Figure 2.

Risk of myocardial infarction (Panel A), stroke (Panel B), cardiovascular death (Panel C), the composite end point (Panel D), and all‐cause mortality (Panel E) in relation to the on‐treatment presence or absence of ECG left ventricular hypertrophy by both Cornell product (CP) and Sokolow‐Lyon voltage (SL) treated as time‐varying covariates in Cox analyses. *Adjusted for randomized treatment, age, sex, prevalent diabetes mellitus, history of stroke, myocardial infarction, ischemic heart disease, heart failure, peripheral vascular disease or prior antihypertensive treatment, baseline serum cholesterol, high‐density lipoprotein cholesterol, glucose and creatinine, and urine albumin/creatinine ratio treated as standard covariates, and on‐treatment heart rate and diastolic and systolic blood pressure treated as time‐varying covariates. CI indicates confidence interval.