Detection of congenital heart disease (CHD) in the first trimester

Abstract

I read with interest the narrative review article on the detection of congenital heart disease (CHD) in the first trimester by Dudnikov, Quinton and Alphonse published in the June issue of Sonography. Undoubtedly, a great deal of work has gone into preparing this article, and I commend the authors for this. Still, I feel that the high detection rates quoted are overstated, particularly for the low-risk population. The authors do state this and some other limitations, but these limitations need more emphasis. In addition, the methodologies of some of the reviewed articles vary greatly and pooling heterogeneous studies increases the risk of significant bias. For example, the detection rate calculation included studies that used only cardiac imaging signs, whereas other studies included non-cardiac signs, for example, nuchal translucency (NT), ductus waveform. Another paper compared 2D imaging and 4D STIC in detecting CHD in a high-risk population. Some congenital cardiac anomalies evolve during pregnancy and are not evident in the first trimester (e.g., aortic senosis). Additionally, other anomalies are present and easily diagnosable (e.g., hypoplastic left heart syndrome) compared to those that are less obvious (e.g., atrioventricular defect). The skill level of sonographers performing first trimester combine screening (FTCS) is also an important consideration. To perform this type of imaging NT certification is required, and to maintain it, sonographers need to demonstrate an ongoing high standard via annual audits. Not all sonographers performing obstetric imaging have this certification. This system does not exist for morphology imaging and could partially explain the discrepancy in detection rates between the two examinations. There is also a need to differentiate between a screening examination (where the heart is thought to be abnormal) and fetal echocardiography (where a cardiac diagnosis is made and counselling of the condition provided) in the first trimester. Most patients referred for early fetal echocardiography are due to concerns at a screening examination, and only a few are because of a previous baby with CHD. In a recent meta-analysis and systematic review, Yu et al. reported a detection rate for CHD at 75% using fetal echocardiography in a high-risk population. This rate also includes false positive and false negative diagnoses. In my 20+ years of performing fetal and paediatric echocardiography with some very experienced paediatric cardiologists, a definitive diagnosis in complex cases is not made until the postnatal period and/or at cardiac surgery. Another potential confounder is the lack of post-mortem data if a pregnancy is terminated based on scan findings alone. Results of detection of CHD at routine FTCS in the low-risk population are quoted at 60%–63%. Of note, none of the reviewed studies was on an Australian or American population where obesity is more prevalent than in Europe. I support screening for CHD at the first-trimester examination. However, I would urge caution in claiming high-detection rates and accurate diagnoses is realistic at early gestations with some lesions and populations. There is a real danger that such claims could add additional workload pressure and increase the litigious risk for sonographers. Yours sincerely