Heart rate variability in congenital heart disease: looking and learning

Among paediatric cardiologists, talk of the autonomic nervous system and its dysfunction typically centres around patients who present with light-headedness, pre-syncope, syncope, postural orthostatic tachycardia syndrome, or other disorders of orthostatic intolerance. Concepts like heart rate variability (HRV), standard deviation of normal to normal RR intervals (SDNN) and other related terms are rarely considered in the setting of congenital heart disease (CHD) [1, 2]. What is often discussed, though, is how patients with repaired STAT Category 1 defects, like secundum atrial septal defect (ASD), do well over the long term [3]. We often think of ASD’s, repaired by a simple atriotomy and patch or, increasingly, transcatheter occluder device placement, as unlikely to have many in the way of future complications (as long as they are closed early). But, this traditional notion is being upended. Increased risks of atrial fibrillation and stroke [4], psychiatric disorders [5] and conduction abnormalities in patients with surgically repaired or transcatheter occluded ASD [6] as compared to the general population are now being recognized. This last aspect of the research is germane, as prior long-term outcome studies only looked at the cohort of repaired ASD patients without comparison to controls. The authors of these newer papers are able to leverage population-level data available to them through the Danish healthcare system.

The same Danish group has now published their findings of autonomic function in patients with repaired/closed ASD as compared to healthy age- and sex-matched controls [7]. In this study, they assess measures of HRV and autonomic function by 48 h Holter monitoring performed at least one decade after either surgical or transcatheter device closure of ASD. Patients with surgically closed ASD had decreased HRV in all 6 domains assessed compared to controls. Those with device closure had evidence of decreased HRV in 3 of 6 domains versus controls. Of the 6 domains, 2 (root mean square sum of differences between successive NN intervals, and percentage of successive RR intervals differing by >50 ms) are related to parasympathetic tone, while the other 4 (SDNN, standard deviation of the average of NN intervals over 5 min periods, or SDANN, mean of the standard deviation of normal NN intervals for 5 min periods, or SDNNi, and total number of NN intervals divided by the modal number of NN intervals, triangular index) are related to measures of both parasympathetic and sympathetic function. Both groups of patients had abnormalities within these 2 test categories. The authors should be commended for several issues. First, they looked at outcomes after a simple cardiac defect, one that would not be thought to be a problem. Second, they age- and sex-matched their study sample with a normal population for a true comparison. Third, they utilized a big data approach, taking advantage of healthcare data widely available for study in their country. And, fourth, they studied a topic that has been thought not to be concerning. Their findings of abnormal autonomic cardiac control in patients with ASD after surgical or device repair years after their procedures should get our attention. In combining the results of multiple studies of HRV, it appears that there is initial worsening of HRV measures in the postoperative period [1], followed by some degree of HRV normalization, at least in patients with ASD several months after repair [8–10]. Then, at some point later, HRV degrades again [7]. As the authors note, the ages of both surgical repair and device closure of the study patients are older than that of patients who now typically undergo these interventions, such that the increased duration of volume loading may have caused irreversible changes to the atria and sinus node. In the end, though, this does not diminish the fact that there is an increased incidence of abnormality, specifically as compared to the general population.

Why would this occur? Does surgical intervention lead to sinus node damage during cannulation? Do suture lines from an atriotomy, ASD patch and/or cannulation repair puncture in the superior caval vein interfere with sinus node function? Is there direct damage to cardiac autonomic innervation? Is it due to persistent right atrial and/or ventricular stretch? Why does it show up later, and when do these changes start? Why would HRV be worse with surgery as compared to device closure? How well do these domains correlate with specific autonomic functioning, and what do they really show? Lastly, what does this mean? Does this mean that these patients are at risk for ‘all-cause mortality?’ Although there are published normal values for HRV for adults, we do not know specifically that the abnormalities found in this study can be translated to the same type of risk seen in adults in congestive heart failure or post-myocardial infarction with HRV abnormalities . But, it does mean that paediatric cardiologists should get used to interpretation of the various domain measurements in 24-h Holter monitoring as it pertains to HRV. It means that we need to be looking at short- and long-term HRV measures more closely across all patients with CHD, and putting these findings into perspective as a tool for risk assessment in patients with adult CHD. At this point, we have a finding that seems to be in search of an intervention. But it does not mean that it is not germane as it stands, nor that it would not be useful later in predicting specific morbidity and mortality as our understanding of its relevance and context grows.

So much of the ability to advance the field of medicine depends upon creativity, an open mind, and an ability to use old information in new ways. As with so much of what we do in medicine, these novel findings of autonomic dysfunction in patients long term after ASD repair again proves the adages, ‘you won’t find it if you don’t look for it’, and ‘the day that we stop learning is the day that we stop being good physicians’.

REFERENCES