Blood Pressure, Hypertension and The Risk of Aortic Dissection Incidence and Mortality: results from the Japan-Specific Health Checkups Study, the UK Biobank Study and a Meta-analysis of Cohort Studies

Circulation. 2021 Nov 8. doi: 10.1161/CIRCULATIONAHA.121.056546. Online ahead of print.

Abstract

Background: Hypertension or elevated blood pressure (BP) is an important risk factor for aortic dissection (AD); however, few prospective studies concerning this topic have been published. We investigated the association between hypertension/elevated BP and AD in two cohorts and conducted a meta-analysis of published prospective studies, including these two studies. Methods: We analyzed data from the Japan Specific Health Checkups (J-SHC) Study and UK Biobank, which prospectively followed 534,378 and 502,424 participants, respectively. Multivariable Cox regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the association of hypertension/elevated BP with AD incidence in the UK Biobank and AD mortality in the J-SHC Study. In the meta-analysis, summary relative risks (RRs) were calculated using random effects models. A potential nonlinear dose-response relationship between BP and AD was tested using fractional polynomial models, and the best-fitting second-order fractional polynomial regression model was determined. Results: In the J-SHC Study and UK Biobank, there were 84 and 182 ADs during 4- and 9-year follow-up, and the adjusted HRs of AD were 3.57 (95% CI, 2.17-6.11) and 2.68 (95% CI: 1.78-4.04) in hypertensive individuals, 1.33 (95% CI: 1.05-1.68) and 1.27 (95% CI: 1.11-1.48) per 20-mmHg increase in systolic BP (SBP), and 1.67 (95% CI: 1.40-2.00) and 1.66 (95% CI: 1.46-1.89) per 10-mmHg increase in diastolic BP (DBP), respectively. In the meta-analysis, the summary RRs were 3.07 (95% CI 2.15-4.38, I2=76.7%, n=7 studies, 2,818 ADs, 4,563,501 participants) for hypertension and 1.39 (95% CI: 1.16-1.66, I2=47.7%, n=3) and 1.79 (95% CI: 1.51-2.12, I2=57.0%, n=3) per 20-mmHg increase in SBP and per 10-mmHg in DBP, respectively. The AD risk showed a strong, positive dose-response relationship with SBP and even more so with DBP. The risk of AD in the nonlinear dose-response analysis was significant at SBP >132 mmHg and DBP >75 mmHg. Conclusions: Hypertension and elevated SBP and DBP are associated with a high risk of AD. The risk of AD was positively dose-dependent, even within the normal BP range. These findings provide further evidence for the optimization of BP to prevent AD.