DISCUSSION
Our study shows that fetal echocardiography can accurately identify the origin of pulmonary vascular supply. Accuracy in defining the anatomy of the pulmonary vasculature and the source of pulmonary blood supply was 82.3% and 88.2%, respectively. Previously published studies report lower rates5,6. Zhoi J et al. and Naimi I et al. reported similar rates with us in recent studies7,8. The improvement seems to be related to the advanced ultrasound technology. Besides, we have been intensely searching for MAPCAs and the origin of pulmonary blood supply since the beginning of the study period. We have no false negative MAPCA-dependent PAVSDs in our study. Therefore, we suggest that cases with MAPCAs can be diagnosed with 100% sensitivity with a meticulous search, including sagittal and coronary planes of the aortic arch and descendent aorta. MAPCAs mostly originate from the descending aorta (Figure 4b,4c, Figure 5c). They can also arise from the aortic arch (Figure 4a, Figure 5a), subclavian arteries, brachiocephalic trunk, internal mammarian artery, and left coronary artery9.
We detected MAPCA originating from the coronary artery in one of the cases (Figure 3a). MPA and the confluence of central PAs were absent on fetal echocardiography. Postnatal angiography (Figure 3c,3d) and surgery (Figure 3e,3f) confirmed the prenatal diagnosis. According to the intraoperative assessment, MPA was absent. The LPA originated from patent DA. The RPA was supplied by a collateral from the left main coronary artery (LMCA) (Figure 3e,3f). There was no confluence of PAs. Coronary to pulmonary collaterals are rare and can be detected in 1.3%-10% of the cases with PAVSD10,11. To our knowledge, this was the first case in the literature diagnosed prenatally and confirmed surgically. It was expected that cardiac ischemia could have occurred due to increased coronary to pulmonary steal secondary to decreased pulmonary vascular resistance after birth. Thanks to the prenatal diagnosis, the newborn was immediately referred to the collaborated cardiovascular clinic on prostaglandin infusion and underwent early surgery to avoid coronary stealing. Therefore, we suggest that the aortic root should also be carefully searched in axial and longitudinal planes in PAVSD cases to detect any collaterals arising from coronary arteries. These newborns may not present cyanosis due to the high flow from coronary arteries to the lungs; however, they may develop acute cardiac failure based on cardiac ischemia.
The impact of MAPCAs on perinatal and postnatal outcomes is debatable in the literature3,7. MAPCA-dependent newborns can present cyanosis in case of narrow collaterals, or those with adequate collaterals may be acyanotic with subtle symptoms for months until stenosis of the collaterals occurs. Multiple, dilated collaterals may lead to pulmonary hypertension, congestive heart failure, or pulmonary parenchymal bleeding. MAPCA-dependent cases usually have non-confluent PAs with arborization abnormalities or no central PAs at all, while DA-supplied cases usually have confluent PAs with complete intrapulmonary arborization8. Consistently, in our study, except the one with DA arising from the 1stbranch of the aortic arch (Figure 6a, 6b), all DA-dependent cases had unifocal, confluent PAs. On the other hand, only 28.5 % of the MAPCA group had confluent PAs. The presence or absence of confluent PAs transforms the method of surgery. Non-confluent PAs have narrow RPA, LPA, and larger MAPCAs which may lead to pulmonary parenchymal disease secondary to incomplete intrapulmonary arborization7. Therefore, assessment of the confluence of PAs and the size and anatomy of PAs on fetal echocardiography is as necessary as identifying the origin of pulmonary blood supply for proper prenatal counseling. Considering the smaller PAs and impaired PA arborization, worse outcomes could have been expected in the MAPCA group compared with the DA group. However, we found that the existence of MAPCAs did not significantly impact postnatal survival. This may be related to the relatively higher accompanying anomalies in the DA group in our study. Moreover, DA-dependent cases needed surgical intervention to prevent desaturation secondary to the stenosis of DA within the first two weeks of life. In contrast, MAPCAs were operated on in later weeks as the stenosis occurred later.
The overall survival was too low compared with the other studies in the literature3,7. Survival was highly disrupted by extracardiac anomalies, particularly in the DA group. Among the babies who survived until the operation, postoperative survival rates for MAPCA and DA groups were 60% and 66.6%, respectively. These rates are also lower than the other studies3,9. Both postoperative deaths were related to extracardiac anomalies in the DA group. Therefore, while counseling the parents, it should be emphasized that extracardiac malformations may prohibit surgery or lead to postoperative mortality. Postnatal surgical series report rather promising outcomes12,13. However, many newborns like the ones in our study do not survive until the operation.
One of the weaknesses of our study is the small sample size. We restricted the study period to the last five years as we have had a dedicated fetal echocardiography clinic since 2017. Another consequence of this short study period is that we could not compare the long-term prognosis between the groups. The strength of the study is that it was conducted in a single center, and the fetal echocardiographies were performed by the same operator. All of the angiographies and the majority of the surgeries were performed by the associate cardiovascular surgery clinic.