Physiological Modeling of Hemodynamic Responses to Sodium Nitroprusside

Background: Computational modeling of physiology has become a routine element in the development, evaluation, and safety testing many types medical devices. Members Food Drug Administration have recently published manuscript detailing validation, sensitivity computational model for blood volume, cardiac stroke pressure, noting that such might be useful development closed-loop fluid administration systems. In present study, we expanded on this to include pharmacologic effect sodium nitroprusside calibrated against our previous experimental animal data. Methods: Beginning with elements original publication, added six new parameters control nitroprusside: two onset time clearance rates, volume (which includes venodilation as “hidden” element), direct arterial pressure. Using model, then predictive performance previously collected study data using infusions simulate shock primary emphasis MAP. Root-mean-squared error (RMSE) was calculated, compared study. Results: RMSE model-predicted MAP actual lower than reported but higher SV CO. The individually fit models showed population average values parameters, suggesting fitting process effective identifying improved parameters. Use partially after removal lowest variance very minor decrement improvement over fully models. Conclusion: clinical effects SNP successfully an <10% mean Model-predicted similar seen base when shifts, heart rate, drug dose inputs.