990 THE JOURNAL OF BONE AND JOINT SURGERY

Prenatal ultrasound diagnosis of club foot

OUTCOME AND RECOMMENDATIONS FOR COUNSELLING AND FOLLOW-UP

E. Bar-On, R. Mashiach, O. Inbar, D. Weigl, K. Katz, I. Meizner

From Schneider Children’s Medical Center, Rabin Medical Center and Tel Aviv University, Tel Aviv, Israel

E. Bar-On, MD, Director of Pediatric Orthopedics

D. Weigl, MD, Orthopedic Surgeon

K. Katz, MD, Orthopedic Surgeon (Professor)Pediatric Orthopedic UnitSchneider Children’s Medical Center, 14 Kaplan Street, Petah Tikva 49202, Israel.

R. Mashiach, MD, Gynaecologist

I. Meizner, MD, Gynaecologist (Professor)Ultrasound Unit, Department of Obstetrics and GynaecologyRabin Medical Center, Beilinson Campus, Petah Tikva 49100, Israel.

O. Inbar, MD, GynaecologistSackler School of Medicine, Tel Aviv University, Ramat Aviv, Tel Aviv, Israel.

Correspondence should be sent to Dr E. Bar-On; e-mail: elbar@barak-online.net

©2005 British Editorial Society of Bone andJoint Surgerydoi:10.1302/0301-620X.87B7. 16076 $2.00

J Bone Joint Surg [Br]

2005;87-B:990-3.

Received 15 October 2004; Accepted 14 December 2004

Club foot was diagnosed by ultrasonography in 91 feet (52 fetuses) at a mean gestational

age of 22.1 weeks (14 to 35.6). Outcome was obtained by chart review in 26 women or

telephone interview in 26. Feet were classified as normal, positional deformity, isolated

club foot or complex club foot.

At initial diagnosis, 69 feet (40 fetuses) were classified as isolated club foot and 22 feet

(12 fetuses) as complex club foot. The diagnosis was changed after follow-up ultrasound

scan in 13 fetuses (25%), and the final ultrasound diagnosis was normal in one fetus,

isolated club foot in 31 fetuses, and complex club foot in 20 fetuses.

At birth, club foot was found in 79 feet in 43 infants for a positive predictive value of 83%.

Accuracy of the specific diagnosis of isolated club foot or complex club foot was lower; 63%

at the initial ultrasound scan and 73% at the final scan. The difference in diagnostic

accuracy between isolated and complex club foot was not statistically significant. In no

case was postnatal complex club foot undiagnosed on fetal ultrasound and all inaccuracies

were overdiagnoses. Karyotyping was performed in 25 cases. Abnormalities were noted in

three fetuses, all with complex club foot and with additional findings on ultrasound.

Club foot is a multiplanar deformity of thelower limb with a prevalence of 1 to 3 per 1000live births.

1

Severity varies widely and includesflexible postural deformity often requiring notreatment; isolated club foot, needing castingand possible surgery, usually with a favourableoutcome or complex club foot, associated withsyndromic, neuromuscular or chromosomalconditions causing major disability.

The widespread use of ultrasonography dur-ing pregnancy and improved techniques havegreatly increased the rate of diagnosis of defor-mities. This has led to the establishment of pre-natal clinics in which expectant parents areinformed about the outcome and long-termconsequences of the condition,

2-6

thereby help-ing them to decide upon the continuation ofthe pregnancy and cope with the deformitypostnatally.

Publications on antenatally diagnosed clubfoot present a wide variation in the accuracyof ultrasonography, the percentage of com-plex cases and agreement on further investiga-tion.

7-16

The present study aimed to providefurther knowledge on these issues.

Patients and Methods

This study included all women referred to theFetal Abnormality Clinic for abnormalities

because of an ultrasonographic diagnosis ofclub foot between 1996 and 2003.

Fifty-six women (57 fetuses) were coun-selled. Four women (five fetuses) were excludedbecause of inadequate documentation, leaving52 women (52 fetuses). In five pregnancies,there was an unaffected twin.

The women underwent a mean of 2.7ultrasonographic scans (1 to 7). The womenwere identified by review of the pregnancyfollow-up notes, birth records (if the womangave birth in our centre), and the children’srecords (if the club foot was treated in ourpaediatric orthopaedic unit). Women whohad follow-up and treatment at other centreswere interviewed by telephone. Data werecollected on details of each sonogram, find-ings at birth or after abortion, subsequenttreatment of the child, and condition at latestfollow-up. Feet were classified as normal,post-deformity, isolated club foot or complexclub foot.

Statistical analysis was performed withFisher’s exact test with values for p < 0.05being regarded as significant.

Results

Club foot was diagnosed in 91 feet at a meangestational age of 22.1 weeks (14 to 35.6). We

PRENATAL ULTRASOUND DIAGNOSIS OF CLUB FOOT 991

VOL. 87-B, No. 7, JULY 2005

diagnosed 43 at the first scan, six at the second, and three atthe third.

At initial diagnosis, 69 feet (40 fetuses) were identified ashaving isolated club foot and 22 feet (12 fetuses) as complexclub foot, a rate of 23%. The diagnosis changed on furtherultrasonographs in 13 fetuses (25%) at a mean gestationalage of 25.3 weeks. In ten of these, either bilateral or moresevere club foot was diagnosed. In the remaining three, thecondition appeared less severe. Overall, eight children whowere initially classified isolated club foot were reclassifiedcomplex club foot. The final prenatal diagnoses were iso-lated club foot in 53 feet (31 fetuses), complex club foot in38 feet (20 fetuses), and one foot was reclassified as normal.

There were 42 live births (30 boys and 12 girls). Theother ten pregnancies were terminated or stillborn. Post-natal data were obtained from maternal, child and autopsycharts in 26 infants and by telephone interview in 26.

The diagnosis of club foot was confirmed postnatally in43 children, a positive predictive value of 83% (52 fetuses)for the initial diagnosis and 84% (51 fetuses) for the finaldiagnosis. The diagnostic accuracy was 63% initially and73% at final diagnosis.

Of the 40 fetuses with an initial diagnosis of isolated clubfoot, 26 had isolated club foot at birth (a positive predictivevalue of 65%). Five of the 14 misdiagnosed fetuses (13%)had normal feet, three (8%) postural deformity and six(16%) complex club foot.

The positive predictive value for the final diagnosis ofisolated club foot was 77% (24 of 31). Of the seven mis-diagnosed children, three had normal feet (10%) and four(13%) had postural deformity.

Of the 11 children with an initial diagnosis of complexclub foot, eight were confirmed at birth and of the 20fetuses with a final diagnosis of complex club foot, 14 hada complex club foot at birth. The positive predictive valuesfor the initial and final scans were 73% and 70%, respec-

tively. Five misdiagnosed children had isolated club foot atbirth and one was normal.

The difference in diagnostic accuracy between isolatedand complex club foot was not significant for either the ini-tial or final scan (Fisher’s exact test).

Ultrasound diagnosed both complex club foot and asso-ciated conditions as follows; five arthrogryposis, four mul-tiple anomalies, two anencephaly, one macrocephaly, oneventricular enlargement, one myelomeningocele, one sco-liosis and growth retardation, one leg-length discrepancyand neck widening, one hydronephrosis and small penis(Fig. 1), one congenital knee dislocation (Fig. 2), one poly-hydramnios and one Rubinstein-Taybi syndrome. The diag-noses were confirmed in 14 of the 20 cases. Three of the fivefetuses diagnosed as arthrogryposis and those with macro-cephaly and Rubinstein-Taybi syndrome were found tohave idiopathic club foot. The fetus diagnosed with ventric-ulomegaly had normal feet.

Foot

Tibia

Fig. 1

An ultrasound at 15 weeks showing a club foot deformity. Additional find-ings were a small penis and hydronephrosis; karyotype was XYY.

Anterior

Posterior

Foot 2

Foot 1

Tibia 2

Tibia 1

Femur 1

Fig. 2a

Fig. 2b

Figure 2a – An ultrasound at 31 weeks showing bilateral club feet and adislocated knee. Figure 2b – Postnatal findings included bilateral club feetand dislocated knees. The diagnosis was Larsen syndrome.

992 E. BAR-ON, R. MASHIACH, O. INBAR, D. WEIGL, K. KATZ, I. MEIZNER

THE JOURNAL OF BONE AND JOINT SURGERY

Amniocentesis and karyotyping were performed in 25pregnancies. Three pathological types were found: one tri-somy 18, one deletion on the long arm of 8q and one XYY.All three had been classified as complex club foot by ultra-sonography before karyotyping because of additional mor-phological anomalies.

Live infants were followed to a mean of 39 months (0 to99). None of the nine children born with postural defor-mity or no deformity developed late deformity.

The 29 children born with isolated club foot were treatedby serial casting. Seven underwent percutaneous heel-cordlengthening, seven a full posterior or posteromedial releaseand one had secondary surgery.

At latest follow-up, all were fully mobile. Three hadresidual deformity possibly needing additional surgery.

Of the six live infants with complex club foot, one wasundergoing serial casting, two were mobile followingextensive surgery, and three died of their underlying dis-ease.

Discussion

The prenatal diagnosis of any fetal deformity causes majoranxiety in expectant parents. Well-informed advice canimprove parental understanding of the condition and alle-viate anxiety. The differentiation between isolated andcomplex clubfoot at prenatal ultrasonographic diagnosis isvery important, as the latter may be associated with othergrave conditions, severe lifelong disability or early demise.Therefore, it is also important to inform parents, duringcounselling, of the limitations of prenatal diagnosis.

In our study, the initial diagnosis of club foot, at a meangestational age of 22.1 weeks, was confirmed at birth in83% of cases, a false-positive rate of 17%. Correspondingrates for the final diagnosis were 84% and 16%. The false-positive rate in previous studies has ranged between 0%and 24%, possibly due to the inclusion of postural deformi-ties as a positive postnatal finding.

7-16

As feet with posturaldeformity do not require treatment, we do not believe theyshould be considered as a positive finding at birth.

The reported accuracy of the specific diagnosis of pos-tural deformity, isolated or complex club foot or normalfeet varies widely between 43% and 100%, with a mean of79%.

7,9,13-16

Our rate of 73% at final diagnosis was close tothis mean. However, in earlier series, between 5% and 29%of cases of complex club foot were misdiagnosed as isolatedclub foot, whereas in our study, no case of complex clubfoot was missed, and all ultrasonographic inaccuracieswere overdiagnoses. The associated condition most over-diagnosed was arthrogryposis, with a 60% (3 of 5) false-positive rate.

The 39% rate of complex club foot in our study was rel-atively low compared to rates between 22% and 80%(mean of 62%) reported by others.

4,7,8,13,15

Our rate mayreflect the widespread use of ultrasonography in Israel,whereas studies from other areas used selected popula-tions.

Changes in diagnosis during pregnancy were addressedby Bakalis et al

7

who reported that 19% of fetuses initiallydiagnosed as isolated club foot were reclassified as complexclub foot when other defects were found on subsequentscans; another 5.5% classified as isolated club foot werefound to have complex club foot postnatally. In our study,the diagnosis changed in 25% of fetuses. Although thechange was found to be correct in seven cases and incorrectin six, continued follow-up increased predictive valuesfrom 63% to 73% and no diagnosis of complex club footwas missed prenatally. These findings emphasise the needfor sequential ultrasonograms after initial diagnosis.

The accuracy of the diagnosis of club foot by stage ofpregnancy has been the subject of several reports. Bar Havaet al

17

described a transient club foot-like deformity in earlypregnancy. Treadwell et al

15

reported a high rate of false-positive results in third trimester sonograms following ear-lier normal results in the second trimester and concludedthat positional factors may cause a false positive sonogramin the third trimester. In our study, eight of nine infantsborn with normal feet had a final diagnosis of club foot bythe third trimester sonogram; however, in all of them, clubfoot had also been diagnosed in earlier examinations atweeks 21 to 27. The other infant had been diagnosed withisolated club foot in weeks 26 and 28, but was rediagnosedcorrectly as normal at 29 and 34 weeks. We, therefore, donot believe the false-positive results were caused by tran-sient malpositioning but rather by other factors whichobscured the diagnosis.

The need for karyotyping after prenatal diagnosis of iso-lated club foot is controversial. Shipp and Benacerraf

18

found that 5.9% of 87 fetuses with isolated club foot hadan abnormal karyotype and concluded that amniocentesisis indicated after that diagnosis. Other investigators cameto the opposite conclusion, having found no pathologicalkaryotypes in those fetuses with isolated club foot whichthey examined.

16,19,20

In the present study, amniocentesis and karyotyping wereperformed in 25 pregnancies and did not yield any addi-tional information regarding club foot. The three abnormalkaryotypes found were in fetuses with additional deformi-ties identified before karyotyping and the results did notalter decision making. Also, a literature and genetic data-base search

21,22

revealed 14 chromosomal aberrations caus-ing club foot. Nine have characteristics easily identifiableon ultrasonography and the remaining four are extremelyrare. We, therefore, question the need to perform amnio-centesis and karyotyping after ultrasonographic diagnosisof isolated club foot.

No benefits in any form have been received or will be received from a commer-cial party related directly or indirectly to the subject of this article.

References

1. Barker S, Chesney D, Miedzybrodzka Z, Maffuli N.

Genetics and epidemiologyof idiopathic congenital talipes equinovarus.

J Pediatr Orthop

2003;23:265-72.

2. Foster BK, Gurness ME, Mulpuri K.

Prenatal ultrasonography in antenatal ortho-paedics: a new subspecialty.

J Pediatr Orthop

2002;22:404-9.

PRENATAL ULTRASOUND DIAGNOSIS OF CLUB FOOT 993

VOL. 87-B, No. 7, JULY 2005

3. Burgan HE, Margaret EF, Foster BK.

Prenatal ultrasound diagnosis of clubfoot.

J Pediatr Orthop

1999;19:11-13.

4. Katz K, Meizner I, Mashiach R, Soudry M.

The contribution of prenatal sono-graphic diagnosis of clubfoot to preventive medicine.

J Pediatr Orthop

1999;19:5-7.

5. Wientroub S, Keret D, Bronshtein M.

Prenatal sonographic diagnosis of musculo-skeletal disorders.

J Pediatr Orthop

1999;19:1-4.

6. Maymon R, Sharon M, Reish O, et al.

Fetal abnormalities leading to termination ofpregnancy: the experience of the Assaf Harofeh Medical Center between the years1999-2000.

Harefuah

2003;142:405-9 (in Hebrew).

7. Bakalis S, Sairam S, Homfray T, et al.

Outcome of antenatally diagnosed talipesequinovarus in an unselected obstetric population.

Ultrasound Obstet Gynecol

2002;20:226-9.

8. Carroll SGM, Lockyer H, Andrews H, et al.

Outcome of fetal talipes following inutero sonographic diagnosis.

Ultrasound Obstet Gynecol

2001;18:437-40.

9. Hashimoto BE, Filly RA, Callen PW.

Sonographic diagnosis of clubfoot in utero.

J Ultrasound Med

1986;5:81-3.

10. Keret D, Ezra E, Lokiec F, et al.

Efficacy of prenatal ultrasonography in confirmedclub foot.

J Bone Joint Surg [Br]

2002;84-B:1015-19.

11. Kevern L, Warwick D, Wellesley D, Senbaga R, Clarke NM.

Prenatal ultra-sound: detection and diagnosis of limb abnormalities.

Pediatr Orthop

2003;23:251-3.

12. Maffulli N.

Prenatal ultrasonographic diagnosis of talipes equinovarus: does it givethe full picture?

Ultrasound Obstet Gynecol

2002;20:217-18.

13. Rijhsinghani A, Yankowitz J, Kanis AB, et al.

Antenatal sonographic diagnosis ofclubfoot with particular attention to the implications of isolated clubfoot.

UltrasoundObstet Gynecol

1998;12:103-6.

14. Tillett RL, Fisk NM, Murphy K, Hunt DM.

Clinical outcome of congenital talipesequinovarus diagnosed antenatally by ultrasound.

J Bone Joint Surg [Br]

2000;82-B:876-80.

15. Treadwell MC, Stanitski CL, King M.

Prenatal sonographic diagnosis of clubfoot:implications for patient counselling.

J Pediatr Orthop

1999;19:8-10.

16. Woodrow N, Tran T, Umstad M, et al.

Mid-trimester ultrasound diagnosis of iso-lated talipes equinovarus: accuracy and outcome for infants.

Aust NZJ Obstet Gynae-col

1998;38:301-5.

17. Bar Hava I, Bronshtein M, Orvieto R, et al.

Caution: prenatal clubfoot can be bothtransient and late-onset phenomenon.

Prenat Diagn

1997;17:457-60.

18. Shipp TD, Benacerraf BR.

The significance of prenatally identified isolated club-foot: is amniocentesis indicated?

Am J Obstet Gynecol

1998;178:600-2.

19. Malone FD, Marino T, Bianchi DW, Johnston K, D’Alton ME.

Isolated clubfootdiagnosed prenatally: is karyotyping indicated?

Obstet Gynecol

2000;95:437-40.

20. Mammen L, Benson CB.

Outcome of fetuses with clubfeet diagnosed by prenatalsonography.

J Ultrasound Med

2003;23:479-500.

21. Lyon Jones K.

Smith’s recognizable patterns of human malformation

. Philadelphia:W.B. Saunders Co., 1997.

22. Buyse ML, ed.

Birth defects encyclopedia

. Dores, Massachusetts: Center for BirthDefects Information Services Inc, 1990.