Author + information
- 1Department of Emergency Medicine, the Second Affiliated Hospital, Xi'an Jiaotong University, Xi'an, China
- 2Shaanxi Engineering Research Center of Medical and Health BIG DATA, Xi'an Jiaotong University School of Management, Xi'an, China
- 3Department of Cardiology, the Second Affiliated Hospital, Xi'an, China
This study aims to identify the phenotypic characteristics for hypertensive patients complicated with coronary heart disease using clustering analysis as a new approach.
Clustering analysis was performed on 26 baseline clinical variables in 718 participants who underwent ambulatory blood pressure monitoring (ABMP) in this study. Circadian blood pressure (BP) patterns were divided into dipper (10% to 20% systolic blood pressure [SBP] fall), non-dipper (0% to 10% SBP fall), and reverse dipper (<10% nocturnal SBP fall), based on the nocturnal reduction of SBP. Main results were given as means (standard deviation) and numbers (percentage) for continuous and binary variables, which were analyzed with ANOVA and χ2 test, respectively. Multinomial logistic regression was also applied to analyze the associations of clinical variables with different cluster group. A calculated difference of p value<0.05 was considered statistically significant for all analyses.
Clustering analysis of clinical variables identified 3 distinct phenotypes of hypertensive patients, in which significantly different characteristics were shown in age, gender, fasting glucose, total cholesterol,low density lipoprotein (LDL), high density lipoprotein, smoking status, BP dipping status, and especially, prevalence of coronary artery disease (CAD). In order to further explore the risk factors for the higher prevalence of CAD in Cluster 2 (100%) and 3 (28.3%), multinomial logistic regression analysis was carried out and revealed that BP reverse dipping (Odds ratio [OR] 2.218; 95% confidence interval [CI] 1.290-3.814; P=0.004), non-dipping (OR 1.665; 95%CI 1.067-2.596; P=0.025), age (OR 0.986; 95%CI 0.973-1.000; P=0.045) and LDL (OR 0.762; 95%CI 0.620-0.936; P=0.010) were significantly different between cluster 3 and cluster 1. Additionally, reverse dipping was also different comparing cluster 2 with cluster 1' (OR 2.084; 95%CI 1.240-3.502; P=0.006).
Our findings extracted two subgroup from essential hypertensive patients presenting high prevalence of CAD. BP reverse dipping worth further investigation as a cardiovascular risk factor for CAD in the future.