44 M.-H. Lee and T. Rosenthal: J Extra Corpor Technol 2026, 58, 43 – 50
CLIA guideline) at the 1st blood gas series after the initiation of CPB with in vitro calibration using its own calibration gas and became more accurate and precise at the 2nd blood gas series with the first in vivo calibration [ 2 – 4, 8 ]. Without in vitro calibration, PaO 2 and PaCO 2 of the CDI system 500 at the 1st blood gas series were unacceptable. However, after the first in vivo calibration, they became acceptable at the 2nd blood gas series [ 14 ]. A study compared CDI system 500, B-Capta, and System M4( Spectrum Medical, Gloucester, United Kingdom) for PaO 2 and other parameters. Unlike the CDI system 500 and B-Capta, System M4 does not make direct contact with arterial blood and thus cannot measure but calculate PaO 2. It showed that while CDI system 500 and B-Capta provided acceptable PaO 2 at the 1st and 2nd blood gas series, System M4 exceeded the acceptable targets at both series [ 4 ].
Similar to System M4, the Quantum perfusion system with Quantum Ventilation 2(“ Quantum System” hereafter; Spectrum Medical, Gloucester, United Kingdom) provides continuous in-line PaO 2 and PaCO 2 without direct contact with arterial blood. Thus, the Quantum System does not measure PaO 2 or PaCO 2. To circumvent this, Quantum System developed formulas to calculate PaO 2 and PaCO 2 with its own non-invasive sensors and embedded software. A small portion of expired gas of the oxygenator is constantly drawn from the active waste gas scavenging system to the sensor in Quantum Ventilation 2 to measure the fraction of CO 2 in the expired gas( FeCO 2). In-line PaCO 2 is calculated with their proprietary formula, which is heavily dependent on FeCO 2. However, the calculated in-line PaO 2 of Quantum System is complex and is mainly dependent on FiO 2 and modified by other non-invasive measurements such as flow, SaO 2, SvO 2, FeCO 2, hematocrit, temperature, and other variables. Several coefficient constants in the formula were determined by the best fitting of experimental data with at least two adult-size oxygenators( Personal Communication).
We have been using Quantum System with the FX05 oxygenator( Terumo Medical Corporation, Somerset, NJ) since August 2019 and observed that the calculated in-line PaO 2 at the 1st blood gas series after the initiation of CPB has been significantly higher than PaO 2 from the point-of-care( POC) blood analyzer( i-STAT, Abbott, Abbott Park, IL). We also noted that the calculated in-line PaO 2 seems to drift upward during the rewarming and rewarmed periods. Similarly, the calculated in-line PaCO 2 seems to drift upward during the rewarming period as well. These deviations prompted us to undertake a quality improvement initiative, focusing on identifying the degree of errors in both calculated in-line PaO 2 and PaCO 2 during the cooling and cooled periods, as well as during the rewarming and rewarmed periods. Since these are not the measured values, we expected a certain degree of errors, which should still be acceptable for safe practice. We were also interested in investigating whether any possible patterns or correlations exist between the errors and other factors, such as patient weight and temperature.
Methods Patient population
We reviewed the EPIC( EPIC Systems, Verona, WI) electronic records of anesthesia and perfusion for 133 consecutive patients who underwent cardiac surgery with CPB using the Quantum System with the FX05 oxygenator at the Children’ s Hospital of Philadelphia( CHOP) from January 3, 2023, to June 6, 2023. Quantum System, equipped with non-invasive sensors and embedded software, served as the CILBGM device. Patients were excluded based on criteria such as missing data, which is necessary to perform the analysis, heart or lung transplants, LVAD insertion, absence or interruption of cooling or rewarming, or if deceased. Fifty-two patients were excluded, resulting in 81 patients to be analyzed. Out of 81 patients, 23 were weight 4 kg, 43 were 4 kg < weight 8 kg, and 15 were 8 kg < weight 14 kg. Twenty-five patients had nadir nasopharyngeal temperature( Nadir T) during CPB 32 ° C, 35 were 32 ° C > NadirT 24 ° C, and 21 were Nadir T < 24 ° C during CPB. Average CPB time was 136.8 ± 57.3 min and average cross-clamp time was 86.9 ± 47.2 min.
Blood gas analysis and data collection
During CPB, pH-stat was used for blood gas management during the cooling and cooled periods and alpha-stat during the rewarming and rewarmed periods [ 15 – 17 ]. During the cooling and cooled periods, pH-stat results of blood gas analysis are recorded in our EPIC electronic record. During the rewarming and rewarmed periods, alpha-stat results are recorded. The cooling period indicates that the patient’ stemper- ature is actively decreasing toward the target temperature, while the cooled period indicates that the patient’ s temperature remained at the target temperature. The rewarming period indicates that patient temperature is actively increasing to reach the venous blood temperature of 36.5 ° C, and the rewarmed period indicates that patient temperature is maintained at the venous blood temperature of 36.5 ° C.
The sweep rate was between 0.5 and 1.5 LPM, which is relatively proportional to the patient’ s weight, with an infusion of CO 2 atarateof30 – 50 mL / min. CO 2 field flooding was a surgeon’ s preference. One of five surgeons routinely employed CO 2 field flooding.
After the initiation of CPB and cooling, the 1st blood gas series was performed after 2 – 5 min of steady state CPB by drawing arterial and venous blood samples from the CPB circuit while pressing the“ Capture All” key on the Quantum monitor at the same time. Blood gas analysis was performed with CD8 cartridges using iSTAT. After the blood gas analysis was completed, the“ Sync” key was pressed to recall the stored values. The first in vivo calibration was performed by replacing the stored values with the iSTAT results. After the first in vivo calibration, arterial blood samples were drawn every 30 min or when necessary for the blood gas analysis, and in vivo calibrations were performed with every blood gas analysis.
During the cooling and cooled period, data from the first three blood gas series, if available, were collected. Many cases had short cooling and cooled periods or proceeded to deep hypothermic circulatory arrest( DHCA). Data is not collected anymore when DHCA is initiated. Thus, we were able to collect the 1st blood gas series from 81 patients, the 2nd blood gas series from 68, and 3rd blood gas series from 42 during the cooling or cooled period( Table 1).