Skip main navigation

Influence of Monitoring Strategy on Assessment of Ablation Success and Postablation Atrial Fibrillation Burden Assessment: Implications for Practice and Clinical Trial Design

Originally published 2022;145:21–30



Various noninvasive intermittent rhythm monitoring strategies have been used to assess arrhythmia recurrences in trials evaluating pharmacological and invasive therapeutic interventions for atrial fibrillation (AF). We determined whether a frequency and duration of noninvasive rhythm monitoring could be identified that accurately detects arrhythmia recurrences and approximates the AF burden derived from continuous monitoring using an implantable cardiac monitor (ICM).


The rhythm history of 346 patients enrolled in the CIRCA-DOSE trial (Cryoballoon Versus Contact-Force Irrigated Radiofrequency Catheter Ablation) was reconstructed. Using computer simulations, we evaluated event-free survival, sensitivity, negative predictive value, and AF burden of a range of noninvasive monitoring strategies, including those used in contemporary AF ablation trials.


A total of 126 290 monitoring days were included in the analysis. At 12 months, 164 patients experienced atrial arrhythmia recurrence as documented by the ICM (1-year event-free survival, 52.6%). Most noninvasive monitoring strategies used in AF ablation trials had poor sensitivity for detecting arrhythmia recurrence. Sensitivity increased with the intensity of monitoring, with serial (3) short-duration monitors (24-/48-hour ECG monitors) missing a substantial proportion of recurrences (sensitivity, 15.8% [95% CI, 8.9%–20.7%] and 24.5% [95% CI, 16.2%–30.6%], respectively). Serial (3) longer-term monitors (14-day ECG monitors) more closely approximated the gold standard ICM (sensitivity, 64.6% [95% CI, 53.6%–74.3%]). AF burden derived from short-duration monitors significantly overestimated the true AF burden in patients with recurrences. Increasing monitoring duration resulted in improved correlation and concordance between noninvasive estimates of the invasive AF burden (R2 = 0.85 and interclass correlation coefficient = 0.91 for serial [3] 14-day ECG monitors versus ICM).


The observed rate of postablation atrial tachyarrhythmia recurrence is highly dependent on the arrhythmia monitoring strategy employed. Between-trial discrepancies in outcomes may reflect different monitoring protocols. On the basis of measures of agreement, serial long-term (7–14 day) intermittent monitors accumulating at least 28 days of annual monitoring provide estimates of AF burden comparable with ICM. However, ICMs outperform intermittent monitoring for arrhythmia detection, and should be considered the gold standard for clinical trials.


URL:; Unique identifier: NCT01913522.

Clinical Perspective

What Is New?

  • Binary efficacy outcomes, such as time to atrial fibrillation (AF) recurrence, appear to underestimate the true effect of catheter ablation on the burden of atrial arrhythmia.

  • The most commonly performed intermittent rhythm monitoring techniques (eg, short-duration [24- and 48-hour] ambulatory Holter ECG monitors) miss a substantial proportion of arrhythmia recurrences and significantly overestimate the true AF burden.

  • Serial long-term noninvasive monitors (eg, four 7-day or two 14-day monitors) provide estimates of AF burden comparable with implantable cardiac monitors, but still miss a significant proportion of arrhythmia recurrences.

What Are the Clinical Implications?

  • Implantable cardiac monitors remain the gold standard for the detection of AF recurrence and quantification of AF burden.

  • In the absence of invasive monitoring, 28 days or more of cumulative intermittent noninvasive monitoring performed using serial longer-term (7-day and 14-day) ambulatory ECG devices provide reasonable arrhythmia detection and quantification of AF burden.

  • Transtelephonic monitoring improved arrhythmia detection when used as an adjunct for short-duration (24- and 48-hour) and intermediate-duration ambulatory monitors (7-day ambulatory ECG), but not for longer-duration ambulatory monitors (14-day ambulatory ECG).

Arrhythmia recurrence remains the primary outcome of clinical trials evaluating pharmacological and invasive therapeutic interventions for atrial fibrillation (AF). Since 2007, the gold standard definition of recurrence has been the binary freedom from any atrial tachyarrhythmia (AF, atrial flutter [AFL], or atrial tachycardia [AT]) lasting >30 seconds.1,2 Although the time to first recurrence of 30 seconds of atrial tachydysrhythmia was established on the basis of consensus, with the objective of standardizing reporting, it does not reflect pathophysiological processes or validated clinically meaningful outcomes. It further implies that the goal of therapeutic intervention should be arrhythmia elimination.

There has been increasing interest in the quantitative evaluation of AF burden, or proportion of the observed time that a patient is in AF. In contrast with binary outcomes, AF burden can be evaluated as a quantifiable entity measured on a continuous scale, which provides a more comprehensive evaluation of effect of therapeutic intervention.2–4 Although clinical implications of AF burden merit further clarification, studies have linked it to quality of life, health care use, and adverse cardiovascular outcomes (eg, stroke, heart failure hospitalization, and death).5–8

Noninvasive intermittent rhythm monitoring is the most widely used method to assess arrhythmia recurrences after catheter ablation. However, noninvasive techniques lack sensitivity for detecting arrhythmia recurrences, which increases the risk of misclassification errors and obscures comparisons between trials that use different methods to ascertain outcomes.9,10 Moreover, AF burden estimates derived from noninvasive intermittent rhythm recordings have been considered unreliable because of low accuracy, low precision, and high degrees of uncertainty.11 In contrast, although it has been suggested that only continuous monitoring obtained from a cardiac implantable electronic device (eg, implantable cardiac monitors [ICMs]) can truly define AF burden, these devices are invasive, costly, and associated with rare but potentially significant complications.

Using data from the CIRCA-DOSE trial (Cryoballoon Versus Contact-Force Irrigated Radiofrequency Catheter Ablation), we sought to assess whether there was a frequency and duration of noninvasive rhythm monitoring that could accurately detect clinically meaningful AF recurrences and approximate the AF burden after catheter ablation derived from continuous cardiac monitoring.


Study Design

The CIRCA-DOSE trial was a multicenter, parallel-group, single-blinded, randomized clinical trial, with blinded end point adjudication conducted at 8 centers in Canada.3 Details of the protocol have been reported previously.3,12 In brief, the study enrolled 346 patients aged >18 years with symptomatic AF refractory to at least 1 class I or class III antiarrhythmic drug referred for a first catheter ablation procedure. Ablation consisted of circumferential pulmonary vein isolation using standard techniques.12 All patients provided written informed consent.

All patients underwent insertion of an ICM at least 30 days before AF ablation. The ICM was used for determination of arrhythmia recurrence, as well as to accurately quantify AF episode duration and burden (defined as percentage of time in AF). Arrhythmia events meeting standardized arrhythmia detection settings were stored for adjudication by an independent, blinded clinical end point committee. Patients were followed for 1 year after the ablation procedure with clinical visits, a 12-lead ECG, and supplementary 24-hour ambulatory ECG monitoring at 3, 6, and 12 months. Automatic transmissions from the ICM were obtained daily.

The primary outcome of the CIRCA-DOSE trial was time to first symptomatic or asymptomatic atrial tachyarrhythmia (AF, AFL, or AT) documented by 12-lead ECG, 24-hour ambulatory ECG monitor, or ICM between days 91 and 365 after ablation, or a repeat ablation procedure at any time. AF burden was a secondary outcome and was defined as the proportion of the monitored interval that a patient was in AF. The study was approved by the institutional ethics and review board of each participating center. The data that support the findings of this study are available from the corresponding author on reasonable request.

Atrial Tachyarrhythmia Recurrence

In every patient, the complete rhythm history was reconstructed using the ICM-derived daily assessment of AF burden. For each patient, an array with 31 536 000 entries (ie, the number of seconds in the 365-day follow-up period) was constructed, with each entry populated by 0 if AT/AF/AFL was not present and by 1 if AT/AF/AFL was present. After construction of complete heart rhythm histories, computational simulation was performed to evaluate different sampling strategies on a per-patient basis. These scenarios included (A) weekly (scheduled) patient activated single-lead ECG recording (eg, transtelephonic monitoring), (B) symptom-based single-lead ECG recording, and (C) timed ambulatory ECG monitoring lasting (C1) 24 hours, (C2) 48 hours, (C3) 7 days, and (C4) 14 days, as well as combinations of single-lead ECG recording(s) and ambulatory ECG monitoring. The ambulatory monitors (24 hours, 48 hours, etc) were simulated by looking at each patient’s rhythm history and considering recurrences only in the window of time during which the patient would have worn the noninvasive monitor. In a participant with adjudicated AF detected by ICM, the intermittent monitoring scenario was considered a true positive if AF was observed on at least 1 of the monitored days, or a false negative if AF was not observed in any of the monitored days.

AF Burden in True-Positive and False-Negative Detections

AF burden was defined as the proportion of the time in AF in relation to the total monitored time. The gold standard AF burden was derived from the ICM and calculated as the time in AF divided by the total follow-up time, excluding a 3-month postablation blanking period, on a per-patient basis. For each noninvasive monitoring scenario, the AF burden was assessed in relation to noninvasive atrial tachyarrhythmia detection and was dichotomized into the following: AF burden in those with noninvasively detected atrial tachyarrhythmia recurrence (true-positive detection), and AF burden in those with atrial tachyarrhythmia recurrence that was not detected with noninvasive monitoring (missed recurrence or false-negative detection). By definition, AF burden in true-negative detections was 0.0%.

Noninvasive AF Burden Estimation

The correlation between the actual ICM-derived AF burden and the noninvasively estimated AF burden was evaluated using different monitoring strategies. The noninvasive AF burden estimate was defined as the time in AF during the monitored interval divided by the total noninvasive monitoring interval.

Statistical Analyses

Normally distributed continuous variables are presented as mean±SD and were compared using t tests, with nonnormally distributed variables presented as median (interquartile range [IQR]) and compared by Wilcoxon rank-sum tests. Categorical variables were presented as frequency and percentage and compared by χ2 tests.

Unadjusted survival curves of detection of atrial tachyarrhythmia on the basis of the monitoring method used were estimated by the Kaplan-Meier method and compared by log-rank tests. Unadjusted hazard ratios and CIs for hazard ratios were derived from Cox proportional hazards models.

The performance of the monitoring strategies was evaluated by means of sensitivity and negative predictive value. The sensitivity was defined as the proportion of patients correctly identified as having an atrial tachyarrhythmia recurrence with the simulated intermittent rhythm monitoring strategy to the true number of patients having AF recurrence (identified from the ICM). The negative predictive value was defined as the number of true negatives (negative AF detection in a participant without any AF) divided by the total number of negative tests (true negatives + false negatives), and the miss rate was defined as 1−sensitivity. The correlation between the actual (ICM-observed) AF burden and the estimated noninvasive AF burden was assessed using linear regression (slope, R2). Interclass correlation (ICC) was used to assess the consistency or reproducibility of quantitative AF burden between ICM and serial noninvasive monitoring strategies, with the ICC classified as poor (ICC<0.5), moderate (ICC, 0.5–0.75), good (ICC, 0.75–0.9), and excellent (ICC>0.9). A 2-sided value of P<0.05 was considered statistically significant. Analyses were performed using Prism 8 (GraphPad Software; San Diego, CA) and MATLAB (MathWorks; Natick, MA).


The study population consisted of all 346 patients with highly symptomatic paroxysmal AF enrolled in the CIRCA-DOSE trial.3 The mean age was 59±10 years, with 67% of participants being men (Table S1). Implantable cardiac monitor placement occurred a median of 73.5 days (IQR, 50–98.25) before ablation. Patients were followed for 12 months after ablation during which time a total of 32 964 ICM recordings were adjudicated by the core laboratory, and a total of 126 290 days of monitoring was performed. Of these 32 964 ICM recordings, 18 310 were adjudicated by the blinded events committee as AF/AFL/AT (13 837 were classified as AF, and 4473 were classified as AFL/AT). Of the 4283 symptomatic activations, 1928 (45.0%) were adjudicated as AF/AFL/AT.

Impact of Monitoring on Detection of Atrial Tachyarrhythmia Recurrence

At 12 months, a documented recurrence of any (symptomatic or asymptomatic) atrial tachyarrhythmia lasting >30 seconds on continuous cardiac monitoring had occurred in 164 patients (47.4%), translating into a 1-year arrhythmia-free survival of 52.6% (Table and Figure 1). In comparison, the 1-year arrhythmia-free survival using 3 intermittent monitoring periods at 3, 6, and 12 months after ablation was 69.4% for 14-day ambulatory ECG monitors, 79.8% for 7-day ambulatory ECG monitors, 88.4% for 48-hour ambulatory ECG monitors, and 92.5% for 24-hour ambulatory ECG monitors (all P<0.0001 versus ICM). Using a 120-second single-lead transtelephonic monitor alone, the 1-year arrhythmia-free survival was 85.3% for symptom-based recordings and 89.3% for weekly (scheduled) recordings. Combining intermittent ambulatory ECG monitoring with transtelephonic monitoring resulted in improved arrhythmia detection (Table), with slightly improved atrial tachyarrhythmia detection when scheduled (weekly) transtelephonic monitoring was performed in addition to symptom-based recordings and ambulatory ECG monitoring.

Table 1. Arrhythmia-Free Survival and Arrhythmia Detection Characteristics on the Basis of the Monitoring Techniques Used

Monitoring techniqueArrhythmia-free survivalSensitivityNegative predictive value
Implantable cardiac monitor52.6%100.0100.0
24-h AECG at 3, 6, 12 mo92.5% (95% CI, 90.2–95.8)15.8 (95% CI, 8.9–20.7)56.9 (95% CI, 54.9–58.3)
48-h AECG at 3, 6, 12 mo88.4% (95% CI, 85.5–92.3)24.5 (95% CI, 16.2–30.6)59.4 (95% CI, 57.0–61.5)
7-d AECG at 3, 6, 12 mo79.8% (95% CI, 75.9–84.4)41.6 (95% CI, 32.9–50.8)65.9 (95% CI, 62.3–69.3)
14-d AECG at 3, 6, 12 mo69.4% (95% CI, 64.8–74.6)64.6 (95% CI, 53.6–74.3)75.8 (95% CI, 70.5–81.2)
Weekly TTM89.3% (95% CI, 86.5–93.1)22.6 (95% CI, 14.6–28.5)58.9 (95% CI, 56.5–60.8)
Symptomatic TTM85.3% (95% CI, 82.0–89.5)31.0 (95% CI, 22.2–38.0)61.7 (95% CI, 58.8–64.1)
Symptomatic TTM + 24-h AECG at 3, 6, 12 mo80.6% (95% CI, 76.8–85.2)40.9 (95% CI, 31.4–48.9)65.3 (95% CI, 61.8–68.5)
Symptomatic TTM + 48-h AECG at 3, 6, 12 mo78.0% (95% CI, 74.0–82.7)46.4 (95% CI, 36.5–54.9)67.4 (95% CI, 63.6–71,1)
Symptomatic TTM + 7-d AECG at 3, 6, 12 mo72.3% (95% CI, 67.9–77.3)58.4 (95% CI, 47.9–67.7)72.8 (95% CI, 68.0–77.5)
Symptomatic TTM + 14-d AECG at 3, 6, 12 mo66.8% (95% CI, 62.1–72.0)70.0 (95% CI, 59.1–80.0)78.7 (95% CI, 73.1–84.7)
Weekly TTM + symptomatic TTM79.2% (95% CI, 75.9–83.9)43.9 (95% CI, 34.0–50.8)66.4 (95% CI, 62.7–69.3)
Weekly TTM + symptomatic TTM + 24-h AECG at 3, 6, 12 mo76.9% (95% CI, 72.8–81.7)48.7 (95% CI, 38.6–57.4)68.4 (95% CI, 64.4–72.3)
Weekly TTM + symptomatic TTM + 48-h AECG at 3, 6, 12 mo75.4% (95% CI, 71.2–80.3)51.9 (95% CI, 41.6–60.8)69.8 (95% CI, 65.6–73.9)
Weekly TTM + symptomatic TTM + 7-d AECG at 3, 6, 12 mo70.8% (95% CI, 66.3–75.9)61.6 (95% CI, 50.8–71.1)74.3 (95% CI, 69.3–79.3)
Weekly TTM + symptomatic TTM + 14-d AECG at 3, 6, 12 mo66.2% (95% CI, 61.5–71.4)71.3 (95% CI, 60.3–81.1)79.5 (95% CI, 73.8–85.5)

Monitoring protocols included serial ambulatory ECG monitors of 24-h, 48-h, 7-d, and 14-d duration at 3, 6, and 12 mo, either alone or in combination with symptom-triggered transtelephonic monitoring (TTM) and scheduled weekly TTM. AECG indicates ambulatory ECG (Holter) monitor; symptomatic TTM, transtelephonic monitor at time of symptoms; and weekly TTM, weekly scheduled transtelephonic monitor.

Figure 1.

Figure 1. Any atrial tachyarrhythmia recurrence. A, Event-free survival for atrial tachyarrhythmia episodes 30 seconds or longer, stratified by arrhythmia monitoring method. B, Event-free survival for serial ambulatory ECG (AECG) monitors in combination with symptom-triggered transtelephonic (TTM) monitoring. C, Event-free survival for serial ambulatory monitors in combination with symptom-triggered TTM and scheduled weekly TTM. *P<0.05; ****P<0.0001; #P<0.0001 vs implantable cardiac monitor.

Compared with ICM, the sensitivity for detecting atrial tachyarrhythmia recurrence ranged from 15.8% (95% CI, 8.9%–20.7%; 24-hour ambulatory ECG monitor) to 64.6% (95% CI, 53.6%–74.3%; 14-day ambulatory ECG monitors; Table), with a negative predictive value ranging from 56.9% (95% CI, 54.9%–58.3%; 24-hour ambulatory ECG monitor) to 75.8% (95% CI, 70.5%–81.2%; 14-day ambulatory ECG monitors; Table). The sensitivity for detecting atrial tachyarrhythmia recurrence increased when transtelephonic monitoring was added to short-duration (24- and 48-hour) and intermediate-duration ambulatory monitors (7-day ambulatory ECG; Table). However, the incremental value of transtelephonic monitoring was modest when combined with longer-duration ambulatory monitors (14-day ambulatory ECG).

The sensitivity for detecting atrial tachyarrhythmia recurrence increased with increasing number of monitored days (Figure 2). For each monitoring strategy, there was an absolute increase in arrhythmia detection (sensitivity) as the number of monitoring events increased. However, the relative improvement attenuated as the number of monitoring events increased beyond 4. For example, considering 14-day ambulatory ECG monitors, the arrhythmia detection sensitivity increased from 44.5% to 66.5% from 1 to 4 monitoring events (Δ=22.0%) but only increased from 66.5% to 81.7% (Δ=15.2%) between 4 and 10 monitoring events. In contrast, for the same number of monitored days, arrhythmia detection was generally better with less frequent prolonged monitoring (eg, one 14-day monitor, sensitivity 44.5%) compared with more frequent but shorter-duration monitors (eg, two 7-day monitors, sensitivity 36.0%; seven 2-day monitors, sensitivity 39.6%; fourteen 1-day monitors, sensitivity 33.4%).

Figure 2.

Figure 2. The sensitivity of intermittent rhythm monitoring to detect atrial tachyarrhythmia recurrence as a function of monitoring frequency (number of random monitoring events, 1–10) and monitoring duration (24 hours, 48 hours, 7 days, and 14 days ambulatory ECG [AECG] monitors). Prolonged monitoring was significantly superior to shorter monitoring (P<0.0001 for all comparisons). The dotted horizontal line represents the sensitivity of 0.25, 0.5, and 0.75. The total number of monitored days required at the maximum frequency of each monitoring strategy is denoted.

Impact of Monitoring on AF Burden Estimates

In patients with documented arrhythmia recurrence after ablation, the median AF burden (percentage of time in AF) detected by ICM was 0.13% (IQR, 0.02%–0.72%; Figure S1).

In those with arrhythmia recurrence documented on serial noninvasive monitoring (true positive), there was an inverse relationship between episodic monitoring duration and the observed median AF burden, ranging from 1.00% (IQR, 0.16%–6.12%; P=0.0014 versus ICM; Figure 3A) when recurrence was documented by serial 24-hour ambulatory ECG, to 0.22% (IQR, 0.05%–1.11%; P=0.3571 versus ICM; Figure 3D) when recurrence was documented by serial 14-day ambulatory ECG monitors.

Figure 3.

Figure 3. The relationship between atrial fibrillation (AF) burden and freedom from recurrent arrhythmia. For each panel, the left panel depicts event-free survival for atrial tachyarrhythmia episodes 30 seconds or longer on continuous cardiac monitoring in comparison with serial noninvasive intermittent rhythm monitoring at 3, 6, and 12 months (A, 24-hour ambulatory ECG [AECG] monitor; B, 48-hour AECG; C, 7-day AECG; D, 14-day AECG). The right panel depicts the AF burden recorded on noninvasive intermittent rhythm monitoring, stratified by the presence or absence of atrial tachyarrhythmia during the intermittent monitoring window (recurrence detected versus missed, respectively). In all cases, the AF burden was greater if atrial tachyarrhythmia was detected (orange, purple, green, red) compared with those in whom atrial tachyarrhythmia recurrence (light blue) was missed. The AF burden was inversely related to intermittent monitoring duration in those with documented recurrence, but not those without documented recurrence. NS, nonsignificant P-value for the AF burden in patients with missed recurrence versus AF burden in missed patients in A. ICM indicates implantable cardiac monitor.

In those without documented recurrence on serial noninvasive monitoring (false negative), the residual (missed) AF burden was low and did not differ between monitoring strategies, ranging from 0.09% (IQR, 0.01%–0.45%) with serial 24-hour ambulatory ECG at 3, 6, and 12 months, to 0.03% (IQR, 0.00%–0.16%) with serial 14-day ambulatory ECG monitors at 3, 6, and 12 months (P=0.0776).

The coefficient of determination (R2) for the linear regression of the distribution of actual (ICM-derived) AF burden versus AF burden estimate derived from serial noninvasive monitoring rose sharply as the total number of monitored days increased (Figure 4A). When ambulatory ECG monitoring was performed at 3, 6, and 12 months, there was a better correlation with the use of serial longer-term monitors (eg, R2 of 0.85 for 14-day ambulatory ECG) relative to shorter-term monitors (eg, R2 of 0.41 for 24-hour ambulatory ECG); however, for the same total number of monitored days, the R2 was lower with serial longer-term ambulatory ECG monitoring (black stars) when compared with short-duration ambulatory ECG monitoring performed at more frequent intervals (black circles).

Figure 4.

Figure 4. Relationship of atrial fibrillation (AF) burden as a function of the total number of noninvasively monitored days. Atrial fibrillation burden derived from invasive cardiac monitoring (ICM) was compared with the burden derived intermittent monitoring, with agreement plotted over the total number of monitored days. A, Correlation coefficient between noninvasive AF burden vs ICM and (B) interclass correlation (ICC) coefficient. The data points are for monitored days with equal intermonitoring intervals throughout the follow-up period. The stars are for timed ambulatory ECG (AECG) monitors at 3 and 12 months; 3, 6, and 12 months; or 3, 6, 9, and 12 months. The red data points in A are for patients with nonzero AF burden only (ie, excluding patients without ICM-detected AF recurrence).

Analysis excluding those patients without AF recurrence (0.0% AF burden) demonstrated a strong association between invasively and noninvasively derived AF burdens (Figure 4A, red). ICC coefficient was good (>0.7) starting with three 48-hour ambulatory ECG monitors and was excellent (>0.9) with three 14-day monitors (Figure 4B). Figure S2 shows representative distributions of ICM-derived versus noninvasive AF burden and corresponding linear correlations.

Considering the number of monitored days in comparison with AF burden, there was no significant difference between the AF burden detected on ICM and the AF burden derived from noninvasive monitors when monitoring was performed for ≥20 days (Figure S3).

Symptomatic Recurrence Detection Versus Monitoring Strategy

Symptomatic activations occurred in 80 patients (23.1%), who transmitted 4283 episodes. Of those 1928 (45.0%) activations were confirmed as AF/AFL/AT in 51/346 (14.7%) patients (1-year symptom-free survival of 85.3%). Serial 14-day ambulatory ECG monitoring at 3, 6, and 12 months had a 1-year symptom-free survival of 90.8% (P=0.0207 versus ICM) with a sensitivity and negative predictive value of 62.7% and 93.9% versus ICM, respectively. A conservative strategy of serial 1-day ambulatory ECG monitoring at 3, 6, and 12 months had a low sensitivity (3.9%) with a 1-year symptom-free survival of 99.4% (P<0.0001 versus ICM).


The present study demonstrates the following key findings: (1) the sensitivity for detecting arrhythmia recurrence after catheter ablation of AF increases with intensity of monitoring, with short-duration (24- and 48-hour) ambulatory monitors missing a substantial proportion of recurrences; (2) AF burden derived from short-duration (24- and 48-hour) ambulatory monitors significantly overestimates the true AF burden; (3) increasing monitoring duration results in more accurate estimates of AF burden; (4) serial long-term noninvasive monitors (eg, four 7-day or two 14-day monitors) provide estimates of AF burden comparable with implantable continuous cardiac monitoring, but the noninvasive monitors still miss a significant proportion of arrhythmia recurrences; and (5) ICM remains the gold standard for arrhythmia detection and AF burden quantification, and should be considered the gold standard for clinical trials.

Defining Success of Ablation

Atrial tachyarrhythmia recurrence has been used as the primary outcome of clinical trials evaluating pharmacological and invasive therapeutic interventions, including catheter ablation. To standardize reporting, arrhythmia recurrence has conventionally been defined as the time to first atrial tachyarrhythmia event lasting >30 seconds.1,2 Although established on the basis of consensus, the use of this outcome suffers from several limitations. First, the duration of atrial tachyarrhythmia used to define a qualifying arrhythmia episode markedly influences the observed incidence of recurrent arrhythmia and perceived diagnostic or therapeutic efficacy. Specifically, in the case of AF ablation, the use of a 30-second threshold significantly underreports AF ablation efficacy, and underestimates the clinical benefits of intervention.13

Second, the definition fails to account for the critical role of the arrhythmia monitoring method used to detect the recrrence. Because it is known that AF detection is proportionate to the duration and intensity of arrhythmia monitoring, it is possible that seemingly significant between-trial differences in success merely reflect the arrhythmia detection protocol used as opposed to true differences in arrhythmia recurrence rates.14 The present study demonstrates that the reported success of ablation, when considered as binary freedom from recurrent arrhythmia, varies from 52.6% to 92.5% depending solely on the arrhythmia monitoring technique used (eg, ICM versus three 24-hour ambulatory ECG monitors). Serial short-duration intermittent rhythm monitors such as 24- and 48-hour ambulatory ECG monitors have low sensitivity (17.9% and 25.5%, respectively) and negative predictive value (62.8% and 65.0%, respectively) for detecting arrhythmia recurrence, suggesting limited utility in defining freedom from recurrent arrhythmia. Although increasing the number of intermittent rhythm monitors improved sensitivity for AF detection, the yield plateaued after 4 short-duration (24- and 48-hour) and 6 longer-duration (7-day and 14-day) monitoring events. Even after 10 discrete monitoring events, short-duration ambulatory monitors fail to identify AF recurrence in more than half of patients. This has implications for clinical studies comparing the efficacy of therapeutic interventions, because a substantial proportion of patients evaluated with short-duration ambulatory monitors will be misclassified, and therapeutic success will be overestimated.

Last, considering recurrent AF as a binary condition inadequately captures clinically relevant outcomes and neglects the complexity of the patient experience. Specifically, patients often report improvements in quality of life in association with significant reductions in the frequency of arrhythmias.8 Thus, from a patient perspective, symptomatic improvement may be considered a procedural “success” even if it is not associated with complete elimination of AF. AF burden, or the proportion of time an individual is in AF, allows for quantitative assessment of treatment outcomes using a continuous scale, which provides a more comprehensive evaluation of treatment effect. A reduction in AF burden has been associated with changes in quality of life, health care use, and adverse cardiovascular outcomes (eg, stroke, heart failure hospitalization, and death).5–8 Although investigators have reported AF burden on the basis of intermittent rhythm monitoring, this approach has not been validated against continuous rhythm monitoring. The present study demonstrates that the most common intermittent rhythm monitoring techniques substantially overestimate arrhythmia burden, with serial short-duration ambulatory monitors demonstrating poor agreement with the true AF burden observed on continuous cardiac rhythm monitoring (R2 of 0.41 and 0.57 for 24- and 48-hour ambulatory ECG monitors used at 3, 6, and 12 months after ablation). However, in contrast with previous studies in the non-AF ablation population, a reasonable agreement was observed between AF burden determined by serial longer-term (7-day and 14-day) monitors and ICMs.9

Postablation Monitoring Protocol

On the basis of the present study, it appears that a total of 28 days of noninvasive monitoring performed using serial longer-term (7-day and 14-day) ambulatory ECG devices provides reasonable arrhythmia detection and quantification of AF burden. At this threshold of monitoring, the sensitivity for AF detection approaches 60% with a negative predictive value of nearly 80%, and an agreement in AF burden determination of nearly 80%. The addition of transtelephonic monitoring either on a weekly basis or symptom-triggered provides modest incremental value (sensitivity increment, 4%–5%).


In the present analysis, we assumed 100% compliance with ambulatory monitoring, which may overestimate the detection characteristics of noninvasive monitoring relative to that observed in clinical practice. The results of this study are directly applicable only to patients with paroxysmal AF; patients with persistent AF respond differently to AF ablation and may therefore benefit from a different postablation monitoring strategy. Furthermore, the probability of AF in the study population was not equally distributed over the follow-up period such that a similar monitoring intensity (number of days monitored) may return slightly different detection characteristics. In other words, three 14-day monitors at 3, 6, and 12 months (monitored days = 42) does not yield the exact same detection as weekly 1-day monitors (monitored days ≈ 42). Moreover, front-loading the monitoring (eg, more intensive intermittent monitoring for the first few months after the blanking period) risks missing AF episodes and underestimating AF burden given the differences in AF density between patients. Last, the correlation between ICM-derived AF burden and noninvasive burden estimates is driven by patients with higher AF burdens (Figure S2). However, patients with a higher AF burden may represent a subset in whom a more accurate noninvasive estimation of AF burden would be clinically desirable.


The results of this study demonstrate that the reporting of atrial tachyarrhythmia recurrence is highly dependent on the monitoring strategy used, such that apparent between-trial discrepancies may reflect the monitoring strategies used as opposed to true differences in outcomes. On the basis of measures of agreement, serial long-term (7–14-day) intermittent monitoring accumulating at least 28 days of annual monitoring provides estimates of AF burden comparable with ICM. However, ICMs outperform intermittent monitoring for arrhythmia detection, and should be considered the gold standard for clinical trials.

Article Information

Supplemental Material

Table S1

Figures S1–S3

Nonstandard Abbreviations and Acronyms


atrial fibrillation


atrial flutter


atrial tachycardia


Contact-Force Versus Cryoballoon Atrial Fibrillation Ablation


interclass correlation


implantable cardiac monitor


interquartile range

Disclosures Dr Andrade reports grants and personal fees from Medtronic, grants from Baylis, and personal fees from Biosense-Webster. Dr Deyell reports grants from Biosense-Webster. Dr Macle reports personal fees from Medtronic, and grants and personal fees from St Jude Medical/Abbott and Biosense-Webster. The other authors report no conflicts.


Supplemental Material is available at

For Sources of Funding and Disclosures, see page 29.

Correspondence to: Jason G. Andrade, MD, 2775 Laurel St, Vancouver BC V5Z 1M9, Canada. Email


  • 1. Calkins H, Brugada J, Packer DL, Cappato R, Chen SA, Crijns HJ, Damiano RJ, Davies DW, Haines DE, Haissaguerre M, et al; Heart Rhythm Society; European Heart Rhythm Association; European Cardiac Arrhythmia Society; American College of Cardiology; American Heart Association; Society of Thoracic Surgeons. HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for personnel, policy, procedures and follow-up. A report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial Fibrillation developed in partnership with the European Heart Rhythm Association (EHRA) and the European Cardiac Arrhythmia Society (ECAS); in collaboration with the American College of Cardiology (ACC), American Heart Association (AHA), and the Society of Thoracic Surgeons (STS). Endorsed and approved by the governing bodies of the American College of Cardiology, the American Heart Association, the European Cardiac Arrhythmia Society, the European Heart Rhythm Association, the Society of Thoracic Surgeons, and the Heart Rhythm Society.Europace. 2007; 9:335–379. doi: 10.1093/europace/eum120CrossrefMedlineGoogle Scholar
  • 2. Calkins H, Hindricks G, Cappato R, Kim YH, Saad EB, Aguinaga L, Akar JG, Badhwar V, Brugada J, Camm J, et al. 2017 HRS/EHRA/ECAS/APHRS/SOLAECE expert consensus statement on catheter and surgical ablation of atrial fibrillation.Heart Rhythm. 2017; 14:e275–e444. doi: 10.1016/j.hrthm.2017.05.012CrossrefMedlineGoogle Scholar
  • 3. Andrade JG, Champagne J, Dubuc M, Deyell MW, Verma A, Macle L, Leong-Sit P, Novak P, Badra-Verdu M, Sapp J, et al; CIRCA-DOSE Study Investigators. Cryoballoon or radiofrequency ablation for atrial fibrillation assessed by continuous monitoring: a randomized clinical trial.Circulation. 2019; 140:1779–1788. doi: 10.1161/CIRCULATIONAHA.119.042622LinkGoogle Scholar
  • 4. Andrade JG, Wells GA, Deyell MW, Bennett M, Essebag V, Champagne J, Roux JF, Yung D, Skanes A, Khaykin Y, et al; EARLY-AF Investigators. Cryoablation or drug therapy for initial treatment of atrial fibrillation.N Engl J Med. 2021; 384:305–315. doi: 10.1056/NEJMoa2029980CrossrefMedlineGoogle Scholar
  • 5. Go AS, Reynolds K, Yang J, Gupta N, Lenane J, Sung SH, Harrison TN, Liu TI, Solomon MD. Association of burden of atrial fibrillation with risk of ischemic stroke in adults with paroxysmal atrial fibrillation: the KP-RHYTHM study.JAMA Cardiol. 2018; 3:601–608. doi: 10.1001/jamacardio.2018.1176CrossrefMedlineGoogle Scholar
  • 6. Brachmann J, Sohns C, Andresen D, Siebels J, Sehner S, Boersma L, Merkely B, Pokushalov E, Sanders P, Schunkert H, et al. Atrial fibrillation burden and clinical outcomes in heart failure: the CASTLE-AF trial.JACC Clin Electrophysiol. 2021; 7:594–603. doi: 10.1016/j.jacep.2020.11.021CrossrefMedlineGoogle Scholar
  • 7. Kaplan RM, Koehler J, Ziegler PD, Sarkar S, Zweibel S, Passman RS. Stroke risk as a function of atrial fibrillation duration and CHA2DS2-VASc score.Circulation. 2019; 140:1639–1646. doi: 10.1161/CIRCULATIONAHA.119.041303LinkGoogle Scholar
  • 8. Samuel M, Khairy P, Champagne J, Deyell MW, Macle L, Leong-Sit P, Novak P, Badra-Verdu M, Sapp J, Tardif JC and Andrade JG. Association of Atrial Fibrillation Burden With Health-Related Quality of Life After Atrial Fibrillation Ablation: Substudy of the Cryoballoon vs Contact-Force Atrial Fibrillation Ablation (CIRCA-DOSE) Randomized Clinical Trial.JAMA Cardiol. 2021; 6:1324–1328.CrossrefMedlineGoogle Scholar
  • 9. Charitos EI, Pürerfellner H, Glotzer TV, Ziegler PD. Clinical classifications of atrial fibrillation poorly reflect its temporal persistence: insights from 1,195 patients continuously monitored with implantable devices.J Am Coll Cardiol. 2014; 63(25 pt A):2840–2848. doi: 10.1016/j.jacc.2014.04.019CrossrefMedlineGoogle Scholar
  • 10. Charitos EI, Stierle U, Ziegler PD, Baldewig M, Robinson DR, Sievers HH, Hanke T. A comprehensive evaluation of rhythm monitoring strategies for the detection of atrial fibrillation recurrence: insights from 647 continuously monitored patients and implications for monitoring after therapeutic interventions.Circulation. 2012; 126:806–814. doi: 10.1161/CIRCULATIONAHA.112.098079LinkGoogle Scholar
  • 11. Charitos EI, Ziegler PD, Stierle U, Robinson DR, Graf B, Sievers HH, Hanke T. Atrial fibrillation burden estimates derived from intermittent rhythm monitoring are unreliable estimates of the true atrial fibrillation burden.Pacing Clin Electrophysiol. 2014; 37:1210–1218. doi: 10.1111/pace.12389CrossrefMedlineGoogle Scholar
  • 12. Andrade JG, Deyell MW, Badra M, Champagne J, Dubuc M, Leong-Sit P, Macle L, Novak P, Roux JF, Sapp J, et al. Randomised clinical trial of cryoballoon versus irrigated radio frequency catheter ablation for atrial fibrillation-the effect of double short versus standard exposure cryoablation duration during pulmonary vein isolation (CIRCA-DOSE): methods and rationale.BMJ Open. 2017; 7:e017970. doi: 10.1136/bmjopen-2017-017970CrossrefMedlineGoogle Scholar
  • 13. Steinberg JS, O’Connell H, Li S, Ziegler PD. Thirty-second gold standard definition of atrial fibrillation and its relationship with subsequent arrhythmia patterns: analysis of a large prospective device database.Circ Arrhythm Electrophysiol. 2018; 11:e006274. doi: 10.1161/CIRCEP.118.006274LinkGoogle Scholar
  • 14. Balabanski T, Brugada J, Arbelo E, Laroche C, Maggioni A, Blomström-Lundqvist C, Kautzner J, Tavazzi L, Tritto M, Kulakowski P, et al; ESC-EHRA Atrial Fibrillation Ablation Long-Term Registry Investigators Group. Impact of monitoring on detection of arrhythmia recurrences in the ESC-EHRA EORP atrial fibrillation ablation long-term registry.Europace. 2019; 21:1802–1808. doi: 10.1093/europace/euz216CrossrefMedlineGoogle Scholar


eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.