Impact of acquisition and interpretation on total inter-observer variability in echocardiography: results from the quality assurance program of the STAAB cohort study

We present analyses based on the echocardiography quality-assurance program of the ongoing STAAB cohort study (characteristics and course of heart failure stages A –B and determinants of progression). In 43 healthy individuals (mean age 50 ± 14 years; 18 females), duplicate echocardiography scans were acquired and mutually interpreted by one of three trained sonographers and an EACVI certified physician, respectively. Acquisition (AcV), interpretation (InV), and inter-observer variability (IOV; i.e., variability between the acquisition-interpretation sequences of two different observers), were determined for selected M-mode, B-mode, and Doppler parameters. We calculated Bland–Altman upper 95% limits of absolute differences, implying that 95% of measurement differences were smaller/equal to the given value: e.g. LV end-diastolic volume (mL): 25.0, 25.0, 27.9; septal e′ velocity (cm/s): 3.03, 1.25, 3.58. Further, 90, 85, and 80% upper limits of absolute differences were determined for the respective parameters. Both, acquisition and inter pretation, independently and sizably contributed to IOV. As such, separate assessment of AcV and InV is likely to aid in echocardiography training and quality-assurance. Our results further suggest to routinely determine IOV in clinical trials as a comprehensive measure of imaging quality. The deriv ed 95, 90, 85, and 80% upper limits of absolute differences are suggested as reproducibility targets of future studies, thus contribut...
Source: The International Journal of Cardiovascular Imaging - Category: Radiology Source Type: research