Relation of left ventricular diastolic properties to systolic function in arterial hypertension

Background-It is unclear whether impairment of left ventricular (LV) diasto lic characteristics is independent of systolic dysfunction. Methods and Results-To address this issue, 159 consecutive hypertensive pat ients (44+/-11 years, 78 obese, 96 women) and 165 normotensive subjects (32 +/-11 years, 84 obese, 110 women) were studied with the use of Doppler echo cardiography. After adjustment for age, body mass index (BMI), and sex, we found that ejection fraction (EF; M-mode, z-derived) was higher in hyperten sive (66.6+/-5.2%) than in normotensive (63.9+/-4.4%, P<0.0001) subjects, w hereas midwall shortening (MS) was lower (hypertensive patients 16.9+/-2.0% , normotensive subjects 17.8+/-2.2%, P<0.02), even after correction for end -systolic wall stress (P<0.05). Isovolumic relaxation time (IVRT) was great er in hypertensive patients (103+/-14 ms) than in normotensive subjects (78 +/-19 ms), as was deceleration time of E velocity and peak A velocity (all P<0.0001). In multivariate analysis, IVRT was unrelated to EF, but a negati ve relation was found with MS (P<0.001), independent of age, BMI, presence of arterial hypertension, LV geometry, and load (multiple R-2=0.58). For co mparable age, sex distribution, BMI, and blood pressure values, hypertensiv e patients with lower afterload-adjusted MS exhibited longer IVRT than pati ents with normal MS (P<0.005). However, IVRT remained higher than in normot ensive control subjects after control for LV geometry and load. Conclusions-Doppler indices of delayed LV relaxation can be detected in the presence of normal or supranormal EF but are independently related to impa ired MS. A less severely abnormal relaxation, however, can be also detected in the presence of normal midwall function, independent of LV geometry and load. Thus, diastolic abnormalities may occur before systolic dysfunction even when it is measured at the midwalI.