SALVAGE OF PSEUDOARTHROSIS AFTER TIBIOTALAR ARTHRODESIS
April 18th, 1997
Stuart E. Levine, MD, Mark S. Myerson, MD, Paul Lucas, MD, and Lew C. Schon, MD
Abstract
We retrospectively reviewed the treatment of a selected group of 23 patients with pseudoarthrosis after ankle arthrodesis who underwent revision arthrodesis at an average of 1.7 years (range, 0.3 to 17.0 years) after the initial, unsuccessful procedure. Fourteen patients underwent isolated revision tibiotalar arthrodesis and nine had a hindfoot arthrodesis (seven tibiotalocalcaneal, two pantalar) performed at the time of the procedure. Rigid internal fixation with screws was performed when possible and, in patients with poor bone quality, an external fixator was used. Autogenous bone grafting was used in 14 patients where bone loss was present. Twenty-one of 23 patients had successful union (average, 14 weeks; range, 6 to 48 weeks). Two patients had persistent ankle nonunions; one was asymptomatic, and one had symptomatic subtalar arthritis. Two patients had successful arthrodesis but persistent pain from reflex sympathetic dystrophy. Overall, 19/23 patients were satisfied with the surgery. We conclude that revision arthrodesis for pseudoarthrosis is a worthwhile procedure.
Introduction
Arthrodesis of the ankle joint was first described in 1879,14 and since then, many techniques have been described for this procedure.1-3,5-7,10-17,19-21,24,26,27,29-31 The rate of pseudoarthrosis after ankle arthrodesis has been reported to range from 0 to 40%.4,24
The management of pseudoarthrosis after ankle arthrodesis is not well described. Kirkpatrick et al.7 reported on 11 patients who underwent revision arthrodesis via a variety of internal fixation techniques with or without bone graft.7 Nine of the revision arthrodeses went on to union, one patient developed a painless fibrous union, and one patient with myelodysplasia developed a recurrent nonunion and required amputation. Kitaoka8 reported a 78% union rate for revision surgery after failed total ankle arthroplasties. Russotti et al.23 described a technique for tibiocalcaneal arthrodesis using a posterior approach and placing iliac crest graft posteriorly to perform an extraarticular arthrodesis. Eight of their 21 fusions were performed for nonunion of ankle arthrodesis.
We report our experience with revision of pseudoarthrosis of ankle arthrodesis in a specific subset of patients who had no complicating features of systemic disease, infection, or avascular necrosis.
Methods and Materials
We retrospectively reviewed the treatment of 23 patients whose ankle arthrodesis was initially complicated by pseudoarthrosis and who subsequently underwent revision arthrodesis from 1986 to 1993 by the senior author (MSM). Office and hospital records as well as all radiographs were reviewed for each patient to determine the diagnosis that necessitated the initial arthrodesis, date of initial and revision surgery, techniques used for initial and revision surgery, and time to radiographic union. We excluded any patient with a history of neuropathy, infection, or systemic arthritides, renal disease, endocrine disorders, steroid usage, arterial insufficiency or extensive avascular necrosis of the talus. One patient had reflex sympathetic dystrophy (RSD) preoperatively and one patient had partial avascular necrosis of the talus secondary to trauma.
Initial Procedure Data
The mean age of the 18 men and five women at time of the initial arthrodesis was 53 years (range, 28 to 74 years). The initial arthrodesis was performed for posttraumatic arthritis in 17 patients and for osteoarthritis without history of major trauma in six patient. Two patients (patients 2 and 16) underwent two revision procedures for pseudoarthrosis, the second of which is the index procedure in this study. Six of the 23 initial arthrodeses were performed at our institution (patients 3, 10, 11, 15, 16, and 17); 17 were performed elsewhere. The initial arthrodesis procedure utilized various techniques and the original fixation was accomplished with screws of different sizes and configurations, staples, blade plates, and external fixation.
Revision Procedure Data
The mean time between the previous, unsuccessful arthrodesis and the revision arthrodesis was 1.7 years (range, 0.3 to 17 years). The patient with the earliest nonunion had gross motion, broken hardware, and marked lucency 4 months after the initial fusion. The revision arthrodesis included the ankle alone in 14 patients, and nine patients underwent additional hindfoot arthrodeses (seven tibiotalocalcaneal, two pantalar). The additional hindfoot joints were included in the arthrodesis if symptomatic arthritis was noted and if pain was alleviated by preoperative injection of lidocaine into the affected joint. Fixation for the revision ankle arthrodesis was obtained with a variety of screw techniques (Figs. 1 and 2). In five patients with poor quality bone, a Calandruccio external fixator (Richards, Memphis, TN) was used; in these patients, the distal pins were inserted into the talus. Although all patients were informed that autogenous bone graft might be required, the use of this graft was determined by the size of the bone defect created by debridement of avascular and necrotic bone.
Postoperatively, patients remained non-weight-bearing for 6 weeks, after which they were allowed to bear weight as tolerated in a short leg walking cast. Radiographs were obtained at 2, 6, and 12 weeks postoperatively, and then approximately every 3 weeks until union. Commercially available removable boots were not used as we believe they allow some motion across the ankle, in contrast to a well-molded cast. The patients with internal fixation remained in a cast until both clinical and radiographic union was determined to be present. The onset of arthrodesis was determined by a combination of radiographic evidence and clinical impressions, including the absence of pain, warmth, and focal swelling. These clinical findings were considered important because radiographic evidence alone was insufficient due to multiple areas of sclerosis and lucency secondary to the previous arthrodesis procedure(s) and to overlap of the fixator pins on the lateral radiograph. Once the fixator was removed, these patients used a short leg weight-bearing cast until conclusive evidence of arthrodesis was present.
All patients were evaluated postoperatively by interview and clinical examination using the American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot score.9
Results
Rate of Union
Radiographic union was ascertained by the appearance of continuous osseous trabeculae across the arthrodesis. The fusion rate for the revision procedures was 91% (21/23 patients) and occurred at an average of 11 weeks (range, 6 to 48 weeks). The two patients (patients 16 and 17) who did not fuse successfully had surgery limited to the tibiotalar joint. Patient 17 had a stable nonunion and was relatively asymptomatic. Patient 16 had a nonunion and additional subtalar arthritis, and symptoms were relieved by injection of local anesthesia into the sinus tarsi, indicating that the subtalar arthritis may have largely contributed to the symptoms.
Complications
Among the 21 patients with successful union, one (patient 15) developed cellulitis postoperatively, which responded to local wound care and oral antibiotics, and one (patient 12) developed subsequent subtalar arthritis and underwent a subtalar fusion, which went on to a painful nonunion. One patient (patient 17), who had RSD syndrome before his revision surgery, continued to be symptomatic from the RSD postoperatively, and one patient (patient 2) developed RSD postoperatively and had continued pain. There were no pin-tract infections in patients treated with external fixation. Four patients underwent hardware removal for symptomatic screw heads: two on the plantar posterior surface of the calcaneus and two over the medial malleolus.
Clinical Results
Twenty-one of 23 patients were examined at an average of 4.4 years (range, 1.8 to 8.3 years) after revision surgery. One patient (patient 13) died 4 years after surgery; data was available from this patient's last examination at 2 years after surgery. One patient (patient 23) was unable to return for further examination due to the travel distance involved. The AOFAS scores of these two patients were based on their last examinations and telephone discussion (with patient 23). Nineteen patients reported having no or mild pain, one had moderate pain, and three (patient 16, who had subtalar arthritis associated with a pseudoarthrosis of the ankle; and patients 2 and 7, who had RSD and severe pain). Seventeen patients were able to participate in daily activities without brace or support, one could participate in daily activities while wearing an ankle foot orthosis, and three (patients 2, 7, and 16) reported limitations of their daily activities. Nineteen of the 23 patients were happy with the results of their procedure and were pleased that they had had the surgery. The four dissatisfied patients included the patient with the subsequent subtalar nonunion (patient 12), the two patients with RSD (patients 2 and 7), and the patient with symptomatic subtalar arthritis associated with a recurring pseudoarthrosis of the ankle (patient 16).
The overall average AOFAS Ankle-Hindfoot Scale9 score was 66 points (range, 16 to 92 points). Three of the lower scores (16, 36, and 49 points) belonged to the patients with RSD and the patient with a subtalar nonunion, respectively. Excluding these three patients, the average AOFAS score was 72 points (range, 39 to 92 points). Preoperative scores were not available for this group of patients.
Discussion
Arthrodesis of the ankle can provide a durable, pain-free limb with good functional results. Unfortunately, not all ankle arthrodeses are successful. In this study, we included failures from nonunions related to mechanical and technical or local conditions (apart from infection) and excluded systemic illnesses. In all our cases, nonunion was the indication for revision. Nonunion after tibiotalar arthrodesis may occur because of infection, inadequate fixation, inadequate bone preparation, poor bone stock, insufficient postoperative immobilization, or avascular bone. In our experience, previous subtalar fusion will increase the forces across the attempted fusion of the tibiotalar joint and may increase the risk for pseudoarthrosis.
In a review of 78 ankle fusions, Frey et al.4 found that nonunion was associated with avascular necrosis, infection, renal failure, diabetes, significant smoking history, and alcohol abuse. In posttraumatic arthritis, severe fracture patterns and open fractures correlated with higher nonunion rates. Patients with decreased sensation, especially that associated with neuroarthropathy, are particularly prone to developing pseudoarthrosis.4,10,20,22,25,28 These patients represent quite a different clinical profile than those with a standard nonunion and were not included.
It is important to stress that not every nonunion requires revision. A relatively painless stable nonunion may require only an ankle-foot orthosis or rocker-bottom shoe. On the other hand, a nonunion combined with infection or associated with diffuse foot pain may be best treated with an amputation, especially if numerous previous surgical procedures have been performed. In cases where several factors co-exist, it is critical to isolate the primary cause of the patient's pain. Before proceeding with revision arthrodesis, radiographs of the hindfoot should be carefully examined for other joints that may be the source of symptoms. If the cause of pain is unclear, selective injection may be useful for localization. If several sites appear to contribute to their pain, then each must be addressed at the time of surgery to achieve success.
In patients who did not require adjacent symptomatic joint arthrodesis, the nonunion site was approached through the previous incisions whenever possible. If there was gross motion and good alignment in the frontal plane, a miniarthrotomy technique was used.18 Varus or valgus malunion in addition to pseudoarthrosis was corrected by using a saw to cut flat surfaces. The deformity could be corrected by excising wedges or translating the talus medially or laterally. If a sizeable bone defect was present medially or laterally, a tricortical iliac crest bone graft was inserted to support and correct the deformity.
We were able to obtain union in 21 of 23 patients (91%) with pseudoarthrosis of the tibiotalar joint. Of the two patients (patients 16 and 17) with nonunions after revision, one is mildly symptomatic and one has symptoms referable to subtalar arthritis. Despite successful ankle arthrodesis, two other patients had poor outcomes. Both developed RSD: one had a failed painful subtalar arthrodesis, and one had persistent intractable subtalar joint pain. These two patients were the most symptomatic of the group.
In this series, 4/23 patients (17%) underwent hardware removal to improve their outcome. This is in distinction to the 5% hardware removal rate in our primary ankle or subtalar fusions. It was apparent that, unlike the primary cases, revision patients require more creative and aggressive hardware placement for adequate purchase. Screw holes previously used for trauma reconstruction and subsequent arthrodesis at times mandated the use of hardware with subcutaneous screw heads that were prominent with shoe wear.
In conclusion, nonunion after failed ankle arthrodesis may be salvaged with revision. Mechanical and technical factors that led to initial failure must be addressed. All interposed fibrous tissue and avascular bone must be removed. Excision of bone is performed as necessary to place the foot in neutral dorsiflexion and 5o of hindfoot valgus. Rigid fixation is essential to prevent recurrence of the nonunion. If the bone quality is poor, external fixation used alone or in conjunction with internal fixation in compression may be necessary.
Every effort must be made to achieve bone-to-bone apposition and to avoid large interposition grafts. Autologous bone graft is placed in any defects that remain between the bleeding bony surfaces. This graft may be harvested from the excised fibula, iliac crest, talus, or calcaneus. Plantigrade positioning of the foot and simultaneous arthrodesis of any symptomatic arthritic joints should lead to a satisfactory result.
References
1. Anger, D.M., Ledbetter, B.R., and Calhoun, J.H. The treatment of ankle fusion with the Ilizarov external fixator. Presented at the 9th Annual Summer Meeting of the American Orthopaedic Foot and Ankle Society, Asheville (NC), July 25, 1993.
2. Campbell, C.J., Rinehart, W.T., and Kalenak, A. Arthrodesis of the ankle. Deep autogenous inlay grafts with maximum cancellous-bone apposition. J. Bone Joint Surg. 56A:63-70, 1974.
3. Charnley, J. Compression arthrodesis of the ankle and shoulder. J. Bone Joint Surg. 33B:180-191, 1951.
4. Frey, C., Halikus, N.M., Vu-Rose, T., and Ebramzadeh, E. A review of ankle arthrodesis: predisposing factors to nonunion. Foot Ankle Int. 15:581-584, 1994.
5. Holt, E.S., Hansen, S.T., Mayo, K.A., and Sangeorzan, B.J. Ankle arthrodesis using internal screw fixation. Clin. Orthop. 268:21-28, 1991.
6. Johnson, E.W., Jr., and Boseker, E.H. Arthrodesis of the ankle. Arch. Surg. 97:766-773, 1968.
7. Kirkpatrick, J.S., Goldner, J.L., and Goldner, R.D. Revision arthrodesis for tibiotalar pseudarthrosis with fibular onlay-inlay graft and internal screw fixation. Clin. Orthop. 268:29-36, 1991.
8. Kitaoka, H.B. Salvage of nonunion following ankle arthrodesis for failed total ankle arthroplasty. Clin. Orthop. 268:37-43, 1991.
9. Kitaoka, H.B., Alexander, I.J., Adelaar, R.S., Nunley, J.A., Myerson, M.S., and Sanders, M. Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int. 15:349-353, 1994.
10. Lance, E.M., Paval, A., Fries, I., Larsen, I., and Patterson, R.L., Jr. Arthrodesis of the ankle joint. A follow-up study. Clin. Orthop. 142:146-158, 1979.
11. Lynch, A.F., Bourne, R.B., and Rorabeck, C.H. The long-term results of ankle arthrodesis. J. Bone Joint Surg. 70B:113-116, 1988.
12. Malarkey, R.F., and Binski, J.C. Ankle arthrodesis with the Calandruccio frame and bimalleolar onlay grafting. Clin. Orthop. 268:44-48, 1991.
13. Mann, R.A., Van Manen, J.W., Wapner, K., and Martin, J. Ankle fusion. Clin. Orthop. 268:49-55, 1991.
14. Marcus, R.E., Balourdas, G.M., and Heiple, K.G. Ankle arthrodesis by chevron fusion with internal fixation and bone-grafting. J. Bone Joint Surg. 65A:833-838, 1983.
15. Marsh, J.L., and Beard, D. Compression ankle arthrodesis utilizing a unilateral external fixation frame. Presented at the 8th Annual Summer Meeting of the American Orthopaedic Foot and Ankle Society, Napa (CA), July 15-19, 1995.
16. Maurer, R.C., Cimino, W.R., Cox, C.V., and Satow, G.K. Transarticular cross-screw fixation. A technique of ankle arthrodesis. Clin. Orthop. 268:56-64, 1991.
17. Mears, D.C., Gordon, R.G., Kann, S.E., and Kann, J.N. Ankle arthrodesis with an anterior tension plate. Clin. Orthop. 268:70-77, 1991.
18. Miller, S.D., Paremain, G.P., and Myerson, M.S. The miniarthrotomy technique of ankle arthrodesis: a cadaver study of operative vascular compromise and early clinical results. Orthopedics, 19:425-430, 1996.
19. Moeckel, B.H., Patterson, B.M., Inglis, A.E., and Sculco, T.P. Ankle arthrodesis. A comparison of internal and external fixation. Clin. Orthop. 268:78-83, 1991.
20. Morgan, C.D., Henke, J.A., Bailey, R.W., and Kaufer, H. Long-term results of tibiotalar arthrodesis. J. Bone Joint Surg. 67A:546-550, 1985.
21. Myerson, M.S., and Quill, G. Ankle arthrodesis. A comparison of an arthroscopic and an open method of treatment. Clin. Orthop. 268:84-95, 1991.
22. Papa, J., Myerson, M., and Girard, P. Salvage, with arthrodesis, in intractable diabetic neuropathic arthropathy of the foot and ankle. J. Bone Joint Surg. 75A:1056-1066, 1993.
23. Russotti, G.M., Johnson, K.A., and Cass, J.R. Tibiotalocalcaneal arthrodesis for arthritis and deformity of the hind part of the foot. J. Bone Joint Surg. 70A:1304-1307, 1988.
24. Scranton, P.E., Jr. Use of internal compression in arthrodesis of the ankle. J. Bone Joint Surg. 67A:550-555, 1985.
25. Scranton, P.E., Jr. An overview of ankle arthrodesis. Clin. Orthop. 268:96-101, 1991.
26. Scranton, P.E., Jr., Fu, F.H., and Brown, T.D. Ankle arthrodesis: a comparative clinical and biomechanical evaluation. Clin. Orthop. 151:234-243, 1980.
27. Sowa, D.T., and Krackow, K.A. Ankle fusion: a new technique of internal fixation using a compression blade plate. Foot Ankle, 9:232-240, 1989.
28. Stuart, M.J., and Morrey, B.F. Arthrodesis of the diabetic neuropathic ankle joint. Clin. Orthop. 253:209-211, 1990.
29. Wang, C.J., Tambakis, A.P., and Fielding, J.W. An evaluation of ankle fusion in children. Clin. Orthop. 98:233-238, 1974.
30. Weltmer, J.B., Jr., Choi, S.H., Shenoy, A., and Schwartsman, V. Wolf blade plate ankle arthrodesis. Clin. Orthop. 268:107-111, 1991.
31. White, A.A. A precision posterior ankle fusion. Clin. Orthop. 98:239-250, 1974.
Figure Legends
Fig. 1. Anteroposterior (A) and lateral (B) radiographs before revision show pseudarthrosis of attempted ankle fusion. Note the broken screw. Anteroposterior (C) and lateral (D) radiographs after revision tibiotalar arthrodesis show successful union.
Fig. 2. This anteroposterior preoperative radiograph (A) demonstrates arthritis and deformity. Anteroposterior (B) and lateral (C) radiographs 9 months after attempted arthrodesis with cannulated screws. The mortise view (D) demonstrates pseudoarthrosis more clearly after screw removal. Lateral (E) and mortise (F) views 9 months after revision demonstrate arthrodesis.
Fig. 3. This man underwent three previous attempts at ankle arthrodesis (A). Note the presence of subtalar arthritis in the ankle mortise view (B). Postoperative anteroposterior (C) and lateral (D) views 4 months after surgery demonstrate arthrodesis.
|
 |
