Few data are available regarding the prevalence and causes of false-negative auscultation (mis-auscultation) of aortic (AR), mitral (MR), or tricuspid regurgitation (TR), and there are no such data that are relevant when the patient's pretest probability of having regurgitation is unknown. The authors therefore studied 294 patients examined by pulsed Doppler echocardiography. On 755 examinations (2.57 examinations per patient), Doppler velocity patterns typical of AR, MR, or TR were found in 63, 96, and 49 patients, respectively. For all three murmurs, mis-auscultation was the rule, rather than the exception, with sensitivities of auscultation ranging from 0 to 37%, depending (but weakly) on the site of the murmur and the years of training of the observer. Specificity of auscultation was high (85% to 100%). The factors associated with the mis-auscultation of AR were poor image quality in the echocardiograms, absence of cardiomegaly, and less experience of the examiner. The probability of missing MR increased in the presence of coronary artery disease (CAD) or if the examiner had less experience. The likelihood of missing TR by auscultation was increased by CAD, obesity, chronic obstructive pulmonary disease, or the absence of cardiomegaly. This study suggests that there is a high prevalence of "silent" murmurs, and that not hearing a regurgitant murmur does not suffice to rule out the presence of regurgitation.