An association between echocardiographic left ventricle longitudinal strain and hypertension in general population depending on blood pressure control

Abstract

Background . Ultrasound assessment of myocardial strain allows non-invasive identification of early stages of heart failure with preserved ventricular ejection fraction. A decline of the left ventricle (LV) global longitudinal strain in hypertension (HT) accompanied by LV hypertrophy (LVH) was reported in clinical samples with manifest HT. Data on the relationship between longitudinal strain and blood pressure (BP) and HT in general population is lacking. Objective . We studied the relationship between LV peak systolic global longitudinal strain ( GLS ) and strain rate ( GSR ) in subjects with high and controlled BP from a general population sample aged over 55 years. Design and methods . The cross-sectional study was based on a population cohort (HAPIEE, Novosibirsk). In a random sample (n = 416, aged 55–84 years old) we performed echocardiography and evaluated GLS and GSR of LV by speckle tracking technique. ANOVA multivariable models were applied. Results. In the studied sample, the prevalence of HT comprised 78,9 %. The mean GLS value was –18,7 ± 3,79 %, and it was lower in men than in women (–18,2 ± 3,85 % vs –19,2 ± 3,66 %, p = 0,005). The mean GSR value was –0,84 ± 0,17s –1 , and did not differ by sex. The absolute value of GLS in HT was lower than in normotensives: –18,5 ± 3,73 % vs –19,9 ± 3,42 %, p = 0,003; this difference was independent of age, sex and LV myocardial mass index (IMM), p = 0,011; but it was attenuated in a multivariable-adjusted model including body mass index (BMI). In HT groups, the GLS was the lowest among those “treated ineffectively” and significantly lower than in normotensives independently of age, sex and myocardium mass index (p = 0,008). The absolute value of GSR in HT was lower than in normotensives: –0,90 ± 0,17 s –1 vs –0,83 ± 0,17s –1 , p < 0,001; and persisted in multivariable models. GSR was the lowest among those “treated ineffectively” and significantly lower than in normotensives in multivariable models independently of age, sex, BMI and myocardial mass index (p = 0,017; 0,002). The average values of LV ejection fraction (Simpson) in all groups were above 50 % without significant inter-group differences (p = 0,904). conclusions. In the studied population sample, GLS and GSR of LV were associated with HT; however, the association between GLS and HT was largely explained by BMI. In hypertensives, the lowest GLS and GSR , as well as higher extent of LVH, were found among those treated ineffectively, which might point at the initial reduction of systolic ventricular function in HT with inadequate BP control.