[Automatic measurement of arterial pressure].

Whenever we have a new technique for the measurement of Blood Pressure (BP), variations to standard devices should be expected. Recent guidelines for measurement and validation of automatic devices by Scientific Societies should be followed if doctors want to improve the accuracy of readings, avoiding mistakes in the diagnosis and monitoring of hypertension. To characterize the variability and reproducibility of the DINAMAP (Dynamic Automatic Mean Arterial Pressure), a random comparison with two observers (A; B) between this device and the sphygmomanometer was performed in a group of 80 patients with different pathologies. To evaluate the variability and reproducibility in the same person we analysed as well the mean variation in a setting of successive readings in 3 persons. The global correlation was good for both systolic (r = 0.9) and diastolic pressures. The behaviour of the DINAMAP for different pressure levels was obtained by the regression analyses of the pressures differences between both devices. The best correlation is found around the 140 mmHg measurements, pressures above or under this level are respectively underestimated and undervalued. We discuss the needs of a standard statistical methodology with physiological meaning for the analysis which enables to understand the multiple variations that doctors can find when performing the comparison tests.