Assessing fluid responsiveness during open chest conditions

BACKGROUND: Measurement of ventilation-induced left ventricular stroke volume variations (SVV) or pulse pressure variations (PPV) is useful to optimize preload in patients after cardiac surgery. The aim of this study was to investigate the ability of SVV and PPV measured by arterial pulse contour analysis to assess fluid responsiveness in patients undergoing coronary artery bypass surgery during open-chest conditions.

METHODS: We studied 22 patients immediately after midline sternotomy. We determined SVV, PPV, left ventricular end-diastolic area index by transoesophageal echocardiography, global end-diastolic volume index and cardiac index by thermodilution before and after removal of blood 500 ml and after volume substitution with hydroxyethyl starch 6%, 500 ml.

RESULTS: Blood removal resulted in a significant increase in SVV from 6.7 (2.2) to 12.7 (3.8)%. PPV increased from 5.2 (2.5) to 11.9 (4.6)% (both P<0.001). Cardiac index decreased from 2.9 (0.6) to 2.3 (0.5) litres min(-1) m(-2) and global end-diastolic volume index decreased from 650 (98) to 565 (98) ml m(-2) (both P<0.025). Left ventricular end-diastolic area index did not change significantly. After fluid loading SVV decreased significantly to 6.8 (2.2)% and PPV decreased to 5.4 (2.1)% (both P<0.001). Concomitantly, cardiac index increased significantly to 3.3 (0.5) litres min(-1) m(-2) (P<0.001) and global end-diastolic volume index increased significantly to 663 (104) ml m(-2) (P<0.005). Left ventricular end-diastolic area index did not change significantly. We found a significant correlation between the increase in cardiac index caused by fluid loading and SVV as well as PPV before fluid loading (SVV, R=0.74, P<0.001; PPV, R=0.61, P<0.005). No correlations were found between values of global end-diastolic volume index or left ventricular end-diastolic area index before fluid loading and the increase in cardiac index.

CONCLUSION: Measurement of SVV or PPV allows assessment of fluid responsiveness in hypovolaemic patients under open-chest and open-pericardium conditions. Thus, measuring heart-lung interactions may improve haemodynamic management during surgical procedures requiring mid-line sternotomy.