The Challenges of Monitoring Outcomes
Isabel Deisenhofer
Over the last few years, catheter ablation of atrial fibrillation (AF) has become an established therapeutic approach that might have the potential to cure this most common human arrhythmia (1, 2, 3, 4, 5, 6). However despite the worldwide use of this technique, it is still very difficult to estimate the true effectiveness of AF ablation. This difficulty is due to the varying definitions of ablation success and the even more disperse methods used to evaluate the effects of ablation. Additionally, a variety of different ablation techniques as shown in the previous chapters of this book with specific complications need to be considered.
Possible endpoints of AF ablation trials include ECG-based “rhythm” endpoints or focus on the clinical improvement of the patient or the quality of life (7). Rhythm endpoints comprise freedom from any atrial tachyarrhythmia, freedom from AF only (atrial tachycardia other than AF would still be counted as success), and/or reduction of AF burden. Possible nonrhythm endpoints of AF ablation may be improvement of (a) New York Heart Association (NYHA) heart failure class, (b) echocardiographic parameters like left ventricular ejection fraction or left atrial size, or (c) quality of life assessed with standard or specially designed questionnaires. Sinus rhythm with (often with a previously ineffective drug) or without antiarrhythmic drug therapy and with one or more ablation procedures, as well as blanking periods immediately after ablation and total time of follow-up are additional variables that should be discussed.
Taking all these considerations into account, it seems rather problematic to compare AF ablation trials, as they generally differ in at least one of these variables. Commonly accepted definitions for ablation success and follow-up evaluation are therefore needed to overcome these problems. As these are not available yet, we can only try to give an overview of currently used definitions of success, follow-up methods, and follow-up timeframes. Completeness cannot be claimed but we will try to concentrate on the variables most important to us.
The Rhythm EndpointsFreedom from any kind of atrial tachyarrhythmia is the strongest endpoint of AF ablation but the most difficult to prove. AF ablation is generally performed in patients
suffering from symptoms during AF. Especially at the beginning of the AF ablation era, only symptomatic patients with paroxysmal AF were included in AF ablation trials (2, 3, 4, 5, 6). When screening for AF recurrences, clinical evaluation of patients’ symptoms seemed to be sufficient. However, it was observed that even previously highly symptomatic patients experienced asymptomatic episodes of AF after ablation. This seemed to occur independent of the chosen ablation technique. The percentage of patients who experienced no AF symptoms after ablation despite occurrence of AF episodes ranges between 20% and 70% (8, 9, 10).
suffering from symptoms during AF. Especially at the beginning of the AF ablation era, only symptomatic patients with paroxysmal AF were included in AF ablation trials (2, 3, 4, 5, 6). When screening for AF recurrences, clinical evaluation of patients’ symptoms seemed to be sufficient. However, it was observed that even previously highly symptomatic patients experienced asymptomatic episodes of AF after ablation. This seemed to occur independent of the chosen ablation technique. The percentage of patients who experienced no AF symptoms after ablation despite occurrence of AF episodes ranges between 20% and 70% (8, 9, 10).
The reasons for this loss of symptoms are not clear. One might be a placebo effect merely due to the invasive procedure. Another could be an ablation-induced modification of the intrinsic cardiac nervous system by damaging (as a “side effect”) mixed ganglionated plexus. A third interesting one is the suspicion that asymptomatic AF episodes occurred before ablation as well as after. The latter has been demonstrated by Hindricks et al. (9) by using 7-day Holter ECG before AF ablation and by Purerfellner et al. (11) by using the arrhythmia Holter of implanted pacemakers before and after AF ablation. The disappearance of symptoms after ablation is obviously not linked to a lower heart rate during AF or an ablation-mediated more-regularized conduction of AF to the ventricles (8). The lesson from all these observations is that freedom from atrial tachyarrhythmia should be documented rather than assumed.
Assessment of Complete Freedom from Any Atrial Tachyarrhythmia
Complete freedom from any atrial tachyarrhythmia can be proven only with a permanent Holter ECG documenting any atrial high-rate episode (12). Implanted devices like pacemakers or ICDs could provide these data, provided that the atrial sensing is (close to) infallible (13,14). In some AF ablation studies, a considerable proportion of included patients had already received an implantable device and thus the optimal monitoring tool, but this is the exception rather than the rule (15).
In patients without an implanted device, there are mainly two methods of intensive, long- term rhythm monitoring: repetitive long-term Holter ECG recordings like 7-day Holter ECG or event monitoring (with daily ECG recordings and additional recording in case of symptoms). Both have been used in AF ablation studies (8, 9, 10,16) and have been found to be comparable regarding the sensitivity of AF episode detection. Strikingly, both revealed that approximately 70% of AF episodes are missed with a conventional 24-hour Holter ECG, which is the most commonly used rhythm assessment tool (16). As pointed out before, another interesting finding is that approximately 20% to 70% of recurrences of AF seem to be asymptomatic (8, 9, 10,17). Thus, ECG event monitor recordings performed only in the moment of symptoms are obviously not sufficient for a dense follow-up of rhythm after AF ablation.
With both methods, the frequency of ECG recordings or the duration of regular telemetric ECG transmissions is very important. The general rule that the more intensive the screening, the more clinical events (i.e., episodes of AF) are found is true for AF in the same way it is for other clinical entities. One main problem from a practical aspect is the quantity of follow-up ECG data that would need to be analyzed by medically trained persons and stored in a comprehensive manner (8,16).
Regardless of the chosen monitoring tool, the question still remains of how long AF has to last to be named AF recurrence. Looking at Holter ECGs, investigators have not reached a consensus as to what is sustained AF after an AF ablation. The range comprises AF episodes lasting more than 30 seconds, 5 minutes, or 30 minutes (2, 3, 4, 5, 6,8, 9, 10,17, 18, 19). Nademanee et al. chose the longest period with episodes lasting
more than 24 hours, but their patient cohort included many patients with persistent or permanent AF (15). A solution might be to take into account the length of the average AF episode in the individual patient before and after ablation. In paroxysmal AF, the aim of the procedure is to eliminate all AF episodes, even those lasting only for seconds. We would therefore recommend using the very strict criterion of 30 seconds in these patients.
more than 24 hours, but their patient cohort included many patients with persistent or permanent AF (15). A solution might be to take into account the length of the average AF episode in the individual patient before and after ablation. In paroxysmal AF, the aim of the procedure is to eliminate all AF episodes, even those lasting only for seconds. We would therefore recommend using the very strict criterion of 30 seconds in these patients.
Regular Atrial Tachycardias after Atrial Fibrillation Ablation
As mentioned before, the freedom from any kind of atrial tachyarrhythmia is the strongest endpoint in AF ablation. The significance of postablation regular atrial tachycardia like atrial flutter instead of AF is not yet clear (19, 20, 21). The occurrence of these tachycardias might still be considered (partial) AF ablation success as the atria are no longer able to fibrillate. In this view, atrial tachycardia would not represent a recurrence of AF and a patient suffering from this kind of tachycardia would be counted as a successful ablation patient. On the other hand, these tachycardias could represent last “remnants” of AF or could be interpreted as a proarrhythmic effect of ablation. The fact is, it is sometimes hard to overcome these tachycardias, that they often persist, and patients are more symptomatic than with AF due to faster atrioventricular (AV) conduction (e.g., 2:1 conduction with a resulting heart rate of 110-150 bpm). Whatever flip side of the coin you want to look at, it has to be kept in mind that anticoagulation is still necessary in patients with persistent atrial tachycardia, even if it is regular.
Atrial Fibrillation Burden
This term has been mainly used by studies evaluating the effect of novel pacemaker stimulation algorithms on the frequency and duration of AF episodes. It represents the percentage of total recording time that a patient is actually in AF (11,13,14). This makes sense by definition only in patients with an implantable device. In patients monitored only with repetitive 7-day Holter ECG, the assessment of AF burden is much more difficult. If the reduction of AF burden is the endpoint of data analysis, a valuable preablation AF burden assessment must be performed, again preferentially with a device—and is much less accurate than with a 7-day Holter ECG (9,12). The measurement of AF burden has become a statistical hazard in patients with few AF episodes. Recording a 7-day Holter ECG seems insufficient to assess the frequency and duration of AF before and after ablation if patients experience only monthly AF episodes (7,12). Even in patients with a (continuously recording) device, AF burden remains a surrogate endpoint in AF ablation as the aim is total elimination of AF.
Clinical Evaluation
The clinical evaluation of ambulatory patients with questionnaires or standard ECG documentation is, in our view, insufficient to prove freedom from AF. Nevertheless, some important information might be gained from a simple routine 12-lead ECG, like accelerated or decreased sinus rate at rest, change of P-wave morphology during sinus rhythm (indicating intra-atrial conduction disturbance), frequent premature atrial contractions, or prolonged PQ or QT interval (e.g., due to antiarrhythmic medication). Clinical evaluation of all patients after AF ablation (e.g., at 1, 3, 6, and 12 months) can provide interesting and important information even though it might be time consuming and costly (6,10). Regular clinical visits after AF ablation are clearly more popular in Europe, probably because of the shorter distances between clinics and patients’ homes (2,5,6,8, 9, 10,16, 17, 18). In the United States, the greater
mean distances between clinics and patients’ residences render clinical visits more difficult (4,22, 23, 24, 25). As an alternative, transmissions of local ECG recordings to the clinic are proposed (22,24). This seems a fairly good way to monitor patient outcomes. At any rate, it is preferable to telephonic follow-up or no follow-up at all.
mean distances between clinics and patients’ residences render clinical visits more difficult (4,22, 23, 24, 25). As an alternative, transmissions of local ECG recordings to the clinic are proposed (22,24). This seems a fairly good way to monitor patient outcomes. At any rate, it is preferable to telephonic follow-up or no follow-up at all.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
