Clinical Chemistry 46: 1384-1386, 2000;
(Clinical Chemistry. 2000;46:1384-1386.)
© 2000 American Association for Clinical Chemistry, Inc.
Stability of Common Analytes in Urine Refrigerated for 24 h before Automated Analysis by Test Strips
Paul Frooma,1,
Barbara Bieganiec1,
Zahava Ehrenrich1 and
Mira Barak1
1
Central Laboratory of Haifa and Western Galilee, Clalit Health Services, Nesher, Israel.
a Address correspondence to this author at: Central Laboratory of Haifa and Western Galilee, Clalit Health Services, Nesher, Israel. Fax 972-4-8209094; e-mail paulf{at}ioh.org.il
 |
Abstract
|
|---|
Background: Central outpatient laboratories might find processing
large numbers of urinary samples that arrive in the late afternoon
inconvenient and refrigerate them overnight before testing.
Furthermore, in certain settings clinics might have difficulty assuring
that the urine arrives at the laboratory during the same day as the
collection. Because the stability of urine samples for delayed
automated dipstick analysis (Supertron) is unknown, after defining
precision, we retested urines refrigerated for 24 h to determine
stability.
Methods: Urinalysis was done twice on the same day and repeated
after the sample was refrigerated for 24 h. Combur-10S (Roche
Diagnostics) dipsticks were read automatically by a Supertron analyzer.
Repeat tests on the same day were compared with tests after storage.
Results: Leukocyte esterase had high precision, but after storage
25% of the positive samples were less reactive (P
<0.005). Precision of hemoglobin retests was also high but declined
significantly after storage for 24 h. Urine protein values
increased after storage. The precision and stability were excellent for
nitrites, glucose, and ketones.
Conclusions: The stability of the automated dipstick urinalysis
varies with the substance tested. After refrigeration for 24 h,
there is a risk of false-positive results for protein, false-negative
results for leukocytes and erythrocytes, and little effect on glucose,
nitrite, and ketone values.
|
Introduction
|
|---|
Supertron, a fully automated urinalysis system, provides
semiquantitative tests for erythrocytes, leukocytes, glucose, protein,
and nitrites (1). A counting chamber method for the numbers
of erythrocytes and leukocytes has shown a concordance of 86–98% and
an excellent correlation with other semi-automated urinalysis systems
(1). Despite a maximum throughput of up to 300 test strips
per hour, processing large numbers of urine samples arriving in the
late afternoon might be difficult. Furthermore, clinics from wide
geographical areas serviced by a central laboratory might have
difficulty assuring that the urine arrives on the same day as
collection. Although urine samples left on the tray for 60 min are
stable (1), the stability of a urine sample refrigerated for
24 h is unknown. The insert recommends testing the urine within
2 h of sampling. To determine stability, we compared the precision
of paired urinalysis tests done at the same time by automated dipstick
analysis (Supertron) to those repeated after a 24-h interval.
|
Materials and Methods
|
|---|
Urine samples from outpatient clinics arrived and were tested
within 4 h of collection. Consecutive samples (n = 190) were
tested and retested within a 1-h period after arrival, and then
retested after 24 h of refrigeration by Supertron automated
analyzer (Hitachi-Boehringer Mannheim, Mannheim, Germany) using
Combur-10 S strips (Roche Diagnostics). To increase the number of
positive tests, an additional 213 consecutive tests were tested a
single time after arrival and retested after 24 h. Thus, we used
190 tests to calculate precision and 403 tests to calculate stability.
The agreement for each element of the urinalysis was determined for
different degrees of positivity. Significant differences between
precision and stability were determined by the
2 test except when there were small numbers,
for which the Fisher exact test was used.
|
Results
|
|---|
The precision for leukocyte esterase tests was excellent (Table 1
), but the stability of positive tests declined significantly
after the storage period (P <0.005). There was little
change in the stability of negative tests. Although 25% of leukocyte
esterase tests delayed for 24 h were false negatives, nitrate
tests were both precise and stable (Table 1
).
There were few specimens with proteinuria concentrations 
5000 mg/L,
but their precision and stability appeared to be high. However, the
test-retest repeatability of specimens with low proteinuria (300
mg/L) was low, and there was a decreased stability of low test results,
with resulting false-positive results (Table 2
). Results of glucose analyses demonstrated both good precision
and stability at high concentrations, but poorer precision and
stability at lower cutoff concentrations (Table 3
). We did not use a 3000 mg/L cutoff because in the precision
measurements, there was only one specimen with a value of 3000 mg/L,
which on retest was 0 mg/L, and for stability testing, there was only
one specimen with a value of 3000 mg/L, which on retest was 0 mg/L, and
one specimen with a value of 0 mg/L, which on retest was 3000 mg/L. The
precision of the urinary red blood cell analysis was similar to that of
the leukocyte esterase, with excellent precision but increasing false
negatives (P <0.05) after 24 h (Table 4
). At the intermediate cutoff value, stability for low values
was significantly less than the precision, producing false-positive
tests. This, however, was not observed for a lower cutoff value.
|
Discussion
|
|---|
Our major finding was that urine testing for red and white blood
cells is precise but unstable if testing is delayed for 24 h. The
result can be falsely negative. Conversely, there were no significant
differences between the precision and stability for urine glucose,
nitrates, and ketones. Immediate test-retest precision was very high
for all substances except for low protein concentrations, with an
increase in false positives for proteinuria after 24 h of
refrigeration.
Semi-automated reflectance readings of urinalysis dipsticks are more
precise than visual readings (2), and studies of precision
using the Miditron Junior for analysis of dipsticks found that it was
precise (3). Studies showed that glucose, protein, and
nitrite measurements in artificially prepared samples also have high
reproducibility (>90%), but that those for hemoglobinuria are less
precise (4).
Glucose was precise only at very high concentrations (>1000 mg/L).
There was, however, imprecision at lower cutoff values, which could not
be accurately defined because of the small number of samples. The lack
of accuracy of glucose in commercially available dipsticks has been
shown previously (5). Compared with the quantitative
hexokinase method, only the Chemstrip could differentiate urine glucose
at 0.3 g/L (upper limit of normal) and 0.6 g/L. Other dipsticks
detected glucose only at concentrations >1 g/L. Part of this lack of
accuracy might be attributable to the imprecision of the dipstick.
Dipsticks require careful storage because those dipsticks exposed to
air have a rapid and cumulative loss of specificity over time for
nitrite (6) and give false-positive results for glucose
after 7 days and false-negative results for blood after 28 days
(7). Protein measurement by dipsticks, however, was accurate
up to 56 days of exposure to air. To avoid these problems, we empty
trays at the end of each workday and store dipsticks in the original
containers.
Our study did not test the validity of dipstick urinalysis. Both
dipstick and microscopic tests require additional confirmatory
evidence. For example, dipstick urinalysis for hematuria after blunt
trauma was considered imprecise when microscopic hematuria was used as
the gold standard (8). Bonnardeaux et al. (9)
also claimed that abnormal dipstick urine tests require microscopic
findings. The problems with both the precision and accuracy of
microscopic urinalysis, however, have ample documentation
(10)(11)(12). Furthermore, hemoglobin from lysed red cells is
detected by dipstick but not by microscopy. Bee et al. (10)
found that blood tested by two commercial strip reagents correlated
poorly with either hemocytometer or sedimentation counts and that the
latter two methods also correlated poorly with each other.
Nevertheless, automated dipstick evidence used to aid the diagnosis of
hematuria-associated conditions or urinary tract infection is precise
but impaired if testing of the urine sample is delayed.
Our results can probably be extrapolated to other multitest reagent
strips because the precision, specificity, and limits of detection of
several products have been shown to be similar (13).
Nevertheless, additional comparative studies are warranted.
|
References
|
|---|
-
Nagel DW, Seiler DJ. Urinalysis with the new fully automated analyzer Supertron. Eur J Clin Chem Clin Biochem 1995;33:147-152.[ISI][Medline]
[Order article via Infotrieve]
-
Peele JD, Gadsden RH, Crews R. Semi-automated vs. visual reading of urinalysis dipsticks. Clin Chem 1977;23:2242-2246.[Abstract/Free Full Text]
-
Gambke B, Kouri T, Kutter D, Nagel D, Vukovich T, Wefers A. Multicentre evaluation of the urine analyser Miditron Junior. Scand J Clin Lab Invest 1997;57:605-611.[ISI][Medline]
[Order article via Infotrieve]
-
Peele JD, Jr, Gadsden RH, Crews R. Evaluation of Ames "Clini-Tek". Clin Chem 1977;23:2238-2241.[Abstract/Free Full Text]
-
Bandi ZL, Myers JL, Bee DE, James GP. Evaluation of determination of glucose in urine with some commercially available dipsticks and tablets. Clin Chem 1982;28:2110-2115.[Abstract/Free Full Text]
-
Gallagher EJ, Schwartz E, Weinstein RS. Performance characteristics of urine dipsticks stored in open containers. Am J Emerg Med 1990;8:121-123.[ISI][Medline]
[Order article via Infotrieve]
-
Cohen HT, Spiegel DM. Air-exposed urine dipsticks give false-positive results for glucose and false-negative results for blood. Am J Clin Pathol 1991;96:398-400.[ISI][Medline]
[Order article via Infotrieve]
-
Goldner AP, Mayron R, Ruiz E. Are urine dipsticks reliable indicators of hematuria in blunt trauma patients?. Ann Emerg Med 1985;14:580-582.[ISI][Medline]
[Order article via Infotrieve]
-
Bonnardeaux A, Somerville P, Kaye M. A study on the reliability of dipstick urinalysis. Clin Nephrol 1994;41:167-172.[ISI][Medline]
[Order article via Infotrieve]
-
Bee DE, James GP, Paul KL. Hemoglobinuria and hematuria: accuracy and precision of laboratory diagnosis. Clin Chem 1979;25:1696-1699.[Abstract/Free Full Text]
-
Gadeholt H. Quantitative estimation of urinary sediment
with special regard to sources of error. Br Med J 1964;I:1547–50..
-
Messing EM, Young TB, Hunt VB, Emoto SE, Wehbie JM. The significance of asymptomatic microhematuria in men 50 or more years old: findings of a home screening study using urinary dipsticks. J Urol 1987;137:919-925.[ISI][Medline]
[Order article via Infotrieve]
-
Smith BC, Peake MJ, Fraser CG. Urinalysis by use of multi-test reagent strips: two dipsticks compared. Clin Chem 1977;23:2337-2340.[Abstract/Free Full Text]
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
P. Froom, R. Etzion, and M. Barak
What Is an Abnormal Test Strip Urinary Erythrocyte Concentration?
Clin. Chem.,
March 1, 2004;
50(3):
673 - 675.
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Penders, T. Fiers, and J. R. Delanghe
Quantitative Evaluation of Urinalysis Test Strips
Clin. Chem.,
December 1, 2002;
48(12):
2236 - 2241.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
