Pediatr Radiol (1998) 28: 967974

Springer-Verlag 1998

Raymond A. Clarke
Gale Catalan
Ashish D. Diwan
John H. Kearsley

Received: 10 October 1997

Accepted: 20 April 1998

R. A. Clarke ( ), J. H. Kearsley
Division of Cancer Services,
St George Hospital Campus,
University of NSW, Kogarah 2217,
Australia
G. Catalan
School of Sociology, Murdoch University,
Murdoch 6150, Australia
A. D. Diwan
Department of Orthopaedic Surgery,
St George Hospital, University of NSW,
Kogarah 2217, Australia

Heterogeneity in Klippel-Feil syndrome:

a new classification

Abstract Background. Klippel-Feil

syndrome (KFS) is characterised by
congenital vertebral fusion of the
cervical spine and a wide spectrum
of associated anomalies. KFS has
often been considered a sporadic
syndrome. However, since the publication of the original KFS classification early this century, a number
of KFS families have indicated heterogeneity complicated by a broad
range of variable expression.
Objective. The two major objectives
of this study were (1) to identify differences and similarities in the postnatal appearance, morphology,
position and inheritance of vertebral
fusions within and between KFS
families and (2) to establish a new
KFS classification focussed on KFS
aetiology.
Materials and methods. Vertebral
fusions were assessed via spinal ra-

Introduction
Congenital fusion of any of the seven cervical vertebrae
is commonly referred to as Klippel-Feil syndrome
(KFS). KFS has an incidence of up to 0.5 % of live births
[14]. Vertebral fusion is most apparent after normal
spinal ossification nears completion in the young child.
KFS can be asymptomatic, however, skeletal, urogenital, cardiac, neurological, auditory, hindbrain, ocular,
craniofacial, otolaryngeal, limb and digital anomalies
have all been associated with KFS in varying degrees
[5]. This variable expression together with KFS heterogeneity have greatly complicated the understanding of
this developmental syndrome [5, 6].

diography. Chromosomal karyotypes were performed using routine

cytogenetics.
Results. The medical histories of
three KFS families are presented.
The postnatal time, position and appearance of vertebral fusions, associated anomalies and mode of
inheritance were different for the
three KFS families. Four classes of
KFS are described in a comprehensive classification table that allays
much of the uncertainty arising from
KFS heterogeneity and variable expression.
Conclusion. We have described four
different KFS classes (KF14) within a comprehensive classification
that addresses KFS genetic heterogeneity. The position of vertebral
fusions in the cervical spine and
their incidence within affected families are delineating features of KFS.

KFS is predominantly sporadic in occurrence and

the early KFS classification by Maurice Klippel and
Andre Feil pre-dated the first descriptions of familial
KFS [7, 8]. Klippel and Feil's classification was based
on the extent of vertebral fusion. Rather than being a
rigid and defining classification parameter, the extent
of fusion can vary widely within affected families
where one assumes a single allelic affect. Confusion
arising indirectly from the old classification has created sufficient impetus to warrant a new KFS classification focussed on the aetiology and genetic origins of
the syndrome. From familial KFS reports, it is now apparent that there exist marked associations between
the position of the most rostral fusion, mode of inheri-

968

tance and some specific KFS associated anomalies [5,

911].
Normal vertebral formation is not uniform along the
developing spinal axis [12, 13]. Somatogenesis/segmentation progresses caudally in the human from the 3rd
week of development and ossification of vertebrae progresses rostrally along the developing cervical spine.
The C2 vertebral body is the last vertebral body to ossify
at around the 4th month of gestation. By Sensenig's
numbering, the four most anterior somites contribute
to the preotic cranium, the fifth is rudimentary, and the
sixth to ninth form the occipital portions of the skull.
The atlas (C1) is formed from the caudal portion of the
tenth and the cranial portion of the eleventh somites.
The cranial portion of the tenth sclerotome forms the
tip of the odontoid process and the ligaments of the atlanto-occipital articulations.
During development C1 loses its centrum which remains attached to the axis (C2) and forms part of the odontoid process. No intervertebral disc develops between
C1 and C2 [2, 13]. Ossification around C2 and the odontoid process continues into childhood. In KFS, fusion of
cervical vertebrae is evident after this normal spinal ossification. Fusion between C2 and the odontoid process
fails in some cases of KFS, resulting in an os odontoideum [14].
C23 fusion has been the most prevalent rostral fusion reported in the KFS literature [5, 10, 11]. Significant intra-familial variation in the extent of vertebral
fusion and variations in the severity of associated anomalies has been particularly apparent within larger more
informative KFS families [5, 11]. KFS may include fusion caudal of the cervical region. However, fusion of
vertebrae in the lower spine, in the absence of cervical
vertebral fusion, is not classified as KFS without supporting evidence. This exclusion directly infers that the
genes responsible for KFS act from the rostral end of
the developing spine.
In this report, we describe three KFS families. Infants from two of these KFS families show distinct variations in the time of appearance of fusions postnatally as
determined by radiography. Within these KFS families,
manifests recessive inheritance, one dominant inheritance and another reduced penetrance. We have used
these three affected families as the models for a new
and comprehensive KFS classification.

Materials and methods

KFS families were referred to us from the Australian Klippel-Feil
Support Group.

Fig. 1 KF101 family pedigree

Radiography
X-rays from KFS families have been collected over a 5-year period. Anterior, posterior and lateral views of the cervical and upper
thoracic spine, with the neck flexed and extended, were completed
for all family members tested (see pedigrees).
Chromosome analysis
Cytogenetic banding studies were performed on cultures of peripheral blood lymphocytes. For high-resolution studies, ethidium bromide (final concentration l mg/ml) was added to the cultures, 2 h
prior to harvest [5].
Patients who gave written consent, in accordance with institutional guidelines, were included in the study. All participants in
the study were asked to donate 20 ml of blood. Data in relation to
their conditions were collected according to standard ethical
guidelines. Ethical clearance was obtained from the St George
Hospital Ethics Research Committee.

Results
Case 1
Proband I (arrow in Fig. 1) was first radiographed when
he was 1 day old (Fig. 2 a). At this early stage the spinal
processes of C14 were fused and there was complete
fusion at C57. There was a very well-defined C45
interspace. Radiographs taken at 12 months of age
(Fig. 2 b) revealed similar levels of fusion compared to
that at 1 day.
The proband displayed the classic triad of vertebral
fusion, very short neck, and low hairline previously described by Klippel and Feil [8]. In addition, the proband
had a cleft palate, large pointed ears and a perimembranous ventriculoseptal defect of the heart that was surrounded by an aneurysm. The maternal grandmother
also had cleft lip and palate without vertebral fusion
and a maternal uncle was reported to have had a very
short neck similar to the proband (Fig. 1). Both the

969

Case 2

b
Fig. 2 a, b Vertebral fusions in the KF101 family. Lateral spinal radiographs of Proband I (individual IV-1, arrow in Fig. 1). a Cervical radiograph at 1 day of age showing malsegmentation of the
cervical spine. Fusion of spinous processes extends caudal of C1.
There is a clear interspace at C45 and fusion of C57 in association with a very short neck. b At 1 year of age the changes are similar to those seen at birth

very short neck and cleft palate are relatively common

anomalies associated with KFS, indicating the possibility of recessive inheritance with variable expression.
The proband and his grandmother had normal chromosomal karyotypes.

Fig. 3 KF201 family pedigree.

Fourteen of the 19 living family
members were assessed using
spinal radiographs. Ten of
14 had at least one vertebral
fusion (always C23) and a
paracentric inversion on chromosome 8. The inversion always segregated with the KFS
phenotype. Deceased individuals classified as affected were
either obligate carriers of this
dominant disease or voice impaired. Individual IV-7 was
deaf and mentally retarded
with a scoliosis. The pedigree is
complete except for the three
unaffected children of unaffected female III-1. This pedigree was originally published
by Clarke et al. [9, 16]

Proband II (arrow in Fig. 3) was first radiographed when

she was 10 weeks old (Fig. 4 a). Her neck did not appear
to be shorter than normal and there was no clear radiological evidence to indicate cervical vertebral fusion.
Follow-up radiographs at 12 and 48 months of age demonstrate progressive ossification of the interspace between the second and third cervical vertebrae (C23
fusion). At 12 months, fusion was evident midway along
the spinous processes between C23 with partial ossification between the vertebral bodies where the intervertebral space had narrowed significantly (Fig. 4 b).
Fusion of the vertebral bodies was complete by
48 months (Fig. 4 c). The proband's ears were relatively
small and low set, she had minor misalignment of the
teeth at the midline and comparatively normal vocal
strength. She also had hindfoot equinus and restricted
arm movements.
Seventy percent of the proband's family also had
KFS [9]. This high frequency of vertebral fusion in the
proband's family was consistent with dominant inheritance [9, 15]. All affected individuals had C23 fusion,
i. e. C23 fusion was dominant, 70 % had C23 and
C45 skipped fusions (Fig. 5) and 20 % had C23, C45
and C67 skipped fusions. Apart from this skipped/positional extension of more caudal fusions, the morphological appearance of the fusions between individuals in the
KF201 family were remarkably similar. In most instances, vertebral bodies were completely fused while
spinous processes were not fused along their entire
length. Ninety percent of proband II's affected family
were vocally impaired in association with laryngeal cartilage malformations [5, 16]. All those with two or
more fusions were vocally impaired. A number of af-

970

Fig. 4 ac Vertebral fusion in the KF201 family. Lateral cervical

radiographs of proband II (individual IV-13, arrow in Fig. 3). a At
10 weeks of age there is hypoplasia at C3 and C4 but no evidence
of vertebral fusion. b At 12 months of age there is narrowing of
the C2/C3 disc space with fusion beginning from midway along
the spinous processes between C23. c At 48 months of age there
is complete fusion of C23 bodies and facet joints. Proband II also
had underdeveloped ears, microstomia, microtia and the chromosome 8 inversion 8 (inv)(q22.2q23.3) [9] common to all other older
members of her family that display dominant expression of the
C23 fusion (Fig. 5) [9, 16]

fected members in the family also had a previous history

of mild conductive hearing impairment or had low set
and/or underdeveloped ears. Thirteen pairs of ribs was
detected in one individual. Affected individuals often
had bilaterally restricted supination and flexion of the
forearms. Here, hindfoot equinus was noted in the
majority of affected individuals. All affected members of the family had chromosome 8 inversion, 8
(inv)(q22.2q23.3) [9].
Case 3
Proband III (arrow in Fig. 6) who was radiographed at
30 years of age had vertebral fusion at C45 (Fig. 7 a).
There was marked reduction in the rostrocaudal width
of the fused vertebrae. Her father had fusion at C56
(Fig. 7 b). KFS was not reported in previous generations.
Fusions in both father and daughter showed complete
fusion along the entire length of the spinous processes
and laminae up to and including the vertebral bodies,
with minor wasp waist line'. Apart from mild neck stiffness, all affected family members had a completely normal' appearance, with no obvious evidence of short
neck or lower hairline. Although the KF301 family

was small, the frequency of vertebral fusion in this family was consistent with dominant inheritance with reduced penetrance. Both affected family members had a
normal chromosome karyotype.

Discussion
Variations in the developmental time of appearance and
morphology of vertebral fusions, as determined from radiographs, was suggestive of different KFS aetiologies.
Fusion of spinous processes (C14), obvious at birth in
proband I, did not progress in the following year to encompass the vertebral bodies. In contrast, the narrow,
yet clear, separation of the C2 and C3 vertebrae of proband II at 10 weeks of age progressed to where ossification of the C23 vertebral interspace was complete by
4 years of age. In proband III and her father, fusion extended from the vertebral bodies to the full length of the
spinous processes. This complete nature of fusion together with reductions in the width of fused vertebrae in father and daughter indicated that these fusions were
congenital and would most likely have been apparent
prenatally. These observations suggested that intrinsically different processes were involved in the vertebral fusion of probands I, II and III. These differences in timing
and gross morphology of fusions together with differences in the incidence and frequency of fusions at specific interspaces and differences in the mode of inheritance
(where applicable) have been addressed in a new KFS
classification (Table 1) that is summarised below:
KF1 C1 fusion is not dominant but is the most rostral fusion recorded in this class. There are often severe associated anomalies, including the very short
neck. Fusion is obvious at birth. Always recessive
mode of inheritance.

971

KF4 Vertebral fusion and ocular anomalies associated with Wildervanck syndrome [17]. Possible Xlinked inheritance.
Control of fusion position
For the KFS families described here and from the more
than 30 KFS families reported in the literature [5, 10,
11, 14, 18, 19], there appears to exist at least three facets
of positional control regulating (1) the most rostral spinal position of fusion, (2) fusion position patterns and
(3) the extent of vertebral fusion within the spine.
The most rostral spinal position of fusion

Fig. 5 Vertebral fusion in the KF201 family. Lateral cervical radiograph of individual III-2 (arrow in Fig. 3). C23 and C45 fusion
of bodies and facet joints and degenerative disc at C34. Anterior
kyphosis at C34. For radiographs of other members of the family
see Clarke et al. [9, 16]

Fig. 6 KF301 family pedigree

KF2 C23 fusion is dominant. C23 is always the

most rostral fusion and fusion is apparent postnatally.
Dominant mode of inheritance.
KF3 C3 (C23 or C34) fusion is not dominant but
is the most rostral fusion recorded in this class. There
is often a single isolated fusion. Reduced penetrance
inheritance.

Proband II and those affected in her extended family all

had C23 fusion, i. e. C23 was dominantly expressed. In
addition, the most rostral spinal fusion in proband II
and her family was always at the C23 position. In all
other KFS families that clearly show a dominant mode
of KFS inheritance, the C23 fusion has also been dominantly expressed as the most rostral spinal fusion [5,
10, 11, 14, 18, 19], though an os odontoidium [14] may
be present. Such dominant KFS families often manifest
a mild KFS phenotype with some craniofacial anomalies
in a proportion of the individuals affected. In the
KF201 family described here, fusion does not involve
the full length of the spinous processes. In the KF2 class
of KFS, neck length has never been severely compromised, which is in distinct contrast to individuals/families where C1 is implicated in fusion(s). KFS families
with recessive or reduced penetrance inheritance have
always had some affected individuals with a fusion sequence that did not include the C23 fusion as the most
rostral fusion, i. e. C23 fusion is not dominant in families manifesting recessive KFS inheritance.
The most rostral spinal position of fusion in proband
I was C12 fusion. As with proband I's family, recessive
inheritance can be very difficult to confirm even with
the aid of a large pedigree. Here KFS familial inheritance was based on the manifestation, in close maternal
relatives, of anomalies commonly associated with KFS,
such as cleft lip and short neck. Proband I was typical
of other cases where C12 fusion has been implicated,
in that all such affected individuals have the very short
neck in addition to other severe manifestations of the
KFS syndrome, e. g. heart, particularly ventriculoseptal
defects, and urogenital abnormalities in addition to
craniofacial involvement.
Proband III had a single isolated congenital fusion
(C45) in the middle of the cervical spine. Proband
III's father had congenital fusion of C56. Therefore,
the most rostral spinal position of fusion in proband
III's family was at C45. Proband III's family was also

972

Table 1 Classification of Klippel-Feil syndrome. Important features are highlighted in bold. KF13 numbering denotes the position of the most rostral fusion in the class, i. e. C1 fusion (O-C1
and/or C12), C23 fusion or C3 fusion as the primary distinguish-

ing clinical criteria in the KF1, KF2 and KF3 classes respectively.
Thoracic and lumbar fusions have no bearing on class. (K & F
Klippel und Feil's original classification, Gunderson Gunderson
et al. [11])

Class

Vertebral fusions

Inheritance

Possible anomalies1

KF1

Only class with C1 fusion

C1 fusion not dominant
Variable expression of other
fusions
Fusions radiologically visible
at birth

Recessive

Very short neck, heart, urogenital, Types I, II and III (K & F)

craniofacial, hearing, limb, digital, Recessive (Gunderson)
ocular defects. Variable expression

KF2

C23 fusion dominant

C23 fusion most rostral fusion
Cervical, thoracic and lumbar
fusion show variable expression
within a family
Fusions apparent postnatally

Dominant
Includes SGM1
gene mutation

Craniofacial, hearing, otolaryngeal, Types I, II and III (K & F)

skeletal and limb defects, etc
C23 dominant (Gunderson)
Variable expression

KF3

Isolated cervical fusions, includes

single fusions at variable positions
Any cervical fusion except C12
Includes single C56 fusion

Recessive or
reduced
penetrance

Craniofacial
Facial dysmorphology
Variable expression

Type II (K & F)
C56 recessive (Gunderson)

KF41

Fusion of cervical vertebrae

Limited data available

Possibly
X-linked
Predominantly
females

Hearing and ocular anomalies

abducens palsy with retractio
bulbi
Heart defects possible

Commonly referred to as
Wildervanck syndrome
(includes Duane's syndrome)

Overlap with other classifications

This classification may be further distinguished by particular associated anomalies

typical of other cases where neither C12 fusion or dominant C23 fusion are implicated [11] in that it is common for such individuals to have an unremarkable
phenotype, without the short neck or other obvious syndrome involvement. Because of this absence of classically associated anomalies, radiologists may question
such individuals' KFS status.
Fusion position patterns
Fusion position patterns in KFS have been reported
previously for the extended family of proband II [5].
Such patterns were defined by the recurring period/distance between fusion positions, e. g. the distance between fusions within a C23, C45 sequence represents
a period of two interspaces. Such patterns can be recognised/established when they are exhibited by all individuals of an affected family. Other KFS families
manifesting dominant expression of the C23 fusion
have also manifest periodic patterns of fusions [10]. To
date, these position patterns have always implicated
the C23 fusion position as the pivotal rostral focus of
the pattern. In addition, fusion frequency within these
affected families generally decreases with the caudal
distance from the most rostral fusion [11]. This reflects
an increased vulnerability of the rostral vertebrae to fusion and suggests implications for the KFS genes within
developmental patterning pathways [5, 2022]. The

dominance of the rostral fusion within these patterns

further emphasises the importance of the most rostral
position of fusion as a delineating characteristic in respect to KFS heterogeneity.
The extent of vertebral fusion
KFS, by definition, is characterised by congenital cervical vertebral fusion, which may be complicated by
delayed/postnatal ossification across the fusion interspace. The extent of vertebral fusion along the spine
was the basis for Klippel and Feil's original classification of KFS into type I extensive cervical fusions,
type II fusion of one or two cervical interspaces and
type III combined cervical and lower thoracic and
lumbar fusion [7, 8]. From descriptions of familial KFS
reported since 1912 it is now obvious that the extent of
KFS vertebral fusion is largely influenced by variable
expression, i. e. some affected members of a KFS family
may have many more vertebrae fused compared to other less affected family members. For example, Proband
II's family manifested minimal C23 fusion in some individuals and multiple cervical fusions in other individuals. Likewise, minimal C23 fusion has been described
in another KF2 class family that includes family members with cervical and thoracic fusions [10].
It is obvious that a KFS classification based on syndrome features that segregate within affected families

973

Fusion position and variable expression

All KFS classes KF14' may at times present with single isolated fusion of cervical vertebrae. KF1 and KF2
can be broadly defined by their expression of C1 fusion
and dominant C23 fusion, respectively, where additional fusions extend caudally with diminishing intrafamilial
frequency. In the absence of ocular defects and the very
short' neck, all individuals with fusions not occurring at
C1 or C23 [11] have been assigned to the KF3 class. In
contrast to the KF1 and KF2 classes, KF3 appears to be
under a complex mode of variable expression where no
particular cervical fusion position (caudal of C23) is
either requisite or predictive. Again, in contrast to the
KF1 and KF2 scenarios, the positional variability of the
most rostral fusion in the KF3 class would appear to indicate that fusion at one interspace in KF3 can occur independently of fusions at any other interspace (caudal
of C23). Concurrent with this variability, however, we
observed a distinct morphological similarity between
the vertebral fusions in proband III and her father, lending support for a functional uniformity in the misexpression of the KF3 allele but at adjacent positions.
KFS associated anomalies

b
Fig. 7 a, b Vertebral fusion in the KF301 family. a Lateral cervical
radiograph of proband III (individual II-1, arrow in Fig. 6) with congenital fusion of C45 with residual calcified disc. Fusion affects
bodies, apophyseal joints and spinous processes. Degenerative disc
at C34 and C67. b Lateral cervical radiograph of the father of
proband III (individual I-1 in Fig. 6) with congenital fusion of C56
indicated by shortened vertebral bodies with residual calcified disc.
Fusion affects bodies, apophyseal joints and spinous processes

will better reflect KFS aetiology and genotype than a

classification based on parameters (e. g. extent of fusion
along the spine) that vary within affected families. Phenotypic variations within an affected family are indicative of those KFS features that differ as a function of
variability of gene/allele expression. The present KFS
appraisal has emphasised an association between the position and incidence of the most rostral fusion and the
mode of inheritance. The KF14 classification (Table 1)
is sufficiently broad and detailed to embrace most cases
of congenital (as distinct from post-traumatic) vertebral
fusion (sporadic and inherited) of the cervical spine.

The type of anomalies, not necessarily their severity,

within KFS families have been relatively consistent in
their expression and thus appear to be a function of the
specific genetic mutation involved [5, 10]. As such, different anomalies associated with KFS should not be the
sole basis for establishing additional KFS classes or the
proliferation of new' syndrome nomenclature [2325].
In this study we have alluded to intrafamilial variability in the expression of craniofacial anomalies including
abnormal ears, cleft palate, laryngeal malformation
and associated vocal defects, restricted limb movement
and heart defects. Such variability has also been typical
in other KFS families [10, 11]. A degree of correlation,
reported previously, between the extent of fusion and
the severity of associated craniofacial anomalies [5] suggested that the severity and/or expression of associated
anomalies can also be a function of variable expression.
Nevertheless, associated anomalies are usually more
severe in the KF1 class.
For proband I, as reported elsewhere in the literature
for the KF1 class, vertebral fusions involving C1 were
associated with the classical KFS symptoms of very
short neck and lower hairline in addition to craniofacial
and other anomalies e. g. heart defects. Abnormalities
of the face affecting the palate and ears in proband I
were also implicated in proband II and have been reported in other KF2 class families [9, 10, 16]. This would
appear to indicate different KFS genes may be closely
related and/or function within overlapping developmen-

974

tal fields impacting on spinal and craniofacial development. Of interest here is the fact that the segmenting tissues that give rise to the vertebrae and face, including
the presomitic mesoderm and the hindbrain, respectively, also share a common developmental time frame and
many common gene determinants involved in body patterning pathways [5, 13, 2022].
The Short Neck
While at odds with Klippel and Feil's original clinical
description of sporadic KFS, a relatively normal neck
length, as described for probands II and III, has been
commonly associated with KFS. In proband II's family,
neck length did appear to vary slightly with the extent
of cervical fusion. However, this phenotype could not
be described as 'no neck'. We feel that the short or no
neck' phenotype of proband I and of other KF1 cases is
more representative of the classic description of this entity by Klippel and Feil. The short neck has been reported in association with a wide range of fusion sequences.

However, this anomaly is clearly related to developmental factors beyond the localised dysplastic effects of vertebral fusion (e. g. Sprengel's shoulder) and most often
the short neck fits best within the KF1 class.

Conclusion
The KF14 classification delineates four KFS classes on
the basis of different vertebral fusion processes, morphologies and positions, different associated anomalies
and different modes of inheritance. This new KFS classification will provide a useful tool for the clinician and
paediatrician by leading to a better understanding of
KFS aetiology while providing a useful framework for
the localisation and characterisation of the KFS genes.
Acknowledgements We would like to thank Dr Carl Bryant for
interpretation of radiographs and help with the text. This work
was funded by the National Health and Medical Research Council
of Australia.

References
1. Brown MW, Tempelton AW, Hodges FJ
(1964) The incidence of congenital fusions in the cervical spine. AJR 92:
12551259
2. Gray SW, Romaine CB, Skandalakis JE
(1964) Congenital fusion of the cervical
vertebrae. Surgery, Gynecology and
Obstetrics: 373385
3. Le Double AF (1912) Trait des variations de la colonne vertebrale de
l'homme. Paris
4. Luftman II, Weintrub S (1953) KlippelFeil syndrome in a full term stillborn
infant. AJR 70: 23
5. Clarke RA, Kearsley JH, Walsh D
(1996) Patterned expression in familial
Klippel-Feil syndrome. Teratology 53:
152157
6. Nguyen VD, Tyrrel R (1993) KlippelFeil syndrome: patterns of bony fusion
and wasp-waist sign. Skeletal Radiol 22:
519523
7. Feil A (1919) L'absence et la diminution des vertbres cervicales (tude clinique et pathogenique): le syndrome de
la reduction numrique cervicale. Sorbonne University, Paris
8. Klippel M, Feil A (1912) Anomalie de
la colonne vertbrale par l'absence des
vertbres cervicales: cage thoracique
remontant jusqua' la base du crne.
Bull Hem Soc Anat Paris 87: 185188

9. Clarke RA, Kearsley JH, Singh S, Yip

M (1995) Familial Klippel-Feil syndrome and paracentric inversion
inv(8)(q22.2q23.3). Am J Hum Genet
57: 13641370
10. Sheffield LJ (1982) A dominantly inherited syndrome of palate and vertebral abnormalities. Annual general
meeting of Human Genetics Society of
Adelaide. Adelaide, Australia
11. Gunderson CH, Greenspan RH, Glaser
GH, Lubs HA (1967) The Klippel-Feil
syndrome: genetic and clinical re-evaluation of cervical fusion. Medicine 46:
491512
12. Meinhardt H (1986) Models of segmentation. In: Ede DA, Lash JW (eds) Somite in 17 developing embryos, vol 118.
Plenum Press, New York, pp 161178
13. Couly GF, Coltey PM, Douarin NM
(1993) The triple origin of the skull in
higher vertebrates: a study in quail
chick chimeras. Development 117:
409429
14. Morgan MK, Onofrio BM, Bender CE
(1989) Familial os odontoideum. Case
report. J Neurosurg 70: 636639
15. Wood S, Othmane B, Bergerheim U
(1993) Report of the first International
Chromosome 8 workshop. Cytogenet
Cell Genet 64: 133146
16. Clarke RA, Davis PJ, Tonkin J (1994)
Klippel-Feil syndrome associated with
malformed larynx. A case report. Ann
Otol Rhinol Laryngol 103: 201207

17. Wildervanck LS (1963) A cervicooculo-acusticus syndrome belonging to

the status dysraphicus. Proceedings of
the second International Congress of
Human Genetics, vol 3. Rome, pp 409
18. Pfeiffer RA, Rott HD, Angerstein W
(1992) An autosomal dominant facioaudio symphalangism syndrome with
Klippel-Feil anomaly: a new variant of
multiple synostosis. Genet Couns 38:
133140
19. Shaver KA, Proud VK, Shaffer MC,
et al (1986) Deafness, facial asymmetry
and Klippel-Feil syndrome in five generations. Am J Hum Genet 39: A81
20. Hunt P, Gulisano M, Cook M, et al
(1991) A distinct Hox code for the
branchial region of the head. Nature
353: 861864
21. McGinnis W, Krumlauf R (1992) Homeobox genes and axial patterning. Cell
6: 283302
22. Stern CD, Keynes RJ (1988) Mechanisms of segmentation (review). Development 103: 413429
23. Connor JM, Smith R (1982) The cervical spine in fibrodysplasia ossificans
progressiva. Br J Radiol 55: 492496
24. Lawry LB (1977) The Klippel-Feil
anomaly as part of the fetal alcohol
syndrome. Teratology 16: 5366
25. Martel W, Holt JF, Cassidy JD (1962)
Roentgenologic manifestations of juvenile rheumatoid arthritis. Am J Rheumatol 88: 400422