Author + information
- Received October 20, 2009
- Revision received March 10, 2010
- Accepted April 13, 2010
- Published online October 26, 2010.
- Takashi Koyama, MD⁎ (, )
- Hiroshi Tada, MD,
- Yukio Sekiguchi, MD,
- Takanori Arimoto, MD,
- Hiro Yamasaki, MD,
- Kenji Kuroki, MD,
- Takeshi Machino, MD,
- Kazuko Tajiri, MD,
- Xu Dong Zhu, MD,
- Miyako Kanemoto-Igarashi, MD,
- Aiko Sugiyasu, MD,
- Keisuke Kuga, MD,
- Yoshio Nakata, PhD and
- Kazutaka Aonuma, MD
- ↵⁎Reprint requests and correspondence:
Dr. Takashi Koyama, Cardiovascular Division, Institute of Clinical Medicine, Graduate School of Comprehensive Human Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8575, Ibaraki, Japan
Objectives We sought to clarify the efficacy of corticosteroid therapy for preventing atrial fibrillation (AF) recurrence after pulmonary vein isolation (PVI).
Background The inflammatory process may cause acute AF recurrence after PVI. However, no studies have examined the relationship between corticosteroid administration and AF recurrence after PVI.
Methods A total of 125 patients with paroxysmal AF were randomized to receive either corticosteroids (corticosteroid group) or a placebo (placebo group). In the corticosteroid group, intravenous hydrocortisone (2 mg/kg) was given the day of the procedure, and oral prednisolone (0.5 mg/kg/day) was administered for 3 days after the PVI. The body temperature and high-sensitivity C-reactive protein level were measured before and on each of the first 3 days after ablation.
Results The prevalence of immediate AF recurrence (≤3 days after the PVI) was significantly lower in the corticosteroid group (7%) than in the placebo group (31%). The maximum body temperature and C-reactive protein during the initial 3 days after ablation and the increase in the body temperature and C-reactive protein level from baseline were significantly lower in the corticosteroid group than in the placebo group. Corticosteroid treatment did not decrease AF recurrences between 4 and 30 days after ablation. The AF-free rate at 14 months post-ablation was greater in the corticosteroid group (85%) than in the placebo group (71%, p = 0.032 by the log-rank test).
Conclusions Transient use of small amounts of corticosteroids shortly after AF ablation may be effective and safe for preventing not only immediate AF recurrences but also AF recurrences during the mid-term follow-up period after PVI.
Pulmonary vein isolation (PVI) is becoming an effective therapeutic option for drug-refractory, paroxysmal atrial fibrillation (AF). However, the rate of AF recurrence after successful PVI remains relatively high (1), and recurrences commonly occur within the first several weeks after the PVI (2–4). There is much evidence of a mechanistic link between inflammatory processes and the development of AF (5,6). Electrical PVI by radiofrequency energy causes extensive myocardial damage, which may result in a systemic and local inflammatory response (7,8). Recent studies have found that inflammatory responses appeared shortly after PVI (7,8) and that those responses were closely associated with AF recurring within the initial 3 days after the procedure (7). Several studies have reported that corticosteroids reduce AF after cardiac surgery by preventing inflammatory reactions (9). Therefore, we hypothesized that corticosteroid therapy might negate AF recurrences immediately after PVI by preventing post-ablation inflammatory responses. This study was designed and undertaken to clarify this point.
Between April 2007 and March 2008, 130 consecutive patients with drug-refractory AF were enrolled in this study (Table 1).All patients were newly enrolled in this study, and none had a history of ablation. All patients had symptomatic, paroxysmal AF, which was defined as AF episodes that spontaneously terminate and last for >30 s and <7 days during treatment with antiarrhythmic drugs (AADs) (7). All AADs were discontinued 5 half-lives before the ablation procedure with the exception of amiodarone, which was discontinued at least 6 weeks before and the other drugs were administered until the end of the study period. The study was approved by an institutional review committee, and all patients gave their written informed consent before participation.
This study was designed as a prospective, randomized, double-blind study. All patients were randomized for treatment with corticosteroids (corticosteroid group) or a placebo (placebo group) on the first day of their hospitalization and were observed prospectively after a single PVI procedure (Fig. 1).
In the corticosteroid group, 1 mg per 1 ml hydrocortisone sodium succinate (Solu-Cortef, Pfizer Manufacturing, Puurs, Belgium) in a 0.9% sodium chloride solution was administered intravenously (2-mg/kg dose) immediately after the procedure. Oral prednisolone (Takeda Pharmaceutical Co., Tokyo, Japan) was also administered (0.5 mg/kg/day) for 3 days after the PVI. A placebo solution (0.9% sodium chloride, 1 ml/kg) and oral drug (lactose) were administered in a manner similar to that in the placebo group patients.
Within each group, the patients were divided into 3 subgroups according to the results of the AF recurrence during the initial 1 month after ablation (7): immediate AF recurrence group, early AF recurrence group, and non-AF recurrence group. Immediate and early AF recurrences were defined as AF episodes that occurred within 3 days and between 4 and 30 days after the PVI, respectively. Non-AF recurrence was defined as no AF after the PVI.
The primary end point of the study was the proportion of patients free of recurrent AF, and the secondary end point was recurrent AF and any serious complications caused by catheter ablation or treatment with corticosteroid.
Extensive PVI was performed using the double Lasso technique (7) without a 3-dimensional mapping system. A 7-F deflectable catheter with an 8-mm distal electrode (Ablaze, Japan Lifeline Co., Tokyo, Japan) was used for ablation in the temperature control mode at a target temperature of 52°C and maximum power output of 30 to 35 W. Two fluoroscopic angles were used for confirming the catheter position. The end point of the PVI was the creation of a bidirectional conduction block from the atrium to the PVs and vice versa (7). After the PVI, if AF was sustained or induced with coronary sinus burst pacing at a cycle length down to 180 ms during the administration of intravenous isoproterenol (1.0 to 3.0 μg/min) and lasted more than 3 min (7,10), additional ablation, consisting of linear ablation of the left atrial roof and/or superior vena cava isolation, was performed. If the AF did not terminate or was inducible after these procedures, sinus rhythm was restored by transthoracic cardioversion (10). A cavotricuspid isthmus block line was created in all patients with confirmation of bidirectional block.
Patients remained hospitalized under continuous rhythm monitoring (CNS-970, Nihon Kohden, Tokyo, Japan) for at least 1 week after the procedure and were followed for 14 months after the PVI procedure. After discharge, all patients underwent follow-up at 2 weeks post-procedure and then every month thereafter. At each hospital visit, the patients underwent 12-lead electrocardiography and intensive questioning regarding any arrhythmia-related symptoms. Holter monitoring for 24 h was performed at 2 weeks and 1, 3, 6, and 14 months after the PVI. Portable electrocardiographic monitoring (HCG-901, OMRON, Kyoto, Japan) was also performed for 3 min twice daily (morning and night for a total 6 min/day) on 3 consecutive days at 2 weeks and 1, 3, 6, and 14 months after the PVI. Twelve-lead electrocardiography, 24-h Holter electrocardiography, and portable electrocardiographic monitoring were also used any time that the patients reported palpitations. If the electrocardiogram showed any AF episodes during follow-up, patients received a diagnosis of clinical recurrence of AF, irrespective of the presence of symptoms (7). Patients with AF recurrence were treated temporarily with Class I, II, III, and/or IV AADs, and the AADs were discontinued if stable sinus rhythm could be maintained. Discontinuation of the AADs was then attempted in those patients in whom the recurrent AF disappeared after temporary treatment; if sinus rhythm could be maintained, the drugs were stopped permanently. Repeat ablation was not performed in any of the patients during the study period.
Other definitions and data analysis
The patients' body temperature (BT) was measured on the morning of the ablation and every 6 h during the first 3 consecutive days after the PVI; the highest BT and maximum difference in the BT (ΔBT), defined as the difference between the highest BT during the first 3 days post-procedure and the BT measured before the ablation, were obtained. The BT was measured by an electrical axillary thermometer (C202, TERUMO, Tokyo, Japan) placed in the axilla of the arm not receiving a fluid infusion. After the patient rested in bed for >20 min, the patient's axilla was wiped free of sweat, and the BT was measured 3 times and averaged. The highly sensitive C-reactive protein (CRP) level was also measured before and for 3 consecutive days after the PVI procedure; the maximum CRP level and rise in the CRP (ΔCRP), defined as the difference between the maximum CRP and CRP before the ablation, were obtained (7). During the first 3 days after ablation, all patients were examined for the presence of frequent atrial premature contractions (APCs), defined as ≥10 APCs within 1 min, and nonsustained AF lasting for <30 s (7).
The recurrence of AF within the first month only was considered transient (3,11). This first 1-month period was applied as a blanking period and was used to classify the patients into the 3 AF recurrence subgroups and to account for the fact that AF recurrences early after the PVI may be transient and may not necessarily imply a failed ablation. Only the prevalence of AF recurrences during the period from 31 days to 14 months after the ablation was examined.
From the results of a single procedure outcome of PVI for paroxysmal AF (12,13) and a relative decrease in AF occurrences after cardiac surgery (9), we assumed that sinus rhythm would be present at 14 months after ablation in 80% of patients who received corticosteroid therapy after the ablation and in 50% of patients who did not. We estimated that a minimum of 52 patients were required in each group to give the study a statistical power of 90% with a 2-tailed alpha value of 0.05. Continuous variables are expressed as the mean ± SD. A Student ttest or analysis of variance with a Tukey analysis was used to analyze the differences between groups, as appropriate. For non-normally distributed data, a Kruskal-Wallis test was used when comparing more than 3 groups, and the Mann-Whitney Utest was used when comparing 2 groups. Categorical variables were compared by a chi-square analysis and Yates correction, if necessary. A univariate and multivariate logistic regression analysis was used to identify predictors of immediate and early AF recurrences. A Cox regression analysis was also used to identify predictors of AF recurrences during the 14-month follow-up period. All parameters with a significance <0.10 in the univariate analysis were entered into the multivariate model. A receiver-operator characteristic curve analysis was used to determine the ability of the BT or CRP level to predict AF recurrences. The time to AF recurrences was estimated by the Kaplan-Meier method, with comparisons made using the log-rank test. A p value <0.05 was considered significant.
Study design and participants
In the corticosteroid group, 5 patients were excluded from this study: 3 refused to participate, and 2 patients with non-pulmonary vein (PV) foci from the superior vena cava that triggered AF before the transseptal puncture, did not undergo PVI. No patients in the placebo group were excluded from this study. As a result, 125 patients were enrolled and randomized under this protocol: corticosteroid group (n = 60) and placebo group (n = 65) (Fig. 1).
Baseline characteristics and acute results of catheter ablation
Although the duration of an AF history was longer in the corticosteroid group than placebo group (p < 0.01), no significant differences were found in terms of the other variables between the 2 groups (Table 1). Extensive PVI and cavotricuspid isthmus ablation were successfully performed in all patients, and bidirectional conduction block was created at all 4 PVs and the cavo-tricuspid isthmus. Linear ablation of the left atrial roof and/or superior vena cava isolation were performed in 10 patients with sustained AF (n = 3) or induced AF lasting more than 3 min (n = 7) (Table 1). A mitral isthmus line was not created in any of the patients. Catheter ablation parameters and prevalence of additional ablation were comparable between the 2 groups (Table 1).
AF recurrence during the initial 1-month follow-up period after ablation
During the initial 1 month after ablation, AF recurrences were observed less frequently in the corticosteroid group (16 patients, 27%) than in the placebo group (32 patients, 49%; p < 0.01) (Fig. 1). A subgroup analysis of the patients in the 2 groups showed the prevalence of early AF recurrences in the corticosteroid group (20%) was comparable to that (18%) in the placebo group (p = 0.9). However, the prevalence of immediate AF recurrences in the corticosteroid group (7%) was less than that in the placebo group (31%; p < 0.001) (Fig. 1).
Inflammatory responses, pericarditis, and atrial arrhythmias during the initial 3 days after ablation
During the initial 3 days after the PVI, the highest BT (p < 0.005) and ΔBT (p < 0.001) were both lower in the corticosteroid group (Fig. 2A).The maximum CRP and ΔCRP in the corticosteroid group were also lower than those in the placebo group (both, p < 0.001) (Fig. 2B). Only 1 placebo group patient reported slight chest pain transiently with deep breaths, but no pericardial effusion or electrocardiographic changes suggestive of pericarditis were found. The prevalence of frequent APCs and nonsustained AF did not differ between the 2 groups (Fig. 2C).
Subgroup analysis of inflammatory responses
In the placebo group, both the highest BT and ΔBT during the initial 3 days after the PVI were greater in the immediate AF recurrence group than the other 2 subgroups (both p < 0.001) (Fig. 3A);however, in the corticosteroid group, these variables were comparable among the 3 subgroups (Fig. 3B). In the placebo group, both the maximum CRP and ΔCRP in the immediate AF recurrence group were the highest among the 3 subgroups (maximum CRP; p = 0.001, ΔCRP; p < 0.001) (Fig. 4A).However, in the corticosteroid group, there was no significant difference in these variables among the 3 subgroups (Fig. 4B). The leukocyte count was comparable among the 3 subgroups of both groups (data not shown).
In the placebo and corticosteroid groups, frequent APCs and nonsustained AF were observed more often in the early AF recurrence group than in the immediate AF recurrence and no AF recurrence groups, and the differences among the 3 subgroups were statistically significant in both groups (Figs. 5A and 5B). No significant differences were found for these variables among the subgroups of the placebo and corticosteroid groups.
Post-ablation course and resolution of AF recurrences
During the ablation procedure and 14-month follow-up, no major complications occurred in any patients. Groin hemorrhages occurred in 2 patients but resolved with manual and bandage compression.
In the placebo group, recurrent AF attacks disappeared in 14 (70%) of 20 patients with immediate AF recurrences and in 4 (33%) of 12 with early AF recurrences between 31 days and 14 months after the PVI (Fig. 1). However, in the corticosteroid group, recurrent AF attacks disappeared in all 4 (100%) patients with immediate AF recurrences and in 6 (50%) of 12 with early AF recurrences during that period. Thus, the prevalence of no AF recurrences without any antiarrhythmic drugs at 14 months post-ablation was greater in the corticosteroid group (85%) than in the placebo group (71%; p < 0.05) (Fig. 1). No sustained atrial tachycardia was observed in any patients in either group during this period.
The prevalence of an AF-free rate at 14 months post-ablation in the subgroup analysis of the patients with immediate AF recurrences was higher in the corticosteroid group than in the placebo group (4 of 4 [100%] vs. 14 of 20 [70%]; p < 0.05) (Fig. 1). However, it was comparable in the patients with early AF recurrences (corticosteroid group, 6 of 12 [50%] vs. placebo group, 14 of 20 [33%], p = 0.43) (Fig. 1).
A Kaplan-Meier survival analysis showed that the AF-free rate without any antiarrhythmic drugs at 14 months post-ablation was greater in the corticosteroid group than in the placebo group (p = 0.032 by the log-rank test) (Fig. 6).
Predictors of AF recurrence after ablation
A logistic regression analysis revealed that the BT (odds ratio [OR]: 32.33; 95% confidence interval [CI]: 7.076 to 147.7; p < 0.001) and CRP (OR: 1.739; 95% CI: 1.007 to 3.005; p = 0.047) were associated with immediate AF recurrences (Table 2).APCs (OR: 15.96; 95% CI: 4.000 to 63.66; p < 0.001), and nonsustained AF during the initial 3 days after ablation (OR: 20.57; 95% CI: 5.184 to 81.62; p < 0.001) were associated with early AF recurrences (Table 3).
In the Cox regression analysis including the highest BT, maximum CRP, frequent APCs, nonsustained AF, and corticosteroid treatment, APCs (hazard ratio: 3.560; 95% CI: 1.809 to 7.007; p < 0.001), and nonsustained AF (hazard ratio: 3.526, 95% CI: 1.856 to 6.697; p < 0.001) were associated with AF recurrences during the follow-up period (Table 4).Corticosteroid treatment decreased AF recurrences during the follow-up period (hazard ratio: 0.458; 95% CI: 0.226 to 0.929; p = 0.031). By a receiver-operator characteristic analysis, a post-CRP level of 1.21 mg/dl was the threshold value for AF recurrences with 91.7% sensitivity and 97.2% specificity (area under the curve: 0.871, p < 0.001) and a BT of 37.6°C with 79.2% sensitivity and 95.0% specificity could distinguish patients with AF recurrences (area under the curve: 0.905, p < 0.001).
The present study showed the following: 1) acute inflammatory responses occurring during the initial 3 days after PVI were greater in patients with immediate AF recurrences than in those with early AF recurrences; 2) corticosteroid treatment inhibited inflammatory responses and decreased immediate AF recurrences; 3) corticosteroid treatment was associated with fewer AF recurrences at 1 month after PVI and during the 14 months after the initial 1-month follow-up; and 4) corticosteroid treatment did not reduce early AF recurrences or affect the AF-free rate at 14 months in patients with early AF recurrences or no AF recurrences within 1 month after PVI.
Inflammation and AF recurrences after ablation
The CRP levels were more than 2-fold higher in patients with AF than in control patients, and the persistent AF patients had higher CRP levels than those with paroxysmal AF (14). A recent study demonstrated that the leukocyte count was an independent pre-ablative predictor of AF recurrences after PVI (5). These findings suggest that non- or pre-operative inflammation may play a prominent role in the etiology, initiation, and maintenance of AF (5,6,15).
It is well-known that AF often occurs several days after cardiac surgery (9) as well as after successful electrical cardioversion (6,16). The serum CRP and cytokine levels correspond to AF recurrence rates after cardiac surgery (17), and CRP levels can be elevated within the first 24 h after AF initiation after successful electrical cardioversion (6). In the PVI procedure, extensive radiofrequency energy delivered throughout the left atrium may result in significant and extensive histopathologic tissue damage (18). Therefore, we think that in the genesis of AF shortly after a successful AF ablation, the inflammation surge caused by the procedure is important, and the transient use of very low-dose corticosteroids is effective for inhibiting immediate AF recurrences after PVI.
In the present study, early AF recurrences occurring within 4 days to 1 month after ablation were not associated with an inflammatory response, which is the same result obtained in a previous study (7). Furthermore, in patients with early AF recurrences, the prevalence of an AF-free rate at 14 months did not differ between the 2 groups. Recovered conduction between the left atrium and PVs and the appearance of non-PV foci triggering AF might be responsible for the early AF recurrences (19).
Effect of corticosteroids on late recurrences of AF after PVI
We found that corticosteroid treatment decreased the overall prevalence of AF recurrence at 14 months post-ablation by preventing immediate AF recurrences caused by inflammatory responses. Because the biological half-life of prednisolone is 12 to 36 h and because it was used only for the initial 3 days after ablation in the present study, the effects of prednisolone, including suppression of the inflammatory response, could last for more than these first 3 days. However, it seems fairly implausible that corticosteroid use shortly after PVI could prevent reconduction between the left atrium and PVs over the long term.
Structural and electrical remodeling commences within a few hours of the onset of AF, whereas reverse remodeling after restoration of sinus rhythm occurs much more slowly (20). This longer AF-free period after ablation in patients with corticosteroids might have enabled the atria to halt any electrical or functional remodeling and allowed reverse remodeling to progress, resulting in a greater AF-free rate at 14 months post-ablation. Atrial tachyarrhythmias also cause inflammation (21), and such induced inflammation might cause AF. Corticosteroid treatment administered shortly after PVI might halt the relationship between inflammation and rapid atrial arrhythmia formation, which may represent a vicious cycle.
Utility and significance of early reablation procedures
The efficacy of an early repeat ablation after the initial ablation procedure has been reported (22). However, several studies (2–4,7) and our present results show that a delayed cure for AF despite early recurrence after PVI is found in a considerable number of patients, and the prevalence in patients with AF recurring within a few days to 1 month after PVI is reported to be as high as 76% (7) and 57% (2), respectively. Therefore, in patients with early recurrence and especially with AF recurrence within a few days after the ablation, relatively quick resolution of the AF episodes can be expected, and an early reablation procedure may not be indicated in most of these patients.
First, this study included a relatively small number of patients and a short follow-up period, and some patients with asymptomatic AF recurrence might have been missed. Second, the effect and risk of various doses of corticosteroids on AF recurrences after PVI were not examined or determined. Third, our findings were not confirmed by a protocol using different ablation methods, ablation catheters, or energy settings. Fourth, the serum CRP and BT are nonspecific markers of inflammation, especially in the acute phase, and the origin of those inflammatory responses after the ablation has not been clarified. Finally, the precise mechanism of corticosteroid use shortly after PVI for preventing AF in the long term could not be clarified.
Transient use of small amounts of corticosteroids shortly after AF ablation may be effective and safe for preventing not only immediate AF recurrences but also AF recurrences during the mid-term follow-up period after PVI.
The authors have reported that they have no relationships to disclose.
- Abbreviations and Acronyms
- antiarrhythmic drug
- atrial fibrillation
- atrial premature contraction
- body temperature
- confidence interval
- C-reactive protein
- odds ratio
- pulmonary vein
- pulmonary vein isolation
- Received October 20, 2009.
- Revision received March 10, 2010.
- Accepted April 13, 2010.
- American College of Cardiology Foundation
- Shah A.N.,
- Mittal S.,
- Sichrovsky T.C.,
- et al.
- Oral H.,
- Knight B.P.,
- Ozaydin M.,
- et al.
- Issac T.T.,
- Dokainish H.,
- Lakkis N.M.
- Oral H.,
- Chugh A.,
- Lemola K.,
- et al.
- O'Neill M.D.,
- Wright M.,
- Knecht S.,
- et al.
- Stabile G.,
- Bertaglia E.,
- Senatore G.,
- et al.
- Chung M.K.,
- Martin D.O.,
- Sprecher D.,
- et al.
- Letsas K.P.,
- Weber R.,
- Bürkle G.,
- et al.
- Tieleman R.G.,
- Van Gelder I.C.,
- Crijns H.J.,
- et al.
- Zaman A.G.,
- Archbold R.A.,
- Helft G.,
- et al.
- Ouyang F.,
- Antz M.,
- Ernst S.,
- et al.
- Allesie M.,
- Ausma J.,
- Schotten U.,
- et al.