Progress in Pediatric Cardiology 34 (2012) 9–14

Contents lists available at SciVerse ScienceDirect

Progress in Pediatric Cardiology

j ourna l homepage: www.e lsev ie r .com/ locate /ppedcard

Marfan syndrome: Progress report

Romy Franken a,b, Alexander W. den Hartog a,b, Michael Singh c, Gerard Pals d, Aeilko H. Zwinderman e,Maarten Groenink a,b,f, Barbara J.M. Mulder a,b,⁎a Department of Cardiology, Academic Medical Center, Amsterdam, The Netherlandsb Interuniversity Cardiology Institute of the Netherlands, Utrecht, The Netherlandsc Department of Cardiology, Children's Hospital, Boston, MA, United Statesd Department of Clinical Genetics and DNA diagnostics, VU University Medical Center, Amsterdam, The Netherlandse Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, Amsterdam, The Netherlandsf Department of Radiology, Academic Medical Center, Amsterdam, The Netherlands

⁎ Corresponding author at: Academic Medical Center,240, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlfax: +31 20 697 1385.

E-mail addresses: r.franken@amc.uva.nl (R. Franken)(A.W. den Hartog), michael.singh@cardio.chboston.org(G. Pals), a.h.zwinderman@amc.uva.nl (A.H. Zwinderma(M. Groenink), b.j.mulder@amc.uva.nl (B.JM. Mulder).

1058-9813/$ – see front matter © 2012 Elsevier Irelanddoi:10.1016/j.ppedcard.2012.05.003

a b s t r a c t

a r t i c l e i n f o

Keywords:

Aortic root dilatationClinical implicationsGhent criteriaLosartan therapyMarfan syndromeRevised Ghent criteria

Marfan syndrome is a multi-system connective tissue disorder, with primary involvement of the cardiovas-cular, ocular and skeletal systems. This autosomal heritable disease is mainly attributable to a defect in theFBN1 gene. Until 2010, the clinical diagnosis of Marfan syndrome was based on the Ghent criteria of 1996.Recently, the Ghent criteria have been revised. The revised guidelines of 2010 place more emphasis on aorticroot dilatation, ectopia lentis and FBN1 mutation testing in the diagnostic assessment of Marfan syndrome.Although the revised Ghent criteria of 2010 are easier to apply, they do raise some issues that need to beaddressed.In addition to adjustments in the diagnosis of Marfan syndrome, there is progress in the understanding of thepathophysiology in Marfan syndrome, leading to new treatment strategies. Losartan, an angiotensin II recep-tor type 1 blocker, has been shown to inhibit transforming growth factor beta signal transduction and there-by prevent aortic root aneurysms in a mouse model of Marfan syndrome. This article will provide a criticalappraisal of the revised Ghent nosology in 2010 and will highlight future perspectives regarding the treat-ment of Marfan syndrome.

© 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

1.1. Diagnosis

Marfan syndrome (MFS) is amulti-systemautosomal dominant her-itable disorder of the connective tissue with a prevalence of 1 per 5000individuals [1]. MFS is mainly caused by mutations in the fibrillin-1gene (FBN1) encoding for extracellular matrix protein fibrillin-1. Pa-tients with MFS may present with aortic dilatation, ectopia lentis,dural ectasia and skeletal features [2–8]. Aortic dilatation is present inthe vast majority of the patients with MFS and is the most importantcause for morbidity and mortality in MFS (Fig. 1) [9–12]. The diagnosisMFS is based on the Ghent criteria of 1996 in which the features havebeen divided into major and minor clinical criteria based on theirfrequency in MFS, in other conditions and in the general population

Department of Cardiology, B2-ands. Tel.: +31 20 566 2193;

, a.w.denhartog@amc.uva.nl(M. Singh), g.pals@vumc.nln), m.groenink@amc.uva.nl

Ltd. All rights reserved.

(Table 1) [13]. In the revised Ghent criteriamoreweight has been givento aortic root dilatation, FBN1 testing and ectopia lentis (Table 2) [14]. Inthis article we will discuss the historical evolution of MFS and provide acritical appraisal of the revised Ghent nosology of 2010.

1.2. Treatment

Treatment of MFS consists mainly of prophylactic surgical replace-ment of aortic root aneurysms (or other dilated parts of the aorta).These procedures have resulted in an increased life expectancy in MFS[15]. Before the aortic aneurysm reaches dimensions which will justifysurgical prophylactic intervention, beta-blocking agents are commonlyused to lower aneurysm expansion rate [16]. Recently, new possibilitiesfor medical therapies have emerged from animal experiments. MFS hasrecently been associated with increased Transforming Growth Factor(TGF)-β signalling [17,18]. TGF-β is a cytokine with diverse cellularfunctions, including cell proliferation and differentiation. It is involvedin cancer pathogenesis, immunity, tissue fibrosis and many other pro-cesses [19]. In addition to the role of fibrillin-1 as a structural compo-nent of the extracellular matrix, it regulates TGF-β activation throughinteractions with TGF-β precursors, keeping it in its inactive form. Dueto defective or deficient fibrillin-1, increased sequestration and activa-tion of TGF-β occurs, causing most of the disease features in a well-

Fig. 1.MR image of aorta in a Marfan patient. The arrow points to the typical aortic rootdilatation seen in most Marfan patients.

10 R. Franken et al. / Progress in Pediatric Cardiology 34 (2012) 9–14

establishedmouse-model of MFS. Many diseasemanifestations, includ-ing aortic root dilatation, mitral valve prolapse, developmental emphy-sema and skeletal muscle myopathy were attenuated after TGF-βinhibition using TGF-β neutralizing antibodies in MFS mice [20]. Inter-estingly, similar protection was achieved using the angiotensin II typeI receptor (AT1R) blocker losartan. This article will highlight future per-spectives regarding the treatment of MFS.

2. Progress in the diagnosis of Marfan syndrome

2.1. History

Marfan syndrome owes its name to Professor Antoine-BernardMar-fan, who described a five-year-old girl with long slender digits andother skeletal abnormalities in the Bulletin of the Medical Society ofParis in 1896 [21]. Ironically,most experts nowagree that the child orig-inally presented by Dr. Marfan was probably not affected by Marfansyndrome, but by ‘contractural arachnodactyly’, a form of connectivetissue disease caused by FBN2 mutations. After this initial delineation,other aspects of MFSwere recognized by VictorMcKusick. He describeda range of pleiotropic and phenotypic variability in MFS, includingcardiovascular aspects, ectopia lentis, skeletal pectus excavatum andkyphoscoliosis [22]. To ensure accurate communication between healthcare providers, researchers and patients, the Berlin nosology werepredefined clinical criteria for MFS in 1986 [23]. However, a pitfall ofthis nosologywas overdiagnosis of patients with a positive family histo-ry of MFS. These patients were sometimes only affected by non-specificconnective tissue findings themselves, without carrying the mutationpresent in more typically affected family members [24]. Therefore, in1996 the Ghent nosologywas reported as new diagnostic criteria to en-sure stringent guidelines for the diagnosis of MFS in unequivocally af-fected individuals and to separate other connective tissue disordersfromMFS [13]. The Ghent criteriamade use of a classification for the dif-ferent clinical features, whichwere divided intomajor criteriawith highdiagnostic specificity andminor criteria in which an organ system is in-volved (Table 1). The presence of a major criterion in two different sys-tems and a minor criterion in a third systemwere required to make thediagnosis of MFS. The prevalence of a FBN1mutation in patients fulfill-ing the criteria for MFS currently is 91–95% [14,25]. Nevertheless, the

sensitivity of the Ghent nosology was relatively low as most featuresof MFS are age dependent, which often postpones the diagnosis ofMFS in children with a positive family history [26]. Another concern re-garding the sensitivity of the Ghent nosology in diagnosing MFS werethe highly variable and diverse features, which are present as well inthe general population and in other connective tissue disorders [27].

2.2. The revised Ghent nosology

An international panel of experts in the diagnosis and managementof MFS aimed to improve the diagnostic criteria by following severalguiding principles: maximal evidence-based decision-making, practicalpatient centric implications, a focus on features that distinguish MFSfrom other disorders and definition of thresholds for diagnosis of MFS.

2.2.1. Purpose of the revised criteriaOne of the aims of the panel in the new criteria was to prevent over-

diagnosis of MFS, in order to protect patients inappropriately fromrestriction of career aspirations, denial of insurance benefits and psycho-social stigmatization. Therefore, one of the main revisions of the revisedGhent nosologywas the requirement of aortic root dilatation in a person-al and/or family history [14]. In absence of aortic root dilatation, identifi-cation of a pathogenic FBN1mutation is required in order to establish thediagnosis MFS. Presence of aortic root dilatation according to Z-score andectopia lentis is now sufficient for establishing the diagnosis of MFS in in-dividual patients. All other clinical features contribute to a ‘systemicscore’ (Table 2), which guides diagnosis when ectopia lentis or aortic dis-ease is absent. Less specific manifestations of MFS have been removed ormade less influential in the diagnostic evaluation. Even the major criteri-on ‘dural ectasia’ from the old Ghent criteria is converted into the system-ic score in the revised criteria. Dural ectasia is now given 2 points out of apossible 20 points, whereas more than 7 points are needed to indicatesystemic involvement.

Furthermore, a more prominent role is provided for moleculargenetic screening of the FBN1 gene and other relevant genes. The diag-nosis ofMFS should be avoided if a patient has specific clinical ormolec-ular observationswhich could reveal alternative and oftenmore seriousdiagnoses [14]. In these cases a delayed or ambiguous diagnosis of MFSis prevented by requiring additional diagnostic tests to prove or excludenewly introduced syndromes, such as ectopia lentis syndrome, Loeys–Dietz syndrome and mitral valve prolapse syndrome. However, in pa-tients under 20 years of age these new syndromes are not eligible, be-cause the phenotype may evolve with time. For these patients theterm ‘potentialMFS’was proposed. These revisions of the Ghent criteriahave ultimately resulted in the revised Ghent nosology of 2010.

2.2.2. Clinical implications of the revised Ghent nosologyRadonic et al. recently reported the practical applicability of apply-

ing the revised Ghent criteria in an established adult Marfan populationof 180patients [28]. In 164 patients (91%) the diagnosis ofMFSwas con-firmed with the revised Ghent nosology. In three patients (2%) the syn-drome was rejected, due to the altered weight given to dural ectasia inthe revised criteria. These patients showed few or no other features ofMFS, besides aortic root dilatation and dural ectasia. In 13 patients(7%) MFS was rejected due to the absence of aortic root dilatation de-fined as a Z-score≥2. In this group, an aortic root diameter of morethan 40 mmwas present in six patients, four ofwhom received an alter-native diagnosis, such as ectopia lentis syndrome andMASS phenotype.The other seven patients with an aortic root diameter of less than40mm had an aortic root shaped typically for MFS; however, justthree of these seven patients received an alternative diagnosis.

Faivre et al. performed a similar kind of study, including 1009 pa-tients with a FBN1mutation and a phenotype compatible with a connec-tive tissue disorder [29]. Each patient had been classified according to theGhent criteria aswell to the revisedGhent criteria. A total of 83% could beclassified as MFS according to the new nosology as compared to 89%

Table 1The Ghent Criteria for diagnosis of Marfan syndrome (1996).

Table 2 The Ghent criteria for diagnosis of Marfan syndrome [1,13]

Criteria for diagnosis of Marfan syndromeFor the index case without family/genetic history:• major criteria in at least 2 different organ systems and involvement of athird organ system

For the index case with family/genetic history:• one major criterion in an organ system and involvement of a secondorgan system

For a relative of an index case:• presence of a major criterion in the family history and• one major criterion in an organ system and involvement of a secondorgan system

Cardiovascular systemMajor criterion (either of the following)• Dilatation of the ascending aorta and involving at least the sinuses of Valsalva• Dissection of the ascending aortaMinor criteria (for involvement only one minor criteria must be present)• Mitral valve prolapse• Dilatation of the main pulmonary arteryb40 years age, in the absence ofan obvious cause

• Calcification of the mitral annulus younger than age 40 years• Dilatation or dissection of the descending thoracic or abdominal aortayounger than age 50 years

Skeletal systemMajor criteria (at least four of the following features) (for involvement:two of the following features)

• Pectus carinatum• Pectus excavatum, needing surgery• Reduced upper-segment to lower-segment ratio or arm span to heightratio >1.05

• Wrist and thumb signs• Scoliosis of >20° or spondylolisthesis• Reduced extension at the elbows (b170°)• Medial displacement of the medial malleolus, causing pes planus• Protrusio acetabulae of any degree (ascertained on radiographs)Minor criteria (for involvement one major criteria and two of the followingfeatures must be present)

• Pectus excavatum of moderate severity• Joint hypermobility• Highly arched palate with crowding of teeth• Facial appearance (dolichocephaly, malar hypoplasia, enophthalmos,retrognathia, down-slanting palpebral fissures)

Ocular systemMajor criterion• Ectopia lentisMinor criteria (for involvement at least two of the following featuresmust be present)• Abnormally flat cornea• Increased axial length of globe• Hypoplastic iris or hypoplastic ciliary muscle, causing decreased miosis

Pulmonary systemMinor criteria (for involvement only one of the following featuresmust be present)

• Spontaneous pneumothorax• Apical blebs

Skin and integumentMinor criterion (for involvement only one of the following features must be present)• Striae atrophicae (stretch marks) without marked weight gain, pregnancy,or repetitive stress)

• Recurrent or incisional herniae

DuraMajor criterion• Lumbosacral dural ectasia

Family/genetic historyMajor criteria (any one of the following)• Having a parent, child, or sibling who meets these diagnostic criteriaindependently

• Presence of a mutation in FBN1, which is known to cause Marfan's syndrome• Presence of a haplotype around FBN1, inherited by descent, known tobe associated with unequivocally diagnosed

Marfan's syndrome in the family

11R. Franken et al. / Progress in Pediatric Cardiology 34 (2012) 9–14

according to the Ghent criteria of 1996. Almost 90% of patients in theadult cohort who received the diagnoses MFS according to the Ghentcriteria of 1996 kept the diagnosis MFS by using the revised criteria; 8%were reassigned to ectopia lentis syndrome and 2% were reassigned toMASS. Both the study of Radonic and the study of Faivre have the sameagreement of approximately 90% among patientswithMarfan syndromeclassified according to the Ghent criteria of 1996 and the revised Ghentcriteria of 2010.

2.3. Critical appraisal

The revised Ghent nosology offers some advantages as compared tothe old nosology. The guidelines seem to be easier to apply by physi-cians, because more weight is given to ectopia lentis and aortic rootaneurysm. Furthermore, several minor criteria which were often sub-jective and difficult to assess have been eliminated. In addition, theuse of the revised criteria narrows the diagnosis of MFS to patientswith fully expressed disease and various new diagnoses are suggestedto decrease the risk of premature diagnosis. By introducing the revisedGhent criteria more specificity had been reached while maintainingclinical relevance. However, the revised Ghent criteria of 2010 alsoraise some issues as compared to the guidelines of 1996 which will bedescribed in the following section.

2.3.1. Z-scores and body surface areaComments can be made regarding one of the major changes in the

new nosology, which includes the use of Z-score measurement for aorticroot dilatation in adults. Normal values for aortic root dilatation dependon age, height and weight (body surface area (BSA)) [30]. The Z-score isa common statistical way of standardizing data on one scale in whicheach Z-score corresponds to a point in a normal distribution (appendix).An aortic root diameter with Z-score≥2 assumes an aortic root diameterwith a value two standard deviations beyond its expected value correctedfor age and BSA. However, instead of a linear distribution between BSAand aortic root diameter, the curve appears to flatten with increasingBSA in subjects above the 95th percentile for height in a nonlinear corre-lation with maximum aortic root diameters of 40 mm [28,31,32]. A largestudy of aortic root dimensions in 2317 highly trained athletes demon-strated an upper normal limit (>99 percentile value) of the aortic root di-ameter of 40 mm in men (n=17) and 34mm in women (n=10) [33].These data suggest that aortic root diameters of more than 40 mm aredilated.

A Marfan population differs significantly from the healthy populationregarding BSA and aortic root size variability [34]. In addition, in MFS pa-tients aortic root dilatation is generally located at the level of the three si-nuses of Valsalva andmay terminate abruptly at the sinotubular junction,resulting in a pear-shaped aortic root (Fig. 2) [35]. Oosterhof et al. havedemonstrated that in 17 patients with a juvenile form of ascending aorticdilatationwithoutMFS the aneurysmwasmostly fusiform,whereas in 78patients with aortic root dilatation in MFS all patients had pear-shapeddilatation [36].

Currently used Z-scores may underestimate aortic enlargement inpatients with a large BSA and may not reflect the potentially progres-sive nature of it. Therefore, we propose to define aortic root dilatation(corrected for BSA and age) as a Z-score>2 and/or an absolute aorticroot diameter of more than 40 mm. However, an appropriate defini-tion of aortic root involvement in adult Marfan patients should alsotake into account the typical pear-shaped aortic root [35,37–39].

2.3.2. FBN1 mutationsTo determine whether a FBN1 mutation is associated with aortic

root dilatation, the clinician needs to have access to a public up-to-date database of known mutations. Unfortunately, such an updateddatabase is not presently available.

In the revisedGhent criteriamore emphasis is placed on FBN1muta-tions. However, the criteria for the pathogenicity of these mutations, as

Table 2The Revised Ghent Criteria for diagnosis of Marfan syndrome (2010).

Table 1 Revised Ghent criteria for diagnosis of Marfan syndrome and relatedconditions [14,29]

Criteria for diagnosis of Marfan syndromeIn the absence of family history of Marfan syndrome:1. Ao (Z≥2) AND EL=MFS*2. Ao (Z≥2) AND FBN1=MFS3. Ao (Z≥2) AND Syst (≥ 7 points)=MFS*4. EL AND FBN1 associated with Ao=MFSEL with or without Syst AND with an FBN1 not previously been associatedwith Ao or no FBN1=ELS

Ao (Zb2) AND Syst (≥ 5 points) with at least one skeletal feature, withoutEL=MASS

MVP AND Ao (Zb2) AND Syst (b 5) without EL=MVPSIn the presence of family history of Marfan syndrome:5. EL AND FH=MFS6. Syst (≥ 7 points) AND FH=MFS*7. Ao (Z≥2 above age of 20 years, ≥ 3 below age of 20 years) AND FH=MFS*

Scoring of systemic features• Wrist AND thumb sign — 3 (wrist OR thumb sign – 1)• Pectus carinatum deformity — 2 (pectus excavatum or chest asymmetry — 1)• Hindfoot deformity — 2 (plain pes planus — 1)• Pneumothorax — 2• Dural ectasia — 2• Protrusio acetabuli – 2• Reduced US/LS AND increased arm/height AND no severe scoliosis – 1• Scoliosis or thoracolumbar kyphosis — 1• Reduced elbow extension — 1• Facial features (3/5) — 1 (dolichocephaly, enophthalmos, downslantingpalpebral fissures, malar hypoplasia, retrognathia)

• Skin striae — 1• Myopia>3 diopters — 1• Mitral valve prolapse (all types) – 1Maximum total: 20 points; score≥7 indicates systemic involvement

Caveat: without discriminating features of Shprintzen-Goldberg syndrome, Loeys-Dietzsyndrome or vascular EDS syndrome. If present then TGFBR1/2 testing, collagen biochem-istry, COL3A1 testing. Other conditions/genes will follow with time.Abbreviations: Ao, aortic diameter at the sinuses of Valsalva above indicated Z-score oraortic root dissection; Z, Z-score; EL, ectopia lentis; MFS, Marfan syndrome; FBN1,fibrilline-1 mutation; Syst, systemic score; ELS, ectopia lentis syndrome; MASS, myopia,mitral valve prolapse, borderline (Zb2) aortic root dilatation, striae, skeletal findings phe-notype; MVP, mitral valve prolapse; MVPS, mitral valve prolapse syndrome; FH, familyhistory; US/LS, upper segment/lower segment ratio.

12 R. Franken et al. / Progress in Pediatric Cardiology 34 (2012) 9–14

formulated by Loeys et al. [14] are incomplete and inaccurate. For exam-ple, the epidermal growth factor-like (EGF) consensus sequence as shownin the revised Ghent criteria paper is represented as (D/N)X(D/N)(E/Q)Xm(D/N)Xn(Y/F), implying that aspartic acid (D) as the first residue ofan EGF-like domain can be replaced by asparagin (N) without effect.However, the negative charge of aspartic acid at this position is essentialfor calcium binding and this type of mutation will undoubtedly lead to

Fig. 2. Echocardiography of dilated sinus of Valsalva.

protein instability. Consequently, any missense mutation involving thefirst aspartic acid of any calcium binding EGF-like domain in FBN1 shouldbe considered pathogenic. A more complete and comprehensive discus-sion of this matter is beyond the scope of the current paper and will bepresented elsewhere.

2.3.3. Differential diagnosis of MFSThe differential diagnosis of MFS includes mitral valve prolapse syn-

drome, ectopia lentis syndrome andMASS phenotype.MASS phenotypeis a syndrome including at least two, but preferably three of the follow-ingmanifestations:myopia,mitral valve prolapse, borderline aortic rootenlargement (Z-scoreb2), skin and minor skeletal features [13]. Theterm ‘borderline aortic root enlargement’ assumes non-progressive aor-tic root dilatation, but it is currently unknown towhat proportion of pa-tients this applies. In patients with mitral valve prolapse syndrome,clinical features include mitral valve prolapse, pectus excavatum, scoli-osis andmild arachnodactyly. However, aortic enlargement and ectopialentis preclude the diagnosis of mitral valve prolapse syndrome. In pa-tients with ectopia lentis syndrome, ectopia lentis is often accompaniedby some skeletal features of MFS and a FBN1mutation. A key character-istic of ectopia lentis syndrome is the absence of aortic disease.

In the revised Ghent criteria, annual cardiovascular monitoring isadvised in individuals whomeet the criteria forMFS, related connectivetissue disorders and individuals with features of connective tissue dis-orders under age 20. The reason for this strict monitoring is the gradualdilatation of the aortic root in MFS which can appear at all ages, there-fore the named syndromes may still evolve to MFS [38,40,41].

However, diagnostic criteria should not be flexible, but reliable andclear, so physicians and patients know what to expect and know howto manage the disease properly. Dynamic diagnosis implies follow-upin patients with a diagnosis other than MFS even without cardiovascu-lar disease, which creates a great burden on healthcare resources. In ad-dition, patients with either ectopia lentis syndrome orMASS phenotypeshould be counselled about the risk of a more severe presentation intheir offspring since aortic enlargement may evolve over time. Thismight represent a psychological and unnecessary burden for patients.Finally, genetic counselling should be uniform in members of thesame family to ensure that patients with the same mutation obtainthe same diagnosis. In order to determine the proportion of patientswho will meet the criteria for MFS later in life, a follow-up study of pa-tients with such MFS-related diagnoses is required.

2.3.4. Optional additions to the revised criteriaFaivre et al. demonstrated that several patients have not been di-

agnosed with MFS by applying the new criteria, but should be careful-ly monitored for the development of aortic root dilatation [29]. Forexample, patients with a systemic score ofb7 without aortic dilata-tion or ectopia lentis, but with a positive family history of MFS anda known pathogenic FBN1 mutation associated with aortic dilatation.These patients were classified as MFS in the study by Faivre et al., sothat members from one family, who are carrying the same FBN1 mu-tation, will have the same diagnosis.

For patients with a systemic score≥7 and a known FBN1mutationwithout family history no specific diagnosis is defined. Faivre et al.have proposed an adaptation to the revised Ghent criteria so thesepatients can be diagnosed as having MFS.

3. Progress in treatment of Marfan syndrome

Currently, several drugs are prescribed in order to slow down aorticdilatation in patients with MFS and their aim is to delay or prevent sur-gical interventions. The most frequently prescribed medicines in MFSare β-blockers. The pharmacologic rationale of the use of β-blockers isbased on reduction of stress on the proximal aorta by decreasinginotropy and chronotropy aswell as lowering of the average blood pres-sure [42,43]. Although the only randomized clinical trial assessing the

Table 3Ongoing Losartan trials.

Institution (study) Start date Follow up Design Age range(years)

Target no.cases

Clinical endpoints(months)

Boston Children's Hospital, Pediatric HeartNetwork. - Lacro et al.

January 2007 36 DB, RCT; losartan vs atenolol 0.5–25 604 Change in AoR

Brigham andWomen's Hospital - Creager et al. October 2007 6 DB, RCT; losartan vs atenolol ≥ 50 50 Aortic biophysical propertiesHeart and Stroke Foundation of Canada January 2008 12 DB, RCT; losartan vs atenolol 12–25 30 Aortic biophysical propertiesSandor et al.

Academic Medical Center (COMPARE) February 2008 36 OB , RCT; losartan vs no losartan ≥ 18 330 Change in AoRMulder et al.

Policlinico St. Matteo Hospital - Gambarin et al. July 2008 48 OB, RCT losartan vs nebivolol or combined 1–55 291 Change in AoRHospital Bichat Paris (Marfan Sartan) September 2008 36 DB, RCT losartan vs placebo ≥ 10 300 Change in AoRDetaint et al.

Ghent Hospital (Ghent Marfan Trial) June 2009 36 DB, RCT losartan vs placebo ≥ 10 174 Change in AoR, CA diameterMoberg et al.

Royal Brompton Harefield Foundation(The AIMS Trial) - Mullen et al.

September 2010 45 DB, RCT irbesartan vs placebo 6–40 490 Change in AoR

Hospital Universitario Vall d'Hebron October 2010 36 DB, RCT losartan vs atenolol 5–60 150 Change in AoR, CA diameterForteza et al.

AoR Aortic root; CA carotid artery; DB double-blind; echo: echocardiography; MRI magnetic resonance imaging; OB open label, blinded endpoints; RCT, randomized controlled trial.

13R. Franken et al. / Progress in Pediatric Cardiology 34 (2012) 9–14

effect of β-blockers has demonstrated a reduction in the rate of aorticroot dilatation [16], β-blockers did not show any beneficial effect onclinical cardiovascular endpoints such as aortic dissection or aortic rup-ture. The beneficial effect of β-blockers is thought to derive from reduc-ing the hemodynamic stress on the aorta; however they do not preventthe development of cystic medial degeneration, the pathological sub-strate for aortic disease in MFS. Cystic medial degeneration in MFSis mostly characterized by vascular smooth muscle cell apoptosis,elastolysis aswell as accumulation of collagen and proteoglycans [44,45].

Several cascades have shown to contribute to the development ofMFS, of which pathways leading to TGF-β signal transduction appearto bemost important. A deficiency of fibrillin-1, due to FBN1mutations,may lead to increased TGF-β signal transduction, which in turn leads tocystic medial degeneration [20,46,47]. Another major pathway leadingto an increased TGF-β signal transduction is the angiotensin II cascade.Angiotensin II has a primary role in the renin–angiotensin system aswell as a contributive role to the development of cystic medial degener-ation in the aortic wall [48–50]. The effect of angiotensin II is mediatedby two receptors. Angiotensin II receptor type 1 (AT1R) increases TGF-βsignal transduction, while in most cell types and tissues angiotensin IIreceptor type 2 (AT2R) signaling opposes AT1Rmediated enhancementof TGF-β signal transduction [51,52]. However, AT2R signaling has alsoshown to induce vascular smooth muscle cell apoptosis and therebymay contribute to cystic medial degeneration [52].

Improved understanding of the histological abnormalities causingMFS has identified several potential pharmacotherapies directlytargeting cystic medial degeneration. Most promising of all these phar-macological treatment strategies is losartan, an AT1R blocker which hasalready been prescribed in other cardiovascular diseases. Losartan, hasdrawn interest for MFS treatment, after Habashi et al. demonstratedno difference in absolute aortic root diameter between losartan treatedMFS mice compared to wild-type mice [20]. In addition, in the losartantreated MFS mice TGF-β signal transduction was diminished and cysticmedial necrosis development as well as elastic fibre fragmentation wasprevented. However, a major limitation of the study by Habashi et al.was the use of a mouse model with one specific FBN1 mutation, whilein humansmore than 1700 different FBN1mutations have been reported[53]. Following this study, Brooke et al. noticed in retrospect a signifi-cant decrease in aortic root growth rate and absolute sinotubularjunction diameter in 18 paediatric MFS patients using AT1R blockers[54]. These compelling findings have resulted in the initiation of multi-ple multicentre randomized controlled trials assessing the effects oflosartan on aortic root growth in humanswithMFS (Table 3). If losartandemonstrates beneficial effect on the aortic root in these trials, itmay bethefirst potential pharmacotherapy to be capable of preventing the aor-tic disease associated with MFS.

4. Conclusion

The revised Ghent nosology of 2010 is an improvement in the di-agnostic assessment of MFS as compared to the Ghent criteria of1996. However, the revised Ghent criteria provoke some critical re-marks. First, the use of Z-score for aortic root dilatation is not validat-ed for patients with a large BSA, which is often the case in patientswith MFS. Besides Z-scores, an appropriate definition of aortic root in-volvement in adult Marfan patients should take into account an uppernormal limit of the aortic root of 40 mm as well as the typical pear-shaped aortic root. Second, proper use of the criteria requires a publicup-to-date database with known mutations correlated to aortic rootdilatation and a better understanding of the pathogenicity of muta-tions in FBN1. As mentioned above, this will be discussed in a separatepaper. Third, in the revised Ghent criteria a gap for two subgroups ofpatients with FBN1 mutations is described, who should be diagnosedas having MFS. For example, patients with a systemic score ofb7,without aortic dilatation or ectopia lentis but with a positive familyhistory of MFS and a pathogenic FBN1 mutation. In addition, patientswith a systemic score≥7 and a known pathogenic FBN1 mutationwithout a family history should also be diagnosed as having MFS.Fourth, genetic counselling should be uniform in members of thesame family to ensure that patients with the same mutation aregiven the same diagnosis.

In all MFS related disorders (MASS phenotype, mitral valve pro-lapse syndrome and ectopia lentis syndrome) annual cardiovascularmonitoring is advised, because they can evolve into MFS. However,according to their definition, these MFS related disorders do not in-clude patients with aortic root dilatation. This annual monitoring isa great burden on healthcare resources, patients and their offspring.

Currently, β-blockers are themost frequent prescribedmedicines toslow down aortic dilatation in patients with MFS. The use of β-blockersis based on the reduction of repetitive stress on the proximal aorta, bydecreasing heart rate as well as lowering of the average blood pressure.However, the knowledge of the pathophysiology of Marfan syndromehas advanced, leading to new treatment strategies. Losartan, an AT1Rblocker, has been shown to inhibit TGF-β signal transduction therebypreventing aortic root dilatation in a mouse model of MFS. These find-ings have led to multiple multicentre randomized clinical trialsassessing the effects of losartan on aortic root growth in MFS patients.

In conclusion, the revised Ghent criteria offer some advantages ascompared to the old nosology, since the guidelines are easier to applyand patients with only fully expressed disease are diagnosed withMFS. However, the revised Ghent criteria raise issues which might re-quire a re-evaluation of the criteria for diagnosing MFS at shortnotice.

14 R. Franken et al. / Progress in Pediatric Cardiology 34 (2012) 9–14

Appendix A

Z-score and mean dependent on age and body surface area:The Haycock formula:

BSA m2� �

¼ 0:024265�Height cmð Þ0:3964 �Weight kgð Þ0:5378

Z-score for aortic root diameter:

Z ¼ AoD mmð Þ−mean mmð ÞSD mmð Þ

Mean aortic root diameter:

−Age b 18years : mean ¼ 1:02� 0:98� BSA m2� �� �

−Age b 40years : mean ¼ 0:97� 0:74� BSA m2� �� �

−Age > 40years : mean ¼ 1:92� 0:74� BSA m2� �� �

BSA=body surface area (m²), Z=Z-score, AoD=aortic root di-ameter (mm), mean=mean aortic root diameter in general popula-tion dependent on age and body surface area (mm), SD=standarddeviation (mm)

References

[1] Judge DP, Dietz HC. Marfan's syndrome. Lancet 2005;366(9501):1965–76.[2] Groenink M, Lohuis TA, Tijssen JG, et al. Survival and complication free survival in

Marfan's syndrome: implications of current guidelines. Heart 1999;82(4):499–504.

[3] Groenink M, de Roos A, Mulder BJ. Biophysical properties of the normal-sizedaorta in patients with Marfan syndrome: evaluation with MR flowmapping. Radi-ology 2001;219(2):535–40.

[4] Jeremy RW, Huang H, Hwa J, et al. Relation between age, arterial distensibility,and aortic dilatation in the Marfan syndrome. Am J Cardiol 1994;74(4):369–73.

[5] Nollen GJ, Groenink M, van der Wall EE, Mulder BJ. Current insights in diagnosisand management of the cardiovascular complications of Marfan's syndrome. Car-diol Young 2002;12(4):320–7.

[6] Nollen GJ, Mulder BJ. What is new in the Marfan syndrome? Int J Cardiol2004;97(Suppl. 1):103–8.

[7] Oosterhof T, Groenink M, Hulsmans FJ, et al. Quantitative assessment of duralectasia as a marker for Marfan syndrome. Radiology 2001;220(2):514–8.

[8] Zadeh N, Bernstein JA, Niemi AK, et al. Ectopia lentis as the presenting and prima-ry feature in Marfan syndrome. Am J Med Genet A 2011;155A(11):2661–8.

[9] Faivre L, Collod-Beroud G, Loeys BL, et al. Effect of mutation type and location onclinical outcome in 1,013 probands with Marfan syndrome or related phenotypesand FBN1 mutations: an international study. Am J Hum Genet 2007;81(3):454–66.

[10] Groenink M, Rozendaal L, Naeff MS, et al. Marfan syndrome in children and ado-lescents: predictive and prognostic value of aortic root growth for screening foraortic complications. Heart 1998;80(2):163–9.

[11] Milewicz DM, Dietz HC, Miller DC. Treatment of aortic disease in patients withMarfan syndrome. Circulation 2005;111(11):e150–7.

[12] Murdoch JL, Walker BA, Halpern BL, Kuzma JW, McKusick VA. Life expectancy andcauses of death in the Marfan syndrome. N Engl J Med 1972;286(15):804–8.

[13] De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnosticcriteria for the Marfan syndrome. Am J Med Genet 1996;62(4):417–26.

[14] Loeys BL, Dietz HC, Braverman AC, et al. The revised Ghent nosology for the Mar-fan syndrome. J Med Genet 2010;47(7):476–85.

[15] Silverman DI, Burton KJ, Gray J, et al. Life expectancy in the Marfan syndrome. AmJ Cardiol 1995;75(2):157–60.

[16] Shores J, Berger KR, Murphy EA, Pyeritz RE. Progression of aortic dilatation andthe benefit of long-term beta-adrenergic blockade in Marfan's syndrome. N EnglJ Med 1994;330(19):1335–41.

[17] Dietz HC, Cutting GR, Pyeritz RE, et al. Marfan syndrome caused by a recurrent denovo missense mutation in the fibrillin gene. Nature 1991;352(6333):337–9.

[18] Ramirez F, Rifkin DB. Cell signaling events: a view from the matrix. Matrix Biol2003;22(2):101–7.

[19] ten Dijke P, Hill CS. New insights into TGF-beta-Smad signalling. Trends BiochemSci 2004;29(5):265–73.

[20] Habashi JP, Judge DP, Holm TM, et al. Losartan, an AT1 antagonist, prevents aorticaneurysm in a mouse model of Marfan syndrome. Science 2006;312(5770):117–21.

[21] Marfan AB. Un cas de déformation congénitale des quarte membres, plusprononcée aux extrémités caracterisée par l'allongement des os avec un certaindegré d'amincissement. Bull Mem Soc Med Hop Paris 1896;13:220–6.

[22] McKusick VA. The cardiovascular aspects of Marfan's syndrome: a heritable disor-der of connective tissue. Circulation 1955;11(3):321–42.

[23] Beighton P, De Paepe A, Danks D, et al. International nosology of heritable disor-ders of connective tissue, Berlin, 1986. Am J Med Genet 1988;29(3):581–94.

[24] Dietz HC, Pyeritz RE, Puffenberger EG, et al. Marfan phenotype variability in afamily segregating a missense mutation in the epidermal growth factor-likemotif of the fibrillin gene. J Clin Invest 1992;89(5):1674–80.

[25] Stheneur C, Collod-Beroud G, Faivre L, et al. Identification of the minimal combi-nation of clinical features in probands for efficient mutation detection in theFBN1 gene. Eur J Hum Genet 2009;17(9):1121–8.

[26] Faivre L,Masurel-Paulet A, Collod-BeroudG, et al. Clinical andmolecular study of 320childrenwithMarfan syndrome and related type Ifibrillinopathies in a series of 1009probands with pathogenic FBN1 mutations. Pediatrics 2009;123(1):391–8.

[27] Aalberts JJ, Schuurman AG, Pals G, et al. Recurrent and founder mutations in theNetherlands: extensive clinical variability inMarfan syndrome patients with a singlenovel recurrent fibrillin-1 missense mutation. Neth Heart J 2010;18(2):85–9.

[28] Radonic T, de Witte P. Groenink M et al. Clin Genet: Critical appraisal of the re-vised Ghent criteria for diagnosis of Marfan syndrome; 2011.

[29] Faivre L, Collod-Beroud G, Ades L. The new Ghent criteria for Marfan syndrome:what do they change? Clin Genet 2011.

[30] Roman MJ, Devereux RB, Kramer-Fox R, O'Loughlin J. Two-dimensional echocar-diographic aortic root dimensions in normal children and adults. Am J Cardiol1989;64(8):507–12.

[31] Kinoshita N, Mimura J, Obayashi C, et al. Aortic root dilatation among young com-petitive athletes: echocardiographic screening of 1929 athletes between 15 and34 years of age. Am Heart J 2000;139(4):723–8.

[32] Reed CM, Richey PA, Pulliam DA, Somes GW, Alpert BS. Aortic dimensions in tallmen and women. Am J Cardiol 1993;71(7):608–10.

[33] Pelliccia A, Di Paolo FM, De Blasiis E. Prevalence and clinical significance of aorticroot dilation in highly trained competitive athletes. Circulation 2010;122(7):698–706 3.

[34] Rozendaal L, Groenink M, Naeff MS, et al. Marfan syndrome in children and ado-lescents: an adjusted nomogram for screening aortic root dilatation. Heart1998;79(1):69–72.

[35] Van Engelen K, Mulder BJM. Connective tissue disorders and smooth muscle dis-orders in cardiology. In: Baars HF, Van Der Smagt JJ, Doevendans PAFM, editors.Clinical cardiogenetics. 1 ed. London: Springer; 2011. p. 267–8.

[36] Oosterhof T, Nollen GJ, van der Wall EE, et al. Comparison of aortic stiffness in pa-tients with juvenile forms of ascending aortic dilatation with versus without Mar-fan's syndrome. Am J Cardiol 2005;95(8):996–8.

[37] Gott VL, Greene PS, Alejo DE, et al. Replacement of the aortic root in patients withMarfan's syndrome. N Engl J Med 1999;340(17):1307–13.

[38] Meijboom LJ, Timmermans J, Zwinderman AH, Engelfriet PM, Mulder BJ. Aorticroot growth in men and women with the Marfan's syndrome. Am J Cardiol2005;96(10):1441–4.

[39] RomanMJ, RosenSE,Kramer-FoxR,DevereuxRB. Prognostic significanceof thepatternof aortic root dilation in the Marfan syndrome. J Am Coll Cardiol 1993;22(5):1470–6.

[40] Nollen GJ, Groenink M, Tijssen JG, van der Wall EE, Mulder BJ. Aortic stiffness anddiameter predict progressive aortic dilatation in patients with Marfan syndrome.Eur Heart J 2004;25(13):1146–52.

[41] van Karnebeek CD, Naeff MS, Mulder BJ, Hennekam RC, Offringa M. Natural histo-ry of cardiovascular manifestations in Marfan syndrome. Arch Dis Child2001;84(2):129–37.

[42] Boucek RJ, Noble NL, Gunja-Smith Z, Butler WT. The Marfan syndrome: a deficien-cy in chemically stable collagen cross-links. N Engl J Med 1981;305(17):988–91.

[43] Boucek RJ, Gunja-Smith Z, Noble NL, Simpson CF. Modulation by propranolol ofthe lysyl cross-links in aortic elastin and collagen of the aneurysm-prone turkey.Biochem Pharmacol 1983;32(2):275–80.

[44] Carlson RG, Lillehei CW, Edwards JE. Cystic medial necrosis of the ascending aortain relation to age and hypertension. Am J Cardiol 1970;25(4):411–5.

[45] Nagashima H, Sakomura Y, Aoka Y, et al. Angiotensin II type 2 receptor mediatesvascular smooth muscle cell apoptosis in cystic medial degeneration associatedwith Marfan's syndrome. Circulation 2001;104(12 Suppl 1):I282–7.

[46] Neptune ER, Frischmeyer PA, Arking DE, et al. Dysregulation of TGF-beta activationcontributes to pathogenesis in Marfan syndrome. Nat Genet 2003;33(3):407–11.

[47] Ng CM, Cheng A, Myers LA, et al. TGF-beta-dependent pathogenesis of mitral valveprolapse in amousemodel ofMarfan syndrome. J Clin Invest 2004;114(11):1586–92.

[48] Dikalova A, Clempus R, Lassegue B, et al. Nox1 overexpression potentiates angio-tensin II-induced hypertension and vascular smooth muscle hypertrophy intransgenic mice. Circulation 2005;112(17):2668–76.

[49] Pueyo ME, Michel JB. Angiotensin II receptors in endothelial cells. Gen Pharmacol1997;29(5):691–6.

[50] Rueckschloss U, Quinn MT, Holtz J, Morawietz H. Dose-dependent regulation ofNAD(P)H oxidase expression by angiotensin II in human endothelial cells: protec-tive effect of angiotensin II type 1 receptor blockade in patients with coronary ar-tery disease. Arterioscler Thromb Vasc Biol 2002;22(11):1845–51.

[51] Habashi JP,Doyle JJ, HolmTM,et al. Angiotensin II type 2 receptor signaling attenuatesaortic aneurysm in mice through ERK antagonism. Science 2011;332(6027):361–5.

[52] Jones ES, Black MJ, Widdop RE. Angiotensin AT2 receptor contributes to cardio-vascular remodelling of aged rats during chronic AT1 receptor blockade. J MolCell Cardiol 2004;37(5):1023–30.

[53] Arslan-Kirchner M, Arbustini E, Boileau C. Clinical utility gene card for: Marfansyndrome type 1 and related phenotypes [FBN1]. Eur J Hum Genet 2010;18(9).

[54] Brooke BS, Habashi JP, Judge DP, et al. Angiotensin II blockade and aortic-root di-lation in Marfan's syndrome. N Engl J Med 2008;358(26):2787–95.