Hi. Good afternoon. Good evening now. My name is Monique Gardner. As said, I am a pediatric cardiac intensivist at the Children's Hospital of Philadelphia, and it brings me a lot of pleasure and gratitude to be able to share this work today on behalf of our collaborative group, including, again, recognizing Dr. Nader Yeha, who's heard his name several times this afternoon already, for his mentorship. I have no relevant financial disclosures nor in our group, but I am grateful for grant support to perform biomarker analysis from the Big Hearts to Little Hearts Foundation. One in 100 live births have congenital heart disease of some form, and a quarter of those with congenital heart disease are neonates that require cardiopulmonary bypass in the first few weeks of life. And these children, frequently born and admitted quickly to an ICU or shortly after birth, have worse outcomes compared to children who need cardiopulmonary bypass surgery later in their lives. They have higher rates of mortality, they have prolonged intensive care unit lengths of stay, and they're at risk even after discharge for multiple readmissions back to the intensive care unit or to the hospital. Studies for risk factors that identify those with worse outcomes have mostly been limited to non-modifiable factors, including demographics or the type of cardiac disease that neonates have. So I wonder, as an intensivist myself, what can we do more to identify these patients? There has been some work within these more surgical groups to identify surgery that patients need and their mortality after as one of the main outcomes to focus on. And this is something called the STAT category, which is a numerical system that measures the complexity of a surgery, STAT 1 category being the lowest complexity with the lowest rates of mortality, such as an atrial septal defect repair. And STAT 5, the highest category, has the highest associated mortality and is the most complex operations, such as the Norwood operation for single ventricle heart disease. And as you can see here, with time over the X-axis and discharge mortality on the Y-axis, mortality is decreasing, but there's a disparate outcome between the most complex surgeries with STAT 5 and STAT 4 at 12% and 6% compared to the more simple straightforward operations of STAT 1 to 3. Most neonates require the more complex operations because of their complex congenital heart disease. So how can we better assess the risk for neonatal cardiopulmonary bypass and poor outcomes? Well, I'm going to hint that in the biomarker group here, we're going to talk a little bit about that. So can we identify patients at risk for poor outcomes when the clinical risk seems the same? When we have these 10 patients that all have the same heart disease and similar demographic factors, they're all babies with hypoplastic left heart syndrome or transposition of the great arteries, it's hard to determine which ones of these are going to be the ones with prolonged intubation, requiring renal replacement therapy, or the one that's at higher risk of mortality. I don't need to tell this group that plasma biomarkers can offer patient-specific and markers of end-organ injury. Preoperative biomarkers have been studied within the congenital heart disease population and ST2, galactin 3, and NT pro BNP have been found to be associated with mortality, prolonged length of stay, and readmission. Most of the work comes out of this one study of 162 patients. But this is a heterogeneous group. You can see that less than 10% of these studies had neonates, and about 15% were the highest stat categories. So this work has not really focused on the patients most at risk of mortality or poor outcome. And we wanted to focus on this population more. So to do that, we wanted to focus on the measurement and collection of pre-cardiopulmonary biomarkers to see if they could be associated with poor post-operative outcome. We performed this as a single-center prospective observational study, and we enrolled neonates with congenital heart disease requiring cardiopulmonary bypass at less than 30 days of age. We measured preoperative biomarkers, 28 of them, with other studies had been important in markers of end-organ injury, inflammation, and coagulation, and either cardiac disease or neonatal disease previously. As important as our measure of the biomarkers, we wanted to focus on a primary outcome, one that didn't just include mortality. As you can see, it's thankfully increasing, and in small cohorts, occurs relatively frequently. And those of us who work in the intensive care unit know that mortality is not the only bad thing that can happen to you while you're being cared for. So earlier this year, we published a new composite outcome that focuses on cardiac care for neonatal disease. And we call this composite outcome ICU-30. The composite outcome is when one of the three following things happens to the patient in the post-operative period after neonatal bypass, a 30-day post-operative mortality, a CICU's length of stay longer than 30 days, or a 30-day readmission back to the CICU, either from home or from within our step-down unit. Functionally, this is a binary categorization of days alive out of ICU, a measurement that's used more typically in adult studies, particularly in heart failure. And what we found is that this outcome occurs relatively frequently in about a third of our neonates who undergo cardiopulmonary bypass. It's relatively easy to collect. And also, it doesn't just mark badness in the first 30 days. Our ICU-30 composite outcome was associated with 6- and 12-month mortality, highlighting a patient population that, even if they survive past the first 30 days, are still at risk for mortality. We use this outcome as our primary outcome for this biomarker study. To date, we've enrolled in 65 subjects and have been able to evaluate 63 samples preoperatively. Similar to our previous cohorts within a larger retrospective study that I just showed, a third of our group had the poor ICU-30 composite outcome, while a majority did not. Of those who had the poor outcome, one patient died, 14 had a prolonged ICU length of stay, and 3 had an ICU readmission within 30 days. When comparing the groups and focusing on the ICU-30 cohort compared to the one that did not have the ICU-30 outcome, you can see there were no significant clinical differences between the groups. To our eye, preoperative risk was the same. Looking a little bit more at some of the operative and postoperative details, there was no difference at the age of operation. There was a difference in the stat category with the ICU-30 patients more likely to have a stat 5 operation, and the patients without the ICU-30 outcome having lower stat category operation. There was no difference in bypass time. There were higher rates of postoperative ECMO and delayed sternal closure, again reflecting complex postoperative courses in the patients with the ICU-30 outcome. And not surprisingly, the patients with the ICU-30 outcome had longer CICU lengths of stay as that was part of the events that could be included in the composite. Now going on to the biomarkers. Of our 28 biomarkers of interest, we found 7 biomarkers that were statistically significant in the patients that had the ICU-30 outcome compared to those without. With IL-6, CCL-12, TNF-alpha, ANG-2, GDF-15, and NGAL being statistically higher preoperatively in patients with the ICU-30 POR outcome, and ADAMS-T13 being statistically lower in patients with the ICU-30 outcome. Going one step further, after standardizing and controlling for stat category, we found four preoperative biomarkers that were still significantly associated with the ICU-30 outcome, including TNF-alpha, IL-6, CCL-2, and NGAL. In summary, when controlling for stat category, four preoperative biomarkers were found to be associated with POR outcome for neonates with cardiopulmonary bypass. They focus on three biomarkers of systemic inflammation, and interestingly, one biomarker of renal injury, NGAL. This is still a small sample size. We have limitations associated with not being able to control for multiple factors at this time. However, as we look forward, we are continuing to enroll a larger cohort that includes both pre- and post-bypass samples that will allow for more advanced statistical techniques, including classification and regression tree analysis, which can combine both clinical factors as well as biomarker levels to start to sub-phenotype the neonatal cardiopulmonary bypass patients. We also hope to apply these findings to a multi-center approach to see if our findings still remain across different practices at centers. And again, to emphasize the work that's been done to date in biomarkers, really we hope to identify risk factors and understand mechanistic pathways for future prognostication and targeted therapeutic interventions in this vulnerable population. Thank you for your time. I'll gladly take questions.

Dr. Monique Gardner

pediatric cardiac intensivist

neonatal cardiopulmonary bypass surgery

preoperative biomarkers

poor outcomes