[Reproducibility of repeated measurements using M-mode echocardiography, radionuclide ventriculoradiography and...

The reproducibility of M-mode echocardiography, radionuclide ventriculography, and systolic time intervals (PEP/LVET) was studied in 16 patients with symmetrically contracting left ventricles and no signs of coronary heart disease. The values were determined four times in these 16 patients: twice each on day 1 and day 8 at an interval of 2-3 h. The mean EF and PEP/LVET values were nearly identical in all 4 repeat studies. There was a high correlation of echocardiographic and scintigraphic EF measurements (r between 0.90 and 0.96); the correlation between PEP/LVET and EF was considerably poorer (r ranging from -0.56 to -0.79). There was no difference in mean serial variabilities of EF for all repeated studies performed on the same day and on separate days for either echocardiography or radionuclide ventriculography. The mean variability of absolute EF for repeated studies was 2.9 +/- 2.4% for M-mode echocardiography and 3.4 +/- 2.4% for radionuclide ventriculography. To be attributed to nonrandom physiological alterations the absolute change in EF in an individual patient should be at least 8% for echocardiography and 9% for radionuclide ventriculography. The mean variabilities of PEP/LVET were different in studies performed on the same day and on separate days (0.03 +/- 0.02 vs 0.04 +/- 0.03, p less than 0.01). In 7 further patients the effect of dobutamine infusion was studied.(ABSTRACT TRUNCATED AT 250 WORDS)