CLINICAL ORTHOPAEDICS AND RELATED RESEARCH
Number 424, pp. 183–190
Klippel-Feil Syndrome
Clinical Features and Current Understanding of Etiology
M. R. Tracy, MD* ,†; J. P. Dormans, MD* ,†; and K. Kusumi, PhD* ,‡ ,§
Klippel-Feil syndrome occurs in a heterogeneous group of
patients unified only by the presence of a congenital defect in
the formation or segmentation of the cervical spine. Numer-
ous associated abnormalities of other organ systems may be
present. This heterogeneity requires comprehensive evalua-
tion of all patients and treatment regimes that can vary from
modification of activities to extensive spinal surgeries. This
also has made delineation of diagnostic and prognostic
classes difficult and has complicated elucidation of the ge-
netic etiology of the syndrome. Furthermore, it is unclear
whether Klippel-Feil syndrome is a discrete entity, or if it is
one point on a spectrum of congenital spinal deformities.
Pedigree analysis has identified a human genetic locus for the
disease. Mouse models suggest members of the PAX gene
family and Notch signaling pathway as possible etiologic can-
didates. Only by identifying the link between the genetic
etiology and the phenotypic pathoanatomy of Klippel-Feil
syndrome will we be able to rationalize the heterogeneity of
the syndrome.
Klippel-Feil syndrome first was described by Klippel and
Feil22 as seen in a 46-year-old tailor evaluated for “pleu-
risy with pulmonary congestion and nephritis.” This pa-
tient had a short neck, low posterior hairline, and limited
range of motion (ROM) of the neck caused by a congenital
Received: May 21, 2003
Revised: October 9, 2003
Accepted: February 18, 2004
From the *Division of Orthopaedic Surgery, The Children’s Hospital of
Philadelphia, Philadelphia, PA; †Department of Orthopaedic Surgery, Uni-
versity of Pennsylvania School of Medicine, Philadelphia, PA; ‡Division of
Human Genetics, The Children’s Hospital of Philadelphia, Philadelphia, PA;
and the §Department of Pediatrics, University of Pennsylvania School of
Medicine, Philadelphia, PA.
This work was supported by grants to K.K. and J.P.D. from the Cervical
Spine Research Society and the Ethel Brown Foerderer Fund for Excellence.
K.K. is a Hitchings-Elion Fellow of the Burroughs Wellcome Fund, and a
Florence R.C. Murray Fellow.
Correspondence to: John P. Dormans, MD, Division of Orthopaedic Surgery,
The Children’s Hospital of Philadelphia, 2nd Floor Wood Center, 34th St. &
Civic Center Blvd., Philadelphia, PA 19104-4399. Phone: 215-590-1527;
Fax: 215-590-1501; E-mail: dormans@email.chop.edu.
DOI: 10.1097/01.blo.0000130267.49895.20
183
© 2004 Lippincott Williams & Wilkins
segmentation defect of the cervical spine.22 It now is
known that fewer than 50% of patients with congenital
defects of the cervical spine have all three signs of this
classic triad (a short neck, low posterior hairline, and lim-
ited ROM of the neck). The spectrum of anomalies known
to be associated with Klippel-Feil syndrome is immense
and variable (Table 1).7,10,17,19,41
The designation Klippel-Feil syndrome includes a het-
erogeneous group of patients unified only by the presence
of a congenital synostosis of some or all cervical vertebrae
(Fig 1).39 The absence of population screening studies has
made it impossible to define the exact incidence and
prevalence of Klippel-Feil syndrome.10 However, it has
been estimated that it occurs in approximately 1:40,000–
42,000 births.40,41 A slight female predominance of ap-
proximately 3:2 has been reported.10,18 Clinical evalua-
tion, natural history, and treatment of the disease are af-
fected by the pathoanatomy of the fusion and the type and
severity of associated anomalies present.
The heterogeneity of patients with Klippel-Feil syn-
drome has made delineation of diagnostic and prognostic
classes difficult and has complicated elucidation of the
genetic etiology of the syndrome. Advances in molecular
genetics have led to reexamination of the classification
schemes currently used for Klippel-Feil syndrome, and the
relationship between Klippel-Feil syndrome and other
congenital spinal deformities. The molecular and clinical
genetics literature addressing the etiology of Klippel-Feil
syndrome is extensive. We present the salient features that
orthopaedists treating patients with Klippel-Feil need to
understand these advances, their impact on diagnosis and
prognosis, and the future directions of research on this
complicated syndrome.
Clinical Evaluation of Patients with
Klippel-Feil Syndrome
Although affected patients have cervical anomalies at
birth, Klippel-Feil syndrome usually is diagnosed at a later
age.10 Neurologic, myelopathic, and biomechanical prob-
 Clinical Orthopaedics
184 Tracy et al and Related Research

TABLE 1. Most Commonly Associated and the disorder is discovered incidentally when radio-
Abnormalities Found in graphs are done for an unrelated reason.7 Regardless of the
Klippel-Feil Syndrome7,10,17,19,41 presentation, after a synostosis of the cervical spine has
Anomaly Percentage of Patients been discovered, high quality radiographs of the patient’s
cervical spine must be obtained to evaluate the nature and
Congenital scoliosis > 50%
Rib abnormalities* 33% extent of the fusion.19,29
Deafness 30% Radiographic evaluation of the cervical spine of infants
Genitourinary abnormalities 25%–35% and children can be difficult, and differences are seen in
Sprengel’s deformity 20%–30% these patients compared with adults. Translation of one
Synkinesia 15%–20%
vertebral body on another may suggest instability associ-
Cervical ribs 12%–15%
Cardiovascular abnormalities 4%–29% ated with an adjacent fused segment or segments. How-
ever, pseudosubluxation of C2 on C3 or C3 on C4 may be
*Excluding cervical ribs normal in children younger than 8 years.11 Also, the ra-
diographic appearance of the cervical spine can change as
lems can lead to presentation. However, the most common a child grows, and incomplete vertebral ossification can
complaints of patients with congenital synostosis of the make fusions difficult to appreciate (Fig 1).10 Finally, the
cervical spine are pain and neurologic symptoms, and de- radiographs of children with fusions secondary to abnor-
creased rotation, flexion, and extension. Patients with at- malities such as juvenile rheumatoid arthritis occasionally
lantoaxial fusions seen incidentally on radiographs often can mimic the appearance of Klippel-Feil syndrome,40 al-
present at younger ages than patients with more caudal though the history and physical examination should allow
fusions.38 Patients with extensive fusions also tend to pre- the clinician to easily distinguish between this condition
sent at a very early age, perhaps because of cosmetic de- and Klippel-Feil syndrome.
formity.10 In some patients, the congenital synostosis is Advanced imaging techniques can provide additional
discovered to be part of a larger, named syndrome such as information not available with standard radiographs. Mag-
Wildervanck, Rokitansky-Kuster-Hauser, or Golden- netic resonance imaging (MRI), including flexion and ex-
har.8,10 In others, the cervical synostosis is asymptomatic, tension MRI, is indicated when preliminary studies sug-

Fig 1A–B. (A) This lateral cervical spine radiograph of a child with Klippel-Feil anomaly shows fusion of adjacent vertebral bodies
(black arrowheads) and adjacent neural arches (white arrowheads). (B) A plain radiograph of a child with Klippel-Feil syndrome
can be difficult because of the dynamic situation of growth. An area between two vertebral bodies may appear open (black
arrowheads), although narrowed, but ultimately may fuse with time.
Number 424
July 2004
gest instability or stenosis.10 Magnetic resonance imaging
can be used to determine if the cervical malformation has
compressed the brain, brainstem, or spinal cord.29 Mag-
netic resonance imaging offers clinicians the ability to
evaluate the space available for the cord, assess the degree
of spinal stenosis caused by the fusion, and detect CNS
abnormalities such as a syrinx, tethered cord, or diastomy-
elia. Judicious use of these advanced imaging techniques,
in combination with plain radiographs, will provide ortho-
paedists with an appropriate understanding of the patho-
logic and surgical anatomy present in most patients with
Klippel-Feil syndrome. Computed tomography (CT) is
helpful in defining bony pathoanatomy but is associated
with radiation exposure to the child. Other radiologic tech-
niques currently available, but of limited usefulness, in-
clude three-dimensional reconstruction CT and fluoro-
scopic imaging.
Because of the numerous possible associated anomalies
that may result in considerable morbidity (Table 1), com-
prehensive evaluation of all patients with Klippel-Feil
syndrome is required. In addition to radiographs of the
cervical spine, all patients diagnosed with Klippel-Feil
syndrome should have imaging of the thoracic and lum-
bosacral spine to look for other spinal anomalies and to
monitor for the development of scoliosis.40 Thorough,
periodic neurologic evaluations are necessary to rule out
cranial nerve abnormalities, cervical radiculopathy, or my-
elopathy.19 Cardiac evaluation, renal ultrasound (some-
times intravenous pyelogram), and audiologic testing are
necessary to rule out disease in these systems.19 Associ-
ated anomalies are seen less commonly in the gastrointes-
tinal, respiratory, and dermatologic systems,8 however cli-
nicians must be prepared to do workups for possible
anomalies.
Natural History and Treatment of
Klippel-Feil Syndrome
There is a spectrum of clinical severity associated with
Klippel-Feil syndrome. Cervical fusions may be asymp-
tomatic and identified only incidentally on radiographs
obtained for other purposes.7 Alternatively, patients may
present with decreased neck ROM related to the extent of
cervical spinal involvement, neck and/or radicular pain
from degenerative changes or hypermobility, cosmetic
concerns secondary to a shortened or webbed neck, or
problems related to anomalies in other organ systems that
are formed embryologically at the same time as the cer-
vical spine (including the inner ear, heart, and kidneys).10
Although fused segments may directly limit neck ROM,
more serious complications, including instability, hyper-
mobility, and symptomatic stenosis, can arise at the inter-
spaces between fused segments. Three specific patterns of
cervical fusion create a high risk for symptomatic insta-
Klippel-Feil: Features and Etiology 185
bility: C2-C3 fusion with occipitocervical synostosis, ex-
tensive fusion over several cervical vertebrae with an ab-
normal occipitocervical junction, and two fused segments
separated by an open joint space.7,18 In each instance,
motion at one level of the vertebral column alters cervical
spine biomechanics such that dangerous instability and
subsequent neurologic compromise can occur. This places
patients at increased risk of neurologic injury and often
presents as neurologic sequelae during the second or third
decade. Alternatively, stenosis and osteoarthritis may de-
velop at interspaces between fused segments.19 As seen,
plain radiographs, CT, and MRI all offer valuable infor-
mation about the pathoanatomy responsible for central and
peripheral neurologic sequelae.10
Most asymptomatic patients with stable fusion patterns
will not have cervical symptoms. Theiss et al39 reported on
32 patients with congenital scoliosis in whom cervical
fusions were seen during the workup for scoliosis. None of
the patients reported cervical symptoms at the time of
diagnosis. During 10 years followup, only seven patients
had cervical symptoms develop. Similarly, Rouvreau et al34
found that only five of 19 children and adolescents with
Klippel-Feil syndrome had neurologic complications de-
velop during an average followup of 12.5 years. However,
if cervical symptoms do arise, symptomatic treatment may
be appropriate. Modification of activities, bracing, and
traction can be used to reduce symptoms, delay surgery,
and prevent neurologic compromise after minor trauma in
these patients.7,19
Patients with progressive symptomatic segmental insta-
bility or neurologic compromise are candidates for surgi-
cal stabilization of the abnormal region of the cervical
spine. Occipitocervical arthrodesis is accomplished most
commonly through a posterior approach. Numerous tech-
niques are available, including onlay of autogenous graft
with or without an occipital periosteal flap.13 Internal fixa-
tion is achieved using wire fixation, prefabricated Luque
rings, or the insertion of plates and screws.16
Most of the techniques available for occipitocervical
arthrodesis were developed for adults. One technique,13
developed specifically for occipitocervical arthrodesis in
children (Fig 2), involves placing four burr-holes in the
occiput in transverse alignment. A corticocancellous graft
from the iliac crest is positioned in the occipital trough and
around the appropriate cervical vertebra. Wires are passed
through the burr-holes and around the graft and are used to
lock the graft into place. This technique allows for in-
creased stability of the fusion and relatively early removal
of the halo immobilization device.13
Regardless of the surgical technique used, occipitocer-
vical arthrodesis usually is accompanied with halo ring
and vest immobilization.13 Complications related to halo
immobilization (pin-site infections) are common in chil-
 Clinical Orthopaedics
186 Tracy et al and Related Research

Fig 2A–D. (A) Four burr-holes in transverse alignment are drilled into the occiput, with a 1-cm osseous bridge between the two
holes of each pair. (B) A corticocancellous ilial graft is shaped into a rectangle with a notch at the base for the spinous process
of C2 or C3. (C) Looped 16- or 18-gauge Luque wires are passed through the burr-holes, while Wisconsin button wires are passed
through the base of the spinous process of either C2 or C3. (D) The wires are crossed, twisted, and cut. (Reprinted with permission
from Dormans JP, Drummond DS, Sutton LN, Ecker ML, Kopacz KJ: Occipitocervical arthrodesis in children. J Bone Joint Surg
77A: 1234–1240, 1995.)

dren and adults. However, these complications do not pre-
vent successful completion of halo immobilization in chil-
dren.12
As with occipitocervical arthrodesis, the techniques
available for atlantoaxial and subaxial arthrodesis vary by
the level of invasiveness, amount of instrumentation used,
and the stability of the resultant fusion. When intact pos-
terior elements are present, wires most commonly are used
to achieve atlantoaxial fusion. However, screw fixation
can be used when the C1 arch is incompetent or when
passage of sublaminar wires would be ill-advised because
of decreased space available for the cord.16
In children, the surgical procedures most commonly
done for subaxial and sublaminar arthrodesis involve in-
Number 424
July 2004 Klippel-Feil: Features and Etiology 187

terspinous or sublaminar wire or cable fixation. Such tech-

niques offer lower costs, decreased operative times, and
lower risks for neurologic and other complications; how-
ever, they require intact posterior elements. When the pos-
terior elements are deficient, lateral mass, transpedicle, or
transarticular screw-plate fixation offer a reasonable alter-
native but have limitations in young children.16 Anterior
decompression or arthodesis occasionally is indicated for a
fixed sagittal deformity.

Embryology and Molecular Genetics of

Spinal Development
Beginning 14 days after conception, the process of gastru-
lation generates the mesenchymal cells that will form the
future head, cardiac, paraxial, and lateral mesoderm.35 Be-
tween Days 20 and 30, the paraxial mesoderm is subdi-
vided (referred to as segmentation by developmental bi-
ologists) in a rostral-to-caudal direction into spherical,
segmental structures called somites (Fig 3). As they ma-
ture, somites divide into three subsections: the sclerotome,
which forms the adult vertebrae; the myotome, which
forms the adult axial musculature; and the dermotome,
which contributes to the adult dermis.21 The sclerotome
undergoes a resegmentation process, such that the caudal
section from one somite joins with the rostral section of
the immediately caudal somite to comprise the vertebral
body (Fig 3).
Klippel-Feil syndrome, and numerous other congenital
spinal deformities, may result from mutations or disrup-
tions in genes regulating segmentation and resegmenta-
tion. Geneticists use multiple tools to identify these dis-
ease genes. One approach involves studying transmission
of a disease in families (pedigree analysis). The mode of
transmission (autosomal dominant, autosomal recessive,
gender-linked, polygenic) is determined, and a chromo-
somal location and genetic locus then can be assigned by
observing how the trait segregates with respect to known
markers. These markers are defined variable DNA se-
quences, including single nucleotide polymorphisms
(SNPs) and microsatellite repeats. With completion of the
human genome sequencing project,24 researchers can de-
termine if any of the genes within the identified chromo-
somal region are likely etiologic candidates.
Unfortunately, the heterogeneous composition of pa-
tients with Klippel-Feil syndrome, small pedigrees, and
the broad spectrum of anomalies associated with predomi-
nantly sporadic cases of Klippel-Feil syndrome make this
a complicated process.4 Online Mendelian Inheritance in
Man lists at least three genetic forms of Klippel-Feil syn-
drome with either dominant or recessive inheritance: Klip-
pel-Feil syndrome (148900); Klippel-Feil deformity, con-
Fig 3A–B. (A) Somitogenesis is the process by which unseg-
mented, presomitic mesoderm (PSM) is subdivided into
somites, which are pairs of segments on either side of the
future spinal cord. As somites mature, the ventral section sub-
divides to form the sclerotome, mesenchymal cells that give
rise to the vertebrae. (B) Cells from the caudal half of one
somite join with those from the rostral half of the adjacent
somite to form the vertebral bodies and arches.
ductive deafness, and absent vagina (148860); and Klip-
pel-Feil deformity, conductive deafness, and facial
asymmetry (148870).30 Autosomal recessive inheritance
of Klippel-Feil anomaly has been described in a large
Brazilian pedigree.8
The medical genetics literature also contains descrip-
tions of defects of known etiology occurring coinciden-
tally with Klippel-Feil anomaly, including a Pro250Arg
mutation in the fibroblast growth factor receptor-3 gene
(FGFR3) in patients with autosomal dominant coronal
synostosis, dysmorphic vertebrae and ribs, and Sprengel’s
anomaly.25 Furthermore, there are several reports of mu-
tations in this gene segregating with traits other than cer-
vical synostosis.27
A study of a large family revealed that Klippel-Feil
syndrome and vocal impairment defects segregate with a
paracentric inversion of chromosome 8. This led to the

188 Tracy et al
proposal of the first human Klippel-Feil syndrome locus,
SGM14,5 It has been proposed that the chromosomal in-
version disrupts the function of the SGM1 gene. This dis-
ruption may have unidentified, downstream consequences
that alter axial patterning of the developing cervical spine
and result in Klippel-Feil syndrome.4 Positional cloning of
the SGM1 gene would provide greater understanding of
the molecular and developmental mechanisms that cause
Klippel-Feil syndrome.
Identification of genes in model systems, such as the
mouse, that cause phenotypic changes similar to those
seen in Klippel-Feil syndrome complement genetic map-
ping efforts. The high degree of similarity between mouse
and human genes allows researchers to use mouse mutants
with spinal defects as models for development of the hu-
man vertebral column. In particular, genes in the Notch
signaling pathway and PAX family are being investigated
for playing possible roles in vertebral segmental disorders.
Discovery of mutations in a candidate gene in multiple,
affected individuals would strongly support an etiologic
role for that gene.
Notch pathway genes regulate cell-fate determination,
embryonic patterning, and somitogenesis.36 Mutations in
the Notch pathway genes Notch1, Dll1, Dll3, Hes7, Psen1,
and Lfng disrupt somite segmentation in mice.33,36 Muta-
tions in the human Notch ligand DLL3 cause the vertebral
malsegmentation disorder, spondylocostal dysostosis.1 In
addition, disruption of the Notch ligand JAG1 defect re-
sults in Alagille syndrome,26,31 a multiorgan disorder with
67% of patients having vertebral disorders.14 Although
Notch pathway genes are important for vertebral segmen-
tation, patients with Klippel-Feil syndrome are only be-
ginning to be examined for potential mutations in these
genes.
In mice and humans, nine PAX genes have been iden-
tified that play important roles in development and share a
conserved paired-box DNA-binding region. In particular,
Pax1 and Pax9 have been shown to regulate somite seg-
mentation in the mouse. Pax1 is strongly expressed in the
developing vertebral column in mice. The mouse Pax1
mutations, undulated and Danforth’s short-tail, have se-
vere malformations of the vertebral bodies, intervertebral
discs, and ribs. In humans, mutations in PAX genes have
been identified in several major human birth defects, in-
cluding aniridia and Waardenburg syndrome.2,9 However,
a human mutation in PAX1 has not yet been identified.
Other developmentally important genes, such as those
in the HOX complexes, are required for the normal speci-
fication and identity of vertebrae.23 For example, inacti-
vation of Hoxd3 in the mouse causes occipitalization of
the atlas,6 whereas a Hoxd4 mutation causes replacement
of the occipital region by supernumerary cervical verte-
brae.28 Although HOX mutations may play a role in the
Clinical Orthopaedics
and Related Research
specification of vertebral identity and in occipital assimi-
lation of C1, evidence from current mouse models does
not suggest that HOX mutations play a role in extensive
vertebral fusions.9
DISCUSSION
The variations in pathoanatomy and associated abnormali-
ties seen in patients with Klippel-Feil syndrome necessi-
tate comprehensive evaluation of all patients and treatment
regimes that can vary from modification of activities to
extensive spinal surgeries. They also have led to a prolif-
eration of classification schemes based on anatomy, prog-
nosis, and genetic features of the syndrome.
The earliest classification schemes were based solely on
the anatomic distribution of the fused segments. Patients
with Klippel-Feil syndrome were assigned to one of three
types: Type I, those with extensive fusions of many cer-
vical vertebrae; Type II, those with fusion(s) at only one or
two cervical interspaces; and Type III, those with fusions
in the cervical spine accompanied by fusions in the lower
lumbar spine.15,41 Subsequent studies showed that patients
with Types I and III most commonly had an autosomal
recessive inheritance pattern, whereas patients with Type
II most often had an autosomal dominant inheritance pat-
tern.15,20,27 Generally, other skeletal abnormalities such as
Sprengel’s deformity and the presence of cervical ribs
most frequently were associated with Type II. Syndromic
anomalies most frequently were associated with Type I.
Patients with Type II were least likely to have increased
curvature develop in the sagittal plane of the spine, with
less than 10° defined as scoliosis,37 whereas patients with
Type I or III had a higher risk for development and pro-
gression of scoliotic curves.40
Spinal kinematics in flexion and extension were used to
create a functional and prognostic classification.32 Com-
parisons of motion at open interspaces were used to
stratify patients with Klippel-Feil syndrome into one of
four classes and identify the likelihood of having neuro-
logic symptoms develop. Patients with hypermobility of
the upper cervical segment were at greatest risk for neu-
rologic sequelae. Patients with hypermobility of the lower
cervical segment were at greatest risk of degenerative
change.
As more genetic information became available, it was
possible to incorporate the mode of inheritance into clas-
sification schemes. This information was combined with
the location of the most rostral fusion found in a family to
form a revised anatomic classification.3 Fusions at C1 with
or without more caudal fusions were described as Klippel-
Feil 1 (KF1), which had autosomal recessive inheritance
Number 424
July 2004
and most frequently was associated with other severe
anomalies. Patients with Klippel-Feil 2 (KF2) had their
most rostral fusion at C2-C3. As with KF1, more caudal
fusions may have been present. Klippel-Feil 2 had an au-
tosomal dominant inheritance pattern with 100% pen-
etrance of the C2-C3 fusion. Isolated fusions of the cervi-
cal spine caudal to C1 and C2-C3 were defined as Klippel-
Feil 3 (KF3), which showed reduced penetrance and either
autosomal dominant or autosomal recessive inheritance.
Klippel-Feil 4 (KF4) was synonymous with Wildervanck
syndrome (congenital cervical synostosis, congenital hear-
ing loss, and the Duane anomaly). This syndrome was
thought to have an X-linked dominant inheritance pattern
with hemizygous lethality.3,27
Genetic advances have led to another reexamination of
the radiologic classification of syndromes characterized by
congenital vertebral malformations (fusions, hemiverte-
brae), including Klippel-Feil syndrome. Because somito-
genesis generates vertebral structures in a rostral-to-caudal
direction, we predict that a continuous spectrum of con-
genital vertebral defects should exist along the craniocau-
dal axis. Furthermore, because segmental defects can
manifest as various types of dysmorphic features (verte-
bral body fusions, neural arch fusions, wedge or hemiver-
tebrae) and cause curvatures in various axes, we think that
the localization and extent of cervical vertebral defects are
more informative indicators of the etiology of Klippel-Feil
syndrome. However, until the genetic etiology of Klippel-
Feil syndrome is clarified more precisely, it is impossible
to know if the immense heterogeneity seen in patients with
the syndrome is a result of disruption of single genes,
complex interactions between variant forms of multiple
genes, or gene-environment interactions leading to anoma-
lies in the axial patterning of the cervical spine.33 Further-
more, it is unclear if the diagnosis of Klippel-Feil syn-
drome is a discrete entity, or if it is one point on a spec-
trum of congenital spinal deformities, which includes
congenital scoliosis and spondylocostal dysostosis or Jar-
cho-Levin syndrome.
Only by identifying the link between the genetic etiol-
ogy and the phenotypic pathoanatomy of Klippel-Feil syn-
drome will we be able to rationalize the heterogeneity of
the syndrome. Pedigree analysis has discovered the first
human Klippel-Feil locus (SGM1), however additional
characterization of this gene and its function are needed.
Although mouse models suggest that HOX genes are un-
likely to cause extensive cervical spine fusions in Klippel-
Feil syndrome, PAX and Notch pathway genes present
promising etiologic candidate genes. We are evaluating
cervical and thoracolumbar radiographs and sequencing
multiple Notch pathway genes in pediatric patients diag-
nosed with Klippel-Feil syndrome to determine if the data
support these hypotheses.
Klippel-Feil: Features and Etiology 189
Acknowledgments
We thank Vanessa Garcia, Carrie Grabowicz, Julia Lou, Akshay
Mehta, Mizuho Mimoto, Megan O’Brien, and Yana Tsygansky
for technical assistance.
References
1. Bulman MP, Kusumi K, Frayling TM, et al: Mutations in the human
delta homologue, DLL3, cause axial skeletal defects in spondylo-
costal dysostosis. Nat Genet 24:438–441, 2000.
2. Chi N, Epstein JA: Getting your Pax straight: Pax proteins in de-
velopment and disease. Trends Genet 18:41–47, 2002.
3. Clarke RA, Catalan G, Diwan AD, Kearsley JH: Heterogeneity in
Klippel-Feil syndrome: A new classification. Pediatr Radiol
28:967–974, 1998.
4. Clarke RA, Kearsley JH, Walsh DA: Patterned expression in famil-
ial Klippel-Feil syndrome. Teratology 53:152–157, 1996.
5. Clarke RA, Singh S, McKenzie H, Kearsley JH, Yip MY: Familial
Klippel-Feil syndrome and paracentric inversion inv(8)(q22.2q23.3).
Am J Hum Genet 57:1364–1370, 1995.
6. Condie BG, Capecchi MR: Mice homozygous for a targeted dis-
ruption of Hoxd-3 (Hox-4.1) exhibit anterior transformations of the
first and second cervical vertebrae, the atlas and the axis. Develop-
ment 119:579–595, 1993.
7. Copley LA, Dormans JP: Cervical spine disorders in infants and
children. J Am Acad Orthop Surg 6:204–214, 1998.
8. Da Silva EO: Autosomal recessive Klippel-Feil syndrome. J Med
Genet 19:130–134, 1983.
9. David KM, Thorogood PV, Stevens JM, Crockard HA: The dys-
morphic cervical spine in Klippel-Feil syndrome: Interpretations
from developmental biology. Neurosurg Focus 6:1–11, 1999.
10. Dietz F: Congenital Abnormalities of the Cervical Spine. In Wein-
stein SL (ed). The Pediatric Spine—Priniciples and Practice. Ed 2.
Philadelphia, Lippincott Williams & Wilkins 239–251, 2001.
11. Dormans JP: Evaluation of children with suspected cervical spine
injury. J Bone Joint Surg 84A:124–132, 2002.
12. Dormans JP, Crisciello AA, Drummond DS, Davidson RS: Com-
plications in children managed with immobilization in a halo vest.
J Bone Joint Surg 77A:1370–1373, 1995.
13. Dormans JP, Drummond DS, Sutton LN, Ecker ML, Kopacz KJ:
Occipitocervical arthrodesis in children. J Bone Joint Surg
77A:1234–1240, 1995.
14. Emerick KM, Rand EB, Goldmuntz E, et al: Features of Alagille
syndrome in 92 patients: Frequency and relation to prognosis. Hep-
atology 29:822–829, 1999.
15. Gundersen CH, Greenspan RH, Glaser GH, Lubs H: The Klippel-
Feil syndrome: Genetic and clinical reevaluation of cervical fusion.
Medicine (Baltimore) 46:491–512, 1967.
16. Heller JG, Klekamp JW, Blechner MH: Posterior Cervical Instru-
mentation. In Emery SE, Boden SD (eds). Surgery of the Cervical
Spine. Philadelphia, WB Saunders 52–75, 2003.
17. Hensinger RN, Lang JE, MacEwen GD: Klippel-Feil syndrome: A
constellation of associated anomalies. J Bone Joint Surg 56A:1246–
1253, 1974.
18. Hensiger RN, MacEwen GD: Congenital Anomalies of the Spine. In
Rothman RH, Simeone FA (eds). The Spine. Philadelphia, WB
Saunders 188–315, 1982.
19. Herman MJ, Pizzutillo PD: Cervical spine disorders in children.
Orthop Clin North Am 30:457–466, 1999.
20. Juberg RC, Gershanik JJ: Cervical vertebral fusion (Klippel-Feil) syn-
drome with consanguineous parents. J Med Genet 13:246–249, 1976.
21. Keynes RJ, Stern CD: Mechanisms of vertebrate segmentation. De-
velopment 103:413–429, 1988.
22. Klippel M, Feil A: The classic: A case of absence of cervical ver-
tebrae with the thoracic cage rising to the base of the cranium
(cervical thoracic cage). Clin Orthop 109:3–8, 1975.
23. Krumlauf R: Hox genes in vertebrate development. Cell 78:191–
201, 1994.
24. Lander ES, Linton LM, Birren B, et al: Initial sequencing and analy-
sis of the human genome. Nature 409:860–921, 2001.

190 Tracy et al
25. Larson AR, Josephson KD, Pauli RM, Opitz JM, Williams MS:
Klippel-Feil anomaly with Sprengel anomaly, omovertebral bone,
thumb abnormalities, and flexion-crease changes: Novel association
or syndrome? Am J Med Genet 101:158–162, 2001.
26. Li L, Krantz ID, Deng Y, et al: Alagille syndrome is caused by
mutations in human Jagged 1, which encodes a ligand for Notch 1.
Nat Genet 16:243–251, 1997.
27. Lowry RB, Jabs EW, Graham GE, Gerritsen J, Fleming J: Syn-
drome of coronal craniosynostosis, Klippel-Feil anomaly, and
Sprengel shoulder with and without Pro250Arg mutation in the
FGFR3 gene. Am J Med Genet 104:112–119, 2001.
28. Lufkin T, Mark M, Hart CP, et al: Homeotic transformation of the
occipital bones of the skull by ectopic expression of a homeobox
gene. Nature 359:835–841, 1992.
29. Manaligod JM, Bauman NM, Menezes AH, Smith RJH: Cervical
vertebral anomalies in patients with anomalies of the head and neck.
Ann Otol Rhinol Laryngol 108:925–933, 1999.
30. National Center for Biotechnology Information Literature Da-
tabases page: Online Mendelian inheritance in man. Available
at http://www.ncbi.nlm.nih.gov/Omim. Accessed January 5,
2004.
31. Oda T, Elkahloun AG, Pike BL, et al: Mutations in the human
Jagged1 gene are responsible for Alagille syndrome. Nat Genet
16:235–242, 1997.
32. Pizzutillo PD, Woods M, Nicholson L, MacEwen GD: Risk factors
in Klippel-Feil syndrome. Spine 19:2110–2116, 1994.
Clinical Orthopaedics
and Related Research
33. Pourquie O, Kusumi K: When body segmentation goes wrong. Clin
Genet 60:409–416, 2001.
34. Rouvreau P, Glorion C, Langlais J, Noury H, Pouliquen JC: As-
sessment and neurologic involvement of patients with cervical spine
congenital synostosis as in Klippel-Feil syndrome: Study of 19
cases. J Pediatr Orthop B 7:179–185, 1998.
35. Sadler TW: Embryonic Period (fourth to eighth week). In Langman
J (ed). Langman’s Medical Embryology. Baltimore, Williams &
Wilkins 58–67, 1985.
36. Saga Y, Takeda H: The making of the somite: Molecular events in
vertebrate segmentation. Nat Rev Genet 2:835–845, 2001.
37. Scoliosis Research Society Public Information Page: Glossary.
Available at http://www.srs.org/patients/glossary.asp. Accessed
January 4, 2004.
38. Talwalker VR, Mencio GA: The Pediatric Spine. In Baratz ME,
Watson AD, Imbriglia JE (eds). Orthopaedic Surgery: The Essen-
tials. New York, Thieme 711–726, 1999.
39. Theiss SM, Smith MD, Winter RB: The long term follow-up of
patients with Klippel-Feil syndrome and congenital scoliosis. Spine
22:1219–1222, 1997.
40. Thomsen MN, Schneider U, Weber M, Johannisson R, Niethard FU:
Scoliosis and congenital anomalies associated with Klippel-Feil
syndrome types I-III. Spine 22:396–401, 1997.
41. Warner WC: Pediatric Cervical Spine. In Canale ST (ed). Camp-
bell’s Operative Orthopaedics. Ed 9. St Louis, Mosby 2815–2847,
1998.