Marfan Syndrome

Overview and Epidemiology

Marfan syndrome is an autosomal dominant connective tissue disorder caused by mutations in FBN1, the gene for fibrillin-1, located on chromosome 15q21.1 (Pepe 2016, PubMed PMID: 27274304). Its cardinal manifestations are proximal aortic aneurysm, dislocation of the ocular lens and long-bone overgrowth (Judge & Dietz 2005, PubMed PMID: 16325700). Its frequency is most often given as about 1 in 5,000 (Pepe 2016; Du 2021, PubMed PMID: 34220303; the 2022 ACC/AHA aortic disease guideline, Isselbacher 2022, PubMed PMID: 36322642); other reviews give 1 in 5,000–10,000 (Pollock 2021; Muiño-Mosquera 2024), and a Danish registry counted 6.5 per 100,000 in 2015 (Du 2021). It occurs with similar frequency in both sexes and in all countries and races (von Kodolitsch 2015, PubMed PMID: 26124674). The syndrome is named after Marfan, who in 1896 described a 5-year-old girl with multiple joint contractures, thoracolumbar kyphoscoliosis, and long, thin limbs, fingers and toes that he called “spider-like” (von Kodolitsch 2015).

About 25% of disease-causing FBN1 variants are de novo (new in the affected person), according to a 2021 ophthalmology review (Akram 2021, PubMed PMID: 37181104). Clinical expression varies widely, both between and within families, and features can appear at any age from birth to about 60 years (Pepe 2016).

Life expectancy has risen. Data reported in 1972 gave a mean age at death of 32 years, with most deaths cardiovascular; a 1995 study of 417 patients from four referral centres found a mean age at death of 41 years and a median cumulative survival of 72 years in 1993, against 48 years in 1972 (Silverman 1995, PubMed PMID: 7810492). The authors attributed the gain partly to cardiovascular surgery, partly to milder cases being diagnosed more often, and partly to general gains in population life expectancy; medical therapy, including beta-blockers, was also associated with longer survival. Dissection of the aortic root remains the leading cause of death (Lacro 2014, PubMed PMID: 25405392).

No newer survival cohort is cited in the sources checked for this page, and current estimates are stated more loosely. A 2021 review estimates that improved cardiovascular management has extended life expectancy by perhaps 30–50 years from the estimated mean of 32 years in 1972 (Pollock 2021); a 2022 review describes life expectancy with proper treatment as close to that of the general population (Chiu 2022); and a 2021 review reports an average lifespan of 50 years among Danish patients (Du 2021). The 2022 ACC/AHA guideline states that prophylactic aortic root replacement prevents type A dissection and improves survival in Marfan syndrome, and describes a 1995 report as a landmark that documented the marked improvement in lifespan with elective aortic repair compared with historical controls (Isselbacher 2022).

Back to Table of Contents

Genetics: FBN1 and the TGF-β Paradox

FBN1 has 66 exons and encodes fibrillin-1, an extracellular-matrix glycoprotein that is a structural component of microfibrils (Dietz 1991, PubMed PMID: 1852208; Pepe 2016). Fibrillin-1 is found throughout the body, which accounts for the wide spread of manifestations (Muiño-Mosquera 2024, PubMed PMID: 39250726). The zonules that suspend the lens of the eye are made of fibrillin (Esfandiari 2019, PubMed PMID: 30820290).

Two lines of explanation run through the research:

• Structural: fibrillin-1 is part of the scaffold of the extracellular matrix; with haploinsufficiency (too little normal fibrillin-1), a thinner fibrillin matrix is thought to reduce aortic wall strength (Pepe 2016).
• TGF-β signalling: in fibrillin-1-deficient mice, TGF-β activation and signalling were markedly dysregulated in the developing lung, and blocking TGF-β around birth rescued alveolar development (Neptune 2003, PubMed PMID: 12598898). In a Marfan mouse model, aortic aneurysm was associated with increased TGF-β signalling and was prevented by a TGF-β-neutralizing antibody or by the angiotensin II type 1 receptor blocker losartan (Habashi 2006, PubMed PMID: 16601194). These were animal studies. A 2016 review notes later human data suggesting that increased angiotensin II, rather than elevated TGF-β, may start aortic dilatation, with TGF-β acting more as a marker of aortic damage (Pepe 2016).

More than 2,000 FBN1 mutations had been published by 2016 (Pepe 2016). Disease-causing variants are spread throughout the gene and are mostly private to one family. Missense variants in the calcium-binding EGF-like domains are the most common type; haploinsufficiency is the cause in 30%–40% of cases, and up to 7% of variants are large or complete deletions of FBN1. Early-onset (neonatal) Marfan syndrome is often linked to de novo missense variants and in-frame deletions clustered within exons 24–32; variants that disrupt cysteine residues are associated with ectopia lentis (Muiño-Mosquera 2024).

FBN1 mutations also underlie other conditions, including the MASS phenotype (myopia, mitral valve prolapse, borderline aortic root enlargement, skeletal findings and striae), acromelic dysplasia, Weill–Marchesani syndrome and stiff skin syndrome (Pepe 2016). Related categories defined alongside the revised Ghent criteria include ectopia lentis syndrome (lens dislocation with an FBN1 variant not known to be associated with aortic disease, or with no FBN1 variant) and mitral valve prolapse syndrome (Chiu 2022, PubMed PMID: 35233355).

Back to Table of Contents

Cardiovascular Manifestations

Cardiovascular collapse from aortic dissection or rupture is the main cause of death in Marfan syndrome (Chiu 2022). The aorta dilates at the level of the sinuses of Valsalva; the proximal aorta bears the pulsatile load of left ventricular ejection, and elastin content there is high (Muiño-Mosquera 2024). By about age 60, close to all patients have some degree of aortic root dilatation, and about three quarters have had aortic root replacement (Pepe 2016).

Aortic root dilation is expressed as a Z-score: the number of standard deviations a measurement lies above or below the population mean (Chiu 2022), indexed to body-surface area. Different published nomograms give different Z-scores for the same measurement, so using one method consistently matters most (Muiño-Mosquera 2024). For scale, in the AIMS trial (mean baseline root diameter about 34 mm) the root grew by a mean of 0.74 mm per year on placebo (Mullen 2019, PubMed PMID: 31836196).

Aortic dissection is the most feared complication. A larger aortic diameter goes with a higher risk of dissection, but some patients dissect below surgical thresholds; dilatation of the sinotubular junction, rapid aortic growth and male sex have been associated with more aortic root replacement, though not necessarily with more dissection (Muiño-Mosquera 2024). In a 10-centre series of 675 patients having aortic root replacement, 46% of the 158 adults with dissection and a documented diameter had an aneurysm of 6.5 cm or less (Gott 1999, PubMed PMID: 10219065). Root dissection is unusual in children but can occur in adolescents (Muiño-Mosquera 2024). The 2022 ACC/AHA guideline reports type B (descending) dissection as the first aortic event in about 10% of patients, sometimes after earlier root replacement (Isselbacher 2022).

Arterial tortuosity is not confined to Loeys-Dietz syndrome. In an MRA study of 90 patients with connective tissue disorders, the vertebral tortuosity index was raised in Marfan syndrome (median 26) as well as in Loeys-Dietz syndrome (median 58) compared with controls (median 4.5); a higher index was associated with younger age at surgery, an index of 50 or more with earlier dissection and death, and age at surgery, dissection or death did not differ between Marfan syndrome and Loeys-Dietz syndrome (Morris 2011, PubMed PMID: 21730308). The 2022 guideline lists marked vertebral arterial tortuosity among the features associated with increased risk of aortic complications in Marfan syndrome, together with a family history of aortic dissection, rapid aortic growth (0.3 cm per year or more) and diffuse dilation of the aortic root and ascending aorta (Isselbacher 2022).

Mitral valve prolapse: in one cohort of 204 patients, mitral valve prolapse was found in 42.6% by age 30 (Du 2021). In children, mitral valve prolapse can be complicated by severe mitral regurgitation and myocardial dysfunction leading to heart failure (Pepe 2016).

Other cardiovascular features reported in the reviews:

• Pulmonary artery — proximal pulmonary artery enlargement and tricuspid valve prolapse are among the features seen in children (Muiño-Mosquera 2024)
• Left ventricular dysfunction — usually attributed to valve regurgitation, but primary heart-muscle disease has been found in the absence of valve disease, and early LV dysfunction can be detected by newer echo techniques (Pepe 2016)
• Arrhythmia — primary cardiomyopathy and ventricular arrhythmia leading to sudden cardiac death are less common but noted complications, reported mainly in adults (Muiño-Mosquera 2024); arrhythmia was one of the two principal causes of late death after aortic root replacement in the Gott series

Imaging: a 2016 review describes standard follow-up as echocardiography every one or two years, and notes that because echo cannot show the whole thoracic aorta, periodic MRI or CT is needed (Pepe 2016). A 2024 European paediatric statement recommends yearly echocardiography in children after diagnosis, biannual surveillance with large diameters (≥45 mm) or rapid growth (>5 mm per year), and MRI in preference to CT angiography to limit radiation (Muiño-Mosquera 2024). The 2022 ACC/AHA guideline recommends a transthoracic echocardiogram at diagnosis and 6 months later to determine the growth rate, then annually if the diameters are stable; considers CT or MRI of the thoracic aorta reasonable in adults after the first echocardiogram; and, after aortic root replacement, recommends MRI (or CT) of the thoracic aorta annually and then every other year if the diameters are normal and unchanged after 2 years (Isselbacher 2022).

Back to Table of Contents

Ocular Manifestations

Ectopia lentis (dislocation of the crystalline lens) is a cardinal feature of the revised Ghent criteria. Reported frequencies vary between sources: one review gives 30% to 72% across studies in its text and 60% to 80% in its abstract (Esfandiari 2019), and another gives up to 70% (Akram 2021). Subluxation is usually upward and toward the temple (superior and temporal), although dislocation into the vitreous or the anterior chamber can also occur (Esfandiari 2019).

Other eye findings include myopia, glaucoma and flattening of the cornea (Akram 2021); an abnormally flat cornea and increased axial length of the globe were minor criteria in the 1996 nosology (Esfandiari 2019). Myopia of more than 3 diopters scores one point in the systemic score.

• Refraction — prescribing glasses with a subluxated lens is difficult; in mild cases correcting myopic astigmatism is all that is needed (Esfandiari 2019)
• Retinal detachment — reported in 5% to 25.6% of patients across studies (Esfandiari 2019), and in around one third in another review (Akram 2021); whether the risk comes from the long, myopic eye or from the FBN1 defect itself is unclear (Akram 2021)
• Glaucoma — tends to be diagnosed at a younger age than in the general population; anterior lens dislocation can occur spontaneously or after trauma, and lens removal is the definitive treatment for it (Esfandiari 2019)

Back to Table of Contents

Skeletal Manifestations

Long-bone overgrowth is one of the cardinal manifestations (Judge & Dietz 2005). Many skeletal features feed the systemic score of the revised Ghent nosology; a 2021 review notes that some features used in older criteria, such as a high-arched palate and hernias, have been dropped from the list altogether (Pollock 2021, PubMed PMID: 34825999).

Body proportions: the earlier nosologies used an upper-to-lower segment ratio below 0.85 or an arm span-to-height ratio above 1.05 (von Kodolitsch 2015); the revised score gives one point for a reduced upper/lower segment ratio together with an increased arm span/height ratio (above 1.05) and no severe scoliosis (Chiu 2022).

Hand signs (scoring 3 points together, 1 point for either alone) (Pollock 2021):

• Thumb (Steinberg) sign — with a fist made over the clenched thumb, the entire distal phalanx of the thumb protrudes beyond the ulnar border of the palm
• Wrist (Walker–Murdoch) sign — when one wrist is encircled with the thumb and little finger of the other hand, the thumb covers the nail of the little finger completely

Scoliosis affects more than 50% of patients when a Cobb angle above 10° is counted; the 2010 nosology uses an angle above 20° for specificity. It contributes only one point, probably because idiopathic scoliosis is common in the general population. Unlike adolescent idiopathic scoliosis, scoliosis in Marfan syndrome has an equal sex ratio and a tendency to hyperkyphosis (Pollock 2021).

Pectus deformities are pectus carinatum (“pigeon chest”, 2 points) or pectus excavatum (funnel chest, 1 point). The effect of chest wall deformity on heart and lung function has been debated; in young people the deformity may improve with age, and reports caution against combining pectus repair with aortic surgery in very young patients (Pollock 2021).

Dural ectasia — widening of the dural sac, most often in the lumbosacral spine — is estimated in 63% to 97% of patients across studies; it scores 2 points. MRI shows the dural sac better than CT and is generally preferred. Back pain, radicular pain, headaches and other symptoms have been reported, but attributing them to dural ectasia is difficult, and even severe ectasia may cause no symptoms. Difficulty with spinal anaesthesia has also been reported (Pollock 2021).

Other features in the systemic score:

• Foot — hindfoot deformity (“hindfoot valgus in combination with forefoot abduction and lowering of the midfoot”, 2 points); plain flat foot, common in the general population (1 point) (Pollock 2021)
• Protrusio acetabuli — prevalence estimates range from 16% to 77.4% depending on definition (2 points) (Pollock 2021)
• Reduced elbow extension — 1 point, though the 2021 review's authors note it is not commonly seen (Pollock 2021)
• Facial features — 3 of 5 (dolichocephaly, enophthalmos, malar hypoplasia, downslanting palpebral fissures, retrognathia) score 1 point (Chiu 2022)
• Skin striae (1 point) and spontaneous pneumothorax from apical lung blebs (2 points) (Pepe 2016)

Back to Table of Contents

Revised Ghent Nosology: Diagnostic Criteria

The Revised Ghent Nosology (2010, Loeys et al., PubMed PMID: 20591885) replaced the 1996 Ghent criteria. It puts more weight on the cardiovascular manifestations, with aortic root aneurysm and ectopia lentis as the cardinal features; without a family history, the two together are sufficient for diagnosis. Without one of them, a bona fide FBN1 mutation or a combination of systemic features (a new scoring system) is required. FBN1 testing is not mandatory but carries greater weight. The panel anticipated that the new criteria may delay a definitive diagnosis but would lower the risk of premature diagnosis or misdiagnosis.

In the absence of family history, the criteria are met by any of four combinations (Chiu 2022; Pepe 2016):

• Aortic root Z-score ≥2 (or dissection) AND ectopia lentis
• Aortic root Z-score ≥2 (or dissection) AND a causative FBN1 mutation
• Aortic root Z-score ≥2 (or dissection) AND systemic score ≥7
• Ectopia lentis AND an FBN1 mutation known to be associated with aortic disease

With a family history of Marfan syndrome, any one of these is sufficient: ectopia lentis; a systemic score ≥7; or aortic root dilatation with a Z-score ≥2 in adults (20 years and older) or ≥3 in those under 20 (Chiu 2022).

The systemic score (maximum 20 points; ≥7 indicates systemic involvement) (Chiu 2022):

• Wrist AND thumb sign: 3 points; wrist OR thumb sign: 1 point
• Pectus carinatum: 2 points; pectus excavatum or chest asymmetry: 1 point
• Hindfoot deformity: 2 points; plain pes planus: 1 point
• Pneumothorax: 2 points
• Dural ectasia: 2 points
• Protrusio acetabuli: 2 points
• Reduced US/LS ratio AND increased arm span/height AND no severe scoliosis: 1 point
• Scoliosis (Cobb angle at least 20°) or thoracolumbar kyphosis: 1 point
• Reduced elbow extension: 1 point
• 3 of 5 facial features: 1 point
• Skin striae: 1 point
• Myopia >3 diopters: 1 point
• Mitral valve prolapse: 1 point

A 2016 review describes how scores group into phenotypes: 4 or less suggests a nonspecific connective tissue disorder, 5–6 the MASS phenotype, and 7 or more potential Marfan syndrome (Pepe 2016).

Back to Table of Contents

Differential Diagnosis

Several heritable connective tissue disorders overlap with Marfan syndrome, and multi-gene sequencing panels are increasingly used to separate them (Pepe 2016).

Loeys-Dietz Syndrome (LDS) — an aortic-aneurysm syndrome first defined in families with mutations in TGFBR1 or TGFBR2. GeneReviews now lists six genes for the autosomal dominant form — SMAD2, SMAD3, TGFB2, TGFB3, TGFBR1 and TGFBR2 — plus a recessive form caused by two pathogenic variants in IPO8, and states that aortic dissection can occur at smaller aortic diameters and younger ages than in Marfan syndrome (Loeys and Dietz, GeneReviews, PubMed PMID: 20301312). The 2006 study of 52 affected families with TGFBR1 or TGFBR2 mutations described the triad of arterial tortuosity and aneurysms, hypertelorism (widely spaced eyes), and bifid uvula or cleft palate; aneurysms were aggressive (mean age at death 26.0 years) and pregnancy-related complications occurred in 6 of 12 women (Loeys 2006, PubMed PMID: 16928994). The 2014 primer contrasts the increased risk of dissection at or above a 5.0-cm aortic root in Marfan syndrome with dissections in LDS types 1, 2 and 3 at 3.9–4.0 cm (MacCarrick 2014, PubMed PMID: 24577266). The 2022 ACC/AHA guideline sets gene-specific surgical thresholds for LDS rather than a single figure; these are listed on the Loeys-Dietz Syndrome page (Isselbacher 2022).

Ehlers-Danlos Syndrome, vascular type (vEDS) — caused by mutations in COL3A1 (type III procollagen). In 220 index patients and 199 affected relatives, patients were at risk of arterial, bowel and uterine rupture; median survival was 48 years, most deaths came from arterial rupture, and complications of pregnancy led to death in 12 of 81 women who became pregnant (Pepin 2000, PubMed PMID: 10706896).

Homocystinuria — an autosomal recessive deficiency of cystathionine beta-synthase (CBS) (Akram 2021) that also causes lens dislocation. In an international survey of 629 patients, untreated B6-nonresponsive patients had an 82% chance of lens dislocation by age 10 and a 27% chance of a thromboembolic event by age 15; spinal osteoporosis and lower intellectual ability were also common. Pyridoxine treatment of late-detected B6-responsive patients slowed the occurrence of first thromboembolic events, and methionine restriction started in the newborn period prevented intellectual disability (Mudd 1985, PubMed PMID: 3872065).

MASS phenotype — aortic root Z-score below 2 AND systemic score ≥5 with at least one skeletal feature, without ectopia lentis (Chiu 2022).

Bicuspid aortic valve (BAV) — ascending aortic enlargement is strongly associated with BAV. A 2016 review reports an association of Marfan syndrome and BAV that brings more severe aortic involvement, and describes FBN1 mutations in some patients with BAV in whom Marfan syndrome had been clinically excluded (Pepe 2016).

Shprintzen-Goldberg syndrome (SGS) — overlaps considerably with Marfan and Loeys-Dietz syndromes, including aortic aneurysm; it is caused by variants in SKI, a repressor of TGF-β activity (Doyle 2012, PubMed PMID: 23023332), or rarely by FBN1 mutations (Chiu 2022).

Back to Table of Contents

Medical Management

Medical treatment aims to slow aortic root growth, since aortic root dilatation remains the best predictor of aortic events and no reliable biomarker exists (Muiño-Mosquera 2024).

Beta-blockers were first proposed in 1971 and are regarded as first-line prophylaxis; they are generally started once Marfan syndrome is diagnosed, especially before puberty, and continued lifelong, including after aortic surgery. A resting heart rate of about 60–70 beats per minute has been suggested for adults (Chiu 2022). A 2024 statement notes that their effect on hard outcomes such as dissection and death lacks robust evidence, but that they are considered standard of care; atenolol and propranolol are the most studied (Muiño-Mosquera 2024). The 2022 ACC/AHA guideline cites an open-label study, followed for more than 10 years, in which propranolol was associated with slower aortic root growth and fewer clinical events than no treatment (Isselbacher 2022).

Angiotensin receptor blockers (ARBs) were tested after losartan prevented aortic aneurysm in mice (Habashi 2006). The main clinical studies:

• Brooke et al. (2008, NEJM, PubMed PMID: 18579813) — a small cohort study of 18 children with severe aortic-root enlargement in whom other medical therapy had failed (17 on losartan, 1 on irbesartan): the rate of root growth fell from 3.54 to 0.46 mm per year after an ARB was started. The authors stated that the findings required confirmation in a randomized trial.
• Pediatric Heart Network trial (Lacro et al., 2014, NEJM, PubMed PMID: 25405392) — 608 participants aged 6 months to 25 years with an aortic-root Z-score above 3 were randomized to atenolol or losartan. The rate of change in Z-score did not differ significantly between groups over 3 years (P=0.08); the Z-score fell in both groups.
• AIMS trial (Mullen et al., 2019, Lancet, PubMed PMID: 31836196) — 192 people aged 6–40 years were randomized to irbesartan or placebo (56% were also taking beta-blockers). The aortic root grew 0.53 mm per year on irbesartan versus 0.74 mm per year on placebo (difference −0.22 mm per year, p=0.030), with no difference in serious adverse events.

• COMPARE (Groenink et al., 2013, Eur Heart J, PubMed PMID: 23999449) — 233 adults, with or without earlier aortic surgery, were randomized to losartan or no additional treatment. Aortic root dilatation over about 3 years was lower with losartan (0.77 versus 1.35 mm), dilatation beyond the root was not significantly reduced, and no significant difference in clinical endpoints (aortic dissection, elective aortic surgery, cardiovascular death) was shown.
• French placebo-controlled trial — in patients already on beta-blockers, losartan lowered blood pressure without slowing aortic growth over 3 years (as described in Pepe 2016).
• Individual-patient meta-analysis (Pitcher et al., 2022, Lancet, PubMed PMID: 36049495) — seven trials and 1,442 patients without earlier aortic surgery. In four trials (676 participants) comparing an ARB with control, an ARB roughly halved the annual increase in aortic root Z-score, including among people also taking a beta-blocker, and the effect was larger in those with FBN1 variants; in three trials (766 participants) comparing an ARB with a beta-blocker, the change was similar in both groups.

The 2022 ACC/AHA guideline recommends either a beta-blocker or an ARB at maximally tolerated doses, unless contraindicated, to reduce the rate of aortic dilation (class 1, level of evidence A), and considers the two together reasonable (class 2a). It notes that randomized trials comparing an ARB with a beta-blocker found no significant difference in aortic root growth or in clinical events such as aortic surgery or dissection (Isselbacher 2022). European guidelines, as summarised in the 2024 paediatric statement, recommend beta-blockers and consider ARBs an alternative. Use of calcium channel blockers has been linked to more acute aortic events (Muiño-Mosquera 2024); in Marfan mice they accelerated aneurysm growth and dissection (Chiu 2022).

Physical activity: for patients with significant aortic disease in general, the 2022 ACC/AHA guideline recommends education and guidance about avoiding intense isometric exercise (such as heavy weightlifting or activities requiring the Valsalva manoeuvre), burst exertion and collision sports, and considers it reasonable to encourage 30 to 60 minutes of mild-to-moderate aerobic activity on at least 3 to 4 days a week when blood pressure is controlled. It states that limited data are available, that heavy lifting with the Valsalva manoeuvre can raise systolic pressure above 300 mm Hg, that expert committees agree light weightlifting and low-intensity aerobic exercise are safe, that no uniform consensus exists on intermediate-level exercise, and that exercise intensity is best individualised (Isselbacher 2022).

For children, the 2024 European paediatric statement does not recommend competitive-level sport except skill-based, low-intensity sports; it considers any recreational sport possible for children 10 years and under; it does not recommend power sports above age 10; and it permits skill, mixed or endurance sports at a recreational level, with intensity depending on aortic size (Muiño-Mosquera 2024). It notes that no data exist on exercise-associated aortic growth, that exercise-related dissection in healthy people is rare and mostly occurs in adults during heavy lifting, and that in Marfan mice mild aerobic exercise slowed aortic root dilatation while no training or high-intensity training worsened it.

Other medicines: stimulant medicines for ADHD can mildly raise heart rate and blood pressure; the 2024 statement describes non-stimulant drugs as considered safer but sometimes less effective, with decisions made case by case (Muiño-Mosquera 2024).

Back to Table of Contents

Surgical Management

Elective replacement of the aortic root before dissection has a low operative mortality. In the 10-centre series of 675 patients, 30-day mortality was 1.5% for elective repair, 2.6% for urgent repair and 11.7% for emergency repair, usually for acute dissection; the authors suggested prophylactic repair well below a diameter of 6.5 cm (Gott 1999, PubMed PMID: 10219065). In-hospital mortality for elective surgery in experienced centres is estimated below 1% (Muiño-Mosquera 2024).

Diameter thresholds reported from guidelines:

• 2022 ACC/AHA guideline: replacement of the aortic root and ascending aorta is recommended at an aortic root diameter of 5.0 cm or more, and is considered reasonable at 4.5 cm or more with features associated with increased dissection risk (listed in the cardiovascular section above), or when the maximal cross-sectional root area (cm²) divided by height (m) is 10 or more, when performed by experienced surgeons in a multidisciplinary aortic team. For a non-dissected aneurysm of the arch, descending or abdominal aorta, surgery at 5.0 cm or more is considered reasonable. The guideline states that the risk of dissection is low below 5.0 cm in patients on optimal medical therapy, and that about 20% of patients in large series had elective surgery below 5.0 cm (Isselbacher 2022)
• European Society of Cardiology (as summarized in 2021): surgery at a maximal aortic diameter ≥50 mm, or ≥45 mm with additional risk factors such as a family history of dissection at a low diameter, progressive aortic regurgitation, desire for pregnancy, or growth of more than 3 mm per year (Du 2021)
• 2024 European paediatric statement: surgery recommended in children at an aortic root diameter of 50 mm, and possibly at 45 mm with additional risk factors such as a family history of dissection, growth over 5 mm per year, or a need for valve surgery (Muiño-Mosquera 2024)

Valve-sparing root replacement keeps the patient's own aortic valve. The 2024 paediatric statement prefers valve-sparing surgery to valve replacement; among valve-sparing techniques it names reimplantation (David procedure), remodelling (Yacoub procedure) and the newer PEARS external support, and notes that reimplantation is preferred in adults because remodelling has a higher rate of re-intervention for aortic regurgitation (Muiño-Mosquera 2024). In a single-centre series of 146 patients followed for a median of 10 years (121 reimplantations, 25 remodellings), 5 needed reoperation on the aortic valve, the 15-year rate of aortic insufficiency was 7.9%, lower after reimplantation, and residual and new aortic dissections were the leading cause of death (David 2015, PubMed PMID: 26403341).

Composite graft replacement (replacement of the root with a prosthetic graft and valve) was the procedure studied in the Gott series. When a mechanical valve is implanted, lifelong anticoagulation is required, and a 2016 review notes that operating unusually early in youth can raise the chance of reoperation (Pepe 2016). In the Gott series there were 114 late deaths; dissection or rupture of the residual aorta (22 patients) and arrhythmia (21 patients) were the principal causes (Gott 1999).

Mitral valve surgery: repair can be performed when severe regurgitation develops, but a 2016 review reports an increased risk of aortic dissection afterward, attributed to the rapid postoperative rise in cardiac output (Pepe 2016).

Descending aorta: as survival has improved, more patients need procedures on the descending thoracic aorta for new or residual dissection or distal aneurysm. A 2015 review of endovascular repair (TEVAR) in Marfan patients cites a study in which retrograde aortic dissection was the most common complication; it cites a position statement that does not recommend TEVAR in connective tissue disease except as a bail-out or bridge to open surgery, or when both landing zones lie within earlier prosthetic grafts, and European guidelines that favour surgery over TEVAR outside emergencies (Parisi 2015, PubMed PMID: 28943617).

Back to Table of Contents

Pregnancy and Special Populations

In the US GenTAC registry, 94 of 184 women with Marfan syndrome had 227 pregnancies; 10 (10.6%) had a pregnancy-related aortic complication (4 type A and 3 type B dissections, 1 coronary artery dissection and 2 with aortic growth of 3 mm or more). The dissection rate during pregnancy and the postpartum period was 5.4 per 100 person-years against 0.6 outside pregnancy (rate ratio 8.4). Three quarters of dissections occurred postpartum, and only 5 of 8 women with a pregnancy-associated dissection knew they had Marfan syndrome (Roman 2016, PubMed PMID: 27515814). The authors concluded that the findings underscore the need for early diagnosis, pre-pregnancy risk counselling and multidisciplinary peripartum care.

• Aortic size — a Dutch prospective study of 33 pregnancies (pre-pregnancy root diameters 25–45 mm) found no significant change in root diameter across pregnancy and no significant difference in long-term growth against matched childless women; the authors concluded pregnancy seems relatively safe up to a root diameter of 45 mm, within the range they observed (Meijboom 2005, PubMed PMID: 15681576). The GenTAC report notes that Pyeritz recommended against pregnancy with aortic diameters over 40 mm (Roman 2016), and European guidance as summarized in 2021 lists desire for pregnancy among the risk factors that lower the surgical threshold to 45 mm (Du 2021).
• 2022 ACC/AHA guideline — it describes a low risk (about 1%) of pregnancy-related type A dissection at an aortic root below 4.0 cm and a much greater risk above 4.5 cm; type B dissection accounts for 20% to 40% of pregnancy-related dissections in Marfan syndrome, often without aortic dilation; and dissection often occurs postpartum, with raised risk up to 12 weeks after delivery. It recommends aortic surgery before pregnancy when the root diameter is above 4.5 cm, and states that surgery may be considered at 4.0 to 4.5 cm, especially with rapid aortic growth (0.3 cm per year or more) or a family history of aortic dissection (Isselbacher 2022)
• ARBs — the US prescribing information for losartan carries a boxed warning for fetal toxicity: drugs acting on the renin-angiotensin system can injure and kill the developing fetus, use in the second and third trimesters reduces fetal kidney function, and the drug is to be discontinued as soon as pregnancy is detected (DailyMed: losartan labels). The 2022 ACC/AHA guideline describes ARBs and ACE inhibitors as contraindicated during pregnancy because of teratogenicity (Isselbacher 2022)
• Delivery and outcomes — in a UK series of 29 pregnancies in 21 women there were no maternal deaths; significant cardiac complications occurred in 5 pregnancies (17%), including one type A dissection. The caesarean rate (41%) did not differ significantly from controls (26%), postpartum haemorrhage was more common (odds ratio 8.46), and babies were more often small for gestational age (24% versus 6%) (Curry 2014, PubMed PMID: 24418012)

Early-onset (neonatal) Marfan syndrome presents at birth with camptodactyly, arachnodactyly, joint contractures, muscle hypoplasia and loose skin. Affected infants often develop severe cardiorespiratory compromise, including congenital emphysema and atrioventricular valve regurgitation, leading to early death; the variants usually lie in exons 24–32 (Muiño-Mosquera 2024).

Children: the 2024 European paediatric statement recommends a beta-blocker or ARB at maximally tolerated doses in children with aortic dilatation, and treatment can be considered without dilatation when other risk factors are present. Because aortic root dilatation has incomplete penetrance, it notes that treating every child would mean unnecessary lifelong treatment for about 20%. Combined therapy is to be considered with fast progression, a root above 40 mm, or a Z-score above 5 (Muiño-Mosquera 2024).

Psychosocial impact: the 2010 nosology panel noted that the diagnosis, whether correct or not, can be stigmatising, hamper career aspirations, restrict life-insurance opportunities and cause psychosocial burden (Loeys 2010). In the Pediatric Heart Network trial population, ADHD significantly affected health-related quality of life (Muiño-Mosquera 2024).

Back to Table of Contents

Key Research Papers

1. Loeys BL et al. (2010) — The revised Ghent nosology for the Marfan syndrome. J Med Genet.
   — PubMed PMID: 20591885
2. Judge DP, Dietz HC (2005) — Marfan's syndrome. Lancet.
   — PubMed PMID: 16325700
3. Dietz HC et al. (1991) — Marfan syndrome caused by a recurrent de novo missense mutation in the fibrillin gene. Nature.
   — PubMed PMID: 1852208
4. Neptune ER et al. (2003) — Dysregulation of TGF-β activation contributes to pathogenesis in Marfan syndrome. Nat Genet.
   — PubMed PMID: 12598898
5. Brooke BS et al. (2008) — Angiotensin II blockade and aortic-root dilation in Marfan's syndrome. N Engl J Med.
   — PubMed PMID: 18579813
6. Lacro RV et al. (2014) — Atenolol versus losartan in children and young adults with Marfan's syndrome. N Engl J Med.
   — PubMed PMID: 25405392
7. Mullen M et al. (2019) — Irbesartan in Marfan syndrome (AIMS): a double-blind, placebo-controlled randomised trial. Lancet.
   — PubMed PMID: 31836196
8. Silverman DI et al. (1995) — Life expectancy in the Marfan syndrome. Am J Cardiol.
   — PubMed PMID: 7810492
9. David TE et al. (2015) — Outcomes of aortic valve-sparing operations in Marfan syndrome. J Am Coll Cardiol. 66(13):1445-53.
   — PubMed PMID: 26403341
10. Gott VL et al. (1999) — Replacement of the aortic root in patients with Marfan's syndrome. N Engl J Med. 340(17):1307-13.
    — PubMed PMID: 10219065
11. De Paepe A et al. (1996) — Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet.
    — PubMed PMID: 8723076
12. Meijboom LJ et al. (2005) — Pregnancy and aortic root growth in the Marfan syndrome: a prospective study. Eur Heart J. 26(9):914-20.
    — PubMed PMID: 15681576
13. Habashi JP et al. (2006) — Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science. 312(5770):117-21.
    — PubMed PMID: 16601194
14. Roman MJ et al. (2016) — Aortic complications associated with pregnancy in Marfan syndrome: the NHLBI National Registry of Genetically Triggered Thoracic Aortic Aneurysms and Cardiovascular Conditions (GenTAC). J Am Heart Assoc. 5(8):e004052.
    — PubMed PMID: 27515814
15. Curry RA et al. (2014) — Marfan syndrome and pregnancy: maternal and neonatal outcomes. BJOG. 121(5):610-7.
    — PubMed PMID: 24418012
16. Muiño-Mosquera L et al. (2024) — Management of aortic disease in children with FBN1-related Marfan syndrome. Eur Heart J. 45(39):4156-4169.
    — PubMed PMID: 39250726
17. Pepe G et al. (2016) — Marfan syndrome: current perspectives. Appl Clin Genet. 9:55-65.
    — PubMed PMID: 27274304
18. Chiu HH (2022) — An update of medical care in Marfan syndrome. Tzu Chi Med J. 34(1):44-48.
    — PubMed PMID: 35233355
19. Pollock L et al. (2021) — The musculoskeletal manifestations of Marfan syndrome: diagnosis, impact, and management. Curr Rheumatol Rep. 23(11):81.
    — PubMed PMID: 34825999
20. Esfandiari H et al. (2019) — Management strategies of ocular abnormalities in patients with Marfan syndrome: current perspective. J Ophthalmic Vis Res. 14(1):71-77.
    — PubMed PMID: 30820290
21. Akram H et al. (2021) — Marfan syndrome and the eye clinic: from diagnosis to management. Ther Adv Rare Dis. 2:26330040211055738.
    — PubMed PMID: 37181104
22. von Kodolitsch Y et al. (2015) — Perspectives on the revised Ghent criteria for the diagnosis of Marfan syndrome. Appl Clin Genet. 8:137-55.
    — PubMed PMID: 26124674
23. Du Q et al. (2021) — The molecular genetics of Marfan syndrome. Int J Med Sci. 18(13):2752-2766.
    — PubMed PMID: 34220303
24. Loeys BL et al. (2006) — Aneurysm syndromes caused by mutations in the TGF-beta receptor. N Engl J Med. 355(8):788-98.
    — PubMed PMID: 16928994
25. Pepin M et al. (2000) — Clinical and genetic features of Ehlers-Danlos syndrome type IV, the vascular type. N Engl J Med. 342(10):673-80.
    — PubMed PMID: 10706896
26. Mudd SH et al. (1985) — The natural history of homocystinuria due to cystathionine beta-synthase deficiency. Am J Hum Genet. 37(1):1-31.
    — PubMed PMID: 3872065
27. Doyle AJ et al. (2012) — Mutations in the TGF-β repressor SKI cause Shprintzen-Goldberg syndrome with aortic aneurysm. Nat Genet. 44(11):1249-54.
    — PubMed PMID: 23023332
28. Parisi R et al. (2015) — Endovascular repair of aortic dissection in Marfan syndrome: current status and future perspectives. Diseases. 3(3):159-166.
    — PubMed PMID: 28943617
29. Isselbacher EM et al. (2022) — 2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease. Circulation. 146(24):e334-e482. Also published in J Am Coll Cardiol 2022;80(24):e223-e393 (PMID 36334952).
    — PubMed PMID: 36322642
30. Groenink M et al. (2013) — Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J. 34(45):3491-500.
    — PubMed PMID: 23999449
31. Pitcher A et al.; Marfan Treatment Trialists' Collaboration (2022) — Angiotensin receptor blockers and β blockers in Marfan syndrome: an individual patient data meta-analysis of randomised trials. Lancet. 400(10355):822-831.
    — PubMed PMID: 36049495
32. Morris SA et al. (2011) — Increased vertebral artery tortuosity index is associated with adverse outcomes in children and young adults with connective tissue disorders. Circulation. 124(4):388-96.
    — PubMed PMID: 21730308
33. MacCarrick G et al. (2014) — Loeys-Dietz syndrome: a primer for diagnosis and management. Genet Med. 16(8):576-87.
    — PubMed PMID: 24577266
34. Loeys BL, Dietz HC — Loeys-Dietz Syndrome. In: GeneReviews. University of Washington, Seattle; 2008, updated 2024-09-12.
    — PubMed PMID: 20301312

Back to Table of Contents

Connections

• Genetics
• Ehlers-Danlos Syndrome — connective tissue disorder, collagen defect
• Loeys-Dietz Syndrome — TGF-β pathway aortic aneurysm, overlapping phenotype
• Noonan Syndrome — heritable connective tissue and cardiac disorder
• Cardiology — Aortic Dissection, MVP, Aortic Root Replacement, Aortic Regurgitation
• Ophthalmology — Ectopia Lentis, Myopia, Retinal Detachment, Glaucoma
• Lab Tests — Genetic Sequencing, Echocardiography, Aortic Z-score

Back to Table of Contents

Featured Videos