ANKLE ARTHRODESIS: RESULTS AFTER THE MINIARTHROTOMY TECHNIQUE
February 7th, 1996
Guy D. Paremain, MD, Stuart D. Miller, MD, and Mark S. Myerson, MD
Abstract
Of 34 ankle fusions (34 patients) performed at our institution between June 1992 and June 1993, 15 utilized the miniarthrotomy technique. This technique involves two 1.5-cm incisions, one medial and one anterolateral, through which the ankle joint cartilage and synovium are debrided. Subchondral bone resection is completed with a high-speed cutting tool, creating a "slurry" that is saved for local bone graft. The ankle is then appropriately positioned (5o of valgus, 0o of dorsiflexion, and neutral rotation), cannulated screws are inserted, the position is checked fluoroscopically, and the wound is closed. The patient receives a short leg cast at 2 weeks and a walking cast at 3 to 5 weeks until there is radiographic and clinical evidence of solid arthrodesis. In our 15 patients, follow-up ranged from 12 to 19 months after surgery and arthrodesis was radiographically evident at a mean of 6.0 weeks (range, 3 to 15 weeks). Complications were limited to a transient synovitis in 7/15 patients, which lasted approximately 3 weeks and was possibly related to the bone slurry.
Although ankle joints with marked malalignment require a more extensive open arthrodesis procedure, this miniarthrotomy technique offers decreased soft-tissue insult, decreased bone stripping, easy application, and rapid healing time for the treatment of severe degenerative changes of the ankle with minimal deformity.
Introduction
Arthrodesis, the treatment of choice for disabling pain from degenerative or inflammatory arthritis of the ankle,{4058, 4066, 4068, 2655} can be successfully achieved by a variety of techniques.{4056, 2635, 2663, 2624, 2639, 2640, 2641, 2642, 2644, 2667, 2669, 2665, 2647, 2649, 2651, 2652, 1450, 2653, 3059, 2655, 4072, 2658, 2659, 2660, 2661, 4063, 2662} As experience with arthrodesis techniques grew, it was found that the application of compression across the arthrodesis site reduced the incidence of complications, especially pseudoarthroses.{2663, 2641, 2644, 2650, 2651, 2653, 3059, 2655} This principle was first implemented through the use of external fixators: Charnley's{2639} device was the model and later the Calandruccio{2663} apparatus provided a less bulky alternative. The popularity of utilizing external fixation devices for this purpose decreased with the subsequent development of internal fixation using compression screws.{2663, 2639, 2665, 2649, 1450, 2653, 2654, 2655, 4072, 2657, 2661, 2662}
According to Mann et al,{4065} ankle fusion techniques can be divided into two categories: realignment and in-situ fusion. The open realignment transfibular arthrodesis technique involves a large exposure of the ankle and meticulous preparation of the bone surfaces and has led to good results, with a satisfactory fusion occurring between 10 and 14 weeks.{4065} In-situ fusion involves arthroscopically assisted techniques of ankle arthrodesis{4059, 2653, 4067} that have produced high rates of fusion with faster healing times but are technically demanding.{4067, 2653, 3059} We describe a compromise technique that offers decreased soft-tissue insult, decreased bone stripping, easy application, and rapid healing time for the treatment of severe degenerative changes of the ankle with minimal deformity.
Materials and Methods
Patient Population
Between June 1992 and June 1993, 34 patients at our institution underwent ankle fusion. All patients had had severe ankle pain for at least 1 year that had not resolved with conservative treatment, including nonsteroidal antiinflammatory medications, local steroids, physical therapy, and/or the use of an ankle-foot orthoses. All patients had intractable pain during walking, which limited their overall level of function.
Of the 34 patients, 19 were excluded from the study because they required treatment with open arthrotomy for severe deformity, neuropathic joints, or avascular necrosis affecting more than 25% of the articular surface of the talus. The remaining 15 patients had minimal deformity (ie <10o varus or valgus, <25% anterior or posterior subluxation, <2 X 1 cm cavitation) that was amenable to "in-situ" fusion; these patients formed our study group.
The average age of the seven men and eight women in the study group was 58 years (range, 36 to 76 years). Most of the patients (9/15) had posttraumatic arthritis, one patient had rheumatoid arthritis, three patients had osteoarthritis, one patient had arthritis following a pyogenic infection in the ankle, and one patient had osteoarthritis associated with an acquired pes planus deformity. Two of these 15 patients had revision of previous fusion.
Each of the 15 patients was managed with a miniarthrotomy technique performed by the same surgeon and clinical and radiological assessment at regular intervals after surgery.
Pre- and postoperative patient assessment was performed retrospectively, based on clinical notes, and each patient was assigned a hindfoot rating according to the 1993 AOFAS rating scale{4317}.
Technique
Surgery was performed under general anesthesia (four patients) or regional ankle block with intravenous sedation (11 patients). The method of anesthesia was decided by the patient and the anesthesiologist. Each patient was positioned supine and a tourniquet applied to the thigh was used (rarely) at the surgeon's discretion. The procedure involves two 1.5-cm incisions, one medial and one anterolateral. The first incision courses medial to the anterior tibial tendon (Fig. 1). The ankle retinaculum is opened in line with the incision, the ankle joint is visualized, and the cartilage of the anterior ankle is resected using a sharp chisel and angled curette. The second incision is then made lateral to the peroneus tertius tendon, taking care to avoid endangering the dorsal cutaneous branch of the superficial peroneal nerve during exposure of the ankle joint. The joint cartilage is similarly debrided. A set of small rongeurs aids in the debridement of synovium and cartilage. After initial resection, a small modified lamina spreader (teeth removed), alternated between medial and lateral incisions, is used to distract the ankle joint and allow additional resection of the joint surface under direct vision. Proper placement of the lamina spreader avoids tilting the talus into plantarflexion. Although only the anterior two-thirds can be adequately debrided by this method, excellent arthrodesis is achieved; to date, the lack of bony fusion across the posterior third of the ankle does not appear to be a problem. The ankle is then copiously irrigated with sterile saline. A pneumatic long burr resector (AM 10-bit, Midas Rex, Fort Worth, TX) is then used to further debride the joint surface and roughen the bone to the subchondral level. The large amount of bone slurry generated by this procedure is collected in a sputum tray for later use.
The ankle is then positioned in 5o of valgus, 0o dorsiflexion, and slight external rotation to match the contralateral ankle. This position is held with 7.0-mm cannulated cancellous screws (Synthes, Paoli, PA) in standard fashion. One guide wire is placed from the anteromedial aspect of the tibia into the body of the talus and a second guide wire is placed from the distal lateral tibia into the talus; the positions are then checked using fluoroscopic imaging to ensure they are within the body of the talus and not into the subtalar joint. The guide wires are then over-drilled and, in sclerotic bone, the talus is completely tapped. The screws are then inserted and the position checked fluoroscopically. For patients with osteopenic bone, a washer may improve compression. The reserved bone slurry is then packed about the joint surfaces. The retinaculum is closed entirely to prevent leakage of this slurry.
Routine closure is followed by a bulky cotton dressing with a medial to lateral coaptation splint and posterior mold. This dressing is changed at 2 weeks to a short leg cast. Postoperatively, patients are given oral narcotics and a nonsteroidal antiinflammatory (typically hydrocodone and ketorolac) as well as oral antibiotics (500 mg four times daily for 1 week). The patient is examined every 3 weeks. When radiographs show some consolidation of bone across the fusion site and the ankle inflammation has clinically decreased (usually between 3 and 6 weeks), a walking cast with a posteriorly placed rubber heel is used until the arthrodesis appears solid by clinical and radiographic exam.
Analysis
Paired Student t-tests were used to analyze the difference between the pre- and postoperative AOFAS scores.
Results
All 15 patients were examined for 12 to 19 months after surgery for clinical (absence of pain with ambulation or on physical examination, no motion, and no warmth or swelling) and radiographic (stable bone apposition with later appearance of osseous trabeculae across the tibiotalar arthrodesis site) union. The arthrodesis site was radiographically stable (with possible osseous bridging) as early as 3 weeks postoperatively (Fig 2); the average time for early signs of fusion to appear was 6.0 weeks (range, 3 to 15 weeks). All 15 patients achieved sound arthrodesis (Fig. 3), and analysis of the AOFAS rating scale for the hindfoot demonstrated a significant (p < 0.001) difference between the preoperative (average, 42.5) and postoperative (average, 81.5) AOFAS scores.
Complications included ankle inflammation and screw head prominence. Seven patients demonstrated warmth and swelling located at, and limited to, the anterior aspect of the ankle 3 to 7 weeks after fusion, despite radiographic appearance of complete arthrodesis; this inflammation completely resolved within 2 to 6 weeks. We believe this inflammation represented an inflammatory ankle synovitis since radiographs demonstrated trabeculation across the joint and there was no associated medial, lateral, or posterior pain. This diagnosis was not proven by arthroscopic re-examination. We attributed this transient synovitis to leakage of the bone slurry from the ankle joint. A review of patients treated later in the series whose arthrodesis sites had been carefully, tightly closed showed no incidence of this inflammation. Two patients required screw removal 1 year after surgery because of screw head prominence.
Discussion
Open arthrotomy has been the standard procedure for obtaining a stable interface for ankle fusion, regardless of the degree of deformity. However, some surgeons have questioned the need for such biologically invasive procedures and extensive soft-tissue stripping in minimally deformed ankles.{4081, 2653, 3059} Open arthrotomy is still the indicated procedure for patients with complex deformities of the tibiotalar joint.{4055, 4058, 4060, 4062, 4064, 2653, 3059, 4067, 4069, 4070, 4071} The optimal position for ankle fusion has been defined{4057, 4061} and, in a patient whose limited deformity does not require gross correction, minimally invasive modalities of ankle fusion, such as the arthroscopic technique, are effective.{4081}
The arthroscopic technique offers a high success rate (100%{2653}) and a faster time to healing than open arthrotomy (8 to 10 weeks versus 3 to 12 months, respectively{2635, 2624, 2638, 2639, 2640, 2642, 2647, 2650, 2651, 2655, 2659, 2662}), perhaps because it only minimally disrupts soft tissue and the periosteal blood supply.{3059} However, the arthroscopic technique is not suitable for complex deformities and the literature lacks a prospective study comparing open to arthroscopic fusion techniques. In addition, arthroscopic ankle surgery can be difficult, time-consuming, and plagued by complications. One study{4067} reported an average of approximately 2.5 hours per procedure. Another report{3613} indicated that, in a series of 518 cases, the complication rate was 9.8%, with the most common problem (50%) being nerve injury; the authors also described problems with instrument failure and with the use of the ankle distractor.
The miniarthrotomy technique for selected cases of ankle fusion{2653, 3059} is associated with minimal complications. In our series, the most common complication was inflammatory ankle synovitis (7/15 patients) secondary to bony slurry leakage; patients treated later in the series experienced this complication less frequently as we increased the care with which we sutured the ankle joint capsule during closure. The technique provides minimal biologic disruption, and the current study demonstrated an excellent fusion rate (100%) in rapid fashion (average, 6.0 weeks). Interestingly, the miniarthrotomy technique was performed under local anesthesia{3061} in 11 of the 15 patients; these patients had their surgery on an outpatient basis and went home on the evening of surgery. The remaining four patients had general anesthesia with an ankle block to minimize postoperative pain; all of these patients were discharged the day after surgery. Average surgical time was much less than that of arthroscopy, approximately 40 minutes. Although postoperative radiographs showed that only the anterior two-thirds of each joint appeared to have fused, the significance of the unfused posterior third of the ankle joint is unclear.
We acknowledge that the "retrospective" calculation of preoperative AOFAS scores was somewhat arbitrary. However, the treatment of our patients predated the development of the AOFAS scoring system. Therefore, we chose to compute a preoperative score from an assessment of the preoperative clinical notes in each patient's records to offer a relative pre- and postsurgical comparison. Prospective studies are currently in progress.
In summary, the miniarthrotomy technique of ankle fusion appears to provide results equal to those obtained with the arthroscopic technique, but with a shorter operative time and fewer complications. Therefore, this technique appears to be an excellent method of ankle fusion for minimally deformed joints.
Acknowledgments
The authors wish to acknowledge the editorial assistance of Elaine P. Bulson in the preparation of this manuscript.
Figure Legends
Fig. 1. Artist's sketch of the miniarthrotomy incisions.
Fig. 2. Patient #5, a 36-year-old woman with posttraumatic arthritis, as demonstrated by preoperative anteroposterior (AP) (A), mortise (B), and lateral (C) weight-bearing views. At 15 days postoperatively, the patient was asymptomatic and a lateral view (D) demonstrated radiographic filling across the anterior ankle joint. At 9 months postoperatively, AP (E), mortise (F), and lateral (G) views showed secure fusion.
Fig. 3. Patient #4, a 69-year-old man with posttraumatic arthritis, as demonstrated by preoperative AP (A) and lateral (B) weight-bearing views. At 11 weeks after fusion, AP (C), mortise (D), and lateral (E) views showed solid fusion.
|
 |
