Abstract BACKGROUND AND AIMS Acute interstitial nephritis (AIN) represents an important cause of acute kidney injury. The most common etiology of AIN is drug-induced disease. Definitive diagnosis of AIN requires renal biopsy. Despite this, there is a lack of consensus regarding precise histological diagnostic criteria, and the clinical significance of common histological findings is uncertain. The aim of this study was to explore the histological characteristics of drug-induced AIN and to evaluate their impact on the renal outcome. METHOD We conducted a retrospective study that included patients with biopsy proven drug-induced AIN, followed in the department of nephrology of Charles Nicolle hospital in Tunis during the period between 1990 and 2019. Histological diagnosis of AIN was based on the presence of an interstitial inflammatory cell infiltrate with or without fibrosis. RESULTS During the period of the study, we collected 69 cases of biopsy proven AIN. The diagnosis of drug-induced AIN was performed in 23 patients (33%), with a medium age of 35 years. In eight patients (35%), nonsteroidal anti-inflammatory drugs were the culprit, followed by antibiotics in six patients (26%) and other drugs in three cases (13%). In six patients (26%), no single drug could be identified as the culprit, and these were classified as multidrug causes. The presence of acute kidney disease was the main indication for the kidney biopsy. All cases were characterized by a prominent tubulointerstitial inflammatory cell infiltrate. Interstitial infiltration was mild in three cases (13%), moderate in five cases (17%) and severe in 15 cases (65%). The interstitial infiltrate consisted predominantly of lymphocytes. The presence of eosinophils has been noted in 10 cases (43%). Interstitial granulomas were noted in two patients (8%). Interstitial edema and tubulitis were observed in 13 (56%) and 6 cases (26%), respectively. In seven patients (30%) interstitial fibrosis was present, and four patients (17%) had mild-to-moderate tubular atrophy, whereas tubular necrosis was present in seven cases (30%). After a mean follow up of 24 months, eight patients (35%) progressed to chronic kidney disease. Histological factors associated with poor renal outcome, were interstitial fibrosis (P ˂ 0.001) and tubular atrophy (P = 0.01). Patients with interstitial edema (P ˂ 0.001) or tubulitis (P = 0.005) had a better renal prognosis. CONCLUSION Renal biopsy is required to establish a definitive diagnosis of drug-induced AIN. Nevertheless, it is not carried out in a systematic way. The composition of cells in the interstitial infiltrate may at times be helpful in determining the etiology of AIN. In particular, the presence of a considerable proportion of eosinophils favors a diagnosis of drug-induced AIN. Moreover, during drug-induced AIN, histological data have a great prognostic value. In our study, the main histological prognostic factors were interstitial fibrosis, tubular atrophy, interstitial edema and tubulits. Further studies are needed to validate these results and establish a score based on histological data in order to predict renal outcome during drug-induced AIN.
Read full abstract