PERIOPERATIVE COMPLICATIONS OF TOTAL ANKLE ARTHROPLASTY
January 1st, 2003
Mark S. Myerson, MD, and Kenneth Mroczek, MD
ABSTRACT
A retrospective radiographic and chart review was performed for the initial 50 patients who underwent Agility (DePuy, Warsaw IN) total ankle arthroplasty by the senior author (M.S.M.). The review focused on the perioperative complications of nerve or tendon lacerations, intraoperative fractures, acute deep infections and wound complications, and component positioning. Major wound complications were defined as those requiring a soft-tissue coverage procedure. Minor wound complications did not require soft tissue coverage and included wound breakdowns, wound edge necrosis, and superficial infections. The immediate mortise and lateral postoperative radiographs were reviewed to measure component positioning. The patients were divided into two groups to compare the initial 25 patients (Group A) with the subsequent 25 patients (Group B). There were no major wound complications in either group. Minor wound complications decreased from six in Group A to two in Group B. There were four lacerations (flexor hallucis longus, posterior tibial tendon, deep peroneal nerve, and superficial peroneal nerve), all occurring in Group A. Five patients sustained intraoperative fractures in Group A, as compared with two fractures in Group B. The number of components varying greater than 4 degrees from neutral as measured by the lateral talar, lateral tibial and mortise tibial component angles decreased by 9 percent from Group A to Group B. The only tibial component to be placed in more than 4 degrees of valgus occurred in Group A. It seems that a notable learning curve exists in the performance of total ankle arthroplasty as demonstrated by a comparison of the initial 25 patients with the subsequent 25 patients performed by one orthopaedic surgeon. This improvement most likely resulted from the use of enhanced techniques and further training with the prosthesis. This information can be used as a teaching tool to decrease the incidence of complications for surgeons performing their initial arthroplasties with this potentially technically demanding procedure.
INTRODUCTION
New designs have made total ankle arthroplasty a viable treatment alternative for patients with ankle arthritis who fail conservative measures.{9249}{10134} The Agility (DePuy, Warsaw IN) ankle arthroplasty involves an arthrodesis of the distal tibiofibular syndesmosis, allowing weight transfer to the fibula and providing a larger surface area for support of the tibial component.{8615}{9371} A well-designed retrospective study of the initial 100 arthroplasties with a 2- to 12-year follow-up provided encouraging results with a low complication rate.{9371} However, a subsequent multicenter study highlighted the potential for a high rate of complications with the initial use of this implant.{10185} Based on these reports, we hypothesized that the incidence of perioperative complications is a function of the surgeon’s experience with the prosthesis. The goal of this retrospective study was to determine which complications occurred in the initial experience of one surgeon with the Agility prosthesis and to determine whether the incidence of these complications decreased with increasing experience with the implant.
MATERIALS AND METHODS
We reviewed our experience with the first 50 patients who underwent an Agility total ankle arthroplasty performed by the senior author, focusing on the operative procedure, immediate postoperative complications, and radiographic positioning of the components. The medical record and immediate postoperative radiographs were reviewed for lacerations of nerves or tendons, fractures, acute deep infections, and wound complications. No changes were made in the technique used for either exposure, the use of intraoperative alignment guides, distraction with a unilateral external fixator, or the sizing of the components used. Major wound complications were defined as those requiring a soft-tissue coverage procedure, and minor wound complications included wound breakdown, wound edge necrosis, and superficial infection.
The position of the components was measured using the methods described by Pyevich et al.{9371} On the lateral radiograph, the tibial component position was determined by the angle between the posterior tibial line and the superior surface of the tibial component (Fig. 1). The placement of the talar component was also measured on the lateral radiograph by determining the angle formed by lines from the anterosuperior aspects of the talus to the posteroinferior talus and then to the posteroinferior edge of the component (Fig. 1). The tibial component position on the mortise radiograph was calculated using the angle between the lateral border on the tibia and the superior surface of the tibial component (Fig. 2). Variation of more than 4° from the neutral position was considered unacceptable.
The patients were divided into two groups to compare the data from the initial 25 procedures (group A) to the subsequent 25 procedures (group B). The average age of patients in Group A and Group B was 57.0 years (range, 27-73 years) and 53.9 years (range, 24-78 years), respectively. Group A was comprised of 10 males and 15 females, and Group B consisted of 6 males and 19 females. The arthroplasties were performed in 12 left ankles and 13 right ankles in Group A and on 11 left ankles and 14 right ankles in Group B. Diagnoses in Group A included posttraumatic arthritis (11), osteoarthritis (7), rheumatoid arthritis (4), hemachromatosis (1), ankle pseudarthrosis after attempted fusion (1), and psoriatic arthritis (1). In Group B, diagnoses were posttraumatic arthritis (13), rheumatoid arthritis (7), osteoarthritis (4), and systemic lupus erythematosus (1).
RESULTS
There were two tendon lacerations (flexor hallucis longus and posterior tibial tendon) and two nerve lacerations (deep peroneal nerve and superficial peroneal nerve) in Group A, with no lacerations occurring in Group B. Intraoperative fracture of either the fibula or medial malleolus occurred in five patients in Group A and in two patients in Group B. There were five intraoperative fractures in Group A, three isolated medial malleolar fractures, one isolated fibular fracture, and one fibular and medial malleolar fracture. In Group B, two isolated medial malleolar fractures occurred. Four of the five isolated medial malleolar fractures from both groups were not displaced and did not appear to require fixation, whereas the patients with isolated lateral malleolar and bimalleolar fractures required intraoperative fixation.
There were no acute deep infections or major wound complications in either group. Minor wound complications (Fig. 3) were treated with daily topical antibiotic dressing changes and oral antibiotic prophylaxis. These minor wound complications included superficial dehiscence of the incision measuring 1-1.5 cm in two patients (Group A) and superficial infection in the remaining patients with these complications. Due to the small numbers, no trend was detectable with regard to patient diagnosis and wound complications.
Group A radiographic measurements revealed unacceptable measurements with variation of more than 4° from the neutral position for the lateral tibial, lateral talar, and mortise tibial angles for 9, 14, and 13 patients, respectively. Group B radiographic values were unacceptable in 7, 11, and 11 patients, respectively, for these angles. The only patient with more than 4° of valgus in the tibial component occurred in Group A (Fig. 4). Twelve patients in Group A and eight in Group B had two or more unacceptable values for the three measurements studied.
DISCUSSION
Wound complications in the immediate postoperative period after total ankle arthroplasty have been noted by many authors.{8579}{8595}{9371} {10185} {8589}{8577} The reported incidence of minor wound complications such as delayed wound healing, edge necrosis, and superficial infection ranges from 2 to 40%, although these reports include different prostheses using various surgical approaches. {8579}{8595}{9371} {10185} {8589}{8577} The two largest series on ankle arthroplasties via an anterior approach {9371}{8595} cite only a 2% and 3% minor wound complication rate in 100 and 204 patients, respectively, whereas smaller studies report a 13% rate in 30 patients and a 39% rate in 36 patients.{10185}{8577} These data suggest that familiarity with technique reduces wound complications if one assumes the authors with the larger series have more experience than the authors with smaller series. We used a single anterior incision for exposure in all cases. In the current study, the rate of minor wound complication decreased from 24% in Group A to 8% in Group B.
Wound complications are potentially disastrous because skin breakdown may lead to a superficial or deep infection. In a report on the Beck-Steffee total ankle implant{8577} utilizing an anterior approach, a 6% deep infection rate was noted, with all patients with deep infection showing prior wound dehiscence. Deep infections are difficult to manage and may necessitate implant removal. Furthermore, wound complications frequently prolong the immobilization period, preventing early motion. This delay in early motion and potential scar formation after the healing of the breakdown may have a negative effect on final range of motion. Prevention of these complications includes utilizing extensile incisions, particularly in posttraumatic patients with tight soft tissues.
Implications for Treatment
Although an extensile approach is advocated, our findings indicated that a single anterior incision may provide adequate exposure. If possible, methotrexate and other immunosuppressant medications should be stopped 1 week preoperatively and reinstated once wound healing is present. Wynn and Wilde{8577} noted that 71% of those who suffered wound dehiscence were steroid-dependent patients with rheumatoid arthritis. We have also learned to avoid unnecessary and excessive retraction by retracting on only one side at a time or at different levels, followed by meticulous wound closure in layers. Most wound complications involve minor edge necrosis and may be treated topically with wound care and oral antibiotic prophylaxis.
Nerve and tendon laceration represent potential complications during ankle arthroplasty. Saltzman and colleagues {10185} reported on 30 ankle arthroplasties comprised of the initial ten cases of three experienced foot and ankle surgeons. Their results of three tendon lacerations in these initial cases were similar to those of the present study. Our results suggest that proper soft tissue release and familiarity with surgical technique allowed us to avoid these lacerations in Group B. Most nerve injuries occur either by excessive stretching through retraction or by improper release and protection distally in the foot. Without appropriate release and protection, the superficial and deep peroneal nerves are at risk of injury distally in the foot by the extremes of the saw blade during the bone cuts, especially the cut of the dome of the talus. The posteromedial tendons are at risk if care is not taken when advancing posteriorly with the saw blade during bone cuts. If one is unsure of the depth, the posterior extent of the cut should be performed slowly, and by repeated smaller excursions of the saw blade, rather than attempting to remove the entire posterior tibia with one cut. This is more of a problem following posttraumatic arthritis where sclerosis and fibrosis are present posteriorly.
A high incidence of intraoperative fractures was reported by Saltzman and colleagues during their initial cases,{10185} and the present study suggests that intraoperative fractures may decrease with experience with the procedure and the prosthesis. Fractures may occur during bone cuts or prosthesis insertion and may be avoided with careful technique. Choosing the appropriate prosthesis size will decrease the chance of intraoperative fracture. Prosthesis size should be determined using preoperative templates and intraoperative fluoroscopy.
Intraoperative fluoroscopy is also helpful to ensure proper cutting block placement prior to the bone cuts and to verify the appropriateness of these cuts afterwards.
The fibula is located posterior to the tibia and obliquely inclined at 30 degrees. As a result, the bulk of the lateral malleolus is located posteriorly. The lateral malleolus may be inadvertently cut while performing the tibial cuts if the fibular position is not noted. Protecting the lateral malleolus with a curved osteotome and placing a lamina spreader between the tibia and fibula to distract the lateral malleolus away from the saw will prevent inadvertent fibular fracture with the saw blade. If we were uncertain as to the position of the fibula in the current study, we found that it could easily be cut freehand, if necessary, after the tibial cut. Prior to the tibial cuts, the surgeon should check that the medial apex of this cut will leave a large enough medial malleolus to prevent fracture. If the cut is too proximal and the remaining malleolus is inadequate, the cutting guide may be lateralized to effectively increase the size of the medial malleolus. Insertion of the components should always proceed slowly, and the components should be removed if difficulty is encountered. Forcing insertion may result in a fracture. An off-center or medialized fin may cause a medial malleolar fracture during insertion. If this is encountered with the tibial component, the fin cut may be enlarged laterally and the medial gap bone grafted after insertion. Often, the medial tibial cut is not vertical and a posteromedial ledge of bone on the medial malleolus will need to be removed with a reciprocating saw so it will not block insertion of the tibial component and result in a medial malleolar fracture.
Proper placement of the component should maximize its chances for long-term survival by decreasing inappropriate stresses on the prosthesis. Pyevich et al. showed that greater than 4 degrees of valgus in the tibial component measured on the mortise radiograph was associated with a significantly higher rate of pain.{9371} In the present study, only one component was placed in greater than 4 degrees of valgus. This component was placed at 6 degrees and occurred in Group A. Although varying a component by more than 4 degrees in other directions has not been shown to be associated with pain, proper alignment within 4 degrees of varus or valgus positioning is considered a key indicator of a successful procedure. In the current study, we were able to achieve improved positioning of the components in the later procedures as reflected in all other measured angles. Larger studies will need to be undertaken in order to determine if such error is significant. Further studies of results with total ankle arthroplasties may discover different angles and measurements that will assist in defining ideal component placement.
We found that care must be taken to avoid valgus tilting of the talus by the medial-placed external fixator during distraction. Adjusting the fixator and releasing some distraction may correct this problem, but if these steps do not correct the valgus talar tilt, the surgeon should take the talar position into account during the bone cuts. Appropriate distraction should always be utilized or not enough bone will be removed, which will lead to limited motion from an overstuffed joint. The alignment guide should be set off the tibia and not the foot. Although this guide is designed to be parallel to anterior tibia in the sagittal plane, this positioning may result in a plantarflexed cut by removing more tibia anteriorly then posteriorly. The current authors have found that lifting the alignment guide off the tibia distally will avoid this plantarflexed cut and provide more motion.
The current study is limited in that it is retrospective and involves only 50 patients, but it suggests that a learning curve exists with the Agility (DePuy, Warsaw IN) ankle arthroplasty. Further study should be revealing, but this preliminary information may help those contemplating their first total ankle arthroplasties with this device.
REFERENCES
1. Acevedo, J. I., and Myerson, M. S.: Reconstructive alternatives for ankle arthritis. Foot Ankle Clin., 4:409-430, 1999.
2. Alvine, F. G.: Total ankle arthroplasty: new concepts and approach. Contemp. Orthop., 22:397-403, 1991.
3. Bolton-Maggs, B. G., Sudlow, R. A., and Freeman, M. A.: Total ankle arthroplasty. A long-term review of the London Hospital experience. J. Bone Joint Surg., 67B:785-790, 1985.
4. Kitaoka, H. B., and Patzer, G. L.: Clinical results of the Mayo total ankle arthroplasty. J. Bone Joint Surg., 78A:1658-1664, 1996.
5. Pyevich, M. T., Saltzman, C. L., Callaghan, J. J., and Alvine, F. G.: Total ankle arthroplasty: a unique design. Two- to twelve-year follow-up. J. Bone Joint Surg., 80A:1410-1420, 1998.
6. Saltzman, C. L.: Perspective on total ankle replacement. Foot Ankle Clin, 5:761-775, 2000.
7. Saltzman, C. L., Alvine, F., Sanders, R., and Gall, R.: Challenges with initial use of a total ankle. Presented at the Annual Meeting of the American Academy of Orthopaedic Surgeons, Orlando (FL), 2000.
8. Unger, A. S., Inglis, A. E., Mow, C. S., and Figgie, H. E., III: Total ankle arthroplasty in rheumatoid arthritis: a long-term follow-up study. Foot Ankle, 8:173-179, 1988.
9. Wynn, A. H., and Wilde, A. H. : Long-term follow-up of the Conaxial (Beck-Steffee) total ankle arthroplasty. Foot Ankle, 13:303-306, 1992.
|
 |
