7262JMed Genet 1994;31:726-730 Syndrome of the month J Med Genet: first published as 10.1136/jmg.31.9.726 on 1 September 1994. Downloaded from

Cleft /foot: clinical and developmental aspects

Paul W Buss

Abstract a useful major feature for syndrome diagnosis. Isolated limb reduction defects occur in Large families with isolated cleft limbs have approximately 1 in 2000 live births within been described which usually show autosomal which central ray anomalies are an im- dominant inheritance and marked variability portant subgroup. Most affected persons in expression and penetrance, although several have mild or moderate functional im- autosomal recessive pedigrees have also been pairment. Considerable psychological reported.8'0 Birch-Jensen" estimated the in- morbidity may also occur. While there cidence of cleft at 1 in 90 000 have been major strides forwards in our births. More recent epidemiological studies understanding of vertebrate limb de- suggest the incidence of central ray anomalies velopment, the mechanisms responsible to be closer to 1 in 10 000 (approximately 0.7 for central ray deformities remain poorly per 10 000 live births).'2 '3 The data of Czeizel understood. Several case reports ofcentral et all4 indicate that three-quarters (77%) of clefting anomalies associated with chro- cases present with unimelic involvement. mosomal rearrangements or interstitial In 1960 Stevenson and Jennings'5 suggested deletions of 7q21.2-q21.3 suggest that this an abnormal segregation pattern in families chromosomal region is important for limb with isolated (pedigree analysis development. showed an excess of affected sons of affected fathers). David,'6 when describing three ped- (J Med Genet 1994;31:726-730) igrees with isolated cleft limbs, suggested the possibility of germinal mosaicism. X linked

inheritance has also been described in a seven http://jmg.bmj.com/ While the term ectrodactyly literally means generation Pakistani kindred.'7 "missing digits", it is now used as a descriptive Ford'8 hypothesised that "ectrodactyly" rep- term for a central ray anomaly that presents resented an anomaly that could be explained clinically as a central cleft of the or feet.' by a structural chromosomal defect although Historically, terms such as "lobster claw"2 and at that time it was acknowledged that tech- split hand3 have been used to describe this niques for chromosome analysis were poor. Since then a number of case reports have in-

defect which results from absence, or re- on September 25, 2021 by guest. Protected copyright. duction, of one or more of the three central dicated that the 7q21.2-q21.3 region is a can- digits (fig 1). The earliest known description didate site for cleft limbs. 19-26 These have ofcleft hand was that by Pare4 although Barsky5 resulted in considerable molecular interest in suggested that this report represented two this region. fingered rather than true cleft hand. Central ray defects, according to Swanson's modified classification, represent terminal Clinical features longitudinal limb anomalies.6 Cleft limb has a Clinical problems associated with isolated cleft number of associations and as a result it can be hand and foot are usually cosmetic and psy-

Department of Neonatal Medicine, University of Bristol, Bristol BS2 8EG, UK P W Buss Correspondence to Dr Buss, Department of Paediatrics, Royal Gwent ~..Ag;4 Hospital, Newport, Gwent Figure 1 Typical central clefts of hands and feet; wide central clefts with normal marginal (radial and ulnar) rays. In NP9 2UB, UK. this case the child had EEC syndrome. Cleft handlfoot: clinical and developmental aspects 727 Table 1 List of conditions in which cleft limb may be a are professional artists, speed typists, and even feature (McKusick No) marathon runners may also be of considerable EEC syndrome (129900) reassurance. Cornelia de Lange syndrome (122430) J Med Genet: first published as 10.1136/jmg.31.9.726 on 1 September 1994. Downloaded from Acrorenal syndrome (102520) Radiology usually reflects the clinical find- (305600) ings but sometimes transverse metacarpals or Ectrodactyly and cleft palate syndrome (129830) Ectrodactyly/mandibulofacial (183700) proximal phalanges affect the appearance. Oc- Ectrodactyly and macular dystrophy (185800) casionally atypical cleft hand and monodactyly Non-specific associations with several anomalies are noted by coexist (fig 2). In the EEC syndrome (the Czeizel et al'4 and also with neural tube defects. ectodermal dysplasia, ectrodactyly, and clefting syndrome), preaxial anomalies also may coexist chological rather than functional. Only in the with a central cleft. While careful examination less common variant monodactylous ec- of the parents' hands may be helpful, there is trodactyly, where radial rays are also hy- little extra information to be gained from rou- poplastic, is function severely compromised. tine radiography of the hands and feet of par- Considerable anatomical variability occurs. ents who are clinically normal. Cleft hand has been described with pre- and postaxial duplication and reduction anomalies.' Absence ofthe central digits is the major feature Differential diagnosis: clinical approach with varying association with both bony and Recognition of a central ray deformity may be soft tissue . Many cases will merely a useful clue for the diagnosis of a number of have one limb affected and others can have underlying disorders (EEC syndrome, Cornelia tetramelic involvement. Birch-Jensen" found de Lange syndrome, acrorenal syndrome, etc) that 58% of cases of cleft hands were bilateral (table 1). When presented with a neonate with with an equal sex distribution. These clinical a central ray defect diagnostic difficulties may findings were confirmed by Barsky.' The ma- arise. In such cases magnification of nails and jority of cases of cleft hand are sporadic. In dental x rays may help in identifying in- Birch-Jensen's study, " 50% had a family history volvement of ectodermal structures (difficult in of limb defects while only 20% of Barsky's the neonate/infant with oligosymptomatic EEC group5 were familial. The study of Czeizel et syndrome). It is also worthwhile reviewing all4 showed a high percentage of presentation these children during the first year to confirm as atypical forms, though this is based on a normal development of ectodermal structures "wide" interpretation as to what actually con- and, if uncertain of this, electron microscopic stitutes atypical split hand/foot. The definition examination of the hair shaft may be helpful of this varies between published reports but in diagnosing ectodermal dysplasia (Buss, un- actually serves relatively little clinical purpose. published data). In the newborn it is important It is also subject to wide differences in clinical to document hair distribution and birth weight. interpretation. A consensus might be that atyp- A renal ultrasound scan may help make the ical cleft limb occurs when there is not total diagnosis of the EEC or acrorenal syndrome. absence of the long finger or when there are, In older children and adults with "isolated" in addition, reduction anomalies affecting the ectrodactyly, one often notes minor associated http://jmg.bmj.com/ marginal rays. This tends to be (though not ectodermal anomalies (of the teeth in par- always) sporadic and unilateral, affecting ticular).' Language milestones should be mainly the upper limbs. checked. Ectrodactyly associated with con- Clinically, few functional problems occur in ductive deafness and mandibular hypoplasia the upper limbs provided opposition is possible. has been described in a father and son by However when a preaxial anomaly coexists or Patterson and Stevenson28 and more recently when monodactyly (a single ) occurs, with sensorineural deafness by Raas-Rothschild marked limitation of function can result. Func- et alP although the latter report looks distinctly on September 25, 2021 by guest. Protected copyright. tional limitation can result when central clefts similar to oligosymptomatic EEC syndrome. are particularly wide or deep and operative The otic capsule is interestingly also of ec- management to close these can often result in todermal origin.30 Opitz and Frias3' described significant improvement in function. Lim- an association with cleft palate as a separate itation of function owing to syndactyly of the entity, the ECP syndrome. Karsch32 recognised fingers may also occur and this also can be overcome surgically.27 In the lower limbs the major practical difficulty encountered is one of obtaining adequately wide fitting shoes. Many persons with "ectrodactyly" ex- perience psychological problems. These often emerge in mid-childhood as a reluctance to use hands, even a need to hide them. Non-verbal communication may be affected by this and common important gestures, such as hand shaking, are often avoided. These problems have frequently not been overcome even in adult life (Buss, unpublished data). Providing adequate reassurance to parents of affected children may allow them to help their children Figure 2 Combined monodactyly with atypical cleft develop a positive body image. Anecdotal re- deformity showing some preservation of marginal rays in ports of cases with tetrameiic involvement who right hand. 728 Buss

Table 2 Review of reported cases of associated with anomalies of chromosome 7q21 Author Clinical description 7q deletionlkaryotype

Del Porto et al" Male infant, low set ears, hypotonia, beaked nose, bilateral del(7) (ql I-q2 1 or q22) J Med Genet: first published as 10.1136/jmg.31.9.726 on 1 September 1994. Downloaded from cleft foot Pfiefer et al2" Female aged 2 years, microcephaly, low set ears, mental del(7) (q I .2-q22. 1) retardation, cleft palate, R cleft hand and foot Tajara et at2' Male aged 3 years, microcephaly, anteverted nares, del(7) (qI 11.2-q22) micrognathia, retardation, clefting of three limbs Morey et at22 Premature male, R cleft hand, of lower limbs, facial del(7) (q2 1 .3-q3 1.3) dysmorphism, ASD. Died at 7 months Sharland et at2' Male infant, tetramelic ectrodactyly, low set ears, Complex rearrangement t(5;7;9) with breaks at 7q21.2 and 7q31.3 Roberts et at24 Male, tetramelic ectrodactyly, mild facial dysmorphism and Complex rearrangement mild developmental delay (intrachromosomal) del(7) (q2 1.3-22. 1) Naritomi et at2' Male, trimelic ectrodactyly, bifid uvula, high arched palate ins(3;7) (q21;q34q22). Break at 7q22q34 Genuardi et al2' 5 generation family, 7 affected, 5 male. Unimelic (3 cases), Balanced translocation t(2;7) tetramelic (4 cases). Two with sensorineural deafness. One (q2 1.1 ;q22. 1) case with tibial hypoplasia

an association with impaired vision owing to lateral cleft limbs, although three members of cataract and macular dystrophy. a family26 showed unimelic involvement. The For parents of isolated cases empirical re- majority of the reported cases have typical cleft currence risks are low but there is a potential limbs. risk to offspring of affected persons of 50%. Few other reports exist that link central ray With possible heterogeneity and altered pene- anomalies with chromosome 7. Klep-de Pater trance risks may be lower.33 Certainly isolated et al5 described a female infant with cleft palate cleft limbs can produce sohie very difficult and digital reduction deformities of the second diagnostic as well as counselling situations. and fifth digits of the right hand (not typical Constriction amputation of digits may be central cleft) and no evidence of ectodermal difficult to distinguish from unimelic atypical dysplasia.35 This case had a karyotype 46,XX, cleft limb. Clues as to a constrictive aetiology del(7) (qi 1 -÷q22). A number ofadditional case might include the observation of amniotic reports of deletions involving the 7q22 band strands at birth or lymphoedema distal to a have not been associated with ectrodactyly.36 possible constriction ring. Soft tissue syn- also occurs in constriction anomalies but is usually distal.34 In such cases no other Aetiology and developmental aspects bony anomalies are usually radiologically pres- The general association of cleft limb with an- ent and occasionally the astute neonatal res- omalies of ectodermal structures is not par- ident will have asked for the placenta to be ticularly surprising as there are similarities in examined showing an amniotic tear. Unilateral limb development to that of several ectodermal limb reduction deficits may also have a vascular structures. Wolpert's experiments which con- origin and the clinician should be aware of

firmed earlier observations of a crucial reliance http://jmg.bmj.com/ relevant points in the history, for example, of embryonic limb ectoderm on the underlying chorionic villus sampling, presence of twins mesoderm are not merely of historical import- (reversed arterial perfusion), and any oc- ance.37 He postulated the presence of a sig- currences that may have compromised fetal nalling mechanism (probably a diffusible circulation. It can be difficult to separate some morphogen) which could be interpreted as a of these potential causes and in the atypical positional signal for regional development isolated unimelic case explanation of a number within the progress zone ofthe developing limb of potential aetiologies may occasionally be bud mesoderm. Historically it has seemed likely on September 25, 2021 by guest. Protected copyright. necessary. that the endogenous signalling morphogen in the limb bud of mammals may be related to retinoic acid (possibly 3'4'didehydroretinoic Cleft hand and foot malformation and 7q acid).38 However an exciting recent report in- abnormalities dicates that a newly identified gene encoding Eight recent case reports of ectrodactyly and the signalling molecule, , may 7q21 .2-q21.3 anomalies are summarised in be involved in establishing polarising activity table 2. It is ofinterest that six of seven isolated in the developing limb bud and hence antero- cases reported are male (note the pre- posterior patterning.39 ponderance of males observed by Stephenson Developmental digit anomalies such as those and Jennings'5 and in the familial cases of seen in central ray anomalies, however, are Czeizel et all4). The significance of this is un- not easily explained by a diffusible morphogen clear. Genuardi et al6 recently reported a family model. Homeobox genes are consistently ex- segregating a balanced translocation with a pressed in the developing limb bud and digits breakpoint at 7q22. 1, in which two out of in a way that reflects their location within a nine translocation carriers had limb deficiencies homeobox cluster (temporospatial colinearity). (similar abnormalities were present in a further These genes are activated sequentially by vary- five persons in whom chromosome analysis was ing concentrations of retinoic acid within the not possible). Overall there is a wide range of limb bud. The expression of genes of the ho- clinical involvement from apparently isolated meobox 4 (HOX D) type is necessary for nor- cleft limbs to ectromelia associated with fatal mal limb development in mice4' and chick.42 anomalies. None of the isolated cases had uni- Transcript expression can be predictably Cleft hand/foot: clinical and developmental aspects 729

The author wishes to acknowledge that his work on cleft limbs altered in experimental animal models by im- was conducted as Action Research Training Fellow in Medical plantation of retinoic acid soaked beads in Genetics at the Institute of , University of Wales College of Medicine, Cardiff. strategic positions within the developing limb J Med Genet: first published as 10.1136/jmg.31.9.726 on 1 September 1994. Downloaded from bud with resultant anomalies43 including mirror 1 Tentamy S, McKusick V. The genetics of hand malformations. hands and . Central ray digit re- New York: Alan A Liss, 1978:53-63. duction anomalies have not yet been observed 2 Goldmann EE. Beitrag zur Lehre von der Misbuldingen der Extremetaten. Beitr K/in Chir 1891;7:239-56. in these experiments. It may be that the an- 3 Meller J. Ein Fall von angeborer Spaltbilding der Hande omaly represents a localised disturbance of und Fusse. Berl Klin Wochenschr 1893;30:232-3. 4 Pare A. Les oeuvres de M Ambroise Pare. Paris: Gabriel Boun, HOX expression at the ectodermal/meso- 1575:5-8. dermal junction. Recent work has shown dis- 5 Barsky A. Cleft hand, classification, incidence and treat- ment. .7 Hand Surg (Am) 1964;46:1707-19. tinct expression ofMSX1 (HOX 7) and MSX2 6 Swanson A. A classification for congenital limb defects. 7 Hand Surg 1978;1:8-12. (HOX 8) at this site in limbs4445 and teeth.46 7 Lewis T, Embleton D. Split hand and split foot deformities Additionally combined expression of both their types, origin and transmission. Biometnrika 1908;6: 26-58. HOX D and of HOX A (localised to 7pl 1) 8 Verma I, Joseph R, Bharjansa S, Sadanand M. Split hand appears necessary for organised cellular differ- and split inherited as an autosomal recessive trait. Clin Genet 1976;9:8-14. entiation in the developing limb bud. The msh 9 Zlotogora J, Nubani N. Is there an autosomal recessive form related MSX1 and MSX2 genes, although of split hand and split foot malformation? . Med eet 1988;25: 138-4i0. widely distributed in the early limb bud, are 10 Friere Maia N. Autosomal recessive form of ectrodactyly later organised at the distal part of the bud and and its implications in genetic counselling. _7 Hered 197 1; 62:53. are responsible for limb growth, possibly by 11 Birch-Jensen A. Congenital deformities of the utpper extremities. a at Copenhagen: Ejnar Munksgaard, 1949. maintaining pool of undifferentiated cells 12 Froster UG, Baird P. Limb reduction defects in over 1 the progress zone.47 Although great strides have million consecutive live births. Teratology' 1989;39: 127-35. 13 Calzolari E, Manioguigi 0, Garani GP, Cocchi G, Magnani been made in our knowledge of vertebrate C, Milan M. Limb reduction defects in Emilia Romagna limb patterning in the context of central ray Italv: epidemiological and genetic study in 173 109 con- secutive live births. .7 Med Cenet 1990;29:353-7. anomalies, it must also be conceded that cur- 14 Czeizel AE, Vitez M, Kodaj T, Lenz W. An epidemiological rently there are no obvious developmental study of isolated split hand/foot in Hungary, 1975-1984. _7 Med Genet 1993;30:593-6. genes which map to the 7q candidate region. 15 Stevenson A, Jennings L. Ectrodactyly - evidence in favour In addition there are no ectrodactylous mouse of disturbed segregation in offspring of affected males. Ann Hum Genet 1960;24:89-96. mutants which map to the homologous region 16 David TJ. Dominant ectrodactyly and possible germinal on mouse chromosome 5. It is interesting to mosaicism. _7 Med Cenet 1972;9:316-20. 17 Ahmad M, Abbass H, Haque S, Gebjard F. X-chro- note, however, that mutation of MSX2 (HOX mosomally inherited split-hand/split foot anomaly in a 8) has been associated with Pakistani kindred. Hunl Geniet 1987;75:169-73. 18 Ford CE. Autosomal anomalies. Second Coniferenice on Cotn- and limb defects (including triphalangeal genital Malfonrmationis, NY Foundation, 1963:25. 19 Del Porto G, D'Alessandro E, De Matties C, LoRe M, thumb and metatarsal shortening).48 DiFusco I. Delzione interstiziale del braccio lungo del Mice have been bred with dactylaplasia cromosoma 7 e sue correlzioni cliniche. Pathologica (Suppl) 1983;75:268-7 1. which closely resembles the human cleft limb. 20 Pfiefer RA. Interstitial deletion of chromosome 7(q 1.2 In these it has been hypothesised that the an- q22. 1) in a child with split hand/split foot malformation. Annii Genet (Paris) 1984;27:45-8. omaly occurs as a result of effects from two 21 Tajara EH, Varella-Garcia M, Gusson ACT. Interstitial long http://jmg.bmj.com/ loci.49 Whether this could be extrapolated to deletion of chromosome 7 and ectrodactyly. Ami _7 Med Genet 1989;32:192-4. human cleft limb is debatable but a similar 22 Morey MA, Higgins RR. Ectro-amelia syndrome associated explanation was put forward by David."6 The with an interstitial deletion of 7q. Ami _7 Med Geniet 1990; 35:95-9. association of cleft limb with absence of long 23 Sharland M, Patton M, Hill L. Ectrodactyly of hands and suggests may oc- feet in a child with a complex translocation including bones50 that the defect 7q21.2. AnzJ Med Genet 1991;39:413-14. casionally be a result of a common defect in 24 Roberts SH, Hughes HE, Davies SJ, Meredith AL. Bilateral regional limb specification. A child exposed in split hand and foot malformation in a bov with a de novo interstitial deletion of 7q2 1.3.3 Med Geniet 1991 28: on September 25, 2021 by guest. Protected copyright. utero to a retinoic acid derivative during the 479-81. 25 Naritomi K, Izumikawa Y, Tohma T, Hirayama K. Inverted crucial time of limb development with absent insertion of chromosome 7q and ectrodactyly. Amz . Med digits and syndactyly has been reported.5' More Genet 1993;46:492-3. 26 Genuardi M, Maria GP, Vincenza S, Bellussi A, Zollini M, recently two case reports of limb anomalies in Neri G. Split hand/split foot anomaly in a family se- a fetus and child with prenatal exposure to gregating a balanced translocation with a breakpoint on 7q22.1. Am 7Med Geniet 1993;47:823-31. isotretinoin have been reported.52 One could 27 Nutt JN, Flatt AE. Congenital central hand deficit. _7 Hand postulate that the mechanism underlying this Surg 1981;6:48-60. 28 Patterson TJS, Stevenson AC. Craniofacial dysostosis and may be a failure ofdifferentiation resulting from malformation of the feet. _7 Med Genet 1964; 1: 112-14. misexpression of homeobox genes. However, it 29 Raas-Rothschild A, Aviram A, Ben-Ami T, et al. Newly recognised ectrodactyly/deafness syndrome. _7 Craniiofac may be worth noting that other teratogens such Genet Dev Biol 1989,9:121-7. as ethanol to cause 30 Mikaelian DO, Der Kalousten VM, Shahin NA. Congenital have also been postulated ectrodactyly with hearing loss. Arch Otolaryngol 1970;92: ectrodactyly,53 in this particular case with no 85-9. 31 Opitz J, Frias JL. The ECP syndrome, another autosomal clear underlying mechanism. dominant cause of monodactylous ectrodactyly. Eur.7 Ped Evidence is clearly accumulating to make 1980j133:217-20. 32 Karsch J. Erbliche augenmissbildung in verbindung mit the 7q21.2-q21.3 locus a prime candidate for spalthand und fuss. ZAugenheilkd 1936;89:274-9. isolated ectrodactyly. While it has also been 33 Emery AEH. A problem for genetic counselling - split hand deformity. Clinl Genet 1977;12:125-7. postulated as a locus for investigation in syn- 34 Gellis S. Constrictive bands in the human. Birth Defects dromes such as the EEC syndrome54 (con- 1977;13(1):259-68. 35 Klep-de Pater JM, Bijlsma JB, Bleeker-Wagemakers EM. tiguous genes), it is likely that cleft limb Two cases with different deletions of chromosome 7. .7 represents a genetically heterogeneous con- Med Genet 1979;15:161-3. 36 Winter R, Tickle C. Syndactylies and polydactylies - em- dition and the 7q genes represent just one bryological overview and suggested classification. Eir _7 defect a common Hum Geniet 1993;1:96-104. potential affecting pathway 37 Wolpert L. Mechanisms of limb development and mal- for limb differentiation. formation. Br Med Bull 1976;32:65-70. 730 Buiss

38 Ragsdale CW Jr, Brockes J. Retinoids and their targets 46 McKenzie A, Leeming GL, Tarrett, Jowett A, Sharpe PT. in vertebrate development. Cuirr Opin Cell Biol 1992;3: Human Hox 7 in regional and temporal expression in 928-34. craniofacial and tooth development. Dcvclop;ncnt 199 1; 39 Riddle RD, Johnson RL, Laufer E, Tabin C. Sonic hedgehog 111:269-85. mediates the polarizing activity of the ZPA. Cell 1993;75: 47 Tickle C. A tool for transgenesis. Natutre 1992;358:188-9. J Med Genet: first published as 10.1136/jmg.31.9.726 on 1 September 1994. Downloaded from 1401-16. 48 Wang Jabs E, Muller I, Li X, ct al. A mutation in the 40 Lewis T. A gene complex controlling segmentation in dro- homeodomain of the human MSX2 gene in a familv sophila. Natlure 1978;276: 565. affected with autosomal dominant craniosynostosis. CclI 41 Dolle P, Izpisua-Belmonte JC, Falkinstein H, Renucci A, 1993;75:443-50. Duboule D. Coordinate expression of murine Hox 5 49 Chai CK. Dactylaplasia in mice. 7 Hcrci 198 1;72:234-7. complex homeobox containing genes during limb pattern 50 Hoyme AD, Lyons K, Nyhan W, Pauli R, Robinow M. formation. Natuire 1989;342:767-9. Autosomal dominant ectrodactylv and abnormalities of 42 Izpisua-Belmonte JC, Tickle C, Dolle P, Wolpert L, Dou- long of upper and lower limbs. _7 Pdiat^r 1987;111: boule D. Expression of the homeobox HOX 4 genes and 538-43. the specification of position in chick wing development. 51 McBride WG. Limb reduction deformities in a child exposed Nature 1991,350:585-9. to isoretinoin in utero at 26-40 days gestation. Launct 43 Bryant SV, Gardiner DM. Retinoic acid, local cell-cell 1985;i: 1276. interactions and pattern formation in vertebrate limbs. 52 Rizzo R, Lammer E, Parsano E, Pavone L, ArglNle JC. Dezv Biol 1992;152:1-25. Limb reduction defects in humans associated with prenatal 44 Benoit R, Lyons G, Simandl BK, Atshushi R, Buckingham isoretinoin exposure. Teratology 1991;44:599 604. M. The apical ectodermal ridge regulates Hox 7 and Hox 53 Hermann J, Pallister PD, Opitz JM. Tetracctrodactyly and 8 gene expresssion in developing chick limb buds. GeneS other skeletal manifestations in the foetal alcohol svn- Dev 1991;5:2363-74. drome. EiirJ7 Pecdiatr 1980;133:221-6. 45 Davidson D, Crawley A, Hill RE, Tickle C. Position de- 54 Qumiseyeh MB. EEC syndrome (ectodermal dysplasia, ec- pendent expression of two related homeobox genes in trodactyly, cleft lip and palate) is on 7q1 1.2 q21.3. Cli//u developing vertebrate limbs. Natutre 1991 ;352:429-31. Genet 1992;42:101. http://jmg.bmj.com/ on September 25, 2021 by guest. Protected copyright.