Kidney Week

Abstract: TH-PO0071

Association of Kidney Interstitial Infiltrate with and Without Tubulitis with Histological Acute Tubulointerstitial Nephritis

Session Information

• AKI: Pathogenesis and Disease Mechanisms
  November 06, 2025 | Location: Exhibit Hall, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Acute Kidney Injury

• 102 AKI: Clinical, Outcomes, and Trials

Authors

• Mukherjee, Josh, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Josh Mukherjee

• Sadarangani, Sagar S., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Sagar S. Sadarangani, MBBS

• Shelton, Kyra A., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Kyra A. Shelton

• Shaw, Melissa M., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Melissa M. Shaw

• Liang, Cathleen G, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Cathleen G Liang

• Koval, Emma L, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Emma L Koval

• Kent, Candice, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Candice Kent, MS

• Perazella, Mark A., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Mark A. Perazella, MD, MS, FASN

• Luciano, Randy L., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Randy L. Luciano, MD, PhD

• Turner, Jeffrey M., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Jeffrey M. Turner, MD

• Moeckel, Gilbert W., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Gilbert W. Moeckel, MD, PhD, FASN

• Kuperman, Michael Benjamin, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Michael Benjamin Kuperman, MD

• Rosenberg, Avi Z., The Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

Avi Z. Rosenberg, MD, PhD

• Kumar, Deepika, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Deepika Kumar, MD

• Wilson, Francis Perry, Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Francis Perry Wilson, MD, MS

• Parikh, Chirag R., The Johns Hopkins University School of Medicine, Baltimore, Maryland, United States

Chirag R. Parikh, MD, PhD, FASN

• Moledina, Dennis G., Yale School of Medicine Department of Internal Medicine, New Haven, Connecticut, United States

Dennis G. Moledina, MD, PhD, FASN

Background

Interstitial infiltrate is a characteristic feature of acute interstitial nephritis (AIN) but occurs in other kidney diseases, while “tubulitis,” the migration of inflammatory cells across the tubular basement membrane, may be more specific to AIN. Here, we test the association of interstitial immune infiltrate with and without tubulitis with histological AIN diagnosis.

Methods

We included two cohorts of participants from the Yale Kidney BioBank who underwent native kidney biopsies between 2020-23 (discovery; n=752) and 2015-18 (validation; n=121). In the discovery cohort, we tested the association of infiltrate with and without tubulitis with final reported AIN diagnosis. We used LASSO feature selection with 100 iterations to select optimal features associated with AIN and report AUCs in both cohorts.

Results

In the discovery cohort, 120 (16%) had infiltrate with tubulitis, 416 (55%) had infiltrate without tubulitis, and 216 (29%) had no infiltrate. Among all histological features tested, infiltrate with tubulitis and interstitial eosinophils were selected in >70% LASSO iterations. Infiltrate with tubulitis and eosinophils were associated with 333-fold and 7.6-fold higher odds of AIN, respectively (Table). A diagnostic model using these two features yielded an AUC of 0.92 (95% CI: 0.89-0.94) in discovery and 0.90 (0.87-0.93) in validation cohort.

Conclusion

Interstitial infiltrate was seen in many diagnoses, while tubulitis was specific to AIN, and combining tubulitis with interstitial eosinophils led to high diagnostic accuracy for AIN. These findings support the use of tubulitis as a central criterion in standardizing histological diagnosis of AIN and reducing diagnostic variability.

Funding

• NIDDK Support

Digital Object Identifier (DOI)

• doi: 10.1681/ASN.2025555bfcpt