This investigation assessed the ability of a non-invasive method to reproduce aortic root pressure waveform and pressures. An external pulse tracing of the subclavian artery was obtained simultaneously with direct aortic root pressures during routine left heart catheterization in 26 patients (aged 39-74 years) with various cardiovascular disorders. Indirect brachial arterial peak-systolic and nadir-diastolic pressures were obtained with oscillometry. The direct and indirect peak-systolic and nadir-diastolic pressures, were separately used to calibrate the pulse tracing. Adequate pulse tracing was obtained in 19 patients (73%). The waveforms agreed well with cross-correlation coefficients for systole and diastole of 0.98. The difference between the pulse trace and the direct pressure curve, when the first was calibrated with the peak-systolic and nadir-diastolic pressures of the latter on average was less than 1 mmHg for systole and diastole. At end-systole the mean difference was 5 mmHg. Oscillometric brachial arterial peak-systolic pressures were (mean +/- SD) 3 +/- 7 mmHg below the corresponding direct measurements, while diastolic pressures were 8 +/- 4 mmHg above. The difference between the pulse trace and the direct pressure curve, when the pulse trace was calibrated with oscillometric pressures, was at end-systole 6 +/- 6 mmHg and for mean arterial pressures 5 +/- 4 mmHg. Thus, the external subclavian arterial pulse tracing provides a non-invasive, clinically feasible access to the aortic root pressure waveform. With optimal calibration, good estimates of aortic root pressures throughout systole and diastole can be obtained, while end-systolic pressure tends to be slightly overestimated.