Acute renal failure: urinalysis and cytodiagnostic microscopy may help diagnose glomerular and interstitial renal disease.

Clinical bottom line (level 2b)

  1. Glomerular disease was more likely in patients with acute renal disease who had renal biopsy, if there was:
    • blood on the dipstick (LR+2.7)
    • proteinuria 2 or more on dipstick (LR+1.9)
    • dysmorphic red blood cells on microscopy (LR+4.7)
  2. It was less likely if there was/were:
    • no blood on dipstick (LR-0.16)
    • no dysmorphic red blood cells on microscopy (LR-0.080)
  3. Interstitial disease was more likely if there were:
    • lymphocytes 10 or more per 10 high power fields (LR+3.0)
    • granular casts 5 or more per 10 high power fields (LR+2.1)
  4. Interstitial disease was less likely if there were fewer lymphocytes than this (LR-0.28) .
Marcussen et al: American Journal of Kidney Disease 1992; 20 (6): 618-628
Expires July 2003

The study

Setting: university hospital, Canada

77 patients (aged mean 40 years, 50%% male) 47 patients (53% male, mean age 43) with native kidney biopsies (mean creatinine 236 micromol/l; urea 11.2 mmol/l); 30 patients (47% male, mean age 37) with renal transplants (mean creatinine 338 micromol/l; urea 12.5 mmol/l)

Independent blinded reference standard, applied in all patients from a consecutive inappropriate spectrum.
Reference standard:
  • renal biopsy under ultrasound guidance, interpreted by a single renal histopathologist
Diagnostic test: cytodiagnostic urinalysis. Well-mixed first morning urine was centrifuged at 1,200 rpm for 10 minutes. Dilution of the pellet before microscopy permitted quantitative analysis of cellular elements. Slides were fixed and stained with Papanicolaou's stain. Glomerular hematuria- if dysmorphic red cells, proteinuria and casts; interstitial- inflammatory cells (lymphocytes, neutrophils or eosinophils) embedded within a cast matrix.

The evidence

pre-test probability of glomerular disease: 61%, (95% CI: 50% to 72%)
pre-test probability of interstitial disease: 31%, (95% CI: 21% to 42%)

diagnostic test glomerular disease no glomerular disease LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
blood on dipstick 42 5 2.7
(1.6 to 4.5) 		81% 0.16
(0.067 to 0.38) 		20%
protein 2 or more on dipstick 38 9 1.9
(1.2 to 2.9) 		75% 0.34
(0.17 to 0.66) 		35%
dysmorphic red blood cells 44 3 4.7
(2.3 to 9.6) 		88% 0.080
(0.026 to 0.24) 		11%
heme-granular casts 25 22 0.66
(0.48 to 0.92) 		51% 2.3
(1.1 to 5.1) 		79%
'glomerular' abnormality 38 9 4.9
(2.2 to 11) 		88% 0.23
(0.12 to 0.42) 		26%
total 47 30


diagnostic test interstitial disease no interstitial disease LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
lymphocytes 10 or more per 10 high power fields 19 5 3.0
(1.8 to 4.9) 		58% 0.28
(0.13 to 0.63) 		11%
heme-granular casts 21 3 1.7
(1.2 to 2.2) 		43% 0.27
(0.089 to 0.79) 		11%
granular casts 5 or more per 10 high power fields 14 10 2.1
(1.2 to 3.6) 		48% 0.58
(0.35 to 0.96) 		21%
total 24 53

Comments

  1. The cytodiagnostic process described differs from that routinely used for urine microscopy (for cytodiagnosis, the urine is centrifuged for longer, at higher rpm, then fixed and stained prior to examination).
  2. Patients undergoing renal biopsy differ greatly from unselected patients with oliguria or acute renal failure.
  3. Patients with renal transplants were a large proportion (46%) of those studied.
  4. Primary renal diagnoses were eventually classified as glomerular in 61%, tubular 4%, interstitial in 24% and vascular 4%. Only a small proportion of patients in this study had tubular lesions (4%), which would be the common differential diagnosis in an oliguric patient with abnormal urinary sediment. Consequently results may not be broadly applicable.

Citation

  1. Marcussen N, Schumann JL, Schumann B, et al: Analysis of cytodiagnostic urinanalysis findings in 77 patients with concurrent renal biopsies. American Journal of Kidney Disease 1992; 20 (6): 618-628
Search Terms: urinaly*, cast*, cytodiagnos*, kidney diseases, York strategy for identification of studies of diagnosis
Contributor: Catherine Clase, Chris Ball and Clare Wotton, April 2000
Reviewer: Catherine Clase

Clinical Question.
Patient acute renal failure
Intervention or Exposure cytodiagnostic urinalysis
Comparison renal biopsy
Outcome diagnosis of glomerular disease