Aims Atrial fibrosis has been found to be associated with recurrent atrial fibrillation (AF) following catheter ablation. Autoantibodies against M2-muscarinic receptors (anti-M2-R) may play a role in the development of AF by inducing left atrial (LA) fibrosis. In this study, we aim to compare anti-M2-R levels between paroxysmal lone AF patients and healthy control subjects and to investigate the relationship between pre-ablation anti-M2-R level, LA fibrosis quantified by delayed enhancement magnetic resonance imaging (DE-MRI), and AF recurrence following cryoablation. Methods and results Thirty-one patients with paroxysmal lone AF (53.4 ± 8.0 years, 61% male), who underwent cryoballoon-based ablation, along with 31 healthy control subjects were included. Enzyme-linked immunosorbent assay tests to measure serum anti-M2-R levels were performed in both groups and DE-MRI was done to quantify LA fibrosis prior to the ablation in the patients. Anti-M2-R levels were higher in the study population when compared with control subjects [212.4 (103.2-655.5) vs. 73.0 (39.5-299.1) ng/mL, P < 0.001]. Anti-M2-R level predicted moderate-extensive LA fibrosis independent of other measures [odds ratio: 1.26 (95% confidence interval (CI): 1.04-1.53), P = 0.017]. At a mean follow-up of 35.2 ± 3.5 months, nine patients (29.0%) had AF recurrence. In the Cox regression model including pre-ablation anti-M2-R level, LA diameter, LA volume index, and moderate-extensive LA fibrosis, only moderate-extensive LA fibrosis predicted late AF recurrence independent of other measures [hazard ratio: 29.41 (95% CI: 3.52-250.00), P = 0.002]. Conclusion Serum anti-M2-R levels may be associated with the severity of LA fibrosis and may be implicated in the pathophysiology of AF recurrence following cryoablation. Detection of anti-M2-R levels may help select appropriate patients for the procedure.