Several sources of error in estimation of left ventricular mass with M-mode echocardiography in elderly subjects

INTRODUCTION M-mode echocardiography estimates of the left ventricular mass (LVM) were greater than magnetic resonance imaging (MRI) estimates. There are substantial differences between the methods both in the means of measuring and the calculation formula. The aim of this study was to investigate whether any difference in estimates of LVM between M-mode echocardiography and MRI is due to the means of measuring or to the calculation formula, using MRI as the gold standard. MATERIAL AND METHODS M-mode echocardiography and MRI were performed on 229 randomly selected 70-year-old community-living subjects. LVM was calculated from echocardiography (LVM(echo)) and from MRI (LVM(MRI)) measurements using standard techniques. Additionally LVM was calculated with the echocardiography formula from echo-mimicking measurements made on MR images (LVM(MRI/ASE)). RESULTS There were significant differences between all three LVM estimates in women, in men, and in the entire population. Echocardiography estimated LVM to be larger than did MRI, and the LVM(MRI/ASE) estimate was larger than the LVM(MRI). The difference between LVM(MRI) and LVM(MRI/ASE) was larger than the difference between LVM(echo) and LVM(MRI/ASE). There was a low correlation between LVM(echo) and LVM(MRI) (R(2) = 0.46) as well as between LVM(MRI/ASE) and LVM(MRI) (R(2) = 0.65). CONCLUSION The means of measuring and the calculation formula both independently add to the error in LVM estimation with M-mode echocardiography. The error of the calculation formula seems to be greater than the error of the means of measuring in a population of community-living elderly men and women.