ACUTE INTERSTITIAL NEPHRITIS
Acute interstitial nephritis (AIN) is a major cause of intrarenal acute kidney injury (AKI) and features diffuse inﬂammation and edema of the tubulointerstitium. It accounts for a small fraction of AKI in general but is seen in up to 25% of patients with AKI who undergo a renal biopsy, generally after more common causes (such as prerenal state and acute tubular necrosis) have been excluded.
The major known causes of AIN fall into three broad categories: drugs, infectious diseases, and autoimmune disorders. Since the implicated drugs and infectious pathogens only cause AIN in a small fraction of patients, the host’s immune response is likely critical to the disease pathogenesis.
Drug reactions account for over two thirds of AIN cases. Although associations with many different drugs have been reported, the most frequent culprits include β-lactam antibiotics, rifampin, sulfonamides, diuretics, proton pump inhibitors, and nonsteroidal anti-inﬂammatory drugs. In the past, the major cause of drug-induced AIN was methicillin, which caused disease in up to one in ﬁve patients, but this antibiotic is no longer used in the United States. Nonetheless, the incidence of drug-induced AIN is rising overall because of increasing drug use, especially in the older population.
Many drugs cause disease by inciting a hypersensitivity-type reaction. β-lactams, for example, can serve as haptens by attaching to proteins on the tubular basement membrane, and forming an antigen that triggers a T-cell response. NSAIDs, however, appear to trigger disease through a largely nonallergic mechanism. Although the exact mechanism is unknown, it has been hypothesized that selective suppression of renal cyclooxygenase enzymes leads to increased metabolism of arachidonic acids toward leukotrienes, which trigger an immune response. NSAIDs may also infrequently induce a hypersensitivity-type response.
Infections account for 15% of AIN cases, and responsible agents can include bacteria (e.g., Campylobacter, Salmonella, Streptococcus, Staphylococcus, Escherichia coli, Brucella), viruses (e.g., cytomegalovirus, Epstein-Barr virus, HIV, herpes simplex virus), fungi (e.g., Histoplasma), and parasites (e.g., Leishmania donovani). Such agents can induce inﬂammation either through direct invasion of the renal parenchyma or through activation of the immune system in remote organs with collateral tubulointerstitial involvement. Infectious agents remain an important cause of AIN in developing nations.
Autoimmune diseases account for 10% of AIN cases, and responsible diseases include systemic lupus erythematosus, sarcoidosis, Behçet disease, and Sjögren syndrome.
The remaining AIN cases are considered idiopathic; however, antitubular basement membrane (TBM) nephritis and tubulointerstitial nephritis/uveitis (TINU) syndrome are now recognized as two causes of previously “idiopathic” AIN. Anti-TBM nephritis usually occurs in early childhood and results from circulating anti-TBM antibodies that target the proximal tubular basement membrane. TINU syndrome was ﬁrst described in the 1970s, and a small number of cases has been reported since that time. The athogenesis is unknown but likely immune-mediated.
PRESENTATION AND DIAGNOSIS
Acute interstitial nephritis typically manifests as AKI following the recent introduction of a new medication. Eighty percent of patients develop symptoms within 3 weeks of drug introduction, although there can be a latent period of several months following onset of NSAID use. The AKI can manifest either as oliguria or as an asymptomatic elevation in serum creatinine con-centration noted on routine serum chemistries. In classic descriptions, the renal injury is accompanied by the triad of fever, rash, and eosinophilia; however, this picture emerged when the major pathogenetic agent was methicillin, which often triggered a hypersensitivity-type reaction. At present, largely because of the growing incidence of NSAID-related AIN, allergic symptoms are less consistent. Fever, rash, and eosinophilia are each seen in about 15% to 25% of patients, and the entire triad is seen in only 10%. In addition to these variable allergic symptoms, a fraction of patients may experience ﬂank pain, gross hematuria, or both. Flank pain likely represents distention of the renal capsule secondary to interstitial edema. Hypertension and gross edema are uncommon.
Urinalysis often reveals proteinuria, which is mild on quantitative analysis (i.e., <2 g/day) and reﬂects impaired tubular reabsorption of ﬁltered proteins. Nephrotic-range proteinuria may rarely be seen in those cases where NSAID exposure causes both AIN and minimal change disease (see Plate 4-8). Microscopic analysis of urine often reveals white blood cells (WBCs), red blood cells (RBCs), and WBC casts. These ﬁndings can facilitate the distinction from acute tubular necrosis (ATN, see Plate 4-3), which is the most common cause of intrarenal AKI and typically features a bland sediment or epithelial cell casts. In addition, the lack of RBC casts or dysmorphic RBCs facilitates the distinction from acute glomerulonephritis. Finally, the presence of proteinuria and an active sediment can be used to exclude prerenal state, which may also occur in the setting of NSAID use secondary to interfe ence with tubuloglomerular feedback (see Plate 3-18).
Eosinophiluria (deﬁned as eosinophils >1% of WBCs seen in urine) occurs in some patients with AIN but can only be detected using special stains, such as Hansel or Wright stains. Moreover, eosinophiluria may be a non- speciﬁc ﬁnding because it can also occur in atheroembolic renal disease, urinary tract infections, and some glomerulonephritides. Thus its diagnostic utility is unclear.
Renal ultrasound results are often normal. Diffuse cortical echogenicity secondary to interstitial inﬂammation has been described, but no studies have validated the sensitivity or speciﬁcity of this ﬁnding. Gallium scan has been proposed as a potentially useful diagnostic tool. Gallium is a radioactive tracer that colocalizes with WBCs and has traditionally been used for the detection of abscesses. In acute interstitial nephritis, there is diffuse, bilateral uptake of gallium, which reﬂects the underlying inﬂammatory process. There are conﬂicting results, however, regarding the sensitivity and speciﬁcity of this procedure for the diagnosis of AIN. Thus it is seldom used in clinical practice.
The distinction between AIN or atheroembolic renal disease can sometimes be challenging because both may cause AKI with eosinophiluria and mild proteinuria, fevers, arthralgias, and rash. The rash in AIN, however, is typically maculopapular and erythematous, whereas atheroembolic disease usually causes livedo reticularis or mottled, violaceous toes. Atheroembolic disease is also more likely in certain high-risk populations, such as elderly patients with vascular disease who have recently undergone an open surgical or percutaneous procedure.
Biopsy is required to conﬁrm the diagnosis of AIN. It is typically performed in patients with unexplained AKI and a cellular urine sediment who do not quickly respond to termination of potentially causative drugs. On light microscopy, AIN features a lymphocyte-pre-dominant interstitial inﬁltrate typically accompanied by edema. The presence of eosinophils suggests a druginduced, allergic cause. Occasionally, granulomas may also be noted. Tubular injury may occur, with passage of lymphocytes across the tubular basement membrane (“tubulitis”), but the glomeruli and blood vessels are typically spared. Inﬂammation is typically much more prominent in the renal cortex than in the medulla.
Initial treatment for acute interstitial nephritis includes discontinuation of all potential offending drugs and eradication of any potential infections. Once an offending drug has been identiﬁed it should never be reintroduced because it will reliably cause future episodes of interstitial nephritis.
In addition, there is recent evidence that early steroid administration in drug-induced disease leads to faster and greater recovery of renal function. Thus, in the absence of any contraindications, a limited course of corticosteroids may be considered.
Most patients will experience complete recovery of renal function. A minority will progress to end stage renal disease and require renal replacement therapy. The duration of renal failure, rather than the peak serum creatinine concentration, appears to be the most important indicator of eventual recovery. Some data also suggest patients of advanced age may have a less favorable prognosis.