If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
In patients with paroxysmal atrial fibrillation or flutter (PAF/PAFL), subjective symptoms are not concordant with real arrhythmic events. It is important to elucidate the incidence of asymptomatic PAF/PAFL in symptomatic patients in a clinical setting.
Methods and results
To identify a possible relationship with subjective symptoms, we reviewed 6319 trans-telephonic electrocardiographic strips (ECGs) recorded from 123 patients in a double-blind, placebo-controlled trial examining dose-response effects of flecainide. During a 4-week observation, 2848 ECGs, comprising 894 (31.4%) symptomatic and 1954 (68.6%) asymptomatic tracings, and, during 31 days of treatment, 3471 ECGs, comprising 874 (25.2%) symptomatic and 2587 (74.8%) asymptomatic tracings, were transmitted. There were significantly fewer symptomatic ECGs during the treatment period (Fisher's exact test, P < 0.01). Asymptomatic episodes accounted for 23.5% of the total in the observation period and 22.4% of that in the treatment period (Fisher's exact test, P = 0.561). But the frequency of asymptomatic PAF/PAFL dropped significantly, from 11.1% to 3.9% (P < 0.01), in the group receiving flecainide 200 mg/day. In the observation period, the positive predictive value of subjective symptoms was about 50%, and the sensitivity in predicting PAF/PAFL, around 80%. During treatment, these values did not change significantly, but negative predictive value (NPV) and specificity increased in the 200 mg flecainide group. In the analysis of each case, significant dose-response profiles were found in NPV (P = 0.019) and specificity (P = 0.041).
Conclusions
Asymptomatic PAF/PAFL was not rare in symptomatic patients, and flecainide reduced both symptomatic and asymptomatic episodes.
]. Paroxysmal atrial fibrillation or atrial flutter (PAF/PAFL) is usually recognized by typical symptoms such as palpitation, chest discomfort caused by the sudden onset of tachycardia, and irregular heart beat. In the Framingham Heart Study [
], the diagnoses in approximately one-third of AF patients were made electrocardiographically in the absence of any subjective symptom. In general, the prevalence of asymptomatic lone AF has been found in about 25% of the study patients in prospective studies [
Asymptomatic atrial fibrillation: demographic features and prognostic information from the atrial fibrillation follow-up investigation of rhythm management (AFFIRM) study.
]. It is important to know the clinical characteristics of asymptomatic arrhythmias in symptomatic patients not only as an aid in the prevention of ischemic stroke associated with PAF/PAFL, but also for the management of patients with PAF/PAFL. Careful attention must therefore be paid to asymptomatic arrhythmias because they may increase the risk of stroke. In the present study, we reanalyzed the trans-telephonic ECGs in a multicenter, placebo-controlled, double-blind trial of flecainide [
for the Flecainide Atrial Fibrillation Investigators Dose-response effect of flecainide in patients with symptomatic paroxysmal atrial fibrillation and/or flutter monitored with trans-telephonic electrocardiography—a multicenter, placebo-controlled, double-blind trial.
] to elucidate the frequency of asymptomatic PAF/PAFL in symptomatic patients and also evaluated the effect of flecainide on asymptomatic PAF/PAFL.
Methods
The relationship between ECG diagnosis and subjective symptoms on a total of 6319 trans-telephonic ECG strips recorded from 123 Japanese patients with a history of symptomatic PAF/PAFL was studied. These patients were included in a multicenter, placebo-controlled, double-blind trial and randomized to placebo or flecainide (25 mg, 50 mg, or 100 mg) twice daily (BID) [
for the Flecainide Atrial Fibrillation Investigators Dose-response effect of flecainide in patients with symptomatic paroxysmal atrial fibrillation and/or flutter monitored with trans-telephonic electrocardiography—a multicenter, placebo-controlled, double-blind trial.
] so as to evaluate the dose-response preventive effect of flecainide on PAF/PAFL. The test drugs and matching placebo were provided by Eisai Co., Ltd. (Tokyo, Japan). The study protocol was approved by the Institutional Review Board at each of 30 centers in Japan. This clinical trial consisted of a 4-week observation period and a 31-day treatment period. During the treatment period, 32 patients received placebo and 91 received flecainide (26 patients on 25 mg, 32 on 50 mg, and 33 on 100 mg BID). The present study represents the analysis for the pre-specified (but not the primary) endpoint. Patients with a history of symptomatic PAF/PAFL who had experienced two or more attacks per month were eligible for participation in this study [
for the Flecainide Atrial Fibrillation Investigators Dose-response effect of flecainide in patients with symptomatic paroxysmal atrial fibrillation and/or flutter monitored with trans-telephonic electrocardiography—a multicenter, placebo-controlled, double-blind trial.
]. A trans-telephonic monitor with a memory device (Cardiophone, Nihon-Kohden, Tokyo, Japan) was given to all participants who submitted written informed consent for this study. Patients were instructed how to record ECGs and were required to transmit routinely recorded ECGs for 30 s once a day during the study period, and also at any time they had a symptom that they felt indicated PAF/PAFL. Also, if a patient had symptoms, ECG transmission was again required during the symptomatic phase and immediately after disappearance of the symptoms. All ECGs were transmitted to the ECG center at Keio University Hospital and then forwarded to each study center by facsimile to confirm the ECG diagnosis and to assess whether it was safe to continue the study. Patients were also required to record their subjective symptoms every day throughout the study period, whenever they were transmitting ECG records. Two cardiologists (HA, SO) reviewed and diagnosed all the ECGs without any information on subjective symptoms. We examined a possible association between ECG diagnoses and subjective symptoms related to PAF/PAFL by scrutinizing the individual ECG strip obtained.
Definitions
“Symptomatic ECG” is defined as an ECG recorded when a patient had any subjective symptom indicative of PAF/PAFL. This type of ECG implies true PAF/PAFL, sinus rhythm, premature atrial complex (PAC), and so on. “Asymptomatic ECG” is defined as an ECG recorded at a time when the patient had no subjective symptoms. This was usually transmitted as a routine daily ECG. “Asymptomatic PAF/PAFL” is defined as an ECG recorded when PAF/PAFL was confirmed without any subjective symptom. Therefore, “asymptomatic PAF/PAFL” is included in “asymptomatic ECG.” Subjective symptoms are classified into palpitation, chest discomfort, chest pain, shortness of breath, dizziness, and fatigue as PAF/PAFL-related symptoms.
Statistics
Data are expressed as mean ± S.D. Demographic and baseline characteristics of each group were summarized using descriptive statistics. Continuous variables and dose-response effects were analyzed using one-way analysis of variance (ANOVA) and categorical variables were analyzed using Fisher's exact test. Positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity of symptoms to predict PAF/PAFL were calculated. Changes in these indices with treatment in each patient were averaged to test the statistical significance by paired-t-test. The dose-response effects of flecainide on these indices were tested with ANOVA using contrast followed by Fisher's least square difference (LSD) method for inter-group comparison (active versus placebo).
Results
The clinical characteristics of the patients studied are shown in Table 1. A total of 2848 ECGs comprising 894 (31.4%) symptomatic and 1954 (68.6%) asymptomatic episodes were transmitted during a 4-week observation period, and 3471 ECGs comprising 874 (25.2%) symptomatic and 2597 (74.8%) asymptomatic episodes were transmitted during the 31-day treatment period. The frequency of symptomatic ECGs was significantly decreased during the treatment period (Fisher's exact test, P < 0.01).
Table 1Baseline clinical characteristics of the study patients
Placebo (n = 32)
25 mg BID (n = 26)
50 mg BID (n = 32)
100 mg BID (n = 33)
P-value
Men (%)
28 (87.5)
22 (84.6)
23 (71.9)
24 (72.7)
0.311
Age (years)
59.4 ± 10.2
61.8 ± 9.8
57.7 ± 10.2
57.6 ± 11.0
0.393
Weight (kg)
67.5 ± 13.5
64.5 ± 9.5
66.3 ± 12.3
63.2 ± 8.9
0.449
Arrhythmia
PAF
27 (84.4)
16 (61.5)
25 (78.1)
27 (81.8)
0.173
PAF and PAFL
5 (15.6)
10 (38.5)
7 (21.9)
6 (18.2)
History of PAF/PAFL (years)
4.1 ± 4.5
4.3 ± 5.8
4.0 ± 3.8
6.0 ± 6.2
0.377
Underlying disease
Total (%)
24 (75.0)
21 (80.8)
23 (71.9)
28 (84.8)
0.604
Valvular disease
3 (9.4)
3 (11.5)
2 (6.3)
0 (0.0)
Cardiomyopathy
2 (6.3)
0 (0.0)
1 (3.1)
0 (0.0)
Angina pectoris
0 (0.0)
1 (3.8)
0 (0.0)
1 (3.0)
Hypertension
16 (50.0)
7 (26.9)
11 (34.4)
13 (39.4)
Pre-excitation syndrome
1 (3.1)
0 (0.0)
0 (0.0)
0 (0.0)
BID, twice daily; PAF, paroxysmal atrial fibrillation; PAFL, paroxysmal atrial flutter. Data are mean ± S.D. P-value is based on one-way ANOVA (for continuous variables) and Fisher's exact test (for categorical variables) (reprinted in part, with permission, from Atarashi et al.
for the Flecainide Atrial Fibrillation Investigators Dose-response effect of flecainide in patients with symptomatic paroxysmal atrial fibrillation and/or flutter monitored with trans-telephonic electrocardiography—a multicenter, placebo-controlled, double-blind trial.
A total of 4442 ECG records showed sinus rhythm; 1176, atrial fibrillation (AF)/atrial flutter (AFL); 600, atrial premature complex (APC); 74, ventricular premature complex (VPC); 37, sinus tachycardia. During the observation period, 626 (22.0%) ECGs showed PAF/PAFL; 339 (11.9%), APC; 43 (1.5%), VPC; 22 (0.8%), sinus tachycardia (ST); the remaining 1823 (64.0%) ECGs demonstrated a normal sinus rhythm. During the 31-day treatment period, there were 550 (15.8%) AF/AFL cases, 261 (7.5%) APC, 31 (0.9%) VPC, 14 (0.4%) sinus tachycardia, and the remaining 2619 (75.5%) were normal sinus rhythm cases.
Subjective symptoms associated with symptomatic ECG
The frequency of subjective symptoms at the time of the ECG recording was decreased, and conversely, asymptomatic ECGs were increased after flecainide treatment in a dose-dependent manner (Fig. 1). The frequency of asymptomatic ECGs was around 70% in each treatment group during the observation period.
Figure 1Dose-dependent effect of flecainide on frequency of subjective symptoms (ECG-related). The open bars and striped bars indicate the frequency of symptoms during the observation period and during the treatment period, respectively. *P < 0.05, **P < 0.01 (vs. observation period: Fisher's exact test).
The most frequent subjective symptom related to PAF/PAFL was palpitation, and the second most frequent was chest discomfort. Other symptoms, including chest pain, shortness of breath, dizziness and fatigue, were relatively less frequent. The subjective symptoms when the ECG showed true PAF/PAFL are summarized in Fig. 2. The frequency of these subjective symptoms appeared to decrease dose-dependently, and the frequency of asymptomatic ECGs (asymptomatic PAF/PAFL) was higher (over 30% of total recorded PAF/PAFL) in the 100 mg BID group than in other groups in both observation and treatment periods (Fig. 2). However, there was no significant difference between observation and treatment periods in any treatment group or in the treatment groups as a whole.
Figure 2Dose-dependent effect of flecainide on frequency of subjective symptoms during episodes of PAF/PAFL or flutter documented by ECG. The open bars and striped bars indicate the frequency of symptoms during the observation period and the treatment period, respectively. *P < 0.05 (vs. observation period: Fisher's exact test).
In PAF/PAFL episodes, 147 of 626 ECG recordings (23.5%) were found to be asymptomatic in the observation period; while 123 of 550 ECG recordings (22.4%) were found to be asymptomatic in the treatment period (NS). But in the flecainide 100 mg BID group, the frequency of asymptomatic PAF/PAFL decreased significantly from 11.1% (69 of 619 recorded as asymptomatic) to 3.9% (37 of 959, as above) (P < 0.01) (Fig. 3).
Figure 3Effect of flecainide on frequency of asymptomatic PAF/PAFL in ECG recorded in an asymptomatic state. The open bars and striped bars indicate the frequency of asymptomatic episodes during the observation period and the treatment period, respectively. **P < 0.01 (vs. observation period: Fisher's LSD).
The associations of subjective symptoms with ECG-recorded PAF/PAFL are summarized in Table 2, Table 3. Flecainide therapy did not significantly affect PPV or sensitivity. In both the observation and the treatment periods, the PPV of subjective symptoms was about 50%, and both sensitivity and specificity were around 80% (Table 2). In by-patient analysis (Table 3), PPV and sensitivity did not demonstrate significant differences between the observation and treatment periods. In the 100 mg BID group, the PPV decreased from 74.1% to 63.3% in the treatment period, but there was no significant difference, but NPV and specificity were significantly increased: from 87.5% to 95.2% (P = 0.016) and from 86.1% to 91.3% (P = 0.050), respectively. There were also significant dose-responses seen in NPV (P = 0.019) and in specificity (P = 0.041).
Table 2Effect of flecainide on correlation of subjective symptoms and PAF/PAFL (analysis by each ECG)
The ECGs made when the patients reported feeling subjective symptoms related to PAF/PAFL and obtained in the observation period and in the treatment period numbered 894 and 874, respectively. The ECG revealed PAF/PAFL in 479 (53.6%), PAC in 154 (17.2%), premature ventricular contraction (PVC) in 22 (2.5%), and ST in 9 (1.0%). The other subjects all had normal sinus rhythms in the observation period. In the treatment period, PAF/PAFL was seen in 427 (48.9%), APC in 154 (17.6%), PVC in 13 (1.5%), and ST in 9 (1.0%). The distribution of arrhythmias in the 4 study groups is summarized in Table 4. The frequency of PAC was decreased while that of normal sinus rhythm was increased, both significantly (P < 0.01), in the treatment period at the maximum dose of 100 mg BID.
Table 4ECG diagnosis when the subjective symptoms were believed to be due to PAF/PAFL in each treatment group
The key findings of this study are that asymptomatic PAF/PAFL is not rare in PAF/PAFL patients who report no symptoms, and that PAF/PAFL is decreased in frequency with flecainide treatment.
Effect of an antiarrhythmic drug on asymptomatic PAF/PAFL
Flecainide treatment contributes to decreasing the frequency of real symptomatic PAF/PAFL. Suppression of symptoms may be associated with elimination of arrhythmias or with a conversion from symptomatic to asymptomatic arrhythmias. In general, antiarrhythmic drugs reduce the ventricular rate in PAF/PAFL [
Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy. A multicenter, double-blind, crossover study of flecainide and placebo with transtelephonic monitoring.
Use of metoprolol CR/XL to maintain sinus rhythm after conversion from persistent atrial fibrillation. A randomized, double-blind, placebo-controlled study.
Role of transtelephonic electrocardiographic monitoring in detecting short-term arrhythmia recurrences after radiofrequency ablation in patients with atrial fibrillation.
Asymptomatic atrial fibrillation: demographic features and prognostic information from the atrial fibrillation follow-up investigation of rhythm management (AFFIRM) study.
]. Conversion of arrhythmias from symptomatic to asymptomatic seems to be associated with a decrease in the duration of arrhythmia or to a slowing of the heart rate, or to both [
Asymptomatic atrial fibrillation: demographic features and prognostic information from the atrial fibrillation follow-up investigation of rhythm management (AFFIRM) study.
] reported that asymptomatic PAF occurred 12 times as often as sustained symptomatic PAF, and the mean heart rates did not differ much between the two groups. Strickberger et al. [
] also reported that there was no significant difference in ventricular rate between the symptomatic and asymptomatic arrhythmic events. The question of whether the ventricular rate affected the subjective symptoms during PAF was still a matter of controversy. Although oral flecainide therapy reduces heart rate during a PAF episode, the effect of flecainide on asymptomatic PAF [
Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy. A multicenter, double-blind, crossover study of flecainide and placebo with transtelephonic monitoring.
] is not reported in these studies. However, the present study indicated that the frequency of asymptomatic PAF/PAFL was significantly decreased after 100 mg BID flecainide therapy. In patients with PAF/PAFL, a strong correlation was noted between symptoms and documented PAF/PAFL [
Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy. A multicenter, double-blind, crossover study of flecainide and placebo with transtelephonic monitoring.
]. A sensitivity of 89% for symptomatic PAF, as well as a specificity of 70%, a PPV of 69%, and an NPV of 89% were found in the previous double-blind trial of flecainide [
Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy. A multicenter, double-blind, crossover study of flecainide and placebo with transtelephonic monitoring.
]. In that trial, patients received placebo or 200–300 mg/day of flecainide, and PAF was recorded in 62.9% of all symptomatic ECGs but in only 10.6% of asymptomatic ECGs. These results were equivalent to our own findings. In the present study, the percentage of asymptomatic PAF/PAFL in the 100 mg BID group was 11.9% in the observation period, a value that was decreased by 3.9% in the treatment period, and this clinical effect was accompanied by a significant dose dependency of NPV and of specificity in the analysis of each case. Our results clearly indicated that flecainide suppressed not only symptomatic but also asymptomatic PAF/PAFL. These findings may reflect the clinical effect of flecainide. It was also reported that azimilide [
], a class III antiarrhythmic drug, reduced asymptomatic PAF using by trans-telephonic ECG. The clinical effects of antiarrhythmic drugs such as flecainide appear to decrease not only symptomatic but also asymptomatic PAF/PAFL.
PAF/PAFL-related subjective symptoms
In the present study, palpitation and chest discomfort were the most common subjective symptoms indicative of PAF/PAFL. In general, PAF develops suddenly and induces severe subjective symptoms, especially in the first episode; however, these subjective symptoms often become less noticeable to patients themselves despite the recurrence or extended duration of the disease, or both. Therefore, an asymptomatic episode of PAF/PAFL is not rare even in symptomatic patients. In the present study, we classified subjective symptoms into palpitation, chest discomfort, chest pain, shortness of breath, dizziness, and fatigue in order to compare the specificity of each subjective symptom of PAF/PAFL. Palpitation, chest pain, tachycardia, shortness of breath, and dizziness [
] were common as PAF-associated subjective symptoms in Caucasian as well as Japanese patients. Dizziness appears to be less frequent in Japanese patients, but there was no significant ethnicity regarding any subjective symptom related to PAF.
Diagnosis of symptomatic ECG
In the present study, in symptomatic ECG, i.e., the ECG recorded when a patient's symptoms are indicative of PAF/PAFL, PAC was the most common arrhythmia, but a normal sinus rhythm with a heart rate below 100/min was more frequent. These results were different from our predictions. During the treatment period, this tendency did not change. Only at the highest flecainide dose of 100 mg BID was the frequency of PAC significantly decreased (P < 0.01), resulting in an increase in the frequency of the normal sinus rhythm. These results suggest that flecainide actually suppresses PAC frequency, but rather increases the frequency of false positive trans-telephonic ECGs.
Clinical implications
In patients with PAF/PAFL, the reduction of symptoms reported above is probably attributable to better control of heart rate and regularity of ventricular responses, resulting in a greater improvement in the quality of life. But the risk of complications such as stroke probably does not differ between symptomatic and asymptomatic PAF because of the continuous advancement of electrical and mechanical remodeling of the atria. Even when PAF/PAFL patients become completely asymptomatic after being prescribed an antiarrhythmic drug, it is difficult to confirm whether they have or have not had an arrhythmic event. Therefore, PAF/PAFL patients with a risk of stroke should not discontinue anticoagulation therapy [
The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002;347:1825–33.
In the present study, all patients were required to transmit ECG recordings only for 30 s routinely once a day, and also at any time when they had symptoms indicative of PAF/PAFL, and again immediately after the symptoms disappeared. Therefore, differing numbers of transmitted ECGs were observed in each treatment group, and the duration of the observation period and the treatment periods were also not the same. The correlation between subjective symptoms and PAF/PAFL was analyzed based on the mean number of episodes in each case. A slowing of the heart rate during PAF/PAFL may have caused arrhythmic episodes to be asymptomatic; however, the heart rate per ECG recording was not investigated in this study.
Conclusions
Despite the limited amount of data obtained in the present study, the results indicate that flecainide has beneficial effects not only on symptomatic but on asymptomatic episodes.
Acknowledgements
The authors thank Mr. CWP Reynolds for his careful revision of the English of this paper.
Sources of funding: This clinical study was supported by a grant from Eisai Co. Ltd., Tokyo, Japan, which had no role in the interpretation of the data obtained.
References
Brand F.N.
Abbott R.D.
Kannel W.B.
Wolf P.A.
Characteristics and prognosis of atrial fibrillation: 30-year follow-up in the Framingham Study.
Asymptomatic atrial fibrillation: demographic features and prognostic information from the atrial fibrillation follow-up investigation of rhythm management (AFFIRM) study.
Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy. A multicenter, double-blind, crossover study of flecainide and placebo with transtelephonic monitoring.
Use of metoprolol CR/XL to maintain sinus rhythm after conversion from persistent atrial fibrillation. A randomized, double-blind, placebo-controlled study.
Role of transtelephonic electrocardiographic monitoring in detecting short-term arrhythmia recurrences after radiofrequency ablation in patients with atrial fibrillation.
The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002;347:1825–33.
00152-4&id=gr1.jpg){kind=link}
00152-4&id=gr2.jpg){kind=link}
00152-4&id=gr3.jpg){kind=link}