Audibility of an artificial third heart sound in relation to its frequency, amplitude, delay from the second heart sound and the experience of the observer.

The possibility of detecting synthetic third heart sounds was studied. A special unit was used that added a sound to a previously recorded phonocardiogram. The sound could be changed in frequency, amplitude and delay from the second heart sound. Four groups of observers--cardiologists, residents, nurses and students--listened to 32 random examples. Detection rate increased with experience of the observer (p less than 0.0001) as well as with amplitude (p less than 0.0001), frequency (p less than 0.0001) and delay of the sound (p less than 0.05). The sensitivity was highest among the cardiologists, but the specificity was not different between the groups. Data from this study indicate that the audibility of the third heart sound depends on several important factors. The sound should usually be audible, but variability of results were considerable even among experienced cardiologists. A quantified phonocardiographic recording should be used for validation.