Circadian rhythm of peripheral and central arterial blood pressures in relation to allcause mortality

Objective: Brachial blood pressure (BP) presents a circadian rhythm. Abnormal rhythms are associated with adverse outcomes. However, it is not clear about the circadian rhythm of central BP and its predictive value. Methods: The study participants were patients referred to the outpatient clinic of Ruijin Hospital for 24-hour ambulatory BP monitoring from the year 2017 to 2020. The 24-hour brachial and central ambulatory BP monitoring were performed with the use of the Mobil-O-Graph (Germany) monitors. Dipping, non-dipping, reverse dipping and extreme dipping were defined as a night-to-day systolic BP ratio of 0.8–0.9, 0.9–1.0, > 1.0, and < 0.8, respectively. In cross-classification analysis of brachial and central dipping status, extreme dipping and reverse dipping were grouped with dipping and non-dipping, respectively. The vital status of patients until Dec 2020 was ascertained according to the vital statistics of the Shanghai Center for Disease Prevention and Control. Results: In 26,023 enrolled patients (men 47.4%, average age 53.0 years) followed up for a median of 1.8 years, 120 deaths occurred. The prevalence of non-dipping (46.7% vs. 43.7%, respectively) and reverse dipping (16.0% vs. 13.1%) was significantly (P < 0.001) higher for central than for brachial BP. The hazard ratios (95% confidence intervals, [CI]) for all-cause mortality were statistically significant for reverse dipping (1.95 [1.18–3.23] and 2.37 [1.43–3.94], respectively), but not other dipping status, versus dipping in brachial and central BP, respectively. In the brachial and central cross-classification analysis, the prevalence of consistent dipping, consistent non-dipping, isolated central non-dipping, and isolated brachial non-dipping was 34.7%, 54.2%, 8.6%, and 2.5%, respectively. Taking the consistent dipping as reference, the hazard ratios (95% CI) for all-cause mortality were 1.83 (0.77–4.31), 1.34 (0.46–3.95), and 1.94 (1.20–3.15) for isolated central, isolated brachial and consistent non-dipping, respectively. Furthermore, cross-classification significantly improved the risk prediction of all-cause mortality with a net reclassification improvement index (95% CI) of 0.193 (0.069–0.304). Conclusions: Non-dipping and reverse dipping were more common for central than for brachial BP, and were associated with a higher mortality risk irrespective of brachial or central BP. Cross-classification of brachial and central dipping status further improved risk stratification.