M.-H. Lee and T. Rosenthal: J Extra Corpor Technol 2026, 58, 43 – 50 49 that the in-line PaO 2 calculation may underestimate oxygen consumption during the rewarming and rewarmed periods.
Quantum System offers an autoregulation function of PaO 2 and PaCO 2, which relies heavily on the accuracy of the calculated in-line PaO 2 and PaCO 2. Ourfindings indicate that the autoregulation function of PaO 2 should not be used before the first in vivo calibration and during the rewarming and rewarmed periods, at least when using the FX05 oxygenator. Since the calculated in-line PaO 2 is significantly higher than the actual PaO 2 during these periods, it is possible that PaO 2 can become dangerously low with the autoregulation function.
We identified a significant error in the calculated in-line PaO 2 exists at the 1st blood gas series, strongly correlated with patient weight( Table 2 and Figure 1A). The combined errors during the rewarming and rewarmed periods moderately correlated with patient weight( R = 0.46, Table 2, Figure 2). A very strong correlation between the calculated in-line PaO 2 and FiO 2 at the 1st blood gas series was observed( Figure 1D). Using these strong correlations, we derived a formula to predict FiO 2 based on patient weight to achieve a target PaO 2 at the 1st blood gas series( see Results). We applied this formula in 15 cases, aiming for a target PaO 2 of 250 mmHg for validation of the formula. The measured PaO 2 with iSTAT ranged from 180 to 290 mmHg, with an average and standard deviation of 243 ± 39 mmHg. In contrast, the calculated in-line PaO 2 of Quantum System ranged from 329 to 394 mmHg, with an average and standard deviation of 355 ± 20 mmHg( data not shown). The measured PaO 2 were all within LOA, suggesting the formula effectively predicts FiO 2 for the target PaO 2 at the 1st blood gas series. Table 5 provides the predicted FiO 2 based on patient weight for achieving the target PaO 2 of 250 mmHg and the predicted in-line PaO 2 corresponding to the predicted FiO 2 at the 1st blood gas series using the FX05 oxygenator.
Given the potential for dangerously low actual PaO 2 during the rewarming and rewarmed periods despite the high calculated in-line PaO 2 of the Quantum System, we emphasize monitoring the in-line SaO 2 during these periods. The in-line SaO 2 of the Quantum System, measured using the spectrophotometric method, has proven accurate throughout CPB [ 24 ]. We recommend increasing FiO 2 if SaO 2 decreases by more than 3 – 4 % after the initiation of rewarming and conducting a blood gas analysis to calibrate the calculated in-line PaO 2.
The calculated in-line PaCO 2 of the Quantum System is acceptable during the cooling and cooled periods without the first in vivo calibration, although SD is largest at the 1st blood gas series and decreases with each calibration( Table 3). The proportion of larger error distribution diminishes as well( Table 4). These observations indicate improved precision with each in vivo calibration, supporting the use of“ Capture All / Sync” with each blood gas analysis.
During the rewarming period, the calculated in-line PaCO 2 drifted upward( Table 4), unrelated to patient weight but correlated with the temperature gradient( Figure 3). This likely occurs as colder venous blood encounters the warmer oxygenator, leading to transiently more CO 2 expiration at the oxygenator, causing upward drift, as the calculated in-line PaCO 2 relies on FeCO 2 measurement.
We initiated this quality improvement initiative with the question,“ Is continuous in-line blood gas monitoring reliable during cardiopulmonary bypass when PaO 2 and PaCO 2 are calculated rather than measured?” Since Quantum System does not measure PaO 2 or PaCO 2, it may not need to comply with CLIA guidelines. However, as a claimed CILBGM device, it should provide values within acceptable targets throughout CPB. We demonstrated that the calculated in-line PaO 2 of the Quantum System is unreliable before the first in vivo calibration and during the rewarming and rewarmed periods. This is likely due to reliance on a single universal formula for different oxygenators, patient sizes, conditions, and surgical procedures without measuring actual PaO 2. By analyzing our data of the Quantum System with the FX05 oxygenator, we identified when and how errors in the calculated in-line PaO 2 and PaCO 2 occur, leading to recommendations for minimizing significant deviations from actual values. However, this does not align with expectations for a CILBGM device. Specific formulas may need development for each oxygenator and varying conditions during CPB, such as different patient sizes, before the first in vivo calibration, cooling, rewarming, etc. The calculated in-line PaO 2 and PaCO 2 should meet the acceptable targets to be reliable throughout CPB as a CILBGM device. Limitations
The findings in this study are limited to the FX05 oxygenator, and the formulas and recommendations provided are specific to this model. We observed similar patterns of error with the FX15 oxygenator, indicating the need for further investigation. Additionally, adult-size oxygenators may warrant similar investigation to understand and mitigate potential errors. iSTAT is a POC blood analyzer, which has demonstrated varying levels of deviation in PO 2 measurement compared to benchtop or laboratory gas analyzers, depending on the study and patient populations. While most studies indicate good correlations or minimal deviations [ 25-28 ], there is a notable exception in a study involving lung donors, which revealed relatively large deviations [ 29 ].
Acknowledgments
We express our gratitude to Molly Dreher from CHOP for the critical review of the manuscript. Additionally, we acknowledge the use of CHOP GPT( chat. chop. edu) solely for proofreading purposes.
Funding
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflicts of interest
The authors declare no conflicts of interest. However, we communicated our findings with Spectrum Medical prior to the submission of the manuscript.
Data availability statement
The data supporting the findings of this study are not publicly available. It may be available upon request to the corresponding author.