Accuracy and Variability of Transthoracic Echocardiography Measuring and Reporting Practices in a Melbourne Tertiary Hospital: A Clinical Audit

Aim We aimed to evaluate inter-observer variability in transthoracic echocardiography (TTE) measurement and reporting practices within a Melbourne tertiary hospital, to guide standardisation and improve consistency in clinical practice. Method Ten de-identified adult TTE studies (taken from February 2025–April 2025) were assessed by departmental imaging cardiologists, cardiologists-in-training, and adult cardiac sonographers on an offline workstation. Continuous variables were summarised by mean and standard deviation, and agreement with intraclass correlation coefficients (ICC). Nominal and ordinal categorical variables were analysed with response frequencies, free-marginal kappa and Kendall’s W. Participants and participant groups were compared using individual and group average Z-scores. Variability between TTE studies was described by the average coefficient of variation (CV). Results Twenty staff participated (nine imaging cardiologists, two cardiologists-in-training, and nine sonographers). Across continuous measures, average ICC was 0.968 (excellent agreement); the lowest ICCs were inferior vena cava (IVC) diameter and aortic valve pressure half-time (0.896 and 0.897, respectively). Categorical agreement was generally substantial to almost perfect, with an average free-marginal kappa of 0.726 and Kendall’s W of 0.882: the lowest values were IVC collapsibility (free-marginal kappa 0.636) and right ventricle size (Kendall’s W 0.588). Imaging cardiologists were closest to the cohort mean (group Z-score 0.01), while registrars showed the largest absolute deviation (|-0.07|). Pathology-level variability was modest (overall average CV 0.13). All reported agreements were statistically significant. Conclusion Most TTE measurements demonstrated high inter-observer agreement. Certain metrics may benefit from structured training interventions targeting standardisation of acquisition and interpretation, with potential for national benchmarking to further enhance consistency in echocardiographic reporting.