Abstract
Aim: The aim of this study was to define the nature of the inflammatory infiltrate that occurs with omeprazole-induced acute interstitial nephritis (AIN) and further to compare and contrast the histology of omeprazole-induced AIN with the interstitial renal lesions seen in PR3 and MPO positive small vessel vasculitis.
Background: Omeprazole is now the most common cause of drug-induced acute interstitial nephritis. However, how omeprazole-induced acute interstitial nephritis is mediated is unknown. We observed in all patients the histological pattern is unlike the classical picture of an acute allergic drug-induced interstitial nephritis as seen with methicillin, but rather it is akin to that seen with PR3/MPO small vessel vasculitis or acute cellular rejection in renal transplant recipients. The nature of the cellular infiltrate in omeprazole-induced acute interstitial nephritis has not been characterized. Given the similarity in appearance to the renal involvement in small vessel vascultits, we postulated that the inflammatory process is predominantly a cell-mediated process involving TH1/TH17 cells.
Methods: Data from 25 biopsy-proven cases of omeprazole-induced acute interstitial nephritis (AIN) and 27 patients with renal involvement with PR3/MPO ANCA positive small vessel vasculitis, who have presented to the renal service (Dunedin Hospital, NZ) over the last 10 years, were reviewed. The lesions were graded according to the updated Banff classification of 1997 to compare the severity of the inflammatory infiltrate. The underlying immune reaction was characterized using immunohistochemical and fluoroscopic chromogenic in situ hybridization (IHC and F-CISH) to detect T-cells (and particular markers associated with Th1, Th2, Th17 and Treg effector T cells) and B-cells. The following antibodies were used CD20 for B-cells, CD4, CD8, IL17A, IL17F and Foxp3 for T-cells.
Results: All patients presented with evidence of acute kidney injury (AKI). Urinalysis demonstrated a sterile pyuria with varying amounts of proteinuria. There were no urinary eosinophils. There were no systemic symptoms to suggest a vasculitis and ANCA (PR3 & MPO titres) were negative in all cases. Histologically, all cases had 4 evidence of inflammatory cells crossing the tubular basement membrane and showing an active tubulitis.
The occurrence of any eosinophils in our patients was 52%. However, in contrast to previous studies only one patient has an eosinophilic infiltrate, the other cases showed only small numbers (3 – 7 eosinophils per high powerfield in some but not all views) of scattered eosinophils in the infiltrate. Similar findings were evident in the vasculitis group. The lack of a substantial eosinophilic infiltrate argues against a pre-dominant Th-2 mediated mechanism for omeprazole-induced AIN.
Our cases showed numerous CD4 positive cells, and fewer numbers of CD20 and CD8 positive cells. CD4 positive cells were present in clusters in 77% of the PPI-IN patients and 67% of the vasculitis patients. CD20 positive cells were present in clusters in 58% of all omeprazole-induced AIN patients and 37% of vasculitis patients. However CD8 positive cells were present in clusters only in 17% of omeprazole-induced AIN patients and 13% of vasculitis patients. Comparing the IL17A and CD4 double staining with the IL17F and CD4 double staining for the omeprazole-induced AIN group the grading of the staining looks very similar (for IL17A and CD4 50% of patients showing clusters, 4% intermediate, 32% of scattered and 14 % of patients without any IL17A and CD4 cells versus the staining for the IL17F and CD4 staining with 48% in clusters, 17% intermediate, 31% scattered and 4% with no staining for IL17F and CD4). Foxp3 regulatory cells were found within the interstitium and tubular epithelium of both vasculitis and omeprazole-induced AIN cases.
Conclusion: This is the largest reported biopsy series of omeprazole-induced AIN. In this series of 25 renal biopsies with omeprazole-induced acute interstitial nephritis, the predominant inflammatory cells were mainly of a Th1-Th17 lineage, suggesting this is the major type of cell-mediated inflammatory process rather than a Th2-mediated response, which has been reported with the classic allergic acute drug-induced interstitial nephritis. With the increasing prevalence of omeprazole-induced AIN, it is now crucial to identify how omeprazole might induce these presumed Th1-Th17 mediated inflammatory pathways of injury within the kidney. Once better defined, more appropriate therapy can be introduced.