Meta-analysis on clinical utility of oesophageal temperature monitoring in AF ablation

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background Atrial fibrillation (AF) ablation can lead to oesophageal thermal injury (ETI) and atrio-oesophageal fistulas can be fatal. Although temperature probes are widely utilized in AF ablations worldwide, in the United Kingdom, AF ablation procedures are performed under local anaesthesia and sedation and not all centres have adopted routine use of temperature probes. This may be due to comparative studies to detect the role of oesophageal temperature monitoring (ETM) in AF ablation showing mixed results. Clinical utility of ETM-guided AF ablation remains debatable, and studies have reported pulmonary vein recurrences in the areas of limited radiofrequency application due to cautious ETM-guided ablation. Purpose This meta-analysis aims to determine the utility of ETM in preventing ETI in the context of radiofrequency catheter ablation of AF. Methods A systematic search of PubMed, Embase databases and Cochrane registry was performed comparing ETI between ETM and non-ETM-guided strategies in AF ablation. Primary outcome was endoscopically detected ETI. Secondary outcomes were AF recurrence, procedure time and ablation time. Statistical analyses including subgroup and covariate analyses were performed using random effect model. Results Baseline: 12 studies (one randomized study, 7 from Germany) satisfied the inclusion criteria; only two studies had patients undergoing AF ablation under GA; 9 studies utilized multi-sensor temperature probes; temperature cut-off ranged from 38.5 ℃ to 42 ℃; posterior wall power setting ranged from 15W to 35W (Figure 1). Primary outcome: ETI were similar in both ETM (n=864) and non- ETM groups (n=639) (RR 1.04, 95% CI 0.34-3.23, I2 =72%) (Figure 2A). Temperature cut-offs, posterior wall power setting, additional linear ablations or body mass index (BMI) had no significant effect on ETI in bivariate or multivariate analysis. Secondary outcome: AF recurrence was statistically similar in both groups (IRR 0.92, 95% CI 0.73-1.17, I2 =0%) but showed a lower trend in non-ETM group (Figure 2B). Ablation time (SMD -0.08, 95% CI -0.37- 0.21, I2 =65%) was numerically lower in the ETM group but was not statistically significant (Figure 2C). Procedure time (SMD 0.96, 95% CI 0.69- 2.61, I2 =97%) was statistically similar across both groups but showed a higher trend in the ETM group (Figure 2D). Major complications of both ETM and non-ETM groups, including atrio-oesophageal fistula, were extremely rare. Conclusion ETM did not make statistically significant difference in the resultant ETI during AF ablation. There was a trend towards longer procedures with marginally lower RF application time. While use of temperature probes continues to be limited in the UK centres, the clinical utility of oesophageal temperature remains questionable and need to be addressed in randomized studies. However our analyses cannot be generalized to high power ablation settings (e.g. 50W or 90W) or RF balloons.