Nephrol Dial Transplant. 2021 Dec 4:gfab346. doi: 10.1093/ndt/gfab346. Online ahead of print.
ABSTRACT
BACKGROUND: Patients with acute interstitial nephritis (AIN) can present without typical clinical features, leading to a delay in diagnosis and treatment. We therefore developed and validated a diagnostic model to identify patients at risk of AIN using variables from the electronic health record.
METHODS: In patients who underwent a kidney biopsy at Yale between 2013-2018, we tested the association of over 150 variables with AIN including demographics, comorbidities, vital signs, and laboratory tests (training set, 70%). We used least absolute shrinkage and selection operator (LASSO) methodology to select pre-biopsy features associated with AIN. We performed AUC analysis with internal (held-out test set, 30%) and external validation (Biopsy Biobank Cohort of Indiana). We tested change in model performance after addition of urine biomarkers in Yale AIN study.
RESULTS: We included 393 patients (AIN = 22%) in training set, 158(AIN = 27%) in test set, 1118(AIN = 11%) in validation set, and 265(AIN = 11%) in Yale AIN study. Variables in selected model included serum creatinine (adjusted odds ratio, 2.31(1.42-3.76)), blood urea nitrogen to creatinine ratio (0.40(0.20-0.78)), and urine dipstick specific gravity (0.95(0.91-0.99)) and protein (0.39(0.23-0.68)). This model showed an AUC of 0.73(0.64-0.81) in the test set, which was similar to the AUC in the external validation cohort [0.74(0.69-0.79)]. The AUC improved to 0.84(0.76-0.91) upon addition of urine interleukin-9 and TNF-α.
CONCLUSIONS: We developed and validated a statistical model that showed a modest AUC for AIN diagnosis, which improved upon addition of urine biomarkers. Future studies could evaluate this model and biomarkers to identify unrecognized cases of AIN.
PMID:34865148 | DOI:10.1093/ndt/gfab346
