Electrocardiographic left ventricular hypertrophy and arterial stiffness: the Ohasama study.

Background: Whether arterial stiffness per se contributes to left ventricular hypertrophy (LVH) independently of blood pressure (BP) remains unknown. We examined the relationship between pulse wave velocity (PWV) and LVH in a large population.

Methods: The PWV was measured between the brachial and ankle regions (baPWV) of 798 individuals. We diagnosed LVH using electrocardiographic criteria: Cornell voltage-duration product >2440 mm x msec or Sokolow-Lyon voltage >38 mm. The participants were initially separated into those with and without LVH [LVH(+) and LVH(-) groups, respectively]. To determine theoretical baPWV, we first constructed a nomogram for the LVH(-) group, calculated the PWV index (measured baPWV - theoretical baPWV) for each individual and then compared the two groups. We also examined the factors associated with LVH(+) using multivariate analyses.

Results: Linear regression analysis revealed that the theoretical baPWV (m/sec) = 0.20 x age (years) + 0.13 x Mean arterial pressure (MAP) (mm Hg) + 0.05 x Heart rate (beats/min) - 11.74 (R(2) = 0.56). The PWV index was greater in the LVH(+) than in the LVH(-) group (P = .025). The baPWV was independently related to LVH(+) along with MAP, medication for hypertension, and for diabetes; a 1 SD (4.3 m/sec) increase in baPWV was associated with a 26% increase in the risk of LVH(+) (P = .022). When LVH(+) risk factors were defined as hypertension, diabetes, and high baPWV (> or =14.6 m/sec), the prevalence of LVH(+) linearly increased with the number of concomitant LVH(+) risk factors (P < .001).

Conclusions: Arterial stiffness is independently related to electrocardiographically determined LVH in the general population.