Evaluation of oesophageal Doppler-derived variables and plethysmographic variability index to predict fluid responsiveness in mechanically ventilated healthy dogs.

OBJECTIVE

To evaluate the plethysmography variability index (PVI) and oesophageal Doppler (OD)-derived dynamic variables [stroke distance variation (StDV) and aortic flow peak velocity variation (PVV)] as predictors of fluid responsiveness.

STUDY

DESIGN

A prospective cohort, diagnostic test study.

ANIMALS: A group of 30 client-owned dogs undergoing general anaesthesia.

METHODS

Dogs received dexmedetomidine (3 μg kg-1) and methadone (0.3 mg kg-1), intramuscularly. Anaesthesia was induced with propofol and maintained with isoflurane. Volume-controlled ventilation was instituted with tidal volume 10 mL kg-1, inspiratory-to-expiratory ratio 1:2. Heart rate (HR) and mean arterial pressure (MAP) were monitored. An OD probe was inserted to measure stroke distance (StD). Once surgery finished, PVV, StDV and PVI were recorded before (T0) and after (T1) the administration of 3 mL kg-1 bolus of hydroxyethyl starch 6% over 5 minutes. Dogs were classified as fluid-responders (R; StD increased ≥ 15%) or non-fluid-responders (NR). Receiver operating characteristic (ROC) curves were plotted for PVV, StDV and PVI.

RESULTS

Of the 30 animals, eight were classified as R and 22 as NR. The areas under the ROC curves for PVV, StDV and PVI were 0.87 [confidence interval (CI) 0.82-1], 0.84 (CI 0.67-1) and 0.78 (CI 0.62-1), respectively. The best cut-offs (positive-negative predictive values) were 9% (54-94%) for PVV (sensitivity, 88%; specificity, 73%), 12% (58-94%) for StDV (sensitivity, 88%; specificity, 77%) and 15% (60-90%) for PVI (sensitivity, 75%; specificity, 82%). No statistical differences were found between R and NR, nor between T0 and T1, for MAP or HR.

CONCLUSIONS AND CLINICAL RELEVANCE: OD-derived values of PVV, StDV and PVI are reasonable predictors of fluid responsiveness in client-owned dogs undergoing general anaesthesia and mechanical ventilation.