Left ventricular systolic and diastolic function measurements using an ambulatory radionuclide monitor: Effects of different time averaging on accuracy

The accuracy of an ambulatory radionuclide detector (VEST) for left ventricular systolic (ejection fraction, EF) and diastolic (peak filling rate, PFR) measurements was assessed at different time averaging of the nuclear and electrocardiographic data. Fifty-one patients, in a total of 67 studies, underwent equilibrium radionuclide angiography (RNA) immediately before a VEST study. VEST data were analyzed using single-beat analysis and different time averaging of 5, 10, 15, 30 and 60 sec. Agreement between VEST and RNA in estimating EF and PFR was evaluated by computing limits of agreement (LA). These were computed as 1.96 times the s.d. of the mean differences between the two methods, expressed in the same unit as EF and PFR. Differences between the two methods were plotted against their mean, allowing investigation of any possible relationship between measurement error and the true value (whose best estimate is the mean between the two methods). The entire statistical analysis was repeated at each different time averaging. LAs for EF measurement by VEST were -10.4:8.8 (single-beat analysis), -11.2:9.9 (5-sec averaging), -5.4:4.8 (10-sec averaging), - 4.9:4.5 (15-sec averaging), -6.2:5.6 (30-sec averaging), -6.9:4.5 (60-sec averaging). Results indicate good agreement between VEST and RNA in measuring EF, at least for time averaging ≥10 sec. LAs for PFR ranged from -0.6:0.6 (single beat) to -1.0:0.6 (60-sec averaging), which was considered a clinically acceptable agreement between VEST and RNA. No relationship between measurement error and true value was found either for EF and PFR.