Global and right ventricular end-diastolic volumes correlate better with preload after correction for ejection fraction

Background Volumetric monitoring with right ventricular end-diastolic volume indexed (RVEDVi) and global end-diastolic volume indexed (GEDVi) is increasingly being suggested as a superior preload indicator compared with the filling pressures central venous pressure (CVP) or the pulmonary capillary wedge pressure (PCWP). However, static monitoring of these volumetric parameters has not consistently been shown to be able to predict changes in cardiac index (CI). The aim of this study was to evaluate whether a correction of RVEDVi and GEDVi with a measure of the individual contractile reserve, assessed by right ventricular ejection fraction (RVEF) and global ejection fraction, improves the ability of RVEDVi and GEDVi to monitor changes in preload over time in critically ill patients. Methods Hemodynamic measurements, both by pulmonary artery and by transcardiopulmonary thermodilution, were performed in 11 mechanically ventilated medical ICU patients. Correction of volumes was achieved by normalization to EF deviation from normal EF values in an exponential fashion. Data before and after fluid administration were obtained in eight patients, while data before and after diuretics were obtained in seven patients. Results No correlation was found between the change in cardiac filling pressures (delta CVP, delta PCWP) and delta CI (R2 0.01 and 0.00, respectively). Further, no correlation was found between delta RVEDVi or delta GEDVi and delta CI (R2 0.10 and 0.13, respectively). In contrast, a significant correlation was found between delta RVEDVi corrected to RVEF (delta cRVEDVi) and delta CI (R2 0.64), as well as between delta cGEDVi and delta CI (R2 0.59). An increase in the net fluid balance with +844 +/- 495 ml/m2 resulted in a significant increase in CI of 0.5 +/- 0.3 l/min/m2; however, only delta cRVEDVi (R2 0.58) and delta cGEDVi (R2 0.36) correlated significantly with delta CI. Administration of diuretics resulting in a net fluid balance of -942 +/- 658 ml/m2 caused a significant decrease in CI with 0.7 +/- 0.5 l/min/m2; however, only delta cRVEDVi (R2 0.80) and delta cGEDVi (R2 0.61) correlated significantly with delta CI. Conclusion Correction of volumetric preload parameters by measures of ejection fraction improved the ability of these parameters to assess changes in preload over time in this heterogeneous group of critically ill patients.