Clinical manifestations and diagnosis of acute interstitial nephritis

INTRODUCTION — Acute interstitial nephritis (AIN) is a kidney lesion that typically causes a decline in kidney function and is characterized by an inflammatory infiltrate in the kidney interstitium [1]. It is most often induced by drug therapy. AIN is also caused by autoimmune disorders or other systemic disease (eg, systemic lupus erythematosus [SLE], Sjögren's disease, sarcoidosis), a variety of infections remote to the kidney (eg, Legionella, leptospirosis, and streptococcal organisms), immunoglobulin G4 (IgG4) disease, and tubulointerstitial nephritis with uveitis (TINU) syndrome [2-10].

An overview of the clinical manifestations and diagnosis of AIN is presented in this topic review. The treatment of AIN is discussed separately. (See "Treatment of acute interstitial nephritis".)

Interstitial nephritis associated with SLE, sarcoidosis, the TINU syndrome, and Sjögren's disease is also discussed elsewhere.

●(See "Lupus nephritis: Diagnosis and classification", section on 'Tubulointerstitial lesions'.)

●(See "Tubulointerstitial nephritis and uveitis (TINU syndrome)".)

●(See "Kidney disease in primary Sjögren's disease".)

●(See "Kidney disease in sarcoidosis".)

ETIOLOGY — In initial reports, the vast majority of cases of AIN resulted from exposure to beta-lactam antibiotics, particularly methicillin. However, drugs other than antibiotics, as well as infections and other underlying conditions, are also recognized as clinically significant causes.

The distribution of causes of AIN has been reported as follows [6,9,11-13]:

●Drugs – 70 to 75 percent (with antibiotics responsible for approximately 30 to 50 percent of these cases)

●Systemic disease including sarcoidosis, Sjögren's disease, systemic lupus erythematosus (SLE), and others – 10 to 20 percent

●Infections – 4 to 10 percent

●Tubulointerstitial nephritis and uveitis (TINU) syndrome – less than 5 percent

Drugs — Virtually any drug can cause AIN, although only a few have been reported with any frequency. While there are single case reports of many drugs associated with AIN, we attribute cause to a previously unreported drug only when the study reports that AIN was confirmed by kidney biopsy and the patient was not concomitantly taking other drugs that can cause AIN. In many cases, a causal link with the drug is established by observing the onset of kidney function impairment after initiating the drug and its resolution after discontinuing the drug.

On the other hand, it may be difficult to identify the culprit drug in some patients with biopsy-proven AIN who are taking multiple medications.

AIN was particularly common with methicillin, occurring in up to 17 percent of patients who had been treated for more than 10 days [5,14,15]. Methicillin is no longer manufactured.

Other common drugs that cause AIN include [2-4,13,15-29]:

●Nonsteroidal antiinflammatory agents (NSAIDs), including selective cyclooxygenase (COX)-2 inhibitors

●Penicillins and cephalosporins

●Rifampin

●Antimicrobial sulfonamides, including trimethoprim-sulfamethoxazole

●Ciprofloxacin and, perhaps to a lesser degree, other quinolones

●Diuretics, including loop diuretics such as furosemide and bumetanide, and thiazide-type diuretics

●Cimetidine (only rare cases have been described with other H-2 blockers such as ranitidine)

●Allopurinol

●Proton pump inhibitors (PPIs) such as omeprazole and lansoprazole

●5-aminosalicylates (eg, mesalamine)

●Anticancer drugs, such as immune checkpoint inhibitors (ICPi; such as ipilimumab and nivolumab)

The development of drug-induced AIN is not dose-dependent, and a recurrence or exacerbation of AIN can occur with a second exposure to the same or a related drug [30].

Several studies have shown an association between PPIs and AIN [26-29]. (See "Proton pump inhibitors: Overview of use and adverse effects in the treatment of acid related disorders", section on 'Kidney disease'.)

AIN can be a common complication among patients with inflammatory bowel diseases (IBD). In most patients with IBD, AIN has been attributed to treatment with 5-aminosalicylates [31,32], although AIN has also been described in treatment-naïve patients [33].

Different types of anticancer drugs can cause AIN [34]. AKI has been reported in approximately 1 to 5 percent of patients receiving an ICPi (ie, ipilimumab, nivolumab, pembrolizumab, atezolizumab); AIN is the most common type of kidney pathology in such patients [35-40]. In the largest observational study of ICPi-associated AKI, 93 percent of the patients biopsied had AIN [39]. Median time of onset of AKI was 14 weeks (interquartile range 6 to 37 weeks). A lower baseline estimated glomerular filtration rate, concomitant use of PPIs, and combination therapy with different ICPis were independently associated with an increased risk of developing AKI. The vast majority of patients stopped the ICPi (97 percent) and were treated with glucocorticoids (86 percent). Complete or partial recovery was observed in 85 percent of patients. Re-introduction of the ICPi led to a recurrence of AKI in only 23 percent of patients. Additional details of kidney toxicity related to ICPis are discussed elsewhere. (See "Toxicities associated with immune checkpoint inhibitors", section on 'Kidney'.)

Infections — Multiple organisms have been associated with AIN including Legionella, Leptospira, cytomegalovirus (CMV), Streptococcus, Mycobacterium tuberculosis, Corynebacterium diphtheriae, Epstein-Barr virus (EBV), Yersinia, polyomavirus, Enterococcus, Escherichia coli, adenovirus, Candida, and others [10,13,41-43]. A histologic variant of AIN that is characterized by granuloma formation has been associated with Mycobacterium, fungi (histoplasmosis, coccidioidomycosis), bacteria (Brucella, Chlamydia, Francisella), spirochetes (Treponema), and parasites (Leishmania, Toxoplasma) [44]. However, the most common etiology of granulomatous interstitial nephritis in westernized countries is medication-related disease [45]. (See 'Histology' below.)

Initial reports suggested that organisms such as Legionella, Leptospira, CMV, and Streptococcus primarily invaded organs remote from the kidney and exerted an inflammatory response in the kidney without invading the kidney [42,43]. However, more recent reports describe the identification of organism-specific antigens or DNA in kidney proximal tubule cells of patients with AIN [41,46-48].

Drug-induced AIN is also relatively common among HIV-infected patients, although infections and immunologic syndromes associated with HIV infection can also induce AIN [49,50].

AIN has rarely been reported in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus infection. With infection, the AIN may have been related to medications rather than the viral infection itself [51,52]. AIN has also been reported after vaccination for SARS CoV-2 although proof of the vaccination as the etiologic stimulus is unproven [53].

Associated with systemic disease — Numerous systemic disorders have been associated with AIN. These primarily include SLE, sarcoidosis, and Sjögren's disease. In a series of 133 patients with biopsy-proven AIN, of autoimmune etiologies, sarcoidosis was the most common [13]. (See "Kidney disease in primary Sjögren's disease" and "Kidney disease in sarcoidosis" and "Lupus nephritis: Diagnosis and classification", section on 'Tubulointerstitial lesions'.)

Patients with SLE and those with granulomatosis with polyangiitis often have interstitial nephritis accompanying the characteristic glomerular disease and may rarely present with AIN, even in the absence of glomerular disease.

Less common causes of AIN include IgG4-related disease [54,55] and hypocomplementemic tubulointerstitial nephritis [56-58]. IgG4-related disease is a systemic disorder characterized by the infiltration of multiple organs by a lymphoplasmacytic infiltrate that is rich in IgG4-positive plasma cells, resulting in diverse clinical manifestations, including autoimmune pancreatitis, enlarged lacrimal and salivary glands and periorbital tissue, and tubulointerstitial nephritis [54,59-65].

A rare cause of AIN results from anti-tubular basement membrane (TBM) antibodies, leading to linear staining with immunoglobulin along the TBMs on immunofluorescence microscopy [66-69]. This can occur in the presence or absence of concurrent anti-glomerular basement membrane antibodies [66,67] and has been described in patients with membranous nephropathy [69]. (See "Membranous nephropathy: Pathogenesis and etiology" and "Anti-GBM (Goodpasture) disease: Pathogenesis, clinical manifestations, and diagnosis", section on 'Kidney biopsy'.)

Tubulointerstitial nephritis and uveitis (TINU) syndrome — Some patients with interstitial nephritis have the TINU syndrome. Patients present with interstitial nephritis and uveitis and occasionally with systemic findings including fever, weight loss, fatigue, malaise, anorexia, asthenia, abdominal and flank pain, arthralgias, myalgias, headache, polyuria, and/or nocturia. The TINU syndrome is discussed elsewhere. (See "Tubulointerstitial nephritis and uveitis (TINU syndrome)".)

Older patients — There may be differences in the distribution of underlying causes of AIN among older and younger patients. In one review, compared with patients ages 18 to 64 years (n = 88), patients older than 65 years (n = 45) were more likely to have drug-induced AIN (64 versus 87 percent) and less likely to have AIN related to autoimmune or systemic diseases (27 versus 7 percent) [70]. There is no known biologic reason for this predilection, which is most likely related to polypharmacy in older individuals. The most common causative agents among older patients were penicillin and omeprazole.

Two other forms of AIN have been described. One of them is characterized by a diffuse interstitial infiltration by immunoglobulin M (IgM)-positive plasma cells [71]. The clinical findings included a mild-to-moderate kidney function impairment and a high prevalence of distal renal tubular acidosis, Fanconi syndrome, and antimitochondrial antibodies. Serum IgM levels were markedly elevated. Corticosteroid treatment led to kidney function improvement in this report. Another study reported 10 cases of tubulointerstitial nephritis caused by circulating antibodies reactive with megalin, an antigen present in the proximal tubule. The patients were older, presented with AKI and subnephrotic proteinuria, and responded poorly to corticosteroids [72].

EPIDEMIOLOGY — AIN is demonstrated in 1 to 3 percent of all kidney biopsies [73,74]. When analysis is restricted to biopsies performed in the setting of acute kidney injury (AKI), the percentage rises to 13 to 27 percent [74,75].

Some studies suggest that the incidence of AIN is increasing, particularly among older subjects [13,70,74]. The reasons for this change are complex and difficult to assess but may include an aging population base, increasing polypharmacy (including nonsteroidal antiinflammatory drugs [NSAIDs] and proton pump inhibitors [PPIs]) in older people, and an increasingly less restrictive policy of kidney biopsies in older patients with AKI. The true incidence of AIN may also be underestimated for the following reasons [26]:

●A confirmatory kidney biopsy is often not done in older and frail patients; empirical treatment is often preferred.

●The prevalence of AIN caused by PPIs, 5-aminosalicylates, and NSAIDs and often without obvious associated symptoms is increasing [26]. Skin rash and eosinophilia are significantly less common with these drugs than in antibiotic-induced AIN. (See 'Clinical manifestations' below.)

CLINICAL FEATURES

Clinical manifestations — With AIN from any cause, patients may present with nonspecific signs and symptoms of acute kidney dysfunction. These may include the acute or subacute onset of nausea, vomiting, and malaise. However, many patients are asymptomatic [10]. Patients may be oliguric or nonoliguric; in a retrospective study that included 60 cases of AIN (92 percent of which were drug induced, with the remainder idiopathic), oliguria was present among 51 percent [11]. Gross hematuria occurs in approximately 5 percent of individuals [11].

Patients usually do not have significant proteinuria, and nephrotic syndrome occurs in <1 percent of patients with AIN [11]. An exception occurs among patients who have nonsteroidal antiinflammatory drug (NSAID)-induced AIN, which may occur concurrently with NSAID-induced membranous nephropathy or minimal change disease. (See 'NSAID-induced AIN and nephrotic syndrome' below and "Membranous nephropathy: Pathogenesis and etiology", section on 'Drugs'.)

Patients may present with symptoms related to the cause of the AIN. Classically, patients with drug-induced AIN were reported to have symptoms and/or signs of an allergic-type reaction, including rash, fever, and eosinophilia [76]. However, in a more recent review of three series that totaled 128 patients with AIN (of whom 70 percent had drug-induced disease), these findings of a typical allergic response were relatively less common at presentation [9]:

●Rash – 15 percent

●Fever – 27 percent

●Eosinophilia – 23 percent

●Triad of rash, fever, and eosinophilia – 10 percent

A similar incidence of findings was reported in two retrospective series, which collected a total of 121 patients [77,78]. Rash, fever, eosinophilia, and the triad were observed in 22, 36, 35, and 11 percent, respectively [11]. Arthralgias were observed in 45 percent of the patients [11].

Thus, the originally described classic triad is less commonly observed than initially reported. This is probably due to the absence of cases of methicillin-induced AIN and (perhaps) the increased inclusion of cases not directly resulting from an allergic response [10]. In addition, some agents, such as NSAIDs and proton pump inhibitors (PPIs), are less commonly associated with fever, rash, and eosinophilia compared with other agents [78,79].

There is no typical range of time of onset for medication-induced AIN. The onset of drug-induced AIN following drug exposure may range from three to five days (as occurs with a second exposure to an offending drug), to as long as several weeks, to many months (as occurs following a first exposure to an offending drug) [2,3]. However, the latent period from initial drug exposure may be as short as one day with rifampin [3] or as long as 18 months with an NSAID [79].

Patients who have AIN that is not related to a drug may have symptoms related to an associated infection or systemic condition such as systemic lupus erythematosus (SLE), sarcoidosis, the tubulointerstitial nephritis with uveitis (TINU) syndrome, and Sjögren's disease. (See "Tubulointerstitial nephritis and uveitis (TINU syndrome)" and "Kidney disease in primary Sjögren's disease" and "Kidney disease in sarcoidosis" and "Lupus nephritis: Diagnosis and classification", section on 'Tubulointerstitial lesions'.)

Patients with immunoglobulin G4 (IgG4)-related interstitial nephritis may have extrarenal signs and symptoms. In a series of 23 patients, nonspecific symptoms such as fever, arthralgias, skin lesions, and edema were present in three, five, one, and two patients, respectively [63].

Overall, 96 percent of patients with IgG4-related AIN had extrarenal lesions, including sialadenitis in 19 (82 percent), lymphadenopathy in 10 (44 percent), autoimmune pancreatitis in 9 (39 percent), dacryoadenitis in 7 (30 percent), and lung lesions (interstitial pneumonia and nodular lesions) in 6 (26 percent) [63].

Laboratory and radiographic findings — In general, patients with AIN present with some combination of the following laboratory findings, with some variation based upon the underlying cause [2-4,9,77]:

●Increased plasma creatinine – Virtually all patients have a rise in the plasma creatinine concentration on presentation [77,78]. If AIN is drug induced, the increase in creatinine is temporally related to administration of the offending drug. Acute kidney injury (AKI) may be severe; in two retrospective series, among 121 patients who presented with AIN, 40 percent required dialysis [77,78].

●Eosinophilia and eosinophiluria – Although eosinophilia (defined by an absolute blood eosinophil count of ≥500 eosinophils/microL) is only found in 25 to 35 percent of AIN cases [9,11,77,78], its finding in a patient with AKI with no other apparent cause should raise the suspicion of drug-induced AIN. Eosinophiluria, defined by eosinophils that account for more than 1 percent of urinary white cells by Hansel stain [5,80], has been associated with AIN [81]. However, urinary eosinophils are not useful in distinguishing AIN from other causes of AKI, and the absence of eosinophiluria does not exclude the possibility of AIN. The lack of clinical utility of eosinophils in diagnosing AIN was best shown in a retrospective study that correlated urinary eosinophils with biopsy-proven AIN [82]. Five-hundred sixty-six patients had both a kidney biopsy and a test for urinary eosinophils performed for AKI. Among 179 patients who had a positive test for urinary eosinophils (defined as ≥1 percent of urinary white cells), only 28 had AIN on biopsy. Conversely, among 387 patients who had a negative test for eosinophils, 63 had biopsy-proven AIN. In this study, urinary eosinophils were found in multiple other kidney diseases, including acute tubular necrosis and crescentic and proliferative glomerulonephritis, and their presence did not alter the pretest probability of AIN on biopsy.

Some reports [78,79], though not all [82], have suggested that eosinophilia and eosinophiluria are less common in AIN induced by NSAIDs compared with other drugs.

●A characteristic urine sediment – The urine sediment usually reveals white cells, red cells, and white cell casts (picture 1A-B). Red blood cell casts, which are typically seen in glomerulonephritis, have also been described in AIN, although this is rare [83]. Some patients will, however, have no urinary sediment findings or isolated microhematuria and leukocyturia. The absence of urinary findings does not exclude a diagnosis of AIN.

●A variable degree of proteinuria – Proteinuria can range from none or minimal to >1 g/day. In two retrospective series that included a total of 121 patients, the mean and median protein excretions were 0.9±1.1 g/day (range 0 to 6 g/day) and 0.70 g/day (interquartile range 0.39 to 1.0 g/day), respectively [77,78]. Older individuals may be more likely to have significant proteinuria [75].

Occasional patients will have nephrotic-range proteinuria [2,3,11]. Concurrent nephrotic syndrome due to minimal change disease or membranous nephropathy can rarely be seen with NSAIDs and, in selected cases, induced by ampicillin, rifampin, interferon, or ranitidine [11,24,79,84,85]. In one study cited above, however, although proteinuria was significantly higher among NSAID-induced AIN as compared with other types of drug-induced AIN, nephrotic-range proteinuria was rare [78]. In addition, although these and other drugs may induce heavy proteinuria [14], an underlying disease (such as diabetic nephropathy or glomerulonephritis due to bacterial endocarditis) may be responsible for at least part of the proteinuria in some patients.

●Evidence of tubulointerstitial damage – Signs of tubulointerstitial damage, such as the Fanconi syndrome and renal tubular acidosis, may be present but rarely dominate the clinical picture [25].

●High fractional sodium excretion – The fractional excretion of sodium (FENa) may be >1 percent, which is in part indicative of tubular damage [2]. Calculators for the FENa are available using either standard units (calculator 1) or SI units (calculator 2) (see "Fractional excretion of sodium, urea, and other molecules in acute kidney injury", section on 'Fractional excretion of sodium in acute kidney injury'). However, lower values may be seen, particularly in patients who are nonoliguric and have less severe kidney failure [86,87].

●Radiographic findings – There are no radiographic findings that are diagnostic for AIN. Radiographic findings, including marked enlargement of kidneys with low-attenuation lesions, may be seen among patients with IgG4-related AIN [55].

IgG4-related disease and hypocomplementemic interstitial nephritis — In addition to the above findings, patients with immunoglobulin G4 (IgG4)-related disease or hypocomplementemic interstitial nephritis usually have elevated serum total IgG and/or IgG4 levels or hypergammaglobulinemia and may have low serum complement concentrations [63,88]. In one series of IgG4-related tubulointerstitial nephritis, among 23 patients, complement C3, C4, or both were reduced in 16 [63].

Microbiologic features unique to different culprit organisms are presented separately. (See appropriate topic reviews.)

DIAGNOSIS — AIN should be suspected in a patient who presents with an elevated serum creatinine and a urinalysis that shows white cells, white cell casts, and, in some cases, eosinophiluria. Red blood cells and, rarely, red blood cell casts can be also observed [89,90], although gross hematuria is distinctly unusual. Drug-induced AIN should be suspected when the onset of characteristic laboratory findings is temporally related to the initiation of a new drug, particularly one that has been previously reported to cause AIN. However, occasional patients have a bland sediment with few cells or casts [18]. Thus, a relatively normal urinalysis should not exclude the diagnosis.

A definitive diagnosis of AIN is made by kidney biopsy. It is often considered unnecessary to make a definitive diagnosis, such as among patients who have clearly documented onset of kidney failure after initiation of a common culprit drug and who improve immediately upon stopping the offending agent.

We suggest a kidney biopsy for the following patients who are suspected of having AIN:

●Patients who have a characteristic urinalysis for AIN but are not being treated with a drug known to cause AIN.

●Patients who are being treated with a drug known to cause AIN but do not have a characteristic urinalysis. Some of the drugs that cause AIN can also produce other forms of acute kidney injury (AKI). As an example, nonsteroidal antiinflammatory drugs (NSAIDs) can exacerbate prekidney disease by inhibiting the production of vasodilator prostaglandins [79]. (See "NSAIDs: Acute kidney injury".)

●Patients who are being considered for treatment with glucocorticoids for AIN (usually drug induced). Among selected patients (such as those at high risk of complications of a biopsy or who do not wish to undergo a biopsy), glucocorticoids may be initiated without first obtaining a kidney biopsy. However, among patients who do not improve after the first five to seven days of empiric glucocorticoid therapy, those without contraindications should have a biopsy in order to exclude other diagnoses or severe interstitial fibrosis. (See "The kidney biopsy", section on 'Contraindications' and "Treatment of acute interstitial nephritis".)

●Patients with putative drug-related AIN who are not treated with glucocorticoids initially and do not have a recovery following cessation of drug therapy [2,77]. We typically perform a kidney biopsy approximately two weeks after cessation of the offending drug if there has been no improvement in the kidney function over that time. However, clinicians wait a variable amount of time to allow for recovery before performing a biopsy. Some wait for three to four days, whereas others wait until the patient is approaching the need for kidney replacement therapy.

●Patients who present with advanced kidney failure, providing the onset of kidney failure is known to be relatively recent (ie, within three months).

●Patients with any features (such as high-grade proteinuria) that cause the diagnosis of AIN to be uncertain.

Patients who have a characteristic urinalysis for AIN but do not have an elevated creatinine may also be considered for biopsy, but such patients rarely come to medical attention since the urinalysis is usually only performed after the detection of an increased serum creatinine. Moreover, it is usually simpler and safer just to remove potential offending medications in such cases.

The approach to the treatment of patients diagnosed with AIN, tubulointerstitial nephritis with uveitis (TINU), and kidney sarcoidosis is presented separately.

●(See "Treatment of acute interstitial nephritis".)

●(See "Kidney disease in sarcoidosis".)

●(See "Tubulointerstitial nephritis and uveitis (TINU syndrome)".)

Histology — The major histologic changes are interstitial edema and a marked interstitial infiltrate consisting primarily of T lymphocytes and monocytes (picture 2A-E) [2,9]. Eosinophils, plasma cells, and neutrophils also may be found. The classic lesion of "tubulitis" is found when inflammatory cells invade the tubular basement membrane (TBM).

Some histologic features may suggest particular variants of AIN. As an example, granuloma formation is particularly characteristic of sarcoidosis, although it may be seen in any form of AIN [91]. In some series, granuloma formation also suggests a greater likelihood of infection-induced AIN compared with AIN without granulomas. In a review of 40 biopsies of patients with granulomatous kidney disease (including 37 patients with interstitial nephritis, two with associated pauci-immune crescentic glomerulonephritis, and one with vasculitis), sarcoidosis was present in 20 patients (50 percent) and drug-induced and Mycobacterium infection present in seven (18 percent) and five (13 percent), respectively [92]. However, in a number of other series, medication-induced interstitial nephritis was the leading cause of granulomatous interstitial nephritis, equaling or exceeding sarcoidosis in all.

Infection is a more likely etiology in areas where infections associated with granulomatous disease are more common than in westernized countries [45]. As described above, other infections that have been associated with granulomatous AIN include fungi (histoplasmosis, coccidioidomycosis), bacteria (Brucella, Chlamydia, Francisella), spirochetes (Treponema), and parasites (Leishmania, Toxoplasma) [44]. (See 'Infections' above.)

Characteristic histologic features that suggest immunoglobulin G4 (IgG4)-related disease include the presence of TBM immune complex deposits and an increase in IgG4-positive plasma cells in the interstitium [55].

Patients with interstitial nephritis related to lupus usually have concurrent glomerular lesions. (See "Lupus nephritis: Diagnosis and classification", section on 'Clinical features' and "Lupus nephritis: Diagnosis and classification", section on 'Histopathologic classification of LN' and "Lupus nephritis: Diagnosis and classification", section on 'Other forms of lupus kidney disease'.)

Differential diagnosis — The differential diagnosis of AIN includes all other causes of AKI. The diagnostic approach to the patient with AKI from any cause is presented elsewhere. (See "Diagnostic approach to adult patients with subacute kidney injury in an outpatient setting".)

In general, the urinary findings will help distinguish AIN from other causes of AKI. The urinalysis, for example, typically shows granular and epithelial cell casts and free epithelial cells in acute tubular necrosis; red cell casts, as well as red and white cells in acute glomerulonephritis; and few, if any, abnormalities in prekidney disease and obstruction.

Among patients with a predominance of white blood cells and white blood cell casts, renal atheroemboli should be considered, particularly among older patients [93]. Similarly to AIN, renal atheroemboli may present with eosinophiluria, eosinophilia, and skin lesions. However, atheroemboli are more commonly associated with a reticular pattern (livedo reticularis) and digital infarcts, whereas AIN is associated with a characteristic rash that is diffuse and maculopapular. The history may also distinguish between AIN and renal atheroemboli. Many atheroemboli occur in the setting of atherosclerotic cardiovascular disease and are preceded by an endovascular procedure. (See "Clinical presentation, evaluation, and treatment of renal atheroemboli" and "Clinical presentation, evaluation, and treatment of renal atheroemboli", section on 'Risk factors'.)

Among patients who have a completely negative sediment, as in all patients with AKI of obscure origin, obstruction should be considered as part of the differential diagnosis. Imaging studies (usually an ultrasound) generally exclude the presence of obstruction, except in rare cases when the diagnosis of AKI is made within the first two to three days (see "Clinical manifestations and diagnosis of urinary tract obstruction (UTO) and hydronephrosis"). In all patients, imaging should be done prior to kidney biopsy.

Some novel noninvasive biomarkers such as tumor necrosis factor-alpha, interleukin-9, and urine ratio of retinol-binding protein-to-creatinine may help distinguish AIN from other diagnosis [94,95].

ESTABLISHING THE CAUSE — Once a diagnosis of AIN is made by biopsy, the underlying cause should be determined. As discussed above, the vast majority of cases of AIN are caused by a drug. A careful review of medications, including the timing of initiation in relation to the onset of acute kidney injury (AKI), may reveal the likely culprit agent.

Less commonly, a drug is not identified or suspected. Among such patients, a histologic diagnosis of AIN should provoke a search for underlying infection and systemic disorders including systemic lupus erythematosus (SLE), sarcoidosis, Sjögren's disease, tubulointerstitial nephritis and uveitis (TINU) syndrome, IgG4 disease, and antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis.

In some cases, histologic features may suggest the underlying disease that is associated with AIN. As an example, granulomas are more likely to be seen in patients with sarcoidosis than in other forms of AIN. Additionally, AIN related to SLE may be accompanied by characteristic glomerular lesions. Immunoglobulin G4 (IgG4)-related disease may be suggested by the presence of tubular basement membrane (TBM) immune complex deposits and an increase in immunoglobulin G4 (IgG4)-positive plasma cells in the interstitium [55]. However, testing for IgG-4 specific plasma cells is not routinely performed at most centers.

Such histologic features, although often suggestive, are not diagnostic, and, in general, the etiology of non-drug-related AIN must be established by means other than the kidney biopsy. Among patients who have AIN that is not believed to be related to a drug, we perform the following tests:

●Chest radiograph to evaluate for sarcoidosis, tuberculosis, and other infections. Among patients in whom the chest radiograph is nondiagnostic, a high-resolution chest computed tomography (CT) should be obtained to evaluate for sarcoidosis.

●Serum levels of angiotensin-converting enzyme (ACE) and measurement of serum calcium and urinary calcium excretion to evaluate for sarcoidosis.

●Serum QuantiFERON gold test or a purified protein derivative (PPD) to exclude tuberculosis, particularly in granulomatous AIN.

●Serologic tests, in select cases, to exclude histoplasmosis, coccidioidomycosis, toxoplasmosis, and Epstein-Barr virus (EBV). Urinary antigen test to exclude Legionella infection and urine culture to exclude leptospirosis.

●ANCA to exclude ANCA-associated vasculitides.

●Antinuclear antibody (ANA) and double-stranded DNA (dsDNA) to exclude SLE.

●C3 and C4 to evaluate for SLE and IgG4-related disease and hypocomplementemic AIN. These tests, however, neither diagnose nor exclude these disorders.

●Anti-Ro/SSA, anti-La/SSb antibodies, C-reactive protein, and rheumatoid factor to exclude Sjögren's disease.

●Serum protein electrophoresis.

●Slit lamp examination in patients with eye pain or redness to evaluate for tubulointerstitial nephritis and uveitis (TINU syndrome).

●The diagnostic evaluation of a particular infection-related AIN should be guided by extrarenal clinical manifestations.

NSAID-INDUCED AIN AND NEPHROTIC SYNDROME — Nonsteroidal antiinflammatory drugs (NSAIDs) may cause AIN with an interstitial infiltrate composed primarily of T lymphocytes, and with the nephrotic syndrome due to minimal change disease or membranous nephropathy [79,96-98].

This disorder is most likely to occur with fenoprofen but probably can be induced by any nonselective NSAID. There have also been case reports of selective cyclooxygenase (COX)-2 inhibitors also being associated with this pattern of injury [21,99,100].

How NSAIDs produce AIN and nephrotic syndrome are not known; it is possible that COX inhibition by the NSAID results in the preferential conversion of arachidonic acid to leukotrienes, which can then activate helper T cells.

Affected patients typically present with hematuria, pyuria, white cell casts, proteinuria, and an acute rise in the plasma creatinine concentration. The full picture of an allergic reaction (fever, rash, eosinophilia, and eosinophiluria) is typically absent, but one or more of these findings may be present. Spontaneous recovery generally occurs within weeks to a few months after therapy is discontinued [79,96]. All NSAIDs should be terminated in patients suspected of having NSAID-induced AIN. Since topically administered NSAIDs can be systemically absorbed, such therapy should also be terminated [101].

There is no definitive evidence that corticosteroid therapy is beneficial in this setting. However, a course of prednisone may be considered in patients whose kidney failure persists more than one to two weeks after the NSAID has been discontinued [2]. (See "Treatment of acute interstitial nephritis".)

Such patients should avoid the subsequent administration of NSAIDs. Relapse may occur with rechallenge [102].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Acute kidney injury in adults".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Acute interstitial nephritis (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Overview – Acute interstitial nephritis (AIN) is a kidney lesion that causes a decline in kidney function and is characterized by an inflammatory infiltrate in the kidney interstitium. Drugs, particularly antibiotics, are the most common cause of AIN. Other causes include autoimmune disorders, infections, sarcoidosis, and tubulointerstitial nephritis with uveitis (TINU) syndrome. (See 'Introduction' above.)

●Etiology

•Drugs – Drugs that most commonly cause AIN are nonsteroidal antiinflammatory drugs (NSAIDs, including selective cyclooxygenase [COX]-2 inhibitors), penicillins and cephalosporins, rifampin, antimicrobial sulfonamides, ciprofloxacin and other quinolones, diuretics, cimetidine, allopurinol, proton pump inhibitors (PPIs), 5-aminosalicylates (eg, mesalamine), and checkpoint-inhibiting monoclonal antibodies. Many other drugs can occasionally cause AIN. (See 'Drugs' above.)

•Infections – Infections that have been associated with AIN include Legionella, Leptospira, cytomegalovirus (CMV), and Streptococcus. Other, less commonly associated organisms include Corynebacterium diphtheriae, Epstein-Barr virus (EBV), Yersinia, and polyomavirus. Mycobacterium infection is associated with a granulomatous variant of AIN. (See 'Infections' above.)

•Autoimmune disorders – Autoimmune disorders that have been associated with AIN include sarcoidosis, Sjögren's disease, systemic lupus erythematosus (SLE), and granulomatosis with polyangiitis. Less common causes of AIN are immunoglobulin G4 (IgG4)-related disease, anti-tubular basement membrane (TBM) antibodies, hypocomplementemic tubulointerstitial nephritis, and the TINU syndrome. (See 'Associated with systemic disease' above.)

●Clinical features

•Clinical manifestations – Patients with AIN present with nonspecific signs and symptoms associated with acute kidney injury (AKI). Signs and symptoms of an allergic-type reaction may be present, including rash, fever, and eosinophilia, although only 10 percent of patients have the full triad of rash, fever, and eosinophilia. (See 'Clinical manifestations' above.)

•Laboratory findings – Patients generally present with a rise in the plasma creatinine concentration, which (if drug induced) is temporally related to administration of the offending drug. Urine sediment usually reveals white cells, red cells, and white cell casts. Urinary eosinophils may be present but lack the specificity and sensitivity to either exclude or diagnose AIN. Protein excretion is usually only mildly or moderately increased. (See 'Laboratory and radiographic findings' above.)

●Diagnosis – AIN should be suspected in a patient who presents with an elevated serum creatinine and a urinalysis that shows white cells, white cell casts, and, in some cases, eosinophiluria. Drug-induced AIN should be suspected when the onset of characteristic laboratory findings are temporally related to the initiation of a new drug, particularly one that has been previously reported to cause AIN. A definitive diagnosis of AIN is made by kidney biopsy. Selected patients with suspected AIN will benefit from kidney biopsy. (See 'Diagnosis' above.)

●Establishing the cause – Patients who have a histologic diagnosis of AIN and are not on a drug known to cause AIN should be evaluated for another underlying etiology. (See 'Establishing the cause' above.)