- Email: [email protected]
Difficulties During Ponseti Casting for the Treatment of Idiopathic Clubfoot
ARTICLE IN PRESS The Journal of Foot & Ankle Surgery 000 (2019) 1−5
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery journal homepage: www.jfas.org
Original Research
Difficulties During the Ponseti Casting for the Treatment of Idiopathic Clubfoot Kyung Rae Ko, MD1, Jong Sup Shim, MD, PhD2, Joo Hwan Kim, MD3, Young Taek Cha, MD3 1
Orthopedic Surgeon, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea Professor, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 3 Orthopedic Resident, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 2
A R T I C L E
I N F O
Level of Clinical Evidence: 3 Keywords: clubfoot Dimeglio score foot length Ponseti method
A B S T R A C T
The Ponseti method has resulted in excellent outcomes in the treatment of idiopathic clubfoot. To evaluate the process of serial casting, most studies previously used the number of casts required to achieve correction. In contrast, difficulties that occurred during serial casting were rarely studied. We retrospectively reviewed 47 patients with severe idiopathic clubfoot (Dimeglio score ≥10) and a follow-up period of ≥4 years by focusing on difficulties during the Ponseti casting and their related factors. All events requiring temporary discontinuation and/or delay of scheduled casting were considered difficulties. Among the 47 patients, difficulties occurred in 18 patients (38.3%). Initial Dimeglio score (14.9 § 2.5 [vs 13.5 § 2.3, p = .042]), its subcomponent of equinus (3.8 § 0.4 [vs 3.3 § 0.6, p = .005]), and the number of casts (6.3 § 2.0 [vs 4.9 § 1.2, p = .009]) in the 18 patients with difficulties were greater than those of the remaining 29 patients. In addition, the foot length was shorter (7.2 § 0.6 cm [vs 7.7 § 0.8 cm, p = .030]) in patients with difficulties. In the regression analysis, both subcomponents of equinus and foot length showed the significance for the occurrence of difficulties. During 8.7 § 3.8 years, further surgery was needed in 35 patients (74.5%) because of residual or recurrent deformity. The ratio was higher in patients with difficulties (94.4% [vs 62.1%, p = .017]). In our experience, difficulties frequently occurred in feet with rigid equinus and small length, and further surgery was required after the initial correction in almost all of the feet with difficulties. © 2019 by the American College of Foot and Ankle Surgeons. All rights reserved.
The Ponseti method (1) has become a standard option of initial treatment for idiopathic clubfoot (2). Despite its noninvasiveness and accessibility to initial correction (3,4), difficulties occasionally occur during serial casting (5). The development of severe skin lesions or uncontrolled edema results in the delay of scheduled casting. Moreover, slippage of applied cast is an emergency that should be managed promptly (Fig. 1). To date, a large body of literature has reported the results of the Ponseti method (1,3−8). To evaluate the process of serial casting, most studies used the number of casts required to achieve an acceptable correction (3,5−7). In contrast, difficulties and other events that occurred during serial casting (i.e., the contents of the process) were rarely studied (5). We believe that more comprehensive efforts to evaluate the process of serial casting are needed in addition to its duration. On this background, we conducted a study on the difficulties during serial casting and their related factors. Because we focused on difficulties, which were rarely observed in mild deformity in our clinical
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Jong Sup Shim, MD, PhD, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, (06351) 81 Irwon-Ro, Gangnam-gu, Seoul, Korea. E-mail address: [email protected] (J.S. Shim).
experience, only severe cases of idiopathic clubfoot were included. In addition to the severity, it was assumed that body size would affect the process of casting. In particular, we assumed that patients with lean body mass and long foot length would be advantageous during the correction without difficulties. Patients and Methods The protocol of this study was approved by the Institutional Review Board of the Samsung Medical Center (IRB No. 2019-03-053). At our institution, the Dimeglio scoring system, a 20-point scale that consists of 4 major components and 4 additional points (9), has been used to assess the initial severity. Feet with a score of <10 points and nonidiopathic deformity were not included. Patients with a history of previous treatment and a follow-up period of <4 years were also excluded. The study period was from 2003 to 2015. After the requirement for informed consent had been waived by the board, we retrospectively reviewed 91 patients with clubfoot who were treated with the Ponseti method at our institution during the study period. Of these, 47 patients (28 boys, 19 girls; 21 bilateral) were included in the study. In all of the feet, initial treatment was performed by the same orthopedic surgeon (J.S. S.) using the Ponseti method (1). The Dimeglio score was checked just before the first session of manipulation and casting. Later sessions were performed at weekly intervals. A short-leg plaster cast was placed first, allowed to dry, and extended up to the proximal thigh to create a long-leg cast with the knee in 90 flexion. Medical records of serial casting and Achilles tenotomy were retrospectively reviewed. In particular, all events requiring temporary discontinuation and/or delay of scheduled casting were considered difficulties. After removal of the final cast, a foot abduction brace was applied. The brace
1067-2516/$ - see front matter © 2019 by the American College of Foot and Ankle Surgeons. All rights reserved. https://doi.org/10.1053/j.jfas.2019.07.022
ARTICLE IN PRESS 2
K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
Fig. 2. Measurement of the foot length on the anteroposterior radiograph. The long axis of the foot and 2 lines, perpendicular to the long axis, were drawn. The marked width indicates the anteroposterior foot length. Fig. 1. Slippage of applied cast in the right foot. Note that the toes of the right foot are not visible compared with the left side. protocol was 23 hours/day during the first 3 months, then 12 hours/day until the patients were 2 years old, and then for prescribed time considering the clinical course. The need for further treatment for residual or recurrent deformity after the initial correction, including additional serial casting, additional Achilles tenotomy, and other surgical procedures, was reviewed. As a parameter of body mass, body weight at birth was used. The foot length was measured by using anteroposterior (AP) and lateral radiographs, which were taken within 48 hours before the beginning of the first session. Both views were obtained with active immobilization and compression by a gloved assistant who kept the foot in passive dorsiflexion by using a pedestal. On AP view, the long axis of the foot that connects the distal tip of the second toe and posterior end of the heel was drawn. Two lines, perpendicular to the long axis of the foot, were drawn through the most anterior and posterior margins of the foot, respectively. The distance between the 2 lines was used as the AP foot length (Fig. 2). On lateral view, 2 lines were drawn perpendicular to the baseline of the foot through the most anterior and posterior margins of the foot. The distance between the 2 lines was used as the lateral foot length (Fig. 3). The average of AP and lateral foot lengths was used as the foot length. We evaluated 4 radiographic angles by using radiographs checked at the final follow-up. The talocalcaneal and talo−first metatarsal angles were measured on weightbearing (standing) AP and lateral views. We described adduction on AP view and cavus on lateral view by using negative talo−first metatarsal angles. All of the radiographic parameters were measured by using the Picture Archiving and Communication System (Centricity Enterprise Web V3.0, GE Healthcare, Buckinghamshire, UK). All parameters, including the foot length, were measured by 2 independent observers, and mean values were used. Each measurement was performed at intervals of a minimum of 1 month by each observer. To evaluate the intraobserver and interobserver reliabilities, intraclass correlation coefficients (ICCs) were calculated and were >0.8 in all parameters, which indicated excellent reliability (10).
available (http://www.randomizer.org) (3,11). As a result, 47 severe clubfeet were included in the analyses. All numerical data are presented as mean § standard deviation (with range) to 1 decimal place. The Mann-Whitney test was used to compare the numerical data between the subgroups. Fisher exact test was performed to compare the ratio between the subgroups. Correlation analyses were performed to analyze the associations among the numerical data. The correlation coefficients (r) of 0.36 to 0.67 (−0.67 to −0.36) were considered moderate positive (negative) correlations (12). Logistic regression analyses were used to describe the significant factors for the occurrence of difficulties. All statistical analyses were performed using SPSS version 23 for Windows (IBM Corp., Armonk, NY), and the significance level was set at p < .05.
Statistical Analysis To avoid violation of the principle of statistical independence, 1 foot was randomly selected in 18 patients with bilateral severe clubfeet, using a program that is freely
Fig. 3. Measurement of the foot length on the lateral radiograph. The baseline of the foot and 2 lines, perpendicular to the baseline, were drawn. The marked width indicates the lateral foot length.
ARTICLE IN PRESS K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
Results The mean patient age was 8.4 § 9.1 days (range 0 to 48), and the mean Dimeglio score was 14.0 § 2.5 (range 10 to 18) at the first casting. The number of casts required to achieve an acceptable correction ranged from 3 to 10 (5.5 § 1.7). A median of 5 casts was required. Achilles tenotomy was needed and performed in 44 feet (93.6%) after serial casting. Regarding the occurrence of difficulties, slippage of applied cast occurred 20 times in 8 feet (17.0%). Delay of scheduled casting occurred 21 times in 14 feet (29.8%) as a result of skin lesions (pressure sores, dermatitis, etc.) and/or edema with a duration of 4.9 § 1.8 days (range 2 to 8 days). As a result, there were 18 feet with difficulties (38.3%, difficult feet [DF]) (Table 1). Of the 47 feet, 12 feet (25.5%) did not require further treatment during the follow-up period of 8.7 § 3.8 years (range 4 to 15.3 years) after the initial correction (Table 2). The mean body weight at birth was 3.2 § 0.4 kg (range 2.6 to 4.5 kg). The mean AP foot length was 7.7 § 0.8 cm (range 6.1 to 10.2 cm), and the mean lateral foot length was 7.3 § 0.9 cm (range 5.5 to 9.9 cm). The foot length, calculated using the average of AP and lateral foot lengths, was 7.5 § 0.8 cm (range 6.2 to 10.0 cm). Comparisons Between the Subgroups Subgroup analyses were performed between the 18 DF and the remaining 29 feet. Initial total Dimeglio score (14.9 § 2.5 [vs 13.5 § 2.3, p = .042]), its subcomponent of equinus (3.8 § 0.4 [vs 3.3 § 0.6, p = .005]), and the number of casts (6.3 § 2.0 [vs 4.9 § 1.2, p = .009]) were greater in the 18 DF. In addition, the foot length was shorter (7.2 § 0.6 cm [vs 7.7 § 0.8 cm, p = .013]) in the 18 DF. Otherwise, body weight at birth was not statistically different between the 2 groups (3.2
3
§ 0.4 and 3.3 § 0.3 kg, p = .504]). Other indicators, including the 4 radiographic angles at the last follow-up, were not statistically different. All except 1 of the 18 DF required further surgery after the initial correction. The ratio of the need in the DF was significantly higher than that in the remaining feet (94.4% [vs 62.1%, p = .017]) (Table 3). In addition to the occurrence of difficulties, the 47 feet were categorized according to the number of casts (>5 [the median in the 47 feet] vs ≤5). As a result, the ratio of the need for further surgery was not significantly associated with the number of casts (p = .288, obtained using Fisher exact test).
Logistic Regression Analysis for the Occurrence of Difficulties Before regression analysis, the associations among initial total Dimeglio score, its 4 subcomponents, foot length, and body weight at birth were analyzed by using the correlation analysis. The foot length showed weak associations with total Dimeglio score and its 4 subcomponents (r = −0.233 to −0.024). The body weight at birth also showed weak associations with total Dimeglio score and its 4 subcomponents (r = −0.166 to 0.094). Total Dimeglio score and the subcomponent of equinus were strongly associated (r = 0.687). The foot length and body weight were positively but weakly associated (r = 0.286) (Table 4). On the basis of the above-mentioned results of comparisons between the 18 DF and the remaining 29 feet and the correlation analyses in the 47 feet, we designed a regression model to describe the significant factors for the occurrence of difficulties by using the subcomponent of equinus and the foot length as independent variables. As a result, both independent variables showed statistical significance. In other words, more rigid equinus and shorter foot length were significantly associated with the occurrence of difficulties after each other’s adjustments (Table 5). We limited further description using a
Table 1 The details of serial casting (47 feet in 47 patients) No. of Casts
No. of Feet
Dimeglio Score
Difficulties
Achilles Tenotomy
≤ 4a
14
4/14 (28.6%)
All
5
18
3/18 (16.7%)
16/18 (88.9%)
6
7
5/7 (71.4%)
6/7 (85.7%)
≥ 7b
8 Total 47
13.9 § 1.8 (range 10 to 17) 13.7 § 2.8 (range 10 to 18) 14.1 § 2.2 (range 11 to 17) 14.9 § 3.0 (range 11 to 18)
6/8 (75.0%) Total 18/47 (38.3%)
All Total 44/47 (93.6%)
a 3 casts in 2 feet; 4 casts in 12 feet. b 7 casts in 1 foot; 8 casts in 1 foot; 9 casts in 5 feet; 10 casts in 1 foot.
Table 2 Details of further treatment after the initial correction (47 feet in 47 patients) Contents
No. of Feet
Time of Surgery
Additional serial casting Additional Achilles tenotomy
2 (4.3%)d 9 (19.1%)
N/A At 12.0 § 9.3 months (range 3.6 to 35.4)
Extensive soft tissue release
9 (19.1%)
a
SSTR § TA tendon transfer § other procedures Awaiting further surgeryb No further treatmentc
21 (44.7%) 7 (14.9%) 12 (25.5%)
At 1.6 § 1.1 years (range 0.8 to 4.3) At 6.0 § 1.6 years (range 4.0 to 11.2) N/A N/A
Abbreviations: SSTR, selective soft tissue release; TA, tibialis anterior; N/A, not applicable. a Performed in 18 feet. b Patients who rejected the recommendation for further surgery, as well as those awaiting further surgery, were included. c Patients awaiting further surgery were not included. d Two feet underwent additional surgery later.
Table 3 Comparisons of clinical characteristics and radiographic evaluation between the 2 groups (47 feet in 47 patients) Indicators
Feet with difficulties (n = 18)
Feet without difficulties (n = 29)
p
Sex (male:female) Bilaterality Side (right:left) Total Dimeglio score Equinusa Varusa Medial rotationa Forefoot adductiona Posterior creaseb Medial creaseb Cavusb Poor muscle conditionb No. of casts Age at the first casting (days) Foot length (cm) Body weight at birth (kg) Noncompliance on bracec Follow-up period (years) Need for further surgery AP TC angle ( )d Lateral TC angle ( )d AP TM angle ( )d Lateral TM angle ( )d
12:6 9 9:9 14.9 § 2.5 3.8 § 0.4 2.7 § 0.8 2.7 § 0.8 2.4 § 0.9 18 15 17 9 6.3 § 2.0 7.5 § 4.9 7.2 § 0.6 3.2 § 0.4 3 8.9 § 3.8 17 20.0 § 7.6 31.1 § 10.4 −2.0 § 8.6 −3.2 § 9.6
16:13 12 19:10 13.5 § 2.3 3.3 § 0.6 2.6 § 0.6 2.4 § 0.8 2.2 § 0.7 28 18 24 16 4.9 § 1.2 9.0 § 11.0 7.7 § 0.8 3.3 § 0.3 3 8.6 § 3.8 18 20.6 § 7.7 33.2 § 5.4 0.7 § 7.8 0.9 § 6.4
.546g .763g .365g .042e .005e .455e .266e .225e >.999g .191g .384g .771g .009e .455e .030e .504e .662g .734e .017g .743e .983e .175e .074e
Abbreviations: AP, anteroposterior; TC, talo-calcaneal; TM, talo−first metatarsal. a Subcomponent of the Dimeglio score. b Number of patients with 1 point. c Defined as reported brace wearing time of <50% of the prescribed time. d Measured on radiographs checked at the final follow-up. e Obtained using the Mann-Whitney tests. g Obtained using Fisher exact tests.
ARTICLE IN PRESS K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
4
Table 4 Correlation analysis among the total Dimeglio score, its 4 subcomponents, foot length, and body weight at birth (47 feet in 47 patients)
Total Dimeglio score Equinus Varus Medial rotation Forefoot adduction Foot length Body weight
Total Dimeglio Score
Equinus
Varus
Medial Rotation
Forefoot Adduction
1
0.687* 1
0.727* 0.450* 1
0.608* 0.203 0.324 1
0.786* 0.583* 0.476* 0.303 1
Foot Length 0.233 0.126 0.024 0.209 0.183 1
Body Weight 0.055 0.036 0.094 0.026 0.166 0.286 1
In each cell, the Pearson correlation coefficient is described. *Coefficient with moderate or strong association.
Table 5 Logistic regression analysis using the occurrence of difficulties as a dependent variable (47 feet in 47 patients) Independent Variable
Exp (B)
95% CI
p
Subcomponent of equinus (points) Foot length (cm)
7.153 0.900
1.560 to 32.794 0.812 to 0.999
.011 .047
Abbreviation: CI, confidence interval.
scatterplot to 27 feet whose subcomponent of equinus was 4 points (ie, reducibility: 90 to 45 ). Among 13 feet shorter than the median (7.4 cm), difficulties occurred in 10 feet (76.9%) (Fig. 4). Discussion The most important finding of this study was that more rigid equinus and shorter foot length at initial presentation were significantly associated with the occurrence of difficulties during the Ponseti casting in idiopathic severe clubfoot. The results of clubfoot correction using the Ponseti method have been frequently reported (1,3,4,6−8). Unlike previous studies, we analyzed the foot length as a clinical factor,
Fig. 4. Scatterplot showing the foot length (mm) and body weight (kg) in 27 feet whose subcomponent of equinus was 4 points. Fifteen feet with difficulties (black circles) and 12 feet without difficulties (with circles) are described. The median foot length (7.4 cm) is marked. Note that difficulties occurred in 10 feet among 13 feet <7.4 cm (76.9%). Otherwise, difficulties occurred in 5 feet among 14 feet >7.4 cm (35.7%).
focused on difficulties during serial casting to comprehensively assess its process, and included only severe cases of idiopathic clubfoot. These were the main achievements and strengths of this study. We believe that more comprehensive efforts to assess the process of serial casting are needed. In particular, the contents of serial casting should be assessed in addition to its duration, which was evaluated by using the number of casts in previous studies (3,5−7). As a result, it was found that not the criteria of the number of casts but the occurrence of difficulties was significantly associated with the need for further surgery. This finding suggests the clinical significance of difficulties. Regarding the foot length as a clinical factor, it was reported that clubfeet with the length of ≥8 cm had better final Dimeglio score than those with the length of <8 cm with a median follow-up of 34 months (5). This is similar to our results that shorter foot length was significantly associated with difficulties during serial casting. The adjustment for the severity, performed in our regression analysis, might strengthen our results. It is generally believed that clubfoot should be treated soon after birth (13−17). However, the upper limit age for the Ponseti method is unclear. It was reported that the treatment using the Ponseti method initiated between 28 days and 3 months showed a lower rate of relapse and better clinical outcome compared with early treatment, before 28 days, with a mean follow-up of 5 years (8). On the basis of our results concerning foot length, the growth of the foot in the delayed group might result in more effective casting. Further study is needed to verify. In this study, the foot length was measured using simple radiographs. Due to the beam inclination, there might be a difference between the measurement and real length. In addition, the measurement might be influenced by foot position when taking the radiographs. To include the reliable value, the foot length was calculated using both AP and lateral views. Regarding the reliability, AP and lateral foot lengths were strongly associated (r = 0.870). Moreover, the ICC, calculated using the 2 values, showed excellent reliability (ICC = .930, 95% confidence interval: 0.874 to 0.961). Even if the subjects were limited to very severe cases (Dimeglio score of ≥15), the association was strong (r = 0.838), and reliability was excellent (ICC = .904, 95% confidence interval: 0.763 to 0.961). We believe that the value used in this study could at least indicate the ranking of the real length. Ponseti et al (18) described the complex clubfoot, which is distinguished by rigid equinus with a deep crease above the heel, significant shortening, a deep plantar crease and high cavus, and severe plantar flexion of metatarsals. The above-mentioned characteristics of complex clubfoot were consistent with those of the DF in this study. In addition to rigid equinus and short length, almost all of the DF scored additional points of posterior crease and cavus in the Dimeglio scoring system. We believe that a distinct entity of idiopathic clubfeet distinguished from the typical idiopathic clubfeet exists and that rigid equinus and short length are its noticeable characteristics. Orthopedic surgeons should pay special attention to feet with these characteristics. Our findings of the need for further surgery due to residual or recurrent deformity suggest that DF would have a relatively more complicated course. Of the 47 feet, further surgery was needed in 35 feet (74.5%)
ARTICLE IN PRESS K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
during the periods of 8.7 § 3.8 years. Comparing this ratio to previous study with reported Dimeglio score and similar periods, it was reported that 68% of patients with the mean initial Dimeglio score of 13.0 § 2.0 had recurrence during 6.8 § 1.4 years (4). Our somewhat higher ratio was explainable by the mean initial Dimeglio score of 14.0 § 2.5. However, heterogeneity of further treatment after the Ponseti method and a lack of a scoring system to assess clinical outcomes (4,8) were the limitations of our study, which restricted accurate comparisons. There were several other limitations. First, only the Dimeglio score was used to rate the severity at presentation. However, an almost perfect correlation between the Dimeglio score and the Pirani score at presentation was found (3). A lack of quantitative measurement of compliance on brace was also a limitation. We believe that the reported time is not reliable, and there is no consensus on the definition of noncompliance (19). Therefore, strict criteria for the compliance (4) and detailed data on brace-wearing time were not included in analyses. In our cohort of idiopathic severe clubfoot, the initial severity of equinus and length of the foot were significantly associated with the occurrence of difficulties during the Ponseti casting. Difficulties frequently occurred in feet with rigid equinus and small length, and further surgery was required after the initial correction in almost all of the DF. Supplementary Materials Supplementary material associated with this article can be found in the online version at https://doi.org/10.1053/j.jfas.2019.07.022.
References 1. Ponseti IV, Smoley EN. The classic: congenital club foot: the results of treatment. 1963. Clin Orthop Relat Res 2009;467:1133–1145. 2. Carroll NC. Clubfoot in the twentieth century: where we were and where we may be going in the twenty-first century. J Pediatr Orthop B 2012;21:1–6.
5
3. Lampasi M, Abati CN, Bettuzzi C, Stilli S, Trisolino G. Comparison of Dimeglio and Pirani score in predicting number of casts and need for tenotomy in clubfoot correction using the Ponseti method. Int Orthop 2018;42:2429–2436. 4. Zionts LE, Ebramzadeh E, Morgan RD, Sangiorgio SN. Sixty years on: Ponseti method for clubfoot treatment produces high satisfaction despite inherent tendency to relapse. J Bone Joint Surg Am 2018;100:721–728. 5. Iltar S, Uysal M, Alemdaroglu KB, Aydogan NH, Kara T, Atlihan D. Treatment of clubfoot with the Ponseti method: should we begin casting in the newborn period or later? J Foot Ankle Surg 2010;49:426–431. 6. Brazell C, Carry PM, Jones A, Baschal R, Miller N, Holmes KS, Georgopoulos G. Dimeglio score predicts treatment difficulty during Ponseti casting for isolated clubfoot. J Pediatr Orthop 2019;39:e402–e405. https://doi.org/10.1097/BPO.0000000000001325. 7. Gao R, Tomlinson M, Walker C. Correlation of Pirani and Dimeglio scores with number of Ponseti casts required for clubfoot correction. J Pediatr Orthop 2014;34:639–642. 8. Liu YB, Li SJ, Zhao L, Yu B, Zhao DH. Timing for Ponseti clubfoot management: does the age matter? 90 children (131 feet) with a mean follow-up of 5 years. Acta Orthop 2018;89:662–667. 9. Dimeglio A, Bensahel H, Souchet P, Mazeau P, Bonnet F. Classification of clubfoot. J Pediatr Orthop B 1995;4:129–136. 10. Fleiss JL. Statistical Methods for Rates and Proportions. Wiley Interscience, New York, 1981. 11. Gray K, Gibbons P, Little D, Burns J. Bilateral clubfeet are highly correlated: a cautionary tale for researchers. Clin Orthop Relat Res 2014;472:3517–3522. 12. Taylor R. Interpretation of the correlation coefficient: a basic review. J Diagn Med Sonogr 1990;6. 35−33. 13. Dobbs MB, Gurnett CA. Update on clubfoot: etiology and treatment. Clin Orthop Relat Res 2009;467:1146–1153. 14. Dobbs MB, Rudzki JR, Purcell DB, Walton T, Porter KR, Gurnett CA. Factors predictive of outcome after use of the Ponseti method for the treatment of idiopathic clubfeet. J Bone Joint Surg Am 2004;86:22–27. 15. Liu Y, Zhao D, Zhao L, Li H, Yang X. Congenital clubfoot: early recognition and conservative management for preventing late disabilities. Indian J Pediatr 2016;83:1266–1274. 16. Morcuende JA, Dolan LA, Dietz FR, Ponseti IV. Radical reduction in the rate of extensive corrective surgery for clubfoot using the Ponseti method. Pediatrics 2004;113: 376–380. 17. Zhao D, Li H, Zhao L, Liu J, Wu Z, Jin F. Results of clubfoot management using the Ponseti method: do the details matter? A systematic review. Clin Orthop Relat Res 2014; 472:1329–1336. 18. Ponseti IV, Zhivkov M, Davis N, Sinclair M, Dobbs MB, Morcuende JA. Treatment of the complex idiopathic clubfoot. Clin Orthop Rel Res 2006;451:171–176. 19. Zhao D, Liu J, Zhao L, Wu Z. Relapse of clubfoot after treatment with the Ponseti method and the function of the foot abduction orthosis. Clin Orthop Surg 2014; 6:245–252.
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery journal homepage: www.jfas.org
Original Research
Difficulties During the Ponseti Casting for the Treatment of Idiopathic Clubfoot Kyung Rae Ko, MD1, Jong Sup Shim, MD, PhD2, Joo Hwan Kim, MD3, Young Taek Cha, MD3 1
Orthopedic Surgeon, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea Professor, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 3 Orthopedic Resident, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 2
A R T I C L E
I N F O
Level of Clinical Evidence: 3 Keywords: clubfoot Dimeglio score foot length Ponseti method
A B S T R A C T
The Ponseti method has resulted in excellent outcomes in the treatment of idiopathic clubfoot. To evaluate the process of serial casting, most studies previously used the number of casts required to achieve correction. In contrast, difficulties that occurred during serial casting were rarely studied. We retrospectively reviewed 47 patients with severe idiopathic clubfoot (Dimeglio score ≥10) and a follow-up period of ≥4 years by focusing on difficulties during the Ponseti casting and their related factors. All events requiring temporary discontinuation and/or delay of scheduled casting were considered difficulties. Among the 47 patients, difficulties occurred in 18 patients (38.3%). Initial Dimeglio score (14.9 § 2.5 [vs 13.5 § 2.3, p = .042]), its subcomponent of equinus (3.8 § 0.4 [vs 3.3 § 0.6, p = .005]), and the number of casts (6.3 § 2.0 [vs 4.9 § 1.2, p = .009]) in the 18 patients with difficulties were greater than those of the remaining 29 patients. In addition, the foot length was shorter (7.2 § 0.6 cm [vs 7.7 § 0.8 cm, p = .030]) in patients with difficulties. In the regression analysis, both subcomponents of equinus and foot length showed the significance for the occurrence of difficulties. During 8.7 § 3.8 years, further surgery was needed in 35 patients (74.5%) because of residual or recurrent deformity. The ratio was higher in patients with difficulties (94.4% [vs 62.1%, p = .017]). In our experience, difficulties frequently occurred in feet with rigid equinus and small length, and further surgery was required after the initial correction in almost all of the feet with difficulties. © 2019 by the American College of Foot and Ankle Surgeons. All rights reserved.
The Ponseti method (1) has become a standard option of initial treatment for idiopathic clubfoot (2). Despite its noninvasiveness and accessibility to initial correction (3,4), difficulties occasionally occur during serial casting (5). The development of severe skin lesions or uncontrolled edema results in the delay of scheduled casting. Moreover, slippage of applied cast is an emergency that should be managed promptly (Fig. 1). To date, a large body of literature has reported the results of the Ponseti method (1,3−8). To evaluate the process of serial casting, most studies used the number of casts required to achieve an acceptable correction (3,5−7). In contrast, difficulties and other events that occurred during serial casting (i.e., the contents of the process) were rarely studied (5). We believe that more comprehensive efforts to evaluate the process of serial casting are needed in addition to its duration. On this background, we conducted a study on the difficulties during serial casting and their related factors. Because we focused on difficulties, which were rarely observed in mild deformity in our clinical
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Jong Sup Shim, MD, PhD, Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, (06351) 81 Irwon-Ro, Gangnam-gu, Seoul, Korea. E-mail address: [email protected] (J.S. Shim).
experience, only severe cases of idiopathic clubfoot were included. In addition to the severity, it was assumed that body size would affect the process of casting. In particular, we assumed that patients with lean body mass and long foot length would be advantageous during the correction without difficulties. Patients and Methods The protocol of this study was approved by the Institutional Review Board of the Samsung Medical Center (IRB No. 2019-03-053). At our institution, the Dimeglio scoring system, a 20-point scale that consists of 4 major components and 4 additional points (9), has been used to assess the initial severity. Feet with a score of <10 points and nonidiopathic deformity were not included. Patients with a history of previous treatment and a follow-up period of <4 years were also excluded. The study period was from 2003 to 2015. After the requirement for informed consent had been waived by the board, we retrospectively reviewed 91 patients with clubfoot who were treated with the Ponseti method at our institution during the study period. Of these, 47 patients (28 boys, 19 girls; 21 bilateral) were included in the study. In all of the feet, initial treatment was performed by the same orthopedic surgeon (J.S. S.) using the Ponseti method (1). The Dimeglio score was checked just before the first session of manipulation and casting. Later sessions were performed at weekly intervals. A short-leg plaster cast was placed first, allowed to dry, and extended up to the proximal thigh to create a long-leg cast with the knee in 90 flexion. Medical records of serial casting and Achilles tenotomy were retrospectively reviewed. In particular, all events requiring temporary discontinuation and/or delay of scheduled casting were considered difficulties. After removal of the final cast, a foot abduction brace was applied. The brace
1067-2516/$ - see front matter © 2019 by the American College of Foot and Ankle Surgeons. All rights reserved. https://doi.org/10.1053/j.jfas.2019.07.022
ARTICLE IN PRESS 2
K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
Fig. 2. Measurement of the foot length on the anteroposterior radiograph. The long axis of the foot and 2 lines, perpendicular to the long axis, were drawn. The marked width indicates the anteroposterior foot length. Fig. 1. Slippage of applied cast in the right foot. Note that the toes of the right foot are not visible compared with the left side. protocol was 23 hours/day during the first 3 months, then 12 hours/day until the patients were 2 years old, and then for prescribed time considering the clinical course. The need for further treatment for residual or recurrent deformity after the initial correction, including additional serial casting, additional Achilles tenotomy, and other surgical procedures, was reviewed. As a parameter of body mass, body weight at birth was used. The foot length was measured by using anteroposterior (AP) and lateral radiographs, which were taken within 48 hours before the beginning of the first session. Both views were obtained with active immobilization and compression by a gloved assistant who kept the foot in passive dorsiflexion by using a pedestal. On AP view, the long axis of the foot that connects the distal tip of the second toe and posterior end of the heel was drawn. Two lines, perpendicular to the long axis of the foot, were drawn through the most anterior and posterior margins of the foot, respectively. The distance between the 2 lines was used as the AP foot length (Fig. 2). On lateral view, 2 lines were drawn perpendicular to the baseline of the foot through the most anterior and posterior margins of the foot. The distance between the 2 lines was used as the lateral foot length (Fig. 3). The average of AP and lateral foot lengths was used as the foot length. We evaluated 4 radiographic angles by using radiographs checked at the final follow-up. The talocalcaneal and talo−first metatarsal angles were measured on weightbearing (standing) AP and lateral views. We described adduction on AP view and cavus on lateral view by using negative talo−first metatarsal angles. All of the radiographic parameters were measured by using the Picture Archiving and Communication System (Centricity Enterprise Web V3.0, GE Healthcare, Buckinghamshire, UK). All parameters, including the foot length, were measured by 2 independent observers, and mean values were used. Each measurement was performed at intervals of a minimum of 1 month by each observer. To evaluate the intraobserver and interobserver reliabilities, intraclass correlation coefficients (ICCs) were calculated and were >0.8 in all parameters, which indicated excellent reliability (10).
available (http://www.randomizer.org) (3,11). As a result, 47 severe clubfeet were included in the analyses. All numerical data are presented as mean § standard deviation (with range) to 1 decimal place. The Mann-Whitney test was used to compare the numerical data between the subgroups. Fisher exact test was performed to compare the ratio between the subgroups. Correlation analyses were performed to analyze the associations among the numerical data. The correlation coefficients (r) of 0.36 to 0.67 (−0.67 to −0.36) were considered moderate positive (negative) correlations (12). Logistic regression analyses were used to describe the significant factors for the occurrence of difficulties. All statistical analyses were performed using SPSS version 23 for Windows (IBM Corp., Armonk, NY), and the significance level was set at p < .05.
Statistical Analysis To avoid violation of the principle of statistical independence, 1 foot was randomly selected in 18 patients with bilateral severe clubfeet, using a program that is freely
Fig. 3. Measurement of the foot length on the lateral radiograph. The baseline of the foot and 2 lines, perpendicular to the baseline, were drawn. The marked width indicates the lateral foot length.
ARTICLE IN PRESS K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
Results The mean patient age was 8.4 § 9.1 days (range 0 to 48), and the mean Dimeglio score was 14.0 § 2.5 (range 10 to 18) at the first casting. The number of casts required to achieve an acceptable correction ranged from 3 to 10 (5.5 § 1.7). A median of 5 casts was required. Achilles tenotomy was needed and performed in 44 feet (93.6%) after serial casting. Regarding the occurrence of difficulties, slippage of applied cast occurred 20 times in 8 feet (17.0%). Delay of scheduled casting occurred 21 times in 14 feet (29.8%) as a result of skin lesions (pressure sores, dermatitis, etc.) and/or edema with a duration of 4.9 § 1.8 days (range 2 to 8 days). As a result, there were 18 feet with difficulties (38.3%, difficult feet [DF]) (Table 1). Of the 47 feet, 12 feet (25.5%) did not require further treatment during the follow-up period of 8.7 § 3.8 years (range 4 to 15.3 years) after the initial correction (Table 2). The mean body weight at birth was 3.2 § 0.4 kg (range 2.6 to 4.5 kg). The mean AP foot length was 7.7 § 0.8 cm (range 6.1 to 10.2 cm), and the mean lateral foot length was 7.3 § 0.9 cm (range 5.5 to 9.9 cm). The foot length, calculated using the average of AP and lateral foot lengths, was 7.5 § 0.8 cm (range 6.2 to 10.0 cm). Comparisons Between the Subgroups Subgroup analyses were performed between the 18 DF and the remaining 29 feet. Initial total Dimeglio score (14.9 § 2.5 [vs 13.5 § 2.3, p = .042]), its subcomponent of equinus (3.8 § 0.4 [vs 3.3 § 0.6, p = .005]), and the number of casts (6.3 § 2.0 [vs 4.9 § 1.2, p = .009]) were greater in the 18 DF. In addition, the foot length was shorter (7.2 § 0.6 cm [vs 7.7 § 0.8 cm, p = .013]) in the 18 DF. Otherwise, body weight at birth was not statistically different between the 2 groups (3.2
3
§ 0.4 and 3.3 § 0.3 kg, p = .504]). Other indicators, including the 4 radiographic angles at the last follow-up, were not statistically different. All except 1 of the 18 DF required further surgery after the initial correction. The ratio of the need in the DF was significantly higher than that in the remaining feet (94.4% [vs 62.1%, p = .017]) (Table 3). In addition to the occurrence of difficulties, the 47 feet were categorized according to the number of casts (>5 [the median in the 47 feet] vs ≤5). As a result, the ratio of the need for further surgery was not significantly associated with the number of casts (p = .288, obtained using Fisher exact test).
Logistic Regression Analysis for the Occurrence of Difficulties Before regression analysis, the associations among initial total Dimeglio score, its 4 subcomponents, foot length, and body weight at birth were analyzed by using the correlation analysis. The foot length showed weak associations with total Dimeglio score and its 4 subcomponents (r = −0.233 to −0.024). The body weight at birth also showed weak associations with total Dimeglio score and its 4 subcomponents (r = −0.166 to 0.094). Total Dimeglio score and the subcomponent of equinus were strongly associated (r = 0.687). The foot length and body weight were positively but weakly associated (r = 0.286) (Table 4). On the basis of the above-mentioned results of comparisons between the 18 DF and the remaining 29 feet and the correlation analyses in the 47 feet, we designed a regression model to describe the significant factors for the occurrence of difficulties by using the subcomponent of equinus and the foot length as independent variables. As a result, both independent variables showed statistical significance. In other words, more rigid equinus and shorter foot length were significantly associated with the occurrence of difficulties after each other’s adjustments (Table 5). We limited further description using a
Table 1 The details of serial casting (47 feet in 47 patients) No. of Casts
No. of Feet
Dimeglio Score
Difficulties
Achilles Tenotomy
≤ 4a
14
4/14 (28.6%)
All
5
18
3/18 (16.7%)
16/18 (88.9%)
6
7
5/7 (71.4%)
6/7 (85.7%)
≥ 7b
8 Total 47
13.9 § 1.8 (range 10 to 17) 13.7 § 2.8 (range 10 to 18) 14.1 § 2.2 (range 11 to 17) 14.9 § 3.0 (range 11 to 18)
6/8 (75.0%) Total 18/47 (38.3%)
All Total 44/47 (93.6%)
a 3 casts in 2 feet; 4 casts in 12 feet. b 7 casts in 1 foot; 8 casts in 1 foot; 9 casts in 5 feet; 10 casts in 1 foot.
Table 2 Details of further treatment after the initial correction (47 feet in 47 patients) Contents
No. of Feet
Time of Surgery
Additional serial casting Additional Achilles tenotomy
2 (4.3%)d 9 (19.1%)
N/A At 12.0 § 9.3 months (range 3.6 to 35.4)
Extensive soft tissue release
9 (19.1%)
a
SSTR § TA tendon transfer § other procedures Awaiting further surgeryb No further treatmentc
21 (44.7%) 7 (14.9%) 12 (25.5%)
At 1.6 § 1.1 years (range 0.8 to 4.3) At 6.0 § 1.6 years (range 4.0 to 11.2) N/A N/A
Abbreviations: SSTR, selective soft tissue release; TA, tibialis anterior; N/A, not applicable. a Performed in 18 feet. b Patients who rejected the recommendation for further surgery, as well as those awaiting further surgery, were included. c Patients awaiting further surgery were not included. d Two feet underwent additional surgery later.
Table 3 Comparisons of clinical characteristics and radiographic evaluation between the 2 groups (47 feet in 47 patients) Indicators
Feet with difficulties (n = 18)
Feet without difficulties (n = 29)
p
Sex (male:female) Bilaterality Side (right:left) Total Dimeglio score Equinusa Varusa Medial rotationa Forefoot adductiona Posterior creaseb Medial creaseb Cavusb Poor muscle conditionb No. of casts Age at the first casting (days) Foot length (cm) Body weight at birth (kg) Noncompliance on bracec Follow-up period (years) Need for further surgery AP TC angle ( )d Lateral TC angle ( )d AP TM angle ( )d Lateral TM angle ( )d
12:6 9 9:9 14.9 § 2.5 3.8 § 0.4 2.7 § 0.8 2.7 § 0.8 2.4 § 0.9 18 15 17 9 6.3 § 2.0 7.5 § 4.9 7.2 § 0.6 3.2 § 0.4 3 8.9 § 3.8 17 20.0 § 7.6 31.1 § 10.4 −2.0 § 8.6 −3.2 § 9.6
16:13 12 19:10 13.5 § 2.3 3.3 § 0.6 2.6 § 0.6 2.4 § 0.8 2.2 § 0.7 28 18 24 16 4.9 § 1.2 9.0 § 11.0 7.7 § 0.8 3.3 § 0.3 3 8.6 § 3.8 18 20.6 § 7.7 33.2 § 5.4 0.7 § 7.8 0.9 § 6.4
.546g .763g .365g .042e .005e .455e .266e .225e >.999g .191g .384g .771g .009e .455e .030e .504e .662g .734e .017g .743e .983e .175e .074e
Abbreviations: AP, anteroposterior; TC, talo-calcaneal; TM, talo−first metatarsal. a Subcomponent of the Dimeglio score. b Number of patients with 1 point. c Defined as reported brace wearing time of <50% of the prescribed time. d Measured on radiographs checked at the final follow-up. e Obtained using the Mann-Whitney tests. g Obtained using Fisher exact tests.
ARTICLE IN PRESS K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
4
Table 4 Correlation analysis among the total Dimeglio score, its 4 subcomponents, foot length, and body weight at birth (47 feet in 47 patients)
Total Dimeglio score Equinus Varus Medial rotation Forefoot adduction Foot length Body weight
Total Dimeglio Score
Equinus
Varus
Medial Rotation
Forefoot Adduction
1
0.687* 1
0.727* 0.450* 1
0.608* 0.203 0.324 1
0.786* 0.583* 0.476* 0.303 1
Foot Length 0.233 0.126 0.024 0.209 0.183 1
Body Weight 0.055 0.036 0.094 0.026 0.166 0.286 1
In each cell, the Pearson correlation coefficient is described. *Coefficient with moderate or strong association.
Table 5 Logistic regression analysis using the occurrence of difficulties as a dependent variable (47 feet in 47 patients) Independent Variable
Exp (B)
95% CI
p
Subcomponent of equinus (points) Foot length (cm)
7.153 0.900
1.560 to 32.794 0.812 to 0.999
.011 .047
Abbreviation: CI, confidence interval.
scatterplot to 27 feet whose subcomponent of equinus was 4 points (ie, reducibility: 90 to 45 ). Among 13 feet shorter than the median (7.4 cm), difficulties occurred in 10 feet (76.9%) (Fig. 4). Discussion The most important finding of this study was that more rigid equinus and shorter foot length at initial presentation were significantly associated with the occurrence of difficulties during the Ponseti casting in idiopathic severe clubfoot. The results of clubfoot correction using the Ponseti method have been frequently reported (1,3,4,6−8). Unlike previous studies, we analyzed the foot length as a clinical factor,
Fig. 4. Scatterplot showing the foot length (mm) and body weight (kg) in 27 feet whose subcomponent of equinus was 4 points. Fifteen feet with difficulties (black circles) and 12 feet without difficulties (with circles) are described. The median foot length (7.4 cm) is marked. Note that difficulties occurred in 10 feet among 13 feet <7.4 cm (76.9%). Otherwise, difficulties occurred in 5 feet among 14 feet >7.4 cm (35.7%).
focused on difficulties during serial casting to comprehensively assess its process, and included only severe cases of idiopathic clubfoot. These were the main achievements and strengths of this study. We believe that more comprehensive efforts to assess the process of serial casting are needed. In particular, the contents of serial casting should be assessed in addition to its duration, which was evaluated by using the number of casts in previous studies (3,5−7). As a result, it was found that not the criteria of the number of casts but the occurrence of difficulties was significantly associated with the need for further surgery. This finding suggests the clinical significance of difficulties. Regarding the foot length as a clinical factor, it was reported that clubfeet with the length of ≥8 cm had better final Dimeglio score than those with the length of <8 cm with a median follow-up of 34 months (5). This is similar to our results that shorter foot length was significantly associated with difficulties during serial casting. The adjustment for the severity, performed in our regression analysis, might strengthen our results. It is generally believed that clubfoot should be treated soon after birth (13−17). However, the upper limit age for the Ponseti method is unclear. It was reported that the treatment using the Ponseti method initiated between 28 days and 3 months showed a lower rate of relapse and better clinical outcome compared with early treatment, before 28 days, with a mean follow-up of 5 years (8). On the basis of our results concerning foot length, the growth of the foot in the delayed group might result in more effective casting. Further study is needed to verify. In this study, the foot length was measured using simple radiographs. Due to the beam inclination, there might be a difference between the measurement and real length. In addition, the measurement might be influenced by foot position when taking the radiographs. To include the reliable value, the foot length was calculated using both AP and lateral views. Regarding the reliability, AP and lateral foot lengths were strongly associated (r = 0.870). Moreover, the ICC, calculated using the 2 values, showed excellent reliability (ICC = .930, 95% confidence interval: 0.874 to 0.961). Even if the subjects were limited to very severe cases (Dimeglio score of ≥15), the association was strong (r = 0.838), and reliability was excellent (ICC = .904, 95% confidence interval: 0.763 to 0.961). We believe that the value used in this study could at least indicate the ranking of the real length. Ponseti et al (18) described the complex clubfoot, which is distinguished by rigid equinus with a deep crease above the heel, significant shortening, a deep plantar crease and high cavus, and severe plantar flexion of metatarsals. The above-mentioned characteristics of complex clubfoot were consistent with those of the DF in this study. In addition to rigid equinus and short length, almost all of the DF scored additional points of posterior crease and cavus in the Dimeglio scoring system. We believe that a distinct entity of idiopathic clubfeet distinguished from the typical idiopathic clubfeet exists and that rigid equinus and short length are its noticeable characteristics. Orthopedic surgeons should pay special attention to feet with these characteristics. Our findings of the need for further surgery due to residual or recurrent deformity suggest that DF would have a relatively more complicated course. Of the 47 feet, further surgery was needed in 35 feet (74.5%)
ARTICLE IN PRESS K.R. Ko et al. / The Journal of Foot & Ankle Surgery 00 (2019) 1−5
during the periods of 8.7 § 3.8 years. Comparing this ratio to previous study with reported Dimeglio score and similar periods, it was reported that 68% of patients with the mean initial Dimeglio score of 13.0 § 2.0 had recurrence during 6.8 § 1.4 years (4). Our somewhat higher ratio was explainable by the mean initial Dimeglio score of 14.0 § 2.5. However, heterogeneity of further treatment after the Ponseti method and a lack of a scoring system to assess clinical outcomes (4,8) were the limitations of our study, which restricted accurate comparisons. There were several other limitations. First, only the Dimeglio score was used to rate the severity at presentation. However, an almost perfect correlation between the Dimeglio score and the Pirani score at presentation was found (3). A lack of quantitative measurement of compliance on brace was also a limitation. We believe that the reported time is not reliable, and there is no consensus on the definition of noncompliance (19). Therefore, strict criteria for the compliance (4) and detailed data on brace-wearing time were not included in analyses. In our cohort of idiopathic severe clubfoot, the initial severity of equinus and length of the foot were significantly associated with the occurrence of difficulties during the Ponseti casting. Difficulties frequently occurred in feet with rigid equinus and small length, and further surgery was required after the initial correction in almost all of the DF. Supplementary Materials Supplementary material associated with this article can be found in the online version at https://doi.org/10.1053/j.jfas.2019.07.022.
References 1. Ponseti IV, Smoley EN. The classic: congenital club foot: the results of treatment. 1963. Clin Orthop Relat Res 2009;467:1133–1145. 2. Carroll NC. Clubfoot in the twentieth century: where we were and where we may be going in the twenty-first century. J Pediatr Orthop B 2012;21:1–6.
5
3. Lampasi M, Abati CN, Bettuzzi C, Stilli S, Trisolino G. Comparison of Dimeglio and Pirani score in predicting number of casts and need for tenotomy in clubfoot correction using the Ponseti method. Int Orthop 2018;42:2429–2436. 4. Zionts LE, Ebramzadeh E, Morgan RD, Sangiorgio SN. Sixty years on: Ponseti method for clubfoot treatment produces high satisfaction despite inherent tendency to relapse. J Bone Joint Surg Am 2018;100:721–728. 5. Iltar S, Uysal M, Alemdaroglu KB, Aydogan NH, Kara T, Atlihan D. Treatment of clubfoot with the Ponseti method: should we begin casting in the newborn period or later? J Foot Ankle Surg 2010;49:426–431. 6. Brazell C, Carry PM, Jones A, Baschal R, Miller N, Holmes KS, Georgopoulos G. Dimeglio score predicts treatment difficulty during Ponseti casting for isolated clubfoot. J Pediatr Orthop 2019;39:e402–e405. https://doi.org/10.1097/BPO.0000000000001325. 7. Gao R, Tomlinson M, Walker C. Correlation of Pirani and Dimeglio scores with number of Ponseti casts required for clubfoot correction. J Pediatr Orthop 2014;34:639–642. 8. Liu YB, Li SJ, Zhao L, Yu B, Zhao DH. Timing for Ponseti clubfoot management: does the age matter? 90 children (131 feet) with a mean follow-up of 5 years. Acta Orthop 2018;89:662–667. 9. Dimeglio A, Bensahel H, Souchet P, Mazeau P, Bonnet F. Classification of clubfoot. J Pediatr Orthop B 1995;4:129–136. 10. Fleiss JL. Statistical Methods for Rates and Proportions. Wiley Interscience, New York, 1981. 11. Gray K, Gibbons P, Little D, Burns J. Bilateral clubfeet are highly correlated: a cautionary tale for researchers. Clin Orthop Relat Res 2014;472:3517–3522. 12. Taylor R. Interpretation of the correlation coefficient: a basic review. J Diagn Med Sonogr 1990;6. 35−33. 13. Dobbs MB, Gurnett CA. Update on clubfoot: etiology and treatment. Clin Orthop Relat Res 2009;467:1146–1153. 14. Dobbs MB, Rudzki JR, Purcell DB, Walton T, Porter KR, Gurnett CA. Factors predictive of outcome after use of the Ponseti method for the treatment of idiopathic clubfeet. J Bone Joint Surg Am 2004;86:22–27. 15. Liu Y, Zhao D, Zhao L, Li H, Yang X. Congenital clubfoot: early recognition and conservative management for preventing late disabilities. Indian J Pediatr 2016;83:1266–1274. 16. Morcuende JA, Dolan LA, Dietz FR, Ponseti IV. Radical reduction in the rate of extensive corrective surgery for clubfoot using the Ponseti method. Pediatrics 2004;113: 376–380. 17. Zhao D, Li H, Zhao L, Liu J, Wu Z, Jin F. Results of clubfoot management using the Ponseti method: do the details matter? A systematic review. Clin Orthop Relat Res 2014; 472:1329–1336. 18. Ponseti IV, Zhivkov M, Davis N, Sinclair M, Dobbs MB, Morcuende JA. Treatment of the complex idiopathic clubfoot. Clin Orthop Rel Res 2006;451:171–176. 19. Zhao D, Liu J, Zhao L, Wu Z. Relapse of clubfoot after treatment with the Ponseti method and the function of the foot abduction orthosis. Clin Orthop Surg 2014; 6:245–252.