Declining Incidence of Postoperative Neonatal Brain Injury in Congenital Heart Disease.
Peyvandi S, Xu D, Barkovich AJ, Gano D, Chau V, Reddy VM, Selvanathan T, Guo T, Gaynor JW, Seed M, Miller SP, McQuillen P.J Am Coll Cardiol. 2023 Jan 24;81(3):253-266. doi: 10.1016/j.jacc.2022.10.029.PMID: 36653093
Take home points:
- Over a 20-year period, patients with complex congenital heart disease (CHD), the prevalence of preoperative white matter injury (WMI) remained stable, but postoperative WMI declined
- Improved postoperative systolic, mean, and diastolic blood pressure in the first 24 hours postoperatively was the most important clinical risk factor in reducing postoperative WMI
- Longitudinal studies will be necessary to determine if this will lead to improved neurodevelopmental outcomes

Commentary from Dr. Jared Hershenson (Greater Washington DC), section editor of Pediatric Cardiology Journal Watch:
As surgical mortality has declined and patients are living longer with even the most complex CHD, the focus has shifted to other morbidities, in particular, neurodevelopmental issues. This includes attention, processing speed, memory, impulsivity, executive function, and decision-making problems amongst others. This has personal and communal impact as there is an association with lower educational attainment and less employment opportunities. Preoperative brain injury (primarily WMI) is likely due to prenatal factors leading to decreased cerebral oxygen delivery, genetic factors, and possibly the time between birth and initial surgery. It has been reported in 10-35% of patients with neonatal complex/cyanotic CHD. Postoperative WMI may be more modifiable. Prevalence of postoperative WMI has ranged from 33-75% of patients. There have been many changes over the years to try to decrease risks, including improved bypass techniques and support times, improved hematocrit, and likely better post-operative ICU care. This study aimed to describe the temporal trends of neonatal brain injury over a 20-year period between 2001-2021 using pre- and post-operative brain MRI, hypothesizing that improved clinical care may decrease the rate of brain injury.
Most patients enrolled were diagnosed with d-TGA or single ventricle physiology (SVP) and expected to require neonatal surgery. Pre-operative MRI was performed as soon as a patient was stable enough to be transported to the MRI scanner. Postoperative MRI were done prior to discharge, with an average of 15 days between studies. 270 patients were enrolled prospectively. 246 patients had a preoperative MRI and 220 had a postoperative MRI. The 20-year period was divided into four 5-year epochs. Primary outcome was presence of WMI, and secondary outcomes included other forms of brain injury (eg. Stroke). The absolute frequency of preoperative WMI or stroke did not change over the study period. However, postoperative WMI decreased and was significantly lower in epoch 4 compared to epoch 1 with an overall decline of ~ 18.7% and a prevalence of ~11% in the most recent epoch. See central illustration. Patients with d-TGA and SVP were stratified and risk factors for WMI studied separately. In both groups, there was no difference in timing of surgery, but both groups had longer bypass and cross-clamp times over the study period. See Tables 4 and 5. In the SVP group, systolic, mean, and diastolic blood pressures were significantly higher in Epoch 4; in the d-TGA group, mean and diastolic BPs were higher. See Figure 2. Some changes in inotropic support were noted, primarily increased use of epinephrine and less milrinone.
Since the study was observational, causal inference is limited; however, the rates of early postoperative hypotension were significantly lower by epoch 4, suggesting that changes in postoperative care and better cerebral perfusion may have an important impact on preventing WMI. However, this is in contrast to previous data suggesting elevated SVR and low cardiac output with poor outcomes. The balance between optimal cerebral perfusion and preventing low cardiac output can be tenuous in the critically ill infant and may require new strategies. Additionally, the rates of preoperative WMI did not change, so further understanding of this process to help prevent prenatal WMI is necessary. Whether the lower postoperative WMI will be associated with true improvement in neurodevelopmental testing remains to be seen.
An accompanying editorial written by Dr. Jane Newburger eloquently places this study in the context of the overall canon of research on ND in CHD. Other studies have not shown improved ND outcomes even with lower postoperative WMI, with other factors (length of stay and socioeconomic status) having a greater impact. Additionally, preoperative WMI may have a greater impact on developmental scores than postoperative WMI. Further longitudinal study and multicenter analysis will be necessary in order to develop evidence-based pathways and interventions to improve ND outcomes.




