ABC | Volume 114, Nº4, April 2020

Original Article Correia et al. CA lesions and AF recurrence - a meta-analysis Arq Bras Cardiol. 2020; 114(4):627-635 would be included for qualitative or quantitative analysis. In the event of disagreement, the authors reached a decision through discussion and consensus. Inclusion criteria for qualitative analysis We included observational studies (with prospective or retrospective design) in humans, whose objective was to study the association between post-ablation LA scarring and AF recurrence after CA. Studies that met the following criteria were included: 1) The study evaluated AF or total arrhythmia recurrence after CA in human subjects; 2) The publication was an original study; 3) The mean follow-up period was equal to or longer than 3 months; 4) The study included more than 20 subjects; 5) The study evaluated LA scarring by LGE-MRI after CA. Inclusion criteria for quantitative analysis Meta-analysis included studies that met the previous qualitative analysis criteria and reported means and 95% confidence intervals (CI) of total LA scarring in patients with and without AF recurrence after CA. Quality assessment Risk of bias in the studies was evaluated by the National Heart, Lung and Blood Institute Quality Assessment Tool for Case Series Studies. 7 Evaluation was independently conducted by two raters (ETOC and LMSB), and, in the event of disagreement, the raters reached a decision by consensus. The following characteristics were assessed: 1) Was the study question or objective clearly stated? 2) Was the study population clearly and fully described, including a case definition? 3) Were the cases consecutive? 4) Were the subjects comparable? 5) Was the intervention clearly described? 6) Were the outcome measures clearly defined, valid, reliable, and implemented consistently across all study participants? 7) Was the length of follow-up adequate? 8) Were the statistical methods well‑described? 9) Were the results well described? Following assessment of those characteristics, the authors assigned a quality rating (good, fair, or poor) to each of the studies. Studies were rated as ‘poor’ if they met fewer than three criteria; ‘fair’ if they met three to five criteria; and ‘good’ if they met more than five criteria. All studies selected met almost all of the criteria and received a good quality rating from both raters. Quality assessment of the included studies is reported in Table 1. Data extraction Using a standard data extraction form, two researchers (ETOC and LMSB) performed data extraction, which was cross- verified by a third researcher (ETM). Extracted data included the following: 1) First author’s last name and publication year; 2) Characteristics of included studies: number of patients, study region, study design, ablation strategy, measurement method of LA scarring, method of AF detection, length of follow-up period, and main findings; 3) Outcome results: means and 95% CI of total LA scarring in patients with and without AF recurrence after CA. Statistical analysis The association between AF recurrence and total LA scarring following RFCA was measured by standardized mean difference (SMD) with 95% CI, and standard errors were determined using the corresponding 95% CI. The inverse variance method was used to weigh studies for combined statistical analysis. Statistical significance was defined as p value < 0.05. Heterogeneity between studies was assessed using Cochran’s Q test and I² statistics and subsequently evaluated by I² values. I² values below 30% were defined as low heterogeneity; values between 30% and 60% were considered moderate heterogeneity; and values above 60% were considered high heterogeneity. 8 The fixed-effects model was chosen due to the small number of studies included and the low heterogeneity. Meta-regression was not carried out due to the small number of studies included. The results are reported in a forest plot with 95% CI. Publication bias was verified using a funnel plot. All analyses were conducted using Review Manager 5.3 software. Results Study selection Initially, a total of 790 studies were identified by the database search, 695 in PubMed and 95 in the Cochrane Central Register of Controlled Trials. Duplicate analysis revealed 28 duplicates, which were subsequently eliminated. After careful reading of the title and abstract, 742 of the 762 studies were excluded, because they were not related to the present review. Twenty studies were analyzed in full text, twelve of which were excluded, because they were not related to the present review. Finally, eight studies 9-16 were included in the qualitative analysis, and four were included in the meta-analysis. 9-11,15 The study selection flow diagram is shown in Figure 1. Characteristics of the included studies Eight studies were included in this review, 9-16 comprising six prospective single center observational studies and two prospective multicenter studies (Table 1). The systematic review included a total of 703 patients, and meta-analysis included 295. The follow-up period ranged from 3 to 12 months. All studies used LGE-MRI to identify post-CA LA scarring. Pulmonary vein isolation (PVI) was the ablation strategy in all of the studies. The studies by Akoum et al. 14 and Hunter et al. 16 used both catheter and cryoballoon ablation. Table 1 and Table 2 summarize the characteristics of all included studies. Total LA scarring post-ablation and AF recurrence Six of the eight included studies 8-12,14 found that the extent of LA scarring was associated with less AF recurrence after CA. In the study by Hunter et al., 16 there was no significant association between identification of ablation lesions and freedom from AF (53% with ablation lesions identified remained free from AF vs. 65% in those with no lesions 628