REVIEW

Economic and care considerations of Marfan syndromeCarl Rudolf Blankarta, Ricarda Milsteina, Meike Rybczynskib, Helke Schülerb and Yskert von Kodolitschb

aHamburg Center for Health Economics, Universität Hamburg, Hamburg, Germany; bUniversity Heart Center Hamburg, University HospitalEppendorf, Hamburg, Germany

ABSTRACTIntroduction: Marfan syndrome is a rare multisystem disease of the connective tissue, which affectsmultiple organ systems. advances in healthcare have doubled the life-expectancy of patients over thepast three decades. to date, there is no comprehensive review that consolidates economic considera-tions and care for marfan patients.Areas covered: Present research suggests that there may be a link between treatment pattern, diseaseprogression and economic costs of Marfan syndrome. It indicates that an early detection of the diseaseand preventive interventions achieve a dual aim. From a patient perspective, it may reduce the amountof emergency surgery or intervention, and inpatient stays. In addition, it slows disease progression,lowers lifestyle restrictions, reduces psychological stress, and improves health-related quality of life.Expert commentary: Early detection and preventive measures are likely to achieve a dual aim bysimultaneously containing costs and reducing the number and length of inpatient stays.

ARTICLE HISTORYReceived 27 June 2016Accepted 21 September 2016Published online6 October 2016

KEYWORDSMarfan syndrome; costs;economy; gene; FBN1;Loeys-Dietz syndrome;healthcare; TGFBR2;decision-making; surgery

1. Introduction

1.1. Prevalence & incidence

The Marfan syndrome (MIM 154700) is a rare disease with aprevalence of up to 1–3 per 10,000 individuals [1–4]. It is amultisystem disease of the connective tissue, which affectsmultiple organ systems, such as the heart and blood vessels,eyes, bones, and lungs resulting in complications. In 50–80%of the individuals, the Marfan syndrome is inherited throughan autosomal dominant genetic pattern. However, in about20–50% of all individuals are de novo mutations meaningwhere no family history is present [1].

Advances of healthcare have doubled the life-expectancyof patients from 30 to 60 years over the past three decades [5].This development can be attributed to a better knowledge ofthe disease and the advancement in diagnostic techniques, forexample noninvasive imaging technology and moleculargenetic diagnostics. Furthermore, early diagnosis, prophylactictreatment, and intervention allow prolonging or halting earlydisease progression. Despite all these advancements, a sub-stantial part of Marfan patients is still diagnosed at a very latestage. This is attributable to a lack of awareness of the diseaseand the large variety of clinical manifestations, which compli-cates diagnosis [4,6]. However, a potential delay in diagnosisleads to later and less effective preventive treatment, worseoutcomes, and higher costs.

1.2. Research objective

With this review, we aim to provide a condense overview overthe Marfan syndrome, its care pathways, and its corresponding

economic impact on patients and society. First, we provide anoverview of the pathogenesis, manifestation, and pathways ofcare. Then, we present the two coexisting modes of diagnosis,i.e. diagnostics based on phenotypes versus diagnostics basedon genetic testing that strongly determine life-time costs andquality of life of Marfan patients. Following that, we approx-imate the economic costs of the disease. We discuss theimpact of different diagnosis and treatment approaches andwe conclude by pointing at lessons learned from this review.

2. Marfan syndrome

2.1. Pathogenesis

Marfan syndrome is a genetic disorder that is caused bymutations in the FBN1 gene on chromosome 15 [6,7]. TheFBN1 gene codes for production of a 350 kD glycoprotein inthe extracellular matrix called fibrilin-1 [1,8,9]. This proteinconnects with other fibrilin-1 proteins and other moleculesto form microfibrils that are vital for the strength and flexibilityof connective tissues. Furthermore, microfibrils stabilize latenttransforming growth factor β-binding proteins (LTBPs) in theconnective tissue. LTBPs keep the transforming growth factor-β (TGFβ) inactive. Due to these features, a failure in the inter-action between fibrillin-1 and LTBPs result in an excess releaseof TGFβ, which in return results in overgrowth, less elasticity,and increased instability of the tissues. In some variants of theMarfan syndrome, called Loeys–Dietz syndromes, phenotypesresult directly from mutations in the TGF β-receptor 1 (TGFBR1)on chromosome 9 and from mutations in the TGF β-receptor 2(TGFBR2) on chromosome 3 [10,11].

CONTACT Yskert von Kodolitsch kodolitsch@uke.de German Aortic Center Hamburg at the University Heart Center, University Hospital Hamburg -Eppendorf, Hamburg, Martinistrasse 52, 20246 Hamburg, Germany

EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH, 2016VOL. 16, NO. 5, 591–598http://dx.doi.org/10.1080/14737167.2016.1240619

© 2016 Informa UK Limited, trading as Taylor & Francis Group

It is important to note that mutations in the FBN1 gene arenot unique to the Marfan syndrome [12]. FBN1 mutations cancause disorders, which are similar to Marfan, such as the MASSphenotype (MIM 604308) [13] or isolated ectopia lentis (MIM129600) [14,15]. Vice versa, not all patients that fulfill the criteriafor a positive diagnosis with Marfan demonstrate fibrillin-1mutations [16,17]. According to a recent systematic review, theshare of patients with causative FBN1 mutation failing to fulfilldiagnostic criteria of Marfan syndrome was 7% according to theformer Ghent-1 nosology [18] and 15% according to the Ghent-2 nosology [4]. Similarly, the same review identified that 15%and 13% patients with final diagnosis of Marfan syndromeaccording to Ghent-1 and Ghent-2, respectively, did not have acausative FBN1mutation [4]. Moreover, mutations in TGFBR1 andTGFBR2 have been described in patients with Loeys–Dietz syn-drome, and non-syndromic familial thoracic ascending aorticaneurysm (MIM 608967)[1,11].

2.2. Manifestations, disease stages, and treatment

Manifestations of the Marfan syndrome in the skeletal, ocular,and cardiovascular system are among the most prominent andwell-known ones [1,2]. The skeletal manifestation may be themost striking and visible one. Classically, Marfan patients areconsidered to be tall and slender, but some may be small orobese [19]. Ocular characteristics are a long and narrow face,crowded teeth, and an abnormal curvature of the spine, knownas scoliosis or kyphosis. Roughly, 60% of all Marfan patientspresent scoliosis [20]. Accordingly, back pain is three timesmore frequent among Marfan patients than in the healthy popu-lation. At the same time, Marfan patients are found to incur lowerbonemineral density especially in the hip and spine, although noincreased rate in fracture could be found if compared to thegeneral population [21,22]. In line with that, some individualsshow a protruding or sunken chest (pectus carinatum or excava-tum) [23]. Furthermore, the dura, a membrane surrounding thebrain and spinal cord, can be abnormally enlarged (dural ectasia)causing pain in the back, abdomen, legs, or head [24–27].

Manifestations in the ocular system are both, highly pre-valent and clinically relevant. In about 60%, patients have alsobeen found to exhibit ectopia lentis, a dislocation of the ocularlens [28,29]. Other ocular defects, such as retinal detachment,early cataracts and glaucoma also rank among potentialophthalmic complications of Marfan syndrome [5,30,31].

Themost relevantmanifestation of theMarfan syndrome is thecardiovascular system including the aorta, the aortic and themitral valve, and the myocardium [1], where aneurysm of theaortic root, regurgitation of the aortic and the mitral valve, andmyocardial dysfunction give rise to aortic dissection, aortic rup-ture, aortic andmitral valve regurgitation, heart failure, ventriculararrhythmia, and sudden cardiac death [32]. Aortic root dilatationand dissection are the most common cause of morbidity andmortality amongMarfan patients [33]. In up to 90% of all patients,mitral valve prolapse is diagnosed [34–40]. In line with that, about85% of patients exhibit aneurysmal disease of the aorta [34,36,41].

Besides direct clinical manifestations, the Marfan syndrome isalso associated with chronic fatigue and pain, psychological

despair and lifestyle changes. This negatively impacts the qualityof life and restricts the autonomy of Marfan patients [4,42].

Patients with the Marfan syndrome require life-long medicalmonitoring, support, and treatment. The type of treatmentdepends on the stage of disease progression. In an early stage,pharmaceutical treatment prevails (Figure 1). Beta-adrenergicblocker (BAB) therapy retards aortic dilatation and is crucial todelay elective surgery especially in young patients with Marfansyndrome. It aims to lower aortic sheer pressure and reducechanges in pressure over time. Angiotensin-conversing enzymeinhibitors (ACEi), angiotensin receptor blockers (ARBs), and cal-cium channel blockers may be reasonable alternatives of med-ication when beta-adrenergic blocking therapy is not possible[1,43,44]. ARB targets TGBβ, and their use has been found tocure mitral valve disease in Marfan mouse models [45,46],although their use was not found superior to BAB in a rando-mized controlled study comparing BAB versus ARB [47].

Besides medical therapy, patients require constant monitor-ing of the progress of their disease. Cardiac manifestationsworsen with age but progression remains still difficult to pre-dict. Monitoring includes frequent specialist visits and annualor bi-annual echocardiograms. In line with that, frequent ske-letal and ocular inspections are necessary to detect potentialdeteriorations. Furthermore, patients are advised to abstainfrom intense and stressful sports to avoid rising blood pres-sure and excessive cardiac stress [48].

When aortic root dilatation progresses, a prophylactic elec-tive surgery of the aorta is needed. Surgical repair is recom-mended once the aorta reaches a diameter of 50 mm [49]. Inadults, surgery is usually performed if growth exceeds a rate of5 mm per year or if a significant aortic regurgitation emerges

Figure 1. Care pathway for Marfan syndrome.

592 C. R. BLANKART ET AL.

[50]. However, there is less evidence on the optimal timing inchildren [51].

Elective surgery is either performed as conduit operationwith a replacement of the aortic root including the aortic valveby a mechanical prosthetic valve, or as a valve-sparring aorticroot reconstruction operation in which the native aortic valveis preserved [52–54]. In contrast to a conduit operation thatincludes a mechanical valve, the aortic root reconstructiontechniques avoid lifetime anticoagulation with warfarin includ-ing its risk of thromboembolism and bleeding [55–57]. Aorticvalve sparing operations can either be carried out as remodel-ing technique according to Yacoub [58] or as a reimplantationprocedure, according to David [52], where long-term resultsshow superior aortic valve durability and longer freedom fromthe need for reoperation for the David operation [59,60].

If patients fail to undergo a prophylactic surgery, for exam-ple because they have missed their regular check-up, orbecause the Marfan syndrome has not been diagnosed, thepatient’s risk emergency surgery which comes with a signifi-cantly higher mortality than a prophylactic surgery [61–63]. Inaddition, cardiac deterioration occurs at a much faster speedcompared to patients who underwent prophylactic surgery.

In addition to aortic surgery, surgeries on the mitral heartvalve, the skeletal system including the sternum and thevertebral column, the ocular lenses, or for pneumothoraxmay become necessary with disease progression and, insome individuals, transplantation of the heart or kidneys maybecome necessary [20,64–66].

2.3. Diagnosis

The diagnosis of Marfan syndrome is based on various itemsas defined in the 2010 revised Ghent nosology (Ghent-2;Table 1) which replaces its predecessor from 1996 (Ghent-1)and the Berlin nosology from 1988 [4,67]. Traditionally, thediagnosis of Marfan syndrome required the evaluation of up to30 clinical features including ocular examinations, echocardio-graphy assessment of family history, an X-ray of the lung, MRIor XCT of the dura and fibrillin-1 mutation testing. All nosol-ogies are understood as an international consensus on rules ofhow Marfan syndrome should be diagnosed. Such nosologyincluding its broad consensus is essential, because there is nosingle objective way or external ‘gold standard’ to definitivelyprove or exclude Marfan syndrome [4]. Accordingly, severalcritiques prevail [2] and these are unlikely to get fully recon-ciled in the near future [68].

The authors of the Ghent-2 nosology listed the importantchanges of Ghent-2 as compared to the revised Ghent-1nosology. These chances included giving more diagnosticweight to aortic root aneurysm or dissection and ectopialentis. Moreover, they attributed a more prominent role tomolecular genetic testing. In addition, they removed numer-ous clinical criteria including dilatation of the main pulmonaryartery, dilatation or dissection of the descending thoracic orabdominal aorta, increased axial length of globe and abnor-mally flat cornea, hypoplastic iris or hypoplastic ciliary musclecausing decreased miosis, joint hypermobility, spondylolisth-esis, highly arched palate, and recurrent or incisional hernia,calcification of the mitral annulus, and apical blebs of the lung.

Moreover, they mitigated the diagnostic relevance of duralectasia, and they added or modifying clinical criteria such asmyopia >−3 diopters, hindfoot valgus, and thoracolumbarkyphosis. Finally, the Ghent-2 nosology listed new alternativediagnoses of Marfan syndrome such as Loeys–Dietz syndrome[71]. In summary, the Ghent-2 nosology made it easier todiagnose Marfan syndrome, where a family history, an echo-cardiogram of the aortic root, the demonstration of ectopialentis, and assessment of ectopia lentis usually suffice to diag-nose Marfan syndrome. In addition, a systematic review of theliterature illustrated a similar diagnostic performance ofGhent-1 and Ghent-2 nosologies [4].

3. Economic considerations on Marfan syndrome

Studies reporting economic data on Marfan disease are veryscarce. Within our review, we could identify in total threestudies that provide insights in the cost of Marfan disease(see Table 1) [43,69,70]. All of the studies originate fromGermany and analyze very different aspects in the treatmentof Marfan. The study by Manow et al. [69] provides insights inthe internal cost structures of an ambulatory care center inGermany. The study by Achelrod et al. [70] uses a geneticmatching approach to determine direct (non)medical andindirect costs attributable to Marfan syndrome from health-care payer and societal perspective. The most recent study byvon Kodolitsch et al. [43] evaluates the marginal costs ofidentifying an additional – otherwise undetected – Marfanpatient in a population of persons referred for confirmationor exclusion of Marfan syndrome.

3.1. Cost structures of an ambulatory care center

Manow et al. [69] provide insights in the cost of treatment in aGerman ambulatory care center. Based on internal servicecharges, the authors show that the reimbursement of sicknessfunds does not match their actual costs. This difference sus-tained even after a considerable increase in reimbursementrates in 2008. The authors calculated that the total facilitycosts for treatment of 184 Marfan patient in a hospital’sambulatory care center accrues to EUR 71 607 per year,while the provider received EUR 55 550 as reimbursement.Even after introduction of the new more generous reimburse-ment scheme, only 84% of the actual costs, i.e. EUR 59 830,were covered.

Besides presenting costs from ambulatory care center per-spective, Manow et al. [69] indicate – but not specify – highcosts arising through informal caregiving, loss of productivity,and traveling. Across all 184 analyzed patients, 36% traveledfor more than 50 km to access treatment at the specializedambulatory care center [69].

3.2. Economic impact for patients and society andresource utilization

Achelrod et al. [70] show that total costs from sickness fundperspective (cost elements shown in Table 1) amount to EUR 4416 compared to EUR 1 920 per patient and year in the controlgroup of non-Marfan patients. Largest cost block are direct

EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 593

medical costs amounting to EUR 4 024. Main elements ofdirect medical costs are inpatient treatment costs (EUR 1337) and care provided by nonphysicians (EUR 1 315), e.g.physiotherapy. The difference between Marfan and controlgroup is defined as excess costs caused by Marfan syndromeof in total EUR 2,496. Main difference of costs from societalperspective and from sickness fund perspective are the addi-tional consideration of indirect costs, i.e. cost of absence fromwork and the cost of premature death, and the considerationof informal family care. Total costs for a Marfan patient fromsocietal perspective are estimated to accrue to EUR 21,865 perpatient and year, while the yearly costs in the control group isEUR 6,137. Indirect costs are estimated to be EUR 10,329, whileinformal family care, an element of direct nonmedical costs, iscalculated to be EUR 7,200 per Marfan patient and year. Theexcess costs caused by Marfan syndrome from societal per-spective amounts to EUR 15,728. If patient level costs areaggregated on national level, the economic impact rangesfrom EUR 24.0 to EUR 61.4 million from the sickness fundperspective, and from EUR 151.3 to EUR 386.9 million fromsocietal perspective for Germany [70].

The authors of this original research paper hypothesize thathigher direct cost of Marfan patients result from more inten-sive resource use across all specialties. For example, Marfanpatients have 236% more medical imaging, 119% more inpa-tient stays, and a 153% longer average length of hospital staycompared to the control group. Furthermore, they have 39%more physician contacts, 33% more prescriptions, and 20%higher prescription costs [70].

Besides reporting average costs from different perspectives,Achelrod et al. [70] show that healthcare resource consump-tion differed by age. For this reason, the authors define the

following age quartiles based on their data: Q1 from0–16 years, Q2 from 17–24 years, Q3 from 25–41 years, andQ4 older than 41 years. From sickness fund perspective, totalcosts per patient and year are high in Q1 (EUR 3,381), Q3 (EUR3,625), and Q4 (EUR 3,530), while they are lower in Q2 (EUR2,415). From societal perspective, total costs peak in Q2 (EUR29,351), while they are medium in Q3 (EUR 17,607) and Q4(EUR 17,873) and lowest in Q1 (EUR 12,509) [70].

3.3. Cost-effectiveness of DNA sequencing

Von Kodolitsch et al. [43] analyze the cost-effectiveness ofhigh-cost FBN1 gene sequencing based on a sample of 300persons referred for confirmation or exclusion of Marfan syn-drome. The nonfinancial data originates from a study ofSheikhzadeh et al. [17] that shows that 24 out of the 171truly affected persons could only be detected using genesequencing. As it is impossible to restrict genetic testing tothe 24 individuals a priori, all of the 300 suspected individualswill have to be genetically tested in order to correctly diag-nose the 24 patients. Considering that the genetic test accruesto EUR 4,603 per suspected individual, the marginal costs ofdetecting one of the 24 otherwise nonidentifiable patientsamounts to EUR 57,538.

A correct and timely diagnosis of Marfan increases thechances that thoracic aortic aneurysms can be treated in anelective inpatient stay which amounts to EUR 18,622 [43]. Adelay in diagnosis bears a high risk of emergency surgery.Emergency interventions are extremely heterogeneous inprice and typically require expensive and lengthy follow-uptreatments [43] which may easily cause total costs of more

Table 1. Overview of studies reporting economic data on Marfan disease.

Manow et al. [69] Achelrod et al. [70] von Kodolitsch et al. [43]

Objective and cost Perspective Provide detailed cost and revenueinformation from providerperspective across one year

Estimate economic impactof Marfan syndromefrom third-party payerand societal perspectiveby age categories

Evaluate value of information ofhigh-cost DNA sequencing fromthird-party payer perspective

Country Germany Germany GermanyTime 2008 2008 2014Cost Categories Direct medical costs based on

hospital internal servicecharges

Direct medical costs based onGerman outpatient feeschedule

Direct medical costs● Outpatient● Inpatient● Pharmaceuticals● Nonphysician care● Devices and appliances● RehabilitationDirect nonmedical costs● Administration● Sick leave

compensation● Other nonmedical

services● Informal family careIndirect costs● Lost production due to

absence, disability, andpremature death

Direct medical costs● DNA sequencing according to the

German● outpatient fee schedule● Pharmaceutical treatment costs● Elective and emergency surgery of

the aorta based on German inpati-ent reimbursement

Sample size 184 persons with Marfandiagnosis

892 individuals withMarfan syndrome and892 control subjects

300 persons referred forconfirmation or exclusion ofMarfan syndrome

594 C. R. BLANKART ET AL.

than EUR 100,000. Therefore, the authors consider the price ofEUR 57,538 as cost-effective.

3.4. Exemplary cost patterns

Although we cannot directly synthesize cost patterns by treat-ment out of the existing literature, we believe that we are ableto derive two exemplary cost curves from the three studies[43,69,70] that are linked to two exemplary treatment patterns(see Figure 2). As the curves are idealized, they represent thebest- and worst-case scenarios. Individual cost curves maysubstantially deviate because Marfan specific treatment isoften highly individualized. The solid line shows the costcurves of a patient that is diagnosed in his early years byexpensive FBN1 gene sequencing, leading to a low cost butadequate pharmacological treatment with mostly out ofpatent drug therapies. Continuous monitoring and regularcheck-ups lead to early detection of a growing aorta andrather inexpensive preventive surgery of the aorta or of themitral valve that usually becomes necessary in the 30s [60].The dashed line shows a patient that is not or – even worse –wrongly diagnosed in the early years. Here, rather low costsaccrue in the early ages but the negligence of adequatetreatment leads to high costs of eventual emergency surgeryof aortic aneurysms in the patients’ 30s–40s with high follow-up costs.

4. Discussion

Present research indicates a link between the treatmentpattern, the disease progression, and the economic costsof the disease. It shows that an early detection of thedisease and preventive interventions achieve a dual aim.From a patient perspective, it slows the disease progression,it lowers lifestyle restrictions, psychological stress, and itimproves the health-related quality of life. In addition,early detection reduces the amount of emergency surgeryor intervention and inpatient stays which improves thepatient’s quality of life. From an economic perspective,

early detection and preventive measures contain costs byreducing the number of inpatient stays as well as the lengthof stay.

However, early diagnosis and prevention requires a highcertainty on the initial diagnosis. As diagnosis is challenging,potential patients should be assessed by multidisciplinaryhealthcare teams [72]. As neither the phenotypic nor thegenetic diagnosis is perfectly unequivocal, it is recommendedto blend both techniques to receive a more comprehensivepicture [2,4,16,73,74]. The assessment by a multidisciplinaryteam requires several visits to specialists and several diagnos-tic procedures, such as an echocardiogram, a CT scan, and anMRI. In addition, genetic testing comes, despite all progress, atvery high costs. However, the higher diagnosis costs may beset off by lower treatment costs due to earlier interventionsand low-cost preventive measures instead of high costly emer-gency surgeries of the aorta [43].

Various deficiencies still impede patients from enjoying aclose-to-normal life [5]. First and foremost, there is an insuffi-cient awareness of the disease implying that the syndrome isoften diagnosed at a late stage or not at all [4,23]. If so,patients risk severe manifestations, such as emergency sur-geries of the aorta, which severely lower their life expectancyand increase the financial burden of the disease. This may notbe achieved by simply increasing medical supply, such as thedensity of physicians as research has shown that the time todiagnosis is immune to the regional healthcare structures [75].This indicates that potential solutions to undertreatment is tobe sought in raising awareness among physicians and patientsalike and in promoting new diagnostic means in the medicalprofession [43]. Following from that, health providers andpatients alike are in need of clear medical guidelines on howto treat the syndrome. For the time being, revisions of theGhent nosology foster the identification of joint standards ondiagnostic prerequisites [4], but there is no harmonized treat-ment path for Marfan patients. Besides a joint understandingthat a diameter of the aorta of 50 mm or more requires asurgical intervention in adult patients [1,76], there are no jointtreatment recommendations. Therefore, standardized treat-ment guidelines appear desirable. However, treatment cannotfollow a ‘one fits for all’ pattern for two crucial reasons: First, inmany patients such ‘uni therapy’ can lead to ‘technical treat-ment failure,’ as we would ignore the patient’s individualphysical, or emotional risks that threaten the success of treat-ment or make it impossible. Second, translation of therapyguideline lead to ‘ethical failure,’ as we would patient’sneeds and individual values.

5. Future research and limitations

Pathogenesis, diagnosis, and treatment patterns of the diseaseare well researched considering that Marfan syndrome is a raredisease. However, evidence on economic issues remainsscarce. Based on the three studies on economic issues ofMarfan syndrome, we believe that there is a link betweenthe treatment pattern, the disease progression, and the eco-nomic costs. For a better understanding of this link and toprove a causal relationship, further research is absolutelyneeded. This research may exploit existing datasets using

Figure 2. Estimated costs patterns depending on diagnosis and treatment pathover lifetime.

EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 595

machine learning techniques to find better ways to clusterdifferent Marfan treatment patterns.

Although there is a joint understanding that preventivesurgery and that an intensified diagnostic procedure may beset off at a later stage [43], cost-effectiveness of those tests ishighly sensible on the right preselection of patients.Unrestricted overuse of gene sequencing must be restrictedto guarantee access to the patients in need. Also, not everycorrectly diagnosed patient is prevented from unplanned sur-gery and not every undiagnosed patient will have to undergoemergency surgery [63]. Here, further research is needed thataccounts for uncertainty of the results. Moreover, the lack ofcost-effectiveness information in general impedes moreexpensive diagnostic and nondiagnostic procedures frombeing reimbursed. Further research in this area is of utmostimportance to demonstrate cost-effectiveness of effectivediagnostic and nondiagnostic procedure to payers.

At last, we call for more cost studies that incorporatedifferent cost perspectives and include further cost items.Although the claims based study from Achelrod et al. [70]includes a broad spectrum of costs and two perspectives,the authors could not account for cost of premature death.In addition, there is some evidence that costs reported inclaims data are underestimating the true efforts taken by theprovider [69].

6. Conclusion

Despite great improvements in the treatment of Marfan in thepast decades, the disease strongly impacts quality of life. Thereview of literature shows that quality of life can be substan-tially improved if the disease is detected early to allow fortimely and preventive action. Genetic testing, advancementsin medical device technology, and excellence in surgical tech-niques are contributing to this goal. Researchers, providers,and policy makers are encouraged to explore these options toimprove quality of life for Marfan patients while containingcosts.

In addition, we point at several persisting limitations in theknowledge of economy and care of Marfan syndrome. First,there is a lack of awareness and a significant share of patientsis diagnosed at a late stage, or not at all [75]. Furthermore,several dimensions of the medical treatment of Marfanpatients remain vague and unclear. This is partly due to thehighly heterogeneous development of the disease in Marfanpatients and partly due to a lack of research. However, medicalinnovations, such as new surgical techniques and new modesof diagnosis, have increased the survival rate of patients sig-nificantly. This should be stressed even further.

7. Expert commentary

The biggest leverage for improving survival, physical fitness, andlife quality for people with Marfan syndrome is to establish adefinitive diagnosis early in life and to provide prophylactic carerather than to fix health catastrophes. It is possible to diagnoserare diseases such as phenylketonuria by simple screening testsright after birth. Unfortunately, there is no such screening test

for Marfan syndrome, and it appears unlikely that such test willbe available in the near future. Therefore, it remains the majorweakness of Marfan healthcare that initial diagnosis still hingeson the personal awareness of individuals to recognize sugges-tive signs and symptoms. Genetic testing is pivotal to confirmthe clinical diagnosis and to dissect Marfan syndrome fromsimilar genetic aortic diseases. Finally, only Marfan centers withcomprehensive personal experience, both with affected childrenand with adults are able to maximize therapeutic success. SuchMarfan centers provide healthcare on a high level of expertise inmany medical disciplines including heart surgery, vascular sur-gery, cardiology, orthopedic surgery, ophthalmic surgery, neu-rology, and psychology. Medical evidence is available todocument that such multidisciplinary healthcare centers max-imize survival, physical fitness, and quality of life for Marfanpatients. In contrast, economic evidence still requires researchto document comprehensively that these centers are also cap-able of minimizing costs of care.

8. Five-year view

In 5 years, there will be three major advances. First, moleculartesting will be in much broader use at significantly lower costs.Second, preventive medicine will be on the rise. We will per-form prophylactic surgery of the aorta and the heart valvesearlier and with still better results than today. We will look ataortic dissection and aortic rupture in Marfan patients as asystematic failure of national healthcare systems. Costs fortreating victims of predictable health catastrophes will onlybe anecdotal. Finally, mechanical heart valves and warfarin willbe history of medicine, because reconstructive valve surgeryor biological heart valves will be state of the art.

Key issues

● Marfan syndrome is a rare multisystem disease of the con-nective tissue, that is caused by mutations in the FBN1 geneand that affects multiple organ systems.

● Medical research provides evidence that a modern treat-ment pattern doubles the life-expectancy and normalizeshealth-related life quality of patients.

● A modern medical treatment pattern comprises adequateclinical screening for Marfan syndrome in the population,early clinical diagnosis at expert centers, molecular geneticsfor diagnostic confirmation, counselling for and managementof life-style, advancedmedication tominimize disease progres-sion, prophylactic aortic and heart valve surgery, and usage ofvalve-sparing rather than valve replacing surgical techniques.

● Economic research suggests that there may be a linkbetween the treatment pattern, the disease progressionand the economic costs of Marfan syndrome. A moderntreatment patter with early diagnosis and preventive mea-sures is likely to contain costs and reduce the number andlength of inpatient stays.

Funding

This paper was not funded

596 C. R. BLANKART ET AL.

Declaration of interest

The authors have no relevant affiliations or financial involvement with anyorganization or entity with a financial interest in or financial conflict withthe subject matter or materials discussed in the manuscript. This includesemployment, consultancies, honoraria, stock ownership or options, experttestimony, grants or patents received or pending, or royalties.

References

1. Judge DP, Dietz HC. Marfan’s syndrome. The Lancet. 2005;366(9501):1965–1976.

2. Dean JCS. Marfan syndrome: clinical diagnosis and management.Eur J Hum Genet. 2007;15(7):724–733.

3. Gray JR, Bridges AB, Faed MJ, et al. Ascertainment and severity ofMarfan syndrome in a Scottish population. J Med Genet. 1994;31(1):51–54.

4. von Kodolitsch Y, De Backer J, Schüler H, et al. Perspectives on therevised Ghent criteria for the diagnosis of Marfan syndrome. ApplClin Genet. 2015;8(June 2015):137—155.

5. Pyeritz RE. Marfan syndrome: 30 years of research equals 30 yearsof additional life expectancy. Heart. 2009;95(3):173–175.

6. Kainulainen K, Pulkkinen L, Savolainen A, et al. Location on chro-mosome 15 of the gene defect causing Marfan syndrome. NewEngland J Med. 1990;323(14):935–939.

7. Dietz H, Pyeritz R. Mutations in the human gene for fibrillin-1(FBN1) in the Marfan syndrome and related disorders. Hum MolGenet. 1995;4(suppl 1):1799–1809.

8. Pereiralrz L., Andrikopoulos K, Tian J, et al. Targetting of the geneencoding fibrillin-1 recapitulates the vascular aspect of Marfansyndrome. Nat Genet. 1997;17(2):218–222.

9. Sakai LY, Keene DR, Engvall E. Fibrillin, a new 350-kD glycoprotein,is a component of extracellular microfibrils. J Cell Biol. 1986;103(6):2499–2509.

10. Singh KK, Rommel K, Mishra A, et al. TGFBR1 and TGFBR2 muta-tions in patients with features of Marfan syndrome and Loeys-Dietzsyndrome. Hum Mutat. 2006;27(8):770–777.

11. Loeys BL, Chen J, Neptune ER, et al. A syndrome of altered cardi-ovascular, craniofacial, neurocognitive and skeletal developmentcaused by mutations in TGFBR1 or TGFBR2. Nat Genet. 2005;37(3):275–281.

12. Faivre L, Collod-Beroud G, Callewaert B, et al. Pathogenic FBN1mutations in 146 adults not meeting clinical diagnostic criteriafor Marfan syndrome: further delineation of type 1 fibrillinopathiesand focus on patients with an isolated major criterion. Am J MedGenet. 2009;149A(5):854–860.

13. Rippe M, De Backer J, Kutsche K, et al. Mitral valve prolapsesyndrome and MASS phenotype: Stability of aortic dilatation butprogression of mitral valve prolapse. IJC Heart & Vasculature.2016;10:39–46.

14. Tsipouras P, Del Mastro R, Sarfarazi M, et al. Genetic linkage of theMarfan syndrome, ectopia lentis, and congenital contractural ara-chnodactyly to the fibrillin genes on chromosomes 15 and 5. NEngl J Med. 1992;326(14):905–909.

15. Robinson PN, Booms P, Katzke S, et al. Mutations of FBN1 andgenotype–phenotype correlations in Marfan syndrome and relatedfibrillinopathies. Hum Mutat. 2002;20(3):153–161.

16. Loeys B, Nuytinck L, Delvaux I, et al. GEnotype and phenotypeanalysis of 171 patients referred for molecular study of the fibril-lin-1 gene fbn1 because of suspected marfan syndrome. ArchIntern Med. 2001;161(20):2447–2454.

17. Sheikhzadeh S, Kade C, Keyser B, et al. Analysis of phenotype andgenotype information for the diagnosis of Marfan syndrome. ClinGenet. 2012;82(3):240–247.

18. De Paepe A, Devereux RB, Dietz HC, et al. Revised diagnosticcriteria for the Marfan syndrome. Am J Med Genet. 1996;62(4):417–426.

19. Yetman AT, McCrindle BW. The prevalence and clinical impact ofobesity in adults with Marfan syndrome. Can J Cardiol. 2010;26(4):e137–e139.

20. Robins P, Moe J, Winter R. Scoliosis in Marfan’s syndrome. Itscharacteristics and results of treatment in thirty-five patients. JBone Jt Surg. 1975;57(3):358–368.

21. Carter N, Duncan E, Wordsworth P. Bone mineral density in adultswith Marfan syndrome. Rheumatology. 2000;39(3):307–309.

22. Le Parc MJ, Plantin P, Jondeau G, et al. Bone Mineral Density inSixty Adult Patients with Marfan Syndrome. Osteoporos Int.1999;10(6):475–479.

23. Sheikhzadeh S, Kusch ML, Rybczynski M, et al. A simple clinicalmodel to estimate the probability of Marfan syndrome. QJM.2012;105(6):527–535.

24. Lundby R, Rand-Hendriksen S, Hald JK, et al. Dural Ectasia in Marfansyndrome: A case control study. Am J Neuroradiology. 2009;30(8):1534–1540.

25. Pyeritz R., Fishman EK, Bernhardt BA, et al. Dural ectasia is acommon feature of the Marfan syndrome. Am J Hum Genet.1988;43(5):726.

26. Stern WE. Dural ectasia and the Marfan syndrome. J Neurosurg.1988;69(2):221–227.

27. Sheikhzadeh S, Sondermann C, Rybczynski M, et al. Comprehensiveanalysis of dural ectasia in 150 patients with a causative FBN1mutation. Clin Genet. 2014;86(3):238–245.

28. Maumenee IH. The eye in the Marfan syndrome. Trans AmOphthalmol Soc. 1981;79:684–733.

29. Lally DR, Monsonego J. Ectopia Lentis in Marfan’s syndrome. N EnglJ Med. 2014;371(19):e28.

30. Nemet AY, Assia EI, Apple DJ, et al. Current concepts of ocularmanifestations in Marfan syndrome. Surv Ophthalmol. 2006;51(6):561–575.

31. Pyeritz RE, McKusick VA. The Marfan syndrome: diagnosis andmanagement. New England J Med. 1979;300(14):772–777.

32. Aydin A, Adsay BA, Sheikhzadeh S, et al. Observational cohort studyof ventricular arrhythmia in adults with Marfan syndrome causedby FBN1 mutations. PLoS One. 2013;8(12):e81281.

33. Aburawi EH, O’Sullivan J. Relation of aortic root dilatation and agein Marfan’s syndrome. Eur Heart J. 2007;28(3):376–379.

34. Brown OR, DeMots H, Kloster FE, et al. Aortic root dilatation andmitral valve prolapse in Marfan’s syndrome: anECHOCARDIOgraphic study. Circulation. 1975;52(4):651–657.

35. Kiotsekoglou A., Sutherland G R,Moggridge J.,et al. Impaired rightventricular systolic function demonstrated by reduced atrioventri-cular plane displacement in adults with Marfan syndrome. EurHeart J Cardiovasc Imaging. 2009;10(2):295–302.

36. Rybczynski M, Bernhardt AMJ, Rehder U, et al. The spectrum ofsyndromes and manifestations in individuals screened for suspectedMarfan syndrome. Am J Med Genet. 2008;146A(24):3157–3166.

37. Bernhardt A.M., Treede H, Detter C, et al. Results of modern mitralvalve repair in patients with Marfan syndrome. Thorac CardiovascSurg. 2014;62(1):35–41.

38. Rybczynski M, Treede H, Sheikhzadeh S, et al. Predictors of out-come of mitral valve prolapse in patients with the marfan syn-drome. Am J Cardiol. 2011;107(2):268–274.

39. Kuhne, K., Keyser B, Groene EF, et al. FBN1 gene mutation char-acteristics and clinical features for the prediction of mitral valvedisease progression. Int J Cardiol. 2012;168:953–959.

40. Rybczynski M, Mir TS, Sheikhzadeh S, et al. Frequency and age-related course of mitral valve dysfunction in the Marfan syndrome.Am J Cardiol. 2010;106(7):1048–1053.

41. Faivre L, Collod-Beroud G, Loeys BL, et al. Effect of mutation typeand location on clinical outcome in 1,013 Probands with Marfansyndrome or related phenotypes and FBN1 mutations: an interna-tional study. Am J Hum Genet. 2007;81(3):454–466.

42. Rand-Hendriksen S, Johansen H, Semb SO, et al. Health-relatedquality of life in Marfan syndrome: a cross-sectional study ofshort form 36 in 84 adults with a verified diagnosis. Genet Med.2010;12(8):517–524.

EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 597

43. von Kodolitsch Y, Blankart CR, Vogler M, et al. Genetics and pre-vention of genetic aortic syndromes (GAS) and of the Marfansyndrome. Bundesgesundheitsblatt - Gesundheitsforschung -Gesundheitsschutz. 2014;58(2):146–153.

44. De Backer J, Renard M, Campens L, et al. Marfan syndrome andrelated heritable thoracic aortic Aneurysms and Dissections. CurrPharm Des. 2015;21(28):4061–4075.

45. Habashi JP, Judge DP, Holm TM, et al. Losartan, an AT1 Antagonist,prevents aortic Aneurysm in a mouse model of Marfan syndrome.Science. 2006;312(5770):117–121.

46. Travis J. Old drug, new hope for Marfan syndrome. Science.2006;312(5770):36–37.

47. Lacro RV, Dietz HC, Sleeper LA, et al. Atenolol versus losartan inchildren and young adults with Marfan’s syndrome. N Engl J Med.2014;371(22):2061–2071.

48. Braverman AC, Harris KM, Kovacs RJ, et al. Eligibility and disquali-fication recommendations for competitive athletes with cardiovas-cular abnormalities: task force 7: aortic diseases, including marfansyndrome: a scientific statement from the American heart associa-tion and American college of cardiology. J Am Coll Cardiol. 2015;66(21):2398–2405.

49. Erbel, R., Aboyans V, Boileau C, et al. 2014 ESC Guidelines on thediagnosis and treatment of aortic diseases: document covering acuteand chronic aortic diseases of the thoracic and abdominal aorta of theadultThe task force for the diagnosis and treatment of aortic diseasesof the European Society of Cardiology (ESC). Eur Heart J. 2014;35(41):2873–2926.

50. von Kodolitsch Y, Robinson P, Berger J, et al. When should surgerybe performed in marfan syndrome and other connective tissuedisorders to protect against type a dissection? In: Bonser RS,Editors. Controversies in Aortic dissection and Aneurysmal disease.London: Springer; 2014. p. 17–47.

51. Zanotti G, Vricella L, Cameron D. Thoracic aortic aneurysm syn-drome in children. Semin Thorac Cardiovasc Surg Pediatr Card SurgAnnu. 2008;11:11–21.

52. David TE, Feindel CM. An aortic valve-sparing operation for patientswith aortic incompetence and aneurysm of the ascending aorta. JThorac Cardiovasc Surg. 1992;103(4): 617–621; discussion 622.

53. Settepani F, Ornaghi D, Barbone A, et al. Aortic valve-sparingoperations in patients with aneurysms of the aortic root or ascend-ing aorta: preliminary results. Interact Cardiovasc Thorac Surg.2005;4(2):137–139.

54. Schäfers H-J, Aicher D, Langer F. Correction of leaflet prolapse invalve-preserving aortic replacement: pushing the limits? AnnThorac Surg. 2002;74(5):S1762–S1764.

55. Graeter TP, Aicher D, Langer F, et al. Mid-term results of aortic valvepreservation: remodelling vs. reimplantation. Thorac CardiovascSurg. 2002;50(1):21–24.

56. Horstkotte D, Schulte H, Bircks W, et al. Unexpected findings con-cerning thromboembolic complications and anticoagulation aftercomplete 10 year follow-up of patients with St. Jude medicalprostheses. J Heart Valve Dis. 1993;2:291–301.

57. Bernhardt AM, Treede H, Rybczynski M, et al. Comparison of aorticroot replacement in patients with Marfan syndrome. Eur JCardiothorac Surg. 2011;40(5):1052–1057.

58. Yacoub M., Fagan A, Stassano P, et al. Results of valve conservingoperations for aortic regurgitation. Circulation. 1983;1983(68):III-321 (abstract).

59. Tian D, Rahnavardi M, Yan TD. Aortic valve sparing operations inaortic root aneurysms: remodeling or reimplantation? AnnCardiothorac Surg. 2013;2(1):44–52.

60. Benedetto U, Melina G, Takkenberg JJM, et al. Surgical manage-ment of aortic root disease in Marfan syndrome: a systematicreview and meta-analysis. Heart (British Cardiac Society). 2011;97(12):955–958.

61. Gott VL, Greene PS, Alejo DE, et al. Replacement of the aortic rootin patients with Marfan’s syndrome. N Engl J Med. 1999;340(17):1307–1313.

62. Gott VL, Cameron DE, Alejo DE, et al. Aortic root replacement in271 Marfan patients: A 24-year experience. Ann Thorac Surg.2002;73(2):438–443.

63. Groenink M, Lohuis TA, Tijssen JG, et al. Survival and complicationfree survival in Marfan’s syndrome: implications of current guide-lines. Heart. 1999;82(4):499–504.

64. Ryosaka M, Omoto K, Nozaki T, et al. Successful kidney transplanta-tion for end-stage renal disease in Marfan’s syndrome. Case RepTransplant. 2013;2013:4.

65. Lipton GE, Guille JT, Kumar SJ. Surgical treatment of scoliosis inMarfan syndrome: guidelines for a successful outcome. J PediatrOrthopaedics. 2002;22(3):302–307.

66. Knosalla C, Weng Y-G, Hammerschmidt R, et al. Orthotopic hearttransplantation in patients with Marfan syndrome. Ann ThoracSurg. 2007;83(5):1691–1695.

67. Groth KA, Hove H, Kyhl K, et al. Prevalence, incidence, and age atdiagnosis in Marfan syndrome. Orphanet J Rare Dis. 2015;10:153.

68. von Kodolitsch Y, Bernhardt AM, Kölbel T, et al. Maximizing ther-apeutic success: The key concepts of individualized medical strat-egy (IMS). Cogent Med. 2015;2(1):1109742.

69. Manow M-L, Paulsen N, Rybczynski M, et al. Analysis of costs andprofits of ambulatory care of Marfan patients after initiation of anovel German legal directive (§ 116 b SGB V). Med Klin. 2010;105(8):529–537.

70. Achelrod, D., Blankart CR, Linder R, et al. The economic impactof Marfan syndrome: a non-experimental, retrospective, popu-lation-based matched cohort study. Orphanet J Rare Dis. 2014;9(1):1–12.

71. Loeys BL, Dietz HC, Braverman AC, et al. The revised Ghent nosol-ogy for the Marfan syndrome. J Med Genet. 2010;47(7):476–485.

72. von Kodolitsch Y, Rybczynski M, Vogler M. The role of the multi-disciplinary healthcare team in the management of patients withMarfan syndrome. J Multidiscip Healthc. 2016. Forthcoming

73. Halliday DJ, Hutchinson S, Lonie L, et al. Twelve novel FBN1 muta-tions in Marfan syndrome and Marfan related phenotypes test thefeasibility of FBN1 mutation testing in clinical practice. J MedGenet. 2002;39(8):589–593.

74. von Kodolitsch Y, Kutsche K. Interpretation of sequence variants ofthe FBN1 gene: analog or digital? A commentary on decreasedfrequency of FBN1 missense variants in Ghent criteria-positiveMarfan syndrome and characterization of novel FBN1 variants. JHum Genet. 2015;60:465–466.

75. Roll K. The influence of regional health care structures on delay indiagnosis of rare diseases: the case of Marfan syndrome. HealthPolicy. 2012;105(2–3):119–127.

76. Jondeau G, Detaint D, Tubach F, et al. Aortic event rate in theMarfan population: a cohort study. Circulation. 2012;125(2):226–232.

598 C. R. BLANKART ET AL.