Limited contribution of left ventricular mass and remodelling to the impact of blood pressure on diastolic function in a community sample.

Abstract

Although the development of left ventricular (LV) dysfunction in hypertension has traditionally been viewed as a transition process from a phase of structural LV remodelling to dysfunction, the extent to which LV mass (LVM) and remodelling account for blood pressure (BP)-associated alterations in LV diastolic function is uncertain. In product of coefficient mediation analysis, we aimed to determine the extent to which LVM index (LVMI) or relative wall thickness (RWT) account for relations between BP and LV diastolic function. In 709 randomly selected participants from a community sample with a high prevalence of hypertension (49.6%), we determined BP and LVMI, RWT and several indices of diastolic function from transmitral blood flow and myocardial tissue Doppler (E/A, e'/a', e' and E/e') and left atrial volume using standard echocardiographic techniques. With adjustments for confounders, LVMI (P < 0.001-0.0001) and RWT (P < 0.05-0.001) were independently associated with E/A, e'/a', e' and E/e'. However, in product of coefficient mediation analysis, LVM and RWT failed to account for most BP-associated changes in diastolic function. Indeed, whilst a one SD increase in DBP or SBP (13 and 22 mmHg, respectively) translated into a 0.07, 0.13 and 0.53 decrease in E/A, e'/a', e' and a 0.73 increase in E/e', respectively, in mediation analysis LVMI accounted for only 0.0005, 0.0017, 0.05 and 0.08 of the impact of a one SD effect of LVMI on E/A, e'/a', e' and E/e', respectively. Similar contributions of RWT as for LVMI to BP-associated LV diastolic functional changes were noted and the contribution of LVMI or RWT to BP-related alterations in diastolic function was similar in those participants not receiving antihypertensive therapy. Although structural LV remodelling is independently associated with changes in LV diastolic function, LVMI and RWT account for only a minor proportion of the impact of BP on diastolic function. Thus, most BP-associated decreases in LV diastolic function are likely to be a transition process independent of LV hypertrophy or concentric remodelling.