Mechanical ventilation guided by uncalibrated esophageal pressure may be potentially harmful

Background: Esophageal balloon calibration was proposed in acute respiratory failure patients to improve esophageal pressure assessment. In a clinical setting characterized by a high variability of abdominal load and intrathoracic pressure (i.e., pelvic robotic surgery), the authors hypothesized that esophageal balloon calibration could improve esophageal pressure measurements. Accordingly, the authors assessed the impact of esophageal balloon calibration compared to conventional uncalibrated approach during pelvic robotic surgery. Methods: In 30 adult patients, scheduled for elective pelvic robotic surgery, calibrated end-expiratory and end-inspiratory esophageal pressure, and the associated respiratory variations were obtained at baseline, after pneumoperitoneum-Trendelenburg application, and with positive end-expiratory pressure (PEEP) administration and compared to uncalibrated values measured at 4-ml filling volume, as per manufacturer recommendation. Data are expressed as median and [25th, 75th percentile]. Results: Ninety calibrations were successfully performed. Chest wall elastance worsened with pneumoperitoneum-Trendelenburg and PEEP (19.0 [15.5, 24.6] and 16.7 [11.4, 21.7] cm H2O/l) compared to baseline (8.8 [6.3, 9.8] cm H2O/l; P < 0.0001 for both comparisons). End-expiratory and end-inspiratory calibrated esophageal pressure progressively increased from baseline (3.7 [2.2, 6.0] and 7.7 [5.9, 10.2] cm H2O) to pneumoperitoneum-Trendelenburg (6.2 [3.8, 10.2] and 16.1 [13.1, 20.6] cm H2O; P = 0.014 and P < 0.001) and PEEP (8.8 [7.7, 15.6] and 18.9 [16.3, 22.0] cm H2O; P < 0.0001 vs. baseline for both comparison; P < 0.001 and P = 0.002 vs. pneumoperitoneum-Trendelenburg) and, at each study step, they were persistently lower than uncalibrated esophageal pressure (P < 0.0001 for all comparisons). Overall, difference among uncalibrated and calibrated esophageal pressure was 5.1 [3.8, 8.4] cm H2O at end-expiration and 3.8 [3.0, 6.3] cm H2O at end-inspiration. Uncalibrated esophageal pressure swing was always lower than calibrated one (P < 0.0001 for all comparisons) with a difference of -1.0 [-1.8, -0.4] cm H2O. Conclusions: In a clinical setting with variable chest wall mechanics, uncalibrated measurements substantially overestimated absolute values and underestimated respiratory variations of esophageal pressure. Calibration could substantially improve mechanical ventilation guided by esophageal pressure.