220 G . S . Matte et al .: J Extra Corpor Technol 2024 , 56 , 216 – 224
A Y-connection would provide the best flow capacity while a traditional venous limb stopcock may work but could be limiting depending on the level of device failure and required flow to the secondary oxygenator to meet overall oxygen transfer needs .
Oxygenator-reservoir failure with concurrent concern for system blood clots
Clotting within the circuit during CPB is a significant concern since emboli can cause neurologic deficits and stroke [ 14 ]. Both intervention options listed in Figure 1 include definitive device change-outs with either the oxygenator reservoir being replaced or the entire CPB circuit . These interventions require relative expediency in action and a period of time off support , which may place the patient at risk if native cardiopulmonary function is not available . An entire circuit change-out exposes the patient to significantly more circuit surface area that may affect the systemic inflammatory response . It may also increase transfusion requirements depending on starting hematocrit , patient circulating blood volume , circuit size , and the ability to recover red cell mass from the original circuit . Changes in circulating medication levels also need to be considered , with redosing for antibiotics and anesthetics likely warranted .
Ultimately , intervention options for confirmed oxygenatorreservoir issues will vary depending on pre-bypass circuit configurations , team preference , and institutional experience .
Programmatic review of oxygenator device issues
Boston Children ’ s Hospital has an internal non-routine event reporting system where all cases of issues during cardiopulmonary bypass and perfusion services are reported [ 15 ]. These events are discussed at a quarterly multidisciplinary conference led by the perfusion team as part of a quality assurance and improvement initiative . Over the past 15 years , we have documented seven instances of suspected oxygenator failure or other oxygenator performance-related issues . Two cases required no circuit intervention during or after bypass as the PaO 2 was deemed sufficient at > 100 mmHg on 100 % oxygen and there was a low chance of resuming CPB . In two cases , the perfusion team changed the oxygenator only after normally weaned from bypass , in case an additional period of CPB was required . One case was treated with an oxygenator-only change-out during a surgically-planned period of circulatory arrest at 18 ° C . The sixth incident involved concern for visible clots in the post-oxygenator / integrated ALF part of the bypass circuit that progressively developed during CPB . The care team elected to perform an entire circuit change-out for that case ( later follow-up revealed the area of concern to be aggregated platelets and not fibrin ). The seventh and most recent case in our experience showed oxygenator failure during the crossclamp period at approximately 6.5 hours on CPB . The perfusionists successfully employed an arterial piggyback method to support the primary oxygenator for that case . There were no instances of emergent , on-bypass , oxygenator change-out in the past 15 years at our institution , which included over 13,000 cardiopulmonary bypass cases .
Recent experience
The seventh reported case of oxygenator failure in our experience , mentioned above , included a 452-minute bypass run on a patient with compromised pulmonary function . That 7 kg patient circuit had a 3 / 16 00 arterial and post-oxygenator recirculation limb , as well as a 1 = 4 00 venous limb on a Stockert S5 heart-lung machine ( LivaNova PLC , London , UK ). It was understood preoperatively that there was a high likelihood that the patient would be transferred to extracorporeal membrane oxygenation ( ECMO ) post-cardiotomy before transfer to the intensive care unit ( ICU ). Two perfusionists were available to support the primary perfusionist while assessing the slowly declining PaO 2 . The surgeon and anesthesiologist were informed of the developing issue and a second Terumo CAPIOX FX-05 oxygenator ( Terumo Cardiovascular Systems , Elkton , MD ) was primed in the pump room with heparinized ( 3 IU / mL ) Plasma-Lyte A 7.4 ( Baxter Healthcare Corporation , Deerfield , IL ). The oxygenator bundle was prepared with 3 / 16 00 inlet and outlet connectors to facilitate a time-efficient oxygenator change-out , as regularly practiced by the perfusion team . The care team verified that the oxygenator was indeed failing per written institutional guidance that has the team rule out numerous potential contributing factors ( Table 1 ) [ 5 ]. Over the course of 37 min ( CPB time 397 – 434 min ), the PaO 2 on 100 % oxygen declined from 250 mmHg to 98 mmHg . This occurred despite serial increases in sweep flow rate , an oxygen source change to a standalone tank , and a Terumo FX-05-specific oxygenator “ wet-out ” procedure of increasing the sweep rate to 5 LPM for only 10 s ( without repeating per the instructions for use ) [ 16 ]. The team decided that since the patient would separate from the bypass in another 15 – 20 min and likely be transitioned to ECMO , an arterial piggyback would be sufficient to support oxygenation . The primed secondary oxygenator was inserted into the standard 3 / 16 00 arterial recirculation line that is included on all of our institution ’ s neonatal and infant circuits . The arterial pump flow was set at the primary device ’ s maximum recommended flow of 1500 ml / min with partial clamping of the post-primary oxygenator recirculation line that led to the secondary oxygenator . This resulted in a patient flow of 860 mL / min ( cardiac index of 2.3 mL / min / m 2 ) and secondary oxygenator flow of 640 mL / min , and we thought this was a reasonable balance between systemic flow and secondary oxygenator flow . An arterial flow probe ( Transonic Systems , Inc ., Ithaca , NY ) was used to guide effective patient flow . The in-line Terumo CDI monitor ( Terumo Cardiovascular Systems , Elkton , MD ) indicated a rising PaO 2 , which settled at > 200 mmHg with 100 % oxygen sweep gas running to both the primary and secondary oxygenator . It was of principal consideration that the clinical care team timed the arterial piggyback intervention to prevent hypoxemia and a subsequent decrease in venous oxygen saturation ( SvO 2 ), cerebral NIRS values and indexed oxygen delivery . Of note , the nadir SvO 2 , cerebral NIRS ( left / right ) and indexed oxygen delivery values during the progressive oxygenator failure period were 70 %, 85 %/ 85 %, and 409 mL / min / m 2 , respectively . The patient was