Steroids prevent early recurrence of atrial fibrillation following catheter ablation: a systematic review and meta-analysis

Previous studies have reported that steroids may reduce the risk of atrial fibrillation (AF) recurrence after catheter ablation, but data regarding this issue have been controversial. Therefore, we conducted a meta-analysis of randomized clinical trials (RCTs) and observational studies to ascertain the association of steroids and AF recurrence after ablation. PubMed, Embase, and Cochrane online databases were searched from inception to December 2017. The primary outcome of the meta-analysis was short-term or long-term AF recurrence following a single ablation procedure with or without the use of steroids. Both fixed- and random-effects models were used to calculate the overall effect estimates. Eight studies (four RCTs and four observational studies), with a total 992 patients, were included in the present study. Our meta-analysis shows that steroid use was associated with reduced AF occurrence at 3 months (odd ratio (OR) = 0.53, 95% confidence interval (CI) = 0.31–0.90, P=0.02) and 12–14 months (OR = 0.67, 95% CI = 0.47–0.95, P=0.02) after radiofrequency (RF) catheter ablation (RFCA). No clear benefit was observed for AF recurrence at 2–3 days, 1 or 24 months of follow-up. Steroid use was associated with decreased risk of early AF recurrence 3 and 12–14 months after ablation. No clear relationship was observed for 2–3 days, 1 and 24 months of follow-up and further data are needed to clarify these results.


Introduction
Atrial fibrillation (AF) is the commonest cardiac rhythm disorder observed in clinical practice but it is challenging to cure [1,2]. Radiofrequency (RF) catheter ablation (RFCA) is an effective treatment for drug-refractory symptomatic paroxysmal or persistent AF patients. However, there are still opportunities for recurrence of AF, especially within the first few weeks after ablation [3]. The relationship between inflammation process and the development of AF has been widely studied [4,5].
Recent study has found that the early recurrence of AF after ablation is associated with an inflammatory response to the ablation itself [6]. Halonen et al. [7] showed that hydrocortisone can reduce the incidence of AF after cardiac surgery by preventing inflammatory reactions. Some studies suggested that treatment with steroids after AF ablation is effective for negating AF recurrence by preventing post-ablation inflammatory responses, whereas others have reported no significant benefit. Moreover, the effectiveness of steroids for prevention of AF recurrence may vary at different time points after ablation [8][9][10][11][12][13][14][15]. Given these conflicting findings, we conducted a systematic review and meta-analysis of randomized clinical trials (RCTs) and observational studies to ascertain the impact of steroids on preventing AF recurrence after RFCA.

Search strategy
PubMed, Embase, and Cochrane online databases were searched from their inception to December 2017. The following search string: (hormone or steroid or corticosteroid or glucocorticoid or hydrocortisone or methylprednisolone) and ('radiofrequency ablation' or 'pulmonary vein isolation' or 'catheter ablation' or 'radiofrequency catheter ablation') and 'atrial fibrillation' were used to identify all the published articles. Searches were not restricted by language. Reference lists of retrieved articles were scanned manually for additional studies.

Study selection
Inclusion criteria were: (i) the study design was randomized controlled trials (RCTs) or cohort studies, (ii) human subjects, (iii) included the characteristics of study patients, (iv) assessed the effect of steroids on AF recurrence after RFCA. Two reviewers (M.L. and M.G.) independently identified duplicates, and then they conducted the initial screening of all titles or abstracts and evaluated all potentially relevant articles based on full-text reviews. Any disagreements between the two investigators were resolved by discussion with a senior reviewer (T.L.).

Quality evaluation
In order to limit the heterogeneity secondary to differences amongst study designs, the quality of each study was evaluated to exclude those studies classified as 'low quality' . This meta-analysis included four RCT studies and four cohort studies. The quality of observational study was assessed using the Newcastle-Ottawa scale (NOS), and NOS ≥ 7 was considered as a high-quality study. We used Cochrane bias risk tools for quality assessment of RCT study, and finally assigned a judgment of high, low, or unclear risk of material bias for each study. Our NOS analysis for observational studies demonstrated that three studies had high quality scores of ≥ 7 [9][10][11] with one study having a moderate score [15] (Table 3). Cochrane bias risk tool for assessing RCTs demonstrated that all four were high-quality studies (Supplementary Figure S1).

Data extraction
Two reviewers extracted data from the eligibility studies according to standard data extraction forms. We extracted and analyzed odd ratios (ORs)/HRs/RRs values and the corresponding 95% confidence intervals (CIs) to evaluate the association between steroids and AF recurrence after RFCA. At the same time, the following data were also collected from each study: study characteristics (first author's last name, year of publication, origin of the studied population, study population, number of patients, follow-up duration, diagnosis and detection methods of AF recurrence, ablation, dose and type of steroids in the steroids group) and patient characteristics (age, male, percentage of paroxysmal AF, hypertension, diabetes, heart failure, antiarrhythmic, echocardiographic parameters, the procedure time of catheter ablation, superior vena cava (SVC) isolation). Any disagreements between the two reviewers (M.L. and M.G.) were resolved by discussion with a senior reviewer (T.L.).

Statistical analysis
Review Manager (version 5.3) was used for statistical analysis. P<0.05 was considered to be statistically significant. Categorical data were described in percentages and continuous variables were reported as mean and S.D. Results of AF recurrence after RFCA were presented as OR and 95% CI. Heterogeneity was evaluated with Chi-square tests and I 2 , and I 2 ≥ 50% was considered to indicate significant heterogeneity. If I 2 < 50% then a fixed-effects model was used, otherwise the random-effects model was used. Subgroup analyses were based on study design (cohort study and RCT study).

Patient baseline characteristics
A flow diagram detailing the search and study selection process is illustrated in Figure 1. Three hundred and eighty-nine citations were initially retrieved from PubMed, Embase, and the Cochrane online database. Thirty-six  records were discarded as duplicates. Subsequent records including 8 animal studies, 32 case reports, 84 review articles, and 218 irrelevant studies were also excluded accordingly. Subsequently, we excuded 3 records [16][17][18] of duplicated data by reading 11 full-text articles. In the end, a total of eight studies involving 992 patients between 2010 and 2016 met our selection criteria for inclusion [8][9][10][11][12][13][14][15]. The baseline characteristics of all patients are summarized in Tables 1 and 2, respectively. Seven of eight studies [8][9][10][11][13][14][15] examined outcomes of AF patients undergoing RFCA. The follow-up period varied from 6 to 38 months. It is worth noting that different steroids, including hydrocortisone, dexamethasone, prednisone, and methylprednisolone were used in these studies. Furthermore, the steroid administration method also varied between articles.

Efficacy of steroids in preventing AF recurrence post-RFCA
As the impact of steroids on AF recurrence after RFCA were reported at different follow-up periods, we evaluated the association between steroid use and AF recurrence post-RFCA at (i) 2-3 days, (ii) 1, (iii) 3, (iv) 12-14, and (v) 24 months, respectively. Only data from the low-dose steroid group in the Kim et al. study were used for this meta-analysis [9].

AF recurrence with steroids 2-3 days after RFCA
Three studies reported AF recurrence 2-3 days after RFCA [8,11,13]. Of these, one was an RCT that showed that steroids reduced AF recurrence [8], whereas the remaining two studies found no observable benefit [11,13]. Though insignificant, pooled analysis of the included studies demonstrated that patients treated with steroids had lower odds of AF recurrence when compared with the control group (OR = 0.46, 95% CI = 0.17-1.24, P=0.13; Figure 2). Similarly, subgroup analyses into two RCTs also showed that steroid use was not significantly associated with decreased  risk of AF recurrence [8,13] (OR = 0.31, 95% CI = 0.08-1.16, P=0.08; Figure 2). The overall heterogeneity was low across subgroups (I 2 = 39%).

AF recurrence with steroids 1 month following RFCA
Three studies reported AF recurrence 1 month after RFCA [8,11,13]. Our analysis showed that the use of steroids post-RFCA is not associated with an increased risk of 1-month AF recurrence (OR = 0.92, 95% CI = 0.41-2.08, P=0.85; Figure 3). Subsequent analyses based on study design also revealed that steroid therapy was not associated with AF recurrence in 1 month after RFCA, whether in RCTs [8,13] (OR = 0.58, 95% CI = 0.28-1.18, P=0.13; Figure  3) or in the cohort study [11] (OR = 1.68, 95% CI = 0.93-3.06, P=0.09; Figure 3). However, the heterogeneity was found to be high across subgroups (I 2 = 80%). To locate the origin of the heterogeneity, sensitivity analysis excluding one study at a time was performed. Doing so did not significantly alter the overall heterogeneity.

Figure 2. Association between steroids and AF recurrence in 2-3 days after ablation
Subgroup analysis based on study design was also presented.

Figure 3. Association between steroids and AF recurrence in 1 month after ablation
Subgroup analysis based on study design was also presented.

Figure 4. Association between steroids and AF recurrence in 3 months after ablation
Subgroup analysis based on study design was also presented.

Figure 5. Association between steroids and AF recurrence in 12-14 months after ablation
Subgroup analysis based on study design was also presented.

Discussion
The main findings of this systematic review and meta-analysis are that steroid use is associated with a decreased risk of AF recurrence 3 and 12-14 months after RFCA. On the other hand, there seems to be no significant association between steroid therapy and risk of AF recurrence in 2-3 days, 1 and 24 months after RFCA. However, we found a significant heterogeneity, which could have been attributed to the type of AF (paroxysmal/persistent), intensity of ECG monitoring on follow-up, lesion sets deployed, doses of steroids used, and the time points on follow-up at which recurrences were evaluated. This heterogeneity is likely responsible for the conclusion that short-term steroid use decreases AF recurrence at 3 and 12 months, but not at 2-3 days and 1 month. Effective doses or dosing intervals to decrease AF recurrence are not fully established, even in cardiac surgery studies.
Inflammation induced by ablation can promote acute AF recurrence [19], which plays an important role during this period, increasing oxidative stress, promoting fibrosis, and facilitating re-entry [20][21][22][23]. It has been demonstrated that colchicine is an effective and safe agent for preventing early AF recurrences after pulmonary vein isolation in patients with paroxysmal AF, associated with reduced CRP and IL-6 levels [24]. The use of steroid to prevent AF has mainly been evaluated in cardiothoracic surgery. Previous meta-analyses have reported that steroid treatment is associated with a decrease in more than 50% postoperative cases [25]. The use of steroids is beneficial in this regard given their anti-inflammatory properties. Indeed, the levels of inflammatory cytokines immediately post-ablation were lower in the steroid group compared with the control group [14]. Moreover, CRP levels 3 days after ablation were lower in the corticosteroid group [8,9]. Furthermore, Koyama et al. [8] reported that AF recurrence 1 month after RFCA was not related to changes in inflammatory markers such as CRP. However, factors other than CRP in the pro-inflammatory signaling cascade can mediate pro-fibrotic changes. Given the main role of inflammation in immediate and early AF recurrence after ablation, treatments targetting this process may have a great potential to ameliorate this adverse event improving prognosis. Steroids can inhibit pro-inflammatory response, suppress VEGF expression and inhibit fibroblast proliferation [26]. These effects can modulate the pro-arrhythmic substrate and reduce conduction defects [27].
The major cause of longer term recurrence may be associated with electrical reconnection between PVs and LA after apparently successful initial isolation. The occurrence of this reconnection results from deficiencies of the index ablation procedure [26]. Rizzo et al. observed that α-lipoic acid (ALA) therapy could reduce inflammation stress, but could not prevent AF recurrence at 12-month follow-up after catheter ablation [28]. In this study, we found that steroid therapy decreased the risk of AF recurrence in 12-14 months after RFCA. However, it seems fairly implausible that steroid use shortly after PVI could prevent re-conduction between the LA and PVs over the long term. First, structural and electrical remodeling takes place within a few hours of AF onset, whereas reverse remodeling after restoration of sinus rhythm occurs much more slowly [29]. The use of steroids may halt electrical or functional remodeling of the atria, thereby permitting reverse remodeling to occur. Second, AF also induces inflammation, which in turn can perpetuate AF [30]. Steroid therapy shortly after RFCA might halt the relationship between inflammation and AF, which may represent a vicious cycle. Third, gaps of sufficient cross-sectional dimensions within linear RF lesions may promote recovery of conduction [31]. Steroids can potentially prevent AF recurrence by suppressing the delayed extension of the RF lesion between the ablation points [12].
Indeed, Andrade et al. [10] suggested that steroids use remained independently associated with the high prevalence of dormant PV conduction unmasked by adenosine. Therefore, a balance of anti-inflammatory action and delayed extension of the RF lesion determines the net pro-or anti-arrhythmic effect of steroids. Moreover, ablation can lead to transient autonomic dysfunction by increasing the sympathetic tone and decreasing the parasympathetic tone [32]. These could promote arrhythmias by both triggered activity and re-entry [33].

Study limitations
Several potential limitations of this meta-analysis should be acknowledged. First, our analysis pooled together both RCTs and cohort studies, but subgroup analyses based on study design were also performed. Second, our study involved a relatively small number of subjects (n=992). Third, the moderate heterogeneity observed was only partially explained by study design, and other factors such as steroid dose may play a role. Steroid dosing was not differentiated in most of the included studies, it was therefore not possible to derive dose-response relationships between steroids and AF recurrence after ablation. Fourth, in most of studies, although we identified AF recurrence by using the ECG and Holter monitoring during follow-up, some asymptomatic intermittent AF events might have been missed. Fifth, less than one-third women were included in the studies. Thereby, data may not be applicable to female patients.