DISCUSSION
Hyperechoic lung images on prenatal ultrasound examination form a heterogeneous group of malformations characterized by an increased echogenicity, usually detected by high-resolution ultrasound during the second trimester of pregnancy 10. Even if this notion is now well accepted, with recent reviews on the subjects13,18, very few series are published. Cavoretto at al.9 described the largest cohort with 193 cases, in which only two types of malformations were identified before and after birth (congenital cystic airway malformations-CCAM- and BPS). The series of Illanes et al. 11 used the same classification with 43 CCAM and 5 BPS, but in half of the operated cases the prenatal diagnosis was not confirmed. More recently, Kane et al. added a third group (hybrid malformations) 12. The confirmation of the diagnosis was good but obtained only for the operated cases (11/13), because no systematic CT scans were performed after birth, leading to an incomplete analysis. In these 3 series no case of CLE was reported, whereas it is now recognized as one of the main causes of prenatal hyperechoic images 19, 20. Based on a systematic post-natal imaging, our series shows in detail that various congenital lung malformations can have a prenatal hyperechoic aspect: CLE, PCM, BC, BPS and hybrid malformation.
Microcystic images were not strongly associated to post-natal cystic images, with sensitivity and a specificity of about 70%. On the other hand, the prenatal detection of a systemic arterial supply had a sensitivity of 96% and a specificity of 83%. Then, while the prenatal diagnosis of a BPS is certain, based on the association of a hyperechoic image and a systemic artery 2, 9, the presence of a hyperechoic zone without the visualization of a systemic artery can lead to various final diagnosis.
Even though compressive signs are exceptional, they can be considered as an indication for a prenatal procedure. In our two cases, they were diagnosed early, at 22 weeks’, according to the published previous cases, stressing the importance of the 2nd trimester evaluation and a close ultrasound monitoring 12. For all these reasons, we propose a cautious prenatal approach for hyperechoic images, avoiding the term of microcystic that appears confusing and strong enough, and focusing on the morphological description of the images and the local consequences.
Concerning the associated prenatal features, mediastinal shift was frequently observed (1/4 of our patients), especially in those with BPS. In agreement with previous series, mediastinal shift was not associated to higher risk of neonatal symptoms and, when isolated, it should not be interpreted as a sign of increased morbidity 21-23. Moreover, among the 7 patients with neonatal symptoms, 5 had no compression signs on the prenatal ultrasound and the final postnatal diagnosis was variable (cf Table 1). These data lead us to think that neonatal respiratory failure or secondary symptoms cannot be anticipated by the prenatal evaluation. Congenital lung malformations should thus be referred to an expert perinatal team to organize the neonatal management.
Systematic post-natal imaging showed that all the malformations were present after birth and during the 3 years mean follow-up. Even though few cases of complete regression of lung malformations have been described and well-documented, it has to be considered as an exceptional event and should not be strengthened during the prenatal counseling. Recent evaluations have shown that the changes in the lung echogenicity in the 3rd trimester (or signal intensity on prenatal MRI) could be the physiological explanation of the apparent regression, reported in the historical series 24. All cases with abnormal echogenicity of the lungs should be controlled after birth, even in cases of disappearance on the late US evaluations.
According to our practice, close to 60% of the patients were operated. The natural history of congenital lung malformations varies widely from asymptomatic cases to severe cardiopulmonary symptoms. Whereas there is an agreement for the surgery of all the symptomatic lung malformations, the management of asymptomatic patients is still controversial25,26. Some clinicians would advise the excision of the malformation, while others could argue that the risks are not high enough to warrant subjecting the child to a major operation and its potential complications 27,28. Apart from the risk of pulmonary infections, the risk of malignancy remains the most worrying factor for surgeons who recommend prophylactic excision of asymptomatic patients with cystic components (PCM and BC), that is our policy29-32. CLE and BPS were thus treated conservatively with secondary symptoms during the follow-up period twice more frequent in BPS.
This large series confirms that the ability to correctly defined congenital lung malformations by prenatal ultrasound is limited18, except when a hyperechoic image is visualized in association with a systemic artery that could highly predict a BPS. Prenatal imaging should then mainly focused on the description of the images and the local consequences more than trying to name a malformation.