Article Text

Download PDFPDF
Mitral valve prolapse and glaucoma: a ‘floppy’ perception?
  1. Francesca N Delling1,2,
  2. Ramachandran S Vasan1,3
  1. 1Framingham Heart Study, Boston, Massachusetts, USA
  2. 2Department of Medicine (Division of Cardiology), Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA
  3. 3Department of Medicine (Sections of Cardiology and Preventive Medicine), Boston University School of Medicine, Boston, Massachusetts, USA
  1. Correspondence to Dr Francesca N Delling, Department of Medicine (Division of Cardiology), Beth Israel Deaconess Medical Center, 330 Brookline Avenue, E/SH-458, Boston 02215, Massachusetts, USA; fdelling{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Mitral valve prolapse (MVP) is defined as the systolic displacement of one or both mitral leaflets into the left atrium. Glaucoma is a neurodegenerative disease characterised by elevated intraocular pressure (IOP), with typical slow, progressive degeneration of retinal ganglion cells and optic-nerve cupping. In primary open-angle glaucoma (OAG), the predominant form of glaucoma in Western countries, the iridocorneal angle is open and normal in appearance but aqueous humour outflow is diminished. How can an association between MVP and OAG even be postulated? Indeed, the two conditions share multiple common features, from a similar prevalence in the general population (approximately 3%)1 ,2 to the potential for significant adverse clinical outcomes (heart failure, need for surgery, sudden cardiac death in MVP and blindness in OAG).1 ,2 Most importantly, myxomatous degeneration is found histologically in prolapsing mitral leaflets in MVP, and in the trabecular meshwork of the iris and the ciliary body in OAG.1 ,3

Extracellular matrix abnormalities in MVP and OAG

The extracellular matrix (ECM) constitutes the fibroskeleton of a normal mitral valve and the normal trabecular meshwork in the eye. Valvular leaflets are populated on the surface by endothelial cells and by interstitial cells (VICs) within the valve.1 Quiescent VICs are non-contractile, fibroblast-like cells that can synthesise and degrade matrix enzymes. Enzymes secreted by VICs include matrix metalloproteinases (MMPs) and tissue inhibitors of MMPs (TIMPs).1 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 characterised by the accumulation of proteoglycans and consequent expansion of the middle spongiosa layer of the valve, and structural alterations of collagen in all components of the valve leaflet. Dysregulation of ECM components plays a key role in mediating these changes. In MVP, VICs acquire properties of activated myofibroblasts responsible for increased concentrations of various proteolytic enzymes, including MMPs, which degrade collagen and elastin at a rate exceeding the rate of production seen in quiescent VICs.1

Similarly, in a murine model of glaucoma,4 MMP2 expression in the iris and the ciliary body and the activity of MMP-2 in the aqueous humour are increased, whereas TIMP1 expression is decreased, leading to abnormal ECM degradation and accumulation of glycosaminoglycans in the iridocorneal trabecular meshwork. In turn, this accumulation of proteoglycans is thought to obstruct the outflow of aqueous humour and increase IOP, leading to a greater mechanical stress on the retinal ganglion cells.

Association of MVP and OAG

Chiang et al5 report a novel association between the presence of MVP and the incidence of OAG on follow-up (∼10 years) by using longitudinal observations from a large administrative database. This retrospective cohort study demonstrates that pre-existing MVP is a significant predictor of the future development of OAG, after adjusting for multiple potential confounding factors. Among the strengths of the investigation are the large sample size, and the relatively rigorous statistical analysis aimed at optimal adjustment for multiple covariates: propensity-score matching followed by further adjustment of residual confounders (ischaemic heart disease, cerebrovascular disease, hyperlipidaemia and hypertension) using standard survival analysis. Of note, the follow-up for the control group was longer—105 months versus 58 months in the MVP group. Hence, sufficient time was provided for OAG to manifest in the matched controls. Finally, the association between MVP and OAG was independent of age, suggesting that the high risk of developing OAG among MVP patients is not a function of age itself or of age-dependent risk factors such as hypertension or ischaemic heart disease. Rather, common genetic pathways may be involved in the co-occurrence of these two conditions.

The manuscript of Chiang et al is not exempt from limitations. The authors acknowledge the lack of information about race, smoking and body mass index (BMI)—all potential risk factors for OAG. They also indicate that asymptomatic individuals with milder forms of MVP may have been excluded because of the focus on clinically symptomatic MVP patients. In addition, the authors fail to mention how MVP was defined on echocardiographic studies; the use of M-mode diagnostic criteria for MVP or 2D echocardiographic diagnosis in non-standard views may result in an overdiagnosis of MVP. Moreover, the authors do not report if any of the patients with MVP also had Marfan syndrome, a condition known to be associated with MVP and glaucoma.6 Prior studies have failed to link non-syndromic or idiopathic MVP (ie, unrelated to connective tissue disorders) with fibrillar or other collagen genes implicated in Marfan syndrome. More recent studies suggest that the link between syndromic and non-syndromic MVP may lie in transforming growth factor beta (TGF-β) overexpression.1 Fibrillin-1-deficient mice show overexpression of TGF-β and Marfan syndrome with myxomatous changes in the mitral/aortic valves.1 Interestingly, Geirsson et al7 demonstrated TGF-β signalling dysregulation in clinical specimens of non-syndromic MVP cases undergoing mitral valve repair. Whether the known association between MVP and glaucoma may be due to a common substrate of Marfan syndrome remains to be investigated. Details about the number of MVP cases with glaucoma and Marfan syndrome (not provided in the report by Chiang et al) would have helped, at least in part, answer this question, and increase our understanding of possible common pathophysiological links between the two conditions.

Genetic underpinnings of OAG and MVP

Mutations in the myocilin gene have been identified in about 4% of adults with OAG and more than 10% of cases of juvenile OAG, a rare autosomal dominant condition with onset of glaucoma between ages 3 and 40 years.2 Myocilin is an ECM protein produced in the ciliary body and trabecular meshwork of the eye that is hypothesised to maintain a normal ECM structure. Mutations in myocilin lead to breakdown of ECM structure in the trabecular meshwork leading to increased IOP and glaucoma. As described above, a similar picture of ECM degradation with accumulation of glycosaminoglycans occurs in MVP. Various loci for autosomal dominant MVP have been published1 but evidence of a specific gene is yet to be reported. TGF-β upregulation appears to have a pivotal role in various biological pathways leading to ECM accumulation in MVP, as demonstrated both in vitro and in knockout mice models.1 Among these pathways is the Wnt/β-catenin signalling, which determines the fate of endocardial cells during valve development.1 Interestingly, myocilin (implicated in OAG) is reported to be a modulator of the Wnt/β-catenin pathway.2

Whereas family studies inform us of rare mutations, genome-wide association studies (GWAS) allow us to identify common genetic variants and represent an alternative approach towards common diseases such as OAG and MVP. Various GWAS have been published for glaucoma, with the largest and most recent meta-analysis reported by Springelkamp et al.8 The authors describe 10 novel loci associated with cupping of the optic nerve, a key determinant of glaucoma. Among the susceptibility loci reported, bone morphogenetic protein 2 (BMP2) on chromosome 20 belongs to the TGF-β superfamily. Interestingly, upregulation of BMP has also been shown to mediate the activation of VICs from healthy quiescent cells to a pathological synthetic phenotype in microarray data of clinical MVP specimens.1 To date, GWAS data for MVP have been scarce. Among the loci reported are LMCD1 and TNS1, both implicated in valve development and function. No single-nucleotide polymorphisms (SNPs) associated with MVP appear to be associated with OAG, suggesting that common variants may be less likely implicated in the co-occurrence of the two conditions.

Future directions for research

Based on the findings of Chiang et al, various questions remain unanswered and represent future research directions:

  1. Given the strong association between glaucoma and MVP, should we screen MVP patients for the presence of glaucoma at a younger age than proposed by current screening guidelines? Does MVP develop before or after OAG, and in the latter case, should we screen for MVP in OAG patients, or in the ones with a family history of OAG? Although the relative risk of having OAG with MVP may seem moderately high (1.88-fold increase), the absolute risk of OAG in MVP is minimal (the incidence rate of 10.17 per 10 000 person-years). In other words, most patients with MVP will not go on to develop OAG, suggesting a minimal yield from a screening strategy.

  2. First-line medical therapy for OAG aims at reducing IOP and includes vasodilators such as prostaglandins and alpha-agonists; the latter agents are associated with systemic (albeit low) absorption. Can these drugs be considered safe in patients with OAG and MVP/severe mitral regurgitation given the significant preload dependence in MVP?

  3. In vitro studies of surgical specimens have shown for the first time that the myxomatous changes characteristic of MVP are pharmacologically preventable,7 which offers some hope for the development of future therapies. Would medical therapies (specifically angiotensin I receptor blockade/TGF-β regulation) potentially developed for treating MVP also have a role in the prevention/treatment of OAG?

In conclusion, the work of Chiang et al has demonstrated a novel and intriguing association between MVP and OAG. Multiple common pathways involving ECM dysregulation in both conditions may explain their co-occurrence. Nevertheless, these pathways are likely to account for only a modest proportion of either MVP or OAG. Additional studies are warranted to confirm the observation reported by Chiang et al and to elucidate the biological basis underlying the association.


View Abstract


  • Contributors FND and RSV contributed to this editorial.

  • Funding This work was supported by the National Heart, Lung and Blood Institute (NHLBI) research grant K23HL116652 (FND) and in part by the NHLBI contract N01-HC-25195 (RSV).

  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Linked Articles