Mid- and long-term reproducibility of noninvasive measurements of spontaneous arterial baroreflex sensitivity in healthy volunteers

Baroreflex sensitivity (BRS) is altered in a variety of circumstances and could be considered as a marker for the prognosis of some cardiovascular diseases. The present study was designed to evaluate the reproducibility of noninvasive measures of BRS, both at mid- and at long-term. Fourteen healthy volunteers were examined on three occasions (first interval = 1 week, second interval = 1 year). Each recording was performed using a noninvasive photoplethysmographic device (Finapres 2300, Ohmeda), both in supine and standing positions. Two different methods of measurement were used: the sequences method and cross spectral analysis. The reproducibility of BRS measures was as satisfactory at mid- as at long-term for the sequences method (intraclass coefficient [ICC] = 0.87 and 0.86, respectively), but it was better at mid- than at long-term for the cross-spectral analysis (ICC = 0.85 and 0.54, respectively). The measures performed in standing position were obviously more reproducible than those made in recumbent position (ICC = 0.87 and 0.70 for the sequences method, 0.85 and 0.71 for the cross-spectral analysis, respectively). Due to the high reproducibility of these noninvasive measures, the number of patients to be included in a pharmacological study was calculated as rather small: for example, only 20 patients are required for detecting a change in upright BRS of 3 msec/mm Hg, at long-term (sequences method). Likewise, the magnitude of the regression to the mean, which has to be expected in patients selected for a follow-up study, turned out to be low: for example, <15% of the difference between the patient group mean value and the reference value, both at mid- and at long-term (standing position, sequences method). We conclude that: 1) The noninvasive measures of BRS in standing position are reproducible enough to allow longitudinal studies to be conducted over either a short or a long period; 2) The long-term reliability of the sequences method seems to be higher than that of the cross-spectral analysis; and 3) Subtle changes in SBR may be noninvasively detected within small patient groups. (C) 1997 American Journal of Hypertension, Ltd.