Influence of non-invasive measurements of arterial blood pressure in frequency and time-domain estimates of cardiac baroreflex sensitivity.

Objective To compare estimates of cardiac baroreceptor sensitivity (BRS) obtained by the Finapres device and from the direct measurement of arterial blood pressure (ABP) values from the ascending aorta, using both spectral analysis and sequence analysis. Design A cohort study of 45 coronary artery disease patients undergoing routine percutaneous coronary procedures. Methods Continuous supine recordings of resting ABP in the finger (Finapres), ascending aorta (Millar catheter-tip transducer) and electrocardiogram were obtained. Beat-to-beat values of systolic ABP (Finapres and aortic) and R-R interval were used to estimate the cardiac BRS from spontaneous sequences and by spectral analysis, using the alpha index for the low-frequency band (0.05-0.15 Hz). The influence of beta-blockers on BRS estimates was also investigated. Results No significant difference was observed between estimates of BRS derived from the Finapres (BRSFIN) and aortic ABP (BRSAO) by the spectral analysis method (Finapres bias 0.30 +/- 2.52 ms/mmHg). For sequence analysis, BRSFIN was significantly higher than BRSAO (7.80 +/- 4.52 versus 6.44 +/- 3.46 ms/mmHg), but the bias (1.36 +/- 3.10ms/mmHg) was not significantly different from spectral analysis. No significant differences in BRS were found between beta-blocker users (n=24) and non-users (n=10) for either the processing method or source of ABP recording. Conclusion Spectral analysis of cardiac BRS showed a better agreement between estimates obtained from the Finapres and aortic ABP.