THE MINIARTHROTOMY TECHNIQUE OF ANKLE ARTHRODESIS: A CADAVER STUDY OF OPERATIVE VASCULAR COMPROMISE AND EARLY CLINICAL RESULTS
March 8th, 1996
Stuart D. Miller, MD, Guy P. Paremain, MD, and Mark S. Myerson, MD
We present a two-part study of the miniarthrotomy ankle fusion technique for minimally deformed joints: a laboratory investigation of ankle vascularity and a clinical review of the early results. In the laboratory portion, five pairs of cadaver legs were injected with radiographic dye injection, after which arthrodesis was performed via either the open technique (one leg of each pair) or the miniarthrotomy technique (the contralateral leg of each pair). Two legs in the open arthrodesis group had disruption of the peroneal arterial circulation but no disruption was identified in the miniarthrotomy group. In the clinical portion of the study, we reviewed the results of 32 patients who had undergone ankle fusion via the miniarthrotomy technique. One patient had a nonunion and two patients had delayed union. The average time to union for the 31 patients was 8 weeks (range, 6 to 22 weeks). These results compare favorably to the arthroscopically-assisted method of fusion and offers another choice for arthrodesis of minimally deformed ankles. We concluded that with minimal disruption, there is less likelihood of arterial injury and devascularization, which may indirectly have a positive effect on the rate of ankle arthrodesis.
The techniques for performing ankle arthrodesis have steadily improved over the past several decades. Compression, first utilizing external fixators and later internal fixation, was found to decrease the incidence of pseudoarthrosis.1,2 Standard ankle fusion techniques afford excellent visualization of the ankle joint and the opportunity to provide excellent apposition of bony surfaces, but the required extensive stripping of bony surfaces and soft tissue has raised concerns about possible loss of vascularity and has led to the development of arthroscopic techniques in the 1980s.3-5 These arthroscopic techniques, suitable for in situ fusions that do not require extensive realignment, result in faster healing times, compared with standard fusion techniques, but are technically quite difficult.6 A recently developed procedure -- the miniarthrotomy technique -- uses two small incisions. This less demanding procedure offers the faster healing times associated with arthroscopic arthrodesis yet provides minimal soft-tissue stripping.
This anatomic and clinical study evaluates the premise that the miniarthrotomy technique damages the vascularity of the ankle joint less than does an open arthroscopic technique.
Materials and Methods
Eight pairs of below-the-knee specimens were obtained from eight cadavers (58 to 86 years old at time of death) 3 to 6 days postmortem. The popliteal artery in each of the 16 legs was cannulated proximal to the knee joint. An irrigating solution was infused to lubricate the endothelium and displace any postmortem clot. Then, under radiographic control to ensure appropriate arterial filling, a radiopaque mixture (0.5 liters of 50% barium sulfate and 7 g of gelatin) was manually infused via a syringe. The infusion was stopped when it was observed that no further vessels were being filled. Visualization of the major vessels required 50 to 100 ml of this barium mixture. Anteroposterior and lateral radiographs, centered on the tibiotalar joint, were then taken of each leg.
Three pairs of legs displayed evidence of severe peripheral vascular disease and failed to fill the three major vascular run-offs (anterior tibial, posterior tibial, and peroneal arteries). They were excluded from the study after unsuccessful attempts to obtain adequate filing by cannulating the arteries beyond the level of stenosis. Each of the remaining 10 legs demonstrated a satisfactory arteriogram and was set aside for 45 minutes to allow the perfusate to set.
Ankle arthrodesis was then performed on one leg of each pair via an open arthrotomy and malleolar ostectomy6: two 8-cm incisions over the medial and lateral malleoli allowed subperiosteal dissection of the malleoli and removal of the malleoli and articulating surfaces of the tibia and talus with a sagittal saw; two 7.0-mm cannulated screws secured the ankle arthrodesis.
The contralateral leg of each pair underwent arthrodesis via the miniarthrotomy technique. Two 1.5-cm long incisions were made, one anteromedial and one anterolateral in the standard locations used for ankle arthroscopy.7 The cartilage was then debrided using a curette, and a high-speed burr was used to shave the articular surfaces of the tibia and talus, including the medial and lateral ankle gutters. In a manner similar to that of an open technique, internal fixation was achieved with two 7.0-mm cannulated percutaneous screws.
Each leg was then placed (under fluoroscopic control) in the preoperative radiograph position, and corresponding postoperative anteroposterior and lateral radiographs were obtained. The radiographs for all 10 legs were then evaluated for destruction of the vessels by making a side-by side comparison of the pre- and postoperative films on a viewing box. Superimposing a tracing of one radiograph on the other made it easier to detect any difference between the two sets of films. Care was taken to note any vessel that might have been destroyed or compromised during dissection.
For the clinical part of the study, we reviewed the results of ankle arthrodesis via the miniarthrotomy technique in 32 patients (13 men, 19 women; average age, 52.5 years; range, 29 to 76 + 13.5 years) over an 18-month period between 1993 and 1994. The indication for arthrodesis was minimal ankle deformity associated with severe pain refractory to conservative treatment (medications, local steroids, physical therapy, and/or orthotic appliances). Of the 32 patients, 27 had posttraumatic arthritis, 1 had rheumatoid arthritis, 4 had degenerative arthrosis. For three patients, the miniarthrotomy procedure was a revision of a prior fusion attempt.
The ankle fusion performed under general anesthesia (10 patients) or regional anesthesia with intravenous sedation (22 patients), based on patient and surgeon preference. The procedure was performed with the patient in a supine position; although a tourniquet was applied to the thigh, it was rarely used.
For the miniarthrotomy technique, two small incisions are made over the ankle: a 1.5-cm anteromedial incision located medial to the anterior tibial tendon at the level of the ankle joint and a 1.5-mm anterolateral incision lateral to the peroneus tertius tendon. After the first incision, the ankle retinaculum is incised along the same line and the ankle joint is visualized. The lateral incision must be made with care to avoid the superficial peroneal nerve. A small lamina spreader is placed alternately in the medial and lateral incisions to allow better exposure. The lamina spreader functions as an internal distractor of the ankle joint, which in turn permits debridement and resection under direct vision. A rongeur is used to remove any debris and hypertrophied synovial tissue. A Weitlander or other soft-tissue retractor aids in visualization. If present, any visible cartilage is then resected using an angled curette. Although this technique does not provide visualization of the most posterior portion of the ankle joint, the rate and success of arthrodesis is are not affected. Small rongeurs are used to debride the synovium and cartilage, paying particular attention to the medial and lateral gutters. The joint is then well irrigated and a pneumatic long burr resector (AM 10-bit, Midas Rex, Fort Worth, TX) is used to debride the joint surface and remove bone to allow for subchondral or bony apposition. Bone slurry created by this resection tends to spill out of the joint, and it may be collected and reserved for later use as a local bone graft.
The ankle is positioned in neutral dorsiflexion with 5o of valgus and slight external rotation to match the opposite ankle. Guidewires for screw insertions are placed and their positions are checked under fluoroscopy. We currently use 6.8-mm cannulated self-drilling and self-tapping screws (Orthopedic Biosystems, Phoenix, AZ). These guide wires can be over-drilled in sclerotic bone and washers may be used in very osteopenic bone.
The first five patients managed with this miniarthrotomy technique received two screws, one from the anteromedial aspect of the tibia into the body of the talus and one from the distal lateral tibia into the talus. This latter screw is by far the more important as it obtains the best purchase in the talus and places load across the joint. For additional stability, a third screw may be placed in the lateral aspect of the joint from the fibula into the talus, as was done in the patients in the latter half of the study (Fig. 1). The final position of the screws is verified as appropriate under fluoroscopy. The bone slurry may then be replaced and packed about the joint. The joint capsule is carefully closed to prevent leakage of the slurry.
Routine skin closure is followed by a bulky cotton dressing with a medial to lateral coaptation type splint and a posterior mold. This dressing is changed to a short leg cast at the first postoperative visit. The patient is given oral narcotics, nonsteroidal antiinflammatory agents, and oral antibiotics.
Evaluation of preoperative radiographs confirm no substantial difference between the location and diameters of the major vessels within each pair of legs; there was some variation of the branching patterns of the minor vessels. Postoperative radiographs demonstrated a good "set" of the barium mixture within the vessels and that no noticeable extravasation had occurred.
The determination of minor differences in the results of the two techniques was difficult; however, the three major vessels were easily discerned. Two ankles, both in the open arthrodesis group, showed damage to the peroneal artery that had occurred during excision of the distal fibula. In one, the peroneal artery itself was damaged; in the other, the enlarged anterior perforating branch of the vessel was divided at the level of fibular excision. Damage to minor vessels was not visualized in either group. There was no visible damage to any vessel in the miniarthrotomy group.
Of the 32 patients managed with the miniarthrotomy technique, 31 demonstrated fusion. The average follow-up for the 31 patients was 23 months (range, 7 to 37 months). Their average time to radiographic fusion, ie the time from surgery to the time the arthrodesis site was radiographically stable with possible osseous bridging, was 8 weeks (range, 6 to 22 + 4 weeks). Clinical and radiographic fusion was judged as absence of pain with ambulation or, on physical examination, lack of obvious motion to the joint, cessation of warmth or swelling at the joint, and radiographic union with stable bone apposition and osseous trabeculae across the tibiotalar arthrodesis site (Fig. 2).
There were two delayed unions and one nonunion. Patient 26 had a delayed union and underwent a revision 34 weeks after his original surgery. This patient smoked two packs of cigarettes daily and did not comply with instructions to remain non-weight-bearing postoperatively; these factors were believed to be partly responsible for the difficulty. Patient 20 had a delayed union; arthrodesis finally occurred at 22 weeks. Patient 25 sustained a nonunion and is currently scheduled for further surgery. These three patients were all noncompliant with instructions for their recovery process and walked extensively on their cast instead of maintaining a non-weight-bearing status. A fourth patient (patient 22) did some walking on her cast, but obtained fusion in 7 weeks.
Complications were minor. Three patients had screws removed 1 year after surgery because of symptomatic prominence of the screw head. Seven patients experienced transient inflammation of the anterior ankle between 4 and 10 weeks, which lasted approximately 4 weeks in each patient. In these patients, the rest of the examination of the ankle was unremarkable, and no warmth or swelling occurred medially, laterally, or posteriorly. Since radiographic evidence of arthrodesis was present in these seven patients, the findings of inflammation were not indicative nor suggestive of delayed bone healing. We attributed this inflammation to leakage of bone slurry from the ankle joint. These seven patients were treated with nonsteroidal antiinflammatory medication; three of the seven required additional injection of steroid into the anterior tissues beneath the joint capsule. All symptoms completely resolved. This problem was eliminated in the more recent patients with closer attention to careful approximation of the ankle joint capsule during operative closure.
The miniarthrotomy technique of ankle fusion has a firm theoretical basis. The preliminary clinical results are very promising and demonstrate the expected early healing time. The technique appears to offer results similar to those of the arthroscopic technique. The rationale for the improved healing time would be the minimally invasive process and avoidance of disruption of small blood vessels characteristic of other procedures using a wider exposure.8
Although this study highlights the potential for vascular disruption with wider exposure, we recognize the inherent limitations of the laboratory portion of the study, which focused on the arterial blood supply to the ankle joint: there were a limited number of specimens examined, and the smaller vessels were not examined with enough sophistication. Although major vessel damage was visible, perhaps digital analysis would improve the resolution, ie the ability to see changes in small vessels. An alternative approach would be to analyze postoperative blood flow using nuclear imaging techniques. Certainly, one may conclude that a less invasive procedure will minimize the likelihood of major vessel damage.
Nuclear medicine analysis of postoperative blood flow may be helpful. A less invasive procedure should lower the chances of damaging major vessels, such as the peroneal artery, which was damaged in two limbs in the open arthrodesis procedure group.
Understanding of the blood supply to the ankle joint is incomplete, and no study has examined the effect of arthrodesis on the blood supply to the ankle, particularly the talus. The extensive collateralization about the ankle joint may minimize the effect of damage to the peroneal artery. Further investigation, similar to the study by Sevitt and Thompson9 of femoral head anastomosis, may be helpful. Such a study would involve sectioning or ligating one or more of the arteries before installation of the barium solution in an attempt to evaluate the flow to the ankle through collateral blood vessels. Alternatively, another study that might be helpful in discerning blood flow to the ankle joint would be magnetic resonance imaging or techniques in bone scan evaluation for areas of avascular necrosis. The evaluation by Scuderi et al.10 of patellar vascular compromise after a lateral release markedly changed the approach to lateral release during total knee arthroplasty. A similar study11 demonstrated improvement of the Hoke triple arthrodesis technique by avoiding avascular necrosis to the talus.
There has been no controlled clinical study of the miniarthrotomy technique versus an open procedure for arthrodesis. The miniarthrotomy technique appears to yield results similar to those of open techniques, but with faster healing times, similar to those of arthroscopic techniques. The miniarthrotomy technique is certainly easy to perform and is, perhaps, less technically demanding than an arthroscopic procedure. In the current study, there were no long-term complications. A transient inflammation of the anterior ankle, which occurred in the first few patients in our study, was of limited duration and resolved completely. We attributed this inflammation to the leakage of bone slurry from the ankle joint. Closer attention to closure of the ankle joint capsule prevented this complication from occurring in subsequent patients.
We no longer perform ankle arthrodesis arthroscopically. The miniarthrotomy approach offers similar advantages without the requirement of specialized equipment, the potential for technical difficulties, and the need for external distraction of the ankle. We are able to perform the miniarthrotomy procedure under local anesthesia in most patients, and the average time to complete the procedure is approximately 50 minutes, with most patients treated on an outpatient basis. Although the indications for using the miniarthrotomy approach are similar to those described for arthroscopic ankle arthrodesis, we are continually developing and expanding its application and uses. For example, the ideal ankle on which to perform this procedure is one with minimal deformity. However, we now recognize that mildly misaligned ankles may be approached with the miniarthrotomy technique by resecting minimal wedges with a high-speed burr. These and other similar techniques may expand the indications for this procedure without compromising the rapid and successful rate of arthrodesis.
1. Charnley J. Compression arthrodesis of the ankle and shoulder. J Bone Joint Surg. 1951; 33B:180-191.
2. Thordarson DB, Markolf K, Cracchiolo A. Stability of an ankle arthrodesis fixed by cancellous-bone screws compared with that fixed by an external fixator. A biomechanical study. J Bone Joint Surg. 1992; 74A:1050-1055.
3. Dent CM, Patil M, Fairclough JA. Arthroscopic ankle arthrodesis. J Bone Joint Surg. 1993; 75B:830-832.
4. Myerson MS, Allon SM. Arthroscopic ankle arthrodesis. Contemp Orthop. 1989; 19:21-27.
5. Ogilvie-Harris DJ, Lieberman I, Fitsialos D. Arthroscopically assisted arthrodesis for osteoarthrotic ankles. J Bone Joint Surg. 1993; 75A:1167-1174.
6. Myerson MS, Quill G. Ankle arthrodesis. A comparison of an arthroscopic and an open method of treatment. Clin Orthop. 1991; 268:84-95.
7. Hartel RM, Van Dijk CN, Van Kampen A, De Waal Malefijt M. Arthroscopic arthrodesis of the ankle -- a new technique [abstr]. Acta Orthop Scand. 1993; 64:10
8. Trueta J. Nonunion of fractures. Clin Orthop. 1965; 43:23-35.
9. Sevitt S, Thompson RG. The distribution and anastomoses of arteries supplying the head and neck of the femur. J Bone Joint Surg. 1965; 47B:560-573.
10. Scuderi G, Scharf SC, Meltzer LP, Scott WN. The relationship of lateral releases to patella viability in total knee arthroplasty. J Arthroplasty. 1987; 2:209-214.
11. Duncan JW, Lovell WW. Hoke triple arthrodesis. J Bone Joint Surg. 1978; 60A:795-798.
Fig. 1. A 41-year-old woman had severe arthritis 3 years after open reduction/internal fixation of a Weber C ankle fracture. She had subsequent surgery to remove the hardware and resect anterior osteophytes. She later underwent a miniarthrotomy fusion procedure. Preoperative radiographs (A and B) demonstrate loss of joint space and widened medial clear space. Radiographs at 6 weeks after fusion (C, D, and E) demonstrate appropriate fusion alignment and screw position.
Fig. 2. A 63-year-old man demonstrated marked loss of joint space and prominent osteophyte formation 10 years after open reduction/internal fixation of a bimalleolar ankle fracture (A, B, and C). He underwent a miniarthrotomy fusion procedure. Postoperative radiographs at 10 weeks (D and E) demonstrate apparent trabeculation across the arthrodesis.