Evaluation of the ability of haemodynamic variables obtained with minimally invasive techniques to assess fluid responsiveness in endotoxaemic Beagles

Objective: To examine the ability of different haemodynamic variables recorded by minimally invasive monitoring techniques to assess fluid responsiveness (FR) in endotoxaemic Beagles. Study design: Prospective terminal experimental study. Animals: A group of six healthy, purpose-bred Beagle dogs (three intact females and males), age 5–9.8 years (range) and weighing 11.4–17.9 kg. Methods: Endotoxaemic shock was induced by injecting 1 mg kg–1 Escherichia coli lipopolysaccharide (LPS) intravenously in six sevoflurane-anaesthetized mechanically ventilated Beagles for another project. After 10 minutes, three Ringer's acetate boluses (10 mL kg–1) were administered each over 10 minutes with collection of haemodynamic data immediately before and after each bolus. Thereafter, arterial hypotension was treated with noradrenaline ± dexmedetomidine until arterial pressures increased to a target value. After a wash-out period of 20 minutes another three boluses of fluid were administered and measurements were repeated equally. For each fluid bolus, FR was considered positive when change (Δ) in stroke volume measured by pulmonary artery thermodilution was ≥15%. To test predictive accuracy for FR, we recorded heart rate, invasive arterial, right atrial and pulmonary capillary wedge pressures, pulse wave transit time with haemodynamic monitors, calculated pulse pressure, shock index and rate over pressure evaluation (ROPE) and measured stroke distance and corrected flow time (FTc) with oesophageal Doppler monitoring. Results: A total of 35 measurements (19 positive and 16 negative responses) were evaluated. A FTc < 330 ms, Δ pulse pressure ≥20%, Δ shock index ≤–14% and ΔROPE ≤–17% were the most significant indicators of positive FR with an area under the receiver operating characteristics curve between 0.72 and 0.74. Conclusions and clinical relevance: In endotoxaemic Beagles, none of the assessed haemodynamic variables could predict FR with high sensitivity and specificity.