J17 - Modeling Mitral Regurgitation: A Lumped-Parameter Approach to Forward Stroke Volume and Diagnostic Precision

Introduction: : Mitral regurgitation (MR) is a prevalent valvular disorder characterized by incomplete closure of the mitral valve during systole, resulting in retrograde blood flow into the left atrium. Affecting up to 10% of individuals over the age of 75, MR poses a growing public health concern amid global aging. Its pathophysiology imposes a chronic volume overload on the LV, which despite preserved or elevated total stroke volume, results in diminished forward SV and effective perfusion. This mismatch often leads to underestimation of MR severity when assessed solely via conventional parameters like EF. Current diagnostic workflows rely heavily on echocardiographic indices ( EF, EROA), which are load-dependent and may miss early LV dysfunction. Although advanced imaging modalities and pressure-volume analysis offer more comprehensive insights, they remain underutilized due to technical complexity and inter-operator variability. With MR patients often remaining asymptomatic until advanced stages, more physiologically reflective and automated diagnostic tools are needed to guide early intervention. Emerging AIframeworks show promise in transforming cardiovascular diagnostics by integrating multi-parametric data and providing interpretable, patient-specific predictions. However, the hemodynamic burden of MR, particularly its impact on forward stroke volume and load-independent cardiac output, remains insufficiently explored in model-based simulations. In this study, we present a computational modeling framework for simulating cardiac dynamics in MR using a closed-loop lumped-parameter model (LPM). The objective is to investigate the discrepancy between EF and forward cardiac output and to demonstrate the need for incorporating forward stroke volume as a critical, AI-compatible metric in MR diagnosis and treatment planning.

Materials and

Methods: : We utilized a validated closed-loop lumped-parameter model (LPM) of the cardiovascular system, modified from the framework proposed by Regazzoni et al. (2022). This model simulates the heart and vasculature using electrical analogs: resistors for flow resistance, capacitors for vascular compliance, and inductors for blood inertia. Each cardiac chamber, left/right atria and ventricles, is modeled with time-varying elastance to represent contractile function dynamically throughout the cardiac cycle. Mitral valve regurgitation was introduced by modulating the leakage resistance (Rm_leak) across the mitral valve. Valve flow was governed by pressure gradients and adjusted resistances that emulate physiological valve behavior. Simulations were conducted across a range of Rm_leak values to represent varying MR severities. Additionally, simulations were performed under bradycardic conditions to study the interaction between heart rate and regurgitant volume. Model validation under healthy conditions yielded physiological benchmarks (e.g., EF = 64.1%, CO = 4.61 L/min) consistent with clinical data. Key hemodynamic parameters such as stroke volume (SV), cardiac output (CO), and pressure-volume loops were extracted for each simulated condition. The effect of MR was quantified through deviations in these parameters, particularly focusing on the divergence between EF and forward SV. The model’s dynamic behavior was visualized via time-resolved pressure-volume loops and flow graphs, providing insight into both acute and chronic hemodynamic adaptations. Results were analyzed to support a mechanistic argument for forward SV as a more reliable metric in MR severity stratification and intervention timing.

Results, Conclusions, and Discussions:: Simulations of increasing mitral regurgitation severity revealed a significant rise in total stroke volume while forward stroke volume declined. Despite preserved or elevated EF (>65% across all MR severities), forward flow (cardiac output) dropped from 6.12 L/min to 4.66 L/min as Rm_leak increased. This highlights a key discrepancy: conventional metrics like EF remain falsely reassuring in the presence of significant regurgitation.
Pressure-volume loops progressively shifted rightward with increasing MR, indicating elevated preload and reduced afterload. Notably, the end-systolic volume remained relatively stable while end-diastolic volume increased, supporting the concept of volume overload compensation. A fixed heart rate simulation confirmed that bradycardia compounds MR’s impact by reducing the effective number of forward ejections per minute.
These findings underscore that EF alone is inadequate to assess functional impairment in MR. Forward SV, as demonstrated in our simulations, is more reflective of systemic perfusion and regurgitant burden. This reinforces the clinical argument for integrating forward flow metrics into both diagnosis and longitudinal monitoring.

Acknowledgements and/or References (Optional)::