Background: Alveolar recruitment maneuvers (ARMs) are known to improve perioperative morbidity but can transiently impact cardiac output (CO). This reproducible hemodynamic perturbation creates a clinical opportunity to test multiple devices during acute changes in CO. The objective of this study was to evaluate the ability of 2 minimally invasive CO monitors, the ECOM (Endotracheal Cardiac Output Monitor) and the esCCO (estimated Continuous Cardiac Output), to measure trends in CO during an ARM in postoperative cardiac surgical patients.
Methods: Twenty-seven mechanically ventilated patients were studied in the postoperative intensive care unit setting. Hemodynamic measurements were made at 3 distinct time points: (1) before an ARM at zero end-expiratory pressure; (2) during an ARM at 15 cm H2O positive end-expiratory pressure; and (3) after the ARM again at zero end-expiratory pressure. Reference CO was obtained from intermittent bolus thermodilution (TDco) using a pulmonary artery catheter. At each of the 3 time points, mean values of 3 CO measurements from each device were collected simultaneously, as well as the corresponding changes in arterial pressure. The coefficient of variation of the 3 sets for each patient at each time point allowed for the calculation of the precision error for each device. Differences between absolute values of CO using the 2 tested methods and TDco were assessed using a Bland-Altman plot. Additionally, the agreement and responsiveness of the changes in CO (ΔTDco, ΔESco, and ΔECco for changes in TDco, esCCO, and ECOM, respectively) and mean arterial pressure (MAP) were assessed using both a 4-quadrant plot with the coefficient of correlation concordance (CCC) and a polar plot diagram. A polar concordance rate above 80% was considered clinically acceptable.
Results: Eighty-one sets of 3 CO values were analyzed. Precision error of TDco was approximately 5.1% (interquartile range: 2.8-7.1). Between esCCO and TDco, the mean bias was +0.7 L/min with limits of agreement of -2.1 L/min and +3.5 L/min. Between ECOM and TDco, the mean bias was +0.2 L/min with limits of agreement of -2.0 L/min and +2.4 L/min. The CCC between ΔECco and ΔTDco (0.82 [95% confidence interval (CI), 0.72-0.89]) was significantly higher (P = 0.0053) than the CCC between ΔESco and ΔTDco (0.42 [95% CI, 0.20-0.59]), but not statistically different (P = 0.16) than the CCC between ΔMAP and ΔTDco (0.69 [95% CI, 0.54-0.80]). Polar plot analysis showed an angular bias with radial agreement limits of -29° ± 38° between ΔESco and ΔTDco and -15° ± 29° between ΔECco and ΔTDco. Four-quadrant concordance rate was 81% (95% CI, 74-88) between ΔESco and ΔTDco and 100% between ΔECco and ΔTDco. Polar concordance rates were 41% (95% CI, 34-48) between ΔESco and ΔTDco and 85% (95% CI, 79-90) between ΔECco and ΔTDco.
Conclusions: Compared to pulmonary artery catheter thermodilution, both ECOM and esCCO underestimate changes in CO during an ARM in postoperative cardiac surgical patients. However, ΔECco is within the angular limits of acceptable agreement and may be as efficient as invasive arterial pressure monitoring to track CO changes. In contrast, esCCO is not able to adequately track CO in these specific conditions.