Effect of Blood Pressure Trajectory on Cardiovascular Disease in Population with High-normal Blood Pressure: a Study Based on the Kailuan Cohort

Background Cardiovascular disease (CVD) is a leading cause of premature death and rising healthcare costs worldwide. An estimated 330 million people in China are affected by CVD, posing a substantial medical and financial burden to the nation. Among Chinese adults aged 18 and over without a prior hypertension diagnosis, 50.9% have high-normal blood pressure (BP). High-normal BP is associated with an increased risk of CVD incidence, and its effective control could prevent over 10% of CVD cases. Objective This study aimed to identify distinct BP trajectory patterns in individuals with high-normal BP using group-based trajectory modeling, investigate the impact of these trajectories on CVD risk, and provide a scientific basis for developing BP management and CVD prevention strategies for this population. Methods Participants from the Kailuan Study who underwent health examinations between 2006 and 2012 and were identified as having high-normal BP at their first examination were enrolled. Baseline data were collected via questionnaires, anthropometric measurements, and laboratory tests. Participants were followed until the first occurrence of a CVD event [myocardial infarction (MI) or stroke] or until December 31, 2022. Group-based trajectory modeling (GBTM) was used to model systolic BP (SBP) and diastolic BP (DBP) trajectories. Kaplan-Meier survival curves and Log-rank tests were used to compare cumulative CVD incidence across trajectory groups. Multivariable Cox proportional hazards regression models, or weighted Cox regression models when the proportional hazards assumption was violated, were used to analyze CVD risk across different BP trajectory groups. Multiplicative interactions between sex/age and BP trajectories were tested using Cox models, followed by stratified analyses. Sensitivity analyses were conducted to verify the robustness of the findings. Results A total of 21 745 subjects were ultimately included, with a mean age of (54.0±11.4) years; 17 556 (80.74%) were male and 4 289 (19.26%) were female. GBTM results identified four distinct SBP trajectories, each characterised by its SBP range and pattern of change over follow-up (i.e., stable, declining, or increasing). Among 7 088 subjects maintained stable SBP around 115 mmHg (1 mmHg=0.133 kPa), termed the "normal-high blood pressure decline group"; 11 662 subjects maintained stable SBP around 130 mmHg, termed the "normal-high blood pressure stable group"; 1 710 subjects exhibited a gradual increase from 129 mmHg to approximately 160 mmHg during the trajectory period, termed the "normal high blood pressure ascending cohort"; 1 285 subjects saw their SBP rise to around 158 mmHg before declining below 140 mmHg, termed the "normal high blood pressure ascending-descending cohort". DBP identified four distinct trajectories, each characterised by its DBP range and pattern of change over follow-up (i.e., stable, declining, or increasing). Among 4 856 subjects, DBP remained stable around 75 mmHg, termed the "normal-high blood pressure declining group"; 13 668 subjects exhibited stable DBP around 83 mmHg, termed the "normal-high blood pressure stable group"; 1 640 subjects exhibited a gradual increase in DBP from 82 mmHg to approximately 100 mmHg during the trajectory period, termed the "normal high blood pressure rising group"; 1 581 subjects experienced an increase in SBP to around 98 mmHg before declining below 90 mmHg, termed the "normal high blood pressure rising-falling group". Participants across different SBP trajectory groups were comparable in age, sex, BMI, educational attainment, smoking, alcohol consumption, physical activity, SBP, DBP, low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), total cholesterol (TC), fasting blood glucose (FBG), high-sensitivity C-reactive protein, estimated glomerular filtration rate (eGFR), isolated systolic hypertension (ISH), isolated diastolic hypertension (IDH), hypertension, diabetes, antihypertensive medication use, and antidiabetic medication use were compared across different DBP trajectory groups. Differences were statistically significant (P