Comparison of actual oxygen delivery kinetics to those predicted by mathematical modeling following stage 1 palliation just prior to superior cavopulmonary anastomosis

Summary Background Optimizing systemic oxygen delivery ( DO 2 ) and hemodynamics in children with hypoplastic left heart syndrome ( HLHS ) is a clinical challenge. Mathematical modeling of the HLHS circulation has been used to determine the relationship between oxygen kinetic parameters and DO 2 and to determine how DO 2 might be optimized. The model demonstrates that neither arterial oxygen saturation (SaO 2 ) nor mixed venous oxygen saturation (SvO 2 ) alone accurately predicts DO 2 . Objectives Oxygen delivery kinetics predicted by previously described mathematical modeling were compared with actual patients' hemodynamic data. We sought to determine which patient derived parameters correlated best with DO 2 . Methods Patients with HLHS who underwent cardiac catheterization prior to surgery to create a superior cavopulmonary anastomosis from 2007 to 2011 were identified. Hemodynamic data obtained were compared with the data derived from the mathematical model. Correlations between SaO 2 , SvO 2 , SaO 2 ‐SvO 2 , SaO 2 /(SaO 2 ‐SvO 2 ), pulmonary‐to‐systemic blood flow ratio ( Q p / Q s ), and DO 2 were evaluated using both linear and nonlinear analyses, and R 2 was calculated. Results/Conclusions Patients' data fit most aspects of the mathematical model. DO 2 had the best correlation with SaO 2 /(SaO 2 ‐SvO 2 ; R 2  = 0.8755) followed by SaO 2 ‐SvO 2 ( R 2  = 0.8063), while SaO 2 or SvO 2 alone did not demonstrate a significant correlation as predicated by the mathematical model ( R 2  = 0.09564 and 0.4831, respectively). SaO 2 /(SaO 2 ‐SvO 2 ) would be useful clinically to track changes in DO 2 that occur with changes in patient condition or with interventions.