First Clinical Results of Novel Haemodynamic Simulation Software for Patient-Specific Qp:Qs Quantification in...

Background/Objectives: Accurate quantification of left-to-right shunt volume is central to clinical decision-making in patients with atrial septal defect (ASD). Conventional echocardiographic Qp:Qs estimation is widely used but limited by operator dependency, Doppler alignment sensitivity, and the quadratic amplification of diameter measurement errors in flow calculations. These factors contribute to clinically relevant variability, particularly in paediatric populations with small vessel dimensions. Simulation-based haemodynamic modelling offers an alternative approach by integrating structural and functional cardiac parameters within a patient-specific computational framework, independent of direct Doppler flow measurements. Methods: This retrospective single-centre study evaluated agreement between conventional Doppler-derived and software-based Qp:Qs quantification using a three-dimensional haemodynamic simulation model. Transthoracic echocardiographic datasets from patients with isolated secundum ASD undergoing defect closure between 2018 and 2024 were analysed. Conventional Qp:Qs was calculated using pulmonary and aortic valve diameter and velocity-time integral measurements, while software-derived Qp:Qs was computed from patient-specific haemodynamic modelling based on cardiac chamber geometry and physiological parameters. Mean values were compared using Student's t-test. Agreement was assessed using Bland-Altman analysis. A predefined ± 20% deviation threshold was considered clinically acceptable according to the study protocol. Results: A total of 98 echocardiographic examinations from 94 patients were included. Mean Qp:Qs was 1.83 ± 0.45 by echocardiography and 1.73 ± 0.47 by simulation modelling, with no statistically significant difference between methods (p Using the predefined ± 20% deviation criterion, 53.1% of examinations fell within the acceptable range. Larger ASD diameter was associated with increased inter-method deviation. Conclusions: This study provides an initial feasibility evaluation of a novel haemodynamic simulation approach enabling patient-specific three-dimensional modelling of cardiac structures and shunt physiology based on routine echocardiographic data. Simulation-derived Qp:Qs estimates demonstrated no systematic difference in comparison with conventional Doppler-based quantification and were feasible within routine clinical workflows. Ongoing prospective validation against invasive haemodynamic reference standards will further define analytical accuracy and potential clinical applicability.