Clinical Orthopaedics and Related Research®

, Volume 472, Issue 4, pp 1329–1336

Results of Clubfoot Management Using the Ponseti Method: Do the Details Matter? A Systematic Review

Authors

  • Dahang Zhao
    • Department of Pediatric Orthopaedics, Xin-Hua HospitalShanghai Jiao Tong University School of Medicine
  • Hai Li
    • Department of Pediatric Orthopaedics, Xin-Hua HospitalShanghai Jiao Tong University School of Medicine
    • Department of Pediatric Orthopaedics, Xin-Hua HospitalShanghai Jiao Tong University School of Medicine
  • Jianlin Liu
    • Department of Pediatric Orthopaedics, Xin-Hua HospitalShanghai Jiao Tong University School of Medicine
  • Zhenkai Wu
    • Department of Pediatric Orthopaedics, Xin-Hua HospitalShanghai Jiao Tong University School of Medicine
  • Fangchun Jin
    • Department of Pediatric Orthopaedics, Xin-Hua HospitalShanghai Jiao Tong University School of Medicine
Survey

DOI: 10.1007/s11999-014-3463-7

Cite this article as:
Zhao, D., Li, H., Zhao, L. et al. Clin Orthop Relat Res (2014) 472: 1329. doi:10.1007/s11999-014-3463-7

Abstract

Background

Although the Ponseti method is accepted as the best choice for treatment of clubfoot, the treatment protocol is labor intensive and requires strict attention to details. Deviations in strict use of this method are likely responsible for the variations among centers in reported success rates.

Questions/purposes

We wished to determine (1) to what degree the Ponseti method was followed in terms of manipulation, casting, and percutaneous Achilles tenotomy, (2) whether there was variation in the bracing type and protocol used for relapse prevention, and (3) if the same criteria were used to diagnose and manage clubfoot relapse.

Methods

We conducted a systematic review of MEDLINE, EMBASETM, and the Cochrane Library. Studies were summarized according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses Statement. Five hundred ninety-one records were identified with 409 remaining after deduplication, in which 278 irrelevant studies and 22 review articles were excluded. Of the remaining 109 papers, 19 met our inclusion criteria. All 19 articles were therapeutic studies of the Ponseti method.

Results

The details of manipulation, casting, or percutaneous Achilles tenotomy of the Ponseti method were poorly described in 11 studies, whereas the main principles were not followed in three studies. In three studies, the brace type deviated significantly from that recommended, whereas in another three studies the bracing protocol in terms of hours of recommended use was not followed. Furthermore no unified criteria were used for judgment of compliance with brace use. The indication for recognition and management of relapse varied among studies and was different from the original description of the Ponseti method.

Conclusions

We found that the observed clinically important variation may have been the result of deviations from the details regarding manipulation, casting, percutaneous Achilles tenotomy, use of the bar-connected brace, and indication for relapse recognition and management recommended for the classic Ponseti approach to clubfoot management. We strongly recommend that clinicians follow the Ponseti method as it initially was described without deviation to optimize treatment outcomes.

Introduction

Congenital talipes equinovarus, or congenital clubfoot, is one of the most common congenital skeletal defects. Clubfoot can be isolated or can occur as part of other disorders such as spina bifida, myelomeningocele, or arthrogryposis. A less-invasive regimen developed by Ponseti has gained acceptance as the gold standard for clubfoot management [9, 10, 24, 29, 37, 39]. This method has been reported to be efficient and effective for clubfoot treatment because of its long-term success and ability to decrease the need for extensive corrective surgery [9, 10, 24, 29, 54]. The advantages of the Ponseti method are the high success rate and prevention and management for relapse [37, 48]. When using this method, attention must be paid to the details of manipulation, casting, bracing, and the treatment of relapses. [9, 10, 24, 29, 37, 48].

Numerous authors have reported a high initial correction rate and excellent long-term results using this method [1, 9, 10, 24, 29, 33, 39, 40, 43, 47]. However, good results were not replicated by some authors who reported higher relapse or surgical rates [7, 17, 26, 35, 36]. Although the Ponseti method is simple, careful attention must be paid to the details in this method. Failure to adhere to details, such as manipulation, type of brace, bracing protocol, and relapse management, might affect whether one obtains a good outcome.

The purposes of our study were to determine (1) what degree the Ponseti method was followed in terms of manipulation, casting, and percutaneous Achilles tenotomy, (2) whether there was variation in the bracing type and protocol used for relapse prevention, and (3) if the same criteria were used to diagnose and manage clubfoot relapse.

Search Strategy and Criteria

A manual systematic literature search of MEDLINE, EMBASETM, and The Cochrane Library without any limit for concepts “clubfoot” OR “congenital talipes equinovarus” OR “CTEV” AND Ponseti was performed by two of the authors (DZ and HL). The language was restricted to English. The search was performed independently with any disagreements in eligibility resolved by consensus discussion among all the authors. Study selection was performed in a stepwise manner, first by title, then abstract, and then full-text review. The inclusion criteria were: (1) clinical diagnosis of idiopathic clubfoot; (2) no treatment before presentation at the authors’ institution; and (3) an average followup of more than 2 years. The exclusion criteria were: (1) clubfoot with any known etiology such as spina bifida, myelomeningocele, and arthrogryposis; (2) any history of treatment before presentation at the authors’ institutions; and (3) mean followup less than 2 years. The last search was completed on July 22, 2013, and the search results on the day of submission were not changed.

Five hundred ninety-one records were identified with 409 remaining after automated deduplication was performed. Three individuals (DZ, HL, LZ) reviewed all these articles independently. First, we excluded 300 citations including 278 irrelevant studies and 22 review articles. Of the remaining 109 articles, 80 were excluded because they did not meet the inclusion criteria, met the exclusion criteria, or were conference abstracts. Seventeen of the remaining 29 studies met our inclusion criteria. Fourteen of these 17 studies [2, 1721, 23, 26, 34, 43, 44, 46, 50, 52] had usable data and were included; the three other studies [30, 31, 49] did not have usable data and were excluded. The remaining 12 studies did not indicate whether all the patients underwent any treatment before being enrolled in the studies, therefore we sent requests for supplemental data from the authors of the studies. Responses were received from the authors of five studies [1, 7, 35, 36, 45] that met the inclusion criteria. Nineteen articles ultimately met our inclusion criteria [1, 2, 7, 1721, 23, 26, 3436, 4346, 50, 52] (Table 1). These studies were summarized according to the PRISMA statement [27]. A flow diagram of the search process is shown (Fig. 1). Most identified studies were considered to be of very low-grade quality according to GRADE criteria [3] owing to their observational nature. There were five case-control studies [2, 7, 19, 21, 43], 12 case series studies [1, 17, 18, 20, 23, 26, 3436, 44, 45, 52], and only two lesser-quality randomized controlled trials [46, 50] in the 19 studies. The mean age of the patients at presentation ranged from 2 days to 8.9 years. The classification system of Diméglio et al. [11] or the Pirani system as reported by Dyer and Davis [12] were used most frequently for initial assessment of clubfoot deformity [1, 2, 7, 17, 18, 20, 23, 34, 35, 4346, 50, 52]. Four to 12 casts were required to get full correction. A percutaneous Achilles tenotomy was performed in 73% of cases. The initial correction rate was 89.2 %. The rates of noncompliance, relapse, and surgical intervention in these studies varied from 7% to 61%, 3% to 62%, and 3% to 39.4%, respectively (Table 1).
Table 1

Characteristics and findings of the studies reviewed

Study/country/year

Number of patients/feet

Age at presentation

Severity

Number of casts needed

PAT (%)

IC (%)

Brace protocol

Non-compliance (%)

Noncompliance or compliance definition

Relapse (%)

Relapse definition

Surgical (%)

Herzenberg et al. [19] USA/2002

27/34

2 months (1–3 months)

N/A

7 (3–10)

91

100

23 hours per day for the first 3 months, then nighttime only for 2 to 4 years

N/A

N/A

3

N/A

3

Ippolito et al. [20] Italy/2003

32/49

Younger than 3 weeks

All feet were Grade III (10–15) Diméglio

6

N/A*

N/A

Until patient was 4 years old

N/A

N/A

41

N/A

N/A

Segev et al. [45] Israel/2005

32/48

91% of patients were treated shortly after birth

11.9 (2–17) Diméglio

N/A

98

94

Full-time for 3 months, then at night until patient was 2 years old

N/A

N/A

N/A

N/A

6

Haft et al. [17] New Zealand /2007

51/73

15 days (6–42 days)

4.6/5.0 Polynesian/White# Pirani

4.5 (3–10)

92

100

Full-time for 3 months, followed by night and naptime wear until patient was 2 years old

49

Full-time brace use for 3 months followed by at least 9 months of night and naptime use

41

Any deformity requiring surgical intervention

27

Lourenço & Morcuende [26] Brazil/2007

17/24

3.9 years (1.2–9 years)

N/A

9 (7–12)

100

100

Full-time basis for a mean of 11.7 months (10 to 12 months)

N/A

N/A

62.5

N/A

33

Abdelgawad et al. [1] USA/2007

89/137

8.6 weeks (0.5–78 weeks)

15 (5–19) Diméglio 5 (1.5–6) Pirani

5.5 (3–10)

76.4

93.4

23 hours during first 3 months followed by night and naptime wear for 3 to 4 years

34

23 hours per day for first 3 months and subsequently for night and naptime only for 3 to 4 years

14

N/A

8

Richards et al. [43] USA/2008

176/267

All younger than 3 months

12.1 (7–19) Diméglio

N/A

73

94.4

Full-time for 3 months, then at night until the patient was 2 years old

61

Noncompliance = not wearing the brace for at least 75% of the number of hours prescribed

37

With fair (limited posterior release) or poor (posteromedial release) outcome

9.4

Sud et al. [50] India/2008

23/36

All younger than 3 months (5–90 days)

14.39 (5–19) Diméglio

6.2 (3–12)

N/A

91.7

Full-time for 2 to 3 months followed by use at night only for 2 to 4 years

N/A

N/A

21.1

Recurrence of any deformity including adduction, varus, cavus, and equinus

8.3

Avilucea et al. [2] USA/2009

100/138

37 days/31days Urban /Rural

4.8/4.7 Urban /Rural Pirani

9.3/9.3 Urban /Rural

93.4

95

Full-time use during first 3 months, then use at night and naptime

32

When orthotic use was discontinued or continued without following the recommendations

26

requiring repeat manipulation and casting or surgical intervention

N/A

Park et al. [35] Korea/2009

33/48

N/A

4.3 (3–6) Pirani

N/A

85

N/A

Full-time bracing for 3 months, then night bracing advised until the patient was 3 years old

N/A

N/A

39.4

Less than 5°dorsiflexion, apparent forefoot adduction, heel varus, forefoot supination walking

39.4

Khan & Kumar [23] India/2010

21/25

8.9 years (7.5–11.1 years)

14.2 (11–18) Diméglio

12.1 (10–14)

100

N/A

N/A

N/A

N/A

24

N/A

19

Janicki et al. [21] Canada /2011

28/39

9.7 weeks (1–30 weeks)

N/A

4.8 (3–7)

74

100

Full-time for first 3 months, then night and naptime use until the patient was 4 years old

N/A

N/A

31

Any deformity that required additional casting and/or surgical procedures

5.1

Hemo et al. [18] Israel/2011

38/57

1.7 weeks (1–12 weeks)

4.5 (2.5–6) Pirani

6.5 (4–10)

100

100

23 hours during first 3 months, then 20 hours uninterrupted wear, 18 hours after 6 months, then night use until the patient was 3 or 4 years old

N/A

Ask about the age at which bracing was discontinued

N/A

N/A

N/A

Clarke et al. [7] UK/2011

50/75

9 days (2–198 days)

84% was Grade III or greater Diméglio

N/A

N/A

N/A

N/A

N/A

N/A

32

Need for recasting or additional operative procedures after the initial treatment

N/A

Verma et al. [52] India/2012

37/55

24.8 months (12–36 months)

4.95 (3.5–6) Pirani

10 (6–12)

80

89.2

23 hours a day for 3 months then night and naptime only

N/A

N/A

27

N/A

7.3

Sætersdal et al. [44] Norway/2012

116/162

2 days (0–9 days)

4.8 (2.5–6) Pirani

7.2 (3–13)

79

N/A

N/A

7

Brace use was terminated before patient was I year old

22

N/A

3.3

Panjavi et al. [34] Iran/2012

78/129

2 weeks 3 days (1–24 weeks)

78 feet (60.5%) Grade III (severe) Diméglio

4.1 (3–7)

93

100

Full-time for 3 months, then night use for 4 years

23.3

Lack of full-time bracing during first 3 months or at night 9 months thereafter

18.6

Any return in each of the four clubfoot components

6.4

Park et al. [36] Korea/2012

52/80

All patients were younger than 2 months

N/A

N/A

80

N/A

Full-time for 3 months, then night use until patient was 3 years old

N/A

Less than 12 hours brace use the first 3 month or less than 4 hours a day thereafter

44.2

N/A

34.6

Selmani [46] Albania/2012

50/76

35.3 days (2–90 days)

5.2 Pirani

7.1 (4–12)

N/A

96

Full-time until patient was walking age, thereafter worn only at night (use shoes similar to those for Kite method by day) until patient was 4 years old

N/A

N/A

13.7

Recurrence of any deformity including adduction, varus, cavus, and equinus

N/A

PAT = percutaneous Achilles tenotomy; IC = initial correction; N/A = not available; *open heel-cord Z-lengthening and posterior capsulectomy of the ankle performed; #mean Pirani scores were 4.6 and 5.0 points for Polynesian and white babies respectively; custom-made ankle-foot orthosis used; data presented by Urban and Rural groups respectively; 96.6% (112/116) was applied the first cast on the second day of life (range, 0–9 days); study compared ankle-foot orthosis with foot abduction orthosis to prevent relapse after Ponseti manipulation and casting, the values are given as foot abduction orthosis group.

https://static-content.springer.com/image/art%3A10.1007%2Fs11999-014-3463-7/MediaObjects/11999_2014_3463_Fig1_HTML.gif
Fig. 1

A systematic search strategy was used to identify the studies included in our review as summarized in this flow diagram.

Results

The details of manipulation, casting, or percutaneous Achilles tenotomy were poorly described in 11 of the 19 studies [7, 17, 18, 20, 21, 23, 26, 35, 44, 50, 52] in which the Ponseti method was used. The core principles and details were not followed in three studies [20, 36, 50], such as conducting open heel cord Z-lengthening and posterior capsulectomy of the ankle [20], pushing the calcaneus into valgus [36], and considering no correction in a patient by 1 year old as failure [50].

The bar-connected brace was prescribed in seven studies as suggested by the Ponseti method [2, 19, 21, 34, 46, 50, 52]. In four studies, the type and design of brace were partly or completely different from those suggested in the original method [20, 21, 26, 44]. In 15 studies, the brace protocols were described precisely [1, 2, 1719, 21, 26, 3436, 43, 45, 46, 50, 52], however, in three of these studies, the protocols were presented differently from the others [18, 26, 46]. The criterion was described in only eight studies for the judgment of bracing compliance [1, 2, 18, 26, 34, 36, 43, 44], and the criterion in each study was different (Table 1).

The definition of relapse was introduced, but varied in eight studies [2, 7, 17, 34, 35, 43, 46, 50] (Table 1). The same indications were used for treatment of relapsed clubfoot in 10 studies [1, 2, 7, 1921, 34, 46, 50, 52], whereas different indications were used in seven studies in which the patients did not have additional manipulations and casts used [17, 23, 26, 35, 36, 43, 45]. No statement was made regarding treatment of relapsed clubfoot in two studies [18, 44]. In addition, the intraarticular surgical rate differed enormously [1, 17, 19, 21, 23, 26, 3436, 4345, 50, 52] (Table 1).

Discussion

Conservative treatment generally is accepted as the first choice for correction of congenital clubfoot, and the most popular approach is the Ponseti method which consists of serial manipulation, casting, and a percutaneous Achilles tenotomy [9, 10, 24, 29, 37, 39, 48]. The Ponseti method also provides instruction for treating relapse [9, 10, 24, 29, 37, 39, 48]. We presumed that no adherence to the details in manipulation, brace type, bracing protocol, and treatment of relapse might produce different clinical outcomes, and therefore we performed a systematic review to investigate whether different studies followed the core principles and details of the Ponseti method. Although there have been two systematic reviews of the Ponseti method and clubfoot treatment [15, 22], our review focuses on adherence of principles and technical details. We found that some details of the Ponseti method were either poorly described or not followed. Furthermore, some studies used different criteria for relapse recognition and management.

Our study has several limitations. First, not all of the 19 identified studies in the review provided all the treatment details of manipulation, casting, brace protocol, and relapse management, so we could judge only the available data. Second, treatment with the Ponseti method may vary among patients depending on age. We aimed to analyze whether the different studies applied similar core principles and details of this method.

We found the details of manipulation, casting, or percutaneous Achilles tenotomy were poorly described in 11 studies [7, 17, 18, 20, 21, 23, 26, 35, 44, 50, 52], and three studies did not adhere to the basic principles of the Ponseti method [20, 36, 50]. Some authors reported high initial correction rates using this method [1719, 21, 26, 34]. Ponseti and Smoley reported the results of treatment for 67 patients (94 clubfeet), in which the abnormal relationship between the talus and calcaneus had not been completely corrected in only five cases after initial treatment [39]. Subsequent studies had 92% to 100% initial correction rates in patients who were younger than 2 years [1, 2, 6, 8, 19, 24, 25, 29, 34]. Patients between 1 and 3 years old who were treated with the Ponseti method had an initial success rate of nearly 90% according to Verma et al. [52]. For patients with a nonidiopathic clubfoot, the Ponseti method is also effective [16]. The initial correction rates were greater than 94% when clubfeet associated with myelomeningocele or arthrogryposis were treated with the Ponseti method [4, 14, 28]. Clubfeet in patients undergoing previous posteromedial release also were responsive to treatment using the Ponseti method [13, 32]. As we understand, strict adherence to the principles and technical details is highly related to the outcome of treatment using the Ponseti method. Although the 19 studies showed similar high initial correction rates (Table 1), details of manipulation, cast molding, or percutaneous Achilles tenotomy were not described or were poorly described in 11 of these studies [7, 17, 18, 20, 21, 23, 26, 35, 44, 50, 52]. In addition, open heel-cord Z-lengthening and posterior capsulectomy of the ankle [20], the calcaneus being touched and pushed into valgus [36], and no correction in a patient by 1 year of age were taken as indications for failure [50]. Moreover, applying a cast with the patient under anesthesia or sedation [34, 52] may lead to skin irritation and ulceration because the patient cannot react when excessive manipulation is exerted. We do not know if doing stretching exercises before weekly casting [18] could improve the outcomes. In older patients, wearing a postoperative cast for 4 weeks [23] might contribute to Achilles tendon healing, however, these are not included in the basic principles of the Ponseti method.

Variation also was found for bracing compliance and adherence. During the maintenance period, a bar-connected brace or foot abduction orthosis is used to avoid relapse [29, 37, 48]. The initial data reported by Ponseti and Smoley showed that the relapse rate was 56% when the brace was worn for approximately 2 years [39]. When the foot abduction orthosis was prescribed to maintain the completely corrected foot at 60° to70° external rotation on the affected side and 30° to 40° on the normal side, and the brace protocol was changed to full-time for the first 3 months and then 12 hours at night and 2 to 4 hours in daytime until the patient was 3 to 4 years old, the recurrence rate showed a radical reduction [29]. In our review, the bar-connected brace which had the same rationale and design as suggested with the Ponseti method, was prescribed in seven studies [2, 19, 21, 34, 46, 50, 52]. In four studies, an ankle-foot orthosis that could not maintain the foot in the position with enough external rotation and dorsiflexion was prescribed [20, 21, 26, 44]. The relapse rate was reported to be as much as 62.5% when a unilateral ankle-foot orthosis was used during the maintenance phase for treatment of neglected clubfoot [26]. However, in another study in which the Ponseti method was used for neglected clubfoot but in which a foot abduction orthosis was used, the relapse rate was 24% [23], which was similar to that for neonatal patients [2, 44, 50, 52]. This suggested the foot abduction orthosis appears to be important in maintaining the correction which could not be achieved with an ankle-foot orthosis [21]. Moreover, we found the brace protocols in the other three identified studies with higher relapse incidences were similar in that they advised only night use until the patient was 2 or 3 years old [35, 36, 43], which indicated recurrence could be associated with an insufficient time wearing the foot abduction orthosis [10, 24, 29, 3739, 53].

Noncompliance in wearing the foot aduction orthosis was reported to be the leading cause of relapse [1, 2, 5, 6, 10, 17, 21, 25, 34, 41, 42, 52]. The noncompliance rates reported in three studies were 49%, 41%, and 36%, while the patients who did not adhere to the bracing protocol were five, 183, and 120 times, respectively, more likely to have a relapse in comparison to children who wore the brace as prescribed [2, 10, 17]. Other authors suggested that noncompliance was associated with a 17 times greater odds of relapse compared with compliance [29]. Although definite brace protocols were prescribed in 15 studies [1, 2, 1719, 21, 26, 3436, 43, 45, 46, 50, 52], the majority of studies we identified did not emphasize the definite age of the patients when the foot abduction orthosis was terminated or the exact hours of night and naptime (some advised night use only) use after the first 3 months [1, 2, 17, 1921, 35, 36, 4345, 50, 52] (Table 1). Ambiguous brace protocols might result in significant deviations when parents follow undefined instructions and make it difficult for the pediatric orthopaedic surgeon to ensure whether a patient is compliant with wearing the foot abduction orthosis. The noncompliance rates for the foot abduction orthosis from the studies we identified in this review were remarkably different from each other [1, 2, 17, 19, 29, 51, 52]. Beyond the effect of different bracing protocols, the significant variations in noncompliance rates may be attributed to the lack of unanimous criteria to judge noncompliance [1, 2, 17, 18, 34, 36, 43, 44] (Table 1). Unfortunately, only eight studies had criteria for brace compliance judgment which were totally different from each other [1, 2, 18, 26, 34, 36, 43, 44] (Table 1). Apart from this, inappropriate wear of the brace also could be considered nonadherence, because the brace cannot maintain the correction effectively [53]. Compliance wearing a foot abduction orthosis does not mean proper application of the brace. Only adherence with brace use and wearing it for enough time can effectively prevent relapse.

The definition of recurrence varied among studies, which reported a wide range of relapse occurrence. Recurrence was defined by Morcuende et al. [29] and Ponseti [37] as the reappearance of any of the components of the deformity. Eight studies identified in our review defined relapse of clubfoot deformity and all defined relapse the same as in the original description of the Ponseti method [2, 7, 17, 34, 35, 43, 46, 50], except for one study in which relapse was defined as a fair or poor outcome [43]. Relapses should be treated with a second series of manipulations and casting with or without percutaneous Achilles tenotomy, and the tendon of the tibialis anterior muscle should be transferred to the third cuneiform in case of dynamic supination of the forefoot during the swing phase of gait [37, 48]. If a patient has a second relapse, the same method of treatment was suggested [37, 48]. In 10 studies, the same indications were used for relapse management as suggested for the Ponseti method [1, 2, 7, 1921, 34, 46, 50, 52] (Table 1), whereas seven studies did not recommend additional manipulation and casting [17, 23, 26, 35, 36, 43, 45] for relapse management. Percutaneous Achilles tenotomy and tibialis anterior tendon transfer are extraarticular procedures which are much less invasive to the foot than release of the tarsal joints. These two procedures were thought to be part of the Ponseti method [37, 48]. Five studies identified in our review had higher surgical rates [17, 26, 35, 36, 43] because of not applying repeated manipulation and casting [17, 35, 36], prescribing an ankle-foot orthosis for older children after initial correction [26], or not considering tibialis anterior tendon transfer as a part of the Ponseti method in treating relapse [43].

Although the Ponseti method has been accepted as the primary means for clubfoot management, there are substantial deviations in the details from different studies. We presumed that the differences in reported clinical outcomes might be attributed to the failure in following the details of the original descriptions of the Ponseti method, especially regarding manipulation, type of brace, brace protocol, and relapse management. We believe that it is important for clinicians and researchers either to adhere to the details and principles of this method, or to compare newer approaches that deviate from the original approach. Consensus is needed through scientific research, communications, and training workshops. We believe that the outcome of clubfoot management can be optimized through this approach when every core detail of the Ponseti method is followed.

Acknowledgments

We thank David A. Spiegel MD and Monica P. Nogueira MD for their input in the discussion and communications which encouraged us to do this investigation.

Copyright information

© The Association of Bone and Joint Surgeons® 2014