CLINICAL ASSESSMENT OF A SYSTEM FOR GENERATING THE CENTRAL AORTIC PRESSURE WAVEFORM AND LEFT VENTRICULAR SYSTOLIC PRESSURE FROM BRACHIAL CUFF SPHYGMOMANOMETRIC PRESSURE PLUS TONOMETRIC RADIAL PRESSURE WAVE CONTOUR
M O'Rourke, X-J Jiang
Medical Professorial Unit, St. Vincent's Hospital, Sydney, Australia
Generation of the ascending aortic (AA) pressure waveform from calibrated radial (RA) tonometric waveform constitutes a potential advance for non-invasive determination of left ventricular load in humans. Small studies, showing correspondence between directly recorded AA pressure waves, and waves estimated from the invasively recorded RA pulse need to be complimented by large non-invasive studies.
Method: This study used a commercial system (Sphygmocor, PWV Medical, Sydney Australia) and comprised 15,608 individual reports from 1,604 different subjects/patients. Estimated AA waves were compared with directly recorded RA tonometric waves, calibrated by brachial cuff sphygmomanometry. Univariate analysis confirmed the relationships between AA augmentation index (AI), (amplitude from wavefoot to second systolic peak divided by amplitude to first systolic peak or shoulder) with RA AI (R=0.807), ejection duration (R=0.468), age (R0.353) and mean pressure (R-0.354), all p<0.001. AA AI was closely related to the difference between brachial and estimated AA systolic pressure ( SP) with AI = 160-2.33 ( SP), R=0.562, p<0.001. SP was zero (peripheral = AR systolic pressure) at AA AI of 160%, and RA AI of 103% (where the second systolic peak of the RA wave approximated the first). SP was weakly related to brachial systolic pressure (R=0.269). However the group absolute value of AA systolic pressure (Y) was closely related to brachial systolic pressure (X) with Y=0.93X-2.7 (R = 0.964). Thus, while for a large group the relationship between recorded and estimated AA (and left ventricular) systolic pressure can be calculated from simple regression, the substantial individual difference between the two values, (and scatter in the group) is dependent on the value of wave augmentation - and hence on the shape of the RA pressure waveform.
Conclusion: For individuals, an accurate estimation of AA and left ventricular systolic pressure requires consideration of both peripheral systolic pressure and contour of the radial waveform. Difference between central and brachial systolic pressure is minimal when the second systolic peak of the radial wave approximates the first systolic peak, and increases progressively as the localised second systolic peak falls in relation to the first.
