Editorial Comment: Is MVP an MVP in Ischemic Cerebral Events?

Support for mitral valve prolapse (MVP) as a source of cardiac embolism has undergone undulating cycles of enthusiasm over the past 25 years. When the association between stroke and MVP was mooted by the pioneering work of Barnett and colleagues,¹ a new referral source opened for cardiologists. Then came the backlash. Many cases appeared to be incidental; studies did not identify an increased risk of cerebral embolism²; the condition was deemed to be a harmless variant, and common in the general population. Revised diagnostic criteria followed, and MVP was all but removed as a “most valuable player” in the team of cardiac abnormalities associated with cerebral embolism. The confounding associations did not help: MVP may be idiopathic and solitary with either familial or sporadic forms. Alternatively it may be secondary to connective tissue disorders or ischemic heart disease (a common cause in those over 50 years of age). In addition, both ventricular and atrial arrhythmias can accompany MVP; there are anatomic variants as well, with different degrees of myxomatous degeneration of the posterior or anterior leaflet, and of papillary muscle involvement and mitral regurgitation. Thus embedded in the condition are many subvariants with a range of embolic potential from benign to extremely severe.

Avierinos and colleagues³ have to some extent pulled a rabbit out of the magician’s hat. Using the Olmsted County database, they were able to overcome some methodological flaws confounding earlier studies. A community reference base was used rather than that derived from hospital or ambulatory care settings, reducing a common source of bias.

MVP is often asymptomatic, but, with an estimated prevalence of about 2%,⁴ an extraordinarily large number of patients would have to be screened in a community to recruit both symptomatic and asymptomatic cases in sufficient numbers to satisfy power considerations—clearly an impractical task. This study, therefore, follows a fairly specific group, selected for echocardiography on the basis of auscultation findings (70%), the remainder having vague cardiorespiratory symptoms or other nonspecific reasons for testing. Inclusion criteria were: no evidence of atrial fibrillation on ECG (although paroxysmal AF may thus be underdetected at entry, a point which should be considered when evaluating AF development during follow up), and no previous cardiac surgery or stroke/TIA. These patients thus represent the tip of the MVP iceberg.

The incidence of confounding risk factors for cerebrovascular events (Doppler vascular ultrasound investigation and history of hypertension, myocardial infarction, heart failure) was low in this group, which was young and followed for a mean of 5.5±3 years. Prognostic indicators for stroke or TIA were: older age (>50 years), atrial fibrillation, cardiac surgery, and mitral valve thickening. Older age is unlikely to be an independent factor: in this group, the etiology of MVP may be ischemic. The incidence of AF associated with MVP increases with age and left atrial dimension, all of which factors are themselves implicated in cardiogenic stroke.⁵ In dachshunds, MVP progressively worsens with age.⁶ Whether the same occurs in humans is unclear.

Cardiac surgery (whether valve repair or replacement is not stated) was performed in one third of patients. For these it could be argued that the prognosis is now related to the cardiac intervention and less to the pure natural history of MVP.

This study helps us identify which of the patients presenting in a clinical setting might be at risk of cerebral vascular events. In this it is valuable. However, recommendations as to treatment are complex. Any modifying effect of concomitant medications on outcome (for example, aspirin or anticoagulation) was not discussed. It would seem prudent that anticoagulation be considered in the higher risk group, notably those over 50 with atrial fibrillation and/or mitral valve thickening, and in those with a previous stroke or TIA. Indeed there is an appreciable recurrence rate once an event occurs (27% over the follow-up period). In patients under 50 without any of these accompanying factors, there may be no need for stroke prophylactic therapy given that the incidence of cerebral embolization is extremely low. In this regard, these results are similar to those from an earlier retrospective case-control report.² Clearly, further studies would be helpful to investigate the effect of intervention on prognosis.