In Reply: Dr Hyman raises 2 relevant questions regarding the utility of the Cornell product, which our study found was associated with the risk of new-onset atrial fibrillation in hypertensive patients with LVH. With respect to whether the correlation of Cornell product LVH with echocardiographic LVH remains intact during treatment, we have previously examined this issue in the echocardiographic subset of the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) study and found a strong relationship between changes in Cornell product and changes in left ventricular mass (LVM).1 In a subset of 584 patients with baseline and 1-year electrocardiograms and echocardiograms, regression of Cornell product LVH (defined as a ≥25% decrease from baseline) was compared with no significant regression of electrocardiographic LVH (<25% decrease from baseline) or progression of electrocardiographic LVH. Regression was associated with stepwise greater absolute decreases in echocardiographic LVM (−32 g [SD, 41 g] vs −29 g [37 g] vs −16 g [33 g], respectively; P<.001) and percentage decreases in echocardiographic LVM (−12.3% [15.6%] vs −11.3% [13.6%] vs −5.7% [14.6%], respectively; P<.001).
Okin PM, Devereux RB. Left Ventricular Hypertrophy Regression and Atrial Fibrillation Incidence—Reply. JAMA. 2007;297(1):40-41. doi:10.1001/jama.297.1.40-b