Orthopedics

Materials and Methods

Between 2003 and 2010, forty-five nonunion or delayed union fractures in 45 patients were treated with this technique and followed for at least 6 months postoperatively. Mean follow-up was 21.3 months (range, 6–69 months). Thirty-three patients were followed for at least 1 year. Thirty-three men and 12 women, with an average age of 42.6 years (range, 17–82 years), were included. Five clavicle, 4 ulna, 2 radius, 17 femur, 15 tibia, and 2 fibula fractures were included. Defects were divided according to Weber’s radiological nonunion classification³: 6 hypertrophic, 7 atrophic, 22 oligotrophic, and 10 defect types. Autologous bone grafting without exchange or the addition of implants was performed in 7 patients. The remaining patients were treated by appropriate fracture fixation, including revision of the implants in combination with autologous bone grafting. Mean duration from trauma to nonunion surgery was 13.9 months (range, 4–76 months). Infection found in 6 patients was eradicated before nonunion surgery.

Surgical Technique

Intraoperatively, the site of delayed union or nonunion is exposed without peeling off the periosteum (Figure 1). The size of the incision and the extent of surgical exposure is minimized as much as possible to preserve blood supply to the nonunion or delayed union site. Rarely, a larger exposure is needed if the previously implanted plates and screws need to be removed.

Figure 1: Exposure of the delayed union or nonunion site. Exposure is done without pealing off the periosteum. The size of incision and extent of surgical exposure is minimized as much as possible to preserve the blood supply.

The procedure is divided into 2 steps (Figure 2). First, both ends of the fracture fragments are chipped into small pieces using an osteotome and a hammer, and pathways into the bone marrow are produced. Second, cancellous bone is harvested from the iliac crest and grafted into the aperture created by the previous bone chipping treatment. For the femur, approximately 2 to 3 cm of both fracture ends are chipped into small pieces (Figure 3). The extent of bone chipping is adjusted according to the size of the fractured bone.

Figure 2: Diagram showing 2 steps of the procedure. Step 1: Bone chipping. Both ends of the fracture fragments are chipped into small pieces and pathways into the bone marrow are produced using an osteotome and hammer without peeling off the periosteum. Step 2: Autologous bone grafting. Cancellous bone harvested from the iliac crest is grafted into the aperture created by the bone chipping.

Figure 3: For the femur, approximately 2 to 3 cm of both fracture ends are chipped into small pieces.

Intraoperative photographs show a tibial nonunion (Figure 4). In this patient, the removal of previously implanted plates was necessary; therefore, a large surgical exposure was created. The implanted plate was removed, and the medial side of the non-union site, which would be covered by a revised plate, was chipped into small pieces (Figure 4A). Cancellous bone grafting to this site was performed, followed by revision plate fixation. The remainder of the non-union site was then chipped into small pieces (Figure 4B), and cancellous bone grafting was performed (Figure 4C). Chipping of the nonunion site and cancellous bone grafting were also performed on the posterior side of the plate (Figure 4D).

Figure 4: Intraoperative photographs of a case of tibial nonunion. The yellow arrow indicates the nonunion site. Previously implanted plates and screws were removed. The medial side of the nonunion site was chipped into small pieces before the revision plate fixation (A). After revision plate fixation, the remaining nonunion site was chipped into small pieces (B). Cancellous bone grafting was performed (C). Cancellous bone grafting was also performed on the posterior side of the plate and under the plate (D).

To obtain optimal results using this technique, the osteotome should be inserted into the bone parallel to the bone axis but not parallel to the fracture line. Bleeding from the bone marrow should be identified, and care should be taken not to let the small bone fragments completely detach. The existing fibrous tissue in the fracture gap should not be curetted but should be chipped.

In addition, a nonunion can be implanted with an intramedullary nail. The whole bone area, including the opposite side, can be chipped via a small skin incision (Figure 5A). In contrast, accessing the opposite side of the surgical exposure from the same incision is difficult due to obstruction of the intramedullary nail. However, sufficient access is available to the nonunion site, including the anterior, lateral, and posterior sides, despite the obstruction by the intramedullary nail (Figure 5B).

Figure 5: A case of nonunion without implantation of an intramedullary nail. The whole bone area, including the opposite side, can be chipped via a small skin incision (A). A case of nonunion implanted with an intramedullary nail is shown. Sufficient access to the nonunion site is available, including the anterior, lateral, and posterior sides, despite the obstruction by the intramedullary nail (B).

Results

Patient details are summarized in the Table. Bony union was successfully obtained in all patients. Of the 45 patients with nonunion or delayed union, 42 obtained bony union uneventfully in single operations and the remaining 3 required second operations. The reasons for second operations included: biologically impaired bone activity due to a previous infection in patient 7 (atrophic nonunion of the tibia), insufficient fixation in patient 15 (oligotrophic nonunion of the femur), and insufficient fixation, diabetes mellitus, and a heavy smoking habit in patient 16 (oligotrophic nonunion of the tibia). Mean duration from surgery to bony union was 5.7 months (range, 2–13 months) in the 42 patients undergoing single operations. Complications, such as infection and major donor site morbidity, were not detected in any patient.

Table: Patient Data

The authors used this technique mainly to treat cases of femoral and tibial nonunion. Five cases are presented.

Case Reports

Patient 23

Exchange nailing and autologous bone grafting was performed for femoral nonunion, and bony union was achieved (Figure 6).

Figure 6: Anteroposterior radiographs of patient 23 showing femoral nonunion (A), exchange nailing and autologous bone grafting using the current technique (B), and bony union (C).

Patient 24

In a femoral nonunion, infection was eradicated by removal of the intramedullary nail, antibiotic-loaded cement implantation, and external fixation. Revision of the intramedullary nail and autologous bone grafting was performed, and bony union was achieved (Figure 7).

Figure 7: Anteroposterior radiographs of patient 24 showing femoral nonunion with infection (A), revision of intramedullary nail, and autologous bone grafting using the current technique (B), and bony union (C).

Patient 27

Revision of distal locking screws and autologous bone grafting was performed for femoral nonunion, and bony union was achieved (Figure 8).

Figure 8: Anteroposterior radiographs of patient 27 showing femoral nonunion (A), revision of distal locking screws and autologous bone grafting using the current technique (B), and bony union (C).

Patient 16

Previous exchange nailing was unsuccessful. Conversion to a locking plate and autologous bone grafting was performed for tibial nonunion, and bony union was achieved (Figure 9).

Figure 9: Anteroposterior radiographs of patient 16 showing tibial nonunion in which previous exchange nailing was unsuccessful (A), conversion to a locking plate and autologous bone grafting using the current technique (B), and bony union (C).

Patient 44

Plate fixation was not revised. Autologous bone grafting was performed for tibial nonunion, and bony union was achieved (Figure 10).

Figure 10: Anteroposterior radiographs of patient 44 showing tibial nonunion (A), autologous bone grafting using the current technique without plate revision (B), and bony union (C).

Discussion

Although new technology to enhance bone healing has been applied to some cases of non-union, such as osteogenic protein-1, which is also known as bone morphogenetic protein-7,^4,5 autologous bone grafting is still a useful and frequently used tool.^6–8 The best results may be achieved by adding autologous bone grafting because it is osteogenic, osteoinductive, and osteoconductive.¹

Matsushita and Watanabe⁹ used a similar technique they called a chipping and lengthening technique for treating delayed unions and nonunions with shortening or bone loss in combination with a lengthening technique using external fixators in the absence of autologous bone grafting. The current technique is different in that the authors not only chip the bone, but also add autologous bone grafting to further enhance the bone healing capacity. In addition, the current technique was used in patients originally treated with various intramedullary nails, plates, and external fixators.

The current technique differs from Judet and Patel’s² decortication technique because decortication only chips off the surface of the bone whereas the current technique chips the entire area of bone. Bone healing capacity is inherent in this technique because stem or progenitor cells with osteogenic capacity will be introduced from the bone marrow, and osteoinductive factors such as bone morphogenic proteins will be introduced from the bone matrix into the fracture site. Bone morphogenic proteins exist in the extracellular matrix of the bone and bind to the collagen network,¹⁰ and they are expected to be released from the extracellular matrix and participate in the healing process as a result of chipping the cortical bone.

Moreover, in the current technique the authors did not curette fibrous tissue at the fracture gap due to results from their previous research revealing that human nonunion tissue contains osteogenic or chondrogenic progenitor cells.¹¹ In these patients, the reservation of intercalary fibrous tissue is helpful in reducing the amount of harvested bone to graft.

The best indication for using this technique lies in the successful treatment of nonunion patients, especially those with oligotrophic and atrophic non-union, which are considered biologically less active. From the authors’ clinical experience, the strengths of this technique are the good nonunion healing rate and the absence of complications. The limitation of the study is the lack of a control group with which to compare results.

This technique provides a simple method for preparing the bed for autologous bone grafting and is a promising approach for enhancing bone healing in non-union and delayed union.