Background: Percutaneous kidney biopsy remains the gold standard for evaluation of AKI, given that clinicians’ pre-biopsy clinical impression of AKI, based on history, exam, and non-invasive testing may not lead to a unifying diagnosis. In this study, we evaluated the concordance of pre-biopsy clinical diagnosis with post-biopsy final diagnosis among patients with clinical AKI. Methods: We leveraged data collected prospectively through the Novel Approaches in the Investigation of Kidney Disease (NAIKiD) Study between 2020 and 2023, in which adult patients admitted to the Johns Hopkins Hospital and scheduled for clinical kidney biopsies consented to provide biosamples paired with data collection. Up to 3 pre-biopsy clinical diagnoses were recorded for each patient, along with up to 3 post-biopsy diagnoses, adjudicated by a study nephrologist post-biopsy. We investigated the concordance of pre- and post-biopsy diagnoses among patients with suspected acute interstitial nephritis (AIN) or acute tubular injury (ATI). Results: Among 164 total participants, 29 patients had a suspected clinical diagnosis of AIN and 47 had a suspected clinical diagnosis of ATI pre-biopsy. Among the participants with suspected AIN, only 7 (24%) had AIN confirmed on biopsy. Of the 22 biopsies without AIN present, alternative diagnoses noted on histology included focal segmental glomerulosclerosis (FSGS), ATI, and various glomerular diseases. Out of 47 participants with suspected ATI, ATI was confirmed on biopsy in 27 (57%) participants. In the 20 biopsies without ATI present, alternative histological findings also included glomerular diseases, diabetic nephropathy, and FSGS predominantly. Conclusions: Patients with suspected ATI or AIN who undergo percutaneous kidney biopsy are often found to have alternative, significant findings present on histology. Among patients without relative or absolute contraindications, kidney biopsy remains an essential part of clinical evaluation, while future research should focus on the development of non-invasive approaches for kidney disease diagnosis.
McGredy et al. (Wed,) studied this question.