ASN's Mission

ASN leads the fight to prevent, treat, and cure kidney diseases throughout the world by educating health professionals and scientists, advancing research and innovation, communicating new knowledge, and advocating for the highest quality care for patients.

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: PO1156

Identifying Hypernatremia Subgroups with Differing Survival by Machine Learning Among Hospitalized Patients

Session Information

Category: Fluid, Electrolyte, and Acid-Base Disorders

  • 902 Fluid, Electrolyte, and Acid-Base Disorders: Clinical

Authors

  • Chong, Grace Yun, Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
  • Thongprayoon, Charat, Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
  • Mao, Michael A., Mayo Clinic Division of Nephrology and Hypertension, Jacksonville, Florida, United States
  • Kattah, Andrea G., Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
  • Keddis, Mira T., Mayo Clinic Division of Nephrology and Hypertension, Phoenix, Arizona, United States
  • Erickson, Stephen B., Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
  • Dillon, John J., Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
  • Garovic, Vesna D., Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
  • Cheungpasitporn, Wisit, Mayo Clinic Division of Nephrology and Hypertension, Rochester, Minnesota, United States
Background

The objective of this study was to characterize patients with hypernatremia on hospital admission into clusters using an unsupervised machine learning approach and to evaluate the mortality risk among these distinct clusters.

Methods

We performed consensus cluster analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 6,297 hospitalized adult patients with hypernatremia present at admission. We calculated the standardized difference of each variable to identify each cluster’s key features. We assessed the association with each hypernatremia cluster with in-hospital and one-year mortality.

Results

There were three distinct clusters of hypernatremia: 1,570 patients (25%) in cluster 1; 2,648 (42%) in cluster 2; and 2,079 (33%) in cluster 3. Figure 1 is a plot of standardized mean differences to visualize key features for each cluster. Compared to cluster 2, the odds ratios for in-hospital mortality were 6.99 (95% CI 4.03-12.13) for cluster 1 and 5.73 (95% CI 3.31-9.90) for cluster 3, whereas hazard ratios for one-year mortality were 3.38 (95% CI 2.69-4.25) for cluster 1 and 4.71 (95% CI 3.82-5.80) for cluster 3.

Conclusion

The characteristics and outcomes of hospitalized patients admitted with hypernatremia were heterogenous. Our cluster analysis identified three clinically distinct phenotypes with differing mortality risks. Identification of heterogeneity in hypernatremic patients using this approach may provide guidance for the management of hospitalized patients with hypernatremia at the time of hospital admission.

The standardized differences across three clusters for each of the baseline parameters. The x axis is the standardized differences value, and the y axis shows baseline parameters. The dashed vertical lines represent the standardized differences cutoffs of <-0.3 or > 0.3.