This study was approved by our institutional review board, and the parents of all participants provided informed consent. We retrospectively reviewed the cases of fifteen consecutive patients (twenty-five feet) with vertical talus associated with neuromuscular and/or genetic disorders treated by a single surgeon (M.B.D.) at St. Louis Children’s Hospital (eight patients) and St. Louis Shriners Hospital for Children (seven patients) between 2001 and 2008. Ten patients had bilateral vertical talus deformities. All five patients with a unilateral vertical talus deformity had this deformity on the left side; three of these five patients had vertical talus on one side and clubfoot on the other. Seven patients were male. Four patients (all of whom had rigid vertical talus deformity at the time of presentation to our clinics) had received treatment prior to referral; the prior treatment consisted of serial casting in two patients and extensive soft-tissue releases in two. Five of the fifteen patients had arthrogryposis (distal arthrogryposis in three and multiple pterygium syndrome in two), six had other genetic abnormalities (diagnosed with chromosomal microarray analysis), and four had myelomeningocele (Table I).
To be eligible for inclusion in the study, a patient had to have a diagnosis of vertical talus confirmed by (1) a lateral radiograph with the foot in maximum plantar flexion, demonstrating dorsal translation of the forefoot on the hindfoot (Fig. 1-A) caused by fixed dorsal dislocation of the navicular on the head of the talus, and (2) a lateral radiograph with the foot in maximum dorsiflexion, demonstrating a persistently decreased tibiocalcaneal angle indicating a fixed equinus contracture of the hindfoot (Fig. 1-B). In addition, patients had to have a minimum of two years of follow-up after treatment, and radiographs made before and after treatment had to be available. Patients with an isolated case of vertical talus were excluded, and all patients were therefore seen by a clinical geneticist and/or a pediatric neurologist with a special interest in musculoskeletal problems to ensure accurate phenotype identification.
A detailed family history was obtained to assess the presence of vertical talus and other orthopaedic anomalies in family members. Pregnancy and delivery histories were obtained for all patients, and any associated systemic anomalies (including orthopaedic, cardiovascular, pulmonary, genitourinary, ophthalmologic, and auditory) were documented. All patients were clinically assessed by the senior author (M.B.D.) at the time of presentation as well as at the time of the latest follow-up. Passive plantar flexion and dorsiflexion of the ankle, subtalar motion, and varus-valgus heel alignment (during weight-bearing) were measured at the time of the latest follow-up by a single examiner with a handheld goniometer. To minimize error in the measurement of ankle dorsiflexion and plantar flexion, the examiner’s index finger was placed around the back of the heel to generate motion (rather than using the forefoot or midfoot as the fulcrum). Directly moving the calcaneus with the index finger ensured that the motion measured was exclusively from the ankle joint and did not include spurious movement through the midfoot. Recurrence of deformity was documented, including the age at the time of recurrence and the additional treatment undertaken.
In addition to the lateral radiographs in maximum plantar flexion and dorsiflexion made at the time of initial presentation, anteroposterior and neutral lateral radiographs of the feet were made before the beginning of treatment, immediately postoperatively, and at the time of the latest follow-up; the latter radiographs were made in a standing position. The severity of the vertical talus deformity was assessed radiographically at the time of presentation as either Coleman type I or type II10. A type-I deformity is isolated to dislocation of the talonavicular joint, whereas a type-II deformity also has dislocation or extreme subluxation of the calcaneocuboid joint (i.e., the long axis of the calcaneus lies plantar to the long axis of the cuboid) as determined on a lateral radiograph made with the foot in plantar flexion. The talocalcaneal and talar axis-first metatarsal angles were measured on the anteroposterior radiograph, and the talocalcaneal, tibiocalcaneal, and talar axis-first metatarsal base angles were measured on the lateral radiograph (Figs. 2-A and 2-B). The lateral talar axis-first metatarsal base angle in maximum plantar flexion was used to differentiate a vertical from an oblique talus; an angle of >35° in maximum plantar flexion was considered to be diagnostic of vertical talus, as described by Hamanishi29. To reduce measurement error, all angles were measured on two separate occasions eight weeks apart by the same investigator. To minimize errors due to foot positioning on the radiographs made at the initial evaluation, a standard and reproducible foot position (simulated weight-bearing) was utilized as described by Becker-Andersen and Reimann15. The mean angles measured on the anteroposterior and lateral radiographs made before treatment, immediately after pin removal, and at the time of the latest follow-up were compared. In addition, the angles were compared with normal values for individuals of the same age, as described by Vanderwilde et al.30.
A ten-point scale designed by Adelaar et al.31 was used to evaluate the outcome at the time of the latest follow-up on the basis of both the clinical appearance of the foot and radiographically measured correction. The clinical appearance of the foot is assigned a maximum of six points on this scale, and radiographic measurements are assigned a maximum of four points. One point is subtracted from the maximum score of ten points for each abnormality that is noted either clinically or radiographically (see Appendix). A score of ten points is considered excellent; a score of seven to nine points, good; a score of four to six points, fair; and a score of zero to three points, poor. An assessment of walking status was also performed at the time of the latest follow-up as described by Hoffer et al.32.
Treatment Technique
The technique followed that described by Dobbs et al.3,22 with a few modifications made because of the rigidity of the vertical talus deformity in this patient population. Manipulation of the vertical talus is performed for several minutes at each visit before cast application, with special emphasis on providing a gentle laterally and dorsally directed pressure on the plantar medial aspect of the head of the talus as the foot is brought into progressively greater plantar flexion and adduction (Fig. 3). The dorsally directed pressure reduces the vertical talus deformity seen on a lateral foot radiograph, and the lateral pressure (which represents a modification of the originally described technique) is important to reduce the increased talocalcaneal angle seen on an anteroposterior radiograph. Serial casting is performed on a weekly basis until either all deformities except the hindfoot equinus are fully corrected or progress toward obtaining correction ceases. In either case, surgery is indicated at that time, and a medial incision is made over the talonavicular joint. No attempt is made to correct the hindfoot equinus with preoperative casting.
If the vertical talus was reduced with casting alone (and a talonavicular capsulotomy is therefore not necessary), the medial skin incision is utilized to ensure accurate placement of the 0.062-in (0.16-cm) Kirschner wire by allowing direct visualization of the starting point for the wire on the navicular or the medial cuneiform as well as palpation of both the talus and the navicular as the wire is driven across the talonavicular joint. Such visualization and palpation are necessary in the young patient in whom the navicular is not yet ossified, making percutaneous wire placement under fluoroscopy difficult. The wire is cut and buried underneath the skin for later removal.
If the vertical talus has been only partially reduced with casting, the medial incision is used to perform a limited capsulotomy of the talonavicular joint and the anterior aspect of the subtalar joint to complete the talonavicular reduction. After open reduction, a double-ended 0.062-in (0.16-cm) Kirschner wire is placed in a retrograde manner across the talonavicular joint and is driven across the talus and out the skin on the posterolateral aspect of the hindfoot. The use of a double-ended Kirschner wire is another modification of the originally described technique. The wire is used as a joystick to ensure that the talus is corrected in both the anteroposterior and the lateral plane. Once reduction is achieved, the wire is driven in an antegrade manner across the talonavicular joint. The wire is cut and buried underneath the skin.
Once the talonavicular joint is stabilized, a percutaneous tenotomy of the Achilles tendon is performed to correct residual hindfoot equinus. The foot is placed in a plaster cast with both the ankle and the forefoot in a neutral position. The cast is changed at two weeks postoperatively in the operating room under fluoroscopy to ensure that the wire has not migrated and to place the foot in 10° of ankle dorsiflexion to ensure correction of the hindfoot equinus while maintaining the forefoot in a neutral position. The patient is also measured for a brace (consisting of two shoes connected to a bar, identical to that used for clubfoot management) prior to reapplying the cast. The brace is then ready to use once the final cast is removed. The wire is removed in the operating room under general anesthesia at six weeks after the index procedure.
The remaining two modifications to the originally described technique involve the postoperative period and consist of use of the bar brace and stretching exercises designed to maintain ankle and foot mobility. The bar brace is the same as that used in the Ponseti method of clubfoot treatment, but the shoes are positioned to point straight ahead rather than in external rotation4. This allows some stretching of the peroneal muscles and is designed to decrease the risk of relapse. The parents are instructed to use the brace full-time (twenty-three hours a day) for two months and then to decrease its use over the course of the next six months. The time in the brace is decreased by several hours every two months until it is worn only when the child is sleeping. The brace is worn during sleep until the child turns two years old. When the child begins standing and/or walking, he or she is fitted with ankle-foot orthoses (AFOs) as needed for support (depending on the level of motor function). Parents are also taught how to perform ankle motion exercises and forefoot exercises that emphasize ankle plantar flexion and forefoot adduction. These exercises are to be performed three to four times a day at the time of diaper changes. Follow-up visits are scheduled one month after initiating use of the bar brace, two months later (at which time the reduction in brace wear is begun), and then at three-month intervals until the age of two years, at six-month intervals for two additional years, and yearly until skeletal maturity. A nurse educator instructs the parents at each visit on how to properly perform the stretching exercises and also telephones the parents during the first week of brace use to identify any problems that may be occurring.
Statistical Analysis
Repeated-measures analysis of variance (ANOVA) was used to compare various radiographic measures before treatment, immediately after treatment, and at the time of the latest follow-up. The mean of the two measurements made at each time point was used in the analysis. If the p value for the overall ANOVA model was significant (<0.05), each pair of time points was compared with use of the Tukey adjusted-least-square means method, and the three resulting p values were calculated. The data analysis was performed with use of SAS software (SAS System for Linux version 9.2; SAS Institute, Cary, North Carolina).
For each angle, the agreement between the initial measurements and the repeat measurements (made by the same examiner eight weeks later) was assessed by two methods. First, the confidence interval for the mean difference between the measurements was calculated, and a paired t test was used to verify that the two sets of measurements were not significantly different. Second, the intraclass correlation coefficient (ICC) and the associated confidence interval were calculated. An unpaired t test was also used to compare radiographic measurements at the time of the latest follow-up with published normative radiographic values for children of the same age, and MedCalc software (Mariakerke, Belgium) was used to calculate p values for the comparisons between the mean measured values and the normative values.
Source of Funding
There was no external source of funding for this study.
Congenital vertical talus is a rare condition that occurs with equal frequency as an isolated deformity5-7,33 or in association with a variety of neuromuscular and/or genetic disorders1,34-36. Regardless of the underlying cause of the vertical talus, treatment has historically involved extensive surgery, with reported complications including amputation37, wound necrosis2, talar necrosis5,26, undercorrection of the deformity10, stiffness of the ankle and the subtalar joint27, pseudarthrosis16, and the need for multiple operative procedures such as subtalar and triple arthrodeses28. Our group recently described a less invasive technique that primarily involves serial manipulation, casting, and minimal surgery to correct isolated vertical talus3,4,22. The success of our technique was subsequently duplicated by other groups in several other series of patients with isolated vertical talus23,24.
To our knowledge, the present study is the first to examine the use of this technique in patients with extremely rigid vertical talus associated with neuromuscular and/or genetic syndromes. The technique was successful in achieving initial correction of the vertical talus deformity in all patients in this population. Achieving initial correction of rigid vertical talus did require some modification of the original technique, as was the case with use of the Ponseti method for rigid clubfoot38-41.
The percentage of patients who achieved full correction of the talonavicular joint with casting alone was smaller in the present study than in the previous groups of patients with isolated vertical talus3,23,24, and many patients required selective release of the talonavicular joint and the anterior aspect of the subtalar joint. Additional modifications in the original technique were made. A dorsally and laterally directed force, rather than solely a dorsally directed force, was exerted on the talus during manipulation. A double-ended Kirschner wire was used as a joystick to help reduce any residual deformity. The remaining modifications to the treatment method involved the post-casting phase. A bar brace with attached shoes pointing straight ahead was utilized until the child turned two years of age to prevent relapse. The brace was worn full-time for two months, and the time in the brace was then reduced by several hours a day every two months over the following six months until it was worn only during sleeping. Many patients also used AFOs while awake to provide support in stance and walking (depending on their overall muscle strength and support needs). Lastly, stretching exercises performed by the parents helped to maintain mobility in the ankle and the foot.
Three patients in this series experienced relapse after initial successful correction. All three of these patients had Coleman type-II deformities on initial presentation, indicating extreme subluxation or dislocation of the calcaneocuboid joint. All three were treated successfully with repeat casting followed by open reduction and pin fixation of both the calcaneocuboid and the talonavicular joint. Although the four other patients with an initial Coleman type-II deformity did not experience relapse, the presence of a type-II deformity should alert the orthopaedic surgeon that the patient may be at increased risk for relapse. The initial treatment in such patients needs to achieve reduction of both the talonavicular and the calcaneocuboid joint.
The principle of avoiding extensive soft-tissue release because of the amount of scarring, stiffness, and pain that can result42 has been well demonstrated in the treatment of clubfoot with both the Ponseti method25,43 and the French method of physiotherapy, taping, and splinting44,45. Both isolated clubfeet and clubfeet associated with neuromuscular and syndromic conditions have been successfully treated with these techniques38-41,46. The principle of avoiding extensive soft-tissue release and its associated complications has also proven successful in the short term for the treatment of isolated vertical talus3,23,24 to provide a more supple foot. Our present report indicates that this principle can also be used to successfully treat vertical talus associated with neuromuscular and syndromic conditions.