Tubulointerstitial Nephritis Workup

In general, proteinuria is usually absent or modest in acute tubulointerstitial nephritis. Urinalysis may show microscopic hematuria and/or sterile pyuria (with or without eosinophils). Although the clinical presentation is often sufficient to make the diagnosis, renal biopsy is required to make a definitive diagnosis.

Patients with acute tubulointerstitial nephritis caused by nonsteroidal anti-inflammatory drugs (NSAIDs) typically present with heavy proteinuria, often in the nephrotic range. Findings on gallium scanning have been reported to be confirmatory in the diagnosis of acute interstitial nephritis. ^{[24, 25]} Thus, a negative finding helps to rule out this diagnosis. However, findings on this test have proved to be too nonspecific, except as a confirmatory tool in suspected cases.

Clinical investigations in chronic tubulointerstitial nephritis may show modest elevation in serum creatinine, evidence of tubular dysfunction (ie, renal tubular acidosis), or Fanconi syndrome (ie, aminoaciduria, glycosuria, hypophosphatemia, hypouricemia). Proteinuria is usually mild, often less than 1 g/d. In contrast to glomerular disease, a significant fraction of the protein is low molecular weight (eg, immunoglobulin light chains, beta2 microglobulin, lysozyme, peptide hormones). These proteins are normally taken up by the proximal tubules and broken down there. Thus, in diseases predominantly involving tubular structures, decreased endocytosis of filtered proteins leads to the characteristic tubular proteinuria.

In patients with suspected lead exposure, an ethylenediaminetetraacetic acid (EDTA) lead mobilization test or determination of tibial bone lead by radiographic fluorescence can confirm lead etiology.

The diagnosis of atherosclerotic kidney disease can usually be made clinically, and radiologic investigations, such as duplex scanning of the renal arteries, digital subtraction angiography, or magnetic resonance imaging (MRI), reveal atherosclerotic stenosis of the renal arteries. Kidney biopsy is seldom necessary and, if performed, shows nonspecific changes of chronic tubulointerstitial nephritis (ie, tubular atrophy, fibrosis, and arterial or arteriolar sclerosis with paucity of cellular infiltration). Because these patients tend to have atherosclerotic complications, they are likely to experience multiple contrast procedures and hence are at risk for acute recurrent contrast nephropathy, which can accelerate progression to end-stage renal disease.

Patients with Balkan endemic nephropathy are identified easily in endemic regions by checking for tubular proteinuria. Beta2 microglobulinuria has proved particularly useful in identifying cases and has been proposed and used as a marker of the disease.

Eosinophilia, when present, can be very helpful in the evaluation of tubulointerstitial nephritis. However, this finding is neither specific nor sensitive enough to establish the diagnosis. Although the true incidence of eosinophilia in acute tubulointerstitial nephritis is unknown, it is estimated to be present in approximately half of patients. Typically, eosinophilia is absent in acute tubulointerstitial nephritis that is induced by nonsteroidal anti-inflammatory drugs (NSAIDs).

A complete set of chemistries, including blood urea nitrogen (BUN) and serum creatinine, provides information on whether kidney insufficiency exists. A low bicarbonate level (total carbon dioxide < 24-23 mEq/L) may indicate acidosis. Low serum potassium levels may indicate a proximal tubular disorder, and elevated serum potassium levels with a low bicarbonate level may indicate type 4 renal tubular acidosis, which can be observed with lead nephropathy and nonsteroidal anti-inflammatory drug (NSAID)–induced analgesic nephropathy, among other conditions.

Urinalysis may reveal proteinuria, hematuria, and the presence of white blood cells (WBCs), with or without bacteria. A microscopic analysis of urine sediment may reveal casts, WBCs, eosinophils, and crystals. If allergic interstitial nephritis is suspected, send a cytospin specimen to determine if eosinophils are in the urine. In nonsteroidal anti-inflammatory drug (NSAID)–induced acute tubulointerstitial nephritis, eosinophiluria is usually absent. Unfortunately, the absence of eosinophiluria does not rule out the diagnosis, and it can be observed in other diseases, including cholesterol microembolism, urinary tract infections, parasitic disorders, and glomerulonephritis.

Quantitative determination of urine protein may also be helpful. Low-molecular weight proteins, such as beta-2 microglobulin, retinol binding protein (RBP), alpha-1 microglobulin, and immunoglobulin light chains, are increased in chronic tubulointerstitial nephritides. Beta-2 microglobulinuria has been found helpful in the diagnosis of Balkan endemic nephropathy and cadmium nephropathy.

Hettinga and colleagues, in a prospective cohort study of 45 young patients with uveitis, found that urinary β2-microglobulin (β2M), urinary protein, and serum creatinine had predictive value for detecting tubulointerstitial nephritis and uveitis (TINU) syndrome. The positive predictive value of increased β2M levels combined with increased serum creatinine was 100% for detecting patients with definitive and/or probable TINU syndrome. ^[26]

Urinary N-acetyl-β-D-glucosaminidase (NAG) and matrix metalloproteinases (MMPs) 2 and 9 were significantly inversely correlated with the rate of decline in estimated glomerular filtration rate (GFR) over a period of 11 to 54 months in a study of 54 patients with drug-induced chronic tubulointerstitial nephritis, 10 patients with IgA nephropathy, and 20 healthy controls. These biomarkers may be able to predict deterioration in drug-induced chronic tubulointerstitial nephritis. The areas under the receiver operating characteristic curve for urinary NAG, MMP-9, MMP-2 and α1-microglobulin for predicting decline in estimated GFR were 0.879, 0.867, 0.735 and 0.709, respectively (all P< 0.05). ^[27]

Ultrasonography is noninvasive imaging technique that is extremely helpful in identifying hydronephrosis in obstructive disease as well as calculi in stone disease. Both radiolucent and radiopaque stones can be visualized with this modality. A combination of ultrasonography and flat plate kidney, ureter, and bladder (KUB) radiography is helpful in the workup and identification of radiopaque versus radiolucent stones.

Normal kidney size by ultrasonographic examination generally favors but does not prove a diagnosis of acute (thus potentially reversible) kidney disease. In contrast, small (shrunken) kidneys with increased echogenicity indicate chronic and irreversible kidney disease.

Once widely used, intravenous pyelography seldom plays a role in the workup of kidney diseases in modern medicine. In many instances, similar information can be obtained by ultrasonography without exposing the patient to potentially nephrotoxic contrast dye.

Computed tomography (CT) scanning provides information similar to ultrasonographic scanning in the workup of kidney disease, generally with greater resolution. However, an ultrasonographic examination is sufficient in most kidney diseases. A high-resolution scan showing microcalcifications in renal papillary tips can be very helpful in diagnosis of analgesic nephropathy. The kidneys may be very small in Balkan endemic nephropathy and aristolochic acid nephropathy.

Consider the possibility of lead nephropathy in patients presenting with chronic renal insufficiency, hypertension, and gout. In the absence of documented episodes of acute symptomatic lead poisoning, the medical history is not reliable in ascertaining the lead etiology in patients presenting with chronic tubulointerstitial nephritis.

Diagnosis of lead nephropathy requires an estimation of the cumulative body stores of lead by either ethylenediaminetetraacetic acid (EDTA) lead mobilization test or by determination of bone lead content by radiographic fluorescence.

The EDTA lead mobilization test is performed by measuring 24-hour urine lead excretion after intravenous or intramuscular administration of 2 g EDTA (calcium disodium versenate). Excretion of more than 0.6 g of lead per 24 hours is considered an abnormal finding.

Blood lead levels, although elevated during acute or recent exposure, are not very helpful in the evaluation of chronic lead poisoning. During acute exposure, lead is concentrated in the red blood cells (RBCs) and later extracted to tissues and bone as the RBCs senesce.

Kidney biopsy is the definitive test for diagnosing acute allergic interstitial nephritis, particularly in cases in which the clinical diagnosis is difficult. Because the differential diagnosis of acute tubulointerstitial nephritis encompasses multiple etiologies, consider kidney biopsy when the diagnosis is not obvious.

Kidney biopsy shows mononuclear and often eosinophilic cellular infiltration of the renal parenchyma with sparing of the glomeruli (see the following images). Sometimes, interstitial changes such as fibrosis and atrophy are also present (eg, Renal biopsies have shown severe interstitial fibrosis in patients with Chinese herb/aristolochic acid nephropathy).

Tubulointerstitial nephritis: Kidney biopsy reveals acute interstitial nephritis. The renal cortex shows a diffuse interstitial, predominantly mononuclear, inflammatory infiltrate with no changes to the glomerulus. Tubules in the center of the field are separated by inflammation and edema, as compared with the more normal architecture in the right lower area (periodic acid–Schiff, 40 X).

View Media Gallery

Tubulointerstitial nephritis. On a kidney biopsy, the diagnosis of acute interstitial nephritis is based on the active inflammatory infiltrate on the right with unaffected glomeruli. Interstitial edema and fibrosis are present on the left side of the field, where some tubules show thickened basement membrane (hematoxylin and eosin, 20 X).

View Media Gallery

Tubulointerstitial nephritis. Kidney biopsy shows acute interstitial nephritis. The mononuclear inflammatory infiltrate contains abundant eosinophils, suggesting an allergic etiology. Severe tubular damage is observed (hematoxylin and eosin, 40 X).

View Media Gallery

Tubulointerstitial nephritis. Kidney biopsy shows acute interstitial nephritis. The inflammatory infiltrate forms an ill-defined granuloma, suggesting allergic or infectious etiologies. A partially destroyed tubule is present (periodic acid–Schiff, 40 X).

View Media Gallery

Findings on kidney biopsy in chronic tubulointerstitial nephritis usually show varying degrees of interstitial fibrosis, tubular atrophy, fibrosis, arteriolar sclerosis, and, occasionally, patchy mononuclear cell infiltration (see the image below). Often, the findings are nonspecific and the etiology is not discernible from the biopsy; some diseases, such as sarcoidosis, show noncaseating granulomas, and, in viral diseases, immunostaining can yield clues to the cause.

Tubulointerstitial nephritis. Kidney biopsy shows chronic tubulointerstitial nephritis. The interstitium is expanded by fibrosis, with distortion of tubules and periglomerular fibrosis. Glomeruli do not show pathologic changes (hematoxylin and eosin, 20 X).

View Media Gallery

Kidney biopsy during the acute phase of cholesterol microembolism shows the characteristic needle-shaped clefts caused by the cholesterol crystals within the small- and medium-sized arterioles accompanied by patchy tubulointerstitial nephritis with mild mononuclear cellular infiltration. (Eyeground examination may also reveal the characteristic cholesterol crystals, also termed Hollenhorst plaques, in retinal vessels, which can support the diagnosis.)

Tubulointerstitial nephritis. Kidney biopsy in interstitial nephritis shows a cholesterol microembolism. The 2 arterioles in the center are occluded by elongated crystals (hematoxylin and eosin, 20 X).

View Media Gallery

Tubulointerstitial nephritis. Kidney biopsy in interstitial nephritis shows a cholesterol microembolism. The arteriole in the center of the field has a thickened wall. The lumen is occluded by elongated spaces, corresponding to dissolved crystals surrounded by cellular reaction. The 2 glomeruli flanking the arteriole are sclerotic and hardly recognizable (hematoxylin and eosin, 40 X).

View Media Gallery

Kidney biopsy is not diagnostic of lead etiology for chronic lead poisoning and shows nonspecific changes such as interstitial fibrosis, tubular atrophy, and vascular sclerosis, findings common to tubulointerstitial nephritides of other etiologies.