ASN's Mission

To create a world without kidney diseases, the ASN Alliance for Kidney Health elevates care by educating and informing, driving breakthroughs and innovation, and advocating for policies that create transformative changes in kidney medicine throughout the world.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005

email@asn-online.org

202-640-4660

The Latest on Twitter

Kidney Week

Abstract: FR-PO077

The Correlation Between Neutrophil-to-Lymphocyte Ratio and Contrast-Induced AKI and Establishment of New Predictive Models by Machine Learning

Session Information

Category: Acute Kidney Injury

  • 101 AKI: Epidemiology‚ Risk Factors‚ and Prevention

Authors

  • Lu, Yi, Ningbo Huamei Hospital University of Chinese Academy of Sciences, Ningbo, Zhejiang, China
  • Zhou, Fangfang, Ningbo Huamei Hospital University of Chinese Academy of Sciences, Ningbo, Zhejiang, China
  • Xu, Youjun, Ningbo Huamei Hospital University of Chinese Academy of Sciences, Ningbo, Zhejiang, China
  • Zhang, Shuzhen, Ningbo Huamei Hospital University of Chinese Academy of Sciences, Ningbo, Zhejiang, China
  • Luo, Qun, Ningbo Huamei Hospital University of Chinese Academy of Sciences, Ningbo, Zhejiang, China
Background

This study intends to explore the correlation between NLR and CI-AKI, and to establish new predictive models of CI-AKI by machine learning.

Methods

The data of patients who underwent elective vascular intervention, coronary angiography and percutaneous coronary intervention in our hospital from January 2016 to December 2020 were retrospectively collected. The patients were divided into AKI group and non-AKI group . The analysis of linear trends was used to assess the correlation between the NLR levels and risk of AKI after the sample was divided into tertiles based on the distribution of controls. Logistic regression was used to analyze the correlation between NLR and CI-AKI, and machine learning methods were used to establish logistic regression (LR), random forest (RF), gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost) and naïve bayes (NB) models. The diagnostic value of the machine learning model was evaluated by receiver operating curve (ROC), and the proportion of feature variables was calculated.

Results

(1) 2230 patients were included in this study, and the incidence of CI-AKI was 5.38%. Compared with patients in non-AKI group, patients in AKI group had higher levels of NLR [3.38(2.60,5.35)vs 2.79(1.98,4.18),P<0.001],and further multivariate logistic regression analysis showed that NLR was an independent risk factor for CI-AKI (OR=1.054, P= 0.048). (2) After dividing patient into tertiles based on NLR, those with higher NLR had higher risk of postoperative AKI than those with lower NLR (2.69% vs 5.95% vs 7.51%, trend P=0.046). (3) Gradient Boosting Decision Tree(GBDT)model has the best predictive performance of CI-AKI (AUC=0.738), followed by RF, NB, XGBoost and LR models respectively (The AUC values are 0.727, 0.725, 0.719, 0.711). Four indicators included in the GBDT model, which were NLR, serum creatinine, fasting plasma glucose, and the use of β-blocker.

Conclusion

There is a significant correlation between NLR and CI-AKI, and the GBDT, RF, NB, XGBoost and LR models established after incorporating this indicator have good effects in predicting and diagnosing the occurrence of CI-AKI.

Funding

  • Government Support – Non-U.S.