBG NVBG VBG
15.7 14.3 31.3 29.7 60 57 55.5 53.8
Time to surgery, mo Wrist grip strength, kgf† Flexion + extension†, degrees Radial deviation, degrees Ulnar deviation, degrees Scapholunate angle, degrees Contralateral scapholunate angle, degrees
Range
N
P
(8.11) (7.1) (25.1) (49.1) (19.3) (23.6) (15.7) (16.3)
18–53 18–56 3–84 6–240 30–120 21–133 47–98 10–110
40 35 40 35 40 35 40 35
0.100
(6.7) (6) (6.7) (8.4) (10.3) (10.1) (6) (5.5)
0–30 0–30 18–46 14–44 46 38 48 42
40 35 40 35 40 35 40 35
0.440* 0.456 0.454
0.354 0.367 0.205 0.201
*Mann–Whitney U test. For the remaining overall analyses, Student t test was used. †Expressed as a percentage compared with the nonaffected side.
belief that a VBG could aid in increasing the local blood supply, leading to the promotion of earlier bone healing. The results of this study demonstrate that union was indeed observed earlier when the scaphoid nonunion was treated with a VBG. For functional outcomes however, including range of motion, wrist strength, and Cooney criteria, no significant differences were found, with the exception of ulnar deviation, which demonstrated a mean of 4 degree difference favoring NVBG. We believe that this ulnar deviation difference is not clinically relevant and the tolerance limits for accuracy of measurement of this range are likely to be greater than this difference. We feel that future studies should assess range of motion as a primary end point to demonstrate whether VBG is related to decreased range of motion.
TABLE 2. Patient Characteristics Characteristics* Gender (male) Heavy laboral activity Handness (right) Affected hand (right) Recognized previous trauma Previous treatment Preoperative pain Nonpain or mild Moderate or intense pain Localization (waist) Localization (proximal pole)
NVBG
VBG
P
38 20 36 20 38 15
(95) (50) (90) (50) (95) (37.5)
33 23 31 18 32 16
(94.3) (65.7) (88.6) (51.4) (91.4) (45.7)
.0.999† 0.170 .0.999† 0.902 0.659† 0.471
8 32 30 8
(20) (80) (75) (20)
9 26 23 2
(25.7) (74.3) (65.7) (5.7)
0.555 0.378 0.175†
*Data expressed as n (%). †Fisher exact test. For the overall analyses, x2 test was used.
Ó 2013 Lippincott Williams & Wilkins
J Orthop Trauma Volume 28, Number 3, March 2014
Dorsal Vascularized Grafting for Scaphoid Nonunion
TABLE 3. Outcomes at Patient Discharge Variable*
Technique
Follow-up, mo
NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG NVBG VBG
Time to union, d Wrist strength, kgf‡ Range of motion‡, degrees Radial deviation, degrees Ulnar deviation, degrees Time to return to work, wk Scapholunate angle, degrees Cooney functional score
Mean (SD) 28.6 29.4 69.7 58 86.1 89.2 78.1 71.5 15.2 12.6 29.4 25.4 9.3 9.3 56.7 54.4 75.2 74.5
(8.3) (7.7) (15.1) (10.3) (12.6) (15) (12.3) (17.8) (5.3) (5.6) (5.8) (8.5) (2.9) (4.1) (6.5) (6.9) (5.6) (7.2)
Range
N
P
20–48 18–46 42–97 28–75 60–127 58–142 50–98 13–104 6–26 6–26 18–40 10–40 4–16 2–18 48–72 44–70 60–90 60–90
38 35 38 35 38 35 38 35 38 35 38 35 38 35 38 35 38 35
0.487† 0.002† 0.384 0.092 0.064 0.033 0.980 0.177 0.662†
*Data were missing for 2 patients due to follow-up loss (NVBG group). †Mann–Whitney U test. For the overall analyses, Student t test was used. ‡Expressed as a percentage compared with the nonaffected side. p , 0.05 considered as significant.
TABLE 4. Outcomes at Patient Discharge Increased density (Yes) Shortening (Yes) Resorption (Yes) Union (Yes) Postoperative pain score Absent or mild pain Moderate or intense pain Laboral status No return With restrictions No restrictions
NVBG n (%)* 9 21 38 32
(22.5) (52.5) (100) (80)
P
VBG n (%) 8 19 35 31
(22.9) (54.3) (100) (88.6)
0.971 0.558 † 0.312
32 (84.2) 6 (15.8)
34 1
0.109‡
9 (22.5) 6 (15) 25 (62.5)
4 (11.4) 7 (20) 24 (68.6)
0.429
*Data were missing for 2 patients because of follow-up loss (NVBG group). †Impossible to calculate. ‡Fisher exact test. For the overall analyses, x2 test was used.
Ó 2013 Lippincott Williams & Wilkins
ACKNOWLEDGMENTS The authors thank Ms. Katelyn Godin for the final revision of the text. REFERENCES
It is important to acknowledge the limitations of this study. We did not perform a final follow-up assessment based on a well-defined schedule and most occurred between the 24th and 30th week after surgery. We measured union by plain radiographs, which are known to incur in false-positive rates.27 CT was used when we had difficulties in assessing the union status. We did not perform the evaluations using blinded and independent assessors for both radiographs and functional outcomes, which may have introduced bias.19,28 Our findings are also limited because we did not consider complications or cost-effectiveness associated with earlier union. Future studies should examine these factors. Last, it would have been beneficial to perform a subgroup analysis on
Variable*
those patients receiving Workers’ Compensation for their fracture as this type of patient has been shown to have a worse prognosis based on their compensation status.29 Our sample was powered to determine differences for time to union; therefore, other differences might not have been encountered. Researchers should consider our methodological limitations when considering further prospective studies. The result of this study demonstrates the benefits of performing a VBG for treating scaphoid nonunion with no recognized avascular necrosis. The procedure yields similar functional results to NVGB. Earlier time to union could be a theoretical benefit in improving function, however, it did not seem to affect function in this study. Therefore, one should balance the real advantage of performing a more technically demanding procedure with the aim of promoting this earlier union.
1. Gray RR, Shin AY. Vascularized bone grafting of scaphoid nonunions. Oper Tech Sports Med. 2010;18:155–162. 2. Dawson JS, Martel AL, Davis TR. Scaphoid blood flow and acute fracture healing. A dynamic MRI study with enhancement with gadolinium. J Bone Joint Surg Br. 2001;83:809–814. 3. Bhat M, McCarthy M, Davis TR, et al. MRI and plain radiography in the assessment of displaced fractures of the waist of the carpal scaphoid. J Bone Joint Surg Br. 2004;86:705–713. 4. Kawamura K, Chung KC. Treatment of scaphoid fractures and nonunions. J Hand Surg Am. 2008;33:988–997. 5. Cooney WP, Dobyns JH, Linscheid RL. Fractures of the scaphoid: a rational approach to management. Clin Orthop Relat Res. 1980;149:90–97. 6. Sauerbier M, Bickert B, Trankle M, et al. Surgical treatment possibilities of advanced carpal collapse (SNAC/SLAC wrist). Unfallchirurg. 2000; 103:564–571. 7. Krimmer H, Krapohl B, Sauerbier M, et al. Post-traumatic carpal collapse (SLAC- and SNAC-wrist)-stage classification and therapeutic possibilities. Handchir Mikrochir Plast Chir. 1997;29:228–233. 8. Pao VS, Chang J. Scaphoid nonunion: diagnosis and treatment. Plast Reconstr Surg. 2003;112:1666–1676; quiz 1677; discussion 1678–1679. 9. Jaroma H, Suomalainen O, Hannikainen E, et al. Surgical treatment for nonunion of the carpal scaphoid. Ann Chir Gynaecol. 1985;74:239–243. 10. Merrell GA, Wolfe SW, Slade JF III. Treatment of scaphoid nonunions: quantitative meta-analysis of the literature. J Hand Surg Am. 2002;27: 685–691. 11. Caporrino FA, Faloppa F, Santos JBG, et al. Tratamento cirúrgico da pseudartrose do escafóide com enxerto ósseo vascularizado da extremidade dorsal e distal do rádio, baseado na artéia supra-retinacular intercompartimental 1,2. Rev Bras Ortop. 2003;38:522–533. 12. Waitayawinyu T, McCallister WV, Katolik LI, et al. Outcome after vascularized bone grafting of scaphoid nonunions with avascular necrosis. J Hand Surg Am. 2009;34:387–394. 13. Kakar S, Bishop AT, Shin AY. Role of vascularized bone grafts in the treatment of scaphoid nonunions associated with proximal pole avascular necrosis and carpal collapse. J Hand Surg Am. 2011;36:722–725; quiz 725. 14. Malizos KN, Zachos V, Dailiana ZH, et al. Scaphoid nonunions: management with vascularized bone grafts from the distal radius: a clinical and functional outcome study. Plast Reconstr Surg. 2007; 119:1513–1525. 15. Moraes VY, Belloti JC, Moraes FY, et al. Hierarchy of evidence relating to hand surgery in Brazilian orthopedic journals. Sao Paulo Med J. 2011; 129:94–98.
www.jorthotrauma.com |
e47
Caporrino et al
16. Alva D, Vijayaraghavan J, Kotecha A, et al. Treatment of scaphoid nonunion with combined vascularized distal radius graft and iliac crest bone graft. Tech Hand Up Extrem Surg. 2008;12:132–135. 17. Beck E. Russe surgical procedures in treatment of scaphoid pseudarthrosis (Russe I, Russe II) [in German]. Handchir Mikrochir Plast Chir. 1992;24:59–61. 18. Mathoulin C, Gras M, Roukos S. Vascularized bone grafting from the volar distal radius for carpal bones reconstruction [in French]. Chir Main. 2010;29(suppl 1):S65–S76. 19. Rizzo M, Moran SL. Vascularized bone grafts and their applications in the treatment of carpal pathology. Semin Plast Surg. 2008;22: 213–227. 20. Zaidemberg C, Siebert JW, Angrigiani C. A new vascularized bone graft for scaphoid nonunion. J Hand Surg Am. 1991;16:474–478. 21. Adams BD, Blair WF, Reagan DS, et al. Technical factors related to Herbert screw fixation. J Hand Surg Am. 1988;13:893–899. 22. Altman D, Schulz K, Moher D, et al. The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med. 2001;134:663–694.
e48
| www.jorthotrauma.com
J Orthop Trauma Volume 28, Number 3, March 2014
23. Russe O. Fracture of the carpal navicular. Diagnosis, non-operative treatment, and operative treatment. J Bone Joint Surg Am. 1960;42-A: 759–768. 24. Matti H. Technik und resultate meiner pseudarthrosenoperation. Zbl Fur Chir. 1936;63:1442–1453. 25. Cooney WP, Bussey R, Dobyns JH, et al. Difficult wrist fractures. Perilunate fracture-dislocations of the wrist. Clin Orthop Relat Res. 1987; 214:136–147. 26. Matsuki H, Ishikawa J, Iwasaki N, et al. Non-vascularized bone graft with Herbert-type screw fixation for proximal pole scaphoid nonunion. J Orthop Sci. 2011;16:749–755. 27. Steinmann SP, Adams JE. Scaphoid fractures and nonunions: diagnosis and treatment. J Orthop Sci. 2006;11:424–431. 28. Straw RG, Davis TR, Dias JJ. Scaphoid nonunion: treatment with a pedicled vascularized bone graft based on the 1,2 intercompartmental supraretinacular branch of the radial artery. J Hand Surg Br. 2002;27:413. 29. de Moraes VY, Godin K, Tamaoki MJS, et al. Workers’ compensation status: does it affect orthopaedic surgery outcomes? A meta analysis. PLoS One. 2012;7:e50251.
Ó 2013 Lippincott Williams & Wilkins
