Feasibility and reproducibility of speckle tracking echocardiography in routine assessment of the fetal heart in a low-risk population

Abstract

Background Assessment of fetal cardiac function can now be undertaken with speckle tracking echocardiography (STE) in a variety of settings. The objective of this study was to evaluate the feasibility and reproducibility of STE in routine fetal cardiac assessments by non-experts in an NHS clinical setting.Methods A prospective longitudinal cohort study was performed from April 2022 to June 2023. For all women attending for ultrasound scan between 18 and 40 weeks’ gestation a three second cineloop of the fetal four chamber view stored if technically possible. Parameters of global cardiac function (ejection fraction [EF], cardiac output [CO], cardiac output per kg [CO/kg], stroke volume [SV] and stroke volume/kg [SV/kg], and fractional area change of the ventricles [FAC]) were analysed off-line using STE software by two researchers independently.Results Of 393 ultrasound scans performed in fetuses fulfilling criteria, 50.9% were able to be analysed using STE. Intraobserver variability was poor to moderate for both operators (Operator 1 correlation coefficient for FAC 0.28, EF 0.17, SV 0.01, SV/Kg 0.00, CO 0.74, CO/Kg 0.39; Operator 2 correlation coefficient for FAC 0.58, EF 0.65, SV 0.67, SV/Kg 0.02, CO 0.66, CO/Kg 0.31). Interobserver variability was also poor to moderate (Operator 1 vs Operator 2 correlation coefficient for FAC 0.01, EF 0.00, SV 0.00, SV/Kg 0.00, CO 0.64, CO/Kg 0.25; Operator 2 vs Operator 1 correlation coefficient for FAC 0.16, EF 0.06, SV 0.72, SV/Kg 0.008, CO 0.70, CO/Kg 0.34).Conclusion Routine speckle tracking echocardiography of global cardiac function in low-risk fetuses has low to moderate feasibility and reproducibility in a real-world NHS setting. A high grade of expertise is likely necessary to use STE in order to achieve high reproducibility. Future research should investigate factors contributing to variability in STE measurements and standardisation of protocols.