Left ventricular filling patterns in patients with systemic hypertension and left ventricular hypertrophy (the LIFE study)
Diabetes Mellitus, Type 2
Hypertrophy, Left Ventricular
Abnormal left ventricular (LV) filling may exist in early stages of hypertension. Whether this finding is related to LV hypertrophy is currently controversial. This study was undertaken to assess relations between abnormal diastolic LV filling and LV geometry in a large series of hypertensive patients with electrocardiographic LV hypertrophy. M-mode, 2-dimensional, and pulsed Doppler echocardiographic recordings of mitral inflow velocity and isovolumetric relaxation time (IVRT) were obtained in 750 patients with stage I to III hypertension and LV hypertrophy determined by electrocardiography (sex-adjusted Cornell voltage duration criteria or Sokolow-Lyon voltage criteria) after 14 days of placebo treatment. The patients' mean age was 67+/-7 years and 44% were women. One hundred forty patients (19%) had normal LV geometric pattern, 79 (11%) had concentric remodeling, 342 (45%) had eccentric LV hypertrophy, and 189 (25%) had concentric LV hypertrophy. A normal LV filling pattern was found in 116 patients (16%), abnormal relaxation in 519 (69%), "pseudonormal" filling was found in 83 (11%), and a restrictive filling pattern in 32 (4%). Prolonged IVRT was associated with LV hypertrophy (p<0.01) as well as elevated relative wall thickness (p<0.05). A stronger difference (p<0.01) in IVRT was found between groups with and without LV hypertrophy. Multiple regression analysis revealed that increased LV mass correlated with prolonged IVRT, whereas LV mass and geometry were not associated with peak early LV filling velocity (E), peak atrial filling velocity (A) ratio or mitral valve E-peak deceleration time, although IVRT was found to be an independent correlate of E/A ratio and deceleration time. Thus, abnormal IVRT was highly prevalent in all LV geometric subgroups among hypertensive patients with electrocardiographic LV hypertrophy, even in those with normal LV geometry determined by echocardiography. We found that IVRT differed significantly among patient groups with different LV geometric patterns, primarily because of the association of IVRT to LV mass.