Arachnodactyly - A key to diagnosing heritable disorders of connective tissue

Abstract

Arachnodactyly literally means spidery fingers, and describes the long, slender fingers typical of patients with Marfan syndrome (MFS). Many clinicians regard arachnodactyly as pathognomonic of MFS; however, this view is misleading as arachnodactyly is a key element of the marfanoid habitus, which is present in several heritable disorders of connective tissue (HDCTs). Other features of the marfanoid habitus include long hands and feet, increased skin stretch, joint hypermobility and characteristic changes in the physiology of the pectum. Here, we focus on the differential diagnosis of diseases with features of the marfanoid habitus. Ectopia lentis (lens dislocation) and aortic root dilation or dissection are cardinal features of MFS. Distinguishing MFS from other HCDTs has important implications for treatment, as cardiovascular and ocular complications commonly seen in patients with MFS are not seen in all HDCTs. Joint hypermobility syndrome and Ehlers-Danlos syndrome are also HDCTs, neither of which is associated with ectopia lentis or aortic changes. Some of the rarer forms of Ehlers-Danlos syndrome are associated with severe vascular, dental and skin pathologies. This Review serves as a guide for correctly diagnosing members of the HDCT family.

Full text

NATURE REVIEWS
|
RHEUMATOLOGY ADVANCE ONLINE PUBLICATION
|
1
Centre for
Rheumatology,
University College
Hospital London, Third
Floor Central, 250
Euston Road, London
NW1 2PQ, UK
(R.Grahame). Whipps
Cross University
Hospital, Barts Health
NHS Trust, London
E111NR, UK
(A.J.Hakim).
Correspondence to:
R. Grahame
r.grahame@ucl.ac.uk
Arachnodactyly—a key to diagnosing
heritable disorders of connective tissue
Rodney Grahame and Alan J. Hakim
Abstract | Arachnodactyly literally means spidery fingers, and describes the long, slender fingers typical
of patients with Marfan syndrome (MFS). Many clinicians regard arachnodactyly as pathognomonic of
MFS; however, this view is misleading as arachnodactyly is a key element of the marfanoid habitus, which
is present in several heritable disorders of connective tissue (HDCTs). Other features of the marfanoid
habitus include long hands and feet, increased skin stretch, joint hypermobility and characteristic changes
in the physiology of the pectum. Here, we focus on the differential diagnosis of diseases with features of
the marfanoid habitus. Ectopia lentis (lens dislocation) and aortic root dilation or dissection are cardinal
features of MFS. Distinguishing MFS from other HCDTs has important implications for treatment, as
cardiovascular and ocular complications commonly seen in patients with MFS are not seen in all
HDCTs. Joint hypermobility syndrome and Ehlers–Danlos syndrome are also HDCTs, neither of which is
associated with ectopia lentis or aortic changes. Some of the rarer forms of Ehlers–Danlos syndrome
are associated with severe vascular, dental and skin pathologies. This Review serves as a guide for correctly
diagnosing members of the HDCT family.
Grahame, R. & Hakim, A.J. Nat. Rev. Rheumatol. advance online publication 12 March 2013; doi:10.1038/nrrheum.2013.24
Introduction
Arachnodactyly—spidery fingers—is an easily recognised
physical sign that might indicate the presence of a herit-
able disorder of connective tissue (HDCT). Arachno-
dactyly should never be considered in isolation, but as a
feature of the marfanoid habitus, a range of skeletal dispro-
portions associated with increased length and decreased
breadth of long bones. Incomplete forms of the marfa-
noid habitus are commonplace and likely to be benign;
however, their presence should trigger an assessment for
features of HDCTs. Although the marfanoid habitus is
primarily associated with Marfan syndrome (MFS), MFS
is only one potential diagnosis. In particular, joint hyper-
mobility syndrome (JHS), a far more prevalent HDCT
than MFS, is also associated with the marfanoid habitus.
The HDCTs comprise a family of genetically-determined
diseases caused by mutations in genes encoding connec-
tive tissue matrix proteins (for example, collagen, elastin,
fibrillin and tenascin). Clinicians often find diagnosis
of these conditions challenging because of overlapping
symptoms and the broad spectrum of physical findings.
The aim of this article is to enable clinicians to make an
accurate diagnosis in cases where arachnodacty ly and
other signs of HDCT have been identified.
Clinical investigations of HDCTs
A number of investigations can be performed in the clinic
to identify potential HDCTs. The need for further testing
is determined by the identity of the putative HDCT.
Arachnodactyly
Arachnodactyly is initially recognized by observing the
size (notably the length and slenderness) of the patient’s
hands and fingers during the course of a consultation
(Figure1a). The acid test for the confirmation of arachno-
dactyly is the performance of the tests for the Steinberg1
and Walker–Murdoch2 signs (Figure1); historically, the
metacarpal index has also been used.3
In the Steinberg test, the patient actively adducts the
thumb across the palm as far as possible and then closes
the fist, thereby enclosing the thumb. The test is positive
if it is possible to extend the thumb so that it projects
beyond the edge of the hypothenar eminence (Figure1b).
The test for the Walker–Murdoch sign requires the
patient to actively encircle their wrist with the contra-
lateral thumb and little finger. A positive test is denoted by
the ability of the two relevant fingernails to overlap by a
complete fingernail length (Figure1c). Prior to the intro-
duction of the Steinberg and Walker–Murdoch tests, the
standard approach to identifying arachnodactyly was to
calculate the metacarpal index. This calculation required
a plain radiograph of the hands; the metacarpal length is
divided by the midpoint girth for each of the digits (omit-
ting the thumb). The mean of the four ratios constitutes
the metacarpal index for that hand. Values >8.6 are con-
sidered to be indicative of arachnodactyly, whilst values
<8.0 are normal. Values between 8.0 and 8.6 do not clearly
indicate the presence or absence of arachnodactyly.
With the advent of the Steinberg and Walker–Murdoch
tests, the metacarpal index became redundant as expo-
sure to ionising radiation would no longer be justified or,
in fact,needed.
Competing interests
The authors declare no competing interests.
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved

2
|
ADVANCE ONLINE PUBLICATION www.nature.com/nrrheum
The marfanoid habitus
Having confirmed the presence of arachnodactyly, the
next stage is to establish the presence of other features of
the marfanoid habitus (Supplementary Figure1). These
features are identified by careful observation supple-
mented by the calculation of four ratios. The first ratio,
the span:height ratio (SHR) is measured by asking the
patient to stand facing and in contact with a wall with
arms, hands and fingers fully extended (horizontally out-
stretched) with the tip of the middle finger of one hand
in contact with a side wall. The distance (measured with
a tape measure) from the tip of the contralateral hand to
the wall is then taken to represent the arm span. Arm span
is divided by the patient’s height to obtain the SHR. The
second and third ratios used are the hand:height ratio
(HHR) and the foot:height ratio (FHR). Hand length is
the distance between the distal palmar crease and the tip
of the middle finger; foot length is the distance between
the base of the posterior edge of the heel and the tip
of the hallux. Hand length and foot length are divided by
the patient’s height to obtain the HHR and FHR, respec-
tively. The final ratio used to determine the presence of
the marfanoid habitus is the length of the upper part of the
body to the length of the lower part of the body (upper
segment [US]:lower segment [LS] ratio; USLSR). The LS
is the distance from the midpoint of the pubic symphy-
sis to the floor with the patient standing erect. The US is
calculated by subtracting the LS from the patient’s height;
thus, USLSR = (height – LS)/LS. Based on the original data
collected by McKusick4 four decades ago, the cutoff levels
for determining the marfanoid habitus are >1.03 for SHR,
>0.11 for HHR, >0.15 for FHR and <0.89 for USLSR.
The marfanoid habitus is not exclusively defined by
these four ratios: other features of the marfanoid habitus
Key points
■Arachnodactyly is a key feature of the marfanoid habitus and thus a reliable
pointer to the presence of a heritable disorder of connective tissue (HDCT)
■Joint hypermobility syndrome is the most commonly seen HDCT
■The differential diagnosis of arachnodactyly is wide and extends to include
many rare genetic syndromes
■Incomplete forms of the marfanoid habitus are highly prevalent and can be
benign, but should lead to an assessment for the presence of an HDCT
■Neither arachnodactyly nor the marfanoid habitus on its own can be regarded
as pathognomonic of Marfan syndrome in the absence of either cardiac or
ocular involvement
are dolichocephaly (skull width/length × 100 <76%),
scolio sis (Bunnell scoliometer reading >5°), pectus
de formity (excavatum or carinatum) (Figure2a), jaw
deformities with overcrowding of teeth, a high-arched
palate (Figure2b) and long feet (often with hammer toes)
that flatten and pronate on weight-bearing. The pene-
trance of these features is variable and not all features are
required to suggest the presence of the marfanoid habitus.
In a University College Hospital London cohort of 574
patients with JHS, 82 (14.3%) had complete marfanoid
habitus, and 43 (7.5%) had an incomplete form (authors’
unpublished data).
Joint hypermobility
Joint hypermobility is the most obvious and most con-
stant outward visible feature of an HDCT and its pres-
ence should always be sought. Joint hypermobility would
be expected to be present in most conditions where the
marfanoid habitus is present, but its degree varies enor-
mously both between the different conditions and within
them, and even within individual members of the same
family. Joint hypermobility is usually recorded by means
of the Beighton score,5 which gives a maximum score of
nine out of nine (Supplementary Box1). The Beighton
score tests the ability of the subject to perform a number
of manoeuvres and the score represents, perhaps sur-
prisingly, not the degree of hypermobility, but the wide-
spread nature of the condition among a selected number
of joints. A score of four or more suggests the presence of
generalized, as opposed to localized, hypermobility. It is
also valuable, however, to look at other joints for hyper-
mobility including the temporomandibular joint, cervical
spine, thoracic spine, shoulders, hips, ankles and feet.
Hyperextensible skin
Dermal collagen represents 70% of the dry weight of
skin, and skin has the advantage of being readily acces-
sible to clinical evaluation. The recognition of increased
skin stretch (hyperextensibility) can have a pivotal role
in the recognition and clinical diagnosis of an HDCT.
The texture is often described as velvety or silky and this
texture can be a striking feature. Thickness is a reliable
indicator of the quantity of collagen (or lack thereof) in
the dermis. The skinfold thickness may be measured by
means of the Harpenden caliper;6 a reduction in thickness
denotes collagen deficiency, which is a feature of several
a cb
Figure 1 | Arachnodactyly. a | Typical appearance of arachnodactyly. b | The Steinberg test for arachnodactyly. Note that the
thumb projects beyond the hypothenar aspect of the hand without assistance or effort. c | The Walker–Murdoch test for
arachnodactyly. With the thumb and little finger encircling the contralateral wrist there is obvious overlap amounting to the
length of a fingernail.
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved

NATURE REVIEWS
|
RHEUMATOLOGY ADVANCE ONLINE PUBLICATION
|
3
HDCTs. Semitransparency is manifest as ‘see-through’
skin, in which the underlying veins and tendons are
more clearly visible than normal. This feature is seen in
JHS (Box1), but more so in the vascular type of Ehlers–
Danlos syndrome (EDS–vascular type; formerly known
as EDS typeIV).
Hyperextensible skin is arguably the most consistently
reliable physical sign discriminating an HDCT from nor-
mality. In the authors’ view, the test is best performed by
raising a fold of skin overlying the third and fourth meta-
carpals of the right hand between the examiner’s thumb
and index finger and carefully observing the pattern of
stretching. In patients with an HDCT the initial stretch-
ing is seen to extend across the whole of the dorsum
of the hand and beyond. The difference in the process of
stretching between a patient with an HDCT (particularly
EDS) and one without lies in the degree of stretch that
takes place in the phase of ‘taking up slack’, the first phase
of skin stretch on the stress–strain curve (Figure3); it is
greater in the former.7, 8
Striae atrophicae are stretch marks that appear during
the adolescent growth spurt. They usually appear over the
lumbosacral, thigh, knee and shoulder regions, generally
between the ages of 11 and 13years. Their appearance is
suggestive of an HDCT, and striae atrophicae are found
notably, but not exclusively, in patients with JHS.
Impaired scar formation is another manifestation of
collagen deficiency. Scars are usually pale, thin, shiny,
easily-puckered and sunken below the level of sur-
rounding skin. Such characteristic scars provide further
ev idence of the possible presence of an HDCT.
HDCTs exhibiting arachnodactyly
MFS, EDS and JHS should all be considered as possible
diagnoses when arachnodactyly is observed. A working
knowledge of the classification criteria for MFS, EDS and
JHS is important in proffering a clinical diagnosis. The
criteria for each of these diseases—the revised Ghent,9
Villefranche,10 and Brighton11 criteria, respectively—
can distinguish between the different common types of
HDCTs associated with arachno dactyly and the marfa-
noid habitus that might present to the clinician (Boxes1,
2 and 3, Table1).12
Marfan syndrome
For the clinician confronted with a patient with arachno-
dactyly or the marfanoid habitus, the first consideration
should be whether or not the patient is likely to have true
MFS—a disease that carries a considerable risk of develop-
ing a potentially fatal progressive aortic root dilatation
leading to dissection or rupture, or both. However, MFS
cannot be diagnosed in the presence of arachnodactyly
and the marfanoid habitus alone.
The classification of MFS has been refined from the
established Ghent 1996 criteria,13 which defined a set
of major and minor manifestations in the presence or
absence of an associated mutation in the gene encoding
fibrillin-1 (FBN1) or a family history of MFS. Several
conditions share certain Marfan-like (marfanoid) mani-
festations, creating the risk of inadvertent erroneous
classification as MFS. The 2010 revised Ghent criteria9
(Box2) emphasize aortic root disease (dilatation and dis-
section) and ectopia lentis (lens dislocation) as cardinal
a b
Figure 2 | Selected features of the marfanoid habitus. a | Pectus excavatum with
furrowed ribs. b | High-arched palate.
Figure 3 | Increased skin stretch in a patient with
Ehlers–Danlos syndrome during phase of taking up slack.
The identification of skin hyperextensibility by clinical
investigation can be essential to the diagnosis of heritable
disorders of connective tissues.
Box 1 | Revised Brighton 1998 criteria for joint hypermobility syndrome11
Joint hypermobility syndrome is diagnosed in the presence of two major criteria, or
one major and two minor criteria, or four minor criteria. Two minor criteria suffice
where there is an unequivocally affected first-degree relative. Joint hypermobility
syndrome is excluded by the presence of Marfan syndrome or EDS (other than
EDS–hypermobility type [formerly known as EDS type III]) as defined by the Ghent
199613 and the Villefranche 199710 criteria, respectively. The first major and
minor criteria are mutually exclusive, as are the second major and minor criteria.
Major criteria
■Beighton score* of four or more (out of nine), either currently or historically
■Arthralgia of duration >3months in four or more joints
Minor criteria
■Beighton score* of 1–3 (or 0–3 if patient is age ≥50 years)
■Arthralgia (duration >3months) in one to three joints or back pain (duration
>3months), spondylosis or spondylolisthesis
■Dislocation or subluxation in more than one joint, or in one joint on more than
one occasion
■Soft tissue rheumatism with more than three lesions (e.g. epicondylitis,
tenosynovitis or bursitis)
■Marfanoid habitus, features of which include: tall, slim build; span:height ratio
>1.03; upper segment:lower segment ratio <0.89; arachnodactyly (positive
Steinberg and Walker–Murdoch signs)
■Abnormal skin: striae, hyperextensibility, thin skin or papyraceous scarring
■Eye signs: drooping eyelids, myopia or antimongoloid slant
■Varicose veins, hernia or uterine or rectal prolapse
*The nine-point Beighton hypermobility score5 is detailed in Supplementary Box 1.
Abbreviation: EDS, Ehlers–Danlos syndrome. Adapted with permission from Hypermobility,
Fibromyalgia and Chronic Pain (Eds Hakim, A., Keer, R. & Grahame, R.) Elsevier 2010.
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved

4
|
ADVANCE ONLINE PUBLICATION www.nature.com/nrrheum
features of MFS; FBN1 mutations and systemic features
(Box3) also contribute to this diagnosis.
Cardiovascular complications of MFS and MFS-like
conditions can be life-threatening. Mitral valve pro-
lapse is a common feature of MFS (and some of the
rarer genetic HDCT variants), but is less common in
patients with EDS or JHS. Mitral valve prolapse can
present as atypical chest pain or symptoms suggestive
of extra or dropped beats and palpitations. Spontaneous
pneumothorax also occurs in patients with MFS but
rarely occurs in patients with EDS or JHS.
The possible presence of these pathologies should be
investigated with echocardiography, Doppler ultrasound
of the abdominal vasculature and, if there is a risk of
aortic dissection or dilation, CT angiography or mag-
netic resonance angiography of the aortic root, arch
and thoracic aorta. Annual echocardiography should be
performed, not only to exclude ‘emergent’ new patholo-
gies but also to assess known aortic dilatation and mitral
valve prolapse in MFS as both are progressive; in indivi-
duals with MFS with normal aortic diameter repeat
imaging is advisable every 2–3years. The threshold for
detailed imaging with magnetic resonance angiography
and preparation for surgical intervention is an aortic
diameter ≥4.5 cm, annual increases in aorta diameter
>0.5 cm, or a diameter at the sinus of Valsalva ≥5.0 cm,
with or without aortic valve dysfunction.
Joint hypermobility and EDS
EDS is a subfamily within the HDCTs. Several types
of EDS have been differentiated based on clinical, bio-
chemical and molecular or genetic differences. They
range from the benign ‘hypermobile’ type at one end of
the spectrum to the life-threatening ‘vascular type’ at the
other (Table1).
Many experts, the authors included, consider EDS–
hypermobility type (formerly known as EDS type III)
to be the same as JHS.14 As in JHS, therefore, the mar-
fanoid habitus is considered to be a common finding
in EDS–hypermobility type; however, the marfanoid
habitus is much less common in the other, rarer forms
of EDS. JHS is a common yet often overlooked15 condi-
tion that can manifest many physical features of HDCT.
JHS is diagnosed using the Brighton criteria (Box1).11
In addition to the signs and symptoms shown in Box1,
JHS is associated with complex systemic symptoms such
as fatigue, autonomic dysfunction, gastrointestinal dys-
motility, anxiety and phobic disorders.16–18 Although
JHS can present with considerable morbidity and signs
that overlap with other HDCTs, it is not associated with
the life-threatening vascular or ocular pathologies of
MFS, or with the more severe skin, vascular and dental
pathologies of the more rare forms of EDS.
The classification tables for EDS, MFS and JHS provide
a concise source of information upon which to build a
clinical history and physical assessment when manag-
ing patients with hypermobility (Boxes1 and 2, Table1).
From this information we have developed a ‘decision tree’
that outlines investigations that should be performed to
differentially diagnose these disorders (Figure4).19
With respect to cardiovascular complications, in
EDS–vascular type there is an increased risk of rupture—
possibly concurrent with dilatation or dissection—of
medium-sized vessels such as renal and splenic arteries.
Other variants of EDS and JHS are not associated with
arterial disease. Mitral valve prolapse is rare in patients
with EDS and JHS and is usually a chance finding of
limited clinical relevance. Spontaneous pneumothorax
rarely occurs in patients with EDS or JHS.
Box 3 | Scoring* of systemic features of Marfan syndrome9
One point is given for each of the following eight features:
■The presence of at least three of the following five facial features:
dolichocephaly, enophthalmus, downslanting palpebral fissures, malar
hypoplasia and retrognathia
■Reduced elbow extension
■Pectus excavatum or chest asymmetry (excluding carinatum, which scores two
points, as below)
■Reduced upper segment:lower segment ratio and increased arm span:height
ratio in the absence of severe scoliosis
■Scoliosis or thoracolumbar kyphosis
■Skin striae
■Mitral valve prolapse
■Myopia >3 dioptres
Two points are given for each of the following five features:
■Pectus carinatum
■Pneumothorax
■Dural ectasia
■Protrusio acetabulae
■Hind foot deformity (one point only is given if plain flat foot [pes planus] )
Three points are given for:
■Positive Steinberg sign and Walker–Murdoch sign (one point is given if only one
of the two signs is present)
*Maximum score is 20. A score ≥7 is indicative of Marfan syndrome. Permission obtained
from BMJ Publishing Group. Loeys, B.L. etal. J.Med. Genet. 47, 476–485 (2010).
Box 2 | Revised Ghent criteria (2010) for the diagnosis of Marfan syndrome9
In the absence of a family history of Marfan syndrome, a diagnosis is made in the
presence of one of four situations:
■Aortic root Z-score* ≥2 (or dissection) AND ectopia lentis, irrespective of
systemic features except where these are indicative of other conditions‡
■Aortic root Z-score* ≥2 (or dissection) AND systemic features score§ ≥7 except
where these are indicative of other conditions‡
■Aortic root Z-score* ≥2 (or dissection) AND a recognized FBN1 mutation
(or definite linkage to a predisposing FBN1 haplotype)
■Ectopia lentis AND FBN1 mutation unequivocally associated with aortic
disease; aortic root dilatation or dissection may be absent at the time
of assessment
With a family history of Marfan syndrome (defined by any of the above), a
diagnosis is made in the presence of one of three situations:
■Ectopia lentis
■Systemic features score§ ≥7 except where these features are indicative of
other conditions‡
■Aortic root Z-score* ≥2 (or dissection) in an individual age ≥20years, or aortic
root Z-score* ≥3 (or dissection) if age <20years
*Z-score is a measure of the aortic root (sinus of Valsalva) diameter standardized for age and
body size. ‡The following conditions should be excluded: vascular Ehlers–Danlos syndrome,
Shprintzen–Goldberg syndrome (FBN1 mutation, craniosynostosis and mental retardation) and
Loeys–Dietz syndrome (TGFBR1 or TGFBR2 mutations, bifid uvula, cleft palate,
craniosynostosis, hypertelorism, cervical spine instability, thin velvety skin, easy bruising and
arterial tortuosity and aneur ysms). §Scoring of systemic features of Marfan syndrome is
detailed in Box1. Abbreviations: FBN1, fibrillin 1; TGFBR1, transforming growth factor β
receptor 1; TGFBR2, transforming growth factor β receptor 2. Permission obtained from BMJ
Publishing Group. Loeys, B.L. etal. J.Med. Genet. 47, 476–485 (2010).
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved

NATURE REVIEWS
|
RHEUMATOLOGY ADVANCE ONLINE PUBLICATION
|
5
Rare genetic syndromes
The marfanoid habitus forms an important common
feature in a number of recently described, though seem-
ingly unrelated, rare genetic syndromes. These syn-
dromes form part of the greater family of HDCTs, and
their discovery provided much of the reasoning behind
the 2010 revision of the Ghent Criteria.9
Two syndromes, homocystinuria20 and ec topia
lentis syndrome (ELS),21 have both marfanoid features
and ectopia lentis. ELS is associated with mutations in
FBN1 and ADAMTSL4. Patients with ELS do not have
aortic root disease and can therefore be distinguished
from patients with either MFS or homocystinuria.
Homocystinuria is an autosomal recessive defect in
methionine metabolism caused by a deficiency in cys-
tathionine β-synthase. This metabolic defect is also
associ ated with a deficiency in methylenetetrahydrofolate
reductase that can be assessed by measuring methio-
nine synthesis. Patients often present with nephro-
lithiasis. Characteristic features of homocystinuria
include arachnodactyly, ectopia lentis and myopia.
How ever, the vascular pathology for patients with homo-
cystinuria is thromboembolism, rather than dilatation
or dissection as seen in MFS. Mental retardation is also
a prominent feature of homocystinuria. The condition
is confirmed by finding homocystine in the serum and
urine. Homocystinuria is treated by oral administration
of vitamin B6 (pyridoxine) and a low-methionine diet.
Three syndromes have features of the marfanoid
habitus as well as some of the cardiovascular features
seen in MFS. Loeys–Dietz syndrome (familial aortic
dissection) is an autosomal dominant genetic syn-
drome, similar to Marfan syndrome, and is caused
by mutations in the genes encoding TGF-β recep-
tor type-1 or type-2 (TGFBR1 and TGFBR2, respec-
tively).22,23 Patients with Loeys–Dietz syndrome
frequently have aggressive arterial aneurysms (mean
age at death=26.1years).24 Characteristic craniofacial
features include ocular hypertelorism, bifid uvula, cleft
palate and craniosynostosis.
Table 1 | Classification of Ehlers–Danlos syndromes by the 1997 Villefranche nosology10
Type (old
classification in
parentheses)
Clinical manifestations IP Protein affected
(% of cases with
protein abnormality)
Gene
Major criteria Minor criteria
Classic (typeI or II) Skin hyperextensibility
Widened atrophic scarring
Joint hypermobility
Easy bruising
Smooth and velvety skin
Molluscoid pseudotumors
Subcutaneous spheroids
Muscular hypotonia
Complications of joint
hypermobility
Surgical complications
Positive family history
AD Type V procollagen
(~90%)
COL5A1
COL5A2
Hypermobility
(typeIII)
Generalized joint
hypermobility
Mild skin involvement
Recurring joint dislocations
Chronic joint pain
Positive family history
AD Tenascin-X (~5%) TNXB
Vascular (typeIV) Excessive bruising
Thin, translucent skin
Arterial, intestinal or uterine
fragility or rupture
Characteristic facial
appearance
Acrogeria
Early-onset varicose veins
Hypermobility of small joints
Tendon and muscle rupture
Arteriovenous or carotid-cavernous
sinus stula
Pneumo (hemo)thorax
Positive family history, sudden
death in close relative(s)
AD TypeIII procollagen COL3A1
Kyphoscoliotic
(typeVI)
Severe muscular hypotonia
at birth
Generalized joint laxity
Kyphoscoliosis at birth
Scleral fragility and rupture
of the ocular globe
Tissue fragility, including
atrophic scars
Easy bruising
Arterial rupture
Marfanoid habitus
Microcornea
Osteopenia
AR Lysyl hydroxylase 1
(type VIA)
D4ST (type VIB)
PLOD1
CHST14
Arthrochalasis
(typeVIIA or VIIB)
Severe generalized joint
hypermobility with recurrent
subluxations
Congenital bilateral hip
dislocation
Skin hyperextensibility
Tissue fragility, including
atrophic scars
Easy bruising
Muscular hypotonia
Kyphoscoliosis
Mild osteopenia
AD TypeI procollagen COL1A1
COL1A2
Dermatosparaxis
(typeVIIC)
Severe skin fragility
Sagging, redundant skin
Excessive bruising
Soft, doughy skin texture
Premature rupture of membranes
Large herniae
AR Procollagen I
N-proteinase
ADAMTS2
Abbreviations: AD, autosomal dominant; AR, autosomal recessive; IP, inheritance patter n. Reproduced from Hypermobility, Fibromyalgia and Chronic Pain
(Eds Hakim. A., Keer, R. & Grahame, R.) Elsevier 2010 with permission.
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved

6
|
ADVANCE ONLINE PUBLICATION www.nature.com/nrrheum
MASS phenotype25,26 is a connective tissue disorder
named anachronistically after the clinical characteristics
seen: mitral valve prolapse, myopia, borderline and non-
progressive aortic enlargement and skin and skeletal sys-
temic features scoring ≥5 as defined in Box3 and at least
one skeletal feature. MASS phenotype is associated with
mutations in FBN1; notably, patients with MASS do not
have ectopia lentis. Mitral valve prolapse syndrome shares
many of the characteristic features of MASS; patients
with mitral valve prolapse syndrome27 carry mutations
in FBN1 and do not have ectopia lentis. MASS and mitral
valve prolapse syndrome can be distinguished by the mild
aortic root dilatation (Z-score <2) seen in patients with
mitral valve prolapse syndrome and by the extent of the
marfanoid features. Patients with mitral valve prolapse
syndrome usually have a systemic features score of <5
whereas patients with MASS have a score of ≥5.
A number of syndromes have features of the marfa-
noid habitus, but have neither ectopia lentis nor any of
the cardiovascular features associated with MFS. Some of
these syndromes have characteristic skeletal morpholo-
gies. In Shprintzen–Goldberg syndrome, cranio synostosis
is associated with severe exophthalmos, maxillary and
mandibular hypoplasia, soft tissue hypertrophy of
the palatal shelves, low-set ears with soft and pliable
au ricles, multiple abdominal hernias, arachnodactyly
and camptodactyly (fixed flexion deformity of the inter-
phalangeal joints of the little finger).28 No FBN1 or
TGFBR1 or TGFBR2 mutations have been found in
patients with Shprintzen–Goldberg syndrome. Van
Den Ende–Gupta syndrome29 (craniofacial and skeletal
manifestations with arachnodactyly) is an extremely rare
autosomal recessive disorder characterized by distinc-
tive craniofacial features, which include blepharophimo-
sis, malar hypopoplasia, maxillary hypoplasia, a narrow
and beaked nose and an everted lower lip. Other features
are arachnodactyly, camptodactyly, peculiar skeletal
abnormalities and normal development and intelligence.
Patients with Beals–Hecht syndrome (congenital contrac-
tural arachnodactyly)30,31 have marfanoid features with
joint contractures associated with mutation in the gene
encoding fibrillin-2 (FBN2).
Patients presenting with mental retardation and the
marfanoid habitus might have Lujan–Fryns syndrome.
This X-linked syndrome is also associated with psycho-
pathology with behavioural abnormalities. Lujan– Fryns
syndrome has been attributed to a missense mutation
in the gene encoding mediator of RNA polymerase II
transcription subunit 12 (MED12).32 This syndrome is
not associated with ectopia lentis or any of the cardio-
vascular features seen in MFS apart from a single case
report involving aortic root dilatation.32
Arachnodactyly
Skin hyperextensibility
Joint hypermobility
Delayed wound healing
Atrophic scarring
Easy bruising
Congentital dislocation of the
hips
Vascular fragility
Marfanoid habitus
Ectopia lentis
Aortic dilatation or dissection
Mitral valve prolapse
Spontaneous pneumothorax
Evaluation for other specic signs:
Ocular pathology
Dysmophic features
Mental retardation
Marfanoid habitus
Ectopia lentis
or
Mitral valve prolapse
or
Neither
EDS? MFS? Other syndromes?
Stickler syndrome, homocystinuria,
MASS phenotype, ELS, LDS,
MVP syndrome, etc
Molecular analysis of FBN1
Vascular imaging
Radiology (e.g. spine for
scoliosis)
Protein chemisty of collagen
type I, III, V
Molecular analysis of COL1A1,
COL1A2, COL3A1, COL5A1,
COL5A2, TNXB, PLOD1,
CH5T14, ADAMTS2
Radiology (e.g. scoliosis, hip
dysplasia)
Vessel imaging for
EDS–vascular type
There is no denitive test
for the diagnosis of
EDS–hypermobility type
Molecular analysis of FBN1,
FBN2, ADAMTSL4, TGFBR1,
TGFBR2, MED12
Analysis for homocystine
Fatigue
Marfanoid habitus
Joint hypermobility
Arthralgia
Anxiety or phobic disorders
There are no diagnostic
tests for JHS
JHS?
Figure 4 | Decision tree for diagnosing patients with heritable disorders of connective tissue. Patients with arachnodactyly
can be subclassified on the basis of the presence of other features, such as joint hypermobility, marfanoid habitus, ectopia
lentis and mitral valve prolapse. These features suggest a possible diagnosis of a heritable disorder of connective tissue,
which can often be confirmed using genotyping, protein chemistry or radiology. Abbreviations: EDS, Ehlers–Danlos
syndrome; ELS, ectopia lentis syndrome; JHS, joint hypermobility syndrome; LDS, Loeys–Dietz syndrome; MASS, mitral
valve prolapse, myopia, aortic enlargement, skin and skeletal features; MFS, Marfan syndrome; MVP, mitral valve prolapse.
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved

NATURE REVIEWS
|
RHEUMATOLOGY ADVANCE ONLINE PUBLICATION
|
7
Although arachnodactyly is not a typical sign of the
Stickler syndrome it is important to be aware of this
member of the HDCT family as, in particular, its ocular
presentation could easily be confused with that of MFS.
Stickler syndrome is subclassified by both ocular and
genetic findings. In most cases there are characteristic
congenital vitreous abnormalities, leading to an increased
risk of retinal detachment and haemorrhage (types I
andII); typeIII Stickler syndrome is a condition in which
the systemic features are present in the absence of ocular
pathology.33 In a recent survey by Stickler et al.,34 of 316
patients with typeI and II Stickler syndrome 95% had
ocular problems, with myopia in 90% of cases, retinal
detachment in 60% and blindness in 4%. Additionally,
90% of patients also had joint pain, usually secondary to
degenerative disease, 84% had problems with facial struc-
tures such as a flat face, cleft palate or small mandible, and
70% had some degree of hearing loss.
Conclusions
Arachnodactyly is an important physical sign that points
the way to the possible identification of an HDCT. It
should never be ignored or discounted until a full exami-
nation of the connective tissues has been performed and
a full clinical evaluation has been undertaken, supple-
mented by further investigations and by genetic testing
where appropriate. The differential diagnosis of arachno-
dactyly is wide-ranging and consequently should never be
regarded as being pathognomonic of Marfan syndrome.
It is hoped that with future advances in exome scan-
ning and other innovations in molecular genetics, it will
soon be possible to improve diagnostic precision by
means of genetic testing in the HDCTs. In the mean-
time, reliance must fall upon careful clinical observation
coupled with the application of existing internationally
agreed classification criteria.
Review criteria
A search for original articles published between 1953
and 2012 was performed in MEDLINE/PubMed using
the search terms “Marfan syndrome”, “Ehlers–Danlos
syndrome”, “marfanoid” and “arachnodactyly”, alone
and in various combinations. All articles identified were
English-language, full-text papers and abstracts. We
also searched the reference lists of identified articles
for further relevant papers. The reference list was last
updated in October 2012.
1. Steinberg, I. A simple screening test for the
Marfan syndrome. Am. J. Roentgenol. Radium
Ther. Nucl. Med. 97, 118–124 (1966).
2. Walker, B.A. & Murdoch, J.L. The wrist sign. A
useful physical finding in the Marfan syndrome.
Arch. Intern. Med. 126, 276–277 (1970).
3. Eldridge, R. The metacarpal index. A useful aid
in the diagnosis of the Marfan syndrome. Arch.
Intern. Med. 113, 248–254 (1964).
4. McKusick, V.A. The Marfan Syndrome. Heritable
Disorders of Connective Tissue, 4th edn 61–223
(C.V. Mosby, St Louis, 1972).
5. Beighton, P.H., Solomon, L. & Soskolne, C.
Articular mobility in an African population. Ann.
Rheum. Dis. 32, 413–417 (1973).
6. Tanner, J.M. & Whitehouse, R.H. The
Harpenden skinfold caliper. Am. J. Phys.
Anthropol. 13, 743–746 (1955).
7. Grahame, R. & Beighton, P. Physical properties
of the skin in the Ehlers–Danlos syndrome. Ann.
Rheum. Dis. 28, 246–251 (1969).
8. Farmer, A.D., Douthwaite, M.A., Gardiner, S.,
Aziz, Q. & Grahame, R. A novel in vivo skin
extensibility test for joint hypermobility.
J.Rheumatol. 37, 1513–1518 (2010).
9. Loeys, B.L. et al. The revised Ghent nosology
for the Marfan syndrome. J.Med. Genet. 47,
476–485 (2010).
10. Beighton, P., De Paepe, A., Steinmann, B.,
Tsipouras, P. & Wenstrup, R.J. Ehlers–Danlos
syndromes: revised nosology, Villefranche,
1997. Ehlers–Danlos National Foundation (USA)
and Ehlers–Danlos Support Group (UK). Am. J.
Med. Genet. 77, 31–37 (1998).
11. Grahame, R., Bird, H.A. & Child, A. The revised
(Brighton 1998) criteria for the diagnosis of
benign joint hypermobility syndrome (BJHS).
J.Rheumatol. 27, 1777–1779 (2000).
12. Beighton, P., Grahame, R. & Bird, H.A.
Hypermobility of Joints, 4th edn (Springer, London,
2011).
13. De Paepe, A., Devereux, R.B., Dietz, H.C.,
Hennekam, R.C. & Pyeritz, R.E. Revised
diagnostic criteria for the Marfan syndrome. Am.
J. Med. Genet. 62, 417–426 (1996).
14. Tinkle, B.T. et al. The lack of clinical distinction
between the hypermobility type of Ehlers–Danlos
syndrome and the joint hypermobility syndrome
(a.k.a. hypermobility syndrome). Am. J. Med.
Genet. A 149A, 2368–2370 (2009).
15. Ross, J. & Grahame, R. Joint hypermobility
syndrome. BMJ 342, c7167 (2011).
16. Hakim, A.J. & Grahame, R. Non-musculoskeletal
symptoms in joint hypermobility syndrome:
indirect evidence for autonomic dysfunction.
Rheumatology (Oxford) 43, 1194–1195 (2004).
17. Zarate-Lopez, N. et al. Unexplained
gastrointestinal symptoms and joint hypermobility:
is connective tissue the missing link?
Neurogastroenterol. Motil. 22, 252–262 (2010).
18. Bulbena, A. et al. Anxiety disorders in the joint
hypermobility syndrome. Psychiatry Res. 46,
59–68 (1993).
19. Malfait, F., Hakim, A.J., De Paepe, A. &
Grahame, R. The genetics of joint hypermobility
syndrome. Rheumatology (Oxford) 45, 502–507
(2006).
20. Homocystinuria due to cystathionine β-synthase
deficiency. Online Mendelian Inheritance in Man
[online], http://omim.org/entry/236200 (2012).
21. Familial ectopia lentis. National Marfan
Foundation [online], http://www.marfan.org/
marfan/2346/Familial-Ectopia-Lentis (2012).
22. Loeys, B.L. et al. Aneurysm syndromes caused
by mutations in the TGF-β receptor. N. Engl. J.
Med. 355, 788–789 (2006).
23. Van Hemelrijk, C. Renard, M. & Loeys, B. The
Loeys–Dietz syndrome: an update for the clinician.
Curr. Opin. Cardiol. 25, 546–551 (2010).
24. Loeys, B.L. & Dietz, H.C. Loeys–Dietz
syndrome. In: GeneReviews™ (eds
Pagon,R.A.et al.) [online], http://www.ncbi.
nlm.nih.gov/books/NBK1133/ (2012).
25. MASS phenotype. National Marfan Foundation
[online], http://www.marfan.org/marfan/2345/
MASS-Phenotype (2012).
26. Glesby, M.J. & Pyeritz, R.E. Association of mitral
valve prolapse and systemic abnormalities of
connective tissue. A phenotypic continuum.
JAMA 262, 523–528 (1989).
27. Familial mitral valve prolapse. Online Mendelian
Inheritance in Man [online], http://omim.org/
entry/157700 (2012).
28. Greally, M.T. et al. Shprintzen–Goldberg
syndrome: a clinical analysis. Am. J.Med. Genet.
76, 202–12 (1998).
29. Van den Ende–Gupta syndrome. Online
Mendelian Inheritance in Man [online], http://
omim.org/entry/600920?search=%E2%80%A2
%09Van%20Den%20Ende-Gupta%20
Syndrome&highlight=van%20endegupta%20
syndrome%20den (2012).
30. Putnam, E.A., Zhang, H., Ramirez, F. &
Milewicz,D.M. Fibrillin-2 (FBN2) mutations result
in the Marfan-like disorder, congenital contractural
arachnodactyly. Nat. Genet. 11, 456–458 (1995).
31. Callewaert, B.L. et al. Comprehensive clinical
and molecular assessment of 32 probands with
congenital contractural arachnodactyly: report of
14 novel mutations and review of the literature.
Hum. Mutat. 30, 334–341 (2009).
32. Van Buggenhout, G. & Fryns, J.P. Lujan–Fryns
syndrome (mental retardation, X-linked, marfanoid
habitus). Orphanet J. Rare Dis. 1, 26 (2006).
33. Rose, P.S. et al. Stickler syndrome: clinical
characteristics and diagnostic criteria. Am. J.
Med. Gen. A 138, 199–207 (2005).
34. Stickler, G.B., Hughes, W. & Houchin, P. Clinical
features of hereditary arthro-ophthalmopathy
(Stickler syndrome): a survey. Genet. Med. 3,
192–196 (2001).
Acknowledgements
The authors would like to thank their patients whose
photographs have been used in this article.
Author contributions
R. Grahame and A.J. Hakim made substantial
contributions to discussions of content and review/
editing of the manuscript before submission, and
R.Grahame researched data for the article and wrote
the article.
Supplementary information is linked to the online
version of the paper at www.nature.com/nrrheum
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved