Construction of a regression model and post-procedure care strategy for predicting atrial fibrillation recurrence risk after radiofrequency ablation using combined P-wave electrocardiographic markers and serum

Background To construct a risk prediction nomogram combining P-wave electrocardiographic indicators with serum brain natriuretic peptide (BNP) levels for predicting recurrence risk after radiofrequency ablation (RFA) for atrial fibrillation (AF), and to propose follow-up care strategies, thereby providing a theoretical basis for clinical prevention of early AF recurrence after RFA. Methods A retrospective study was conducted on 200 AF patients who underwent radiofrequency ablation at our hospital between March 2023 and December 2024. All patients were confirmed to be in sinus rhythm at the time of preoperative electrocardiographic recording. Based on recurrence status at 3-month follow-up, patients were divided into recurrence ( n = 62) and non-recurrence ( n = 138) groups. Recurrence was assessed through a standardized rhythm monitoring protocol including scheduled electrocardiograms, 24 h Holter monitoring, and symptom-driven event recordings. The last enrolled patient completed follow-up on March 31, 2025. Results Comparisons of AF type, left atrial diameter (LAD), P-wave duration (PWD), maximum P-wave duration (Pmax), P-wave dispersion (Pd), and BNP levels between groups showed statistically significant differences ( P < 0.05). Logistic regression analysis confirmed that AF type, LAD, PWD, Pmax, Pd, and BNP were independent risk factors for post-AF radiofrequency ablation recurrence (OR > 1, P < 0.05). A risk prediction nomogram model was constructed. In the derivation cohort, the area under the receiver operating characteristic curve (AUC) was 0.959 (95% CI: 0.925–0.993). Apparent calibration was acceptable (Hosmer-Lemeshow test P = 0.726). Risk stratification analysis identified a total nomogram score of 85 points as an optimal threshold to distinguish low-risk from medium-to-high-risk patients. Formal internal validation was not performed. Conclusion The derivation-cohort risk prediction nomogram based on P-wave ECG indicators combined with serum BNP demonstrates potential value for estimating early recurrence after AF radiofrequency ablation. Clinicians should interpret the model cautiously and develop personalized follow-up care strategies based on individual risk profiles to reduce recurrence risk and improve patient quality of life.