To determine whether it is possible to assess baroreceptor sensitivity by measuring changes in blood velocity in the carotid artery and changes in the heart rate, we performed a series of 108 experiments in 19 hypertensives aged 20-57 years (mean 46.6 +/- 8.6). In each experiment, we took simultaneous measurements of carotid artery blood flow velocity (Doppler technique), the brachial intra-arterial blood pressure and the heart rate, during a rapid and transient increase in blood pressure induced by injections of phenylephrine. We then calculated the maximum slope of the regression lines correlating blood velocity with the heart period (Trieste method) and blood pressure with the heart period (Oxford method). We obtained good accuracy from the Trieste method compared with the Oxford method, as assessed by the mean of the sum of the squares (difference + 5%, NS). After the administration of 4 mg oral lacidipine to 13 essential hypertensives, aged 37-54 years (47.6 +/- 5.3), baroreflex sensitivity was not changed, as assessed by either method (Oxford method 10.1 +/- 5.5 versus 9.8 +/- 6.2 ms/mmHg; Trieste method - 0.57 +/- 0.32 versus - 0.49 +/- 0.31 ms/Hz). The coefficients of variation for the two methods, calculated for the measurements taken before and after the administration of lacidipine, were not statistically different (Oxford method 25.0 +/- 18.4 versus 36.2 +/- 16.0; Trieste method 36.7 +/- 19.2 versus 39.7 +/- 19.2). The new non-invasive Trieste method thus showed the same accuracy and precision as the invasive Oxford method in measuring baroreflex sensitivity and can be used in pharmacological studies.