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


The Latest on Twitter

Kidney Week

Abstract: FR-PO385

Evaluation of Glomerular Filtration Rate Change as an Indicator of Development of Incident CKD Using Support Vector Machine: Community-Based Prospective Cohort Study

Session Information

Category: Chronic Kidney Disease (Non-Dialysis)

  • 301 CKD: Risk Factors for Incidence and Progression


  • Kanda, Eiichiro, Tokyo Kyosai Hospital, Meguro, TOKYO, Japan
  • Epureanu, Bogdan I., University of Michigan, Ann Arbor, Michigan, United States
  • Suwa, Kaname, Saitama Health Promotion, Saitama, Japan
  • Nakajima, Kei, Kanagawa University of Human Services, Yokosuka, Japan

Chronic kidney disease (CKD) is a risk factor for cardiovascular disease and death. To decrease the number of CKD patients, it is important to identify people at high risk of developing CKD among healthy people from the public-health viewpoint. We investigated the role of estimated glomerular filtration rate (eGFR) change in the development of CKD in a community-based prospective cohort study using the support vector machine (SVM).


A total of 3295 healthy people (male, 72.7%) were enrolled in this prospective cohort study for 6 years (Saitama Cardiometabolic Disease and Organ Impairment Study) in Japan. The outcome event was incident CKD 6 years later. Subjects were categorized on the basis of eGFR change over 3 years by 10%.


The mean±SD age was 38.8±10.1 years; eGFR, 81.8±17.3 ml/min/1.73m2; and eGFR change over 3 years, -15.0±18.5%. Multivariate logistic regression models showed the relationship between eGFR change and incident CKD (reference 0 %): -10%, adjusted odds ratio 3.89 (95%CI, 2.0, 7.56); -20%, 40.3 (20.5, 79.0); -30%, 124.8 (59.9, 260.1); -40%, 297.9 (119.8, 741.1). Interaction between eGFR change and eGFR was also observed (p=0.0001). A receiver operating characteristic curve showed a cutoff value of -12.4% for CKD prediction. The use of SVM enabled the identification of high-risk patients and showed that cutoff values differ depending on eGFR: eGFR 90, eGFR change -40%; 80, -30%; 70, -20% (Figure 1).


eGFR change tends to be associated with the risk of CKD. The cutoff values for the prediction of CKD may differ depending on eGFR in the general population. The composite index of eGFR and eGFR change may have high potential use for detecting high-risk people for CKD.