Ankle Fracture Management in a Patient with Diabetes Mellitus
A 72-year-old patient’s fibula fracture has failed to heal 3 months after surgery. Not only that, but a full-thickness skin ulcer has also developed over the medial malleolus. What went wrong, and what should be done during the revision procedure to ensure fracture union?
Brandon J. Yuan, MD
A 72-year-old male patient presents with a right ankle fracture dislocation sustained in a slip and fall. This patient takes insulin to manage type 2 diabetes mellitus, and his medical history is significant for peripheral neuropathy, plantar ulcerations on the left foot, nephropathy, and retinopathy. His recent hemoglobin A1C was 7.6%.
Figure 1 shows the comminuted and displaced fibula fracture with ligamentous medial injury (deltoid ligament injury). Initially, this was a closed fracture. The treating surgeon, however, noted an area of significant ecchymosis over the medial malleolus.
Figure 1. Radiographs showing comminuted and displaced fibula fracture with ligamentous medial injury.
Patients with diabetes, particularly those with peripheral neuropathy, have a significantly increased risk of nearly every feared postoperative complication of ankle fracture surgery, including: 
- Wound complications
- Loss of reduction
- Need for repeat surgery
Multiple authors have advocated for increased rigidity of internal fixation, prolonged postoperative immobilization, and weight-bearing restrictions for these patients. 
The treating surgeon elected for early open reduction and internal fixation (ORIF) of the fibula fracture. In the absence of significant soft tissue injury or swelling, early surgery can help to maintain anatomic reduction of the tibiotalar joint and prevent repeated dislocations of the ankle joint, which could lead to significant articular injury as well as soft tissue damage. External fixator should be utilized when the reduction cannot be maintained anatomically with a splint and when there is a question regarding the suitability of the soft tissue envelope for ORIF.
The initial postoperative radiographs (Figure 2) demonstrate near anatomic reduction of the fibula with a one third tubular plate and anatomic reduction of the tibiotalar joint.
Figure 2. Postoperative radiographs following ORIF of the fibula fracture.
The patient was kept non-weight-bearing in a cast for 3 months postoperatively. Radiographs obtained at the 3-month follow-up visit (Figure 3) show a readily apparent fibular fracture line, lateral subluxation of the talus, and medial clear space widening. In addition, the plate is bent through the open hole at the fracture site, allowing for increased valgus angulation of the fibula through the fracture. Failure of the lag screws can be seen as well.
Figure 3. The fibular fracture line is readily apparent in radiographs from the 3-month follow-up visit.
In addition, the increased tension over the already compromised medial soft tissues had resulted in the formation of a full-thickness skin ulcer over the medial malleolus (Figure 4). However, there was no drainage from this area, no fluctuance, and no abscess formation.
Figure 4. Full-thickness skin ulcer over the medial malleolus.
What Went Wrong?
The treating surgeon was on the right track and thinking about the increased risk for fixation failure in this patient:
- The patient was kept non-weight-bearing in a cast for a prolonged period after surgery (3 months).
- The fixation construct was extended more proximally than is typical, and a total of 3 lag screws (2 independent and 1 through the plate) were placed.
Although the postoperative protocol was appropriate, the fixation construct was not designed to appropriately resist the significant deforming forces on the fibula.
When considering the method for fixation of any fracture, it is important to take into account the deforming forces present. This patient’s fracture will naturally try to assume the position that the fibula was in at the time of injury, which is primarily due to lateral translation of the talus. The increased proximal fixation in the plate does increase the force necessary for the plate to fail by pulling off the bone proximal to the fracture, but it does nothing for improving the resistance to bending forces on the plate at the fracture itself (Figure 5, black arrow).
In addition, the strength of the interface between the 2 most distal screws and the bone (Figure 5, circle) is less than critical, as the necessary function of the plate is to buttress the fibula against lateral translation of the talus. Thus, the deforming force will have minimal effect on pulling the plate away from the bone distal to the fracture.
Besides the rigidity of the plate, the only resistance to the main deforming force on the fibula is the few threads of the lag screws biting into the distal aspect of the proximal segment (Figure 5, red arrows).
Because the one third tubular plate is relatively flexible, the failure seen in the radiographs can be predicted. The lag screws will lose fixation and the plate will bend through the empty screw hole at the fracture site.
Figure 5. Radiographs showing failure of the initial fixation construct.
Options for the Revision Procedure
At this point, the patient presented to the author’s clinic for a consultation regarding a revision procedure.
Delaying revision surgery may be tempting because of the presence of the open medial wound, which represents a risk for deep infection. However, this would be a mistake: The medial wound is present primarily because of the failed fixation and tension on the medial soft tissues created by lateral subluxation of the talus.
Inflammatory markers were obtained and were mildly elevated. In addition, the patient underwent a standard work-up for fracture non-union, including white blood cell count and endocrine labs (vitamin D, thyroid stimulating hormone, parathyroid hormone, alkaline phosphatase, calcium, and phosphorus). All were within normal limits.
Feasible options for the revision include:
Removal of hardware and application of an ankle–spanning external fixator. This is a particularly good option if infection is a strong concern. It involves removal of the failed hardware and reduction of the tibiotalar joint through an external fixator, which relieves the tension on the medial soft tissues while minimizing the risk of new infection. This option can be utilized as a bridge to definitive internal fixation or, in rare cases, as a definitive measure. However, obtaining union of the oblique fibular non-union with an external fixator would be difficult.
Removal of hardware and revision internal fixation in a single procedure. This option is acceptable if there is no concern for significant deep infection due to the medial wound. It has the advantage of being the most effective way to rigidly hold the tibiotalar joint reduction while obtaining union of the fibula fracture.
Reduction of the Fibula Fracture and Tibiotalar Joint
The second option was chosen, and the patient underwent revision reduction and internal fixation in a single stage. His previous lateral incision was utilized for exposure. The non-union site was debrided, and the fracture reduced (Figure 6). This afforded excellent reduction of the tibiotalar joint as well. Autograft bone from the ipsilateral calcaneus was used to augment the fracture site.
Figure 6. Reduction of the fibula fracture and tibiotalar joint.
The final fixation construct is shown in Figure 7. The bone graft harvest site from the calcaneus is visible on the lateral radiograph.
Two important points regarding the fixation construct:
A more-rigid, pre-contoured, non-locking plate was utilized in the revision procedure. As was noted previously, the weak point in the previous construct primarily related to the flexibility of the one third tubular plate. The thickness and rigidity of the plate is important; however, the use of locking screws was unnecessary in this case. The quality of the bone in the distal fibular diaphysis is quite good, and the use of non-locking screws is not only adequate, but also necessary to maximize the full buttress effect of the plate at the apex of this fracture. If the plate is not fully compressed down to bone with a cortical screw at the apex of the fracture, the buttress effect of the plate will be lost. The use of locking screws in the proximal aspect of the plate could improve the resistance to “lift-off” of the plate from the bone proximal to the fracture, but as demonstrated in the failure radiographs, that is not the weak point of this construct.
The most important cortical screws – those near the apex of the fracture and crossing the fracture (Figure 7, red arrows) – were augmented by anchoring them through the tibia. The purpose of these screws is similar to that of the lag screws in the prior construct: As the talus attempts to translate laterally, these screws resist the deforming force through their fixation in the bone. Placing these screws through the fibula and into the tibia dramatically increases their pull-out strength and, thereby, the overall strength of the construct.
Figure 7. Final fixation construct.
The patient was kept non-weight-bearing in a cast for 12 weeks. Radiographs obtained at that time demonstrated osseous union of the fibula (Figure 8). The medial wound was healed. At 2 years after surgery, radiographs demonstrate union of the fibula and maintenance of the reduction of the tibiotalar joint, albeit with moderate degenerative changes (Figure 9).
Figure 8. Radiographs showing osseous union of the fibula 12 weeks after the revision procedure.
Figure 9. Radiographs showing union of the fibula and reduction of the tibiotalar joint 2 years after the revision procedure.
- Patients with diabetes who sustain an ankle fracture, particularly those with neuropathy, are at significantly increased risk of multiple complications, including infection, loss of reduction, non-union, need for repeat surgery, and amputation.
- Increased fixation strength is often necessary to prevent loss of reduction and non-union. Study the injury radiographs: They are a roadmap for the deforming forces that will be applied to the fixation construct. Ensure that the construct design will optimally resist these deforming forces. The indiscriminate application of longer plates, more screws, or locking screws is not always the answer for the increased fixation strength necessary.
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.
- Hak DJ, Egol KA, Gardner MJ, Haskell A. The “not so simple” ankle fracture: avoiding problems and pitfalls to improve patient outcomes. Instr Course Lect 2011; 60:73-88.