Relationship and predictive value of serum L1CAM,FABP4,and NOX2 levels for postoperative atrial fibrillation in patients undergoing cardiac valve replacement

GAO Zhi-bing; WANG Deng-feng; HAN Dong; ZHANG Chao

doi:10.13820/j.cnki.gdyx.20251206

Abstract：Objective To investigate the relationship and predictive value of serum L1 cell adhesion molecule(L1 CAM),fatty acid-binding protein 4(FABP4),and NADPH oxidase 2(NOX2)levels in patients with valvular heart disease(VHD)undergoing cardiac valve replacement(CVR)who develop postoperative atrial fibrillation(POAF).Methods A total of 127 patients with VHD who underwent CVR between January 2023 and March 2025 were enrolled as the observation group.Based on electrocardiographic monitoring,they were divided into the POAF group(n=57)and non-POAF group(n=70).An additional 127 healthy volunteers were included as the normal control(NC)group.Ser-um levels of L1CAM,FABP4,and NOX2 were measured using ELISA.Multivariate logistic regression was applied to identify independent risk factors for POAF.Receiver operating characteristic(ROC)curves were used to evaluate the pre-dictive value of these biomarkers.Results Compared with the NC group,serum L1CAM levels were significantly de-creased,while FABP4 and NOX2 levels were significantly increased in VHD patients(P＜0.05).Compared with the non-POAF group,the POAF group showed larger left atrial diameter,higher proportion of NYHA class Ⅳ,increased FABP4 and NOX2 levels,and decreased L1CAM levels(all P＜0.05).High L1CAM expression was an independent pro-tective factor,whereas high FABP4 and NOX2 expression were independent risk factors for POAF(P＜0.05).The pre-dictive AUCs of L1CAM,FABP4,and NOX2 for POAF were 0.810,0.820,and 0.807,respectively,while their com-bined detection yielded a higher AUC of 0.950,superior to individual predictors(all P＜0.05).Conclusion Serum L1CAM is downregulated,whereas FABP4 and NOX2 are upregulated in VHD patients undergoing CVR.These three bio-markers are important determinants of POAF,and their combined detection demonstrates superior predictive value for post-operative risk stratification.