Session IX - Tibia
The Use of a Locking Custom-Contoured Blade Plate for Peri-Articular Nonunions and Fractures
Edward J. Harvey, MD, FRCS (C); M. Bradford Henley, MD; Marc F. Swiontkowski, MD; Stephen K. Benirschke, MD, Harborview Medical Center, Seattle, WA
Summary: The treatment of peri-articular nonunions or fractures of the axial skeleton is complicated by proximity of the joint, soft tissue compromise, infection and previous surgery. Peri-articular nonunions are difficult to treat with internal fixation because of the small osteoporotic remnant on the articular side. One method of addressing these problems is with a locking custom-contoured blade plate. Periarticular nonunions and fractures of the tibia and humerus were treated with blade plates fashioned from standard compression plates. Forty procedures in 35 consecutive patients over a 7-year period were retrospectively reviewed.
There were 35 patients of average age 45.6 years (range 25-80). Average follow-up was 3.9 years (range 0.25 to 7). The average number of operations before blade plate implantation was 3.7 (range 0 to 16). Seventeen blade plates were performed in patients with clinical infection and all progressed to union. Nine tissue transfers were used as adjunctive therapy in 8 patients. Five (12.5%) of 40 plates had plate failure before union, of which 4 united after a second custom-contoured blade plate procedure. One of the failures united after late removal, bone graft and lag screws. Two fractures of the proximal humerus had late movement of fracture fragments before healing. Union was achieved for 97.2% of the operative sites with the use of the blade plate.
Surgeon-contoured blade plates are an option for peri-articular nonunions and fractures even in the presence of infection. They result in a high union rate and a low complication rate.
Introduction: Peri-articular fractures and nonunions pose an interesting challenge to the orthopaedic surgeon. These are relatively uncommon problems1,2,7. A small peri-articular osseous fragment is difficult to stabilize. The presence of osteoporosis compounds this challenge beyond the scope of normal osteosynthesis regimens. Standard modes of fixation have included plate fixation, through single or combined procedures, classical external fixation, often spanning the affected joint, or, more recently, hybrid or small wire fixators. Adjacent joint stiffness will increase the stress on the implant. These treatment modes have been associated with complications which include but are not limited to infection, wound dehiscence, failure of fixation, joint stiffness and delayed union or nonunion.
Small pin fixator systems have become a standard of care for periarticular pathology, particularly for very short segment periarticular fragments4,6. The Ilizarov method or hybrid constructs have reported union rates as high as 96% in small studies. Even in these successful studies, complications include nerve palsy, patient non-compliance with fixator care, leg-length inequality and infection or inflammatory problems at the pin sites. The present series of metaphyseal nonunions illustrates a technique using standard compression plates custom-contoured by the surgeon into angled blade plates.
Materials and Methods: This is a consecutive series of 40 custom blade plates placed at 36 operative sites in 35 patients treated at Harborview Medical Center between 1990 and 1997. There were 38 nonunions and 2 acute fractures treated with the customized blade plate technique. All nonunions were the sequellae of trauma. Four failures were treated with a second blade plate and are included as 4 nonunions in the group of 38 total nonunions. There were 13 females and 22 males. One female had blade plates for nonunions of both the distal and proximal tibia. Custom blade plates were implanted in 17 distal tibiae, 19 proximal tibiae, and 4 proximal humeri. The average age of the patients was 45.6 years (range 25-80). Average follow-up was 3.9 years (range 0.3 to 7.0). The average number of operations before blade plate implantation was 3.7 (range 0 to 16). High- energy injury was the most common etiology, with 35 motor vehicle or motorcycle accidents. Four patients sustained their fracture in a fall from a height and one person fractured his tibia while skiing. Seventeen patients had a deep infection during the course of their treatment. Eight patients received a total of nine tissue transfers as an adjunctive soft tissue treatment in the perioperative period of blade plate stabilization.
One patient who received blade plates during the period of this study was excluded. This patient had an amputation at day 10 after a latissimus dorsi free-tissue transfer necrosed. There was insufficient time for healing evaluation to include this patient.
Operative Procedure: Patients were evaluated with biplanar radiographs. Computed tomography (CT) was used if the plane of nonunion could not be determined on the orthogonal biplanar films. The contour of the blade plate was determined from radiographs of the nonunion and the contralateral normal extremity. Correction of deformity was planned from these studies. Although the plate can be shaped in the operating room, most surgeons chose to bend the plate in the machine shop the day before the surgery. The plates recommended for this construct are the 3.5 mm DCP or LCDCP and the 4.5 mm narrow DCP or LCDCP. One patient was treated with a tubular plate which resulted in one of the five failures. It was determined from the radiographs and/or CT scan where the bend in the plate should occur. The bend was usually through the third hole in the plate or between the third and fourth hole. The angle of this bend depended on the anatomical site but approximated 90 degrees (+/-20 degrees). One of the surgeons (MBH) used a hydraulic press to form this bend between the holes to avoid creating a stress raiser. The plate was then shaped to the contour of the bone's normal anatomy. A bone model taken to the machine shop was helpful with this step. The terminal portion of the blade plate was then beveled to a chisel point. This was usually done with a table top wheel grinder. Fully threaded cortical screws were passed from the side plate to the blade portion to assure the surgeon that the screws would pass easily. Two screws, if possible, were threaded in this triangulated manner. The use of an oversized screw (4.5 screw in the 3.5 plates; 6.5 in the 4.5 plates), will ensure that there is an interference fit between the screw and the plate. Oversized screws were used unless the plate was in a subcutaneous position. The plate in this study was judged to be locked if at least one of these screws engaged a hole in the blade portion of the bent plate. This "locking" of the plate effectively tensions the plate and anchors it to the periarticular bone fragment. This prevents the blade plate from changing position in relatively osteoporotic bone when the side plate is fastened to the bone in a compressive mode.
In the operating room, the exposure of the operative site was carried out using conventional surgical exposure technique. Hypovascular tissue was removed at the discretion of the surgeon and autologous bone grafting was carried out for any large volumetric defects. A distractor was used when needed to correct deformity or gain length. There was an attempt to insert the blade plate portion of the plate parallel to the subchondral bone of the neighboring joint. A seating chisel was used infrequently, as the bevel of the blade facilitated the insertion into osteoporotic bone. The side plate portion was placed flush with the diaphyseal bone and was often sunk below the cortex of the metaphyseal portion at the plate's axilla. The locking of the blade with the most distal screw(s) was done initially to anchor the periarticular fragment and tension the blade portion of the plate. This was done using fluoroscopic guidance to ensure that the screw(s) did engage holes in the terminal blade portion of the plate. When the screws engage the distal portion, the blade tip is tensioned towards the nonunion, and this can be observed fluoroscopically. The side plate portion was then used in standard compression mode to compress the nonunion site. Lag screws across the fracture or nonunion but outside the plate were added at the discretion of the surgeon based on the nonunion plane.
Results: Forty customized blade plates were inserted in 35 patients. There were 36 operative sites with four of the sites being treated with a second blade plate after failure of the first plate. The patients had an average age of 45.6 years (range 25-80). Average follow-up was 3.9 years (range 0.3 to 7.0). The average number of operations before blade plate implantation was 3.7 (range 0 to 16).
Thirty-five of the 36 operative sites (97.2%) achieved clinical union with use of the blade plate. Five blade plate constructs failed clinically. One plate failed at one year, two others failed at six months and one failed at 10 weeks. These four cases were treated with a second blade plate without bone graft. Three of these united within 3 months of the second plating. The fifth failure was treated with removal of hardware, iliac crest bone grafting and lag screw insertion at three years after initial blade plate insertion. This case had the only 1/3 tubular plate used initially as a blade plate in this study. This hypertrophic nonunion united after three months.
One patient had a fatigue fracture of the blade plate noted on follow-up radiographs but was healed clinically and radiographically, and, therefore the hardware was not revised. Three patients have had the plate removed because of its subcutaneous prominence. Two patients had implant loosening associated with 10 degrees and 25 degrees of varus in the proximal humerus before healing. Both of these patients have no clinical limitations due to their healed position.
Eight patients were treated with free-tissue transfers for infection at our institution. One patient received two flaps. Patients with a history of osteomyelitis were treated in a staged fashion. These patients received excisional debridement and antibiotic beads followed at six weeks with bead removal, blade plate insertion and/or tissue transfer and/or bone grafting. Only autologous bone graft was used. Many patients had received previous bone grafts prior to their customized blade plating.
Of the 36 operative sites treated initially with custom blade plates in this study, union occurred at all but one site (35/36) for a union rate of 97%. The one nonunited site healed with insertion of lag screws.
Discussion: Metaphyseal nonunions are rare 2,5,7. The metaphyseal blood supply and large area of bone apposition will usually ensure fracture healing. The occurrence of a nonunion in these regions presents a challenge to the treating physician. Often recognition of the problem is delayed. Average time from initial treatment to diagnosis of nonunion is reported as 36 weeks by Ebraheim et al 5. This delay in diagnosis and the resultant disuse osteoporosis of the periarticular fragment compound the challenge for the surgeon. The goal of early motion at the adjacent joint to improve functional outcome dictates adequate fixation in the fracture fragment while still allowing joint motion 2,7. Hybrid or Ilizarov technique small-wire fixators have been advocated by many as the best treatment option 4,6. However, DiPasquale et al. 4 report complications of nerve palsy (6%), leg- length inequality (24%), pin tract inflammations or infections (82%) (21% of these deep infections), and nonunion (6%). Barbieri et al. 1 had complications of infection, skin slough, nonunion and loss of reduction with hybrid fixation for acute fractures. Patients may not accept small wire fixators for protracted periods of time. Treatment of metaphyseal nonunions with locked nails has been reported by McLaren at al 8. These nonunions were mainly at the diaphyseal-metaphyseal junction. The use of a nail device for fractures and nonunions in the periarticular region is difficult unless the fracture is sufficiently distant from the joint to permit interlocking of the nail. Due to the danger of pan-medullary osteomyelitis, the use of a nail is discouraged for infected nonunions.
The use of a blade plate for these periarticular challenges avoids many of the problems with other fixation devices. Patient tolerance and compliance with this fixation device is not an issue as it is with small tensioned wire external fixators. The expertise needed for application of small pin fixators is not possessed by all orthopaedic surgeons. Carpenter and Jupiter 2 have reported on a group of 16 patients with nonunions for which they used a standard 95-degree blade plate with good success. They achieved a union rate of 87% (14/16). They thought that the blade plate could not be used with infection or segmental bone loss. They converted their two nonunions to fixators for salvage rather than using a second blade plate. They did need to custom-contour an unreported number of blade plates because of the inability to fit a standard blade plate to the periarticular fragment.
The custom-contoured blade plate construct in this study allows stabilization of the fracture site due to its ability to maintain solid fixation on the small osteoporotic fragment. Desai et al. 3 have shown in mechanical testing that this blade plate is significantly stiffer than both a cloverleaf plate or a custom-contoured plate without the triangulated locking screw. This internal fixation device allows correction of deformity. It also allows early physiotherapy of the adjacent joint, thereby promoting retention of motion. The presence of active infection was not a contraindication to our use of this technique. Union was obtained in three of the five failures with a second blade plate, and one case healed with removal of hardware, bone graft and addition of lag screws. This series of 36 operative sites with nonunions and fractures resulted in a 97.2% union rate with the blade plate technique. The technique is a reliable method of stabilization for periarticular fractures and nonunions.
Figures Legend:
Figure 1: 48 yo male in MVA sustained open pilon and talus fractures. Treated initially with ORIF fibula, limited internal fixation of pilon, ORIF talus. Placed in external fixation for maintenance of alignment of tibia.
Figure 2: At 3 months post injury, failure to maintain alignment, delayed union of pilon fracture. External fixation pin sites inflamed. Placed in cast for healing.
Figure 3: At 4 months post injury, patient had ORIF of pilon nonunion with DC plate. Failure to maintain alignment has resulted in increasing valgus and dorsiflexion at the nonunion site. Patient was referred for persistent nonunion.
Figure 4: At 7 months, patient received a custom contoured blade plate without bone graft. Radiographs at 9 months post injury show medial bridging of nonunion line. Maintenance of alignment without recurrence of angulation is accomplished.
References
1. Barbieri R, Schenk R, Koval K, Aurori K, Aurori B: Hybrid external fixation in the treatment of tibial plafond fractures. Clin Orthop 332: 16-22, 1996.
2. Carpenter CA, Jupiter JB: Blade plate reconstruction of metaphyseal nonunion of the tibia. Clin Orthop 332: 23-28, 1996.
3. Desai BM, Finkelstein JA, Harrington R, Benirschke SK, Chapman JR. The truss construct for fracture fixation. In press: J Orthop Trauma, 1997.
4. DiPasquale D, Ochsner MG, Kelly AM, Maloney DM: The Ilizarov method for complex fracture nonunions. J Trauma 37(4): 629-633, 1994.
5. Ebraheim NA, Skie MC, Heck BE, Jackson WT: Metaphyseal nonunion: A diagnostic dilemma. J Trauma 38(2): 261-268, 1995.
6. Ebraheim NA, Skie MC, Jackson WT: The treatment of tibial nonunion with angular deformity using an Ilizarov device. J Trauma 38(1): 111-117, 1995.
7. Mandt PR, Gershuni DH: Treatment of nonunion of fractures in the epiphyseal-metaphyseal region of long bones. J Orthop Trauma
8. McLaren AC, Blokker CP: Locked intramedullary fixation for metaphyseal malunion and nonunion. Clin Orthop 265: 253-260, 1991.