I Have a 34-Year-Old Male With a Nonunion of a Plated Midshaft Tibia Fracture. How Do You Approach This Problem?
Stuart M.
Gold,
MD
When I am referred a nonunion for consultation and treatment, I try to determine if I am dealing with a biologic or a mechanical problem. In many cases, this can be determined by the x-ray alone; however, obtaining an accurate history from the onset of the injury always provides critical information.
The significant questions I ask include mechanism of injury (high or low energy); open or closed fracture; and whether significant soft-tissue, vascular, or neurologic injury occurred. Additionally, I inquire about the initial treatment management the patient underwent, ie, splinting versus temporary external fixation, the length of time the temporary fixation was in place, and if there were any pin track issues. I determine if there were any wound issues or wound drainage at any time, including the initial or later phases of treatment, as well as the development of any deformity during the treatment period. Finally, the length of time from the initial injury to presentation and the number and types of surgeries as well as possible bone grafts and/or types of bone grafting material are important queries.
Following all of these critical history questions, I carry out a thorough physical examination to assess deformity, ie, angulation, shortening, malrotation, and most importantly, the mechanical axis of the limb and the stiffness of the nonunion. I then carefully determine the status of the soft-tissue envelope, most notably around the nonunion as well as proximal and distal to the nonunion, to determine if it will tolerate violation with an extensive open surgery. The function of both the knee and the foot also need to be carefully assessed.
Radiographs are then reviewed to help assess the biology for any evidence of callus or any evidence of infection or for bone that clearly does not have healing potential. Comparison films from the contralateral limb are helpful, as are long mechanical axis films from the hip to the ankle. If a deformity is present, the specifics such as bone loss, shortening, or whether a gap is present need to be assessed. Additionally, overall shortening or angulation or rotational deformity needs to be determined to see if they can all be corrected at the time of the definitive treatment of the nonunion.
Other forms of testing can be helpful, but are rarely definitive with respect to determining if an infection is present. These include laboratory tests such as white blood cell counts, ESR, C-reactive protein, as well as more invasive tests such as technetium bone scans, Indium-111 scans, and magnetic resonance imaging (MRI) scans; however, it is my opinion that infection is usually best determined by history, physical examination, and deep cultures; the other methods are only occasionally helpful.
In this case with a 34-year-old male with a nonunion of plate at the midshaft tibia, the key issue is to determine if it is a biologic or mechanical problem. My specific treatment plan is directly related to the issues of infection as well as the status of the soft-tissue envelope surrounding the nonunion. If this is purely a mechanical problem, there is no sign of infection, but there is good healing potential, my choice of refixation will then depend on the soft-tissue envelope. During open plate removal, a gram stain, deep cultures, and a frozen section are taken, which provide me with the definitive answer with respect to infection.
Assuming no infection, my preferred approach is for an intramedullary, reamed nail locked at both ends, correcting any mechanical axis deformities at the time of nailing. Additionally, I place a bone graft either using an autogenous cancellous iliac crest or a combination of a demineralized bone matrix or cancellous chips with marrow aspirate. I believe that opening the nonunion site to remove the plate affects the local biology enough to warrant bone grafting, even in the setting of a hypertrophic nonunion. If a nail, for whatever reason, was not feasible, then replating with a longer plate would be a second option; again, assuring mechanical axis is aligned as well as providing a graft as noted above.
If found to be an infected nonunion, my approach is different in that the hardware needs to be removed, followed by a resection of all the necrotic or infected bone, assuring that healthy bone with healing potential is evident on both sides of the gap. My treatment then depends on the virulence of the organism and the residual bone defect after debridement. If I had a gap of less than 3 cm, I place an antibiotic nail for 6 to 12 weeks, to be followed by an exchange nail and bone grafting using the grafting options as noted above. If the gap was greater than 3 cm and since I am comfortable with bone transports, I proceed with a circular frame and bone transport. The gap would be filled with antibiotic impregnated cement, a combination of vancomycin and tobramycin in the appropriate ratio for 6 to 8 weeks then followed by removal of the antibiotic impregnated beads and then start of the bone transport. At that point, I decide if the docking site requires a bone graft prior to its completion of docking.
In conclusion, the key reasons I have used a circular fixator for nonunions of the tibia include infected gaps greater than 3 cm, nonunions associated with shortening, and significant deformities with poor soft-tissue envelopes. As many circular fixators as I apply, I still nail and plate a significantly greater number of fractures and non/malunions per year.
Suggested Reading
Gold SM, Wasserman R. Preliminary results of tibial bone transports with pulsed low intensity ultrasound (exogen). J Ortho Trauma. 2005;19:10-16.
Mckee MD, Schemitsch EH, Waddell JP. The use of an antibiotic-impregnated osteoconductive, bioabsorbable bone substitute in the treatment of infected long bone defects. Early results of a prospective trial. J Ortho Trauma. 2002;16:622-627.
Stoffel K, Dieter U, Stachowiak G, Gachter A, Kuster MS. Biomechanical testing of the LCP: how can stability in locked internal fixation be controlled? Injury. 2003;34(supp 2):B11-B19.