Pulse pressure variation does not reflect stroke volume variation in mechanically ventilated rats with lipopolysaccharide-induced pneumonia

1. The present study examined the relationship between centrally measured stroke volume variation (SVV) and peripherally derived pulse pressure variation (PPV) in the setting of increased total arterial compliance (CA rt ). 2. Ten male Wistar rats were anaesthetized, paralysed and mechanically ventilated before being randomized to receive intrapulmonary lipopolysaccharide (LPS) or no LPS. Pulse pressure (PP) was derived from the left carotid artery, whereas stroke volume (SV) was measured directly in the left ventricle. Values of SVV and PPV were calculated over three breaths. Balloon inflation of a catheter positioned in the inferior vena cava was used, for a maximum of 30 s, to decrease preload while the SVV and PPV measurements were repeated. Values of CA rt were calculated as SV/PP. 3. Intrapulmonary LPS increased CA rt and SV. Values of SVV and PPV increased in both LPS-treated and untreated rats during balloon inflation. There was a correlation between SVV and PPV in untreated rats before (r = 0.55; P = 0.005) and during (r = 0.69; P < 0.001) occlusion of the vena cava. There was no such correlation in LPS-treated rats either before (r = -0.08; P = 0.70) or during (r = 0.36; P = 0.08) vena cava occlusion. 4. In conclusion, under normovolaemic and hypovolaemic conditions, PPV does not reflect SVV during an increase in CA rt following LPS-induced pneumonia in mechanically ventilated rats. Our data caution against their interchangeability in human sepsis