Clinical and diagnostic value of the novel algorithm for categorisation of neonatal interatrial communication; findings at follow-up in preschool children

Abstract Background and aim Differentiation between a patent foramen ovale (PFO) and an atrial septal defect (ASD) of the secundum type is challenging, particularly in newborns. A PFO is considered normal, often undergoes spontaneous closure and is rarely associated with complications. An ASDs on the other hand tend to remain patent and may lead to right sided volumen overload, atrial fibrillation, and stroke, often necessitating surgical repair. To determine the need for follow-up in newborns with interatrial communications (IACs), a clear diagnostic differentiation between ASD and PFO is essential. We recently developed an algorithm to facilitate differentation between ASD and PFO in newborns. In this study, we reassessed preschool children diagnosed according to this algorithm with an ASD or a PFO neonatally, to investigate potential differences in patentcy between the two IAC subtypes. Methods Children participating in a large, multicenter, prospective population study were enrolled for a follow-up echocardiogram if they had a PFO or an ASD detected at the neonatal echocardiogram. The recently developed algorithm classifies IACs into three subtypes of PFO and three subtypes of ASD. According to the algorithm an ASD was defined as a defect with a diameter ≥4 mm, a location in the inferior part of the septum, or if multiple communications were present. Children with neonatal IACs were included until reaching n=200 in each of the six groups of IAC subtypes, or all children in a group (n<200) were invited. A single operator, blinded to the neonatal IAC subtype, assessed the follow-up echocardiograms for patency of an IAC. Patent IACs were reclassified as ASD or PFO using the same algorithm. Results We included 904 children with IAC (median age 5.2 years [interquartile range 4.5-5.5], 54% female). Neonatally, ASDs were present in 292 newborns and PFOs in 612 newborns. The follow-up echocardiograms showed patency of the IAC in 62 of the children with a neonatal ASD (18.5%) and in 54 of the children with a neonatal PFO (10.1%), (OR 2.0 [95% CI 1.3;3.0], p-value<0.001). Overall, patency was sligtly more frequent in girls (15.1%) compared to boys (10.1%), (OR 1.6 [95% CI 1.0;2.4], p-value<0.05). The risk of having a patent IAC, that classified as an ASD at follow-up, was 7.5% in the group of children with a neonatal ASD, compared to 1.1% in the group with a neonatal PFO, (OR 7.0 [95% CI 2.9;19.7], p-value<0.0001), figure. Conclusions Newborns diagnosed according to the new IAC algorithm with an ASD were twice as likely to have a patent IAC at follow-up at five years of age compared to children diagnosed with a neonatal PFO. At follow-up, the risk of having a patent IAC, that was classified as an ASD, was very low for children with a neonatal PFO. These findings support the diagnostic and clinical value of the new IAC diagnostic algorithm to differentiate between neonates with IACs that may need follow-up.