PROGNOSTIC VALUE OF CHANGES IN ARTERIAL STIFFNESS IN ESSENTIAL HYPERTENSION

Abstract

Objective: To analyze the relationship between changes in carotid-femoral pulse wave velocity (c-f PWV) with the development of cardiovascular complications. Design and method: One thousand thirty one hypertensive patients with at least one c-f PWV determination were evaluated. C-f PWV was determined with SphygmoCor® and using the distance subtraction method. Atrial fibrillation, heart failure, stroke, ischemic heart disease, peripheral arterial disease and cardiovascular death were considered as cardiovascular complications. Ttest for independent samples was used to compare c-f PWV between groups and the delta over time between two consecutive measurements. Cox regression models adjusted by age, sex, BMI, SBP, hypertensive treatment and time lapse between the two measurements were used to analyze the associated risk of c-f PWV measurements with cardiovascular complications. Harrell's C statistic was used to evaluate the predictive value for CV complications of different c-f PWV measurements. Stata IC v14 was used for the statistical analysis. Results: Average follow-up for the total sample was 5.06 ± 2.7 years, range 0.3y-11.2y. At least one c-f PWV was available in 795 patients; two measurements in 174; three in 46, four in 13 and five in only 3. Those 174 patients with at least two PWV measurements experienced 12 cardiovascular complications during a follow-up period of 2.4 years (5 AF, 5 stroke, 1 HF and 1 CV death). There was a significant interaction for PWV values, between age and SBP. 1st and 2nd c-f PWV and the delta value were significantly higher in those with CV complications. After the adjustment by co-variables, only the 2nd measurement remained significant (10 vs 8.43 m/s, p-value 0.0078). The survival curves were significantly different among tertiles of PWV and the delta, especially for the 2nd PWV determination, which also showed the highest predictive value (Harrell's C = 0.86). According to our data the optimal cutpoint for the 2nd PWV was 9.10 m/s.Conclusions: Although the sample is too small to extrapolate the results, most patients with CV complications had an increase in PWV during follow-up higher than 1 m /s. Our data partly support the use of repeated PWV measurements.