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 X

Kidney Week

Abstract: TH-PO052

External and Temporal Validation of Acute Tubulointerstitial Nephritis (AIN) Diagnostic Index: An Electronic Health Record-Based Multivariable Diagnostic Model for Assessing Probability of Acute Interstitial Nephritis on Kidney Biopsy

Session Information

Category: Acute Kidney Injury

  • 101 AKI: Epidemiology, Risk Factors, and Prevention

Authors

  • Shelton, Kyra A., Section of Nephrology, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
  • Menez, Steven, Division of Nephrology, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, United States
  • Shaw, Melissa M., Clinical and Translational Research Accelerator, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
  • Kent, Candice, Section of Nephrology, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
  • Bitzel, Jack, Division of Nephrology, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, United States
  • Thiessen Philbrook, Heather, Division of Nephrology, Internal Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, United States
  • Wilson, Francis Perry, Section of Nephrology, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
  • Hu, David, Division of Nephrology, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, United States
  • Wen, Yumeng, Division of Nephrology, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, United States
  • Parikh, Chirag R., Division of Nephrology, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, United States
  • Moledina, Dennis G., Section of Nephrology, Department of Internal Medicine, Yale School of Medicine, New Haven, Connecticut, United States
Background

Kidney biopsy is essential in cases of suspected acute tubulointerstitial nephritis (AIN), as there are limited diagnostic tests and clinical features associated with the disorder. Previously, we developed a multivariable model for biopsy-confirmed AIN containing serum creatinine, blood urea nitrogen to creatinine ratio, urine dipstick specific gravity and protein with an AUC of 0.73, which outperformed the clinicians’ prebiopsy AUC of 0.60. Here we perform validation of this diagnostic index.

Methods

In two geographically and temporally distinct cohorts of patients who received a kidney biopsy at either Johns Hopkins Hospital (JHH) or Yale between 2019-2023, we tested discrimination (area under receiver operating characteristics curve, AUC) and calibration of the diagnostic index. We used model weights from the previous model and followed TRIPOD guideline for validation of multivariable models.

Results

We included 1,973 participants, with 1,444 from JHH and 529 from Yale. The prevalence of AIN at Yale (21%) was similar to what we previously observed (23%) whereas that of JHH was lower (5%). We noted similar AUC at JHH (panel A) and Yale (panel B) as the development set. However, the JHH cohort showed poor calibration (panel C) that improved after application of an intercept correction factor, accounting for the lower prevalence of AIN in this cohort (panel D).

Conclusion

The diagnostic model retained discrimination in these two distinct cohorts but required recalibration based on the prevalence of biopsy proven AIN. This indicates that the model may be a useful tool in identifying specific clinical features associated with AIN and assisting providers in discerning whether an AIN diagnosis is warranted based on laboratory and dipstick values. Calculator: https://ainriskprediction.shinyapps.io/ain_calc/.

Funding

  • NIDDK Support