Non-invasive prediction of fluid responsiveness during major hepatic surgery

Abstract

The aim of this study was to evaluate potential predictors of fluid responsiveness obtained during major hepatic surgery. The predictors studied were invasive monitoring of intravascular pressures (radial and pulmonary artery catheter), including direct measurement of respiratory variation in arterial pulse pressure (PPVart), transoesophageal echocardiography (TOE), and non-invasive estimates of PPVart from the infrared photoplethysmography waveform from the Finapres (PPVfina) and the pulse oximetry waveform (PPVsat). We conducted a prospective study of 54 fluid challenges (250 ml colloid) given for haemodynamic instability in eight patients undergoing hepatic resection. Fluid responsiveness was defined as an increase in stroke volume index (SVI) >or=10%. The following variables were recorded before each fluid challenge: right atrial pressure (RAP), pulmonary artery occlusion pressure (PAOP), PPVart, PPVfina, PPVsat, and the TOE-derived variables left ventricular end-diastolic area index (LVEDAI), early/late (E/A) diastolic filling wave ratio, deceleration time of the E wave (MDT) of mitral flow and the systolic fraction of the pulmonary venous flow (SF). Only PPVfina, PPVart (both P<0.001), PPVsat (P=0.02), LVEDAI and MDT (both P=0.04) were different in responder vs non-responder fluid challenges. The areas under the receiver operating characteristic (ROC) curves were 0.81 (PPVfina), 0.79 (PPVart), 0.70 (LVEDAI), 0.68 (PPVsat and MDT), 0.63 (RAP), 0.62 (E/A), 0.55 (PAOP) and 0.42 (SF). The areas under the ROC curves for RAP, E/A, PAOP and SF were significantly less than that for PPVfina (P<0.05 in each case). Only PPVart (r=0.59, P=0.0001) and PPVfina (r=0.56, P=0.0001) correlated with the fluid challenge-induced changes in SVI. PPVart and PPVfina predict fluid responsiveness during major hepatic surgery. This suggests that intraoperative monitoring of fluid responsiveness may be implemented simply and non-invasively.