K . Kohlsaat et al .: J Extra Corpor Technol 2023 , 55 , 175 – 184 177
MI ), a custom tubing pack ( LivaNova PLC , London , UK ) and an HPH400 hemoconcentrator ( Medivators Inc . Minneapolis , MN ). The circuit prime volume , inclusive of the 70 mL active ultrafiltration circuit and 75 mL in the reservoir was 285 mL for circuits with a 3 / 16 00 arterial boot and 300 mL for circuits with a 1 / 4 00 arterial boot . All patients received 30 mg / kg of methylprednisolone at the onset of bypass per our institutional standard as well as antibiotic coverage . A natural grouping of patients that received PBUF versus those that did not occur as PBUF was performed at the perfusionist ’ s discretion . The SP control group included all patients who received a standard blood prime . The PBUF group received a standard blood prime followed by the PBUF procedure , which was performed before the initiation of CPB .
A standard blood prime consisted of the perfusionist flushing the circuit with carbon dioxide before proceeding to prime the circuit with Plasma-Lyte A 7.4 ( Baxter Healthcare , Deerfield , IL ) heparinized to 3 IU / mL . Then , the majority portion of the crystalloid in the circuit was displaced with blood-bankreconstituted whole blood ( 1 unit of supernatant-removed-RBCs mixed with one unit of plasma ), also heparinized to 3 IU / mL . Sodium bicarbonate ( 6 [± 1 ] mEq ) and calcium gluconate ( 600 [± 100 ] mg ) were then added to normalize pH and ionized calcium levels , respectively . Blood gas analysis was performed on the prime blood before patient use .
The PBUF prime group received a standard blood prime followed by the PBUF procedure . The PBUF process included the addition of 200 mL of heparinized Plasma-Lyte A and 80 mL of 0.45 % saline to the cardiotomy venous reservoir . This excess volume was then filtered off , bringing the reservoir level back to 75 mL . Additional sodium bicarbonate and calcium gluconate were added to normalize values and blood gas analysis confirmed the results before patient use . Please see the Limitations section for the post-study updated protocol that includes PBUF of the entire unit of reconstituted blood with sequestering of volume not needed in the prime for transfusion later during the case .
Statistical analysis
Categorical variables are summarized with frequencies and percentages , and continuous variables with ranges or interquartile ranges as noted . Patient characteristics , clinical variables , and outcomes were compared for patients receiving and not receiving PBUF using Fisher ’ s exact test or the Wilcoxon rank sum test . Summaries were also stratified by case complexity . Analyses were conducted in Stata version 16 ( StataCorp , College Station , TX ).
Results Baseline patient characteristics
Between August 2017 and September 2021 , 704 patients under 1 year of age underwent open-heart surgery for a congenital heart defect where PBUF was utilized . In contrast , 417 patients under 1 year of age during this period did not undergo PBUF . Baseline patient characteristics are described in Table 1 . While there are no statistically significant differences in these baseline patient characteristics , it is notable that the median age at surgery was 3 months and 47 % of patients were female in both cohorts . Preoperative risk factors were present in 308 ( 44 %) of the PBUF cohort and 163 ( 39 %) of the SP cohort ( Table 1 ). At the recorded index surgery , 149 ( 21 %) of the PBUF cohort and 97 ( 23 %) of the SP cohort had previously undergone a prior cardiac operation with or without CPB ( Table 1 ). The median cardiopulmonary bypass times ( CPBT ) were similar with a CPBT of 141 [ IQR 103 – 197 ] mininthe SP group and 145 [ IQR 104 – 198 ] min in the PBUF group . Cross-clamp time ( CCT ) was also similar between groups at 91 [ IQR 54 – 128 ] min in the SP group and 93 [ IQR 60 – 125 ] min in the PBUF group .
Primary outcome
The SP and PBUF groups had similar prime values after each received a standard blood prime ( Table 2 ), as expected . The PBUF cohort had statistically significant shifts towards physiologic values for sodium , glucose , potassium , lactate , and osmolality after that procedure was added ( Tables 3a and 3b ). While osmolality in the two cohorts was similar at the initial priming of the CPB circuit ( Table 2 ), significant differences were evident after the completion of PBUF ( Tables 3a and 3b ), which was anticipated . The first values on CPB continued to show the same trend of statistically significant differences for pH , sodium , glucose , lactate ( Figure 1 ), and osmolality with the PBUF group having more physiologic values ( Table 4 ). The last CPB values showed that most differences had equilibrated with the exception of glucose ( Table 5 , Figure 1 ). The pH of the SP cohort displayed a wider range of values throughout the measured time points compared to the PBUF cohort . Normality in a majority of measured blood parameters was evident by the end of CPB ( Table 5 ).
Secondary outcomes
There were no statistically significant differences in clinical outcomes between the two cohorts for time to extubation , CICU LOS , unplanned reoperation , re-exploration for bleeding , mediastinitis , CNS complications , use of ECMO , pacemaker requirement , and mortality ( Table 6 ). There were also no statistically significant differences in postoperative hospital LOS between the groups ( Figure 2 ). Intraoperatively , the median dose of phenylephrine used while on bypass was 60 mcg for both cohorts ( Table 7 ). The ultrafiltration total ( modified ultrafiltration + conventional ultrafiltration ) was no different between the two cohorts , and postoperative urine output was similar in the first 24 h ( Table 7 ). Therewerenosignificant differences in the total dose / maximal infusion rate of inotropes perioperatively ( Table 7 ). Discussion
The process of blood priming CPB circuits is not standardized across institutions providing cardiopulmonary bypass for congenital heart surgery patients . Bank blood varies in degree