Newman's Notions | May 29, 2024 | FREE
Most ACP Hospitalist content is available exclusively to ACP Members. This article is free to the public.

The rise and fall of the urine eosinophil

A resident teaches an attending something about high-value kidney assessment.

Resident: "Mr. Johnson has developed an acute kidney injury during this hospitalization. I reviewed his medication list, and I am concerned that he may have acute interstitial nephritis."

Attending: "Let's order urine eosinophils. I've relied on them since I went to medical school in the 1980s. They are one of my favorite tests, second only to a Wassermann."

Resident: "Let's discuss first."

Illustration by David Rosenman
Illustration by David Rosenman

Acute interstitial nephritis (AIN) is present in approximately 6% to 30% of all acute kidney injuries in the hospital. Technically, any drug can cause AIN, but common medications associated with AIN include NSAIDs, penicillin and cephalosporins, proton-pump inhibitors, diuretics, sulfonamides, and immune checkpoint inhibitors. It can also result from systemic disease (i.e., lupus or sarcoidosis) or infections (i.e., Legionella or leptospirosis).

Urinalysis is part of the conventional workup for acute kidney injury. Urinalysis may be the oldest form of laboratory medicine, used as far as back as 6,000 years ago. For example, Hindu civilization noted sweetness in some patients' urine, which attracted black ants, the earliest reports of glucosuria. Hippocrates described bubbles on the surface of fresh urine as a sign of long-term kidney disease and also associated urinary sediment with fever. In the Middle Ages, uroscopy reached its peak.

Attending: "That's pretty cool history. You know what I like."

Resident: "I'm not finished."

As early as the late 1960s, urine eosinophils were reported in kidney transplant rejections, suggesting they were related to kidney inflammation. The initial recorded use of urine eosinophils as a marker of AIN was in 1978. In the 1980s, in a study using Wright stain, urine eosinophils were found in 65 out of 470 patients with kidney injury; however, only nine of the 65 had a diagnosis of AIN, which was made on clinical grounds and not even confirmed with renal biopsy. Subsequently, Hansel stain was found to improve the sensitivity of urine eosinophils for AIN diagnosis from 25% to 62.5%.

After these two studies, the use of urinary eosinophils to diagnose AIN gained traction, even though many subsequent studies with larger sample sizes demonstrated that neither the presence nor absence of urine eosinophils provides any useful diagnostic information. In one study with 566 patients with acute kidney injury who underwent kidney biopsies, the presence of urine eosinophils was 30% sensitive and 68% specific. In the 179 patients who did have eosinophils on urinalysis, AIN was diagnosed in about 15% based on biopsy, and in the 387 patients without eosinophils, AIN was diagnosed in 16% on biopsy.

Furthermore, urine eosinophils can also be present with other causes of renal injury, such as acute tubular necrosis and acute glomerulonephritis. This emphasizes the importance of evaluating the entire clinical picture, rather than relying on a single diagnostic test. Instead of obtaining urine eosinophils, if there is high suspicion of AIN and other etiologies of acute kidney injury have been ruled out, consider stopping any potential offending medications. The definitive diagnosis of AIN is based on a renal biopsy, and the risk-benefit ratio should be considered prior to ordering one.

Attending: "OK, that's fascinating. You sure know AIN. So you want to consult nephrology to get a kidney biopsy? Go ahead! But first make sure to check for urine eosinophils."

Resident: (Rolls eyes.)