Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Level of evidence

Comments

Ponseti versus PMR (comparative studies) 2 RCTs / 3 cohort studies

Church et al, 2012

Retrospective cohort; 2 cohorts compared

Group-1 (PMR)

N=26 patients

N= 43 clubfeet

73% male

Age 9.2 ± 1.3 y

Group-2 (Pons)

N=22 patients

N= 35 clubfeet

41% male

Age 6.3 ± 1.4 y

Time period =

1999 – 2003 (initial treatment)

Setting/country:

hospitals; USA

Source of funding, declaration of proprietary or financial interest: no conflicts of interest or sources of funding

Inclusion criteria:

diagnosis of isolated clubfoot treated from birth and minimum follow-up (post treatment) of 5 years; time period 1999–2003

recruited at 2 sites: Ponseti site versus surgical site; all children who fitted the criteria were contacted

Two groups were age-matched but not matched according to severity (pre-operative data were not available for operative site)

Exclusion criteria:

None mentioned

Groups comparable at baseline? No; possible difference in severity; age range similar but mean age differs; treated at 2 different sites

Diagnosis: based on

absence of other identified diagnoses; only idiopathic clubfeet included

I: Ponseti

N=22 patients

N= 35 clubfeet

Ponseti

Ponseti with average 5 casts for 5 weeks; foot abduction brace (FAB); 18/22 (82%) had percutaneous Achilles tenotomies

Relapse requiring surgery: 3/35 (9%; 2 with TATT, 1 TAL) ; total relapse, requiring additional casting or surgery 5/35 feet (14%)

C:Surgical (PMR)

N=26 patients

N= 43 clubfeet

In addition, normative data were collected from a convenience sample of subjects with no reported foot deformity, motor disorder, or pain, and no observable gait deviations

Surgical (PMR)

Initially casted (3–4 casts), and then traditional PMR (Turco); casted with pins for 4 weeks and followed up with splinting

Relapse requiring surgery: 14/43 feet (33%) required repeat surgical intervention 11/14 feet (79%; 26% of all clubfeet) required major surgery

Minor = joint-sparing; tendon lengthenings and transfers

Major = joint-opening procedures or boney realignment procedures

Endpoint of follow-up: minimum 5 years

Group-1 (PMR)

Age at operation

10 months (range, 2-33 m);

Age at end follow-up 9.2 ± 1.3 y

Group-2 (Pons)

Treated from birth; Age at end follow-up 6.3 ± 1.4 y

For how many participants were no complete outcome data available?

Not mentioned (only patients with ≥5 years follow-up were included ..)

Reasons for incomplete outcome data described?

Not relevant

Significant differences between groups?

Not relevant

Treatment quality; gait, pedobarograph, physical examination,

questionnaires, radiographs; relapse (requiring surgery)

Gait: kinematic analysis; multi-segment foot model

Pedobarograph: heel pressure and medial and lateral column pressures

Physical examination:

passive ROM, flexibility, static rotational alignment, manual muscle-testing strength

Questionnaires:

pediatric Outcomes Data Collection Instrument (PODCI; normal=100; higher=better); Activities Scale for Kids (ASK; normal=100; higher=better); clubfoot disease-specific instrument (DSI; 10-40; higher=better QOL); Diméglio scale (0-20; lower=better)

Radiographs: standing anteroposterior and lateral radiographs (: angles)

Gait

 Pedobarograph

Physical examination

Radiographs

Ponseti group had significantly better plantarflexion and dorsiflexion range, greater push-off power, less residual in-toeing; PMR group had significantly increased varus and decreased medial foot pressure, greater stiffness in dorsiflexion, plantarflexion, ankle inversion, and midfoot abduction and adduction; greater equinus and cavus and increased foot internal rotation profile (for details see full text)

Questionnaires

median (IQR)

PODCI – global funct.

I: 99 (5)

C: 97 (8)

P=0.008

ASK

I: 98 (6)

C: 93 (8)

P= NS

DSI

I: 12 (3)

C: 16 (6)

P<0.001

Diméglio scale

I: 4 (3)

C: 15 (7)

P<0.001

Relapse

Requiring surgery

I: 3/35 feet (9%)

C: 14/43 (33%)

11/14 relapsed feet in C (PMR) required major surgery (26% of total); in I (Ponseti), only minor surgery (2 TATT and 1 TAL)

Low number of patients; intermediate (no long-term) follow-up; retrospective comparison of 2 cohorts recruited and treated at different sites (potential confounding); groups differ (age at end of follow-up); functional evaluation

No reasons to upgrade the level of evidence

See GRADE-profile per intervention (comparison) and outcome measure (taking all relevant studies together)

Outcome measure = Treatment quality

Level of evidence = LOW

Authors conclude that Ponseti-treated group required fewer repeat surgical interventions, walked with a better quality of gait, showed better static alignment and foot mobility, and had better global functioning; residual deformity persists, but is much less severe in the Ponseti group

Note: almost no baseline characteristics of the treatment groups are mentioned (considerable risk for confounding), and no data on brace compliance

Intermediate length of follow-up, i.e. not into skeletal maturity

Halanski et al, 2010

Prospective cohort

N=55 patients

N= 86 clubfeet

Time period =

Nov 2001 – Jan 2005

Setting/country:

National children’s hospital; New Zealand

56% bilateral; 69% male; 60% Polynesian; 47% with pos family history; average Pirani score of 5.2; average follow-up of 3.7 y (range 2.2-5.7 y)

Source of funding, declaration of proprietary or financial interest: no external funding

Inclusion criteria:

newborns referred for clubfoot treatment; idiopathic clubfoot; only patients with ≥2 years follow-up were included in analysis

Families chose randomization (9 families) or selected their choice of treatment (majority) based on information supplied by clubfoot nurse coordinator, handouts and websites; data analyzed together (prospective cohort); clinical data collected prospectively; the two treatments managed by two separate surgeons (one favoring Ponseti, the other favoring surgical treatment)

Exclusion criteria:

None mentioned

Groups comparable at baseline? Yes; with respect to sex, ethnicity, family history, bilaterality, mean age at 1^st cast, and Pirani score

Diagnosis: not specified; only idiopathic clubfeet included

I: Ponseti

N= 26 patients

N= 40 clubfeet

Weekly intervals with above-the-knee casting (Ponseti); followed by percutaneous Achilles tenotomy and final cast for three weeks; abduction orthosis (open-toed, high-top shoes; M.J. Markell Shoe) attached at shoulder width Denis Browne bar

Parents instructed to ensure full-time brace wear for 3 months, followed by night and naptime wear until at least age of 2 years

Compliance defined as full-time brace wear for 3 months followed

by at least 9 months of night and naptime use

Initial percutaneous Achilles tenotomy was not considered a primary surgical procedure but part of Ponseti method

Characteristics (mean)

age at 1^st cast = 21 d

number of casts = 6

95% of clubfeet had Achilles tenotomy

age at tenotomy = 2.4 m

C:Surgical (PMR)

N= 29 patients

N= 46 clubfeet

Feet treated with below-the knee plaster casts with upward directed

pressure to progressively correct the equinus deformity; changed weekly or biweekly until six months of age

When indicated (subjective; by surgeon; 93% of the cases), surgery was scheduled; majority underwent posterior or posteromedial release and subsequent cast changes (Cincinnati incision); typically one subsequent cast or splint change; after final cast or splint removal, patients followed clinically without additional bracing; patients in whom deformity was corrected after below-the-knee casting alone were managed with a similar orthosis as those in the Ponseti group

Characteristics (mean)

age at 1^st cast = 18 d

number of casts = 13

age prim. surgery = 6.7 m

prim. surgery = 43 (93%)

of which, major = 42 (91%) with posterior release (11) or posteromedial release (31)

Endpoint of follow-up: minimum 2 years; average of 3.7 years

For how many participants were no complete outcome data available?

Not mentioned (only patients with ≥2 years follow-up were included ..)

Reasons for incomplete outcome data described?

Not relevant

Significant differences between groups?

Not relevant

Treatment quality; proportion (%), p-values

Treatment failure, defined as surgical recurrence (minor / major); complications; risk factors (multivariate model)

Surgical recurrence: recurrence of foot deformity requiring operative interventions (note that initial surgery in control group is not considered a surgical recurrance)

Severity of surgical recurrence:

Minor= extra-articular soft-tissue procedure such as open Achilles tendon lengthening;

Major= intra-articular surgical procedure, including corrective osteotomies

Revision surgery= feet undergoing an initial surgical procedure that subsequently were found to have a recurrence requiring surgery

Surgical recurrence

Feet (%) / patients (%)

Total recurrence

I: 15(38%) / 12(46%)

C: 14(30%) / 8(28%)

P: 0.5 (NS) / 0.17 (NS)

NS: non-specific

Minor recurrence

I: 11(28%) / 9(35%)

C: 2(4.3%) / 2(6.9%)

P: 0.003 / 0.03

Major recurrence

I: 4(10%) / 3(12%)

C: 12(26%) / 7(24%)

P: 0.003 / 0.03

Revision surgery

I: 1(2.5%) / 1(3.8%)

C: 14(30%)* / 8(28%)

P: <0.001 / 0.03

* one foot operated twice and counted twice

Model for recurrence

logistic regression

Family history:

OR= 3.81 [1.11; 13.04]

Other variables: treatment type, race, Pirani, statistically non-significant (only p-values given; P>0.15)

Separate model for Ponseti to test influence of brace compliance (non-significant; P=0.67)

Post-operative complications

I: 2 patients

C: 2 patients

P: NS

Post-operative URTI, cellulitis, wound slough and / or infection

Patients with < 2 years follow-up were excluded (selection bias); there were no set criteria for defining an operative recurrence (potential bias); the minor/major surgical recurrence classification system is biased to the Ponseti method because the surgeon attempted to avoid intra-articular surgery unless absolutely necessary (and this was not the case for the surgeon of the surgical group; potential selection bias); low number of patients; no long-term follow-up; no functional evaluation

No reasons to upgrade the level of evidence

See GRADE-profile per intervention (comparison) and outcome measure (taking all relevant studies together)

Outcome measure = Treatment quality

Level of evidence = LOW

Authors conclude that both treatments had a relatively high recurrence rate, but that Ponseti group was managed with less operative intervention and required less revision surgery

A recent parallel study shows that Ponseti is more cost-effective than the surgical method (Halanski et al, 2009)

The authors observed ‘poor compliance with orthotic wear’, even ‘despite lowering recommended total brace wear from the age of 3-4 years (recommended by Ponseti) to 1 year’; this lack of compliance must be taken into account in clinical decision-making

Note that difference in functional outcome between the two interventions is not known, and long-term functional data are needed

Note that this study was performed in New Zealand, and 1/5 of the population claims a Polynesian ethnic background; the estimated incidence of 6.8 clubfeet per 1000 in Polynesian populations, is considerably higher than the incidence of 1 per 1000 estimated for European populations; local orthopaedic

community believes that clubfeet in the Polynesian population may be more resistant to treatment

Note that in this study, below-the-knee casts were applied in the surgical group, which may have reduced the need for full PMR, and may also have affected the need for subsequent surgery

Note the 30% to 40% relapse rate, higher than reported in the literature; may relate to the New Zealand population

Adegbehingbe et al., 2010

RCT

Double-blind

N=105 patients

N=153 clubfeet

ITT

Time period =

Oct 2007 – Nov 2009

Setting/country:

University Teaching Hospital; Nigeria

46% bilateral; sex not mentioned; severity (Diméglio-Bensahel score; max 20):

0-5 (34%)

6-10 (33%)

11-15 (23%)

16-20 (10%); age newborns and children upto 360 months; non-diopathic (23%)

Source of funding, declaration of proprietary or financial interest: not mentioned

Inclusion criteria:

All cases of consenting clubfoot, at any age, idiopathic, syndromic/asyndromic, unilateral or bilateral clubfoot, with or without previous treatment and/or post surgical recurrence; patients with minimum follow-up of three months post last casting were included

randomization using computer-generated random numbers; assessors and patients/parents blinded to allocations; assigned to study group by chief nursing officer blinded to managing team and study protocol; consultant orthopaedic surgeon as blinded clinical outcome evaluator; orthopedic managing teams were blinded to clubfoot patient’s selection

Exclusion criteria:

patients not primarily treated at participating institution; patients

discharged against medical advice; patients with acquired clubfoot

Groups comparable at baseline? No; despite randomization, large difference in age (I versus C, 58% versus 28% with age < 6 months)

Diagnosis: not specified; both idiopathic, and syndromic/asyndromic clubfeet included

I: Ponseti

N= 55 patients

according to Ponseti; manipulation performed and retention casts changed weekly; correction of neglected idiopathic clubfeet by Ponseti with

minimal modification; foot abducted to 30-40 degrees; Equinus corrected via percutaneous tenotomy of the tendo Achilles

Locally fabricated abduction foot braces made available; patient’s compliance braces was generally good (no details provided)

Initial percutaneous Achilles tenotomy (performed in only 5.5% of patients) was considered a primary surgical procedure (: prim. outcome measure)!

C: Non-Ponseti (“PMR”)

N= 50 patients

routine clubfoot manipulation and soft tissue (tendon and ligament) stretching; mothers taught to do manipulation; Robert Jones strapping, Denis Browne splints and Kite’s methods were non-operative treatments utilized.

Clubfeet that failed ended up receiving (extensive) soft tissue/bony surgery (66% had surgery; 34 % had extensive surgery); considered a primary surgical procedure (: prim. outcome measure)!

Endpoint of follow-up: minimum 3 months after last casting (inclusion criterium)

Fup varies from <12 m to upto 120 months

For how many participants were no complete outcome data available?

None?: (but note difference in length of follow-up between groups ..)

Reasons for incomplete outcome data described?

Not relevant

Significant differences between groups?

Yes; much longer follow-up in non-Ponseti group: 84% > 12months, versus 36% in Ponseti group

Treatment quality; proportion (%), p-values

Treatment failure, defined as requiring extensive soft tissue (ST) surgery; number of casts required; costs of care; complications; recurrence; succesful treatment (Diméglio score ≤6)

Note that initial surgery (Achilles tenotomy, PMR) is part of primary outcome measure

Minor surgery = achilles tenotomy;

Major surgery = extensive soft tissue surgery, anterior tibialis tendon transfer, posterior capsulotomy, triple arthrodesis

Clubfoot treatment judged satisfactory if Diméglio-Bensahel score < 6 at the last follow-up clinic (satisfactory ‘early outcome’)

Costs: judged low when patient received max 6 casts with or without minor surgery

Extensive ST surgery

Patients (%)

I: 0/55 (0%)

C: 17/50 (34%)

P<0.000

No Surgery required

Patients (%)

I: 52/55 (94.5%)

C: 17/50 (34%)

P<0.000

Number of casts

Patients with >6 casts

I: 13 / 55 (23.6%)

C: 35 / 50 (70.0%)

P< 0.000

Satisfactory treatment

Diméglioscore≤6

I: 55 / 55 (100%)

C: 29 / 50 (58%)

P< 0.000

 Recurrence

I: 2 / 55 (3.6%)

C: 24 / 50 (48%)

P< 0.000

Treatment complicat

Ugly scar, recurr(?), blister, infection

I: 2 / 55 (3.6%)

C: 21 / 50 (42.0%)

P< 0.003

Patients with < 3 months follow-up were excluded (selection bias); Large differences between treatment groups at baseline (age) and at end of follow-up (length of follow-up); mixed cases (idipathic, non-idiopathic, neglected clubfeet); treatment in control group not well-defined (mix of non-Ponseti treatments); setting (Nigeria) has low relevance for situation in NL; low number of patients; no long-term follow-up; no functional evaluation

reasons to downgrade the level of evidence

See GRADE-profile per intervention (comparison) and outcome measure (taking all relevant studies together)

Outcome measure = Treatment quality

Level of evidence = LOW

Authors conclude that Ponseti method has reduced clubfoot surgery frequency and reduced cast utilization, resulting in reduced total care costs

Note that the difference in functional outcome between the two interventions is not known, and long-term functional data are needed

Note that this study was performed in Nigeria

The authors observed ‘maintaining correction with the foot abduction bar is paramount to succesful treatment’, and ‘relapsing cases in Ponseti group were related to delay in procurement and use of abduction foot braces’(no details mentioned)

Authors note ‘Competent

surgical management of clubfeet should yield about 90%

satisfactory results at skeletal maturity’

‘Good results imply a flexible, plantigrade foot with the ability to bear weight on the forefoot through strong gastrocneumius muscle contraction, minimal cavus and a slightly externally rotated foot-progression axis’

‘When the re-operation rate is more than 5 to10%, the surgical method employed requires re-evaluation’

Clarke et al, 2011

(re-uses data from Uglow and Clarke, 2000)

Retrospective

Cohort

Historical control (prospective)

Period-I: June 2002 – Dec 2004

N=70 patients

N=107 clubfeet

Period-II: 1988 – 1995

N=86 patients

N=120 clubfeet

Data on period-II from Uglow and Clarke, 2000

General hospital; UK

Inclusion criteria: children with clubfeet and ≥2 years of follow-up after beginning treatment, or relapse

Exclusion criteria: genetic deformity, or neurologic involvement

Groups comparable at baseline? Yes; similar Diméglio scores (but note considerable loss to follow-up; no details on these patients)

Diagnosis: not specified; idiopathic clubfeet

I: Ponseti

N= 50 patients

N=75 clubfeet

Standard Ponseti treatment, including percutaneous tenotomy (in 85% patients); ‘most patients treated within the first few weeks after birth’

Tenotomy is considered part of treatment

Same center / same surgeons

Diméglio score:

At baseline; number feet (% total feet)

1 (1.3%)   grade-1

11 (14.7%) grade-2

46 (61.3%) grade-3

17 (22.7%) grade-4

No statistically significant differences between intervention and control groups

C: surgical (PMR)

N= 68 patients

N= 91 clubfeet

Physiotherapy, passive stretching and serial plaster casts; followed by

2-stage PMR; at mean age 8.9 months; plantar medial release + posterolateral release 2 weeks later; followed by above the knee plaster casts; splints at night, usually upto age 2 years

2-stage PMR is considered part of treatment

Same center / same surgeons

Diméglio score:

0 (0%)     grade-1

11 (12.1%) grade-2

54 (59.3%) grade-3

26 (28.6%) grade-4

Endpoint of follow-up: minimum 2 years or relapse (inclusion criterium)

Fup varies from ‘2.5-year presentation period’ (Ponseti) to ‘7-year presentation period’ (PMR)

For how many participants were no complete outcome data available?

I: 11 out of 61 patients had follow-up <2 years (18%); C: 18 out of 91 feet had fup <2 years (20%); no details provided on these patients

Reasons for incomplete outcome data described?

No

Significant differences between groups?

.....

Treatment quality; rate (%), p-values

Treatment failure defined as relapse; relapse = requiring recasting or additional operative procedures (after initial Ponseti/tenotomy, or 2-stage PMR)

Relapse rate as function of baseline Diméglio grade (severity)

Overall:

I: 24/75 (32%)

C: 28/91 (31%)

P=NS

Grade-2

I: 2/11 (18.2%)

C: 0/11 (0%)

P=NS

Grade-3

I: 16/46 (36.2%)

C: 11/54 (20.4%)

P=NS (P= 0.106)

Grade-4

I: 6/17  (35.3%)

C:17/26 ( 65.4%)

P= NS (P= 0.053)

‘near significance’

All p-values non-significant

Multiple testing (not adjusted for in statistical methods); much longer follow-up period in control group (PMR); use of historical control group; small sample size

No reasons to upgrade the level of evidence

See GRADE-profile per intervention (comparison) and outcome measure (taking all relevant studies together)

Outcome measure = Treatment quality

Level of evidence = LOW

Authors conclude that the Ponseti method is as valid as the 2-stage operative method (based on a comparison of relapse rates)

Note: (1) no functionale outcomes of the treatment methods were collected; (2) no data are provided on compliance to bracing; (3) follow-up is (much) longer in the surgical group which may bias data on relapse rate

Authors suggest that surgery should still be chosen in certain circumstances, but restricted to grade 2 or 3 feet (and not be used in grade 4 feet)

Findings for grade-4 clubfeet (very severe) was near statistical significance, in favor of Ponseti treatment

Relation Diméglio score (severity) and relapse: not significant in Ponseti group, but highly significant in PMR group (P= 0.001)

Farsetti et al, 2009

Follow-up study on Ippolito et al, 2003

Not analyzed: also see Ippolito et al, 2003, studies compare PMR with a modified Ponseti method which involves mild PMR instead of tenotomy

EXCLUDED

EXCLUDED

Zwick et al, 2009

RCT

Unblinded

ITT

N= 19 patients

N= 28 clubfeet

Single center; university hospital; Austria

Inclusion criteria: otherwise healthy infants younger than 2 weeks born with

clubfoot/clubfeet and without any other congenital anomalies

Exclusion criteria: none mentioned

Groups comparable at baseline? Yes; similar Pirani scores (severity)

Diagnosis: not specified; idiopathic clubfeet

I: Ponseti

N= 9 patients (ITT)

N=12 clubfeet (ITT)

Standard Ponseti treatment, including percutaneous tenotomy (in all patients); 6 to 8 plaster casts; Orthotic management for 24 months (Tibax bar with GloboPed shoes and a splint); molded insoles when walking

Per protocol analysis (PP):

N= 7 patients

N=10 clubfeet

Parental reasons for opting out the 2 patients were discomfort of the infant in the orthoses

C: surgical (PMR)

N= 10 patients (ITT)

N= 16 clubfeet (ITT)

PMR

Plaster casts for 5.5 to 11.4

months followed by PMR; posteromedial release plaster casts after surgery for 6 months; rigid knee-ankle-foot orthoses for nighttime splinting to the age of 36 months; molded insoles when walking

No protocol violations

Endpoint of follow-up: mean 3.5 years (range 3.3-3.8 years)

For how many participants were no complete outcome data available?

None (no loss to follow-up)

Reasons for incomplete outcome data described?

Not relevant

Significant differences between groups?

Not relevant

Treatment quality; foot function (score), health status (score), radiographic outcome (angle): median scores, p-values

Foot function measured using Functional Rating System (FRS) of Laaveg and Ponseti: Satisfaction (max 20 points), Function (20), Pain (30), Position of heel when standing (10), Passive motion (dorsiflexion [5], Varus-valgus motion of heel [3], Anterior inversion-eversion of foot [2]), Gait (10), Total score (100); treatment succes defined as FRS ≥ 75 (out of 100 points)

Health status at 3.5 years of age according to PODCI score (Pediatric Outcomes Data Collection Instrument) for transfer and basic mobility, sports and physical functioning, pain and comfort, happiness, and the global function scale

Radiographic outcome measures: from radiographs, using handhold goniometer; Lateral tibiocalcaneal angle, Anteroposterior talocalcaneal angle, Lateral talocalcaneal angle

Foot function (FRS)

Passive dorsiflexion

I: 4

C: 2

P= 0.006

Varus-valgus motion

I: 2

C: 1

P= NS

inversion-eversion

I: 1.5

C: 1

P= 0.026

Total score

I: 94.5 (95 in PP)

C: 84 (84 in PP)

P= 0.002

P= 0.0001 in per protocol (PP) analysis

Health status (PODCI)

I: 94.4

C: 94.8

P= NS

Radiographic outcome

No significant differences

Similar outcomes in per protocol (PP) analysis

Randomization procedure not well-described (‘using a randomization table’); no blinding; very small number of patients; trial stopped early for ethical reasons (because of apparent efficacy in intervention group); multiple testing (significance level < 0.0004 with Bonferroni adjustment); independent outcome accessor (but not blinded?); relatively short follow-up

reasons to downgrade the level of evidence

See GRADE-profile per intervention (comparison) and outcome measure (taking all relevant studies together)

Outcome measure = Treatment quality

Level of evidence = LOW

None of the patients experienced foot pain, and all feet had a plantigrade position when standing; the FRS category for gait was comparable in both groups; health status and radiographic variables (foot angles) were also similar between groups

The Ponseti group scored significantly higher in FRS total score (in parental satisfaction, and in categories for passive motion: passive maximal dorsiflexion, and total anterior inversion-eversion mobility)

Overall, foot function can be classified as good for the control group (PMR), and excellent for the Ponseti group

Very small study; stopped early for ethical reasons before reaching sufficient power (at 23 clubfeet per treatment)

Non-comparative studies (including studies with healthy controls): Ponseti  5 cohort studies

Bor et al, 2009

Retrospective cohort

N= 74 patients

N= 117 clubfeet

Time period = 1997 – 2002 (initial treatment)

2 Pediatric orthopedic units; Israel and USA

Inclusion: 88 patients with idiopathic clubfoot between 1997 and 2002

Exclusion: 14 because loss to follow-up

74 patients (117 clubfeet); 48 boys; 43 bilateral, 31 unilateral

Graded (Pirani 6-point severity score)

Diagnosis: not specified; idiopathic clubfeet

Pirani score

6-point initial severity;

Mean (SD)

5.1 (1.3)

I: Ponseti

N= 74 patients

N= 117 clubfeet

Ponseti treatment

strict Ponseti protocol; including percutaneous Achilles tenotomy (if required; in 86% patients)

C: no controls

Endpoint of follow-up: mean 6.3 y (range 5– 9 y)

Mean age at presentation 67 days

14 patients (16%) lost to follow-up (were excluded)

Treatment success; brace use; ROM; parent’s satisfaction

Treatment succes defined as requiring no or only minor additional surgery (apart from tenotomy); minor (TATT), moderate (posterior release), major (PMR)

Good brace use defined as full-time use for >3 months followed by >9 months nighttime use (based on questioning parents )

Parent’s satisfaction using 10-item clubfoot Disease Specific Instrument (DSI)

Brace compliance

31/70 (44.3%) poor

39/70 (55.7%) good

ROM

Final dorsiflexion

97/109 feet ‘adequate’

Treatment success

99/111 (89.2%) ‘good’

Did not require surgery or only minor surgery

Parent’s satisfaction 10-item DSI

 Mean score (1=best, 4=worst)

Satisfaction index

1.2 (range 1-4)

Brace compliance and need for additional surgery

Poor compl: 45.2%

Good compl: 20.5%

P=0.03

No comparison of treatments (not rated with GRADE); association analyses were not corrected for confounders; use of parent-reported data for brace use; intermediate follow-up (not into adulthood); no functional outcome measures

Outcome measure = Treatment quality

Level of evidence = VERY LOW

Authors conclude that 89% achieved a good outcome (requiring no additional surgery or only minor additional surgery), and that parent’s satisfaction is high

In addition, good brace use (compliance) was associated with a reduced need for additional surgery

Changulani et al, 2006

Retrospective cohort

N= 66 patients

N= 100 clubfeet

Time period = Nov 2002 – Nov 2004

Children’s hospital; UK

Inclusion criteria: 66 consecutive children with clubfeet; Referred from maternity units, general practitioners, and other orthopaedic centres

Often (32%) after other methods of treatment

had resulted in little or incomplete correction

Exclusion: children with other congenital deformities, syndromes or neurological causes of club feet

I: Ponseti

N= 66 patients

N= 100 clubfeet

Ponseti treatment

Ponseti; as soon as possible after birth; aboveknee plaster cast at weekly intervals for 4-12 weeks; Achillis tenotomy if necessary (85%); foot-abduction orthosis (Denis Browne splint), full-time 3 months followed by night- time up to 4 years

C: no controls

Endpoint of follow-up: mean 18 months (range 6-30 months)

Mean age at presentation 12 weeks (range 1-60 weeks)

No loss to follow-up

Treatment success (Pirani score); non-response; relapse; compliance

Poor compliance defined as failure to use splint continuously for at least 10 hours a day (parents answer)

Non-response

4/100 (4%) required PMR (‘resistant clubfeet’)

96(96%) clubfeet were responsive

Relapse

31/96 (32%)

16/96 repeat casting

15/96 (16%) PMR

Compliance

21/31 (68%) of relapsed clubfeet had poor compliance (no statistics mentioned)

Pirani (mean[ range])

Before/after follow-up

Hind-foot contracture

2.5[2-3] / 0.5[0-1]

Mid- foot contracture

2.5[2-3] / 0

Total Pirani score

5.0[4-6] / 0.5[0-1]

No comparison of treatments (not rated with GRADE); use of parent-reported data for brace use; short follow-up; small study; no functional outcome measures

Outcome measure = Treatment quality

Level of evidence = VERY LOW

Authors conclude that in more than 80% of feet, the deformity was corrected avoiding PMR

Percutaneous tenotomy required in 85%; damage to neurovascular structures in 1 child

Most cases of relapse seem related to poor compliance: parents

reported difficulty in keeping foot in splint; some found it difficult to apply the splint single-handed and others reported that their child was unable to kick the legs freely and was unable to roll over in bed; these problems caused frustration in the family and often sleepless nights leading to poor compliance

Ponseti et al, 2006

Retrospective cohort

N= 50 patients

N= 75 clubfeet

Time period = Oct 1992 – Feb 2004

Orthopaedic dept; multicentre (USA, UK, Bulgaria, Germany)

Inclusion criteria: congenital idiopathic

clubfoot treated Oct 1992 – Feb 2004; with complex clubfeet

complex clubfeet:

in severe equinus and supination, short and stubby, with adducted metatarsals in plantar flexion and deep creases across sole of foot and above heel

Exclusion: patients

with arthrogryposis, syndromic, or neuromuscular clubfeet

762 patients; 50 patients with 75 complex clubfeet; 31 (62%) male; 31 (62%) had initial (plaster cast) treatment elsewhere with 11 (22%) percutaneous Achillis tenotomy

I: Modified Ponseti

N= 50 patients

N= 75 clubfeet (complex)

Modified Ponseti

Requires precise identification of subtalar joint and head of talus; plaster splint applied over calf, heel, and sole; knee immobilized in at least 110° flexion by plaster splint; patients required average of 5 casts (range, 1–10 casts) for full correction; all patients had tendo Achillis tenotomy*

*in previously treated patients, a 2^nd tenotomy was not performed if foot could be dorsiflexed few degrees above neutral (5°)

C: no controls

Endpoint of follow-up: mean 20 months

Mean age at presentation 3 months weeks (range 1-9 months)

Mean age at end of follow-up 23 months (6-46 months)

No loss to follow-up

Treatment success: correction of deformity components, degree of ankle dorsiflexion, need for extensive soft tissue releases, relapse; complications

Correction deformity

All feet were well corrected with mean ankle dorsiflexion of 15° (range, 10°–25°)

Relapse

7/50 (14%)

2/50 (4%) 2^nd relapse

all relapses related to problems with shoe fit

with standard foot abduction brace

No relapses occurred after using a new, premolded foot-ankle abduction brace

Relapses treated succesfully with 2^nd series manipulations

and casting, and 3 patients with 2nd tenotomy

Complications

11 (22%) : erythema,

slight swelling forefoot and toes, mild rocker-bottom deformity, midfoot hyperabduction, downward cast slippage

No infections or profuse bleeding observed after the tenotomy

No comparison of treatments (not rated with GRADE); concerns ‘complex clubfeet’ (without using severity classification system); no radiographic follow-up; short follow-up; small study; no functional outcome measures

Outcome measure = Treatment quality

Level of evidence = VERY LOW

Authors conclude that a modified Ponseti

treatment protocol for complex clubfeet successfully corrected

the deformity without the need for extensive corrective surgery

Note: short follow-up and no functional outcome measures.

Authors note that ‘natural history of relapses in clubfoot is that most (80%) will happen during the first 2 years of life’

Bracing: all relapses related to problems with shoe fit

with standard foot abduction brace

Morcuende et al, 2004

Retrospective cohort

N= 157 patients

N= 256 clubfeet

Time period = Jan 1991 – Dec 2001

Orthopaedic dept, university hospital; USA

Inclusion criteria: consecutive patients; congenital idiopathic clubfoot;under supervision of senior author (Ponseti)

157 patients; 256 clubfeet; 107 (68%) male; 128 (81%) < 6 months; 113 (73%) had initial (plaster cast) treatment elsewhere with 14 (9%) percutaneous Achillis tenotomy

95% patients came to the clinic with all of the components of the deformity uncorrected

I: Ponseti

N= 157 patients

N= 256 clubfeet

Ponseti

guidelines treatment developed by senior author (Ponseti); including simple percutaneous tendoachilles tenotomy under local anesthesia (86% if cases); all casts in all patients under supervision of senior author; foot-abduction brace (bar with shoes); full-time for 2-3 months and at night and during naptime for 3 to 4 years

Relapses after initial correction are treated with additional manipulation and serial casting in marked foot abduction; tendoachilles tenotomy is indicated when dorsiflexion of the ankle

is <15°

No severity classification is used because the deformity is not necessarily related to the resistance to correction

Number of casts required to obtain full correction used as proxy for severity of the deformity

C: no controls

Endpoint of follow-up: mean ~20 months

Mean age at last follow-up 26 months (range: 6 months to 8 years)

128/157 (81%) presented at age < 6 months

No loss to follow-up

Treatment success: correction deformity, ankle dorsiflexion, relapse; complications

Noncompliance defined when brace was not used for at least 10 hours a day (parent self-report)

Relapse defined as appearance of any of the components of the deformity, including cavus, adductus, varus, and/or equinus

Correction deformity

In all but 3 (98%);

ankle dorsiflexion posttenotomy was 20° (range: 0–35°)

The 3 patients were previously treated in

other institutions and presented with very severe deformities

Relapse

17/157 (11%)

4/157 (2.5%) required extensive corrective surgery (1 PMR, 3 posterior releases with tendoachilles tenotomy)

Relapse not significantly related to age at presentation, previous unsuccessful treatment at other institution, or number of casts required for correction (used as a

measure of severity; all P> 0.05).

Relapses were associated with noncompliance with foot-abduction brace (P= 0.001);

noncompliance

associated with a 17 times greater odds of relapse (15 of 17) compared with compliance (6 of 140; P= 0.0001)

Complications

12/157 (8%) patients had cast complication: erythema, slight swelling of toes, or downward slippage of cast; no infections, skin necrosis, or neutenotomy

No comparison of treatments (not rated with GRADE); short follow-up; relatively large study; no functional outcome measures

Outcome measure = Treatment quality

Level of evidence = VERY LOW

Authors conclude that Ponseti method is a safe and effective

treatment that

radically decreases need for extensive corrective surgery; technique can be used in children up to 2 years of age even after previous unsuccessful nonsurgical

treatment

Authors believe that the discrepancy in the results of treatment with earlier studies is attributable to deficient understanding of the nature of the deformity and the normal functional anatomy of the foot: ‘lack of understanding has led to poor corrective manipulation and casting techniques, in particular application of counterpressure on the calcaneocuboid joint during manipulation’

Authors also stress the importance of parent education to increase brace compliance (which is a key determinant of the relapse rate)

Cooper and Dietz, 1995

Retrospective cohort

N= 45 patients

N= 71 clubfeet

Time period = 1950 – 1967

(initial treatment)

N = 97 healthy control patients

Orthopaedic dept, university hospital; USA

Inclusion criteria: consecutive patients; congenital idiopathic clubfoot;treated under supervision of Ponseti

126 patients; < 4 months old when first seen; 61 could be contacted, 4 refused to collaborate, 12 were not evaluated for time of financial restrictions;

Remaining 45 patients, 71 clubfeet

 34 (76%) male; 26 (28%) bilateral clubfeet

65/71 (92%) percut. Achillis tenotomy:

5 no tenotomy,

27 tenotomy,

28 tenotomy + additional surgery

I: Ponseti

N= 45 patients

N= 71 clubfeet

Ponseti

Patients managed under direction of Ponseti; including tendoachilles tenotomy (92%); foot-abduction brace (bar with shoes); full-time for 2-3 months and at night and during naptime for 2 to 6 years

C: healthy controls

N= 97 control patients

Randomly selected from waiting room of opthalmology outpatient clinic; within age-range of clubfoot patients (patients with congenital deformity of the foot were excluded); Note: groups are not comparable In a number of (demographic) characteristics

Endpoint of follow-up: mean ~34 years

Mean age at last follow-up 34 years (range, 25-42 y)

Presented at age < 4 months

For how many participants were no complete outcome data available? 81 (62%), mainly due to loss to follow-up

Treatment success: questionnaire (pain, function, satisfaction, etc); physical examination (incl gait analysis); radiography (angles); electrogoniometric analysis; pedobaric analysis

In addition, simple association analyses performed (excellent/good outcome versus poor outcome); Note: no proper correction for confounders

Foot function questionnaire (pain, limitation of foot function) = ‘modified Laaveg and Ponseti’

I: 28 (62%) ‘excellent’

C: 61 (63%) ‘excellent’

I: 7 (16%) ‘good’

C: 21 (22%) ‘good’

I: 10 (22%) ‘poor’

C: 15 (15%) ‘poor’

No significant differences

Note: ‘mild residual deformity was common, but only minimum signs of early degenerative osteoarthrosis were found’

No comparison of treatments (not rated with GRADE); great loss to follow-up; confounding (association analyses); relatively small study; very long follow-up; functional outcome measures

Outcome measure = Treatment quality

Level of evidence = VERY LOW

Authors conclude that with the use of pain and functional limitation as the outcome measure, 78% of the clubfoot patients had an excellent or good outcome, compared to 85% of individuals who did not have congenital deformity of the foot

Based on simple (!) association analyses, the authors suggest that a sedentary occupation, and avoidance of overweight may improve overall long-term results

The functional outcome could not be predicted from the radiographic results

Notes: no data on compliance with bracing; no proper correction for confounders in association analyses

Excellent = does not limit activities of daily living, never painful

Good = occasionally limits activities, only painful after strenuous activities

Poor = limits activities, or is painful during daily activities&a