Improved ECG models for estimation of left ventricular hypertrophy progression and regression incidence by redefinition of the criteria for a significant change in left ventricular hypertrophy status

Three sets of left ventricular hypertrophy (LVH) criteria were evaluated in terms of their utility for LVH progression and regression incidence estimation. Incidence comparisons were performed with and without taking short-term variability into consideration using electrocardiographic records taken at the baseline and the first annual follow-up examination of 11,969 men, aged 37 to 55 years, at the baseline of the Multiple Risk Factor Intervention Trial. Short-term total technical and biologic variability (95% range) was determined from the records of 104 men invited for a repeat electrocardiographic recording at 2-week or shorter intervals from the first recording. The results indicated that after elimination of the false fraction due to normal variability, the residual incidence estimates are so low that their utility and statistical power in monitoring changes in LVH status is very limited. It is concluded that a redefinition of LVH progression and regression criteria is needed to enhance their utility in clinical trials.