Spratt James C S, Leslie Stephen J, White Audrey, Fenn Lynn, Turnbull Colin, Northridge David B
Department of Cardiology, Western General Hospital, The University of Edinburgh, Edinburgh EH4 2XU, UK.
Int J Cardiovasc Imaging. 2004 Apr;20(2):107-11. doi: 10.1023/b:caim.0000014047.59389.1f.
To assess the effect of tissue harmonic imaging (THI) on assessment of left ventricular mass index (LVMI) measurements by M-mode trans-thoracic echocardiography, when compared with magnetic resonance imaging (MRI).
20 hypertensive male subjects were studied. LVMI was measured in all subjects by both gradient-echo MRI (Lscelsint Prestige 1.9 T) and by transthoracic echocardiography (ATL HDI 5000). M-mode echocardiography recordings were taken for each patient, two with fundamental imaging (FI) and two using THI in a randomised order and the images unlabelled. Recordings were analysed off-line, by a blinded observer. LVMI by MRI was calculated using Simpson's rule on serial short axis slices of 8 mm thickness. Data are expressed as mean +/- SD.
There was a difference in LVMI measurements between FI and THI (LVMI) (79 +/- 20 vs. 93 +/- 25 g2; p < 0.001). A lower mean difference was obtained by THI, compared to FI, when compared with MRI (2 +/- 15 vs. -32 +/- 22 g2; p < 0.001) suggesting that FI underestimates LVMI. Inter-observer variability was similar between THI and FI (4.5 +/- 15 vs. 6.4 +/- 15 g2; p = 0.46).
In hypertensive males, M-mode echo derived from FI underestimated LVMI. These results imply that widely accepted reference ranges for LVMI using FI are not applicable when THI is used.