Priming Cardiopulmonary Bypass in Pediatric Surgery: Reply

We thank Dr. Schmartz et al.¹  for their interest in our randomized clinical trial comparing fresh frozen plasma versus crystalloid for priming cardiopulmonary bypass (CPB) circuit in pediatric surgery.²  The authors are right in noting that our studied population does not belong to the group of patients at the highest risk of postoperative bleeding, i.e. neonates and small infants. However, up to now, no single double-blind study has evaluated this issue in older infants and small children. From an ethical point of view this question needed to be analyzed in this population before conducting a similar study in a very high-risk population. As a matter of fact, the recent published guidelines on patient blood management in pediatrics undergoing cardiac surgery concluded that in infants and children, no recommendations can be made regarding the addition of fresh frozen plasma to the CPB because of the absence of evidence in the matter.³ 

Dr. Schmartz et al. ask why we did not use conventional and/or modified ultrafiltration that both help in preserving the coagulation factors in the autologous blood returned to the children. Ultrafiltration techniques are routinely performed in many centers performing congenital heart surgery. These techniques, especially modified ultrafiltration, have shown many advantages including reduced positive fluid balance and reduced transfusion of allogeneic blood products.⁴  However, studies showing benefits from modified ultrafiltration are all greater than 10 yr old,⁴  and it has been shown that modified ultrafiltration is no longer necessary when small circuit sizes and priming volumes are used.⁵  In addition, modified ultrafiltration can result in significant complications.⁵  We routinely use normothermic CPB with warm blood cardioplegia. We therefore do not apply modified ultrafiltration and use conventional ultrafiltration only sporadically. At this stage the administration of cell salvage is the best option in our population.

Dr. Schmartz et al. ask whether cell-salvage blood was discarded in some patients. This question refers to our statement that cell-salvage blood at the end of procedure could be administered when necessary. We of course did not discard this autologous blood, but the administration took place either in the operating theatre or in the intensive care unit whenever it was necessary.

Dr. Schmartz et al. point out that there is no clear evidence from the literature that the balanced crystalloid as used in our trial is as effective and as safe as balanced hydroxyethylstarch solutions. We have highlighted in our manuscript why we opted not to use hydroxyethylstarch solutions. A meta-analysis of randomized controlled trials in pediatrics has moreover shown that these solutions significantly decrease platelet count as compared with other fluids.⁶  This may result in increased postoperative bleeding and would have adversely influenced our results. Moreover, comparing fresh frozen plasma to a solution that is not the standard of care in our hospital would not have been approved by our local ethical committee.

Dr. Schmartz et al. note that our perioperative transfusion trigger for red blood cells is higher than what is currently recommended.³  These current guidelines are based on two randomized studies.^7,8  Both studies were conducted with moderate to deep hypothermic CPB, and the randomization was aimed to obtain a hematocrit level of 20% versus 30%⁷  or 25% versus 35%⁸  at the onset of low-flow CPB. The question is whether the results of these studies can be extrapolated to our normothermic CPB management. Interestingly, the same group combined the data from these two trials to investigate the relationship between continuous hematocrit levels and postoperative outcomes.⁹  They concluded that a hematocrit level at the onset of low-flow CPB of approximately 24% or higher is associated with higher Psychomotor Developmental Index scores and reduced lactate levels. Kussman et al.¹⁰  analyzed the regional cerebral oximetry data of the trial that compared a hematocrit of 25% with a hematocrit of 35%.⁸  They showed that perioperative periods of decreased cerebral oxygen delivery, as indicated by cerebral oximetry, are associated with lower Psychomotor Developmental Index scores and greater risk of hemosiderin on brain magnetic resonance imaging. We routinely use intraoperative regional cerebral oximetry and treat any cerebral oxygen desaturation based on established algorithms. We therefore have higher transfusion criteria.

Dr. Schmartz et al. assert that we should have included a broader population by stratifying the studied population according to the size of CPB circuit. As stated earlier, we aimed to investigate the CPB prime strategy in older infants before conducting the same study in neonates and younger infants. In conclusion, we agree that further well conducted studies are required in higher-risk patients. Meanwhile, our group has answered an important clinical question in a specific category of pediatric patients undergoing open heart surgery.

Supported by a research grant from the Belgian Society of Anesthesiology and Reanimation and the Department of Anesthesiology of Cliniques Universitaires Saint Luc, Brussels, Belgium.

The authors declare no competing interests.