Discussion
The recognition of need for atrial septal intervention after HP remains relatively controversial, with little evidence to support a consistent approach when there is such wide heterogeneity of the interatrial communication (2-4).  It is well recognised that initially unrestrictive interatrial communications may become increasingly restrictive, and also that a degree of atrial restriction may be acceptable, and indeed important, for those where biventricular repair may be possible in the future (7). It is an important decision as if a child is well-balanced even in the face of flow acceleration seen on echocardiography, they may become less well balanced after a septal intervention.
There are many factors affecting echocardiographic assessment of the atrial septum in HLHS. A colour and 2D approach will highlight aliasing, but this can be present even with a generous communication in the context of high pulmonary blood flow. Sometimes application of the pulmonary artery bands may reduce the pulmonary flow sufficiently to reduce the perceived transatrial gradient. Pulmonary vein Dopplers were not included in this study as they were not reliably measured in all studies and measurement can be positional. Additionally, we have noted that the A wave reversal can be absent in severe restriction only becoming obvious after relief of the obstruction which may be to do with reduced pulmonary flow in the face of obstruction.
The Doppler of the left pulmonary artery band appears to be the most affected by increasing atrial restriction.  The left pulmonary artery often falls in a better plane for Doppler interrogation compared to the right and therefore Doppler traces may be more accurate. It may also relate to the left pulmonary artery band being generally looser. Access to the left pulmonary artery can be challenging (a short intra-pericardial length)(8) both during HP and when it comes to pulmonary artery reconstruction at the subsequent stage.  A looser band may therefore be more representative of acute changes in the pulmonary vascular bed, although it should be noted that a similar waveform is seen when the band is too loose.
As the HP population is small, there are a relatively small number of patients in this cohort which did not therefore allow comparison between the groups with and without forward flow. Not all measurements were available for all scans at all time-points, and some (albeit few) scans were performed >48 hours after the procedure.  Additionally, some patients had the atrial septostomy as a proactive or opportunistic intervention, when another intervention was performed.  Invasive catheter data was not available for all patients as many had septostomies performed under echocardiographic guidance only, if there had been more invasive data, we may have been able to correlate this with some of the echocardiographic parameters.