RePeTAr: An In Silico Lumped-Parameter Hemodynamic Model for Deep Venous Arterialization

Objective Deep venous arterialization (DVA) is used for selected patients with no-option chronic limb-threatening ischemia (CLTI), but the hemodynamic consequences of a proposed DVA configuration are difficult to estimate. We describe RePeTAr, a zero-dimensional, lumped-parameter electrical-resistance model, and report an exploratory in silico analysis of three virtual configurations. The model is not computational fluid dynamics (CFD), does not solve the Navier–Stokes equations, and has not been clinically validated. Methods RePeTAr (Rapid Prediction of Deep Venous Arterialization Outcomes) represents mean arterial pressure as the pressure source; shunt and venous-network resistances as serial elements; and tissue and steal-pathway resistances as parallel elements. The current mobile application is available on iOS and Android platforms. Three illustrative configurations were entered into the application. Model outputs were total circuit flow, model-estimated tissue-branch flow, model-estimated steal-branch flow, and steal fraction. No real patient data, measured tissue perfusion, or clinical outcome data were used. Results For the three predefined virtual configurations, total circuit flow was 450.10, 308.55, and 192.10 mL/min, respectively. Corresponding model-estimated tissue-branch flow was 326.69, 179.99, and 102.45 mL/min; model-estimated steal-branch flow was 123.42, 128.56, and 89.65 mL/min; and model-estimated steal fraction was 27.42%, 41.67%, and 46.67%. The branch flows summed to total circuit flow, subject to rounding. These numerical differences are arithmetic outputs of the prespecified inputs; they do not demonstrate prediction of DVA success, limb salvage, thrombosis, or amputation. Conclusions RePeTAr is a transparent, exploratory lumped-parameter model that rapidly reports internally consistent branch-flow estimates for illustrative DVA configurations. Its potential clinical utility, parameter calibration, and any clinically actionable thresholds require evaluation against prospectively collected patient-level data.