PPV and SVV have been widely used to predict fluid responsiveness based on heart–lung interaction. The principle of heart–lung interaction is that cyclic changes in intrathoracic pressure and transpulmonary pressure affect cardiac preload. Therefore, these effects can be seen when ventilating with a sufficient tidal volume of at least 8 mL/kg of predicted body weight [12]. During low tidalvolume ventilation, PPV and SVV may be falsely low because the tidal volume may be insufficient to produce a significant change in intrathoracic pressure [13, 14]. Therefore, it is recommended to use a tidal volume of at least 8 mL/kg with cutoff points for static SVV and PPV of 10% and 13%, respectively. In some situations, a tidal volume of 8 mL/kg might injure the lungs; therefore, we reduced tidal volume to 6 mL/kg, which resulted in the inability of SVV and PPV to predict fluid responsiveness. For example, patients with acute respiratory distress syndrome (ARDS) who received lung-protective ventilation have been excluded from studies using PPV and SVV to predict fluid responsiveness because of the decrease in transmission of intrathoracic pressure to the cardiovascular system [15]. However, Myatra et al. demonstrated that in patients with ARDS with low tidal volume ventilation, a tidal volume challenge from 6 to 8 mL/kg predicted fluid responders with larger areas under the receiver operating characteristic curves compared with using static PPV and SVV at a tidal volume of 8 mL/kg; the cutoff values were 3.5% and 2.5%, respectively. Jun et al. also demonstrated the predictive ability of a tidal volume challenge from 6 to 8 mL/kg in robotic-assisted laparoscopic surgery in the Trendelenburg position, with cutoff values of 1% for ∆PPV and 2% for ∆SVV [9]. Messina et al. performed a tidal volume challenge from 6 to 8 mL/kg in elective neurosurgery and found that tidal volume challenge predicted fluid responsiveness with cutoff values of 13.3% for PPV and 12.1% for SVV [16]. Additional previous studies illustrated the ability of tidal volume challenge to predict fluid responsiveness, using variable cutoff values. The main factor explaining these different cutoff values may be differences in patients’ chest wall compliance. Liu et al. inserted esophageal balloons in patients with ARDS and found that pleural pressure change (ΔPpl) was the most important determinant of PPV among other respiratory variables (plateau pressure, change in airway pressure, tidal volume, respiratory elastance, ΔPpl, and chest wall elastance/respiratory elastance (Ecw/ERS)) in both responders and nonresponders [17]. Moreover, the authors emphasized that ΔPpl was attenuated primarily by a low Ecw/ERS ratio and, to a lesser extent, by low tidal volume. Therefore, PPV and SVV in patients with low Ecw/ERS was less reliable than in patients with a high Ecw/ERS, with a proposed cutoff of 0.28, according to Liu et al.‘s study, and tidal volume challenge in low Ecw might result in an insufficient increase in ΔPpl.
In the present study, we performed tidal volume challenge in patients with intermediate to high risk of PPCs receiving lung-protective ventilation using a tidal volume of 6 mL/kg of predicted body weight. To our knowledge, this is the first study to evaluate increasing tidal volume from 6 to 8 mL/kg and then from 8 to 10 mL/kg, to maximize the efficacy of the tidal volume challenge. We found different results compared with previous studies, and we hypothesized that tidal volume challenges using increases from 6 to 8 mL/kg, 8 to 10 mL/kg, and 6 to 10 mL/kg did not cause adequate pleural pressure change. Because our patient population constituted postoperative patients with intermediate to high risk of PPCs (primarily atelectasis), the average respiratory compliance was 38.2±9.7 mL/cmH2O; therefore, more than half of the patients had values below the normal range. Atelectasis might have caused decreased lung compliance in our patients. We assumed that our patients had normal chest wall compliance according to results from previous studies evaluating low chest wall compliance; for example, our patients were not obese or septic, and the administered perioperative fluid volume was less than 3L [18]. Our patients might have had normal or good chest wall compliance, but low respiratory compliance from increased lung stiffness secondary to atelectasis; therefore, the usefulness of the tidal volume challenge was limited in these patients. Differences in the cutoff values reported after tidal volume challenges in different studies may be explained by differences in patients’ chest wall and respiratory elastance, which were not measured in our study. Messina et al. reported much higher cutoff values after the tidal volume challenge compared with the studies of Myatra et al. and Jun et al. [8, 9]. Patients from Messina et al.‘s study underwent cranial surgery and had an average respiratory compliance of 65 [58–73] mL/cmH2O, suggesting that the sensitivity of the tidal volume challenge was lower in patients with good respiratory compliance. We concluded that a change of 2 or 4 mL/kg of tidal volume from baseline might not cause sufficient changes in pleural pressure to affect cardiac preload, especially in patients with good chest wall compliance and low total respiratory compliance.
The main limitation of this study is that we did not measure pleural pressure change; therefore, our conclusion regarding the inability of the tidal volume challenge to predict fluid responsiveness in this study was based on knowledge from previous studies. Another limitation is that we intermittently administered sedative agents during the procedure, which might have induced cardiovascular effects; for instance, vasodilatation. As a result, the interval between the beginning of the study and the fluid loading to identify fluid responsiveness might be a confounder because of changes in fluid responsive status related to the sedative drugs. This limitation can be minimized by performing the tidal volume challenge under constant-level sedation or anesthesia.