A multi-model study of serum uric acid/albumin ratio in predicting acute kidney injury after PCI in STEMI patients
Objective To explore the potential of serum uric acid/albumin ratio(sUAR)as a predictive model for acute kidney injury(AKI)after PCI in patients with acute ST-segment elevation myocardial infarction(STEMI).Methods A total of 166 STEMI patients admitted to Duchang Hospital from July 2021 to July 2023 were retrospectively selected and divided into two groups:the occurrence group(n =34)and the non-occurrence group(n =132)based on whether AKI occurred after PCI.Base-line data,biochemical indexes,and sUAR were compared between the two groups.Univariate and multivariate logistic regression were utilized to analyze the risk factors for AKI following PCI,and a nomogram prediction model was developed.The ROC curve was developed to analyze the predictive efficiency of the model.Results There were significant differences in age,hypertension,Killip classification,NLR,sUAR,LVEF,contrast agent dose,PCI operation time,and multi-vessel lesions between the two groups(P＜0.05).Older age(OR=1.066),Killip grade≥2(OR=7.174),elevated NLR(OR=4.440),increased sUAR(OR=2.071),high contrast agent dose(OR=1.104),and prolonged PCI operation duration(OR=1.044)were identified as the independent risk factors for AKI following PCI(P＜0.05).The AUC values for the NLR,sUAR,"NLR+sUAR"and no-mogram prediction models were 0.807(95%CI:0.717～0.897),0.810(95%CI:0.729～0.892),0.877(95%CI:0.808～0.946),0.940(95%CI:0.901～0.979),respectively.Bootstrap(B =1 000)internal validation indicated that the bias-corrected prediction curve was closely aligned with the ideal line,and the nomogram risk prediction model had good predictive a-bility.The decision-making curve analysis revealed that the model's threshold probability ranged from 0.01 to 0.90 with a net re-turn more than 0.Conclusion AKI after PCI in STEMI patients are closely related to such indicators as age,Killip classifica-tion,NLR,sUAR,contrast agent dose,and PCI operation duration.The nomogram prediction model demonstrates higher predic-tive efficiency for AKI after PCI compared to the single model and it holds better clinical application value.
万方数据
维普
