How Reliable Are the Functional Hemodynamic Tests in Predicting Fluid Responsiveness in Patients Undergoing Protective Ventilation During Prone Spine Surgeries?

To the Editor We read with great interest the study by Messina et al,1 which assessed the feasibility and reliability of functional hemodynamic tests in the assessment of fluid responsiveness in patients undergoing prone spine surgeries. We appreciate the research work done by the authors and have a few queries regarding this study. In statistical analysis, the authors mentioned that the receiver operating curves (ROCs) would be constructed for pulse pressure variation (PPV) and stroke volume variation (SVV) recorded at baseline T0 (ie, before the application of the hemodynamic tests). However, in the results, it is mentioned that ROC of PPV and SVV recorded at T3 (ie, before the fluid challenge) did not predict fluid responsiveness.1 We would like to know whether the authors had constructed a ROC curve for PPV and SVV at T3, which they failed to mention in the statistical analysis part, or they considered ROC of the variables at T0. If they had utilized the ROC curves for PPV and SVV measured at T3, there is a concern that the carryover effects of the tidal volume challenge test performed just before T3 could have influenced these variables. In this study, the tidal volume challenge test was performed by increasing the tidal volume from 6 to 8 mL/kg. This could lead to increased alveolar recruitment as compared to baseline. This can affect the pulmonary dynamics and the cardiopulmonary interaction, resulting in a change in PPV and SVV, as evidenced by their study. However, reducing the tidal volume to 6 mL/kg within a 1-minute interval may not necessarily derecruit the already inflated alveoli, thereby influencing the variables recorded at T3. Thus, the PPV and SVV recorded at T3 (after cessation of tidal volume challenge test) may not be representative of the actual baseline, and comparing it with post-fluid challenge for fluid responsiveness could lead to significant bias. Instead, considering the PPV and SVV recorded at T0 as baseline would have negated this probable bias. In this study, the hemodynamic monitoring was done with a MostCare device (Vytech Health, Padova, Italy), which uses a patented algorithm to evaluate hemodynamic parameters from the arterial line, without any previous calibration.1,2 A study by Grensemann et al3 evaluated the efficacy of calibrated and uncalibrated pulse contour–derived cardiac index values in prone positioning, and concluded that the uncalibrated system showed a high degree of error, which was unacceptable. Moreover, it is to be noted that 80% of this study population were hypertensives, and the MostCare device uses a complex calculation of systemic vascular impedance, which could be affected in patients with decreased arterial compliance as in chronic hypertensives.2 Another confounding variable that possibly affects the hemodynamic variables is the intraoperative use of vasopressors. In patients on vasopressors, a system that receives an input from the arterial line can underestimate cardiac output measurements because the results are coupled with the mean arterial pressure, thereby resulting in a higher degree of error.3,4 In this study, the authors failed to mention the intraoperative use of vasopressors. Diabetics, hypertensives, and patients with chronic renal dysfunction were included in this study. These conditions are associated with autonomic dysfunction, which can affect dynamic indices, and it is vital to rule out the autonomic dysfunction before enrolling them.4 Also, they had not considered excluding patients on α-blockers. Patients presenting with upper motor neuron type of bladder paresis due to spinal trauma are often administered α-adrenergic blockers, which could affect the dynamic indices of fluid responsiveness.4 Neeraja Ajayan, MDAjay Prasad Hrishi, DMManikandan Sethuraman, MD, PDCCDivision of NeuroanesthesiaDepartment of AnesthesiologySree Chitra Tirunal Institute for Medical SciencesThiruvananthapuram, Kerala, India[email protected]