The comparison between stroke volume variation and filling pressure as an estimate of right ventricular preload in patients undergoing renal transplantation.

Toyoda, Daisuke; Fukuda, Mitsue; Iwasaki, Ririko; Terada, Takashi; Sato, Nobukazu; Ochiai, Ryoichi; Kotake, Yoshifumi

PURPOSE The purpose of this prospective, observational study was to respiratory variation of stroke volume (stroke volume variation, SVV) against central venous pressure (CVP) and pulmonary artery diastolic pressure (PADP) as an estimate of right and left ventricular preload. METHODS With IRB approval and informed consent, 31 patients undergoing living related renal transplantation were analyzed. Under general anesthesia with positive pressure ventilation, stroke volume index and SVV were continuously monitored with FloTrac/Vigileo monitor. Right ventricular end-diastolic volume index (RVEDVI) as well as CVP and PADP were continuously monitored with volumetric pulmonary artery catheter. Data of every 30 min interval were used for analysis. The relationship between RVEDVI and CVP, PADP, SVV was analyzed with non-linear regression and the goodness-of-fit was assessed with coefficient of determination (R(2)) of each regression curve. The ability of CVP, PADP and SVV to correctly differentiate RVEDVI <100, <120 or >138 ml/m(2), which were used to guide fluid administration, was also assessed with ROC analysis. RESULTS Three hundred forty-eight data sets were obtained and analyzed. The goodness of fit between RVEDVI and SVV (R(2)) = 0.48) was better than that between RVEDVI and CVP or PADP (R(2)) = 0.19 and 0.33, respectively). The area under the ROC curve of SVV was significantly high compared to CVP or PADP. CONCLUSIONS This study confirmed the theoretical framework of right ventricular preload and ventricular filling pressure and respiratory variation of stroke volume. The result also suggests that SVV can correctly predict preload status compared to pressure-based indices.