Epidemiology and Pathophysiology of Mitral Valve Prolapse: New Insights into Disease Progression, Genetics, and Molecular Basis

Introduction

Mitral valve prolapse (MVP) is a common disorder afflicting 2- 3% of the general population.^{1, 2} It is characterized by typical fibromyxomatous changes in the mitral leaflet tissue with superior displacement of one or both leaflets into the left atrium.^{3, 4} Based on a prevalence of 2-3%, MVP would be expected to affect approximately 7.8 million individuals in the United States and over 176 million people worldwide. MVP can be associated with significant mitral regurgitation (MR), bacterial endocarditis, congestive heart failure, and even sudden death.^5-7

MVP is a clinical entity that is not fully understood, despite being known for more than a century. A ‘mid-systolic click’ was first described in 1887 by Cuffer and Barbillon.⁸ In 1963 Barlow demonstrated the presence of MR by angiography in patients with the ‘click-murmur’ syndrome.⁹ Criley subsequently coined the term mitral valve prolapse.¹⁰

MVP may be familial or sporadic. Despite being the most common cause of isolated MR requiring surgical repair,¹¹ little is known about the genetic mechanisms underlying the pathogenesis and progression of MVP. Studies on the heritable features of MVP have been limited by the analysis of relatively small pedigrees and by self-referral and selection biases, including a preponderance of data from hospital-based cohorts.^{12, 13} Nonetheless, a majority of data favors an autosomal dominant pattern of inheritance in a large proportion of individuals with MVP.^{12, 13} Despite the variability in the clinical features, familial MVP might be considered a prevalent Mendelian cardiac abnormality in humans. While filamin A has been identified as causing an X-linked form of MVP,¹⁴ the causative genes for the more common form of autosomal dominant MVP have yet to be defined. In this review we summarize our current knowledge regarding the diagnosis, epidemiology, prognosis, pathophysiology, and genetics of MVP, with a focus on potential future research directions.

Diagnosis of MVP

Physical examination and two-dimensional (2D) echocardiography are the diagnostic gold standards for MVP.^{15, 16} Various symptoms (including atypical chest pain, exertional dyspnea, palpitations, syncope, and anxiety) and clinical findings (low blood pressure, leaner build, and electrocardiographic repolarization abnormalities) have been associated with MVP and their constellation termed the ‘mitral valve prolapse syndrome’.^{1, 17} Of the numerous reported correlates, only the association with leaner body mass has been reproducibly associated with MVP in the literature.¹ Abnormal autonomic function has been reported as the mechanism explaining symptoms in patients with MVP,¹⁸ but given its absence in asymptomatic MVP patients, it remains unclear whether MVP is directly related to autonomic dysfunction or any reported association is purely incidental.¹⁹ Hypomagnesemia and dysregulation of the renin-angiotensin-aldosterone system have also been demonstrated in MVP syndrome, albeit in small patient samples.^{20, 21}

The classic auscultatory finding in MVP is a dynamic, mid-to-late systolic click frequently associated with a high-pitched, late systolic murmur. A careful physical examination is highly sensitive for making a diagnosis of MVP, but its specificity is limited (using echocardiography as the gold standard).²² Redundant leaflets or chordae may produce an audible click without echocardiographic evidence of leaflet prolapse, giving false positive physical findings. Finally, echocardiographic prolapse may exist without significant auscultatory findings.²² Patients with physical examination findings that suggest MVP should undergo confirmatory testing with 2D echocardiography.

In the early days of 2D echocardiography, the diagnosis of MVP occurred with prevalence ranging from 5 to 15%, and even as high as 35% of those undergoing imaging.^23-25 In part, this over-diagnosis was due to the erroneous assumption that the mitral valve (MV) was planar; thus, any sonographic view that showed excursion of the leaflets superior to the mitral annulus was deemed pathological. Pivotal echocardiographic work in the late 1980s redefined normal mitral anatomy.²⁶ Using three-dimensional (3D) echo imaging, Levine and colleagues established that the mitral annulus was in fact saddle-shaped. Therefore, in the anterior-posterior axis, the mitral annulus is concave upward, whereas medially-to-laterally, the annulus is concave downward. This mitral geometry creates the possibility that in a sonographic four-chamber view the leaflets can appear to ‘break’ the annular plan (creating the appearance of prolapse), when in reality they are normal. Echocardiographic MVP has since been defined as single or bi-leaflet prolapse of at least 2 mm beyond the long-axis annular plane,²⁶ with or without mitral leaflet thickening (Figure 1A). Prolapse with thickening of the leaflets greater than 5 mm is termed ‘classic’ prolapse, whereas prolapse with lesser degrees of leaflet thickening is regarded as ‘non-classic’ prolapse.²⁶

Figure 1.

Prolapse of the intermediate posterior mitral valve scallop (P2) shown in a long axis view of A) a 2D transthoracic echocardiogram, B) a 2D transesophageal echocardiogram (TEE) with C) associated severe, eccentric, anteriorly directed mitral regurgitation, and in a D) 3D TEE surgical view. 2D, two-dimensional; 3D, three-dimensional; AO, aorta; LV, left ventricle; and RV, right ventricle.

Transthoracic echocardiography (TTE) may not adequately visualize the entire MV anatomy. Anatomically, the posterior and anterior leaflet of the MV each may be divided into three sections. Carpentier’s widely recognized nomenclature describes three posterior leaflet scallops, the lateral (P1), middle (P2), and medial (P3) — and three anterior segments — lateral (A1), middle (A2) and medial (A3) (Figure 1D).^{27, 28} Most cases of prolapse involve the posterior middle scallop, which is easily identified on long-axis TTE images (Figure 1A). However, the posterior lateral scallop (P1) is not clearly seen on long-axis images, but is best visualized in the apical four-chamber view. As noted above, superior leaflet displacement in a four-chamber view should not be regarded as diagnostic of prolapse. Thus, TTE can confirm the diagnosis of MVP, but may not be able to exclude prolapse of all scallops. While the Carpentier nomenclature is based on leaflet indentation, in the Duran classification scallops are grouped based on chordal attachments. ²⁹ Specifically, the anterior leaflet is divided into two segments (A1, A2) and the posterior into four segments (P1, PM1, P2, PM2). Segments A1, P1 and PM1 attach to the anterolateral papillary muscle, and segments A2, P2, and PM2 to the posteromedial papillary muscle. The modified Carpentier classification³⁰ is a combination of the Carpentier and the Duran nomenclatures. Although the Duran and the modified Carpentier are anatomically more precise than the classic Carpentier scheme, they are less widely utilized.

By taking into account several planes of imaging, 2D transesophageal echocardiography (TEE) is more effective in identifying prolapsing MV segments (Figure 1B).²⁸ 3D TEE has the additional advantage of simulating the surgeon’s view of the MV, with the aortic valve at the 11 o’clock position (Figure 1D), and has become an essential tool in the intra-operative setting.³¹

Cardiac Magnetic Resonance (CMR) represents a novel, albeit still not widely used noninvasive imaging method that identifies MVP with a sensitivity and specificity of 100%, using 2D TTE as the gold standard (Figure 2A).³² In addition, CMR can quantify MR using phase contrast velocity mapping.³³ Because CMR can reliably provide quantitative determination of ventricular volumes and function, it is becoming an important clinical tool for follow-up of patients with MVP-related moderate to severe MR, and surgical decision making.³⁴ Finally, CMR provides novel insight into the biology of the MV and its linked myocardium, through improved spatial resolution provided by 3D acquisition of images with delayed gadolinium enhancement.³² Such enhancement occurs when the kinetics of gadolinium excretion is different in two adjacent compartments, so that over time, one compartment enhances more than the other. This has been a powerful tool for delineating infarcted and scarred myocardium, which excrete gadolinium slower than viable tissue. The presence of gadolinium enhancement has been shown in both the MV and in the papillary muscle tips in patients with MVP, but not in normal controls (Figure 2B).³² It has been speculated that the papillary muscle is altered in MVP by repetitive traction exerted by the prolapsing leaflets,³⁵ which has been shown experimentally to lower the threshold for arrhythmias.³⁶ Although more frequent complex arrhythmia on 24-hour ambulatory Holter monitor has been demonstrated in MVP patients with scarring of the papillary muscles,³² its clinical significance remains to be established.

Figure 2.

Cardiac magnetic resonance Steady State Free Processing (SSFP) long-axis view of bileaflet mitral valve prolapse (panel A). Short axis view with 3D late-gadolinium enhancement (LGE) showing fibrosis of the papillary muscle tips. Adapted from reference 32.

Clinical Classification and Prevalence of MVP

MVP can be distinguished into primary or ‘non-syndromic’ MVP, and secondary or ‘syndromic’ MVP. In the latter case, MVP occurs in the presence of connective tissue disorders such as Marfan syndrome, Loeys-Dietz syndrome, Ehlers-Danlos, osteogenesis imperfecta, pseudoxanthoma elasticum, and the recently reported ‘aneurysms—osteoarthritis syndrome’.^37-42 MVP has also been observed in hypertrophic cardiomyopathy, and may contribute the pathophysiology of obstruction typical of this myopathy.⁴³

Non-syndromic MVP

Based on revised echocardiographic diagnostic criteria,²⁶ the prevalence of MVP and its clinical associations were examined in the community-based Framingham Heart Study (FHS).¹ The sample analyzed consisted of 3491 participants in whom routine 2D echocardiograms were available and adequate for the evaluation of the MV. Forty-seven individuals (1.3%) had classic and 37 (1.1%) had non-classic MVP, yielding an estimated overall prevalence of 2.4%. The prevalence of MVP was fairly evenly distributed among individuals in each decade of age from ages 30 to 80 years. With respect to gender, MVP was equally distributed between men and women. These findings differed from older studies based on M-mode diagnostic criteria and/or observations of pedigrees^{12, 13} that reported that MVP preferentially afflicted women and older individuals. Although the genetic predisposition to develop MVP may be present at birth, MVP is not found in newborns,⁴⁴ and its prevalence is low among children (0.3%),⁴⁵ and young adults (0.6%).⁴⁶ These findings suggest that MVP is a progressive disease affecting predominantly middle-aged individuals.¹ Participants with MVP in FHS were leaner compared to those without MVP.¹ An important limitation of the FHS sample is that it is predominantly white. A similar prevalence of MVP was described in a population-based sample of American Indians (the Strong Heart Study),² and in a different sample of Canadians of South Asian, European and Chinese descent (the SHARE study).⁴⁷ While these recent studies were based on revised echocardiographic criteria, the prevalence of MVP in African Americans was based on older M-mode criteria and non-standard 2D echocardiographic views.⁴⁸ A systematic review of the published literature did not reveal prior studies that have evaluated the prevalence of MVP in Hispanic samples.

Tricuspid valve prolapse has been observed in up to 40 to 50% of patients with primary or ‘non-syndromic’ MVP,⁴⁹ but isolated tricuspid prolapse has been rarely reported.

Syndromic MVP: Marfan syndrome and other connective tissue disorders

The prevalence of at least mild MV pathology in Marfan syndrome (MFS) has been estimated to be approximately 75%, whereas the prevalence of more severe myxomatous MV thickening with prolapse is closer to 25% in these individuals.⁵⁰ The prevalence of MVP in patients with Ehlers-Danlos syndrome using standard echocardiographic criteria appears to be much lower (6%).³⁷ The prevalence of MV disease also appears to be lower in the Loeys-Dietz syndrome (relative to MFS).⁵¹ One group reported a direct comparison of MVP prevalence in 71 individuals with TGFBR2 mutations (characteristic of Loeys-Dietz syndrome) with that in 243 people with FBN1 mutations (typical of MFS), and in 50 unaffected family members.⁵¹ The investigators observed a substantially higher prevalence of both MVP and MR in the cohort with FBN1 mutations than in the group with TGFBR2 mutations (45% and 56% vs. 21% and 35%, respectively). Among affected individuals with the aneurysms—osteoarthritis syndrome, MV abnormalities were common and ranged from mild to severe; 10 of 22 (45%) had MVP and 6 of 22 (27%) had MR.⁴² The presence of MVP has also been described in osteogenesis imperfecta and pseudoxanthoma elasticum,^{38, 52} although the true prevalence of the disease is unclear as standard diagnostic criteria were not used in the initial imaging studies performed on these patients.

Hypertrophic cardiomyopathy

The largest study assessing the prevalence of MVP in hypertrophic cardiomyopathy (HCM) observed it in 3% (of 528 people with HCM), which might suggest that HCM and MVP are two distinct conditions that may coexist in some cases.⁵³ Yet, the prevalence of other MV abnormalities (leaflet elongation and increased thickness) is much higher in HCM, estimated at 66% in one study.⁵⁴ This suggests that MV abnormalities are intrinsic to HCM,⁵⁴ either as a primary trait, or occur as a secondary adaptive response to 1) shear stress in a turbulent outflow tract or 2) paracrine effects arising in the adjacent hypertrophic ventricle (see below).⁵⁵

Prognosis of MVP

A controversial past

The prognosis of MVP has varied in the published literature. In the community-based FHS sample, MVP was described as a benign entity, with a low occurrence of adverse sequelae.¹ Specifically, none of the individuals with MVP had a history of heart failure, one (1.2 percent) had atrial fibrillation, one (1.2 percent) had cerebrovascular disease, and three (3.6 percent) had syncope, as compared with the prevalence of these outcomes in the subjects without prolapse of 0.7, 1.7, 1.5, and 3.0 percent, respectively. The frequencies of chest pain, dyspnea, and electrocardiographic abnormalities were similar among individuals with and without MVP. Individuals with MVP had a greater degree of MR than those without prolapse, but typically the valvular regurgitation was classified as trace or mild.¹ In prior studies,^{5, 6} MVP was portrayed as a disease with frequent and serious complications, including stroke, atrial fibrillation, heart failure, and MR requiring surgery. These discrepancies may be due to selection biases inherent in evaluating symptomatic patients at referral tertiary care centers, compared to observations made on healthier asymptomatic volunteers.^{1, 5, 6} Changes in diagnostic criteria for MVP over time may have further exacerbated these differences in prevalence of MVP.²⁶ Subsequently, a community-based study from the Mayo Clinic conducted in a primary care setting has underscored the clinical heterogeneity of MVP, including a widely varying prognostic spectrum.⁵⁶ Based on primary (depressed left ventricular ejection fraction, moderate/severe MR) and secondary (age > 50 years, mild MR, left atrial enlargement, atrial fibrillation, and flail leaflet) risk factors, different groups of MVP with varying prognosis were identified with regard to cardiovascular morbidity and mortality.⁵⁶ Overall, young (<50 years), medically treated patients presenting with normal left ventricular function and no symptoms have excellent survival, even with severe mitral regurgitation.^{56, 57} The benefit of early surgery (ie, valve repair in asymptomatic patients) versus a watchful wait was suggested in observational studies,⁵⁸ but remains controversial.⁵⁷

Impact of mitral regurgitation

The common denominator of the studies evaluating prognosis of MVP is the role of MR at the time of diagnosis in determining the risk for adverse events (such as congestive heart failure, atrial fibrillation, ischemic neurologic event, endocarditis), and the need for surgery on follow-up.^{1, 5-7, 59, 60} The Mayo clinic series highlighted that over a follow-up period of 1.5 years, MR volume increased by more than 8 mL in 51% of 74 individuals with MVP. In this clinical series, the progression of the valvular lesion (particularly a new flail leaflet), and an increase in the mitral annular diameter were the two independent predictors of an increase in the regurgitant volume over time.⁶⁰ Although mitral leaflet thickness >5 mm on M-mode echocardiography has been associated with increased risk for sudden death, endocarditis, and MR in patients with classic prolapse in some series,^{5, 6} a more recent larger series using 2D echocardiography reported that mitral leaflet thickness was not an independent predictor of mortality and valvular morbidity.⁷ In this community-based study of 833 individuals diagnosed with asymptomatic MVP and followed longitudinally in the Olmsted County, cardiac mortality was best predicted by the presence of MR and left ventricular dysfunction at the time of diagnosis. Risk factors for cardiac morbidity (defined as the occurrence of heart failure, thromboembolic events, endocarditis, atrial fibrillation, and need for cardiac surgery) included age ≥50 years, left atrial enlargement, MR, presence of a flail leaflet, and prevalent atrial fibrillation at the time of the baseline echocardiogram.⁷

Impact of a flail leaflet

The presence of a flail MV leaflet has been associated with a widely varying prognosis.⁶¹ Survival in medically treated asymptomatic patients with MVP presenting with a flail leaflet and normal left ventricular function is excellent.⁶¹ Thus, such patients are at relatively low risk of cardiovascular morbidity. The indications for valve surgery in this group include the development of atrial fibrillation (4 percent per year) and heart failure (5.7 percent per year). Older age, presence of symptoms, or a left ventricular ejection fraction < 60% at the time of initial diagnosis increase the risk of developing heart failure and atrial fibrillation, and are markers of the need for valve surgery and mortality.^{61, 62} As for chronic severe MR in general, management decisions for patients with flail leaflet are based largely upon the presence or absence of clinical symptoms, the functional state of the left ventricle, as well as the feasibility of successful MV repair.¹⁶

Sex-related differences in outcomes

As noted above, the prevalence of MVP was similar in the two sexes in the FHS, a referral-free, community-based sample.¹ Conversely, in the Olmsted county population, characterized by a mixed spectrum of community-dwelling and referred patients, using similar echocardiographic criteria, women were diagnosed with MVP more often than men and at a younger age.⁵⁶ However, complications (such as the development of a flail leaflet) have been reported more frequently in men.⁶² The Mayo clinic group also underscored the anatomical and functional differences among the two sexes in the context of MVP. Women present with more anterior and bi-leaflet prolapse, more thickened leaflets, less flail leaflets, and with less MR compared to men.⁶³ These milder clinical features in women have led to the speculation that an interplay may exist between loading conditions on the valve (such as higher blood pressure in men) and development of complications in MVP. However, women represent a large proportion of patients with moderate or severe MR. In these severe forms, assessment of left ventricular enlargement in women is challenged by the differences in weight and height among the two sexes, and the frequent use of an absolute rather than a body size-adjusted left ventricular size measurement. Consequently, for the same degree of MR, women undergo mitral surgery less frequently and later than men. As a consequence, women exhibit excess long-term mortality but equivalent survival after valve surgery compared to men.⁶³ Thus, there are important sex-related differences in the morphology, presentation, and the prognosis of MVP.

Pathology and Pathophysiology

Myxomatous valve degeneration

MVP is characterized by progressive increases in the area and length of the MV tissue, and typically progresses with a natural history spanning decades, causing leaflets to thicken anatomically and prolapse superiorly into the left atrium beyond the mitral annulus in systole, leading to MR (Figure 1C). Histologically, the mitral leaflets in MVP are characterized by myxomatous degeneration. A detailed explanation of myxomatous changes requires an understanding of the histology and the development of the normal MV.

Normal mitral valve histology and alterations in MVP

The extracellular matrix (ECM) constitutes the fibro-skeleton of a normal MV. Normal valve tissue is divided into three layers: the atrialis on the atrial side, the spongiosa, which is the middle layer, and the fibrosa on the ventricular side (Figure 3).⁶⁴ The atrialis is a dense sheet of elastic fibers and provides elasticity to the valve leaflet. The spongiosa is rich in glycosaminoglycans and proteoglycans within a fine, interweaving, spongy elastin network. It functions to resist compression between the outer layers, gives flexibility to the valve leaflet and dampens the vibrations resulting from valve closure. The fibrosa, which is the thickest part of the leaflet, is made primarily of organized collagen fibers that give the valve its tensile strength.⁶⁴

Figure 3.

Schematic of normal mitral valve histology. ECM, extracellular matrix; GAG, glycosaminoglycans; VEC, valvular endothelial cells; VIC, valvular interstitial cells. Adapted from reference 66.

Valvular leaflets are populated on the surface by endothelial cells (VECs) and by interstitial cells (VICs) within the valve (Figure 3). VICs originate from endothelial progenitor cells and are considered the principal initiators of collagen synthesis and degradation in the valve leaflets.⁶⁵ Quiescent VICs are non-contractile, alpha-smooth muscle actin-negative, fibroblast-like cells that can synthesize and degrade matrix enzymes.⁶⁵ Enzymes secreted by the VICs include matrix metalloproteinases such as collagenase (MMP-1) and gelatinases (MMP2 and MMP9), as well as tissue inhibitors of MMPs (TIMPs).⁶⁵ Quiescent VICs maintain a tight balance between degradation and synthesis of the matrix proteins, thus allowing normal valve leaflet strength and function.

Myxomatous degeneration is characterized by the expansion of the middle spongiosa layer of the valve (due to an accumulation of proteoglycans), structural alterations of collagen in all components of the leaflet, and by structurally abnormal chordae.^{3, 4} Dysregulation of ECM components plays a key role in mediating these changes. In MVP, the VICs acquire properties of activated myofibroblasts characterized by the expression of vimentin and alpha-smooth muscle actin, but not SM1 or SM2 (markers of differentiated smooth muscle cells).^{3, 66} Activated myofibroblasts are responsible for increased concentrations of various proteolytic enzymes, including matrix metalloproteinases, which degrade collagen and elastin at a rate exceeding the rate of production seen in quiescent VICs.⁶⁷ In addition, cells staining for the pan-hematopoietic marker CD45+ are also be found in myxomatous valve tissue,⁶⁶ and may represent fibrocytes capable of differentiating into myofibroblasts that can both secrete matrix and degrade collagen and elastin.

The chordae tendineae in myxomatous valves do not appear to have increased cellularity, although they do contain increased amounts of glycosaminoglycans.⁶⁸ In a study by Grande-Allen et al, myxomatous chordae contained significantly more chondroitin/dermatan 6-sulfate and slightly more hyaluronan than control chordae. In contrast to leaflets, which contained predominantly hyaluronan, the predominant glycosaminoglycan class in chordae was chondroitin/dermatan sulfate. Myxomatous chordae contained more water and less collagen than control chordae. These findings may account for the reduced tensile strength at which the valve leaflets, and especially the chordae fail in MVP.⁶⁹ Chordal rupture is a frequent pathological findings in MVP, and may be secondary to mechanical weakening of the chordae combined with the abnormal hemodynamic stresses arising from the redundancy of the valve leaflets.^{68, 69}

Finally, degenerative processes have two main histologic phenotypes recognized in the surgical literature:⁷⁰ diffuse myxomatous degeneration (or “Barlow’s disease”), and fibroelastic deficiency (or “FED”). Barlow’s disease is characterized by thickened and diffusely redundant myxomatous leaflet tissue with disrupted collagen and elastic layers leading to prolapse of most of the mitral leaflet segments, severe mitral annular enlargement, and elongated (rarely ruptured) chordae.⁷⁰ Patients usually present young or are middle-aged at the time of surgery after a long history of murmur and/or MR. FED is characterized by decreased connective tissue, deficient in collagen, elastin and proteoglycans; thin, smooth and translucent leaflets without excess tissue and only moderate annulus dilatation; and thin, slightly elongated chordae. Patients frequently present at an older age with chordal rupture and flail leaflet after a shorter (if any) clinical history. While most of the mitral leaflet is thin, localized myxomatous degeneration and thickening occur within the flail scallop, mainly of the posterior leaflet.⁷⁰ Although Barlow’s and FED are treated with very different surgical approaches, it is unclear whether they represent two histopathological features of the same syndrome or two genetically distinct entities.

Molecular biology of valve changes in MVP

Some clues to the various signaling pathways involved in abnormal valve biology in MVP can be gleamed from our understanding of normal heart valve development. Early septation of the cardiac tube into distinct chambers is achieved through regional swellings of the extracellular matrix, known as cardiac cushions, which form the primordial valves.⁶⁶ Reciprocal signaling between the endocardial and myocardial cell layers in the cardiac cushion (mediated in part by members of the TGF-}β family) induces a transformation of the endothelial cells (VECs) into interstitial or mesenchymal cells (VICs).⁷¹ This transformation is also known as ‘endothelial to mesenchymal transition’ or EMT.⁶⁶ Sox9 is a transcription factor activated when the endothelial cells undergo mesenchymal transformation, and Sox9-deficient mesenchymal cells fail to express ErbB3, an enzyme required for the proliferation of cardiac cushion cells.⁷² The mesenchymal cells then migrate into the cardiac cushions and differentiate into the fibrous tissue of the valves.⁷²

Several genes have been shown to play pivotal roles in the formation of the heart valves: calcineurin, with the signaling and downstream activation of a family of transcription factorsnamed Nuclear factor of activated T-cells (NFAT); the absence of NFAT activation leads to fatal defects in cardiac valve formation;⁷³ Wnt/beta-catenin signaling, which determines the fate of the endocardial cells during valve development;⁷⁴ fibroblast growth factor (FGF)-4, the homeobox gene Sox4, and the downstream modulator of TGF-β superfamily signaling SMAD6,^{67, 75} the disruption of which leads to abnormally thickened, gelatinous valves. Defects in one or more of these genes and their signaling cascades may also conceivably lead to myxomatous change and mechanically weakened valves in adult life.⁶⁶ Similar to syndromic MVP (see below), TGF-β up-regulation appears to have a pivotal role among various biological pathways in the pathogenesis of primary or non-syndromic MVP. Specifically, TGF-β is known to activate VICs towards a pathologic synthetic phenotype, as shown both in animal models⁷⁶ and in human in-vitro studies. Geirsson et al.⁷⁷ demonstrated TGF-β signaling dysregulation in clinical specimens of sporadic MVP cases undergoing MV repair. TGF-β-induced extracellular matrix production in cultured valvular interstitial cells was dependent on SMAD2/3 and p38 signaling, and was inhibited by angiotensin II receptor blockers. In another study of human MVP surgical specimens by Hulin et al.,⁷⁸ up-regulation of TGF-β2 was secondary to reduced expression of metallothioneins, genes involved in the response to oxidative stress. In turn, TGF-β2 up-regulation lead to down-regulation of genes of the ADAMTS family (responsible for degradation of proteoglycans), ultimately causing excessive ECM remodeling. Finally, up-regulation of Bone Morphogenetic Protein (BMP4) has also been shown to mediate the activation of VICs from healthy quiescent cells to a pathologic synthetic phenotype in microarray data of clinical MVP specimens.⁷⁹

EMT can be induced in vitro and is increased in vivo in response to mechanical stretch, suggesting that EMT occurs not only in normal valve development, but also plays a role in adaptation to pathophysiologic conditions.⁶⁶ The ability of valves to remodel and “reset the clock” in response to reduction of mechanical stretch can be derived from the clinical context: mitral valve repair is typically very durable, suggesting that an annuloplasty ring, by reducing the mechanical load on the valves and chordae improves long-term valve function.

The mesenchymal differentiation potential of the VECs can be directed towards their transformation into osteogenic and chondrogenic phenotypes, reflecting an ability of these cells to generate VICs that reside in specific regions of the valve.⁶⁶ The multi-lineage differentiation potential of the VECs combined with a robust capacity for self-renewal strongly suggests that at least a subset of the VECs are likely progenitor cells.⁶⁶ Such progenitor cells may be essential for health and longevity of the valve, and may also become activated during the disease process. Whether these cells can be harnessed or manipulated to prevent or limit valve disease will be an exciting direction for future translational research.

Altered ECM turnover is also crucial in the pathogenesis of ruptured chordae tendineae in MVP. Although the heart is a vascular-rich organ, most of the cardiac valve complex is avascular (similar to cartilage and tendons).⁸⁰ Work by Kimura et al. showed a differential local expression of tenomoduline (a recently isolated anti-angiogenic factor) measured in the chordae tendinae.⁸¹ Specifically, tenomoduline is locally absent in the ruptured zones of the chordae, favoring abnormal vessel formation also in combination with enhanced expression of vascular endothelial growth factor-A (VEGF-A). Moreover, in contrast to what observed in normal or non-ruptured areas, higher numbers of inflammatory cells positive for CD11b, CD14 and vimentine and with an augmented expression of MMP-2 and 13 were detected in association to the downregulation of tenomoduline.⁸¹

Molecular biology of syndromic MVP

Syndromic MVP associated with connective tissue disorders has been shown to manifest similar myxomatous changes to primary or non-syndromic MVP.⁸² Marfan syndrome (MFS) is associated with mutations in the fibrillin-1(FBN1) gene (chromosome 15q15-q21).⁸³ It can also be caused by inactivating mutations of the transforming growth factor-β receptor 2 (TGFBR2) gene, located on chromosome 3p24.2-p25.⁸⁴ A recent study addressed the importance of TGF-β dysregulation in the connective tissue for the development of MVP in MFS. Ng et al⁸⁵ tested the hypothesis that the FBN1-TGF-β pathway is implicated in the pathogenesis of MVP in a murine model of MFS. Increased TGF-β signaling was observed in the prolapsing valves of FBN1-deficient mice. Treatment with a TGF-β-neutralizing antibody successfully normalized both the length and the thickness of the MV leaflets in these mutant mice, further supporting the hypothesis that the valve abnormalities in this mouse model were caused by increased TGF-β levels.⁸⁵ The responsiveness of the MV length and thickness to excess TGF-β activity has implications for the identification of novel genes that may be implicated in the pathogenesis of MV disease. Furthermore, these findings raise the possibility that myxomatous MV disease could be potentially treated with therapies that reduce TGF-β over-expression. Although treatment with TGF-β-neutralizing antibodies successfully improves many phenotypic manifestations of MFS in murine models, this therapy cannot be readily translated to the treatment of the human form of the disease in the absence of FDA-approved ‘humanized’ antibodies that block TGF-β. Because of extensive interactions between angiotensin II and TGF-β signaling pathways, mice with FBN1 mutation were treated with losartan, a selective angiotensin II type 1 receptor (AT1) antagonist, in an attempt to reduce TGF-β signaling.⁸⁶ Remarkably, the mutant mice treated with losartan had dramatic improvement in their rate of aortic root growth compared to mutant mice treated with beta-adrenergic receptor blockers in doses with similar hemodynamic effects.⁸⁶ In the recently published COMPARE trial,⁸⁷ losartan reduced aortic root dilatation rate compared to placebo in adult humans with MFS. Another trial comparing the effect of losartan versus atenolol on aortic dilatation and MVP in children and young adults with MFS is ongoing.^{88, 89}

Among other connective tissue disorders commonly associated with MV disease, the Loeys—Dietz syndrome typically involves marked arterial tortuosity and distinct craniofacial abnormalities such as hypertelorism, and cleft palate. It is caused by heterozygous mutations in either the TGFBR1 or TGFBR2 genes that encode subunits of the TGF-β receptor.³⁹ Immunostaining of diseased tissues from affected individuals with the syndrome shows evidence of increased TGF-β activity, such as increased nuclear accumulation of phosphorylated SMAD2 and increased connective tissue growth factor, which is induced by TGF-β.³⁹ In the aneurysms— osteoarthritis syndrome, characterized by aortic aneurysms, arterial tortuosity, craniofacial features (similar to the Loeys-Dietz), and early-onset osteoarthritis,⁴² inactivating heterozygous mutations in MADH3 encoding SMAD3 (positive regulator of TGF-β signaling) have been identified, once again pointing to the link between TGF-β over expression and myxomatous MV disease.

Non-myxomatous valve elongation: hypertrophic cardiomyopathy and mitral valve disease

Elongation and pathological thickening of the MV is commonly seen in HCM, a genetic disorder typically caused by mutations in sarcomere genes and characterized by unexplained thickening of the LV walls and by LV outflow tract obstruction.⁵⁵ For years it has been recognized that the simple impingement of the MV on the interventricular septum (due to high systolic velocities in the vicinity of the leaflets generated by the predominant upper septal hypertrophy) contributes to systolic anterior motion of the mitral valve or SAM (Venturi effect).⁵⁵ Whereas the Venturi effect likely contributes to the propagation and worsening of SAM, it is not adequate to initiate SAM, indicating that complementary mechanisms centered on structural MV abnormalities likely contribute to SAM.⁵⁵ Experimental in vitro studies and computational models have demonstrated that isolated anterior and internal displacement of the papillary muscles combined with leaflet elongation (especially of the posterior leaflet) is able to recreate SAM.⁵⁵ The degree of SAM is related to leaflet length, even in the absence of septal hypertrophy. Typically, basal and mid-anterior MV leaflet elongation causes SAM with prolapse, while distal anterior leaflet elongation creates SAM with a mobile flap. Leaflet elongation without papillary muscle displacement creates prolapse.⁵⁵ Several mechanisms may contribute to MV disease in hypertrophic cardiomyopathy, including the primary sarcomeric gene mutation, the response to shear stress in a turbulent outflow tract, and concomitant but unrelated familial MV disease.⁵⁵ Paracrine factors such a periostin have also been implicated in the pathogenesis of MV disease in HCM.⁹⁰ Periostin is a TGF-β1-inducible secreted protein originally identified in mouse osteoblasts.⁹¹ In the heart, periostin is physiologically expressed in embryonic cardiac valves, while it is re-expressed abundantly in adult left ventricle after pressure overload or myocardial infarction.⁹¹ Markedly elevated levels of periostin are indeed secreted by non-myocyte elements in the left ventricular walls and expressed in HCM mice.⁹⁰ Periostin promotes VIC proliferation, differentiation, and matrix production, therefore potentially driving leaflet elongation in HCM.^{55, 90}

Genetics of Non-Syndromic MVP

A familial basis for MVP has long been recognized, with an autosomal dominant mode of inheritance, a variable penetrance influenced by age and sex, and a marked heterogeneity of clinical presentation even among the affected members within a family.^{12, 13, 92}

Because MVP is found in many, but certainly not all patients with Marfan syndrome (MFS), it was suggested that primary MVP may be due to a mutation of FBN1. However, studies have failed to link non-syndromic familial MVP with variants in fibrillar or other collagen genes.^{93, 94} Negative genetic linkage results may have been related to a lack of systematic examination of the entire human genome and to phenotypic ambiguity. More recently, our understanding of the 3D shape of the MV has improved the specificity of MVP diagnosis and in turn the yield of genetic studies.²⁶ Based on this newer MVP phenotype, three loci for autosomal dominant, non-syndromic MVP have been identified on chromosomes 16, 11, and 13.^95-97 While filamin A has been identified as causing an X-linked form of MVP,¹⁴ the genes for the more common form of autosomal dominant MVP have yet to be defined (Table 1).

Table 1.

Summary of linkage studies of non-syndromic MVP

	MMVP1	MMVP2	MMVP3	XMVD
Number of probands	17	1	1	N/A
Number of pedigrees	4	1	1	1
Total number of subjects in pedigrees with echo + DNA	79	28	47	92
Number of affected individuals	25	12	9	21
Ethnicity	Ashkenazi Jewish (pedigree 1), Western France (pedigrees 2,4), Eastern France (pedigree 3)	Western European descent	Western European descent	French origin
LOD score	>5	>3	>3	>6
Chromosome	16	11	13	X
Gene map locus	16p12.1-p11.2	11p15.4	13.q31.3-q32.1	Xq28

MMVP, myxomatous mitral valve prolapse. XMVD, X-linked myxomatous valvular dystrophy. N/A, non applicable. Adapted from reference 8.

In 1999, the first genetic locus for MVP was mapped to chromosome 16p11.2-p12.1 (MMVP 1) in a family with the trait segregating in an autosomal dominant fashion.⁹⁵ Genetic linkage studies yielded maximum multipoint LOD scores of 5.4 and 5.6. This was confirmed by haplotype analysis demonstrating that a chromosomal region of about 5 centimorgans (cM) containing the locus (a genetic distance equivalent to 5 million DNA base pairs) was present in all affected individuals.⁹⁵ In 2003, Freed et al identified a second locus for mitral valve prolapse (MMVP 2) at chromosome 11p15.4.⁹⁶ The ‘risk haplotype’ comprised a 4.3-cM region on this chromosome. In 2005, a new locus for autosomal dominant MVP was mapped to chromosome 13q31.3-q32 with a multipoint LOD score of 3.17 (MMVP 3).⁹⁷ Haplotype analysis showed that a portion of the chromosome containing the locus was present in all affected members of the family. The discovery of MMVP3 not only confirmed the genetic heterogeneity of MVP, but also provided important clinical lessons. Specifically, phenotyping of the chromosome 13 pedigree revealed a spectrum of expression that included valve morphologies previously considered to be normal variants but now for the first time recognized as having the same genetic substrate in the familial context.⁹⁷ ‘Prodromal’ morphologies shared two salient features with fully diagnostic MVP: an anteriorly displaced coaptation point, and posterior leaflet asymmetry (Figure 4). The term ‘prodromal’ used in the initial literature may not be ideal outside the familial context, as its prognostic significance is unclear at this time. Therefore, these morphologies are better defined as ‘Abnormal Anterior Coaptation’ or AAC to underline their similarity with fully diagnostic MVP. Individuals with minimal systolic displacement (MSD) shared the posterior leaflet asymmetry with the individuals with full blown MVP, but their coaptation point was posterior (as in normal individuals). Individuals with AAC and MSD shared either the complete or a major portion of the at-risk haplotype with fully diagnostic MVP. This same AAC morphology was also observed in the family linked to chromosome 11.⁹⁷ Quantitatively, the height of coaptation relative to the annulus or LV diameter (P/D or C/LVID in Figure 4; see legend for abbreviation expansion) correlated well with the ratio of posterior to anterior leaflet length (r =0.83 to 0.85) in the members of the family with a genetic linkage signal on chromosome 11. Thus, the spectrum of valvular abnormalities (AAC or MSD) may represent, in the familial context, early disease expression in gene carriers, a stage of progression, or the result of disease modifying factors.

Figure 4.

Two-dimensional parasternal long axis images of posteriorly coapting leaflets (anterior leaflet [AL]; posterior leaflet [PL]) in a normal individual A) versus increased coaptation height and an elongated posterior leaflet in an individual with ‘Abnormal Anterior Coaptation’ or AAC features B) and in a patient with bileaflet mitral valve prolapse (MVP) into the left atrium (LA) C). Schematics D) showing projections of anterior (A) and posterior (P) leaflets onto the mitral annular diameter (D). C indicates the coaptation height relative to the annulus and is calculated as P/D or C/LVID. AO, aorta; LV, left ventricle; LVID, left ventricular internal diameter; and RV, right ventricle.

Recognizing early forms of MVP may be important because the condition often manifests clinically in the fifth or sixth decade of life as a severe cardiac event.^{5, 6} It is conceivable that earlier targeted intervention to reduce hemodynamic stresses on the MV leaflets in genetically susceptible individuals (as shown in a murine model of MFS with aortic dilatation and in surgical specimens of non-syndromic MVP),^{77, 85, 86} may potentially prevent progression of MVP and severe MR requiring surgery, although this premise remains to be tested.

A rare form of myxoid heart disease, the X-linked ‘myxomatous valvular dystrophy’ (XMVD), was first described over three decades ago.^{14, 98} It is characterized by multivalvular myxomatous degeneration, although the histopathological features of the MV do not differ significantly from the severe form of autosomal dominant MVP. In one large family XMVD co-segregated with hemophilia A.⁹⁸ This relatively large familial study that included over 92 individuals over five generations, with a full penetrance for men (men were either clearly affected or not), was the first investigation that mapped this rare clinical dystrophy to a sex chromosome, Xq28. Thus, the genetic linkage analysis, facilitated by the X-linked mode of inheritance and by the linkage with a mild form of hemophilia A in this pedigree, permitted a rapid mapping of this XMVD gene, with a highly informative LOD score of 6.57 (Table 1). Coupling genealogical surveys of the larger family with linkage analyses, Kyndt et al. refined the previously mapped locus on chromosome Xq28 to a 2.5-Mb region.¹⁴ Screening of candidate genes revealed a P637Q missense mutation in the Filamin-A gene in the affected members of the larger family.¹⁴ Mutational analyses of the Filamin-A gene in the other families identified 3 additional Filamin-A gene mutations: 2 more missense mutations (G288R, V711D) and a 1944 bp in-frame deletion.¹⁴ Male and female carriers were both affected, but the affected females had a less severe phenotype.

Filamins are large cytoplasmic proteins that play an important role in cross-linking cortical actin filaments into a dynamic three-dimensional structure, and thereby transmit extracellular signals through their interactions with the integrin receptors.⁹⁹ These proteins not only serve a structural role in cytoskeletal organization, but also appear to serve as hubs or docking platforms for second messengers important in signal transduction. The filamin group of proteins contains three members: A, B, and C. Filamins-A and B are reported as ubiquitously expressed in tissues whereas filamin-C expression is restricted to the cardiac and the skeletal muscle.⁹⁹ The filamins are present as homo or heterodimeric Y-shaped proteins with each chain consisting of an actin-binding region at the amino terminus. The core of the protein consists of 24 highly homologous Ig-like repeats followed by a carboxyl integrin binding domain.⁹⁹ Gene knockout studies have indicated the importance of these proteins in diverse developmental processes, and filamin-A appears to be the major family member responsible for cardiac and vascular development.¹⁰⁰ Hemizygous mice for the filamin-A null allele show embryonic lethality and a wide range of cardiovascular malformations including: incomplete septation of the outflow tract that leads to a common arterial trunk and double outlet, abnormally thickened and malformed outflow tract valves, atrial and ventricular septal defects, type B interruption of the aortic arch, abnormal endothelial organization in blood vessels, abnormal vascular permeability, and thickening of the MV.¹⁰⁰ Compensation of the other filamin genes does not seem to occur nor have cardiovascular defects been described in mouse mutants for either the filamin-B or the filamin-C genes.^{101, 102} Filamin-A appears as a functional hub in many signaling pathway that may contribute to the development of valvular disease. Filamin A coordinates localization and activation of TGF-β receptor—activated SMADs, particularly SMAD2, to act as a positive regulator of TGF-β signaling.¹⁰³ One potential mechanism underlying cardiac valvular dystrophy could involve increased TGF-β signaling secondary to perturbed filamin A-SMAD interactions with consequent dysregulation of EMT.¹⁰⁴ Filamin A mutations may provide a link between MFS and non-syndromic MVP, conditions that are both characterized by increased TGF-β signaling. Specifically, myxomatous mitral valves found in fibrillin-1— deficient mice (which model MFS) display excessive TGF-β activation and up-regulated expression of Filamin-A.^{85, 105}

Remaining Questions

MVP Pathophysiology and Genetics

In this review, we present a unifying theory for the pathogenesis of MVP based on our knowledge of the biology of valves and the dynamic interplay of differentiating cells and growth factors. Yet, a full understanding of the molecular basis of MVP requires two additional steps (Table 2): 1) identification of the genetic variants responsible for non-syndromic autosomal dominant MVP, including the role of both susceptibility and modifier genes; 2) functional studies with animal models to corroborate the clinical relevance of identified mutations.

Table 2.

Summary of Future Research Directions

MVP Pathophysiology & Genetics
Identify the genetic variants responsible for non-syndromic autosomal dominant MVP
Develop functional studies with animal models to corroborate the clinical relevance of identified genetic variants
Develop new non-surgical therapies based on a better understanding of genetic determinants of MVP and related biochemical pathways
MVP Epidemiology
Better characterize the natural history of non-diagnostic MVP morphologies
Identify genetic and environmental risk factors responsible for accelerated MVP progression
Examine if specific risk factors act differently at different stages of MVP progression
Evaluate the duration of the different stages of MVP progression
Assess if response to future non-surgical therapies (? Losartan) varies based on different MVP stages (non-diagnostic MVP morphologies versus full-blown MVP)

MVP = mitral valve prolapse.

MVP appears to be the result of multiple genetic pathways, as illustrated by the identification of several genes in syndromic MVP^{83, 84} and three loci for non-syndromic MVP.^95-97 The identification of Filamin-A mutations in an X-linked form of valvular dystrophy¹⁴ highlights the importance of the cytoskeleton not only in providing structural integrity but also in critical cellular signaling pathways, specifically the TGF-β pathway. Advances in DNA sequencing technologies may lead to the identification of the MMVP1, MMVP2 and MMVP3 genes in the near future. Large-scale collections of MVP patients and genome-wide association studies will allow identification of additional MVP genes, and finally elucidate the pathways leading to the occurrence of MVP. Identification of the genes involved in the development of MVP is important because the disease typically manifests later in life, and earlier intervention in susceptible individuals may potentially prevent progression to a clinically severe stage, a premise that remains to be tested. In-vitro studies of surgical specimens have shown for the first time that the myxomatous changes characteristic of MVP are pharmacologically preventable,⁷⁷ which offers great hope for the development of therapies based on future genetic discoveries.

Mouse models have proven essential for demonstrating the importance of filamin A mutations in cardiac valve development.¹⁰⁰ Further opportunities for understanding human MV disease may derive from a study of the King Charles spaniel, a species in which MVP naturally occurs similarly to humans with a frequency of 1-5%.¹⁰⁶ In addition, the zebrafish has recently been identified as an ideal model for genetic knockdown experiments and understanding of valve development.¹⁰⁷ While zebrafish display differences in atrio-ventricular canal and valve morphology compared to other model organisms (such as mice and chicken), they share the same molecular and cellular mechanisms for valve development with humans.¹⁰⁷ Thus, several signaling pathways implicated in MVP also contribute to atrioventricular canal formation in zebrafish (e.g., TGF-β/SMAD, Notch1b, and VEGF/Calcineurin/NFAT signaling pathways).¹⁰⁷

Identification of mutations responsible for autosomal dominant MVP and their corroboration through functional studies will provide a potential screening tool to help clinicians identify asymptomatic patients who may progress to significant MR, perhaps among the diagnostic MVPs without MR or among AAC individuals who do not meet standard diagnostic criteria but share features of excessive leaflet motion with the fully affected family members (Figure 5).⁹⁷ Advances in the understanding of MVP pathophysiology outlined in this review may lead to new medical therapies aimed at preventing progression of disease by targeting different cells and signaling pathways (Figure 5): from excessive TGF-β signaling (similarly to what has been shown in the Marfan syndrome and in cultured human MV cells of non-syndromic MVP, in which angiotensin I receptor blockade leads to regulation of TGF-β and limitation of clinical disease progression), to manipulation of progenitor cells (among the endothelial cells responsible for both valvular embryogenesis and response to hemodynamic stress in adult life), to the development of tissue-engineered heart valves. Such medical interventions may hold the potential for limiting disease progression at an early stage in non-diagnostic MVP phenotypes, or avoid surgical treatment and clinical complications in fully diagnostic MVP individuals (Figure 5).

Figure 5.

Proposed temporal spectrum of mitral valve prolapse (MVP) progression with potential interventions/non-surgical therapies depending on progression stage (bottom section). MV, mitral valve; MSD, minimal systolic displacement; A, P, projections of anterior and posterior leaflets onto the mitral annulus; C; mitral leaflet coaptation height; MR, mitral regurgitation; TGF, transforming growth factor.

Progression of MVP

Although the natural history of MVP has been studied since the 1980s, these investigations have focused on individuals with fully diagnostic MVP and its clinical outcomes, including cardiac death, heart failure, endocarditis or severe MR requiring surgery.^{5, 6} The natural history of early echocardiographic stages of MVP leading to diagnostic MV leaflet displacement and to subsequent clinical outcomes have yet to be described in both tertiary-care and community-based studies (Table 2). In the familial context, previously non-diagnostic morphologies that share features of excessive leaflet motion with fully diagnostic MVP have been shown to represent mild or early stages of phenotypic expression in gene carriers.⁹⁷ The existence of early MVP morphologies in the general population would raise the possibility of echocardiographic screening and provide an opportunity for potential intervention at an early phase of disease. As suggested in Figure 5, progression of MVP may occur in stages over a lifetime beginning from a genetic substrate leading to mild, non-diagnostic valve morphologies, developing into full expression of the MVP phenotype, and culminating in severe MR requiring valve surgery. The duration of the individual stages of the disease can range from months to years, the shorter progression duration being a consequence of the development of a flail mitral leaflet. Various risk factors including genetic modifiers, environmental factors (smoking, diet, body mass index, hypertension), and non-modifiable characteristics such as race or gender may all influence the progression from one stage of MVP to the other and the duration of the different stages.

Several questions regarding the epidemiology of MVP still remain unanswered (Table 2). Specifically, it is unknown: if early, non-diagnostic MVP morphologies progress within and/or outside the familial context; if some risk factors more than others contribute to progression and at which stages; and finally, if response to future non-surgical therapies varies as MVP progresses from early disease to significant MR. Further longitudinal studies with a focus on specific clinical, demographic, and ethnic subgroups are needed to answer these important questions.

Conclusions

MVP is a common clinical phenotype and remains the most common valvular pathology requiring surgery. Multiple loci for autosomal dominant non-syndromic MVP and a gene responsible for a rare, X-linked form of MVP have been discovered. Studies in a mouse Marfan model and in clinical specimens of excised myxomatous mitral valves have underlined the role of excessive TGF-β signaling in the development of degenerative MV disease and the potential of angiotensin I receptor blockade in limiting MVP progression. These discoveries overall have exponentially accelerated our understanding of MVP. However, many questions remain unanswered in relation to both MVP pathophysiology and epidemiology, and future studies are needed to address these important issues.