Accuracy of implantable loop recorders for detecting atrial tachyarrhythmias after atrial fibrillation catheter ablation

Background Implantable loop recorders (ILRs) can provide an enhanced possibility to detect atrial fibrillation (AF), but the accuracy, especially the positive predictive value (PPV), is controversial. This study aimed to evaluate the accuracy of ILRs for detecting AF through a comparison with Holter. Method and results Thirteen patients who underwent AF ablation were enrolled. ILRs were implanted in all patients, who were scheduled to have Holter monitorings after the procedure. The incidence of AF was compared between the two modalities and analyzed for any correlations. A total of 51 Holters (67,985.5 min) and concomitant ILRs were available for the comparison. The judgment of the presence of AF did not perfectly correlate between the ILR and Holter (Kappa = 0.866, P < 0.001). In the ILR data, the sensitivity of detecting AF on the Holter was 81.6% (95% confidence interval [CI] 0.812–0.820; P < 0.001). The specificity was 99.9% (95% CI 0.998–0.999; P < 0.001). When the ILR detected AF, the PPV was 99.5% (95% CI 0.994–0.995), but the ILR did not detect AF, and the negative predictive value was 94.2% (95% CI 0.941–0.944). A separate analysis of AF/atrial tachycardia (AT) showed that the AT detection rate of the ILR was 2.3%. Conclusion The ILR had a low false positive value and high PPV for AF events. However, it was limited in identifying AT.