Effect of obesity on electrocardiographic left ventricular hypertrophy in hypertensive patients : the losartan intervention for endpoint (LIFE) reduction in hypertension study.

Obesity may limit sensitivity of ECG voltage criteria for left ventricular hypertrophy (LVH) because of the attenuating effects of increased body mass on precordial voltages. However, obesity is associated with an increased prevalence of anatomic LVH, making more accurate ECG criteria in obese patients a clinical priority. ECG LVH by Cornell voltage-duration product and/or Sokolow-Lyon voltage criteria was used to select patients for the Losartan Intervention For Endpoint (LIFE) Reduction in Hypertension Study. Clinical and ECG data were available in 8417 patients (54% women; mean age, 67+/-7 years); 2519 were overweight and 1573 were obese by gender-specific body mass index criteria. Increased body mass index had significant but directionally opposite effects on ECG LVH by these 2 criteria. Compared with normal-weight patients, obese and overweight patients had lower Sokolow-Lyon voltage and a lower prevalence of ECG LVH by Sokolow-Lyon criteria (10.9% versus 16.2% versus 31.4%; P<0.001). In contrast, obese and overweight patients had higher mean values of the Cornell product and higher prevalences of ECG LVH by this criterion (75.1% versus 69.9% versus 60.7%; P<0. 001). After adjustment for age, gender, race, myocardial infarction, and diastolic and pulse pressure with the use of logistic regression analysis, increased body mass remained highly predictive of the presence of ECG LVH. Compared with normal-weight patients, obese patients had a >2-fold higher risk of ECG LVH by the Cornell product but a 4-fold lower risk of ECG LVH by Sokolow-Lyon voltage; overweight status was associated with intermediate risks, with a 151% greater likelihood of ECG LVH by the Cornell product but only 44% of the risk of LVH by Sokolow-Lyon voltage criteria compared with normal-weight individuals. Thus, Sokolow-Lyon voltage criteria underestimate the prevalence of anatomic LVH in the presence of obesity, whereas Cornell product criteria for ECG LVH appear to provide a more accurate measure of LVH in obese and overweight patients.