Complete Subtalar Release in Resistant Clubfeet:
A Critical Analysis of Results in 146 Cases

Nikolay J.Rumyantsev, M.D., and Victor E. Ezrohi, M.D.

PDF Version

Study conducted at Regional Children's Hospital, St. Petersburg, Russia

Summary: A series of 101 patients (146 feet) with resistant clubfoot corrected by complete subtalar release is presented. A detailed rating system was used to evaluate the results. Mini­mum follow-up was 2 years. Fourteen feet (9.6%) had under­gone additional surgical procedures at the time of review, and 101 feet (69%) had excellent or good functional ratings at that time. The mean total ankle motion was 34.2° (range, 8-56°).

The better results occurred in feet without previous surgery. Final ankle range of motion was increased by using a special flexed-knee cast with reserve space above the foot. Longer follow-up is needed to determine the optimal age for surgery.

Key Words: Clubfoot—Complete subtalar release—Rating system.

There are many treatment regimens for clubfoot. Some authors recommend manipulations with minimal multi-staged surgery [2], whereas others recommend neonatal corrective surgery [7]. It is impossible to compare ob­jectively various treatment programs, because their au­thors use different criteria to evaluate results. Only a few investigators described long-term results of complete subtalar release [1,3,5,9]. The number of feet in these published series ranged from 17 to 55.

A detailed functional rating system for clubfoot was used in this study. Positive and negative points were assigned, with major complications scoring most nega­tively. Priority in evaluation was given to clinical crite­ria. Range of ankle motion was documented by radio­graphs. The purpose of this study was the analysis of clubfeet corrected by complete subtalar release. The ad­vantages and disadvantages of this surgical approach are presented.

Materials and methods

At Regional Children's Hospital (St. Petersburg, Rus­sia), complete subtalar release [4] was first performed in 1989. Between that time and May 1993, 129 patients with 189 clubfeet underwent this procedure. Of these, 14 patients with 22 teratologic or neuromuscular clubfeet were excluded from review, leaving a group of 115 patients with 167 clubfeet. Of this group, 14 patients (21 feet) were lost to follow-up. This article presents results in 146 patients with idiopathic clubfeet who returned for examination.

All feet were initially treated by serial long-leg plaster of paris casts for a minimum of 5 months. Nonsurgical treatment was successful in 40% of cases (165 feet); 248 feet failed cast treatment and underwent surgery. The indications for complete subtalar release were the pres­ence of hindfoot varus after previous treatment and un-corrected horizontal calcaneal rotation beneath the talus, which causes toeing-in.

The Cincinnati incision and soft-tissue releases were performed as described by McKay [4] and Simons [8] with the variations listed subsequently. In feet with rigid forefoot adduction after correction of the calcaneocuboid and talonavicular joints in which the calcaneal-second metatarsal angle was >30°, capsulotomies of the navicu-lar-first cuneiform and first cuneiform-first metatarsal joints were added. The tendo Achilles was lengthened in the coronal plane. The interosseous talocalcaneal liga­ment was preserved when calcaneal rotation could be easily corrected. At wound closure, the lower skin mar­gin was displaced medially in relation to the upper mar­gin. One talonavicular and two talocalcaneal pins were inserted in all feet. The talonavicular pin was directed from posterior to anterior. One calcaneocuboid pin was used in cases with marked medial displacement of the cuboid. In feet with marked forefoot adduction, an addi­tional pin was directed from the calcaneus into the first metatarsal to maintain correction of the forefoot. We supervised or performed all procedures.

At the end of the procedure, a long-leg, flexed-knee plaster cast was applied. The foot was positioned in external rotation (thigh-foot angle, 10-15°) and neutral dorsiflexion or slight equinus. Extra space was created above the foot to provide early ankle motion. The cast application (with reserve space) was performed as fol­lows (Fig.I): a special elastic pad was placed on the dorsal aspect of the foot before the cast application and fixed with a cotton bandage. The plaster cast was ap­plied. After the cast was set, the elastic pad was removed, leaving a reserve space on the dorsal aspect of the foot' The cast was changed after 24 h and again 2 weeks later. Foot displacement into dorsiflexion occurred spontane­ously and also with manipulations. Manipulations into dorsiflexion were started after the swelling decreased, on about the fifth to seventh day after surgery. A thumb was positioned below the foot to provide passive dorsiflex­ion. This manipulation was done by the surgeon or parent several times a day. The patient also could actively dor-siflex the ankle.

FIG. 1.Cast application with reserve space above foot. A, B: A special elastic pad is placed on the dorsal aspect of the foot (frontal and lateral views). C: After the hardening of the cast, the elastic pad is withdrawn, leaving a reserve

The foot was gradually brought into neutral position. The next cast change took place 4 weeks after surgery. At this point, pins and sutures were removed, and careful manipulation of the foot (especially into plantarflexion) was performed by the surgeon. Another long-leg, flexed-knee cast with reserve space was applied for 3 more weeks. Finally, the foot was positioned in a short-leg cast for 4-6 weeks.

A detailed rating system for functional results was used, with 150 points indicating a normal foot (Table 1).

Table 1. Rating system for clubfoot surgery

Criteria	Level	Points
Gait abnormality	Absent Only while running Constant	10 0 -10
Shoe type	Regular Regular+ orthopedic Orthopedic only	10 5 0
Pain	Never With heavy activity With routine activity	10 5 0
Ankle dorsiflexion (passive motion)	15-25 over 90 5-14 over 90 0-A over 90 <90	20 10 0 -20
Position of heel when standing	0-5 valgus 6-10valgus 11-20 valgus >20 valgus 0-5 varus >5 varus	10 5 -5 -25 -5 -25
Appearance of forefoot	Neutral <5 ADD/ABD 5-15 ADD/ABD >15 ADD/AB	10 5 0 -10
Ankle motion by radiograph	>4 31^t0 21-30 11-20 <11	30 20 10 0 -20
Calf atrophy	Absent Mild Severe	5 3 0
Cavus/planus	Absent Present Rocker-bottom or dorsal navicular subluxation	10 0 -10
Foot-knee realignmenti	External rotation 0-15 Neutral Internal rotation orexternal rotation >15	10 0 -10
Flexion of great toe	Present Absent	5 0
Strength of tricep surae	Weight supported on toes, one foot only Weight supported on toes, both feet Weight not supported on toes	10 5 0
Functional abilities	Can heel- and toe-walk Can heel- or toe-walk Cannot heel- and toe-walk	10 5 0
ADD, adduction; ABD, abduction. General foot position in relation to the knee joint and lower limb as a whole was assessed according to presence or absence of toe-in gait, foot-progression angle, and thigh-foot angle were also evaluated.

Marked residual or secondary deformities (e.g., heel val­gus >20°) were assigned negative points. The results were assessed, according to the scores, as follows: ex­cellent (Fig. 2), 111-150 points; good, 71-110 points;

fair, 31-70 points; and poor, <31 points. Foot-progression axis, thigh-foot angle, and transmalleolar

angle as described by Staheli et al. [10] also were docu­mented.

Our radiographic assessment included an anteroposte-rior (AP) and two lateral radiographs, as described by Simons [9]. Critical analysis was emphasized on radio­graphs of feet with residual or secondary deformities. Calcaneocuboid evaluation was made according to Thometz and Simons [12].

Results

Of the 101 patients in this study, 54 (53.4%) were boys, and 47 (46.6%) were girls. Of the 146 clubfeet, 70 (48%) were right and 76 (52%) left feet. Forty-three patients (42.6%) had bilateral clubfeet.

Five patients had undergone six prior operative pro­cedures (posterior release in two feet, posteromedial re­lease in three feet, and medial release in one foot). The average age at operation was 13.2 months (range, 5-62). Average postoperative follow-up was 34.1 months (range, 24-62; SD, 8.8).

Seventy feet (47.9%) had external foot-knee rotation, 52 feet (35.6%) had central (neutral) rotation, and 24 feet (16.5%) had internal rotation.

Average range of ankle motion (by radiographs) was 34.2° (range, 8-56°).

At follow-up, equinus was present in 3.4% of feet, hindfoot varus in 2.7%, and heel valgus >10° in 8.9%. Additional findings were calcaneal gait in 10.9% of feet, cavus in 21.9%, forefoot supination in 12.3%, and re­sidual forefoot adduction (averaging 11.2°) in 12.3%.

Calcaneocuboid relationships before surgery were classified, according to Thometz and Simons, as follows:

grade 0, 39 feet; grade I, 93 feet; and grade 14 feet. Postoperatively, the classifications were as follows:

grade 0, 117 feet; grade I, 27 feet; and grade II, 2 feet. Feet with minimal Calcaneocuboid malalignment had a good clinical appearance, and an insignificant radio-graphic talonavicular overcorrection was frequently seen.

The mean rating at follow-up was 102.1 points (range, -22 to 148; SD, 28.5). The results were assessed as excellent in 22.2% of feet, good in 46.8%, fair in 27.3%, and poor in 3.7%. To compare our results with those of other authors, we used the Magone et al. rating system [3]. Our average rating using this system was 83.3 points. Complete subtalar release in the series of Magone et al. yielded an average rating of 78.8 points.

COMPLICATIONS AND SUBSEQUENT SURGERY (Table 2)

Serious wound problems occurred in three feet. Wound dehiscence developed on the posteromedial as­pect of the foot medial to the Achilles tendon. Usually it was seen on postoperative days 7-10. The distance be­tween skin margins did not exceed 12 mm. These cases were managed as follows: the talonavicular pin was im­mediately removed. A long-leg, flexed-knee cast with the foot in external rotation was applied with the "win­dow" over the wound. Antibiotic therapy and laser radiation of the wound were used. Laser treatment was given for 5 min a day for 10-14 days. Technical speci­fications for our laser therapeutic apparatus "Atoll" are as follows: He-Ne laser, radiation wavelength, 0.6328 (Jim; average radiation power, 20 mW; density of stream, 0.002 W/mm (square). During the healing period (3-8 weeks), the cast was not removed. At follow-up, these wounds were healed without scarring. We believe that these complications were the result not of infection, but of lack of vascularity. All three feet were rated as good at final follow-up.

FIG. 2.A 5-year-old boy with excellent result of surgical treatment (rating, 138 points). A: External foot rotation in relation to the knee joint is 10°. B: Lateral view with normal appearance of longitudinal arch (absence of cavus or planus deformities). C: Position of heel when standing is 0°. D: Weight supported on toes, one foot only. E: Neutral position of the forefoot. F: Ankle motion by radiographs is 32°.

Residual equinus was present in five feet. In these cases, postoperative casting had failed to bring the foot into dorsiflexion. AP radiographs showed normal bony relationships, but lateral dorsiflexion radiographs dem­onstrated decreased talocalcaneal and tibiocalcaneal angles. All these feet underwent posterior release com­bined with Ilizarov distraction. The Ilizarov apparatus provided gradual correction of resistant equinus defor­mity over 2-3 weeks. The corrected position of the foot was maintained in the apparatus for 6-8 weeks. Ages at surgery were 19, 26, 29, 37, and 54 months. Three of these feet had good and two had fair results. The mean range of ankle motion in this group was 17.6%, with mean dorsiflexion of 9.3°.

Clinically marked cavus was associated with dorsal na-vicular subluxation in three feet. Surgical correction in these cases included extensive talonavicular and navicular-cuneiform capsulotomies, plantar dissection, and gradual navicular depression using the olive-wire technique with the Ilizarov distractor. This was combined with elongation of the medial column of the foot. Good results were ob­tained in all three feet (mean rating, 99.7 points).

Calcaneal gait caused by overlengthening of the Achilles tendon occurred in 16 feet. The mean foot dor­siflexion in this group was 33°, and patients were not able to toe-walk. None of these feet has required addi­tional surgery to date for this complication.

Table 2. Subsequent surgery: clinical summary

Ten feet (6.8%) had marked forefoot adduction >10°. This complication caused an in-toeing gait even when correction of subtalar calcaneal rotation was achieved. Of these feet, seven had significant medial displacement at the calcaneocuboid joint, as well as medial navicular translation on the talar head from one quarter to one half. In this group, medial release with extensive calcaneocu­boid capsulotomy was performed through a lateral inci­sion, and the Ilizarov distractor was applied. Two feet underwent simple medial release, and one foot under­went medial release together with application of the Ilizarov apparatus. At final follow-up, seven feet in this group showed good results, and three feet showed fair results.

Valgus angulation of the hind part of the foot >10° occurred in 13 feet (8.9%). All these feet had normal talocalcaneal angles on the AP radiographs, but talocal-caneal divergence was increased to +3 and +4 (according to Simons' scheme). Eight of these feet had lateral na­vicular subluxation from one quarter to one half on the talar head. The other five feet showed no significant lateral navicular translation. In our opinion, the valgus heel is a result of lateral displacement of the calcaneus beneath the talus. The total release of all talocalcaneal ligaments (including the interosseous ligament) and in­accurate talocalcaneal pinning cause this complication. Five of seven children with a valgus heel had hyperlaxity of their joints. All these feet may require a Grice proce­dure in the future.

Discussion

Many have analyzed the results of clubfoot correction, but authors use different criteria to evaluate results. It is therefore difficult to compare objectively various treat­ment programs and surgical procedures.

Measurement of total ankle range of motion is impor­tant in the assessment of corrected feet. Only a few au­thors [3,5,6,9] verified these data by radiographs. Quan­titative evaluation differs in various investigations. Si­mons [9] indicated that ankle range of motion of at least 25° must be present for a satisfactory result. McKay [5] defined range of motion <35° as an unsatisfactory result. Magone et al. [3] assigned decreasing points for motion from 40° to 0. Stauffer et al. [11] reported that during the stance phase of gait, the average total ankle range of motion was 24.4°, with average dorsiflexion of 10.2° and average plantarflexion of 14.2°. Therefore a normal foot must be plantigrade and should dorsiflex. It is crucial to evaluate not only total range of motion but its compo­nents as well, as emphasized by McKay [5] and Brougham and Nicol [1].

The mean ankle motion in our series was 34.2°. We believe that postoperative manipulation of the foot at the time of cast change produces increased ankle range of motion. In our opinion, positioning the foot in an above-the-knee plaster cast in mild equinus at the end of the procedure produces a plantigrade foot in the future. In our experience, casting in maximum dorsiflexion causes postoperative scarring of the anterior ankle capsule, re­sulting in anterior ankle contracture. A cast with reserve space above the foot also increases the total ankle range of motion.

From our point of view, clubfoot surgery is based on (a) soft-tissue releases of the bones, (b) correction of bony malposition, (c) accurate pin fixation, and (d) res­toration of muscle balance if necessary.

The circumferential Cincinnati incision provides the surgeon with the best exposure for soft-tissue release and bony repositioning.

Like Porter [6], we believe that failure of clubfoot correction rests more in the surgeon's hand than in the child's foot. For example, incomplete subtalar release does not allow the surgeon to correct calcaneal rotation beneath the talus. The normalization of the talonavicular

and calcaneocuboid joints requires a complete release of these joints. Otherwise, the forefoot adduction will per­sist. Full bone repositioning is possible only with bilat­eral or circumferential surgical approaches.

The type of bony repositioning depends on the sur­geon's clubfoot philosophy. Marked medial displace­ment of the cuboid required surgery in 14 feet in our series. Posteromedial release would not be successful in treatment of these feet.

In the cases of accurate pinning, the previous bony incongruity resolves because of the remodeling potential of bone and cartilage. Failures in bone fixation lead to overcorrection or secondary foot deformities. For ex­ample, inaccurate talonavicular pinning may be mani­fested as a cavus foot with dorsal navicular subluxation or as a planovalgus foot with lateral navicular translation. Inaccurate talocalcaneal pinning when the calcaneus is displaced laterally beneath the talus causes severe valgus deformity.

All the components of clubfoot surgery are intercon­nected. If the joint releases are incomplete, the surgeon's options for bone repositioning are limited. On the other hand, incomplete joint release may also prevent overcor­rection.

We believe that overcorrection and secondary multi-planar foot deformities are the main problems with ex­tensive clubfoot surgery. Therefore, treatment of these complications is of special interest for clubfoot surgeons.

Acknowledgment:We thank Drs. L. Suvorova and Carol Mowery for help in translation and manuscript preparation.

References

• 1. Brougham DI, Nicol RO. Use of the Cincinnati incision in con­genital talipes equinovarus. J Pediatr Orthop 1988;8:696-8.
• 2. Laaveg SJ, Ponseti IV. Long-term results of treatment of congen­ital club foot. ] Bone Joint Surg [Am] 1980;62:23-81.
• 3.Magone JB, Torch MA, Clark RN, Kean JR. Comparative review of surgical treatment of the idiopathic clubfoot by three different procedures at Columbus Children's Hospital. J Pediatr Orthop 1989;9:49-58.
• 4.McKay DW. New concept of and approach to clubfoot treatment: section II: correction of the clubfoot. J Pediatr Orthop 1983;3:10-21.
• 5.McKay DW. New concept of and approach to clubfoot treatment: section III: evaluation and results. J Pediatr Orthop1983;3:141-8.
• 6.Porter RW. Congenital talipes equinovarus: II: a staged method of surgical management. J Bone Joint Surg [Br\1987;69:826-31.
• 7.Ryoppi S, Saranen H. Neonatal operative treatment of clubfoot. J Bone Joint Surg [Br] 1983;65:320-5.
• 8.Simons GW. Complete subtalar release in club feet: part I: a pre­liminary report. J Bone Joint Surg [Am] 1985;67:1044-55.
• 9.Simons GW. Complete subtalar release in club feet: part II: com­parison with less extensive procedures. J Bone Joint Surg [Am] 1985;67:1056-65.
• 10.Staheli LT, Corbett M, Wyss C, King H. Lower extremity rota­tional problems in children. J Bone Joint Surg [Am] 1985;67:39-47.
• 11.Stauffer RN, Chao EYS, Brewster RC. Force and motion analysis of the normal, diseased and prosthetic ankle joint. Clin Orthop 1977;127:189-96.
• 12.Thometz JG, Simons GW. Deformity of the calcaneocuboid joint in patients who