Urine analysis: urine dipsticks ruled out blood and leukocytes when completely negative.

Clinical bottom line (level 4)

  1. Dipstick urine tests for haematuria and pyuria were useful for ruling in or ruling out the conditions on microscopy.
Bonnardeaux et al: Clinical Nephrology 1994; 41 (3): 167-172
Expires February 2003

The study

Setting: teaching hospital, Canada

- patients (aged mean 55 years, 51% female) any in- or out-patients providing a mid-stream urine sample which tested abnormal on a dipstick

Independent blinded reference standard, applied in all patients from a consecutive inappropriate spectrum.
Reference standard:
  • microscopy by two nephrologists. The sample was centrifuged, 0.4 ml of sediment was resuspended, examined under low power and the high power view was estimated. K interobserver red cells=0.94; k interobserver white cells=0.96
Diagnostic test: dipstick for protein, ketones, glucose, leukocyte esterase, occult blood, nitrites. Positive if there was a trace or more on any test.
  • 5486 samples were collected.

The evidence

pre-test probability of >5 leukocytes: 24%, (95% CI: 22% to 26%)
pre-test probability of red blood cells in urine: 11%, (95% CI: 10% to 12%)

diagnostic test >5 leukocytes 0-5 leukocytes LR
(95% CI)		 post-test probability
large abnormality on dipstick 96 22 14
(8.8 to 22) 		81%
moderate abnormality 130 46 9.0
(6.5 to 12) 		74%
small abnormality 157 192 2.6
(2.2 to 3.1) 		45%
trace abnormality 109 380 0.91
(0.75 to 1.1) 		22%
negative 74 1153 0.20
(0.16 to 0.25) 		6%
total 1793 566


diagnostic test >5 erythrocytes 0-5 erythrocytes LR
(95% CI)		 post-test probability
large abnormality 90 21 34
(22 to 54) 		81%
moderate 56 50 9.0
(6.3 to 13) 		53%
small 43 162 2.1
(1.6 to 2.9) 		21%
trace 40 360 0.89
(0.66 to 1.2) 		10%
negative 34 1508 0.18
(0.13 to 0.25) 		2%
total 263 2101


diagnostic test any casts no casts LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
any dipstick abnormality 290 2069 1.2
(1.1 to 1.2) 		12% 0.26
(0.16 to 0.44) 		3%
total 304 2511


diagnostic test red blood cell casts no red blood cell casts LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
any dipstick abnormality 49 2310 1.2
(1.1 to 1.2) 		2% 0.12
(0.017 to 0.85) 		0%
total 50 2765

  • Haematuria tests were not altered by changes in glucose, ketones or proteinuria.
  • Pyuria tests were altered if glucose >5.5 mmol/l, or if ketones >1.5 mmol/l.

Comments

  1. The population was so broad as to make comment about use of the dipstick to specific conditions impossible.
  2. Nephrologists examined a proportion of normal urine specimens every day. They calculated 43% of urine samples taken in the hospital to be abnormal during the study.

Citation

  1. Bonnardeaux A, Somerville P, Kaye M: A study on the reliability of dipstick urinalysis. Clinical Nephrology 1994; 41 (3): 167-172
Search Terms: bibliographic reference
Contributor: Catherine Clase, Chris Ball and Clare Wotton, April 2000
Reviewer: Catherine Clase

Clinical Question.
Patient patient providing mid-stream urine which tested abnormal on dipstick
Intervention or Exposure dipstick tests
Comparison microscopy
Outcome leukocytes or red blood cells in urine