The prevalence of mitral valve prolapse and the frequency of associated complications is currently still not known with certainty. In post-mortem studies, myxomatous changes of the mitral valve are found in less than 5%. The relationship between characteristic auscultatory findings, detectable in 6 to 18% of young asymptomatic subjects, and angiographic criteria for mitral valve prolapse, observed in up to 30% and more of those undergoing routine cardiac catheterization, is similarly unclear. Establishing the diagnosis based on M-mode echocardiographic criteria has yielded problems, in particular, a frequency too high for apparently healthy subjects. By means of two-dimensional echocardiography, displacement of a mitral leaflet could be detected more frequently in the four-chamber view than in the parasternal long-axis view, a finding which renders both the diagnostic comparability and the assumption of a planar mitral annulus questionable. Accordingly, a saddle-shaped mitral annulus has been postulated. The hypothesis of the saddle-shaped form has been repeatedly tested and confirmed: on a valve model, in patients without mitral valve disease by means of two-dimensional echocardiography as well as by means of three-dimensional reconstruction of two-dimensional echocardiographic images and, lastly, in animal experiments with surgical implantation of radioopaque markers with fluoroscopic observation. Patients can be divided into one of three categories according to the position of the leaflet with respect to the highest or lowest point of the mitral annulus: level of coaptation of the leaflets beneath, completely within or above the highest and lowest points of the annulus. A subgroup of patients in the latter category can be regarded as abnormal.(ABSTRACT TRUNCATED AT 250 WORDS)