Prognostic value of inflammatory indices for atrial fibrillation recurrence after cryoablation: a cohort study

Inflammatory markers have emerged as potential prognostic markers of atrial fibrillation (AF) recurrence following cryoablation. However, comparative analyses of multiple systemic indices are limited. This study aimed to evaluate four inflammation-derived biomarkers-the neutrophil-to-lymphocyte ratio (NLR), systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), and panimmune-inflammation value (PIV)-for their prognostic value in post-cryoablation AF recurrence.

We conducted a retrospective cohort of 757 patients undergoing first-time cryoablation at Nanjing First Hospital (January 2017-December 2023). We investigated the associations between the four systemic inflammatory markers and AF recurrence. Baseline characteristics were collected from medical records, and inflammatory marker levels were calculated from routine blood tests. Multivariable Cox proportional hazards models were used to estimate adjusted hazard ratios; restricted cubic splines (RCS) assessed potential nonlinearity; and time-dependent receiver operating characteristic (ROC) analyses quantified predictive performance at 12 and 24 months.

Compared with tertile 1, tertile 3 showed higher multivariable-adjusted hazards of recurrence (HR: NLR = 4.112, SII = 4.010, SIRI = 5.137, PIV = 5.298; all P < 0.001). The RCS revealed inflection points (logNLR = 1.0, logSII = 6.0), beyond which the risk slopes intensified. Time-dependent ROC analyses showed the highest AUCs for logPIV (AUC = 0.764 at 12 months; 0.741 at 24 months) compared with the other indices (AUC range = 0.715-0.742), with an optimal cutoff yielding 79.2% sensitivity and 68.3% specificity.

Systemic inflammation indices-particularly the pan-immune-inflammation value (PIV)-show prognostic association with AF recurrence after cryoablation and may inform preprocedural risk stratification and postablation surveillance. Given the observational design, these findings are associative and do not evaluate whether biomarker-guided selection or management improves outcomes. External calibration and validation-including in radiofrequency (RF) and pulsed-field ablation (PFA) cohorts-are needed to establish generalizability and clinical utility.