Impact of Regional Cerebral Perfusion on Outcomes Among Neonates Undergoing Norwood Operation.

Migally K, Rettiganti M, Gossett JM, Reemtsen B, Gupta P.

World J Pediatr Congenit Heart Surg. 2019 May;10(3):261-267. doi: 10.1177/2150135118825274.

PMID: 31084315

Similar articles

Select item 31084314


Take Home Points:

  • The use of regional cerebral perfusion (RCP) during the Norwood surgery is a technique utilized by many (but not all) surgeons to improve cerebral protection to the neonatal brain during aortic arch reconstruction.
  • This study utilized the public use data from the Pediatric Heart Network Single Ventricle Reconstruction trial to evaluate outcomes whether RCP affected outcomes such as mortality, need for heart transplant, prolonged mechanical ventilation and prolonged hospital stay … but not neurologic outcomes. And the dataset used to come to these conclusions was not designed to evaluate RCP vs. DHCA.

 Commentary from Dr. Timothy Pirolli (Dallas), section editor of Congenital Heart Surgery Journal Watch:  The traditional method of protecting the brain during aortic arch surgery has been deep hypothermic circulatory arrest (DHCA). However, in recent years, an increasing number of surgeons have been using regional cerebral perfusion (or selective antegrade cerebral perfusion) during arch surgeries to maintain some perfusion to the brain for a hypothetical improvement in outcomes, especially neurologic outcomes. The Single Ventricle Trial was conducted between 2005 and 2009 at 15 North American centers and was designed as a randomized controlled trial to evaluate modified BT shunts and Sano RV-PA conduits during the Norwood procedure for hypoplastic left heart syndrome.  The dataset from this study was made available for public use in 2013.


This present study was designed by researchers at the University of Arkansas to use this public dataset to evaluate whether patients who received RCP (as opposed to DHCA) at the time of their Norwood had different outcomes with respect to mortality, need for heart transplant, prolonged mechanical ventilation and prolonged hospital stay. These endpoints were evaluated between the two study groups (RCP vs. DHCA) from the initial study (Sano vs. BT shunt). However, the study was not designed to evaluate the difference between DHCA and RCP. This is clear because the evaluation of neurologic outcomes is not one of the primary endpoints of this present study (even though it is the main reason why surgeons may employ RCP vs. DHCA). The researchers also examined a host of other factors to determine the need for heart transplant or mortality, but only a few were statistically significant and those that were significant were not surprising. (Table 1).


Of the 549 patients qualified for inclusion in this study, 252 (45.9%) received RCP during their Norwood procedure. The information in Table 2 shows the variables that were related to the use of RCP that were statistically significant. Essentially, RCP was associated with longer CPB times, increased use of ultrafiltration and open chest after Norwood. It also showed that surgeons performing ≤ 10 Norwoods/year were more likely to use RCP.  Table 3 shows that these longer surgeries and the use of RCP had no effect on the study outcomes, which is the key point of this study. Figure 2 depicts the increasing effect no mortality of increasing RCP and DHCA times.


What do we learn from this study?  Well, when it comes to evaluating the principal difference between why a surgeon uses RCP and DHCA, which is neurologic outcomes, we learn nothing.  The “secondary” outcomes that are evaluated here show there are no differences between the use of RCP and DHCA from this dataset. The authors also emphasized the finding that surgeons who do fewer Norwoods are more likely to perform RCP during their surgeries. As a surgeon, this makes perfect sense since the cases (should) take you longer if you do fewer of them. Also surgeons that are performing the bulk of the Norwood procedure at any center in this study are likely to be at a more senior level in their career, faster, and (in general) used to performing the surgery under DHCA.  So this finding is little more than an unsurprising footnote. The inherent bias of some (especially experienced) surgeons to use DHCA vs. RCP really undermines the entire point of this study. There is no good way to parse out this factor from this dataset.


Is there really a difference in the selected outcomes in this study?  No, not from what this dataset says. But the study was not randomized (or even powered) to look at RCP vs. DHCA, so it is unclear how valid the researchers’ findings are.  And, again, there is zero mention of neurologic outcomes (not even stroke) which also is a major flaw. To break the results down to their core, we learn that during that 4 year period of the SVR trial there was no major difference between the chosen outcomes between Norwoods with RCP vs DHCA at these 15 institutions. And this fact may, or may not, be applicable to present-day practices. As usual, a better trial is needed to answer this question.

Tables and Figures: