Aims: The clinical significance of common histological parameters in acute interstitial nephritis (AIN) is uncertain. We aimed to evaluate the utility of histology in predicting clinical outcomes in patients with AIN.
Methods and results: Adult renal biopsies yielding a diagnosis of AIN between 2000 and 2015 were re-examined. Patients were divided into groups based on: (i) the percentage of non-fibrotic cortex containing inflammation (NFI score) (NFI-1 = 0-24%; NFI-2 = 25-74%; NFI-3 = 75-100%) and (ii) the percentage of cortex containing tubular atrophy (TA score) (TA1 = 0-9%; TA2 = 10-24%; TA3 = 25-100%). The primary outcome was a composite of ≥50% reduction in serum creatinine (sCr) or an estimated glomerular filtration rate (eGFR) > 60 ml/min/1.73 m2 1 year post-biopsy. From a total of 2817 native renal biopsies, there were 120 patients with AIN and adequate data for analysis. Of these, 66 (56%) achieved the primary outcome. On univariable logistic regression, NFI-3 was associated with a 16 times increased likelihood of achieving the primary outcome compared to NFI-1 [odds ratio (OR) = 16, 95% confidence interval (CI) = 5.2-50)]. In contrast, TA3 was associated with a 90% reduced likelihood of achieving the primary outcome compared to TA1 (OR = 0.10, 95% CI = 0.0-0.3). Maximal clinical utility was achieved by combining TA and NFI into a single prognostic 'TANFI' score, which had an independent predictive effect on the primary outcome in a multivariable regression model consisting of age, sex, baseline sCr and identified drug cause.
Conclusions: In patients with biopsy-proven AIN, a lower percentage of cortical tubular atrophy and, paradoxically, a higher percentage of inflammation in non-fibrosed cortex were associated with an increased likelihood of a positive clinical outcome.
Keywords: acute interstitial nephritis; acute kidney injury; corticosteroids; renal histology.
© 2019 John Wiley & Sons Ltd.