Prediction of postnatal circulation in pulmonary atresia/critical stenosis with intact ventricular septum: systematic review and external validation of models.

Categories:


Prediction of postnatal circulation in pulmonary atresia/critical stenosis with intact ventricular septum: systematic review and external validation of models.

Villalaín C, Moon-Grady AJ, Herberg U, Strainic J, Cohen JL, Shah A, Levi DS, Gómez-Montes E, Herraiz I, Galindo A.Ultrasound Obstet Gynecol. 2023 Jul;62(1):14-22. doi: 10.1002/uog.26176. Epub 2023 Jun 5.PMID: 36776132 

Commentary from Dr. Anna Tsirka (Hartford, CT, USA), section editor of Pediatric and Fetal Cardiology Journal Watch

Summary: This systematic review of the literature evaluated studies that aimed to predict the presence of biventricular (BV) versus univentricular (UV) circulation in fetuses with PA/IVS or critical PS/IVS. They concluded that there is great heterogeneity in predictive criteria among the different studies with large variations in outcomes. Multivariate models that include TV z score, RV/LV length ratio, TV/MV ratio, degree of TR and presence of ventricular coronary connections have good specificity but underestimate the outcome of UV circulation. Given these findings, fetal intervention may need to be limited to those cases where UV circulation is near certain, till better discrimination can be achieves with larger multi center studies.

Introduction

Pulmonary atresia (PA)- critical pulmonary stenosis (CS) with IVS is a disease with variable severity, and diverse outcomes. RV dependent coronary circulation (RVDCC) may complicate management. Physiologic outcomes include biventricular circulation, 1.5 or single ventricle (UV) physiology. Predicting need for UV palliation remains challenging to date. Accurate prediction of the need for single ventricle palliation would allow for improved fetal counseling and is crucial in determining referrals for potential fetal intervention.

Methods

This study is a systematic review of the literature regarding prenatal factors that may predict UV circulation in fetuses  with PA or CS/IVS. Cohort and case control studies published between 2000 and 2020 with a minimum sample size of 10 patients that were diagnosed between 20-28 weeks of gestation were included in the analysis. Those fetuses were singleton gestations and did not undergo fetal intervention.

Standard methods of quality evaluation of the studies and the data collection were utilized.

Results

321 manuscripts were initially identified and after critical review

11 studies were included in the analysis. The sample size of those studies ranged between 15 and 38. Outcomes evaluated were UV outcome, BV outcome, non BV outcome and RVDCC. 9 studies reported on both PA and CS/IVS, while two studies only reported on PA/IVS. Because these studies were non uniform  in their outcome and analysis, the sample was validated using 71 cases identified from the International Fetal Cardiac Intervention Registry (IFCIR). These were fetuses who did not receive an intervention. Table 2 reveals their characteristics based on postnatal circulation.

About a third of the cases had UV circulation. Those fetuses had lower TV diameter z score, lower PV annulus z score, lower TV/MV  and RV/LV ratio. The UV circulation group were less likely to have TR and more likely to have ventriculo-coronary connections.

Predictive models were developed based on the IFCIR sample for the 9 studies reviewed (2 studies evaluated parameters that could not be validated). The most reported predictive parameters of UV circulation were TV annulus z score <-3 , TV/MV ratio <0.7, RV/LV length ratio< 0.6. The most highly predictive parameter was the presence of ventricle coronary connections, as shown below in Figure 2. The duration of RV inflow did not appear to be a significant predictor.

Multiparametric models were more specific in their estimations, however in general they underestimated the risk of UV outcome.

Discussion:

The main finding of this systematic review was the lack of standardized criteria for prognostic models with significant heterogeneity of results. There were very large variations in the rate of UV outcome, from 25 to 70%, probably reflecting institutional preferences, selection bias and different termination of pregnancy policies. 

The use of multi parametric models has shown AUC of over 80% for prediction of UV circulation, nevertheless, these models still underestimate the risk for UV circulation.

The authors conclude that large prospective studies are needed to standardize prognostic criteria for UV, as those fetuses would benefit the most from fetal intervention.

Pediatric Cardiac Professionals