Carotid-to-femoral pulse wave velocity (PWV) is the gold standard for the assessment of aortic stiffness. It is calculated by the ratio of pulse transit time (PTT) between two arterial sites and the distance between them. The precision of PTT estimation depends upon the algorithm that determines characteristic points at the foot of the pulse waveforms. Different algorithms yield variable PTT values thus affecting the precision of PWV and subsequently its diagnostic and prognostic accuracy. Our aim was to apply in vivo a new 'diastole-patching' algorithm and investigate whether it improves the precision of PWV measurement. Two repeated PWV measurements were performed in a general population (340 subjects) by a reference apparatus (SphygmoCor) which uses the tangential method for PTT estimation. PTT was re-estimated by the 'diastole-patching' algorithm. We computed statistical parameters of agreement, consistency, precision and variability between the two PWV measurements. The 'diastole-patching' method yielded more precise and reproducible measurements of PWV compared to the tangential method at the total population. In those cases where the reference method provided PWV measurements with difference > 1 m s(-1), the 'diastole-patching' algorithm further improved the precision of PWV. These findings may have direct implications concerning the enhancement of the diagnostic and prognostic value of PWV.