Sources of discrepancy between heart end diastolic volume measured by ultrasound dilution (UD) and transpulmonary thermodilution (TT)

Abstract

Introduction: Quantitative assessment of patient volume status is crucial for therapy titration. Literature suggests that blood volumes measured by dilution methods may be better indicators of preload than the cardiac filling pressures. New UD method (COstatus, Transonic Systems Inc., USA) and TT method (PiCCO, Pulsion, Germany) both estimate heart diastolic volume: TEDV (Total End Diastolic Volume) by COstatus and GEDV (Global End Diastolic Volume) by PiCCO. Hypothesis: It is known that UD and TT produce different results related to the end diastolic heart volume. The purpose of this study is to compare GEDV and TEDV with referenced data from literature and analyze the major sources of discrepancy Methods: Thirty patients, 10 -Sepsis, ALI/ARDS, 17 -Septic Shock, 2-Hemorrhage and 1-Pulmonary Edema, were studied per the approved protocol. A triple lumen central venous catheter and a 5Fr PiCCO femoral artery catheter were inserted into all the patients for their care in the ICU. A sterile tube set (AV loop) was connected between the arterial and venous catheters through stopcocks for COstatus measurements. 20ml iced 5% dextrose solution was injected into the venous catheter for GEDV measurements while 25 ml body temperature saline was injected into the venous limb of the AV loop for TEDV measurements. 2-3 discrete measurements from each method were obtained and averaged for comparison. Results: : In this study GEDV measurements were 2 times higher than TEDV (Table 1). Reference data for Left Ventricle End Diastolic Volume (LVEDV) measured by Transesophageal echocardiography; Transthoracic echocardiography and 3D Echocardiography was collected from 6 publications. Total 387 measurements from normal adults (including athletes), septic patients, patients with dilated cardiomyopathy and patients with hypertrophic cardiomyopathy were summarized. The mean value for 4∗ LVEDV ranged between 226 -379 ml/m2. Only in patients with hypertrophic cardiomyopathy the mean value was 600 ml/m2. In mathematical modeling used by PICCO the value for GEDV is calculated by subtracting pulmonary thermal volume from intrathoracic thermal volume. In COstatus measurements of TEDV the mathematical model is based on the assumption that main factor that forms the spread of saline dilution curve is the mixing in heart chambers. Conclusions: Different mathematical models of indicator passage through cardiopulmonary system used by COstatus and by PICCO leads to different values for heart volumes. COstatus readings for heart end diastolic volumes were mostly in the expected range. PICCO readings were 2-3 times higher. Future assessment of volume by dilution technique should include simultaneous measurement with a reference imaging method.