Acute renal failure: acute interstitial nephritis: few urinary eosinophils made it less likely.

Clinical bottom line (level 4)

  1. In patients with an active urinary sediment, <1% urinary eosinophils on Hansel's stain made acute interstitial nephritis unlikely (LR-0.11) .
  2. >1% urinary eosinophils made acute interstitial nephritis more likely but was not diagnostic (LR+6.1) .
Nolan et al: New England Journal of Medicine 1986; 315: 1516-1519
Expires July 2003

The study

Setting: renal services of university hospital, USA

92 patients (aged ?, ?% male) active urinary sediment

Independent ?blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
Reference standard:
  • clinical criteria (some combination of rash, fever, peripheral eosinophilia, azotaemia, urinalysis demonstrating haematuria and pyuria in the absence of muddy brown granular casts and rapid improvement in renal function after drug withdrawal or corticosteroid therapy); renal biopsy (two patients underwent biopsy)
Diagnostic test: urine for eosinophils by Hansel's stain. White blood cells were examined in centrifuged urine, 200 cells were counted and >1% eosinophils were considered positive.

The evidence

pre-test probability of acute interstitial nephritis: 12%, (95% CI: 5% to 19%)

diagnostic test acute interstitial nephritis no acute interstitial nephritis LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
>1% eosinophils 10 12 6.1
(3.5 to 11) 		45% 0.11
(0.016 to 0.69) 		1%
total 11 81

Comments

  1. The finding of greater than 5% eosinophils was not associated with any greater specificity.
  2. False positive urinary eosinophils were noted in rapidly progressive glomerulonephritis, postinfectious glomerulonephritis and urinary tract infection.
  3. The properties of Wright's stain for eosinophils were also examined and this was found to be much less sensitive, identifying eosinophils in only 2 of 11 patients with clinical interstitial nephritis.
  4. Interobserver agreement on the presence of urinary eosinophils was assessed in a blinded manner. K was 0.7 (good agreement).

Citation

  1. Nolan CR, Anger MS, Kelleher SP: Eosinophiluria- a new method of detection and definition of the clinical spectrum. New England Journal of Medicine 1986; 315: 1516-1519
Search Terms: reference list of retrieved paper
Contributor: Catherine Clase, Chris Ball and Clare Wotton, April 2000
Reviewer:

Clinical Question.
Patient active urinary sediment
Intervention or Exposure Hansel's stain
Comparison clinical diagnosis
Outcome diagnosis of acute interstitial nephritis