Influence of equipment and settings on myocardial performance index repeatability and definition of settings to achieve optimal reproducibility

Abstract

To compare left myocardial performance index (MPI) values and reproducibility using different settings and ultrasound equipment in order to standardize optimal machine settings. Left MPI was prospectively evaluated by one observer performing conventional Doppler in 62 fetuses (28-36 weeks of gestational age) using different settings (changing sweep speed, gain and wall motion filter (WMF)) and two different ultrasound devices (Siemens Antares, Siemens; Voluson 730 Expert, GE Medical Systems). Intraclass coefficients of agreement (ICCs) were calculated using Bland-Altman analysis. Using baseline settings on the Siemens, mean (SD) MPI was 0.44 (0.05) with an ICC of 0.81. Decreasing the sweep speed resulted in decreasing average MPI values (0.43) and decreasing ICC (0.61). Lowering gain also influenced average MPI values (0.46) and ICC (0.76). Raising gain resulted in similar MPI values (0.45) with better ICC (0.90) compared with baseline settings. Raising wall motion filter (WMF) provided the best ICC (0.94) compared with the other settings. Changing the ultrasound equipment resulted in an ICC of 0.64. The optimal settings to achieve the highest reproducibility in measurement of MPI were sweep speed 8, gain 60 dB and WMF 281 Hz for Siemens Antares and sweep speed 5, gain -10 dB and WMF 210 Hz for Voluson 730 Expert. Changing ultrasound settings or equipment may affect the calculation and repeatability of measurement of MPI values. Strict standardization of methods decreases the variability of this parameter for fetal cardiac function assessment.