Correcting Congenital Talipes Equinovarus in Children Using Three Different Corrective Methods

Abstract Equinus, varus, cavus, and adduction are typical signs of congenital talipes equinovarus (CTEV). Forefoot adduction remains a difficulty from using previous corrective methods. This study aims to develop a corrective method to reduce the severity of forefoot adduction of CTEV children with moderate deformities during their walking age. The devised method was compared with 2 other common corrective methods to evaluate its effectiveness. A Dennis Brown (DB) splint, DB splint with orthopedic shoes (OS), and forefoot abduct shoes (FAS) with OS were, respectively, applied to 15, 20, and 18 CTEV children with moderate deformities who were scored at their first visit according to the Diméglio classification. The mean follow-up was 44 months and the orthoses were changed as the children grew. A 3D scanner and a high-resolution pedobarograph were used to record morphological characteristics and plantar pressure distribution. One-way MAVONA analysis was used to compare the bimalleolar angle, bean–shape ratio, and pressure ratios in each study group. There were significant differences in the FAS+OS group compared to the DB and DB+OS groups (P < 0.05) for most measurements. The most salient differences were as follows: the FAS+OS group had a significantly greater bimalleolar angle (P < 0.05) and lower bean–shape ratio (P < 0.01) than the other groups; the DB+OS and FAS+OS groups had higher heel/forefoot and heel/LMF ratios (P < 0.01 and P < 0.001) than the DB group. FAS are critical for correcting improper forefoot adduction and OS are important for the correction of equinus and varus in moderately afflicted CTEV children. This study suggests that the use of FAS+OS may improve treatment outcomes for moderate CTEV children who do not show signs of serious torsional deformity.


INTRODUCTION
C ongenital talipes equinovarus (CTEV), or clubfoot, is a common foot deformity that involves a complex threedimensional musculoskeletal abnormality. 1 The deformity has 4 main components: equinus, varus, cavus, and adduction. 1,2 If the deformity is not corrected promptly, the ambulatory ability of children will be seriously affected. Nonoperative treatments are typically considered the first choice for treating CTEV in young children. 3 During the prewalking period, the Ponseti method is usually regarded as the standard initial treatment for CTEV. 1,4-6 For short-term effect of the Ponseti treatment, corrective bracing is used following initial correction. 7 According to published reports, 8 the highest rate of recurrence occurs between 1.5 and 4 years of age. Therefore, correction is still needed even for children with CTEV who are beginning to walk.
Foot abduction braces are typically used for continuous correction. 7 Ponseti and Smoley 9 reported an early analysis of a Denis Browne (DB) splint, which is a commonly used corrective device. This brace consists of open-toed, high-top, straightlace shoes attached in external rotation to a bar, 10 and is used to hold the affected foot at approximately 708 and the unaffected foot at approximately 408 of external rotation. 7 DB splints must be worn every night for 2-4 years. 7,11,12 Some studies concluded that the DB splint could correct equinus and varus, but residual adduction of the foot would still exist. 3,13 Furthermore, given that the shoes are attached to a bar, there is a high rate of noncompliance due to difficulties with use for years on end. 4 Orthopedic shoes (OS) are considered more convenient than DB splints and can be used for walking. 14 Moreover, weight-bearing is also important for effectively correcting CTEV. OS are custom-made shoes with inserts that are molded to the shape of the hind foot so as to hold it firmly. OS have been reported with positive results for correcting equinus and varus in weight-bearing correction. 15,16 However, considering that OS are usually set in a neutral alignment and fail to provide abduction correction, they could not be used to stretch medial structures, 17 and residual adduction is also present after treatment. 18 Reiman 19 proposed a dynamic splint to correct adduction. An elastic cord on the splint's lateral side acts to support the cuboid and exert the necessary counter-pressure for abduction of the forefoot. However, considering its complexity and difficulty with use, the splint can only be used at night, and is thus not commonly used in clinical practice.
This paper introduces a corrective method that consists of the daytime and nighttime use of orthoses. Orthopedic shoes are used during the daytime, while forefoot abduction shoes are used at night. This paper will also explore the outcome of this new corrective method in comparison to 2 other common corrective methods for children with CTEV.

METHOD Corrective Treatment Methods
DB splints comprise a pair of boots connected by a rigid bar, as shown in Figure 1A. A kind of forefoot abduction shoe (FAS) specially designed to correct adduction deformities is shown in Figure 1C. Abduction of the forefoot was controlled by a spring placed on the lateral side of the FAS, as shown in Figure 1D. The FAS is a 2-piece orthosis with an adjustable spring between the lateral hindfoot and lateral forefoot. An ankle-fixing strap is placed inside the shoe to secure the heel. The shoe was initially set with a 208 to 258 outflare according to the principle of the Bēbax shoe. 27 When the spring contracts, the FAS retains the outflare, and the foot remains in abduction. For unilateral children the unaffected foot is free and for bilateral children a pair of FAS was needed at night. Given that the DB and FAS are only used at night, orthopedic shoes (OS) were used while walking, as shown in Figure 1B. The OS were used with an orthopedic insole, and the hard heel cup and upper shoe helps to keep the heel in a neutral position. To evaluate the function of different orthoses, these 3 different corrective methods were evaluated on children who were in the early stages of walking: DB, DBþOS, and FASþOS.

Subjects
This is a prospective, single-blinded, randomized, controlled trial. A total of 113 children with CTEV were recruited after Ponseti treatment. All children were participating following successful initial management using the Ponseti method and had been braced with a DB splint during their prewalking stages. The feet were examined and scored at the first visit according to the Diméglio classification, 20 which has been considered the most reliable classification method. 21 There was no significant difference in equinus, varus, and addutus deformities between them based on the scores (P > 0.05). After the Diméglio classification, 53 children with moderate deformities were recruited. All the children in this study had finished Ponseti treatment and were wearing DB splints for the initial period of correction, which effectively controlled torsional deformity. When they began to walk, 15 children continued to use a DB splint at night. A total of 20 children used DB splint for nighttime use and OS for daytime use. A total of 18 children accepted FAS for nighttime use and OS for daytime use. For unilateral children, the unaffected foot was not restricted or constrained. The data were collected when the children were between 4-5 years of age. Informed consent was obtained from their parents. Follow-up visits were done, and the orthoses were changed as the children grew. The mean follow-up time was 44 months. Table 1 shows the demographics of the children after they underwent corrective treatment. No significant differences were found in age, height, and weight among the three groups (P > 0.05). Informed written consent was obtained from the parents of each subject in accordance with clinical protocols. This study was approved by the Science and Ethics Committee

Procedure
All participants underwent three-dimensional foot scanning and pedobarography. All data were captured by an experienced pedorthic doctor and a bioengineer in the gait analysis laboratory at Rokab Pedorthic Center, Beijing, China.
A Delcam Iqube scanner (Delcam, Birmingham, UK) and a malleolar jig were used to collect and calculate the malleolar angles ( Figure 2A). The malleolar angle is determined from a scanned image of the plantar aspect of the foot while standing still ( Figure 2B). The malleolar jig was adjusted to fit and was aligned with the transmalleolar axis of each subject, and the medial and lateral malleoli were registered to the scanned image of the foot. 22 From this image, the bean-shape ratio ( Figure 2C), which was calculated by width to length, 24 could be obtained for each feet.
A FreeMed 1 baropodometric platform, as shown in Figure 3A, was used to measure the dynamic foot pressure (Rome, Italy). The platform surface was 240 Â 50 cm, with an active surface of 244 Â 74 cm and thickness of 0.8 cm (Sensormédica 1 , Italy). The reliability of this baropodometric platform has been shown in previous studies. 23 Dynamic measurements required the children to walk at their natural self-selected speed. Four recordings of each affected foot were taken, and the average was considered the final measurement. The FreeStep system 1 was used to provide information during gait.

Data Processing
The bimalleolar angle and bean-shape ratio were used to quantify the level of forefoot adduction. 2,24,25 The bimalleolar angle is the anteromedial angle which is formed between the bimalleolar axis and the longitudinal axis of the foot passing through the second toe. The bean-shape ratio assesses the curvature of the foot, thus quantifying forefoot adduction and hindfoot varus. The average and maximum peak pressure on each region ( Figure 3B) were calculated according to methods detailed in the literature. 26 The MFF/LFF ratio, heel/forefoot ratio, and heel/LMF ratio were calculated to evaluate the degree of equinus and varus deformity in CTEV children. 24,26 Statistical Analysis Statistical analyses were performed using the SPSS for Windows version 19.0 (IBM Corp, Armonk, NY, USA). Continuous variables were reported as means AE SD. The unaffected foot was not used for calculation during statistical analysis. One-way MAVONA analysis was used to evaluate the effects of different corrective methods on the bimalleolar angle, bean-shape ratio, and pressure distribution. Leastsignificant-difference tests were used for posthoc comparisons. Power analysis was performed using PASS software (Jerry Hintze, Kaysville. UT). The significance level was set at 0.05.

RESULTS
A total of 53 children were included in this study. There were no significant differences in varus, equines, and adduction deformities between the 3 groups according to the Diméglio score (P > 0.05) at initial inception. A total of 15, 20, and 18 children comprised the groups DB, DBþOS, and FASþOS, respectively. The study began in 2010 and the mean follow-up time was 44 months. No patients withdrew from the treatment during the study period. The FASþOS group showed better compliance than the other 2 groups, as reported by the children's parents. Although physiotherapy and orthoses were continuous, 5 children in the DB group and 1 child in the DBþOS group exhibited severe equinus, varus, and adduction deformities after treatment. Surgery may still be needed to correct these severe deformities.
MANOVA revealed that the different corrective methods had a primary influence on the bimalleolar angle and beanshape ratio, as shown in Table 2. For the 3 groups, DB, DBþOS, and FASþOS, the bimalleolar angle was found to successively increase [F (2,80) ¼ 3.598, P ¼ 0.032] and bean-shape ratio to successively decrease [F (2,80) ¼ 6.852, P ¼ 0.002], respectively. A significant difference in bimalleolar angle was found between the DB and FASþOS groups (posthoc comparison: P ¼ 0.009). Also, a significant difference in bean-shape ratio was found between the DB and DBþOS groups (posthoc comparison: P ¼ 0.049), as well as between the DBþOS and FASþOS groups (posthoc comparison: P < 0.001).
As shown in Table 4, MANOVA revealed that different corrective methods had a primary influence on the peak pressure ratios of heel/forefoot and heel/LMF. For the DB, DBþOS, and FASþOS groups, both the heel/forefoot and heel/LMF ratio successively increased [F (2,80) ¼ 4.975, P ¼ 0.009 and F (2,80) ¼ 14.878, P < 0.001]. A significant difference in heel/ forefoot ratio was found between the DB and DBþOS groups (posthoc comparison: P ¼ 0.006), as well as between the DB and FASþOS groups (posthoc comparison: P ¼ 0.005). A significant difference in the heel/lateral arch ratio was found among the DB, DBþOS, DBþOS groups (posthoc comparison: P ¼ 0.003, P < 0.001, P ¼ 0.007). For the significantly different variables (bimalleolar angle, bean-shape ratio, heel/forefoot, and heel/LMF), the power ranged from 0.65 to 0.99.

DISCUSSION
When treating CTEV, the corrective method used, or whether correction was used at all, will have a great impact Our results demonstrate that the FASþOS group exhibits superior correction of abnormal adduction (has the greatest bimalloeolar angle of 77.558 and the lowest bean-shape ratio of 0.27). A bimalleolar angle of less than 808 is considered abnormal. 25 Although the use of FASþOS resulted in an angle of less than 808, it achieved a higher value than the other 2 groups and a significant difference was observed between the FASþOS and the DB groups (posthoc comparison: P ¼ 0.009). The bean-shape ratio is a comprehensive index that indicates forefoot adduction as well as hindfoot varus. 3 The more curved the foot in the transverse plane, the higher the ratios were. If values are greater than 0.267, a bean-shaped foot is indicated, and if the value is greater than 0.34, a marked deformity is denoted. 24 The bean-shape ratio in the FASþOS group was 0.27, which is the closest to the normal value of 0.23 AE 0.02. A significantly greater bimalloeolar angle and lower bean-shape ratio was observed in the FASþOS group compared with the DB and DBþOS groups (posthoc comparison: P < 0.001, P ¼ 0.049). The result demonstrates that FAS may be critical to correcting forefoot adduction, and OSþFAS has a positive effect in correcting forefoot adduction and varus. Some studies have proposed that if the medial soft tissue is released, the release of the medial plantar fascia, tendon, and ligament can help to correct varus. 28 The bean-shape ratio has been considered a sensitive indicator of relapse. 24 Our results suggest that FASþOS placed the foot into a position with a low beanshape ratio, which could help decrease adduction relapse in patients with CTEV.
Our results also demonstrate that OS have a positive effect on the correction of equino-varus deformities in the OS and FASþOS groups. These results are reflected by the heel/forefoot and heel/ LMF ratios. If the value of heel/forefoot ratio is less than 0.8, an equinus deformity is considered. If the value is less than 0.4, severe equinus deformity is considered. 25 The average values of heel/forefoot ratios of the DB and DBþOS groups were 0.44 and 0.72, respectively. A significant difference was observed in the DBþOS group when compared with the DB group (posthoc comparison: P ¼ 0.006). The results demonstrated that OS is critical to the correcting equinus in CETV. The heel/LMF ratio is considered a sensitive indicator of equino-varus deformity. 24 If the ratio is less than 2.5, an equino-varus deformity is typically diagnosed. If the ratio is less than 1.0, a severe equino-varus deformity is considered. 25 The average values of the heel/LMF ratios of the DB and DBþOS groups were 0.77 and 1.45, respectively. A significant difference was observed among the DB and DBþOS groups (posthoc comparison: P ¼ 0.003, and P ¼ 0.007). The results also demonstrate that OS have a positive effect on the correction of equino-varus deformities in CETV.
Compared with previous research, 26 this study found lower values for both heel/forefoot and heel/LMF pressure ratio in the DB group. This difference may be attributed to the younger age of the children in our study, and their noncompliance to the DB splint. Noncompliance may be considered to be the greatest barrier to successful DB splinting. 3,12 A lack of adherence to the bracing protocol increases the risk of relapse. In this study, most DB group parents found it difficult to persist in donning the DB This work presents essential information on foot morphological characteristics and pressure characteristics in children with CTEV treated with different corrective techniques. This quantitative description is not only objective but also reproducible, and could provide more useful parameters than single recurrence rates or clinical evaluation. However, this study has some limitations: The corrective methods used do not consider severe foot deformities. Also, FAS is used for controlling adduction, it is not effective for controlling other types of deformity. FASþOS design applies on those children who do not have serious torsional deformity.
The bias of the patients recruited in the study is extremely wide, even though it is a randomized study. There are many other conditions such as muscle strength, cooperation with physiotherapy, and cognitive skills which may have an effect on the results.

CONCLUSION
This article developed a simple manner of correction that consists of the daytime and nighttime use of orthoses for moderate CTEV children. OS used during the day mainly control equinus and varus, while FAS are used at night to correct forefoot adduction. This novel method achieves better corrective results for controlling equinovarus, especially in forefoot adduction, in comparison to other conventional treatment regimes. Compliance is also improved. Therefore, this approach may be a more appropriate treatment option for CTEV children with moderate foot deformities and without serious torsional deformity.