Can ambulatory blood-pressure monitoring provide reliable indices of arterial stiffness?

Background: The use of ambulatory recordings of blood pressure (BP) was proposed to estimate arterial stiffness (AS). We compared the relative value of the ambulatory AS index (AASI), and of the slope of pulse pressure (PP) according to mean BP (MBP) obtained from 24-h ambulatory BP monitoring, to the monitoring of the arrival time of Korotkoff sounds (QKD interval) in the prediction of cardiovascular (CV) events. Methods: Twenty-four-hour ambulatory BP and QKD monitoring were recorded at baseline, before antihypertensive treatment of hypertensive patients in our Bordeaux cohort. From these recordings, the AASI, the PP/MBP slope, and the theoretical value of the QKD for a systolic pressure of 100 rnm Hg and a heart rate of 60 beats/min (QKD(100-60)) were calculated. The patients were then given antihypertensive treatment and followed by their family physicians. who were unaware of the QKD, AASI, and PP/MBP slope results. Regular updates on patients were obtained. The reproducibility of measurements was studied in 38 normal subjects evaluated on two occasions. Results: The reproducibility of the AASI and the PP/MBP slope was less than that of BP over 24 h and of QKD(100-60). The cohort comprised 469 patients. With an average follow-up of 70 +/- 39 months, 62 CV complications, including 13 deaths, were recorded. In the monovariate analysis, age, PP over 24 h, QKD(100-60), AASI, and the PP/MBP slope were significantly related to the occurrence of complications. In the multivariate analysis, when age and PP over 24 were included in the model, only QKD(100-60) remained significantly linked to CV events. Conclusions: Our data support the value of the AASI as an indirect estimate of AS and as an element in the evaluation of CV risk in hypertensive patients. However, the reproducibility of this index is less, and its predictive value for complications is poorer, than that of QKD(100-60), a parameter that we believe is more closely linked to AS.