OBJECTIVETo assess the significance of stroke volume variation (SVV) and intrathoracic blood volume index (ITBI) on the responsiveness to volume loading in mechanically ventilated canine with hemorrhagic shock.

METHODSHemorrhagic shock canine model was established with the modified Wiggers' method. The heart rate (HR), mean artery pressure (MAP), central venous pressure (CVP), pulmonary artery wedge pressure (PAWP), intrathoracic blood volume index (ITBI) and stroke volume variation (SVV) were investigated by Swan-Ganz catheter or PiCCO monitor. Graded volume loading (VL) was performed. Successive responsive VLs were performed (increase in SV > 5% after VL) until continuous change in SV < 5% (unresponsive) was reached.

RESULTSFourteen canines were studied and a total of 134 VLs were performed. In 94 VLs, an increase in SV of more than 5% was reached. In the other 40 VLs, increase in SV was less than 5%. The change of HR, MAP, ITBI, SVV in responsive were more than those of unresponsive after VL. The change of CVP, PAWP in responsive were less than those of unresponsive. Significant correlation was found between DeltaSV after VL and the baseline values of ITBI, SVV. No correlation was found between DeltaSV and HR, MAP, CVP, PAWP. Significant correlations were also found between DeltaSV and DeltaCVP, DeltaPAWP, DeltaITBI, DeltaSVV after fluid loading. No correlation was found between DeltaSV and DeltaHR, DeltaMAP. By using receiver operating characteristic analysis, the area under the curve were 0.872 for SVV and 0.689 for ITBI, more than those of HR, MAP, CVP, PAWP statistically. As SVV value of 9.5% or more will predict an increase in the SV of at least 5% in response to a VL with a sensitivity of 92.6% and a specificity of 82.5%.

CONCLUSIONSSVV and ITBI were more useful indicators than CVP and PAWP on the assessment of responsiveness to volume loading. SVV as a functional preload parameter and for on-line monitoring may help to improve the hemodynamic management.