Proximal Tibia Non-Union: When an Autograft Does Not Work

    A Type IIIC open tibia fracture sustained in a high-speed motor vehicle collision has not progressed toward bony union, despite multiple attempts at treatment. Will a staged approach to management finally lead to fracture union?


    Brandon J. Yuan, MD

    Case Presentation

    A 34-year-old female patient was brought to the emergency department with a Type IIIC open right tibia fracture and volume-expanding pelvic ring injury sustained in a high-speed motor vehicle collision (Figure 1).

    Figure 1. Radiographs demonstrate the presenting injuries: Open type IIIC right tibia fracture and volume expanding pelvic ring injury.

    Management of life- and limb-threatening injuries according to advanced trauma life support (ATLS) protocols and volume resuscitation were prioritized in the emergency department. In this patient’s case, treatment included stabilization of the volume-expanding pelvic ring injury and stabilization and revascularization of the right lower extremity.

    The pelvis was stabilized with a binder and the patient was then taken to the operating room for emergent vascular repair, debridement, external fixation, and fasciotomy of the right lower extremity.

    Initial Management

    During the initial debridement, the surgeon opted for limited internal fixation of the proximal tibia fracture in addition to knee-spanning external fixation. An antibiotic bead pouch was placed to assist in dead space management and to decrease bacterial load (Figure 2). For open injuries, particularly those near a vascular repair, the use of an adjunctive plate is an excellent option to provide additional skeletal stability as a bridge to the eventual definitive fixation.

    Figure 2. Initial treatment included limited internal fixation of the proximal tibia fracture, knee-spanning external fixation, and use of an antibiotic bead pouch.

    The patient’s pelvis was managed with open reduction and internal fixation, but for the purposes of this article, we will focus on the tibia fracture.

    The patient’s wounds were debrided every other day for 6 days until stable. The surgeon had planned for definitive internal fixation after the soft tissues were stable for coverage.

    The external fixator was removed, as was the provisional plate. The non-displaced intra-articular fracture was compressed with 3.5-mm lag screws placed from medial to lateral through the large open wound. The metaphyseal fracture was stabilized with an intramedullary nail. At the conclusion of the procedure, an additional 3.5-mm LC-DC plate was added to the anteromedial surface of the tibia to serve as additional stabilization of this very proximal metaphyseal fracture (Figure 3).

    Figure 3. Imaging shows the use of lag screws, an intramedullary nail, and an LC-DC plate to compress and stabilize the tibial fracture.

    Surgeons typically avoid mixing fixation approaches (open reduction and plating versus indirect reduction and medullary fixation). In this case, however, the soft tissue injury provided wide and direct exposure of the fracture. The addition of direct reduction and plating methods to the medullary nail did not necessitate any additional periosteal stripping or exposure, as that had already been done at the time of the original injury. Thus, the addition of the plate served to provide additional fixation to this very unstable metaphyseal fracture without additional soft tissue insult.

    The comminuted proximal fibula fracture was associated with a partial peroneal nerve palsy, which was managed expectantly with a brace during recovery.

    Soft tissue coverage was completed the following day with a vascularized latissimus transfer performed by a plastic surgeon (Figure 4). The time between definitive fixation and definitive soft tissue coverage should be minimized as much as possible. Serial debridement of the fracture and soft tissue should continue until the patient and the soft tissue bed are amenable to soft tissue coverage. Definitive fixation and soft tissue coverage are typically done within 48 hours.

    Figure 4. Soft tissue coverage was completed with a vascularized latissimus transfer.

    6 Months Later: Delayed Union

    This patient’s free tissue flap and skin graft sites healed without incident. She was initially kept non-weight-bearing for 3 months after surgery and then transitioned to partial weight-bearing. Six months after the index procedure, radiographs (Figure 5) and a CT scan demonstrated lack of any progression toward bony union, suggesting a delayed union.

    Figure 5. As the radiographs show, the fracture had made no progress toward bony union.

    Options at this point include the following:

    Continued observation. Given the degree of soft tissue stripping that occurred at the time of injury, however, it was unlikely that the patient would go on to develop sufficient healing callus to unite this fracture prior to catastrophic implant failure.

    Revision internal fixation with bone graft. This option has the advantage of removing the previous implants and allowing for biologic augmentation while restarting the clock on potential mechanical failure of the implants. However, this option would also be a much larger surgery and the implants showed no signs of significant loosening.

    Bone grafting. Bone graft can be placed through elevation of the muscle flap, and, if necessary, the anteromedial fixation through the small fragment plate could be revised concurrently.

    Prior to any revision procedure, the surgeon must rule out underlying infection or an endocrine abnormality that may be contributing to delayed bone healing. Preoperative labs include:

    • C-reactive protein (CRP) level
    • Erythrocyte sedimentation rate (ESR)
    • White blood cell (WBC) count
    • Endocrine panel: vitamin D, thyroid stimulating hormone, parathyroid hormone, alkaline phosphatase, calcium, and phosphorus

    All were within normal limits in this patient, and 7 months after the index procedure, the patient underwent bone grafting of the non-union.

    The flap was elevated by a plastic surgeon and the anterior medial plate was removed. The non-union was debrided and cultured and filled with autograft bone from the patient’s ipsilateral iliac crest. The plate was replaced with new screws at the conclusion of the procedure (Figure 6).

    Figure 6. Radiographs obtained following removal of the anterior medial plate, bone grafting, and application of new screws.

    S. epidermidis Cultured on POD5

    Unfortunately, on POD5, the patient’s operative cultures grew Staphylococcus epidermidis that was resistant to oxacillin. The patient was started on a course of oral suppressive antibiotics for an infected tibial non-union. It was unfortunate that autograft bone had been placed into a potentially infected tissue bed; however, given the negative preoperative work up, it did not appear that the patient had an infection at the time of surgery.

    Options at this point included:

    • Revision surgery to remove all implants from the leg and treat the infection with a temporary antibiotic device (such as an antibiotic nail)
    • Continued attempt to suppress the infection with oral antibiotics until fracture union, followed by implant removal

    Given that the infection appeared to be relatively clinically indolent, albeit already antibiotic-resistant, the decision was made to retain the implant and continue antibiotic suppression.

    Six months later, and 12 months after the index injury, the patient showed no progress toward osseous union. The bone graft had resorbed and the alignment was stable (Figure 7). The presence of infection had likely contributed to the failure of the bone grafting procedure.

    Figure 7. No progress toward osseous union can be seen on radiographs 1 year following the index injury.

    Staged Approach to Treatment

    Now that the patient has failed an attempt at autograft and has developed an infected non-union, further treatment must proceed via a staged approach.

    In the first stage of treatment, the goal will be to eradicate infection while providing enough skeletal stability to allow for soft tissue healing. This will require the following:

    • Removal of all prior implants
    • Debridement of the fracture, as well as the intramedullary canal
    • Placement of local antibiotics
    • Intravenous (IV) antibiotics

    After eradication of the infection, the second stage of treatment will address the remaining issues:

    • Stabilization of the non-union
    • Filling of any osseous defects with bone graft (acutely or via bone transport)

    For the first stage, all prior implants were removed and the non-union and medullary canal were debrided, requiring elevation of the flap (Figure 8). Cultures were obtained intraoperatively, the defect was filled with antibiotic-impregnated cement, and the fracture was stabilized with an external fixator.

    Figure 8. Implants were removed, the non-union and medullary canal were debrided, the defect was filled with antibiotic-impregnated cement, and the fracture was stabilized with an external fixator.

    Surprisingly, the cultures were only positive for methicillin-susceptible Staphylococcus aureus.

    Multiple options were available for management of the bone defect and stabilization of the tibia:

    Circular external fixation and distraction osteogenesis would allow for stabilization of the tibia with a frame, and retrograde bifocal bone transport could be utilized to manage the bone defect. Disadvantages of this approach include the need for prolonged external fixation and management of the pin tracts around and through the patient’s large free tissue flap. Some of this could be prevented with cable transport, however. No definitive instrumentation of the infected portion of the non-union would be necessary with this approach, a clear advantage.

    Internal fixation and induced membrane bone grafting would obviate the need for prolonged external fixation. However, this non-union has already failed autograft bone grafting once, albeit in the presence of an infection at that time.

    Internal fixation with vascularized bone grafting would also preclude the need for prolonged external fixation and would have the advantage providing vascularized bone graft to the non-union site.

    The third option was chosen based on the relatively small size of the defect and the failure of the prior autograft procedure.

    A Successful Revision Procedure

    The patient was treated with IV antibiotics for 4 weeks. Revision surgery was performed 6 weeks after the debridement and included the following steps (Figure 9):

    • The cement spacer was removed and the non-union was debrided and cultured. The tibia was initially stabilized with a laterally based proximal tibia locking plate.
    • Next, vascularized medial femoral condyle autograft from the contralateral lower extremity was harvested and implanted into the defect, along with autograft cancellous bone from the distal femur.
    • The anastomosis was completed by the plastic surgeon and the fixation was augmented with a large fragment plate placed over the anteromedial surface of the tibia and containing the vascularized graft.

    Final radiographs showed the fixation and filling of the defect with a graft (Figure 10). Intraoperative cultures at the time of definitive surgery demonstrated no growth.

    Figure 9. In the revision procedure, tibia was initially stabilized with a laterally based proximal tibia locking plate. After the autograft was harvested and implanted into the defect, the fixation was augmented with a large fragment plate placed over the anteromedial surface of the tibia.

    Figure 10. Following the revision, radiographs show fixation and filling of the defect with the graft.

    The patient was kept non-weight-bearing for 3 months, and then progressed to full weight-bearing. Radiographs obtained 1 year after the revision procedure demonstrated stable fixation and incorporation of the vascularized graft (Figure 11). The patient had returned to full-time employment and complained of pain only when walking very long distances. Her peroneal nerve palsy resolved spontaneously.

    Figure 11. Radiographs obtained 1 year after the revision procedure demonstrated stable fixation and incorporation of the vascularized graft.

    Key Points

    High-energy fractures around the tibia warrant special consideration.

    • This patient presented with an open fracture and vascular injury requiring repair. Emergent treatment should prioritize vascular repair, skeletal stability, and fasciotomy, as in this case.
    • The use of a plate to augment the external fixation in a widely open Type IIIC fracture can help to ensure fracture stability around a vascular repair or otherwise compromised soft tissues.

    When faced with an impending non-union, several factors must be considered:

    • How stable is the current fixation? Patients with fixation showing signs of early failure or less-than-ideal stability may benefit from earlier intervention.
    • What is the biologic potential of the fracture? Closed fractures with a healthy soft tissue envelope may be monitored, as their potential for continued osseous healing is greater than fractures that have been previously devitalized or have a compromised soft tissue structure.
    • Has infection been ruled out? Although the diagnostic yield of laboratory studies such as CRP, ESR, and WBC count is not perfect, they offer more information if infection is suspected.

    Infected tibial non-unions, particularly those with bone loss, require a staged approach to treatment.

    • The first stage must involve eradication of infection through local and IV antibiotics as well as address soft tissue defects and provide skeletal stability.
    • Many options are available for the second stage of treatment, including circular external fixation and bone transport or internal fixation combined with induced membrane bone grafting versus vascularized bone grafting.

    Author Information

    Brandon J. Yuan, MD, is an Assistant Professor in the Division of Orthopedic Trauma, Mayo Clinic, Rochester, Minnesota.

    Disclosures: Dr. Yuan has no disclosures relevant to this article.