Value of estimated pulse wave velocity to identify left ventricular hypertrophy prevalence: insights from a general population

Abstract

Background: Aortic stiffness shares a similar profile of risk factors with left ventricular hypertrophy (LVH) and can also lead to LVH by itself. Published data have demonstrated the correlation between aortic stiffness and LVH. Recent data have revealed estimated pulse wave velocity (ePWV) to be a simple and cost-effective marker of the severity of aortic stiffness. Our analysis aimed to explore the association between ePWV and LVH prevalence, and to investigate the incremental value of ePWV for the identification of LVH prevalence.

Methods: The present analysis based on a cross-sectional survey which included 11,597 participants from rural areas of southeastern China between Sep 2020 and Feb 2021. ePWV was formulated based on mean blood pressure and age according to a published algorithm.

Results: The prevalence of LVH was 14.56%. With the adjustment of age, sex, education, income and physical activity level, current drinking and smoking status, BMI, waist circumference, serum creatinine, total cholesterol, high density cholesterol, mean blood pressure, fasting plasma glucose, anti-hypertensive therapy, anti-diabetic therapy, lipid-lowering therapy, and cardiovascular disease history, every standard deviation increment of ePWV associated with a 2.993 times risk of LVH prevalence. When dividing ePWV into quartiles, the top quartile had a 4.520 times risk of LVH prevalence when compared with the bottom quartile. Furthermore, smooth spline analysis displayed that the association was linear in the whole range of ePWV (p for non-linearity = 0.073). Additionally, subgroup analysis revealed the association was robust to sex, obesity and diabetes, and younger people and hypertensive population were more vulnerable to the increase of ePWV than their corresponding counterparts. Finally, ROC analysis showed a significant advancement when introducing ePWV into established risk factors (0.787 vs. 0.810, p for comparison < 0.001), and reclassification analysis also confirmed significant improvement from ePWV to identify LVH prevalence (category-free net reclassification analysis = 0.421, p < 0.001; integrated discrimination index = 0.023, p < 0.001).

Conclusion: Our analysis demonstrated a linear association between ePWV and LVH prevalence. Furthermore, our results suggest younger people and hypertensive population are more likely to have LVH prevalence with the increase of ePWV. More importantly, our findings implicate the incremental value of ePWV to optimize the identification of LVH prevalence in a general Chinese population.

Keywords: Aortic stiffness; Estimated pulse wave velocity; Left ventricular hypertrophy.

Fig. 1

Flow chart of the enrolling process

Fig. 2

Smooth spline analysis of the association between ePWV and the risk of the presence of LVH. Smooth spline analysis was conducted through generalized addictive model with the adjustment of Clinical risk factors: age, sex, education level, income level, physical activity, current smoking, current drinking, BMI, WC, Scr, TC, HDL-c, FPG, MBP, anti-hypertensive therapy, anti-diabetic therapy, lipid-lowering therapy, and CVD history. In the plot, the risk of LVH prevalence increased proportionally with the increment of ePWV, and P for non-linearity was insignificant, suggesting the association between ePWV and LVH prevalence was linear in the whole range of ePWV

Fig. 3

Subgroup analysis of the association between ePWV and LVH prevalence. The model in each stratum was adjusted for age, sex, education level, income level, physical activity, current smoking, current drinking, BMI, WC, Scr, TC, HDL-c, FPG, MBP, anti-hypertensive therapy, anti-diabetic therapy, lipid-lowering therapy, and CVD history except for the variate that was used to define subgroups (in HTN subgroups, MBP and anti-hypertensive therapy were not adjusted; in DM subgroups, FPG and anti-diabetic therapy were not adjusted). Subgroups of sex, obesity and diabetes showed insignificant interaction with the association between ePWV and LVH prevalence (p for interaction > 0.05). Significant interaction existed between age, hypertension, and the association between ePWV and the LVH prevalence. Younger people and hypertensive population were more vulnerable to the increase of ePWV than their corresponding counterparts regarding the risk of LVH prevalence