297
lACC Vol. 17, No.2
February 1991:297-303
COOPERATIVE STUDIES
Flecainide Acetate Treatment of Paroxysmal Supraventricular Tachycardia and Paroxysmal Atrial Fibrillation: Dose-Response Studies EDWARD L. C. PRITCHETT, MD, STEVEN D. DATORRE, MD, MARC L. PLATT, MD, SALLY E. McCARVILLE, MS, ANDRINA J. HOUGHAM, BA, FOR THE FLECAINIDE SUPRAVENTRICULAR TACHYCARDIA STUDY GROUP* The dose-response relations for efficacy and tolerance of the antiarrhythmic drug flecainide acetate were studied in 28 patients with paroxysmal supraventricular tachycardia (Group 1) and 45 patients with paroxysmal atrial fibrillation or flutter (Group 2). Recurrent symptomatic tachycardia was documented with use of transtelephonic electrocardiographic recording. Patients received flecainide in doses of 25, 50, 100 and 150 mg twice daily and placebo for 1 month treatment periods. Among 14 patients in Group 1 who qualified for efficacy analysis, 4 (29%) had no tachycardia while taking placebo. The number with no tachycardia increased with progressively larger
flecainide doses; with the 150 mg twice daily dose, 12 (86%) of 14 patients had no tachycardia (p < 0.01 for overall differences among all treatments). Among 28 patients in Group 2, 2 (7%) had no tachycardia while taking placebo. The number with no tachycardia also increased with progressively larger flecainide doses; with the 150 mg twice daily dose, 17 (61 %) of 28 patients had no tachycardia (p < 0.01 for overall differences among all treatments). Noncardiac adverse experiences were the leading cause of premature study discontinuation during flecainide treatment periods (five patients in Group 1 and six patients in Group 2).
Paroxysmal supraventricular tachycardia and paroxysmal atrial fibrillation are common cardiac arrhythmias. Flecainide acetate is one of the few antiarrhythmic drugs for which randomized, double-blind, placebo-controlled studies (l,2) have demonstrated efficacy in preventing symptomatic recurrences of these arrhythmias. However, the novel studies that demonstrated efficacy for this compound did not address the issue of selecting a proper starting dose or assess the effect of efficacy and tolerance of various doses (that is, dose-response studies were not performed). This multicenter study was undertaken to establish the dose-response relations for efficacy and tolerance of f1ecainide in patients with symptomatic paroxysmal supraventricular arrhythmias.
dia defined by a mean heart rate > 120 beats/min, a QRS configuration that was normal in duration or showed aberrant conduction, 80 beats/min and a QRS configuration that was normal in duration or showed aberrant conduction. The baseline ECG during fibrillation showed fine fibrillatory waves and an irregular ventricular rhythm in contrast with flutter, which appeared as regular flutter waves on the baseline ECG with an atrial rate between 240 and 360 beats/min. Patients were excluded if an episode of tachycardia precipitated angina, heart failure or neurologic symptoms. Patients also were excluded if they had an antitachycardia pacemaker. Women were excluded if they were pregnant or lactating. This protocol was reviewed and approved by the respective institutional review boards for clinical investigation at the participating centers. All patients gave written consent to participate. Screening protocol. Before patients entered the study, their previous antiarrhythmic therapy was discontinued for a minimum of 4 half-lives, except that patients in Group 2 (paroxysmal atrial fibrillation or flutter) were allowed to continue taking digitalis at a constant dose throughout the study. During a screening period and for the remainder of the
Methods Study patients. Patients eligible for this protocol were men or women who were 2:18 years of age. Patients were classified into two groups on the basis of electrocardiogram (ECG) recorded during their symptomatic tachycardia. Group 1patients had paroxysmal supraventricular tachycar•A complete listing of the F1ecainide Supraventricular Tachycardia Study Group can be found in the Appendix. This study was supported by a grant from 3M Pharmaceuticals, St. Paul, Minnesota. Manuscript received April 18, 1990; revised manuscript received August 8, 1990. accepted September 5. 1990. Address for reprints: Edward L. C. Pritchett, MD. Box 3477, Duke University Medical Center, Durham. North Carolina 27710. ©1991 by the American College of Cardiology
(J Am
Coll CardioI1991;I7:297-303)
0735-1097/91/$3.50
298
PRITCHETT ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
JACC Vol. 17, No.2 February 1991:297-303
study, patients documented their cardiac rhythm during symptoms using a small portable recording ECG device (Cardiobeeper Memory Monitor) (3). Electrocardiograms were transmitted to a central receiving service (Survival Technology). To qualify for this study, patients had to record and transmit ECGs obtained during two episodes of tachycardia on separate calendar days during a 28 day screening period. Treatment protocol. Patients who qualified in the screening phase entered the dose-response phase, in which we gave patients a placebo and four different doses of flecainide in ascending order (25, 50, 100 and 150 mg twice daily). Each treatment period lasted a maximum of 31 days. The placebo treatment period was inserted randomly as the first treatment period or between two of the flecainide treatment periods; placebo and flecainide were given in a double-blind fashion. Patients returned for examination at the end of each treatment period. The investigators also called each patient at biweekly intervals during the treatment periods to answer questions, ask about adverse experiences and encourage compliance. Data analysis: efficacy. We analyzed the effect of each flecainide dose and placebo on four variables derived from the ECG data: 1) the number of patients with no episodes of tachycardia during each treatment, 2) the time to the first episode during each treatment, 3) the time interval between episodes during each treatment, and 4) the heart rate during tachycardia with each treatment. We defined the minimally important therapeutic effect on heart rate during tachycardia as a reduction 2': 15%. The study protocol specified that tachycardia recorded in the 1st 3 days of any treatment period was not to be included in the analysis to allow flecainide washout during the placebo treatment period or to achieve steady state blood concentrations during a drug treatment period (3). The protocol also specified that patients would advance to the next treatment period after 28 observation days or after four separate episodes of tachycardia (occurring after the 3rd treatment day) were recorded. In some patients the arrhythmia began during the washout phase and lasted for days or weeks; in others, it began at the end of one treatment period and continued into the next treatment period. For patients experiencing these types of arrhythmia, the starting date for each observation period was adjusted to the first date after day 3 that normal sinus rhythm was recorded. The primary efficacy analyses included all patients who complied with the study drug protocol and had efficacy data for all five treatment periods. Groups 1and 2 were analyzed separately. Differences between patients who were included or excluded from the analysis were compared by using chi-square analysis or Fisher's exact test for categoric variables and Student's t tests for continuous variables. Twotailed p values ::50.05 were considered statistically significant. A variety of tests were used to evaluate efficacy in the dose-response study. Cochran's test for a randomized com-
plete block design was used to compare the number of patients with no episodes of tachycardia across all five treatment periods; McNemar's test was used to compare placebo with each flecainide dose separately (4). The percent of patients reporting no episode during each treatment period was tested for linearity of dose-response using a repeated measures analysis of variance (5). In comparing the time to the first episode of tachycardia among the various treatments, the paired Prentice-Wilcoxon test (6) for censored paired data was used to test for significant differences between placebo and each flecainide dose; Bonferroni's adjustment for four comparisons (7) was used to maintain an overall significance level of 0.05 (that is, p = 0.05/4 = 0.0125). If a patient had no episode of tachycardia during a treatment, the time to the first episode was defined as the number of observation days plus 1 and was considered censored. To determine the interval between episodes, only the first four episodes were counted and only those episodes occurring in the 28 observation days were counted. For patients with one to four episodes of tachycardia, the interval between episodes was defined as the number of days of treatment divided by the number of episodes; if a patient had five or more episodes, the interval between episodes was defined as the time to the fifth episode divided by 4. If a patient had no episodes of tachycardia, the interval was defined as >28 days and was censored. The paired Prentice-Wilcoxon test with Bonferroni's adjustment (6,7) was used to test for significant differences between each flecainide dose and placebo. The heart rate during tachycardia for each patient was averaged for each treatment period. Student t tests were then used to compare placebo and each treatment dose; only patients experiencing tachycardia in both the placebo and the treatment periods were used in these pairwise comparisons. Bonferroni's adjustment for four comparisons (7) was performed to maintain the overall level of significance. Data analysis: adverse experiences. The percent of patients reporting any noncardiac adverse experience was compared across treatment groups. Cardiac adverse experiences were tabulated but were not analyzed statistically with respect to dose-response relations because of the small number reported. Data analysis: projected discontinuations due to noncardiac adverse experiences. This analysis included all patients, not just those used in the primary efficacy analysis. Adverse experiences for each patient were reviewed, and patients were placed into one of the following categories for each dose level: discontinued because of noncardiac adverse experience, discontinued because of inadequate response or continued in the study. Patients who were discontinued as a result of noncardiac adverse experiences while receiving a low dose of flecainide were placed in the noncardiac discontinuation category for all higher doses of flecainide because we assumed that a patient who had an adverse experience requiring discontinuation at a low dose would have the same adverse experience at all higher doses. For patients discon-
lACC Vol. 17, No, 2
PRITCHETI ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
February 1991:297-303
~
100
~
80
.PSVT (N a 14) • PAF or PAFI (N-28)
o
.. -
c 60 .c~
j
40
~CD
20
~
o
Placebo
25 mg bid 50 mg bid 100 mg bid 150 mg bid
Treatment
Figure 1. Percent of patients with no tachycardia, Among patients included in the primary efficacy analysis, there was a steady increase in the number reporting no episode of tachycardia (Attacks) as the flecainide dose was increased. PAF or PAFI = paroxysmal atrial fibrillation or flutter (Group 2); PSVT = paroxysmal supraventricular tachycardia (Group 1).
tinuing because of inadequate response, no assumptions were made concerning their response at higher flecainide dose levels. The percent of patients discontinuing treatment at each dose was then projected.
Results Demographic observations. A total of 73 patients (28 in Group 1and 45 in Group 2) qualified by having two episodes of tachycardia during the screening period and entered the dose-response phase. Among the 28 patients in Group 1 (supraventricular tachycardia), there were 17 men and 11 women; the mean age was 44.8 ± 15.9 years. Twenty-three of the 28 patients were white. The mean height was 172 ± 8.4 em; weight was 78.5 ± 17.1 kg. Patients had used a mean of two antiarrhythmic drugs (range zero to eight) before enrolling in this study. Among Group 1 patients, 17 entered all five treatment periods, but 3 were excluded from the efficacy analysis because of noncompliance with the protocolor unanalyzable data. There were no significant differences between the 14 patients included and the 14 patients excluded from in the efficacy analysis. Among 45 patients in Group 2(atrial fibrillation or flutter), there were 27 men and 18 women; the mean age was 59.8 ± 11.5 years. Forty-three of the 45 patients were white. The mean height was 173.2 ± 11.6 em; weight was 78.4 ± 19.7 kg. Group 2 patients had used a mean of 3.5 antiarrhythmic drugs (range 1 to 7) before entering this study. Among Group 2 patients, 33 entered all five treatment periods, but 5 patients were excluded from the efficacy analysis because of noncompliance with the protocol or unanalyzable data. There were no significant differences between the 28 Group 2 patients included in and the 17 patients excluded from the efficacy analysis. Antiarrhythmic efficacy (Fig. 1). Four (29%) of the 14 patients in Group 1 reported no episode of supraventricular tachycardia while receiving placebo, and this number increased with each increment in flecainide dose (Fig. 1); while
299
taking flecainide at a dose of 150 mg twice daily, 12 (86%) of 14 reported no episode of tachycardia. Two (7%) of 28 patients in Group 2 had no episode of tachycardia while receiving placebo, and this number increased to 17 (61%) during treatment with the highest dose of flecainide. Cochran's test showed that there were significant (p < 0.01) differences among treatments for both Group 1 and Group 2 patients. The repeated measures analysis of variance showed a significant (p < 0.001) linear dose response in the percent of Group 1 and Group 2 patients remaining free of tachycardia. However, McNemar's test for pairwise differences among treatments with Bonferroni's adjustment for four comparisons indicated that none of the pairwise comparisons between flecainide and placebo were significant for Group 1. In Group 2, a significantly higher percent of patients had no tachycardia while receiving the 100 and 150 mg twice daily flecainide treatments than while they were receiving placebo. In the analysis of time to the first episode of tachycardia, 50% of the observations in Group 1 and >30% of the observations in Group 2 were censored (that is, patients had no tachycardia during the 28 observation days). For patients in Group 1, the median time to the first episode during placebo treatment was 11 days. The median was more than doubled (to >24 days) at all flecainide dose levels except the lowest dose of 25 mg twice daily. However, no pairwise Prentice-Wilcoxon comparison was statistically significant when Bonferroni's adjustment for four comparisons was used. For patients in Group 2, the median time to the first episode was 3 days for placebo treatment; the median was almost doubled (to 5 days) with the 50 mg twice daily dose; and median times for the 100 and 150 mg twice daily doses were >25 and 14 days, respectively (p < 0.0125 for each comparison). Estimation of the median interval between episodes, like the time to the first episode, was hampered by many censored observations. With placebo treatment, the median interval between episodes was 19.8 days for Group 1and 7.2 days for Group 2. As assessed by paired Prentice-Wilcoxon analysis, no flecainide treatment was significantly different from placebo for Group 1, probably because of the small number of observations. In Group 2 there were more observations and the median time between episodes during flecainide therapy was significantly longer than that during placebo administration for the 100 mg (28 days, p < 0.0125) and 150 mg (>25.5 days, p < 0.0125) twice daily doses. Flecainide did not have a major effect on the heart rates recorded during recurrent arrhythmias (Fig. 2). In Group 1, the heart rate measured during placebo therapy did not differ significantly from that at any of the flecainide doses, primarily because only a few patients had episodes of tachycardia at the higher flecainide doses, and there was little power for comparison with placebo. In Group 2, the mean heart rate recorded during recurrent tachycardia while patients were taking placebo was 132 ± 21 beats/min (n = 26). The rate was significantly lower only with the 100 mg twice daily
300
lACC Vol. 17, No.2 February 1991 :297-303
PRITCHETT ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
~
c:
~en
250.-----------------,
.PSVT PAF or PAFI
200
e 1ii
150
ell
1ii
100
li'i
50
ex::
ell
:J:
Placebo
25 mg bid 50 mg bid 100 mg bid 150 mg bid
Treatment
Figure 2. Heart rate during tachycardia. In patients with paroxysmal supraventricular tachycardia (PSVT), the small number of events recorded with higher doses of flecainide weakened the statistical significance of the decreases observed with the 100 and 150 mg twice daily (bid) doses. For patients with paroxysmal atrial fibrillation (PAF) or paroxysmal atrial flutter (PAF!), the 100 mg twice daily dose was significantly lower (p < 0.0125) compared with placebo; however, the percent reduction did not meet the stated criterion of 15% to be clinically important.
flecainide dose (119 ± 18 beats/min, n = 13, p < 0.0125). Although this dose significantly slowed the rate compared with placebo, even this decrease failed to meet our criterion of a 15% reduction to be considered clinically important. Adverse experiences. The most frequent noncardiac adverse experiences for Group 1 and Group 2 patients by treatment period are shown in Table 1. For patients in both groups, a significant dose-response relation was observed for the number of patients reporting at least one adverse experience. Four of 73 patients (2 each in Group 1 and Group 2) developed cardiac adverse events while taking flecainide. One patient (Group 1) developed incessant supraventricular tachycardia while taking 25 mg of flecainide twice daily and one (Group 2) developed incessant atrial flutter with a rapid ventricular rate while taking the 50 mg twice daily dose. The third patient (Group 2) developed ventricular tachycardia while taking flecainide, 100 mg twice daily; this patient had coronary artery disease, angina, a history of myocardial infarction and nonsustained ventricular tachycardia before entering the study and had undergone coronary artery bypass surgery. The fourth patient (Group 1) developed symptoms of dyspnea and pedal edema suggesting heart failure while taking 50 mg of flecainide; he completed the trial and his heart failure symptoms resolved during continued treatment with open label flecainide. Of the total of 73 patients who entered the dose-response phase of the study) 28 were discontinued before completing the protocol. In Group 1, 11 (39%) of 28 patients were discontinued (6 while receiving flecainide and 5 while receiving placebo). Seventeen (38%) of 45 patients in Group 2 were discontinued prematurely (13 while receiving flecainide and 4 while receiving placebo). The reason most often cited for premature discontinuation was a noncardiac adverse experience. Reasons for discontinuation for each treatment level in each group are shown in Table 2.
Projected discontinuations due to noncardiac adverse experiences. These data estimale the percent of patients who would develop noncardiac adverse experiences requiring discontinuation from the study if dosing were started with each of the dose levels. Among 24 patients in Group 1 who could be evaluated during treatment with placebo, 1 patient (4.2%) was discontinued because of a noncardiac adverse experience. Two (7.7%) of26 patients evaluated at the 25 mg twice daily flecainide dose also were discontinued because of a noncardiac adverse experience (Fig. 3). Among the 23 Group 1 patients who were l:valuated at the 50 mg twice daily dose, none were discontinued; however, we projected that the 2 patients who were dscontinued at the 25 mg twice daily dose also would have bl:en discontinued at this dose, yielding a projected cumulative discontinuation rate of 8.7% at the 50 mg twice daily dos{: (Fig. 3). Similar cumulative projections for the 100 and 150 mg twice daily doses yielded discontinuation rates of 13% and 22.7%, respectively. Among 41 Group 2 patients who were evaluated while receiving placebo, 1 (2.4%) was discontinued because of a noncardiac adverse experien:::e, as was 1 (2.4%) of 42 patients evaluated at the 25 mg twice daily flecainide dose. In Group 2, projected cumulative discontinuation rates for noncardiac adverse experiences for the 50, 100 and 150 mg twice daily doses were 2.6%, 5.1% and 17.6%, respectively.
Discussion This study clearly defines the dose-response relation of flecainide in paroxysmal supraventricular arrhythmias in relation to both efficacy and tolerance variables. Indeed, this may be the only study of an antiarrhythmic drug in which dose-response relations are defined by using symptomatic arrhythmias as the primary measure of drug effect. Typically, dose-response relations bave been defined on the basis of elimination of asymptomatic premature ventricular beats (8).
Dose-response relation: effic~lcy. Understanding the doseresponse relations for an antiarrhythmic drug requires information about both efficacy (Fig. 1 and 2) and adverse experiences (Fig. 3). Our definition of the minimally effective dose was a dose that approximately doubled the following measures of efficacy in this study: 1) the number of patients with no tachycardia (Fig. 1); 21 the median time to the first recurrence of tachycardia; and 3) the median time interval between episodes of tachycardia. The 50 mg twice daily dose of flecainide appears to meet ttis criterion for both Group 1 (supraventricular tachycardia) and Group 2(atrial fibrillation or flutter) patients. For example, the percent of patients with no tachycardia with placebo was 29% and 7% for Group 1 and Group 2, respectively. The 50 mg twice daily dose increased these values to 57% and 21%, respectively, suggesting that this dose was slightly more potent in Group 2 than in Group 1 by this measure. Conversely, the median time to the first episode of tachycardia in Group 1 was increased from 11 days to >24 days with the 50 mg twice
301
PRITCHETI ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
JACC Vol. 17, No.2 February 1991 :297-303
Table 1. Most Frequent Noncardiac Adverse Experiences* Patients With Paroxysmal Supraventricular Tachycardia (Group I) Flecainide (mg twice daily)
Noncardiac Adverse Experience Headache Dizziness Nausea Dyspnea Flushing Fatigue No. of patients reporting 2: 1 such experience
25 (n = 26)
Placebo (n = 24)1
50 (n = 21)
150 (n = 19)
100 (n = 21)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
I
(4) (4) (4)
(4) (15) (0) (0) (0)
2
3 0 0 0 0 1
(0) (0) (5)
3 3 3 1 2 0
(16) (16) (16) (5)
(0)
0 3 0 1
(10) (5) (0) (14) (0) (5)
(14) (0)
(0) (8)
1 4 0 0 0 0
(29)
6
(23)
7
(33)
6
(29)
10
(53)
1 1 0 0 2 7
(0)
I
(0)
(II)
(0)
Patients With Paroxysmal Atrial Fibrillation or Flutter (Group 2) Flecainide (mg twice daily)
Noncardiac Adverse Experience
25 (n = 43)
Placebo (n = 42)t
50 (n = 40)
150 (n = 33)
100 (n = 38)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
Dizziness Fatigue Vision abnormal Dyspnea Myalgia Headache Nausea Tinnitus Edema Tremor Chest pain
1 2 2 0 3 2 2
2 2
0 1
(0)
(2)
3 3 3 2 0 3 2 1 1 0 1
(8) (8) (8) (5) (0) (8) (5) (3) (3) (0) (3)
(13)
1 0 2 2 2
(5) (5) (2) (2) (0) (5) (5) (5) (2)
5 3 4 2
1 0 1
(2) (5) (5) (0) (7) (5) (5) (2) (2) (0) (2)
2 1 1 1 2
(3) (5) (3) (3) (3) (5)
9 4 3 3 3 0 2 2 2 2 1
(27) (12) (9) (9) (9) (0) (6) (6) (6) (6) (3)
No. of patients reporting 2: I such experience
13
(31)
10
(23)
12
(30)
14
(37)
18
(55)
I
I
I
I I
(8) (II)
(5) (3)
'This table reports the actual number of patients who received each treatment and the actual number who reported noncardiac adverse experiences with that treatment. Patients could report more than one noncardiac adverse experience at each dose. tOne patient did not report the presence or absence of adverse experiences while receiving placebo.
daily dose; in Group 2 the median time increased from 3 to 5 days, suggesting that this dose was slightly more potent in Group 1 than in Group 2 by this measure. Recognizing the lack of precision in our measurements, we recommend flecainide at 50 mg twice daily as the starting dose for patients with paroxysmal supraventricular tachycardia or paroxysmal atrial fibrillation or flutter. The simplicity of using the same dose for patients with both types of paroxysmal arrhythmia is appealing. Additional improvement can be expected in both patient groups by dose increases. A flecainide dose of 50 mg twice daily appears to be approximately equipotent with verapamil at 480 mg/day (9).
Dose-response relation: projected discontinuations. The analysis of study discontinuations (Fig. 3) suggests that
< 10% of patients would discontinue therapy as a result of noncardiac adverse events at the recommended starting dose of flecainide. The most frequently reported noncardiac adverse experiences among all patients were dizziness, headache, nausea, vision abnormality, dyspnea and fatigue (Table 1); these findings are consistent with previous reports (10) of the use of flecainide in patients with ventricular arrhythmias. Cardiac adverse experiences also were reported. Two patients (one in each group) reported that they had more prolonged or more severe episodes of tachycardia than they had previously experienced. In addition, one patient with paroxysmal atrial fibrillation developed ventricular tachycardia during flecainide treatment; however, circumstances suggest that this arrhythmia could have been due to associ-
302
PRITCHEIT ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
JACC Vol. 17, No.2
February 1991 :297-303
Table 2. Reasons for Discontinuation From the Study Flecainide Dose (mg twice daily) Placebo
25
50
100
150
Total
Patients With Paroxysmal Supraventricular Tachycardia (Group I) Cardiac adverse experience Noncardiac adverse experience Personal Other* Total Total no. receiving each dose
1
I
I I
2 5
2 0 0 3
0 0 0 0 0
0 1 0 0 1
25
26
21
21
19
1
o
2
I I
4 1
7
o 2
o o 2
2 6 I
2 11
Patients With Paroxysmal Atrial Fibrillation or Flutter (Group 2) 0 1
Cardiac adverse experience Noncardiac adverse experience Personal Inadequate response Lost to follow-up Othert Total
4
3
1 0 1 0 0 0 2
41
42
40
I
0 I I
Total no. receiving each dose
0 1 0 1 0 I
0 0 0 3
o
38
33
o
o 5
4 1 I 2 17
*One patient was discontinued from the study because of prestudy laboratory abnormality, and one patient was dbcontinued because he failed to make follow-up calls. tOne patient was discontinued for receiving excluded concomitant medications. One patient was discontinued for a protocol violation.
ated coronary artery disease, although we called it a cardiac adverse experience to be conservative in our interpretation of the therapeutic role of flecainide. The occurrence of proarrhythmia as aggravation of supraventricular arrhythmias and as the new occurrence of ventricular tachycardia has been reported (l, II) in small numbers of patients taking flecainide. Limitations of the study. The current study has some limitations. The dosing scheme would have been improved if the sequence of treatments had been completely randomized. However, we believed that it was unreasonable to randomize patients to begin treatment with flecainide at 150 Figure 3. Projected cumulative study discontinuations due to noncardiac adverse experiences. These calculations suggested that if all patients initiated therapy with the 150 mg twice daily (bid) dose, 22.7% of patients with paroxysmal supraventricular tachycardia (PSVT) and 17.6% of patients with paroxysmal atrial fibrillation (PAF) or paroxysmal atrial flutter (PAF!) would be discontinued from therapy for noncardiac adverse experiences.
! 25...------------------, o
::: tll :::l
c ::: c
20
.PSVT .PAF or PAFI
0 ...... 15
~~
C
10
'C
~ ~
..
l1.
5
0
Placebo
25 mg bid
50 mg bid 100 mg bid 150 mg bid
Treatment
mg twice daily when they had never been exposed to lower doses. Also, the emphasis in the study on developing efficacy data made the dose-response data for tolerance somewhat less rigorous. For example, the discontinuations shown in Figure 3 were projected rather than actually observed for the 50, 100 and 150 mg twice daily doses of flecainide. However, these projections:end to give a conservative picture of the therapeutic role of flecainide, which is appropriate for a new indication. Other limitations were the ~;mall number of patients who completed the study and thl~ relatively short treatment periods, which made it difficult to estimate the true effect of the higher flecainide doses on the time to the first episode of tachycardia and intervals between episodes. These considerations limited the statistical power of some pairwise comparisons of flecainide with placebo; however, overall tests for significance allowed us to reject the null hypothesis that all doses tested were the same. The small number of patients completing the study was partially due to difficulty in enrolling patients. A high degree of cooperation was required from the patients to record and transmit their arrhythmias and the patients had to be willing to endure numerous episodes of tachycardia to complete the s::udy. If one estimates 2 episodes in the screening period, a few episodes that were not counted during the washout days and up to 4 episodes in each of the treatment period~., then patients easily could have had> 12 episodes during the study. Conclusions. Safety issues remain a concern with flecainide and other class IC antiarrhythmic drugs. In the Cardiac Arrhythmia Suppression Trial (CAST) (12), patients
303
JACC Vol. 17. No.2 February 1991:297-303
PRITCHETI ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
with recent myocardial infarction and frequent premature ventricular beats who were taking ftecainide or the related drug encainide had an increased risk of arrhythmic death compared with patients treated with placebo. No deaths were observed in our study, but the small number of subjects enrolled (n = 73) and the short length of the study (5 months) limited the ability of our study to detect events that occurred with low frequency. The exact role of ftecainide in the management of paroxysmal supraventricular arrhythmias will be determined by continued close observation of its effectiveness and safety as clinical experience accumulates.
Oregon Cardiovascular Teachings, Ltd, Eugene, Oregon: Leonard G. Christie. Jr., MD, Kathe L. Carlson, RN West Penn Hospital, Pittsburgh, Pennsylvania: Barry L. Alpert, MD, Connie M. Fecik, RN Latter Day Saints Hospital, Salt Lake City, Utah: Jeffrey L. Anderson, MD, Fidela Moreno, MD, Linda Johnson. RN, Kaye Summers, RN Hunter Holmes McGuire Veterans Administration Medical Center, Rich· mond, Virginia: Kenneth A. Ellenbogen, MD, Mary Lynn Martin, RN 3M Pharmaceuticals, St. Paul, Minnesota: Michael T. CuUen, Jr., MD, Bruce P. Ekholm, MS, Terrance L. Fox, MS, Patricia A. Fredell, Ronald W. Hawkinson, MS, Mary Jane Maser, Sharon L. Parrish,lnara Porietis, Judy A. Sellers, RN, Kathleen A. Wagenknecht, Richard R. Wilson, MD
Appendix Investigators from the Flecainide Supraventricular Tachycardia Study Group who participated in this study were:
University of Alabama, Birmingham, Alabama: G. Neal Kay. MD, Vance J. Plumb, MD, Andrew E. Epstein, MD, Rosemary S. Bubien, RN. MSN The Hospital of the Good Samaritan, Los Angeles, California: Anil K. Bhandari MD, Cheryl Leon, RN Cardiology and Invasive Cardiac Electrophysiology, Torrance, California: Marc L. Platt, MD, Annelie M. Brinkley, LVN Torrance Memorial Medical Center, Torrance, California: Benjamin L. Rosin, MD, Evan McCabe, RN, MN Indiana University School of Medicine, Indianapolis, Indiana: James 1. Heger, MD, William M. Miles, MD, Lawrence S. Klein, MD, Douglas P. Zipes, MD. Linda Crouter Johns Hopkins Hospital, Baltimore, Maryland: Thomas Guarnieri, MD, Steven DaTorre. MD. Rozann DeBorde, RN University of Massachusetts Medical Center, Worcester, Massachusetts: Charles Haffajee, MB Chir, Karen Rufino. RN. Charlene Degon. RN Henry Ford Hospital Heart and Vascular Institute, Detroit, Michigan: Charles R. Webb. MD, Karen A. Bielinski, RN, BSN Minneapolis Heart Institute, Minneapolis, Minnesota: Irvin F. Goldenberg, MD, Wes R. Pedersen, MD, Adrian Almquist, MD. Jonathan W. Rogers, MD, Richard R. Nelson, MD, James A. DanieL MD. Susanne K. Roeller, PharmD. Donna Hunn. RN, Faye Bul/emer, RN Hennepin County Medical Center, Minneapolis, Minnesota: David Dunbar. MD, Morrison Hodges. MD, Karin Fjeldoo-Sperbeck, RN Mount Sinai Medical Center, New York, New York. J. Anthony Gomes, MD, Stephen Winters, MD, Elena Pe, RN University of North Carolina, Chapel Hill, North Carolina: James R. Foster, MD. Alan Woelfel, MD, Linda Cullinane. RN Duke University Medical Center, Durham, North Carolina: Edward L. C. Pritchett, MD, Elizabeth A. McCarthy, RN Creighton Cardiac Center, Omaha, Nebraska: Syed M. Mohiuddin, MD. Tom T. Hee, MD. Daniel E. Hil/eman, PharmD, Lois A. Stengel, BSN
References I. Anderson JL, Gilbert EM, Alpert BL, et al. Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy: a multicenter, double-blind, placebo-controlled, crossover study of flecainide and placebo using transtelephonic monitoring. Circulation 1989;6: 1557-70. 2. Henthorn RW, Waldo AL, Anderson JL, et al. Flecainide for prevention of paroxysmal supraventricular tachycardia: a multicenter, double-blind, placebo-controlled, crossover study using transtelephonic monitoring (abstr). J Am Col/ CardioI1988;1I:77A. 3. Pritchett ELC, Lee KL. Designing clinical trials for paroxysmal atrial tachycardia and other paroxysmal arrhythmias. J Clin EpidemioI1988;41: 851-8. 4. Conover WJ. Practical Non-parametric Statistics. 2nd ed. New York: Wiley. 1980:130-3, 199-205. 5. Robson DS. A simple method for constructing orthogonal polynomials when the independent variable is unequally spaced. Biometrics 1959;53: 187-91. 6. O'Brien PC, Fleming Te. A paired Prentice-Wilcoxon test for censored paired data. Biometrics 1987;43: 169-80. 7. Encyclopedia of Statistical Sciences. Vol. I. New York: Wiley, 1982: 294-7. 8. Morganroth J, Pool P, Miller R, Hsu PH, Lee I, Clark DM. Dose· response range of encainide for benign and potentially lethal ventricular arrhythmias. Am J CardioI1986;57:769-74. 9. Pritchett ELC, Hammill SC, Reiter MJ, et al. Life table methods for evaluating antiarrhythmic drug efficacy in patients with paroxysmal atrial tachycardia. Am J CardioI1983;52: 1007-12. 10. Gentzkow GD, Sullivan JY. Extracardiac adverse effects of flecainide. Am J CardioI1984;53:IOIB-5B. 1I. Falk RH. Flecainide-induced ventricular tachycardia and fibrillation in patients treated for atrial fibrillation. Ann Intern Med 1989;111: 107-1 I. 12. CAST Study Group. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;31:406-12.
lACC Vol. 17, No.2
February 1991:297-303
COOPERATIVE STUDIES
Flecainide Acetate Treatment of Paroxysmal Supraventricular Tachycardia and Paroxysmal Atrial Fibrillation: Dose-Response Studies EDWARD L. C. PRITCHETT, MD, STEVEN D. DATORRE, MD, MARC L. PLATT, MD, SALLY E. McCARVILLE, MS, ANDRINA J. HOUGHAM, BA, FOR THE FLECAINIDE SUPRAVENTRICULAR TACHYCARDIA STUDY GROUP* The dose-response relations for efficacy and tolerance of the antiarrhythmic drug flecainide acetate were studied in 28 patients with paroxysmal supraventricular tachycardia (Group 1) and 45 patients with paroxysmal atrial fibrillation or flutter (Group 2). Recurrent symptomatic tachycardia was documented with use of transtelephonic electrocardiographic recording. Patients received flecainide in doses of 25, 50, 100 and 150 mg twice daily and placebo for 1 month treatment periods. Among 14 patients in Group 1 who qualified for efficacy analysis, 4 (29%) had no tachycardia while taking placebo. The number with no tachycardia increased with progressively larger
flecainide doses; with the 150 mg twice daily dose, 12 (86%) of 14 patients had no tachycardia (p < 0.01 for overall differences among all treatments). Among 28 patients in Group 2, 2 (7%) had no tachycardia while taking placebo. The number with no tachycardia also increased with progressively larger flecainide doses; with the 150 mg twice daily dose, 17 (61 %) of 28 patients had no tachycardia (p < 0.01 for overall differences among all treatments). Noncardiac adverse experiences were the leading cause of premature study discontinuation during flecainide treatment periods (five patients in Group 1 and six patients in Group 2).
Paroxysmal supraventricular tachycardia and paroxysmal atrial fibrillation are common cardiac arrhythmias. Flecainide acetate is one of the few antiarrhythmic drugs for which randomized, double-blind, placebo-controlled studies (l,2) have demonstrated efficacy in preventing symptomatic recurrences of these arrhythmias. However, the novel studies that demonstrated efficacy for this compound did not address the issue of selecting a proper starting dose or assess the effect of efficacy and tolerance of various doses (that is, dose-response studies were not performed). This multicenter study was undertaken to establish the dose-response relations for efficacy and tolerance of f1ecainide in patients with symptomatic paroxysmal supraventricular arrhythmias.
dia defined by a mean heart rate > 120 beats/min, a QRS configuration that was normal in duration or showed aberrant conduction, 80 beats/min and a QRS configuration that was normal in duration or showed aberrant conduction. The baseline ECG during fibrillation showed fine fibrillatory waves and an irregular ventricular rhythm in contrast with flutter, which appeared as regular flutter waves on the baseline ECG with an atrial rate between 240 and 360 beats/min. Patients were excluded if an episode of tachycardia precipitated angina, heart failure or neurologic symptoms. Patients also were excluded if they had an antitachycardia pacemaker. Women were excluded if they were pregnant or lactating. This protocol was reviewed and approved by the respective institutional review boards for clinical investigation at the participating centers. All patients gave written consent to participate. Screening protocol. Before patients entered the study, their previous antiarrhythmic therapy was discontinued for a minimum of 4 half-lives, except that patients in Group 2 (paroxysmal atrial fibrillation or flutter) were allowed to continue taking digitalis at a constant dose throughout the study. During a screening period and for the remainder of the
Methods Study patients. Patients eligible for this protocol were men or women who were 2:18 years of age. Patients were classified into two groups on the basis of electrocardiogram (ECG) recorded during their symptomatic tachycardia. Group 1patients had paroxysmal supraventricular tachycar•A complete listing of the F1ecainide Supraventricular Tachycardia Study Group can be found in the Appendix. This study was supported by a grant from 3M Pharmaceuticals, St. Paul, Minnesota. Manuscript received April 18, 1990; revised manuscript received August 8, 1990. accepted September 5. 1990. Address for reprints: Edward L. C. Pritchett, MD. Box 3477, Duke University Medical Center, Durham. North Carolina 27710. ©1991 by the American College of Cardiology
(J Am
Coll CardioI1991;I7:297-303)
0735-1097/91/$3.50
298
PRITCHETT ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
JACC Vol. 17, No.2 February 1991:297-303
study, patients documented their cardiac rhythm during symptoms using a small portable recording ECG device (Cardiobeeper Memory Monitor) (3). Electrocardiograms were transmitted to a central receiving service (Survival Technology). To qualify for this study, patients had to record and transmit ECGs obtained during two episodes of tachycardia on separate calendar days during a 28 day screening period. Treatment protocol. Patients who qualified in the screening phase entered the dose-response phase, in which we gave patients a placebo and four different doses of flecainide in ascending order (25, 50, 100 and 150 mg twice daily). Each treatment period lasted a maximum of 31 days. The placebo treatment period was inserted randomly as the first treatment period or between two of the flecainide treatment periods; placebo and flecainide were given in a double-blind fashion. Patients returned for examination at the end of each treatment period. The investigators also called each patient at biweekly intervals during the treatment periods to answer questions, ask about adverse experiences and encourage compliance. Data analysis: efficacy. We analyzed the effect of each flecainide dose and placebo on four variables derived from the ECG data: 1) the number of patients with no episodes of tachycardia during each treatment, 2) the time to the first episode during each treatment, 3) the time interval between episodes during each treatment, and 4) the heart rate during tachycardia with each treatment. We defined the minimally important therapeutic effect on heart rate during tachycardia as a reduction 2': 15%. The study protocol specified that tachycardia recorded in the 1st 3 days of any treatment period was not to be included in the analysis to allow flecainide washout during the placebo treatment period or to achieve steady state blood concentrations during a drug treatment period (3). The protocol also specified that patients would advance to the next treatment period after 28 observation days or after four separate episodes of tachycardia (occurring after the 3rd treatment day) were recorded. In some patients the arrhythmia began during the washout phase and lasted for days or weeks; in others, it began at the end of one treatment period and continued into the next treatment period. For patients experiencing these types of arrhythmia, the starting date for each observation period was adjusted to the first date after day 3 that normal sinus rhythm was recorded. The primary efficacy analyses included all patients who complied with the study drug protocol and had efficacy data for all five treatment periods. Groups 1and 2 were analyzed separately. Differences between patients who were included or excluded from the analysis were compared by using chi-square analysis or Fisher's exact test for categoric variables and Student's t tests for continuous variables. Twotailed p values ::50.05 were considered statistically significant. A variety of tests were used to evaluate efficacy in the dose-response study. Cochran's test for a randomized com-
plete block design was used to compare the number of patients with no episodes of tachycardia across all five treatment periods; McNemar's test was used to compare placebo with each flecainide dose separately (4). The percent of patients reporting no episode during each treatment period was tested for linearity of dose-response using a repeated measures analysis of variance (5). In comparing the time to the first episode of tachycardia among the various treatments, the paired Prentice-Wilcoxon test (6) for censored paired data was used to test for significant differences between placebo and each flecainide dose; Bonferroni's adjustment for four comparisons (7) was used to maintain an overall significance level of 0.05 (that is, p = 0.05/4 = 0.0125). If a patient had no episode of tachycardia during a treatment, the time to the first episode was defined as the number of observation days plus 1 and was considered censored. To determine the interval between episodes, only the first four episodes were counted and only those episodes occurring in the 28 observation days were counted. For patients with one to four episodes of tachycardia, the interval between episodes was defined as the number of days of treatment divided by the number of episodes; if a patient had five or more episodes, the interval between episodes was defined as the time to the fifth episode divided by 4. If a patient had no episodes of tachycardia, the interval was defined as >28 days and was censored. The paired Prentice-Wilcoxon test with Bonferroni's adjustment (6,7) was used to test for significant differences between each flecainide dose and placebo. The heart rate during tachycardia for each patient was averaged for each treatment period. Student t tests were then used to compare placebo and each treatment dose; only patients experiencing tachycardia in both the placebo and the treatment periods were used in these pairwise comparisons. Bonferroni's adjustment for four comparisons (7) was performed to maintain the overall level of significance. Data analysis: adverse experiences. The percent of patients reporting any noncardiac adverse experience was compared across treatment groups. Cardiac adverse experiences were tabulated but were not analyzed statistically with respect to dose-response relations because of the small number reported. Data analysis: projected discontinuations due to noncardiac adverse experiences. This analysis included all patients, not just those used in the primary efficacy analysis. Adverse experiences for each patient were reviewed, and patients were placed into one of the following categories for each dose level: discontinued because of noncardiac adverse experience, discontinued because of inadequate response or continued in the study. Patients who were discontinued as a result of noncardiac adverse experiences while receiving a low dose of flecainide were placed in the noncardiac discontinuation category for all higher doses of flecainide because we assumed that a patient who had an adverse experience requiring discontinuation at a low dose would have the same adverse experience at all higher doses. For patients discon-
lACC Vol. 17, No, 2
PRITCHETI ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
February 1991:297-303
~
100
~
80
.PSVT (N a 14) • PAF or PAFI (N-28)
o
.. -
c 60 .c~
j
40
~CD
20
~
o
Placebo
25 mg bid 50 mg bid 100 mg bid 150 mg bid
Treatment
Figure 1. Percent of patients with no tachycardia, Among patients included in the primary efficacy analysis, there was a steady increase in the number reporting no episode of tachycardia (Attacks) as the flecainide dose was increased. PAF or PAFI = paroxysmal atrial fibrillation or flutter (Group 2); PSVT = paroxysmal supraventricular tachycardia (Group 1).
tinuing because of inadequate response, no assumptions were made concerning their response at higher flecainide dose levels. The percent of patients discontinuing treatment at each dose was then projected.
Results Demographic observations. A total of 73 patients (28 in Group 1and 45 in Group 2) qualified by having two episodes of tachycardia during the screening period and entered the dose-response phase. Among the 28 patients in Group 1 (supraventricular tachycardia), there were 17 men and 11 women; the mean age was 44.8 ± 15.9 years. Twenty-three of the 28 patients were white. The mean height was 172 ± 8.4 em; weight was 78.5 ± 17.1 kg. Patients had used a mean of two antiarrhythmic drugs (range zero to eight) before enrolling in this study. Among Group 1 patients, 17 entered all five treatment periods, but 3 were excluded from the efficacy analysis because of noncompliance with the protocolor unanalyzable data. There were no significant differences between the 14 patients included and the 14 patients excluded from in the efficacy analysis. Among 45 patients in Group 2(atrial fibrillation or flutter), there were 27 men and 18 women; the mean age was 59.8 ± 11.5 years. Forty-three of the 45 patients were white. The mean height was 173.2 ± 11.6 em; weight was 78.4 ± 19.7 kg. Group 2 patients had used a mean of 3.5 antiarrhythmic drugs (range 1 to 7) before entering this study. Among Group 2 patients, 33 entered all five treatment periods, but 5 patients were excluded from the efficacy analysis because of noncompliance with the protocol or unanalyzable data. There were no significant differences between the 28 Group 2 patients included in and the 17 patients excluded from the efficacy analysis. Antiarrhythmic efficacy (Fig. 1). Four (29%) of the 14 patients in Group 1 reported no episode of supraventricular tachycardia while receiving placebo, and this number increased with each increment in flecainide dose (Fig. 1); while
299
taking flecainide at a dose of 150 mg twice daily, 12 (86%) of 14 reported no episode of tachycardia. Two (7%) of 28 patients in Group 2 had no episode of tachycardia while receiving placebo, and this number increased to 17 (61%) during treatment with the highest dose of flecainide. Cochran's test showed that there were significant (p < 0.01) differences among treatments for both Group 1 and Group 2 patients. The repeated measures analysis of variance showed a significant (p < 0.001) linear dose response in the percent of Group 1 and Group 2 patients remaining free of tachycardia. However, McNemar's test for pairwise differences among treatments with Bonferroni's adjustment for four comparisons indicated that none of the pairwise comparisons between flecainide and placebo were significant for Group 1. In Group 2, a significantly higher percent of patients had no tachycardia while receiving the 100 and 150 mg twice daily flecainide treatments than while they were receiving placebo. In the analysis of time to the first episode of tachycardia, 50% of the observations in Group 1 and >30% of the observations in Group 2 were censored (that is, patients had no tachycardia during the 28 observation days). For patients in Group 1, the median time to the first episode during placebo treatment was 11 days. The median was more than doubled (to >24 days) at all flecainide dose levels except the lowest dose of 25 mg twice daily. However, no pairwise Prentice-Wilcoxon comparison was statistically significant when Bonferroni's adjustment for four comparisons was used. For patients in Group 2, the median time to the first episode was 3 days for placebo treatment; the median was almost doubled (to 5 days) with the 50 mg twice daily dose; and median times for the 100 and 150 mg twice daily doses were >25 and 14 days, respectively (p < 0.0125 for each comparison). Estimation of the median interval between episodes, like the time to the first episode, was hampered by many censored observations. With placebo treatment, the median interval between episodes was 19.8 days for Group 1and 7.2 days for Group 2. As assessed by paired Prentice-Wilcoxon analysis, no flecainide treatment was significantly different from placebo for Group 1, probably because of the small number of observations. In Group 2 there were more observations and the median time between episodes during flecainide therapy was significantly longer than that during placebo administration for the 100 mg (28 days, p < 0.0125) and 150 mg (>25.5 days, p < 0.0125) twice daily doses. Flecainide did not have a major effect on the heart rates recorded during recurrent arrhythmias (Fig. 2). In Group 1, the heart rate measured during placebo therapy did not differ significantly from that at any of the flecainide doses, primarily because only a few patients had episodes of tachycardia at the higher flecainide doses, and there was little power for comparison with placebo. In Group 2, the mean heart rate recorded during recurrent tachycardia while patients were taking placebo was 132 ± 21 beats/min (n = 26). The rate was significantly lower only with the 100 mg twice daily
300
lACC Vol. 17, No.2 February 1991 :297-303
PRITCHETT ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
~
c:
~en
250.-----------------,
.PSVT PAF or PAFI
200
e 1ii
150
ell
1ii
100
li'i
50
ex::
ell
:J:
Placebo
25 mg bid 50 mg bid 100 mg bid 150 mg bid
Treatment
Figure 2. Heart rate during tachycardia. In patients with paroxysmal supraventricular tachycardia (PSVT), the small number of events recorded with higher doses of flecainide weakened the statistical significance of the decreases observed with the 100 and 150 mg twice daily (bid) doses. For patients with paroxysmal atrial fibrillation (PAF) or paroxysmal atrial flutter (PAF!), the 100 mg twice daily dose was significantly lower (p < 0.0125) compared with placebo; however, the percent reduction did not meet the stated criterion of 15% to be clinically important.
flecainide dose (119 ± 18 beats/min, n = 13, p < 0.0125). Although this dose significantly slowed the rate compared with placebo, even this decrease failed to meet our criterion of a 15% reduction to be considered clinically important. Adverse experiences. The most frequent noncardiac adverse experiences for Group 1 and Group 2 patients by treatment period are shown in Table 1. For patients in both groups, a significant dose-response relation was observed for the number of patients reporting at least one adverse experience. Four of 73 patients (2 each in Group 1 and Group 2) developed cardiac adverse events while taking flecainide. One patient (Group 1) developed incessant supraventricular tachycardia while taking 25 mg of flecainide twice daily and one (Group 2) developed incessant atrial flutter with a rapid ventricular rate while taking the 50 mg twice daily dose. The third patient (Group 2) developed ventricular tachycardia while taking flecainide, 100 mg twice daily; this patient had coronary artery disease, angina, a history of myocardial infarction and nonsustained ventricular tachycardia before entering the study and had undergone coronary artery bypass surgery. The fourth patient (Group 1) developed symptoms of dyspnea and pedal edema suggesting heart failure while taking 50 mg of flecainide; he completed the trial and his heart failure symptoms resolved during continued treatment with open label flecainide. Of the total of 73 patients who entered the dose-response phase of the study) 28 were discontinued before completing the protocol. In Group 1, 11 (39%) of 28 patients were discontinued (6 while receiving flecainide and 5 while receiving placebo). Seventeen (38%) of 45 patients in Group 2 were discontinued prematurely (13 while receiving flecainide and 4 while receiving placebo). The reason most often cited for premature discontinuation was a noncardiac adverse experience. Reasons for discontinuation for each treatment level in each group are shown in Table 2.
Projected discontinuations due to noncardiac adverse experiences. These data estimale the percent of patients who would develop noncardiac adverse experiences requiring discontinuation from the study if dosing were started with each of the dose levels. Among 24 patients in Group 1 who could be evaluated during treatment with placebo, 1 patient (4.2%) was discontinued because of a noncardiac adverse experience. Two (7.7%) of26 patients evaluated at the 25 mg twice daily flecainide dose also were discontinued because of a noncardiac adverse experience (Fig. 3). Among the 23 Group 1 patients who were l:valuated at the 50 mg twice daily dose, none were discontinued; however, we projected that the 2 patients who were dscontinued at the 25 mg twice daily dose also would have bl:en discontinued at this dose, yielding a projected cumulative discontinuation rate of 8.7% at the 50 mg twice daily dos{: (Fig. 3). Similar cumulative projections for the 100 and 150 mg twice daily doses yielded discontinuation rates of 13% and 22.7%, respectively. Among 41 Group 2 patients who were evaluated while receiving placebo, 1 (2.4%) was discontinued because of a noncardiac adverse experien:::e, as was 1 (2.4%) of 42 patients evaluated at the 25 mg twice daily flecainide dose. In Group 2, projected cumulative discontinuation rates for noncardiac adverse experiences for the 50, 100 and 150 mg twice daily doses were 2.6%, 5.1% and 17.6%, respectively.
Discussion This study clearly defines the dose-response relation of flecainide in paroxysmal supraventricular arrhythmias in relation to both efficacy and tolerance variables. Indeed, this may be the only study of an antiarrhythmic drug in which dose-response relations are defined by using symptomatic arrhythmias as the primary measure of drug effect. Typically, dose-response relations bave been defined on the basis of elimination of asymptomatic premature ventricular beats (8).
Dose-response relation: effic~lcy. Understanding the doseresponse relations for an antiarrhythmic drug requires information about both efficacy (Fig. 1 and 2) and adverse experiences (Fig. 3). Our definition of the minimally effective dose was a dose that approximately doubled the following measures of efficacy in this study: 1) the number of patients with no tachycardia (Fig. 1); 21 the median time to the first recurrence of tachycardia; and 3) the median time interval between episodes of tachycardia. The 50 mg twice daily dose of flecainide appears to meet ttis criterion for both Group 1 (supraventricular tachycardia) and Group 2(atrial fibrillation or flutter) patients. For example, the percent of patients with no tachycardia with placebo was 29% and 7% for Group 1 and Group 2, respectively. The 50 mg twice daily dose increased these values to 57% and 21%, respectively, suggesting that this dose was slightly more potent in Group 2 than in Group 1 by this measure. Conversely, the median time to the first episode of tachycardia in Group 1 was increased from 11 days to >24 days with the 50 mg twice
301
PRITCHETI ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
JACC Vol. 17, No.2 February 1991 :297-303
Table 1. Most Frequent Noncardiac Adverse Experiences* Patients With Paroxysmal Supraventricular Tachycardia (Group I) Flecainide (mg twice daily)
Noncardiac Adverse Experience Headache Dizziness Nausea Dyspnea Flushing Fatigue No. of patients reporting 2: 1 such experience
25 (n = 26)
Placebo (n = 24)1
50 (n = 21)
150 (n = 19)
100 (n = 21)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
I
(4) (4) (4)
(4) (15) (0) (0) (0)
2
3 0 0 0 0 1
(0) (0) (5)
3 3 3 1 2 0
(16) (16) (16) (5)
(0)
0 3 0 1
(10) (5) (0) (14) (0) (5)
(14) (0)
(0) (8)
1 4 0 0 0 0
(29)
6
(23)
7
(33)
6
(29)
10
(53)
1 1 0 0 2 7
(0)
I
(0)
(II)
(0)
Patients With Paroxysmal Atrial Fibrillation or Flutter (Group 2) Flecainide (mg twice daily)
Noncardiac Adverse Experience
25 (n = 43)
Placebo (n = 42)t
50 (n = 40)
150 (n = 33)
100 (n = 38)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
No.
(%)
Dizziness Fatigue Vision abnormal Dyspnea Myalgia Headache Nausea Tinnitus Edema Tremor Chest pain
1 2 2 0 3 2 2
2 2
0 1
(0)
(2)
3 3 3 2 0 3 2 1 1 0 1
(8) (8) (8) (5) (0) (8) (5) (3) (3) (0) (3)
(13)
1 0 2 2 2
(5) (5) (2) (2) (0) (5) (5) (5) (2)
5 3 4 2
1 0 1
(2) (5) (5) (0) (7) (5) (5) (2) (2) (0) (2)
2 1 1 1 2
(3) (5) (3) (3) (3) (5)
9 4 3 3 3 0 2 2 2 2 1
(27) (12) (9) (9) (9) (0) (6) (6) (6) (6) (3)
No. of patients reporting 2: I such experience
13
(31)
10
(23)
12
(30)
14
(37)
18
(55)
I
I
I
I I
(8) (II)
(5) (3)
'This table reports the actual number of patients who received each treatment and the actual number who reported noncardiac adverse experiences with that treatment. Patients could report more than one noncardiac adverse experience at each dose. tOne patient did not report the presence or absence of adverse experiences while receiving placebo.
daily dose; in Group 2 the median time increased from 3 to 5 days, suggesting that this dose was slightly more potent in Group 1 than in Group 2 by this measure. Recognizing the lack of precision in our measurements, we recommend flecainide at 50 mg twice daily as the starting dose for patients with paroxysmal supraventricular tachycardia or paroxysmal atrial fibrillation or flutter. The simplicity of using the same dose for patients with both types of paroxysmal arrhythmia is appealing. Additional improvement can be expected in both patient groups by dose increases. A flecainide dose of 50 mg twice daily appears to be approximately equipotent with verapamil at 480 mg/day (9).
Dose-response relation: projected discontinuations. The analysis of study discontinuations (Fig. 3) suggests that
< 10% of patients would discontinue therapy as a result of noncardiac adverse events at the recommended starting dose of flecainide. The most frequently reported noncardiac adverse experiences among all patients were dizziness, headache, nausea, vision abnormality, dyspnea and fatigue (Table 1); these findings are consistent with previous reports (10) of the use of flecainide in patients with ventricular arrhythmias. Cardiac adverse experiences also were reported. Two patients (one in each group) reported that they had more prolonged or more severe episodes of tachycardia than they had previously experienced. In addition, one patient with paroxysmal atrial fibrillation developed ventricular tachycardia during flecainide treatment; however, circumstances suggest that this arrhythmia could have been due to associ-
302
PRITCHEIT ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
JACC Vol. 17, No.2
February 1991 :297-303
Table 2. Reasons for Discontinuation From the Study Flecainide Dose (mg twice daily) Placebo
25
50
100
150
Total
Patients With Paroxysmal Supraventricular Tachycardia (Group I) Cardiac adverse experience Noncardiac adverse experience Personal Other* Total Total no. receiving each dose
1
I
I I
2 5
2 0 0 3
0 0 0 0 0
0 1 0 0 1
25
26
21
21
19
1
o
2
I I
4 1
7
o 2
o o 2
2 6 I
2 11
Patients With Paroxysmal Atrial Fibrillation or Flutter (Group 2) 0 1
Cardiac adverse experience Noncardiac adverse experience Personal Inadequate response Lost to follow-up Othert Total
4
3
1 0 1 0 0 0 2
41
42
40
I
0 I I
Total no. receiving each dose
0 1 0 1 0 I
0 0 0 3
o
38
33
o
o 5
4 1 I 2 17
*One patient was discontinued from the study because of prestudy laboratory abnormality, and one patient was dbcontinued because he failed to make follow-up calls. tOne patient was discontinued for receiving excluded concomitant medications. One patient was discontinued for a protocol violation.
ated coronary artery disease, although we called it a cardiac adverse experience to be conservative in our interpretation of the therapeutic role of flecainide. The occurrence of proarrhythmia as aggravation of supraventricular arrhythmias and as the new occurrence of ventricular tachycardia has been reported (l, II) in small numbers of patients taking flecainide. Limitations of the study. The current study has some limitations. The dosing scheme would have been improved if the sequence of treatments had been completely randomized. However, we believed that it was unreasonable to randomize patients to begin treatment with flecainide at 150 Figure 3. Projected cumulative study discontinuations due to noncardiac adverse experiences. These calculations suggested that if all patients initiated therapy with the 150 mg twice daily (bid) dose, 22.7% of patients with paroxysmal supraventricular tachycardia (PSVT) and 17.6% of patients with paroxysmal atrial fibrillation (PAF) or paroxysmal atrial flutter (PAF!) would be discontinued from therapy for noncardiac adverse experiences.
! 25...------------------, o
::: tll :::l
c ::: c
20
.PSVT .PAF or PAFI
0 ...... 15
~~
C
10
'C
~ ~
..
l1.
5
0
Placebo
25 mg bid
50 mg bid 100 mg bid 150 mg bid
Treatment
mg twice daily when they had never been exposed to lower doses. Also, the emphasis in the study on developing efficacy data made the dose-response data for tolerance somewhat less rigorous. For example, the discontinuations shown in Figure 3 were projected rather than actually observed for the 50, 100 and 150 mg twice daily doses of flecainide. However, these projections:end to give a conservative picture of the therapeutic role of flecainide, which is appropriate for a new indication. Other limitations were the ~;mall number of patients who completed the study and thl~ relatively short treatment periods, which made it difficult to estimate the true effect of the higher flecainide doses on the time to the first episode of tachycardia and intervals between episodes. These considerations limited the statistical power of some pairwise comparisons of flecainide with placebo; however, overall tests for significance allowed us to reject the null hypothesis that all doses tested were the same. The small number of patients completing the study was partially due to difficulty in enrolling patients. A high degree of cooperation was required from the patients to record and transmit their arrhythmias and the patients had to be willing to endure numerous episodes of tachycardia to complete the s::udy. If one estimates 2 episodes in the screening period, a few episodes that were not counted during the washout days and up to 4 episodes in each of the treatment period~., then patients easily could have had> 12 episodes during the study. Conclusions. Safety issues remain a concern with flecainide and other class IC antiarrhythmic drugs. In the Cardiac Arrhythmia Suppression Trial (CAST) (12), patients
303
JACC Vol. 17. No.2 February 1991:297-303
PRITCHETI ET AL. FLECAINIDE IN SUPRAVENTRICULAR ARRHYTHMIAS
with recent myocardial infarction and frequent premature ventricular beats who were taking ftecainide or the related drug encainide had an increased risk of arrhythmic death compared with patients treated with placebo. No deaths were observed in our study, but the small number of subjects enrolled (n = 73) and the short length of the study (5 months) limited the ability of our study to detect events that occurred with low frequency. The exact role of ftecainide in the management of paroxysmal supraventricular arrhythmias will be determined by continued close observation of its effectiveness and safety as clinical experience accumulates.
Oregon Cardiovascular Teachings, Ltd, Eugene, Oregon: Leonard G. Christie. Jr., MD, Kathe L. Carlson, RN West Penn Hospital, Pittsburgh, Pennsylvania: Barry L. Alpert, MD, Connie M. Fecik, RN Latter Day Saints Hospital, Salt Lake City, Utah: Jeffrey L. Anderson, MD, Fidela Moreno, MD, Linda Johnson. RN, Kaye Summers, RN Hunter Holmes McGuire Veterans Administration Medical Center, Rich· mond, Virginia: Kenneth A. Ellenbogen, MD, Mary Lynn Martin, RN 3M Pharmaceuticals, St. Paul, Minnesota: Michael T. CuUen, Jr., MD, Bruce P. Ekholm, MS, Terrance L. Fox, MS, Patricia A. Fredell, Ronald W. Hawkinson, MS, Mary Jane Maser, Sharon L. Parrish,lnara Porietis, Judy A. Sellers, RN, Kathleen A. Wagenknecht, Richard R. Wilson, MD
Appendix Investigators from the Flecainide Supraventricular Tachycardia Study Group who participated in this study were:
University of Alabama, Birmingham, Alabama: G. Neal Kay. MD, Vance J. Plumb, MD, Andrew E. Epstein, MD, Rosemary S. Bubien, RN. MSN The Hospital of the Good Samaritan, Los Angeles, California: Anil K. Bhandari MD, Cheryl Leon, RN Cardiology and Invasive Cardiac Electrophysiology, Torrance, California: Marc L. Platt, MD, Annelie M. Brinkley, LVN Torrance Memorial Medical Center, Torrance, California: Benjamin L. Rosin, MD, Evan McCabe, RN, MN Indiana University School of Medicine, Indianapolis, Indiana: James 1. Heger, MD, William M. Miles, MD, Lawrence S. Klein, MD, Douglas P. Zipes, MD. Linda Crouter Johns Hopkins Hospital, Baltimore, Maryland: Thomas Guarnieri, MD, Steven DaTorre. MD. Rozann DeBorde, RN University of Massachusetts Medical Center, Worcester, Massachusetts: Charles Haffajee, MB Chir, Karen Rufino. RN. Charlene Degon. RN Henry Ford Hospital Heart and Vascular Institute, Detroit, Michigan: Charles R. Webb. MD, Karen A. Bielinski, RN, BSN Minneapolis Heart Institute, Minneapolis, Minnesota: Irvin F. Goldenberg, MD, Wes R. Pedersen, MD, Adrian Almquist, MD. Jonathan W. Rogers, MD, Richard R. Nelson, MD, James A. DanieL MD. Susanne K. Roeller, PharmD. Donna Hunn. RN, Faye Bul/emer, RN Hennepin County Medical Center, Minneapolis, Minnesota: David Dunbar. MD, Morrison Hodges. MD, Karin Fjeldoo-Sperbeck, RN Mount Sinai Medical Center, New York, New York. J. Anthony Gomes, MD, Stephen Winters, MD, Elena Pe, RN University of North Carolina, Chapel Hill, North Carolina: James R. Foster, MD. Alan Woelfel, MD, Linda Cullinane. RN Duke University Medical Center, Durham, North Carolina: Edward L. C. Pritchett, MD, Elizabeth A. McCarthy, RN Creighton Cardiac Center, Omaha, Nebraska: Syed M. Mohiuddin, MD. Tom T. Hee, MD. Daniel E. Hil/eman, PharmD, Lois A. Stengel, BSN
References I. Anderson JL, Gilbert EM, Alpert BL, et al. Prevention of symptomatic recurrences of paroxysmal atrial fibrillation in patients initially tolerating antiarrhythmic therapy: a multicenter, double-blind, placebo-controlled, crossover study of flecainide and placebo using transtelephonic monitoring. Circulation 1989;6: 1557-70. 2. Henthorn RW, Waldo AL, Anderson JL, et al. Flecainide for prevention of paroxysmal supraventricular tachycardia: a multicenter, double-blind, placebo-controlled, crossover study using transtelephonic monitoring (abstr). J Am Col/ CardioI1988;1I:77A. 3. Pritchett ELC, Lee KL. Designing clinical trials for paroxysmal atrial tachycardia and other paroxysmal arrhythmias. J Clin EpidemioI1988;41: 851-8. 4. Conover WJ. Practical Non-parametric Statistics. 2nd ed. New York: Wiley. 1980:130-3, 199-205. 5. Robson DS. A simple method for constructing orthogonal polynomials when the independent variable is unequally spaced. Biometrics 1959;53: 187-91. 6. O'Brien PC, Fleming Te. A paired Prentice-Wilcoxon test for censored paired data. Biometrics 1987;43: 169-80. 7. Encyclopedia of Statistical Sciences. Vol. I. New York: Wiley, 1982: 294-7. 8. Morganroth J, Pool P, Miller R, Hsu PH, Lee I, Clark DM. Dose· response range of encainide for benign and potentially lethal ventricular arrhythmias. Am J CardioI1986;57:769-74. 9. Pritchett ELC, Hammill SC, Reiter MJ, et al. Life table methods for evaluating antiarrhythmic drug efficacy in patients with paroxysmal atrial tachycardia. Am J CardioI1983;52: 1007-12. 10. Gentzkow GD, Sullivan JY. Extracardiac adverse effects of flecainide. Am J CardioI1984;53:IOIB-5B. 1I. Falk RH. Flecainide-induced ventricular tachycardia and fibrillation in patients treated for atrial fibrillation. Ann Intern Med 1989;111: 107-1 I. 12. CAST Study Group. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;31:406-12.
