Idiopathic atrial fibrillation: a rose by any other name?

This editorial refers to ‘Idiopathic atrial fibrillation revisited in a large longitudinal clinical cohort’ by B. Weijs et al., on page 184 The category of idiopathic, sometimes called ‘lone’, atrial fibrillation (AF) is a mythic creature. A sort of ‘leftover’ category of ‘AF with no apparent explanation or underlying aetilogy’, idiopathic, or lone AF is often a source of fascination and frequently mentioned. Nevertheless, it is not an entity with a clear definition that is widely acknowledged and accepted. There are several problems in trying to come up with a suitable definition, particularly over different epochs of time. Even within frequently used definitions there are variations, uncertainties, and inconsistencies. It remains uncertain whether defining idiopathic AF is in itself a useful exercise. It is noteworthy that the number of medical conditions, anthropometric measures, and genotypes associated with AF and our diagnostic capabilities have all increased dramatically over the last few decades. Thus, the task of defining idiopathic AF seems much like that of the mythic Sisyphus, who endlessly rolled a rock to the top of a hill in Hades only to have it return to the bottom, requiring him to start all over again. Just when we have a definition of idiopathic AF with which we are content, a new underlying aetiology or a better diagnostic test for known underlying aetiologies comes along that requires starting all over again. As a consequence, the proportion of cases of new-onset AF that truly have no explanation has become smaller and smaller (3% in the present report1) as time passes. One wonders whether this diagnosis will disappear altogether at some point in the future. Nonetheless, in this issue of the journal the definition of idiopathic AF is re-explored using patients enrolled in the Euro Heart Survey on AF.1 It has frequently but not always been customary to include exclusion for age, often age over 60 years, in the definition of idiopathic AF.2 One presumed reason for including such a provision in the definition is to exclude as yet unknown ‘causes’ or silent underlying aetiologies for AF that are, like AF, frequently found within the aging process, given the strong relationship between age and incident AF.3 Another reason for excluding elderly patients is that frequently the definition of idiopathic AF is applied retrospectively to a cohort after the data have already been collected. The data set therefore may not include underlying aetiologies that are actually known to be associated with AF but were not included as data were being collected, as the data set was not planned with this particular purpose in mind. Nevertheless, it is certainly seems plausible that AF may emerge in patients older than 60 years without any apparent ‘cause’ and in the absence of any of the known underlying aetiologies for AF. Furthermore, designating anyone older than 60 years as ‘elderly’ seems rather quaint and outdated in the modern era. In the report of Weijs et al. it turns out that about half of the patients in the Euro Heart Survey on AF who otherwise qualified for the designation of idiopathic AF were older than 60 years.1 The age limit of 60 years does seem excessively conservative in the modern era. It seems perfectly reasonable to remove this particular age restriction from the definition of idiopathic AF. Whether there should be absolutely no age restriction whatsoever is arguable, however. What other problems of definition are to be considered with respect to idiopathic AF? By far the most frequently encountered problems are found swirling around the apparent need to exclude a history of, or the presence of, hypertension again because of the strong association between hypertension and AF.4 The absence of hypertension in 1971 is not the same as the absence of hypertension in 2011. The definition of hypertension itself over the decades has been anything but constant. It is arguable that a single pressure for all ages is a suitable criterion for exclusion of hypertension. How many measurements of blood pressure must exceed the current arbitrary pressure limit to make the diagnosis of hypertension is debatable. Alternatively, how many blood pressures must be measured to exclude hypertension? Is labile or ‘white coat’ hypertension the same as constant hypertension with respect to pathogenesis of AF? The questions seem endless. Here again, the authors come up against the limitations of the data available in the Euro Heart Survey on AF. The definition for hypertension they chose seems reasonable enough: a history of hypertension [defined as antihypertensive drug use (diuretics, angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists, aldosterone receptor antagonists, dihydropyridine calcium-channel blockers) or a systolic blood pressure ≥140 mmHg or a diastolic blood pressure ≥90 mmHg on the qualifying visit, or left ventricular hypertrophy interventricular septum width >10 mm, posterior wall width >10 mm].1 However, can we be sure that all hypertensives have been excluded with this definition? I think probably not. It is interesting that part of their definition of hypertension was the echocardiographic finding of left ventricular hypertrophy, a known consequence of hypertension and a frequent indicator of hypertensive heart disease. Paradoxically they did not exclude left atrial enlargement (volume, diameter, area) on the echocardiogram in their definition of idiopathic AF. In fact, they made a major point about not using left atrial size in the definition. The argument they put forward for doing so is that an enlarged left atrium can be a consequence rather than a ‘cause’ of AF. They are undoubtedly correct in this supposition. However, this leads into another of those endless arguments for which there is no correct answer. In patients with AF is an enlarged left atrium always the consequence of AF? Certainly one sees patients in situations when one can be fairly confident that they are presenting with their first and only episode of AF and the left atrium is already enlarged and thus logically part of the ‘cause’ for AF. Left atrial enlargement is a frequent fellow traveller with hypertension. Thus, it seems more reasonable to take the position that left atrial enlargement can be both a ‘cause’ and a ‘consequence’ of AF. Finally, the measurement they used was a simple left atrial diameter (albeit normalized for body mass index). In reading the reports of echocardiograms I am impressed by the inconsistency and lack of agreement between simple left atrial diameter and left atrial volume, particularly when normalized for body mass index, in determination of whether the left atrium is actually enlarged. For these reasons, I think the authors are on a bit of a slippery slope in ignoring left atrial size in their definition of idiopathic AF. How convincing is the authors' argument that age and left atrial size doesn't count in defining idiopathic AF? Even within their own data a skeptic might point out a couple of their observations raise doubts. In spite of the authors assertion that there was no difference in left atrial diameter between patients above and below the 60-year-old threshold, there does seem to be a strong trend (P= 0.07). What would the data look like if they had used left atrial volume, or a slightly different age cut-off, or had more patients? Assuming there is some fundamental mechanistic difference between paroxysmal and persistent AF, one could also argue that the observation that older patients had more persistent AF than paroxysmal AF also seems to be trying to tell us something. Once again Sisyphus comes to mind. What is the purpose of the task? Getting to the top of the hill and staying there seems to be out of the question. As in previous studies the authors found that those patients who met their definition of idiopathic AF had a lower risk of morbidity over the next year. In this case, a lower risk of stroke was found compared with patients with uncomplicated hypertension and no other apparent underlying aetiologic condition (including left ventricular hypertrophy) for AF, although the numbers are small. Finding low risk of morbidity in idiopathic AF is consistent with other lines of evidence, suggesting much of the serious morbidity and probably mortality associated with AF is more likely due to associated medical conditions rather than AF itself.5 Rather than tweaking the definition of idiopathic AF and describing the resultant population and their risk of morbidity, a potentially more interesting exercise is to try to determine why these patients actually have AF. Could there be a genetic mechanism? Would examination of such patients with new diagnostic techniques such as cardiac magnetic resonance imaging determine the presence of a common denominator that ‘causes’ all AF even that considered to be idiopathic? An interesting emerging candidate in this respect is fibrosis of the atria.6,7 Extrapolating from the role of fibrosis in the so-called isolated diastolic heart failure,8 an understanding of the various factors that initiate fibrosis in the heart, could lead to new upstream therapies aimed at primary prevention of AF in targeted patients. Factors that initiate fibrosis in the heart may include some previously unsuspected culprits such as remote transient viral infections or again, a genetic predisposition to atrial fibrosis. We are a long way from understanding the fundamental basis for AF and until we have such knowledge primary prevention of AF remains a noble but elusive goal. Idiopathic AF may be a rarer and rarer problem as time passes but understanding why it happens could be an important step in developing new paradigms for the treatment of AF. As so eloquently stated by the poet Robert Frost we truly ‘have miles to go before we sleep’. Conflict of interest: Disclosures for past two years: Honoraria as a member of Data Monitoring Committee—Boerhinger Ingelheim, Medtronic, Bristol Myers Squibb/Pfizer, Sanofi Aventis, Biotronik; All <$10,000.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.