Congenital Heart Disease, Atrial Fibrillation, and Ischemic Stroke Risk

Over the past several decades, major advances in the medical and surgical management of infants and children with congenital heart disease (CHD) have led to the improving survival of these vulnerable patients beyond childhood. Later in life, however, dysrhythmias become a prevalent comorbidity. Adults with CHD tend to develop atrial fibrillation (AF) at younger ages than the general population because of abnormal cardiac anatomy, displaced or malformed sinus nodes or atrioventricular conduction systems, other cardiac diseases, medications, and genetics. ¹ A prior large study found that atrial arrhythmias increased the risk of death and heart failure but not ischemic stroke in adults with CHD. ² Despite this finding, ischemic stroke remains a major concern for adults with CHD and AF.

Extensive prior studies have led to the development of guidelines and clinical scores to stratify stroke risk related to AF in the general population. ³ Clinical scores, such as congestive heart failure, hypertension, age ≥75 (doubled), diabetes, stroke (doubled), vascular disease, age 65 to 74 and sex category (female) (CHA₂DS₂‐VASc), help medical providers determine when to initiate anticoagulation for stroke prevention. ³ , ⁴ , ⁵ Currently, only limited data are available to guide decision‐making about anticoagulation for stroke prevention in adults with CHD and AF. ⁶ , ⁷ The CHA₂DS₂‐VASc score does not take into account the presence or severity of CHD, and it is thought to have only modest stroke predictive value in adults with CHD. ⁶ , ⁷ Nonetheless, guidelines suggest that it should be 1 of the factors that influence decision‐making for anticoagulation of patients with CHD in clinical practice. ⁸ Whether AF confers similar ischemic stroke risk in people with CHD as in the general population remains a gap in knowledge.

In this issue of the Journal of the American Heart Association (JAHA), Holmgren et al report the results of a Swedish, population‐based retrospective study that takes a step toward addressing this gap. ⁹ The authors performed a register‐based study using diagnostic codes to identify patients born between 1970 and 2017 with AF and CHD from the Swedish National Patient Registry and the Swedish Cause of Death Register. For each patient with AF and CHD, age‐ and sex‐matched controls with AF and without CHD were identified. Cases and controls were followed up from the onset of AF until ischemic stroke, death, or the end of the study. By study definition, the primary outcome of ischemic stroke had to occur after the date of AF diagnosis. Patients who sustained an ischemic stroke before AF was diagnosed were excluded.

Of 71 942 patients with CHD, 951 developed AF. Of 714 462 controls without CHD, 606 developed AF. Among these patients with AF, ischemic stroke was infrequent overall: only 28 ischemic strokes were identified among patients with AF and CHD, and only 3 ischemic strokes were identified among controls with AF and no CHD. The stroke risk was greater for patients with CHD and AF despite a higher proportion of treatment with anticoagulation therapy both before and after the onset of AF. The hazard ratio (HR) for ischemic stroke after the onset of AF was estimated to be >3 times higher in people who had noncomplex CHD. In patients with complex CHD, the risk was >6 times higher. The magnitudes of these HRs were even greater (adjusted HR of 8.3 for patients with complex CHD) in a statistical model that included several components of the CHA₂DS₂‐VASc score, including sex, age, hypertension, and heart failure. Diabetes, ischemic heart disease, and myocardial infarction were not included in the models because of a low number of these comorbidities.

Children were eligible for inclusion in the study, but most of the patients were adults, and the study results are primarily applicable to older age groups with CHD. The cohort born after 1990 (who were aged <28 years in 2017, at the end of the study) had few ischemic strokes. The lower incidence rate of ischemic stroke for those born after 1990 may reflect lower stroke risk at younger ages or could reflect improving stroke prevention strategies among people with CHD in more recent decades.

One of the strengths of the study was its large size, with the inclusion of patients from hospitals nationwide over an entire population rather than a potentially skewed subset of patients seen at tertiary care centers or a single institution. Health care in Sweden is decentralized, and health care expenditure is relatively high compared with other countries, and these factors should be kept in mind before generalizing these results to other populations.

The study had several additional limitations that should be kept in mind. The HRs were calculated from only a small number of cases and controls who had ischemic stroke. With this small number of events, a few misclassified or excluded patients with ischemic stroke attributable to AF could swing the results by a large margin. By study design, patients who had an ischemic stroke diagnosed before the onset of AF were excluded. This was potentially significant because 43 patients with CHD and 6 patients without CHD were excluded for an ischemic stroke that occurred before AF was diagnosed. Some of these patients could have had an AF‐related stroke, with AF detected during diagnostic workup prompted by stroke. The precision of the HRs, therefore, should be interpreted cautiously, and future studies are needed to confirm the findings.

The implications of the study by Holmgren et al are that ischemic stroke risk in adults with AF and CHD could be 3 to 8 times higher than in adults with AF in the general population, and that the risk of stroke is modified by the severity of the underlying heart lesion. The reasons for increased stroke risk in the setting of AF and CHD are not clear. One explanation is that CHD exposes people to non‐AF stroke risks more frequently. The study by Holmgren et al adjusted for heart failure identified by diagnostic codes when it was present before AF onset, but this strategy likely missed heart failure in some patients. The authors did not adjust for cardiac catheterization, cardiac surgery, ventricular assist devices, or extracorporeal life support.

In addition to extrinsic cardiac devices and procedures, stroke risk factors inherent to the heart may be more common in people with CHD. The mechanisms of AF‐associated stroke are not straightforward but rather tightly intertwined with underlying endothelium dysfunction, atrial fibrosis, impaired myocyte function, and other atrial cardiopathy. ¹⁰ How these vary in different congenital heart lesions is complex but must be considered as future stroke prevention strategies are developed.

AF is estimated to affect more than half of adults with CHD by the age of 50 years. ¹ With the increasing number of adults with AF and CHD, clinical decision‐making tools and antithrombotic stroke prevention strategies for this group of patients must be optimized. Prospective studies are needed to verify its findings, but the study by Holmgren et al suggests that AF‐related stroke risk assessment should not be directly extrapolated from the general population to people with CHD. If verified, a further implication is that stroke prevention strategies developed for AF in the general population may not work well if applied to adults with AF and CHD. More research is needed to understand underlying stroke mechanisms in this growing group of patients. Improving stroke prevention for adults with AF and CHD should be considered a specific, distinct research priority.

DISCLOSURES

Dr Fox receives grant funding from the National Institutes of Health (NIH) (1R01NS119896‐01A1 and 1UG3NS119702) and the American Heart Association/Bugher Foundation (23BFHSCP1176240 and 814 692). She served on a data safety and monitoring board for the NIH–National Heart, Lung, and Blood Institute (the pumps for kids, infants and neonates Trial). She receives royalties from UpToDate for pediatric stroke topics. She served as a consultant for Competitive Drug Development International. Dr Kamel served in a principal investigator role in the atrial cardiopathy and antithrombotic drugs in prevention after cryptogenic stroke trial, which received in‐kind study drug from the BMS‐Pfizer Alliance for Eliquis and ancillary study support from Roche Diagnostics; is a Deputy Editor role for JAMA Neurology; served in clinical trial steering/executive committee roles for the stroke of known cause and underlying atrial fibrillation (Medtronic), a study of milevexian versus apixaban in participants with atrial fibrillation (Janssen), and the fourth left atrial appendage occlusion study (Boston Scientific) trials; served in consulting or end point adjudication committee roles for AbbVie, AstraZeneca, Boehringer Ingelheim, and Novo Nordisk; and has household ownership interests in TETMedical, Spectrum Plastics Group, and Ascential Technologies.