The basic goals of treating nonunions of diaphyseal forearm fractures are anatomic alignment of the bone, rigid internal fixation, osteogenic stimulation by application of autogenous bone graft at the nonunion site, and restoration of full limb function.28,32,37 Numerous surgical procedures have been described for treatment of nonunions of diaphyseal forearm fractures including intramedullary (IM) nailing, plate osteosynthesis, bone grafting, and Ilizarov bone transport.4,6,7,10,11,23,33
Plate and screw fixation with or without bone grafting usually is adequate for hypertrophic nonunions if the bone is not osteoporotic and if the fragments are large enough for firm screw fixation.29 Good results have been obtained with plating of diaphyseal forearm nonunions.4,10,23 However, plate and screw fixation requires periosteal stripping for plate application, increasing the chance of delaying union.21,29 Futhermore, bone quality may prove insufficient to achieve good screw fixation in older patients or patients with disuse osteoporosis.15,23,29,38 Iinterlocking IM nailing has been used successfully and extensively to treat femoral, tibial, and humeral shaft nonunions.16,17,20,42,43 It offers advantages such as minimal damage to the soft tissues and risk of infection, more stable fixation by transfixing screws, and an increase of osteogenic activity.16-18,20,36,42,43 Like other long bones, the ulna has a shape that is particularly favorable to nailing. Prebending a straight nail before insertion and using the static inter- locking nailing technique can help restore and maintain radial bow.9 However, little has been published about treating nonunions of a diaphyseal fracture of the radius and/or ulna with an interlocking IM nail.11,19
The purpose of this review of a series patients with nonunions of diaphyseal forearm fractures treated with interlocking IM nails was to document the success of healing with this technique. Our secondary purposes were to document the complications that occur with this technique and to evaluate whether interlocking IM nailing can be recommended as an alternative to plate fixation for treatment of diaphyseal forearm nonunions.
MATERIALS AND METHODS
From April 2002 to May 2003, 19 patients with 38 nonunions of diaphyseal forearm fractures with failure of a previous internal or external fixation procedure were treated with interlocking IM nails and bone grafting at our institution. Four patients were not included in this study: two had bilateral forearm nonunions, and two had repeat surgery and massive segmental bone loss treated with a free vascularized segmental fibular autogenous graft. Fifteen patients with 26 diaphyseal forearm nonunions were included in this retrospective study. Patients were eligible for inclusion in this study if they had an established nonunion of a diaphyseal forearm fracture, a normal contralateral forearm, and no additional surgery to gain bony union after the original fixation. Patients were excluded if there was a previous history of infection during treatment or if there was more than 50 mm bone loss after freshening of the bone ends. Nonunion was defined as either failure of fracture fixation with loss of bony fixation, implant breakage, or lack of progression of healing 6 months after the original surgery. There were 11 men and four women with an average age of 42.3 years (range, 26-58 years). The right forearm was involved in nine patients and the left in six patients. The mechanisms of injury were industrial accidents for nine patients and vehicle accidents for six. The patterns of injury were diaphyseal fractures of both forearm bones in 14 patients and an isolated fracture of the ulnar diaphysis in one patient. There was a dislocation of the distal radioulnar joint in one patient. One patient sustained an injury of the ipsilateral upper extremity with a perilunate dislocation of the wrist. Two patients had associated nerve injuries, including an injury of the ulnar nerve in one and injury of the radial and ulnar nerve in another. Ten fractures were associated with an open wound, and according to the criteria of Gustilo and Anderson, there were one Grade I and nine Grade II fractures (Table 1).13
Four patients had isolated ulna nonunions, and 11 patients had radius and ulna nonunions. The nonunions were classified radiographically as hypertrophic in one bone, oligotrophic in three bones, and atrophic nonunion in 22 bones.40 All nonunions of forearm shaft fractures included in our review were the consequence of previously inadequate surgical treatment of the fractures. The initial operation comprised plating in 19 bones, IM pin fixation in six bones, and external fixation in four bones. No patient had a deep infection develop.
The interval between the injury and the index procedure for treatment of the nonunions averaged 15 months (range, 6-42 months).
The surgical technique consisted of removing all necrotic bone, inserting an autogenous bone graft, and applying the ForeSight® nail system (Smith and Nephew, Memphis, TN) to stabilize the nonunion.
For patient with IM pin and/or external fixation, the definitive operation was done 2 weeks after removal of the previously placed hardware to avoid infection. For patients with plates, the old hardware was removed during the definitive procedure. Radiographs of the contralateral forearm were taken preoperatively to act as a template for the injured extremity.
Under regional anesthesia with tourniquet control, the patient was placed in the supine position on a radiolucent operating table. After the old hardware was removed, the bone ends were exposed and freshened with necrotic bone being removed until bleeding bone was present. A radiograph was taken to measure the final bone gap accurately. The average bone loss in all patients was 20 mm (range, 10-30 mm). There was an average loss of 20 mm (range, 10-25 mm) in the radius and an average loss of 22 mm (range, 10-30 mm) in the ulna. The length of the iliac crest graft required to restore the bone to its original length and shape was calculated by comparing the radiograph of the contralateral bone. The iliac crest graft was harvested and entirely filled the bone defect. Autogenous corticocancellous bone graft from the iliac crest was applied to the nonunion site in five radii and 10 ulnae, and autogenous cancellous bone graft was used in six radii and five ulnae. Also, bone graft saved from entry portal reaming and manual reaming was applied about the nonunion site.
The operative technique for the ForeSight® nail was that described by Crenshaw.8 All nonunions were treated with open reaming technique. For the ulna, a 1-cm incision was used at the olecranon with the elbow flexed at 90°. A 1.9-mm Kirschner wire (K wire) was advanced into the medullary canal, which then was over-reamed with a 6-mm reamer for a distance of 2.5 cm. Manual reamers (0.5-mm increments) were used to enlarge the canal to 0.5-1 mm over the selected nail diameter to prevent nail incarceration or bone end distraction. For the radius, with the wrist and forearm pronated, a 2-cm incision was made on the radial side of Lister's tubercle. Entry into the radius was at the distal radius beneath the extensor carpi radialis brevis tendon, 5 mm from articular surface. A 1.9-mm K wire was used and the canal then was over-reamed with a 6-mm cannulated reamer. The canal was enlarged and the last reamer was left in place to maintain reduction of the radius. The radiograph of the contra- lateral forearm was used for measurement of ulnar nail length (subtracting 2 cm for radial nail length). Nails were prebent to conform to the radial bow and the gentle S shape of the ulna before insertion. For isolated nonunions of the ulna, the ulna nail was inserted and interlocked with a fully threaded 2.7-mm self- tapping screw for the driving end and a 2.7-mm unicortical screw for the nondriving end, respectively. For nonunions of the radius and ulna, the ulnar nail was inserted first, followed by the radial nail.
A well-molded long arm cast was placed on all patients with the forearm in the supine or neutral position and the elbow at 90° flexion for 4-6 weeks. Thereafter, a removable sugar-tong orthosis was worn, and active range of motion (ROM) exercises were allowed until bridging callus was present (Fig. 1).
One independent observer who was not involved in the clinical care of the patients performed a complete followup assessment. The mean followup was 31 months (range, 25-38 months). Union was assessed radiographically. Radiographic union was defined as restoration of the continuity of the involved forearm bone with bridging callus across three of four visible cortices on anteroposterior (AP) and lateral views. A forearm goniometer was used to measure rotation of the forearm with the elbow flexed at 90°. Range of motion of the elbows and wrists also was measured. The unaffected forearm was used as a control for each patient. The location and amount of the maximum radial bow were measured according to the method of Schemitsch and Reichards.32
The functional outcome was assessed with the rating system of Anderson et al,1 which is based on union of the fracture and motion of the forearm, elbow, and wrist. According to this system, an excellent result is defined as a united fracture with less than 10 loss of elbow or wrist motion and less than 25% loss of forearm rotation; a satisfactory result is defined as healed fracture with less than 20°loss of elbow or wrist motion and less than 50% loss of forearm rotation; an unsatisfactory result is defined as a healed fracture with greater than 30 loss of elbow or wrist motion and greater than 50% loss of forearm rotation; and failure is defined as nonunion or unresolved chronic osteomyelitis.
Patient-rated outcome was assessed with the limb-specific questionnaire (Disability of the Arm Shoulder Hand questionnaire [DASH]).14,39 The DASH ascertains overall function of the upper extremity. It is a 30-item upper extremity scale that measures function, symptoms, and quality of life issues related to disorders of the upper extremity. It is scored as a percentage of 100 as recommended by the Upper Extremity Collaborative Group. A score of zero points indicates a perfectly functioning arm, whereas a score of 100 points indicates complete impairment of the upper extremity.
Statistical analysis was performed using the SPSS® system software package (SPSS Inc, Chicago, IL). An independent sample t test was used to determine differences in union time between radial and ulnar nonunions and the open and the closed fractures. Analysis of variance was used to study the difference of motion of the forearm among the normal side, and preoperatively and postoperatively injured side. A p value of 0.05 or less was considered significant.
Radiographic union was achieved in 25 diaphyseal forearm nonunions (96%). The average healing time for the radius and ulna were equal, with a mean time to union for the radius of 14 weeks (range, 12-16 weeks) and a mean time to healing of 15 weeks (range, 12-20 weeks) for the ulna. Nonunions that occurred in closed fractures healed in an average of 14 weeks (range, 12-16 weeks), and non- unions in open fractures healed in an average of 15 weeks (range, 12-20 weeks).
The final ROMs averaged 66° (range, 34°-88°) pronation, 61°(range, 20°-88°) supination, 48°(range, 10°-80°) wrist extension, 66°(range, 30°-90°) wrist flexion, 9°(range, 0°-20°) elbow flexion contracture, and 131°(range, 100°-150°) elbow flexion. Compared with the normal arm, the mean loss of flexion and extension motion at the wrist was 27°. The mean loss of the arc of motion at the elbow was 18°, and the mean loss of rotation of the forearm was 39°. The ROM of the uninjured contralateral arm was greater in all planes (p = 0.012; 0.001; 0.001).
The mean amount and location of the maximum radial bow in the uninjured contralateral arm were 16 mm and 60%, respectively. The mean amount and location of the difference of the maximum radial bow between the injured side and uninjured side were 2.2 mm and 7%, respectively. Seven patients had an amount or location of the maximum radial bow of 2.8 mm or 8.9% more or less than that of the uninjured contralateral arm.
Using the rating system of Anderson et al,1 two patients (13%) had excellent results, six patients (40%) had satisfactory results, six patients (40%) had unsatisfactory results, and one patient (7%) had a failed result. The failed result might be attributable to initiating ROM too aggressively before adequate healing had occurred. The mean DASH score was 35 points (range, 16-56 points), indicating moderate residual impairment.
There were three postoperative complications. The incidence of complications was 12% (three of 26 fractures). A hypertrophic nonunion that did not heal occurred in a 54-year-old woman who smoked tobacco. Her ulna non- union still was not united radiographically 30 months after the index operation, but she did not wish additional surgery. She had mild to moderate pain during everyday activities (Fig. 2).
The other two complications were the loosening of two screws at the nondriving end of the ulna nail. One of the screws was removed because of wrist pain. There were no mechanical irritations at the distal radius or olecranon. No injury of the posterior interosseous nerve was present with proximal interlocking of the radial nail. No deep infection occurred in any of the 15 patients, and no ectopic bone formation occurred.
Different operative methods, including plating and medullary nailing combined with bone grafting, have been used to treat nonunions of diaphyseal forearm fractures.2,10,23,25 Plate and screw fixation is used most commonly for diaphyseal nonunions.29 However, plate osteosynthesis requires extensive soft tissue dissection, which may compromise the blood supply of the healing fracture.3,27 Osteoporosis at the nonunion site may decrease the strength of purchase of the screws holding the plate, resulting in inadequate fixation.23 Interlocking IM nailing has been successful, with extensive use in treating non- unions of long bones such as the humerus, femur, and tibia.16,17,22,41,42 Compared with an interference-fit forearm nail, an interlocking IM nail can provide additional antirotational control and prevent the nail from backing out.24,30 Also, unlike a compression plate, an IM nail is stress sharing rather than a stress shielding, which leads to peripheral periosteal callus that may produce a stronger fracture union.24 Although reaming and nailing temporarily disturb blood supply to the endosteum, it has been shown to elicit a periosteal vascular reaction that stimulates bone formation.18,36 Moreover, the ulna has a shape that is particularly favorable to nailing. Prebending a straight nail before insertion and using the static interlocking technique can help restore and maintain radial bow.9
The cohort of patients chosen in the current series had less-complicated diaphyseal forearm nonunions with less than 30-mm bone loss after freshening of the bone ends, no previous history of infection during treatment, and no additional intervention after the nonunions. However, our results showed a disadvantage in terms of functional recovery of interlocking IM nailing compared with plating for forearm shaft nonunions. With average bone loss of 22 mm (10-30 mm), we had a union rate of 96% and a 12% incidence of complications. With similar bone defects, Ring et al25 achieved a union rate of 100% and a complication rate of 11% using plate and screw fixation and autogenous cancellous bone grafting Using stable internal plate fixation and autogenous intercalary bone grafts, Moroni et al23 reported good results in 24 patients with isolated radius and ulna nonunions with segmental bone loss of 36 mm (range, 20-105 mm). Their patients had a union rate of 96%, a complication rate of 33%, and an infection rate of 13%. Davey and Simonis10 treated 19 patients with established nonunions of the radius and/or ulna with average bone loss of 40 mm (range, 20-100mm). They used a slightly modified Nicoll's technique by inserting a block of corticocancellous graft into the defect and securing it by a tibial plate. This yielded a 95% union rate with a 16% complication rate. The single failure was attributed to the excessive length of the bone defect (100 mm). Despite the severity of the cases, the final functional outcomes reported for patients treated with plating techniques using the rating system of Anderson et al1 are still superior to our results.23,25 Moreover, the mean DASH score of 35 points in the current series indicated moderate disability.
The reason that the functional results in the current series were not superior to those obtained with a plating technique might be related to prolonged immobilization, inadequate restoration of the radial bow, and use of an open IM nail technique. First, nonunions of diaphyseal forearm fractures treated with an interlocking IM nail, compared with plating, needed longer times to be stabilized.10,23 With compression plate fixation, early active motion is possible. This helps prevent muscle atrophy and joint stiffness, which often are responsible for unsatisfactory results.2,12 Some patients in our series had loss of joint motion and/or forearm rotation after the index procedure. Second, although bone length could be restored in these nonunions using bone grafts and interlocking screws, prebending a straight nail before inserting it might not fully restore the normal radial bow of the nonunited bone because of the small diameter of the IM nail (4 or 5 mm), the change of nail contour during insertion, and the wide medullary canal of the radius, especially in the distal third of the radius. A disproportion between the size of the nail and the medullary canal can allow side to side and rotary movements if the nail is too small.9 Schemitsch and Reichards32 showed that alteration of the maximum radial bow of 2.8 mm more or less than that of the uninjured contralateral arm or the location of the maximum radial bow of 8.9% more or less than that of the uninjured contralateral arm could significantly influence forearm function. Finally, although percutaneous IM nailing can decrease damage to the soft tissue thereby reducing the risk of infection,30 nailing with an open technique may delay healing of nonunions as a result of additional damage to the periosteal blood supply. Therefore, using open technique might yield unfavorable final results.5
Precautions should be taken to prevent or minimize complications. When using an interlocking IM nail for treatment of forearm nonunions, one must avoid a disproportion between the size of the nail and the medullary canal to prevent side to side and rotary movements if the nail is too small and additional comminution or fracture if it is too large.9 Although iatrogenic posterior interosseous nerve injury seldom occurs, we think that the proximal locking screw of the radial nail is best inserted within 30 mm from the radial head with the forearm in neutral rotation.35 We found a higher frequency of nonunion in the ulna compared with the radius, as has been reported.23,34 This might be because of torsional stresses applied to the relatively stationary ulna during pronation and supination of the forearm.34 In this series the single case of an ulnar nonunion was associated with inadequate fixation from this torsional motion. The low infection rate might be related to case selection and the protocol of an external fixator and IM pin removal 2 weeks before the definitive procedure.
Ring et al25,26 concluded that insertion of an autograft is necessary in patients with failed internal fixation. An advantage of an iliac crest autograft is the transplantation of bone graft without the risk of an immune reaction or of transferring disease with osteoconductive and osteoinductive capacities.31 In the current series, the iliac graft was used in all cases.
There are several limitations of this study including the small cohort of patients, the retrospective methodology, the selection bias of the patients, and the different implants used for primary fracture fixation. Additionally, the amount of graft material used was not standardized. Intramedullary nailing of diaphyseal forearm nonunions with small bone defects can achieve predictable bone unionin as much as 96% of cases. However, in terms of functional recovery, interlocking IM nailing has disadvantages compared with plating. Interlocking IM nailing of nonunions of the diaphysis of the radius or ulna with an open reaming technique should not be considered an adequate alternative to plate fixation for these injuries.
We thank Dr. Jiang Jia for assistance with data collection.
1. Anderson LD, Sisk D, Tooms RE, Park WI III. Compression-plate fixation in acute diaphyseal fractures of the radius and ulna. J Bone Joint Surg Am
2. Barbieri CH, Mazzer N, Aranda CA, Pinto MM. Use of a bone block graft from the iliac crest with rigid fixation to correct diaphyseal defects of the radius and ulna. J Hand Surg Br
3. Bostman O, Varjonen L, Vainionpaa S, Majola A, Rokkanen P. Incidence of local complications after intramedullary nailing and after plate fixation of femoral shaft fractures. J Trauma
. 1989;29: 639-645.
4. Brakenbury PH, Corea JR, Blakemore ME. Non-union of the isolated fracture of the ulnar shaft in adults. Injury
5. Browner BD. Pitfalls, errors, and complications in the use of locking Kuntscher nails. Clin Orthop Relat Res
6. Chirstos CG. Forearm bone non-union and its management. Ethiop Med J
7. Christensen NO. Kuntscher intramedullary reaming and nail fixation for nonunion of the forearm. Clin Orthop Relat Res
. 1976;116: 215-221.
8. Crenshaw AH Jr. Surgical technique: ForeSight Nail System
. Memphis, TN: Smith & Nephew; 1997
9. Crenshaw AH, Zinar DM, Pickering RM. Intramedullary nailing of forearm fractures. Instr Course Lect
10. Davey PA, Simonis RB. Modification of the Nicoll bone-grafting technique for nonunion of the radius and/or ulna. J Bone Joint Surg Br
11. Esser RD. Treatment of a bone defect of the forearm by bone transport: a case report. Clin Orthop Relat Res
12. Grace TG, Eversmann WW Jr. Forearm fractures: treatment by rigid fixation with early motion. J Bone Joint Surg Am
13. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am
. 1976;58: 453-458.
14. Hudak PL, Amadio PC, Bombardier C. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand)[correcterd]. Upper Extremity Collaborative Group (UECG). Am J Ind Med
15. Ingle BM, Hay SM, Bottjer HM, Eastell R. Changes in bone mass and bone turnover following distal forearm fracture. Osteoporos Int
16. Johnson EE, Marder RA. Open intramedullary nailing bone-grafting for non-union of tibial diaphyseal fracture. J Bone Joint Surg Am
17. Johnson KD. Management of malunion and nonunion of the tibia. Orthop Clin North Am
18. Klein MP, Rahn BA, Frigg R, Kessler S, Perren SM. Reaming versus non-reaming in medullary nailing: interference with cortical circulation of the canine tibia. Arch Orthop Trauma Surg
19. Lavelle DG. Delayed union and nonunion. In: Canale ST, ed. Camp- bell's Operative Orthopaedics
ed. St Louis, MO: Mosby; 1998:2618-2619.
20. Lin J, Hou SM, Hang YS. Treatment of humeral shaft delayed unions and nonunions with humeral locked nails. J Trauma
. 2000; 48:695-703.
21. Marsh JL, Buckwalter JA, McCollister-Evarts C. Delayed union, non-union, malunion and avascular necrosis. In: Epps CH, ed. Complications in Orthopaedic Surgery
ed. Philadelphia, PA: JB Lippincott; 1994:183-211.
22. Martinez AA, Herrera A, Cuenca J. Good results with unreamed nail and bone grafting for humeral nonunion: a retrospective study of 21 patients. Acta Orthop Scand
23. Moroni A, Rollo G, Guzzardella M, Zinghi G. Surgical treatment of isolated forearm non-union with segmental bone loss. Injury
24. Rand JA, An KN, Chao EY, Kelly PJ. A comparison of the effect of open intramedullary nailing and compression-plate fixation on fracture-site blood flow and fracture union. J Bone Joint Surg Am
25. Ring D, Allende C, Jafarnia K, Allende BT, Jupiter JB. Ununited diaphyseal forearm fractures with segmental defects: plate fixation and autogenous cancellous bone-grafting. J Bone Joint Surg Am
26. Ring D, Barrick WT, Jupiter JB. Recalcitrant nonunion. Clin Orthop Relat Res
27. Ring D, Jupiter JB, Sanders RA, Quintero J, Santoro VM, Ganz R, Marti RK. Complex nonunion of fractures of the femoral shaft treated by wave-plate osteosynthesis. J Bone Joint Surg Br
28. Rodriguez-Merchan EC, Forriol F. Nonunion: general principles and experimental data. Clin Orthop Relat Res
29. Rodriguez-Merchan EC, Gomez-Castresana F. Internal fixation of nonunions. Clin Orthop Relat Res
30. Rosen H. Nonunion and malunion. In: Browner BD, Jupiter JB, Levine AM, Trafton PG, eds. Skeletal Trauma
. 2nd ed. Philadelphia, PA: WB Saunders Company; 1998:619-660.
31. Salkeld SL, Patron LP, Barrack RL, Cook SD. The effect of osteogenic protein-1 on the healing of segmental bone defects treated with autograft or allograft bone. J Bone Joint Surg Am
. 2001;83: 803-816.
32. Schemitsch EH, Richards RR. The effect of malunion on functional outcome after plate fixation of fractures of both bones of the forearm in adults. J Bone Joint Surg Am
33. Shelton WR, Sage FP. Modified Nicoll-graft treatment of gap non- unions in the upper extremity. J Bone Joint Surg Am
34. Stern PJ, Drury WJ. Complications of plate fixation of forearm fractures. Clin Orthop Relat Res
35. Tabor OB Jr, Bosse MJ, Sims SH, Kellam JF. Iatrogenic posterior interosseous nerve injury: is transosseous static locked nailing of the radius feasible? J Orthop Trauma
36. Templeman D, Thomas M, Varecka T, Kyle R. Exchange reamed intramedullary nailing for delayed union and nonunion of tibia. Clin Orthop Relat Res
37. Tooms RE, Calandruccio JH. Complications of treatment of injuries to the forearm. In: Epps CH Jr, ed Complications in Orthopaedic Surgery
ed. Philadelphia, PA: JB Lippincott Company; 1994:307-320.
38. Trotter DH, Dobozi W. Nonunion of the humerus: rigid fixation bone grafting, and adjunctive bone cement. Clin Orthop Relat Res
39. Turchin DC, Beaton DE, Richards RR. Validity of observer-based aggregate scoring systems as descriptors of elbow pain, function, and disability. J Bone Joint Surg Am
40. Weber BG, Brunner C. The treatment of nonunions without electrical stimulation. Clin Orthop Relat Res
41. Wiss DA, Stetson WB. Nonunion of the tibia treated with a reamed intramedullary nail. J Orthop Trauma
42. Wu CC. Humeral shaft nonunion treated by a Seidel interlocking nail with a supplementary staple. Clin Orthop Relat Res
. 1996;326: 203-208.
43. Wu CC, Shih CH, Chen WJ, Tai CL. Effect of reaming bone grafting on treating femoral shaft aseptic nonunion after plating. Arch Orthop Trauma Surg