The association between obstructive sleep apnea (OSA) and atrial fibrillation (AF) is strong and is now well established. However, studies on the role of OSA on AF recurrence after catheter ablation have yielded conflicting results. The aim of the present study was to investigate the role of OSA on AF recurrence after catheter-based pulmonary vein isolation. We performed a data search on the PubMed, Web of Science, and the Cochrane databases for studies published by August 2010. In addition, we manually searched the conference proceedings of the European Society of Cardiology, American College of Cardiology, and American Heart Association for related abstracts. After the initial search returned 402 reports, we identified 6 studies with a total of 3,995 patients that met our inclusion criteria. Overall, patients with OSA have a 25% greater risk of AF recurrence after catheter ablation than those without OSA (risk ratio 1.25, 95% confidence interval 1.08 to 1.45, p = 0.003). Subgroup analysis showed that OSA diagnosed using polysomnography is a strong predictor of AF recurrence (risk ratio 1.40, 95% confidence interval 1.16 to 1.68, p = 0.0004) but not when OSA was diagnosed using the Berlin questionnaire (risk ratio 1.07, 95% confidence interval 0.91 to 1.27, p = 0.39). In conclusion, patients with OSA have significantly greater AF recurrence rates after pulmonary vein isolation. In addition to other factors, a diagnosis of OSA merits special consideration when evaluating patients for catheter-based AF ablation.
The relation of obstructive sleep apnea (OSA) and atrial fibrillation (AF) is well established, and a strong correlation between these 2 entities has been demonstrated. However, previous studies of OSA as a predictor for AF recurrence after catheter ablation have yielded conflicting results. In this comprehensive meta-analysis and systematic review, we sought to examine the present evidence and investigate whether the presence of OSA increases the risk of AF recurrence after catheter ablation.
Methods
We performed the present analysis according to the guidelines of the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. Two investigators (C.N., T.L.) performed a search independently using the on-line databases of PubMed, Web of Science, and Cochrane Central Register of Controlled Trials, using the keywords “sleep apnea,” “sleep disordered breathing,” and “atrial fibrillation” for studies published before August 2010. In addition, we also manually searched the conference proceedings from the European Society of Cardiology, American College of Cardiology, and American Heart Association for any related abstracts. The inclusion criteria for our analysis included studies with the following characteristics: (1) human subjects; (2) retrospective/prospective cohort studies; (3) catheter ablation for AF performed; (4) assessed for the presence of OSA; (5) compared AF recurrence after single catheter ablation for patients with and without OSA; and (6) follow-up interval >6 months.
To limit the heterogeneity secondary to differences among study designs, the quality of each study was evaluated according to the guidelines developed by the United States Preventive Task Force and the Evidence-Based Medicine Working Group. A point score system was applied according to the quality of the study. The following characteristics were assessed: (1) clear description of inclusion and exclusion criteria; (2) study sample representative for mentioned population; (3) clear description of sample selection; (4) full specification of clinical and demographic variables; (5) follow-up interval of ≥6 months; (6) reporting loss of follow-up; (7) clear definition of OSA; (8) clear definition of outcomes and outcome assessment; (9) temporality (assessment of OSA before catheter ablation); and (10) adjustment of possible confounders on multivariate analysis. If a study did not clearly mention 1 of these key points, we considered that it had not been performed. Therefore, the reported characteristics might have been underestimated.
Two blinded investigators (C.N and T.L) independently performed data extraction using a standard data extraction form to determine the eligibility for inclusion. The following collected information was tabulated: (1) publication details, including first author’s surname, publication year, and study population; (2) characteristics of the studied population, including sample size, age, gender, diagnostic method for OSA and AF, and duration of follow-up; (3) catheter ablation strategy; (4) end point evaluation (methods of AF detection); and (5) blanking period and AF recurrence rates.
The dichotomous data of AF recurrence for OSA and non-OSA groups were used to calculate the risk ratio (RR) and 95% confidence intervals (CIs) for AF recurrence after catheter ablation. The Mantel-Haenszel method for the random effects model was used to pool the RRs for our study outcome. To evaluate the heterogeneity across studies, we used I 2 derived from the chi-square test, which describes the percentage of the variability in effect estimates resulting from heterogeneity, rather than sampling error (chance). An I 2 >50% indicates at least moderate statistical heterogeneity. The sensitivity analysis was also done in a random predefined manner to assess the effect of different types of AF, the number of enrolled patients, and the follow-up period and to check the consistency of the overall effect estimate. We also performed subgroup analysis on the different diagnostic methods for OSA. Publication bias was evaluated using the funnel plot. Statistical significance was defined as a 2-tailed p value of 0.05. All statistical analyses were performed using Review Manager (RevMan, version 5.0, Copenhagen, Denmark: The Nordic Cochrane Centre and the Cochrane Collaboration, 2008).
Results
A flow diagram of the data search and study selection is presented in Figure 1 . A total of 402 studies were found using our search criteria. We identified 202 duplicate studies, which were discarded. The remaining 200 abstracts were screened further, and we excluded 190 studies, because they were either unrelated, irrelevant, review articles or animal studies. Ten full-text studies were then retrieved for detailed evaluation. Of the 10 studies, 4 did not meet the inclusion criteria and were excluded. The remaining 6 studies were included in our meta-analysis, with a total of 3,995 patients: 958 with OSA and 3,037 without OSA. The follow-up periods varied from 7 to 32 months. The characteristics and quality assessments of each study are listed in Table 1 , and the demographics of the patients in each study are listed in Table 2 .
| Investigator | Patients (n) | Ablation Strategy | Diagnostic OSA Method | Mean Follow-Up | Methods of AF Detection | Blanking Period | Quality Score |
|---|---|---|---|---|---|---|---|
| Patel et al | 3,000 | PVI plus left atrial linear ablation | Polysomnogram | 32 mo | Event monitor and 48-h Holter monitor | NA | 7 |
| Matiello et al | 174 | PVI plus left atrial linear ablation | Polysomnogram | 12 mo | Holter monitor | First 3 mo | 10 |
| Chilukuri et al | 109 | PVI | BQ | 11 mo | ECG and telephone follow-up; event monitor for symptomatic patients | First 3 mo | 9 |
| Tang et al | 178 | PVI | BQ | 344 days | ECG and 24-h Holter monitor | First 3 mo | 9 |
| Chilukuri et al | 210 | PVI | BQ | 25 mo | ECG and telephone follow-up; event monitor for symptomatic patients | First 3 mo | 7 |
| Jongnarangsin et al | 324 | PVI plus CAFE ablation | Polysomnogram | 7 mo | ECG and 30-day autotrigger event recorder | First 8 wk | 8 |
| Investigator | Patients (n) | Mean Age (y) | Men (%) | Paroxysmal AF (%) | Structural Heart Disease (%) | HTN (%) | DM (%) | BMI (%) | Mean LAD (mm) | Mean LVEF (%) |
|---|---|---|---|---|---|---|---|---|---|---|
| Patel et al | 3,000 | 55.7 | 77.2% | 53.4% | 10.9% (CAD) ⁎ | 42.9% | 13.3% | 27.1% | 42.6 | 53.6% |
| Matiello et al | 174 | 52.5 | 77.6% | 56.3% | 17.9% | 42.0% | NA | 27.3% | 42.1 | 58.5% |
| Chilukuri et al | 109 | 60 | 78.0% | 68% | 11% (CAD) ⁎ | 39% | 8% | 28% | 45 | 58% |
| Tang et al | 178 | 57.2 | 68.5% | 100% | 11.8% | 49.4% | 12.4% | 25.6% | 37.2 | 63.6% |
| Chilukuri et al | 210 | 58 | 80% | 57% | NA | 50% | 9% | 29% | 4.6 | 58% |
| Jongnarangsin et al | 324 | 57 | 76% | 72% | 11% (CAD) ⁎ | 46% | NA | 30.0% | 44 | 56% |
Of the 6 studies, 4 demonstrated that the presence of OSA increased the risk of AF recurrence after catheter ablation, and 2 showed no significant difference using multivariate analysis. Our meta-analysis demonstrated that OSA patients have a 25% greater risk of AF recurrence after catheter ablation compared to controls (RR 1.25, 95% CI 1.08 to 1.45, p = 0.003; Figure 2 ). The heterogeneity test showed significant differences among the individual studies (chi-square = 9.77, p = 0.08, I 2 = 49%). A sensitivity analysis ( Table 3 ) and subgroup analysis ( Figure 2 ) were performed to find the origin of heterogeneity. After removing the study by Tang et al, which had enrolled only patients with paroxysmal AF, the analysis did not find significant influences on heterogeneity across studies or overall results. After excluding the study by Chilukuri et al, which had the least number of enrolled patients, no significant differences were found in the heterogeneity among the remaining 5 studies (chi-square = 5.83, p = 0.21; I 2 = 31%), and the overall outcome remained the same (RR 1.31, 95% CI 1.15 to 1.50 p <0.0001). Excluding the study with the shortest follow-up period also reduced the heterogeneity (chi-square = 7.18, p = 0.13; I 2 = 44%), without changing the overall effect (RR 1.25, 95% CI 1.08 to 1.45, p = 0.02). Therefore, the differences in study population size and follow-up period might have accounted for the heterogeneity in our study.
| Excluding Study | Remaining Studies (n) | RR (95% CI); p Value | p heterogeneity ; I 2 |
|---|---|---|---|
| Tang et al (all patients with paroxysmal AF) | 5 | 1.27 (1.08–1.38); 0.003 | 0.06; 56% |
| Chilukuri et al (smallest sample size) | 5 | 1.31 (1.15–1.50); <0.0001 | 0.21; 31% |
| Jongnarangsin et al (shortest follow-up period) | 5 | 1.21 (1.04–1.40); 0.01 | 0.13; 44% |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
