CAROTID-TO-FEMORAL PULSE TRANSIT TIME VARIABILITY PREDICTS ALL-CAUSE AND CARDIOVASCULAR MORTALITY IN CHINESE AND SWEDISH ELDERLY

Abstract

Listen

Translate article

Objective: Acquisition of pulses at carotid and femoral arteries allows measurement of carotid-femoral pulse wave velocity (cf-PWV) via determination of pulse transit time (PTT) and pulse travelled distance. PWV predicts adverse outcome and contributes to risk stratification. However, little is known about the PTT variability during the cf-PWV measurement and its predictive value. Design and method: Community-based prospective cohorts of elderly population from Shanghai, China and Malmö, Sweden were included in the study. The cf-PWV, cf-PTT and its coefficient of variation (CV) were measured with the Sphygmocor device. Vital status in both cohorts and cause of death in the Malmö subjects were ascertained. Cox proportional hazard model and Fine-Gray competing risk model were applied for analysis. Results: Of the 1249 (median age, 75.6 years; 42.5% men) and 2999 (72.4 years; 39.6% men) participants from Shanghai and Malmö who were followed up for a median of 8.9 and 6.6 years, respectively, 292 (23.3%) and 388 (12.9%) died, respectively. The cf-PWV (median: 9.4 vs. 8.7 m/s in Shanghai and 11.2 vs. 9.9 m/s in Malmö) and CV of cf-PTT (median: 8.6 vs. 7.4 % in Shanghai and 9.0 vs. 7.9 % in Malmö) were significantly higher in the deceased people compared to those survived. Age, body mass index, heart rate, mean blood pressure and pulse pressure were determinants (P < 0.05) of the CV of PTT in both cohorts. Each 1-standard deviation increase in the log-transformed CV of cf-PTT was associated with a 14% (95% CI, 1–28%) increased risk of all-cause mortality in Shanghai subjects and 25% (13–37%) and 59% (34–89%) increased risk of all-cause and cardiovascular mortality in Malmö subjects. Adding the CV of cf-PTT to the models including cf-PWV had significant net reclassification improvement (0.202 [0.062–0.321]; P = 0.007) and integrated discrimination improvement (0.013[0.001–0.039]; P = 0.02) for cardiovascular mortality in Malmö subjects. Conclusions: The variability of cf-PTT during the cf-PWV measurement was predictive of mortality and improved risk prediction in the elderly.