|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this report is to review our experience with talonavicular joint arthrodesis for the treatment of severe valgus foot deformities in older children and adolescents with cerebral palsy (CP).
The clinical, radiographic, and gait parameters results after talonavicular joint arthrodesis were retrospectively reviewed in 32 patients (59 feet) with valgus deformities of the foot. The surgery was performed as part of multiple simultaneous surgeries for the treatment of gait disorders. The mean age of the patients was 13.9 years (range 9–20 years) and the mean follow-up was 40 months (range 18.3–66.7 months).
The clinical and radiographic measurements improved significantly (P = 0.000). There were no significant changes in gait parameters. Symptoms were relieved in most patients with symptomatic preoperative feet. The most frequent complication was pseudoarthrosis, which occurred in seven feet. We found a high rate of satisfaction of patients (or parents) and most of them recommended the procedure to other patients with the same condition.
Talonavicular joint arthrodesis is a reliable technique that provides both functionally and cosmetically good results with a low rate of complications in the treatment of severe pes planus valgus in older children and adolescents with CP. Careful examination should rule out concomitant ankle valgus deformities. A stable fixation of the arthrodesis is recommended.
Pes planus valgus is reported to occur in 25% of patients with cerebral palsy (CP) . This condition affects 64% of diplegic and quadriplegic children  and is a major cause of instability during stance . This deformity can be considered as a flexible lever arm dysfunction , and, sometimes, can cause pain and/or skin pressure or breakdown in the region of the talar head . Surgical correction of the deformity is then recommended.
In the older child and in the adolescent, pes planus valgus becomes severe, therefore, some authors recommend subtalar arthrodesis, os calcis lengthening [5–7], triple arthrodesis, or some variations of these techniques [8, 9]. Some papers report that, in CP, the results of these techniques have a high failure rate and are not as efficient as in other conditions [10–12]. In our experience, the clinical and radiographic correction obtained with these techniques is often unsatisfactory, obtaining rigid and deformed feet, unable to provide stability during stance . One of the authors (C. A. Turriago) noticed that the longitudinal arch of the spastic pes planus could be adequately restored when reducing and shortening the internal column with talonavicular joint arthrodesis. The mid-foot instability and hind-foot valgus are corrected as well. Although talonavicular joint arthrodesis has been useful in the treatment of acquired flat foot in the adult [14–16], arthritis and arthrosis of the talonavicular joint [16–19], residual deformities of club foot , and sequels of trauma or talus necrosis , there are no clinical reports about its use in the spastic flat foot in CP. We report our experience with this technique.
Between February 2002 and December 2005, as part of multilevel simultaneous surgery for the treatment of gait disorders, 38 ambulatory patients (68 feet) with spastic CP underwent talonavicular joint arthrodesis in order to treat their pes planus valgus deformity. We included 59 feet in 32 patients (26 of them bilateral); six patients were discarded because of incomplete data. Sixteen (30 feet) patients were females and 16 (29 feet) males. The average age was 13.9 years (range 9–20 years). Thirty feet were left and 29 were right. The average follow-up was 3 years and 4 months (range 18.3–66.7 months). The concomitant simultaneous procedures are listed in Table 1. Twenty-one feet had previous surgeries: five os calcis lengthening (Evan’s procedure), 11 subtalar joint arthroereisis, and five posterior tibial tendon lengthening. Thirteen extremities had previous Achilles tendon lengthening.
Twenty-nine patients had spastic CP; 12 of them were diagnosed as quadriplegic, 16 diplegic, and one triplegic. Two patients had mixed CP and one had dystonic CP.
Data were obtained from clinical history, preoperative and postoperative weight-bearing X-ray films, gait parameters from computerized gait analysis (Ariel Dynamics and BTS), and a satisfaction questionnaire answered by patients and/or parents.
The information was analyzed with the SSPS v10 statistical program, applying Wilcoxon’s test of ranges or Fisher’s test.
The talonavicular joint is approached through a 3-cm medial incision over the joint following skin folds. After identification and preservation of the tibialis posterior tendon and saphenous vein, the joint capsule is generously removed and articular surfaces are exposed and resected with an oscillating saw. Care should be taken to remove just enough bone to obtain good cancellous surfaces for fusion so as not to overly shorten the medial column, else the forefoot will go into adductus. Remaining cartilaginous tissue is totally removed with a curette. Special care is required when restoring foot alignment by properly reducing the talonavicular joint and adequately pronating the forefoot. When the correct position is achieved, the arthrodesis is then fixated with a 3-mm percutaneous Steinmann pin or a 4.5 cannulated cortical screw, with the help of an image intensifier. The foot is immobilized in a short leg cast for 6 weeks. When a Steinmann pin is used, removal is performed after 6 weeks. Proper ankle foot orthosis are adapted and gait rehabilitation is started.
The preoperative clinical description of the foot was as follows: planovalgus 23 feet (39%), equino-valgus 23 feet (39%), calcaneo-valgus 13 feet (22%). Postoperatively, eight feet (13.5%) were described as planus valgus. Preoperatively, all feet had midtarsal hypermobility and none had this feature postoperatively (Fig. 1).
The preoperative hindfoot valgus (measured in a static barefoot video) averaged 20.96° (range 12°–34°) and the postoperative hindfoot valgus was 9.5° (range 3°–35°), P = 0.000 (Wilcoxon’s test of ranges). Ankle valgus (≥10°) was diagnosed in six feet and was considered as the cause of the residual hindfoot valgus. This finding was not noticed preoperatively as a contributing factor of the deformity.
Thirty-five feet had no preoperative or postoperative pain. Twenty-four feet had pain before surgery. Nineteen feet improved, four showed no improvement, and one had severe pain postoperatively. This last case had inappropriate reduction of the talonavicular joint and the valgus worsened from 25° to 35° (talocalcaneal X-ray AP view); this foot required surgical revision. One of the feet with no improvement of the pain had a painful pseudoarthrosis, with no correction of the deformity, and also required revision.
Forty-six feet had no preoperative or postoperative skin pressure or breakdown; 13 feet had preoperative evidence of increased pressure over the tarsal head and 12 of them improved after surgery.
Twenty-eight of the 32 patients (87.5%) were satisfied with the surgical correction and four patients (or parents) were partially satisfied and none were unsatisfied. Thirty-one patients (or parents) indicated that they would recommend the surgery to other patients in the same situation. One stated that he was not sure he would recommend it.
Preoperatively, the talocalcaneal angle on the X-ray AP view averaged 42.9° (range 26°–55°). The postoperative angle averaged 21.7° (range 10°–31°), P = 0.000. The preoperative lateral talocalcaneal angle averaged 51.9° (range 12°–90°) and the postoperative averaged 32.2° (range 13°–58°), P = 0.000 (Figs. 2 and and33).
The preoperative AP first metatarsal–talus angle on the AP X-ray view averaged 25.7° (range 14°–40°) and the postoperative angle averaged 5.3° (range −11° to 26°), P = 0.000. The preoperative first metatarsal–talus angle in the lateral X-ray view averaged 22.6° (range 4°–46°) and the postoperative angle averaged 7.5° (range −9° to 25°), P = 0.000 (Fig. 4).
No significant changes were seen in the gait parameters. The preoperative gait speed averaged 0.51 m/s (range 0.11–1.04 m/s) and the postoperative speed averaged 0.49 m/s (range 0.09–0.85 m/s), P = 0.2. The preoperative step length averaged 0.3 m (range 0.04–0.68 m) and the postoperative length averaged 0.32 m (range 0.03–0.78 m), P = 0.3. The preoperative single stance percentage during the gait cycle averaged 28.8 (range 9.6–49.1%) and postoperative, it averaged 26.7% (range 7.6–78%), P = 0.14.
Percutaneous Steinmann pins were used to fix the talonavicular arthrodesis in 35 feet and 4.5 cortical screws in 24 feet.
Pseudoarthrosis: seven feet developed pseudoarthrosis (11.8%) of the talonavicular joints. Six of these cases were fixed with Steinman pins and one with a cortical screw. There was no statistical significance between the occurrence of pseudoarthrosis and the fixation method (P = 0.639, Fisher’s test).
Over- or under-correction: two feet were considered to be under-corrected and one over-corrected. There was no correlation with the fixation method and under-correction (P = 0.3).
Seven feet required revision surgery: five because of pseudoarthroses and two due to insufficient correction.
Surgical correction of pes planus is recommended when it interferes with the function of the foot as a lever arm or support structure, or causes symptoms such as pain or skin pressure or breakdown in the region of the talar head. In the older child and adolescent, spastic flat foot becomes severe and its surgical treatment is difficult; various techniques have been described in order to obtain functional and cosmetically acceptable results [5–8, 22, 23]. Triple arthrodesis is one of the most commonly used techniques for the treatment of this condition. Monson and Gibson  found that the result of triple arthrodesis is not as good in CP as in other diseases, such as polio sequels, cavus deformities, and flat foot secondary to other disorders. Tenuta et al.  found that triple arthrodesis in patients with CP, with a mean follow-up of 17.8 years, had degenerative changes at the ankle joint in 43% of the feet, and that the worst results were correlated to residual deformity, especially pes planovalgus. Stein et al.  found that poor results of triple arthrodesis were correlated to residual mid-foot deformities, especially valgus flat foot deformities. In our experience, too, the result of conventional triple arthrodesis in CP was disappointing .
Although talonavicular joint arthrodesis has been reported to be useful in the treatment of acquired flat foot in the adult [14, 15], arthritis and arthrosis of the talonavicular joint [16–19], residual deformities of club foot , and after trauma or talus necrosis , there are no clinical reports of its use in the spastic flat foot.
In general, the clinical appearance of the foot showed postoperative improvements with respect to both stability and hindfoot valgus, and most patients (or parents) were satisfied with the procedure and recommended it to other patients with the same disorder.
The radiographic evaluation showed improvement in the relationship between the talus and the calcaneus, and especially of the talo-first metatarsal angle on the AP X-ray view.
The effects on gait parameters showed no statistical significance and were difficult to evaluate because of the association with other simultaneous procedures. Many of these patients preoperatively had a severe compromise and crouch gait. Even if gait parameters do not improve, it does not mean that function does not improve (e.g., reverting crouch gait many times does not correlate well with gait parameters, but the prognosis and function improves).
Complications such as pseudoarthrosis and the loss of correction had no statistical correlation with the fixation method (Steinmann pins or cortical screws). Although not all of the patients with postoperative pseudoarthrosis referred pain, most of them did, and some showed loss of correction. We encourage the use of a secure fixation with one or two 4.5 cannulated screws with image intensifier control.
In six feet (10%), residual valgus could be attributed to pre-operative ankle valgus, which was missed on preoperative evaluation. As such, we feel strongly that the preoperative evaluation of the spastic flat foot should carefully establish the origin of the valgus (rule out more than 10° ankle valgus) and the origin of the mid-foot instability, since it can be distal to the talonavicular joint. If an ankle valgus deformity is diagnosed, it should be simultaneously corrected. Also, contractures of the peroneus and gastrocnemius muscles should be simultaneously treated.
Our results suggest that the goal of obtaining a stable lever arm with simultaneous recovery of the foot’s arch, correction of the midtarsal instability, and hindfoot alignment can be obtained with talonavicular arthrodesis in severe pes planus in older children and adolescents with CP. We found no evidence of recurrence of the deformity or the development of instability on distal articulations, but our average follow-up was only 40 months, so our findings should be considered as preliminary results. A longer follow-up is also desirable, since triple and isolated arthrodesis of the foot have been associated with long-term arthrosis of the nearby joints [11, 19, 24–26].
The authors thank Dr. Fernando Ortiz MD for his help in the statistical analysis of this report and Dr James Gage MD and Dr. Pablo Rosselli MD for their valuable comments.