In patients in the prone position, the reliability of pulse pressure variation and stroke volume variation (PPV and SVV) and the use of functional hemodynamic tests to predict fluid responsiveness have not previously been established. Perioperatively, in this setting, optimizing fluid management can be challenging, and fluid overload is associated with both intraoperative and postoperative complications. We designed this study to assess the sensitivity and specificity of baseline PPV and SVV, the tidal volume (V_T) challenge (V_TC) and the end-expiratory occlusion test (EEOT) in predicting fluid responsiveness during elective spinal surgery.

METHODS:

The study protocol was started during a period of intraoperative hemodynamic stability after prone positioning and before the administration of any vasopressor: (1) at baseline, the controlled ventilation was set at 6 mL/kg of predicted body weight (PBW) (T₀); (2) patients underwent the first EEOT (EEOT₆) by interrupting the mechanical ventilation for 30 seconds; (3) the ventilation was set again at 6 mL/kg PBW for 1 minute (T₁); (4) the V_TC was applied by increasing the V_T up to 8 mL/kg PBW for 1 minute; (5) the ventilation was kept at 8 mL/kg PBW for 1 minute (T₂); (6) a second EEOT (EEOT₈) was performed; (7) the V_T was reduced back to 6 mL/kg PBW for 1 minute (T₃); (8) a fluid challenge of 250 mL of Ringer’s solution was infused over 10 minutes. After each step, a complete set of hemodynamic measurements was recorded.

RESULTS:

Neither PPV and SVV values recorded at T₃ nor the EEOT₆ or the EEOT₈ predicted fluid responsiveness. The change in PPV after V_TC application predicted fluid responsiveness with an area under the curve of 0.96 (95% confidence interval, 0.87–1.00), showing a sensitivity of 95.2% and a specificity of 94.7%, using a cutoff increase of 12.2%. The change in SVV after V_TC application predicted fluid responsiveness with an area under the curve 0.96 (95% confidence interval, 0.89–1.00) showing a sensitivity of 95.2% and a specificity of 94.7%, using a cutoff increase of 8.0%. A linear correlation between stroke volume index changes after fluid challenge administration and the changes in PPV and SVV after V_TC application was observed (r = 0.71; P < .0001 and r = 0.68; P < .0001, respectively).

CONCLUSIONS:

In prone elective neurosurgical patients, the baseline values of PPV and SVV and the EEOT fail to predict fluid responsiveness, while the V_TC is a very reliable functional hemodynamic test and could be helpful in guiding intraoperative fluid therapy.