Original Article Ann Clin Biochem 2000; 37: 355±359

Effect of urine storage on urinary uric acid concentrations

Jun-ichi Yamakita, Tetsuya Yamamoto, Yuji Moriwaki, Sumio Takahashi,
Zenta Tsutsumi and Toshikazu Hada
From the Third Department of Internal Medicine, Hyogo College of Medicine, Mukogawa-cho 1-1,
Nishinomiya, Hyogo 663-8501, Japan

SUMMARY. Accurate determination of serum and urinary uric acid concentrations

is essential for the diagnosis and classi®cation of gout according to uric acid
metabolism derangement. Urine and/or serum samples are often kept at either 4 C Ê
Ê
or 3 20 C until assayed, when a large number of samples are handled
simultaneously. Our preliminary study indicated a signi®cant decrease in urinary
uric acid concentration after preservation, regardless of the storage temperature.
Uric acid crystals were often observed in these cases which showed a marked
decrease in urinary uric acid concentration after storage. In the present study, we
sought the factor(s) that might cause this decrease in urinary uric acid concentration,
as well as measures to overcome the problem. High urinary uric acid concentration
and low pH proved to play major roles in the decrease in urinary uric acid
concentration after storage. In contrast, dilution of the urine samples before storage
resulted in no signi®cant change in urinary uric acid concentration. Based on these
results, we recommend diluting urine before storage for determination of uric acid
concentration and avoiding underestimation.

INTRODUCTION signi®cantly decreased in some preserved sam-

ples. Therefore, the present study was con-
Diagnosis of gout and its derangement of uric ducted to clarify the factor(s) causing this
acid metabolism is usually based on the
measurement of serum uric acid concentration,
uric acid clearance, and 24-h urinary uric acid
excretion. Based on these values, gout patients
are classi®ed as overproducers of uric acid, who
excrete over 1000 mg/day of urinary uric acid,1
or underexcreters of uric acid, whose fractional
clearance of uric acid (uric acid clearance/
creatinine clearance2 100) is below 4%.2 It is
standard practice to administer xanthine oxi-
dase inhibitor (allopurinol) to overproducers
and uricosuric agents to underexcreters. For this
purpose, accurate measurements of serum and
urinary uric acid concentrations are important.
However, in the process of treating large
amounts of urine and serum, the samples are
Ê
often kept at either 4 C or 3 20 C until Ê
analysis. We incidentally found a discrepancy
in urinary uric acid concentrations between
fresh and preserved samples during the analysis FIGURE 1. Correlation of uric acid concentrations
of uric acid excretion in 78 patients with gout between fresh urine samples and preserved ones (stored
(see Fig. 1); urinary uric acid concentration was Ê Ê
at 4 C and warmed by incubation at 37 C) of 78
patients with gout. A linear regression line (r=1) was
Correspondence: Dr Yuji Moriwaki. observed in most cases, but in some others the dots
E-mail: tetsuya@hyo-med.ac.jp deviated from the line.

355
356 Yamakita et al.

discrepancy and to propose solutions for the

problem.

SUBJECTS AND METHODS

Eighty-six healthy men took part in the study,
after informed consent was obtained. Spot urine
was collected in a plastic tube, and the samples
were divided into three groups. In one group,
urinary uric acid concentration was measured
immediately after voiding, while in the remain-
Ê
ing two groups, urine was kept at either 4 C or
Ê
3 20 C overnight. The stored urine samples
Ê Ê
were incubated in 37 C or 60 C containers for
10 min before measurement, and then determi-
nation of urinary uric acid concentration was
performed at room temperature by the uricase
method3 using a commercially available kit (Uric
acid B-test Wako, Wako Pure Chemical In-
dustries, Osaka, Japan). The degree of impreci-
sion (percentage coef®cient of variation; %CV)
with this kit is below 1´5% according to the
manufacturer’s instruction. Values are expressed
as the mean6 SD. Observed differences between
groups were assessed by Student’s t-test. A
stepwise regression analysis was used to assess
the association between the decrease in urinary
uric acid concentration after storage and other
measured variables. P values below 0´05 were
considered statistically signi®cant.
RESULTS
Correlation of uric acid concentrations between
fresh urine samples and preserved ones
The relationship of uric acid concentrations
from fresh urine samples and preserved ones
Ê
(stored at 4 C and warmed by incubation at
Ê
37 C) was analysed in 78 patients with gout. As
shown in Fig. 1, a linear regression line (r6 1)
was observed in most cases, but there were
discrepancies in urinary uric acid concentrations
between fresh and preserved samples in some
cases. Urinary uric acid concentration was
signi®cantly decreased in these preserved sam-
ples. A similar tendency was observed in samples
from 86 healthy subjects (see Fig. 2).
Effect of storage on the urinary uric acid
concentrations
The mean spot urinary uric acid concentrations
were 3´606 1´81 mmol/L in fresh samples,
2´946 1´23 mmol/L in ones stored at 4 C
and warmed by incubation at 37 C,
Ê
3´286 1´05 mmol/L in ones stored at 4 C and
Ê
warmed by incubation at 60 C (see Fig. 3a),
FIGURE 2. Correlation of uric acid concentrations
between fresh urine samples and preserved ones (stored
Ê Ê
at 4 C and warmed by incubation at 37 C) of 86 healthy
subjects. As with Fig. 1, a linear regression line (r=1)
was observed in most cases, but in some others the dots
deviated from the line.
3´376 1´62 mmol/L in ones stored at 3 20 C and Ê
warmed by incubation at 37 C, and Ê
3´316 1´53 mmol/L in ones stored at 3 20 C Ê
Ê
and warmed by incubation at 60 C (see Fig. 3b).
Urinary uric acid concentration was found to be
signi®cantly decreased after preservation, regard-
less of storage and incubation temperatures
(P< 0´01). Needle-like crystals were often ob-
served in cases that showed a marked decrease in
urinary uric acid concentration after storage (see
Fig. 4). Infrared absorption spectrophotometry
demonstrated that the crystals were almost
entirely (98%) composed of uric acid.
Comparison of various biochemical parameters
between groups with marked and slight decreases
in uric acid concentration after storage
In 20 cases (group A), a marked decrease in
urinary uric acid concentration (> 0´59 mmol/
L; 5´586 1´61 mmol/L in fresh urine and
2´746 1´66 mmol/L in Ê
4 C-stored urine,
P< 0´0001) was observed, while in the remain-
ing 66 cases (group B) the decrease was only
slight (< 0´59 mmol/L). To determine the causa-
tive factors involved in this difference, several
biochemical components, such as uric acid
Ê concentration, pH and electrolytes, were com-
Ê pared between the two groups, using fresh
samples. As shown in Table 1, the urinary
concentrations of uric acid, urea nitrogen,

Ann Clin Biochem 2000: 37

 Effect of storage on urinary uric acid concentrations 357

Ê
FIGURE 3. Effect of storage on urinary uric acid concentrations. (a) Samples stored at 4 C: lane 1=fresh samples;
Ê Ê Ê
lane 2=incubation at 37 C; lane 3=incubation at 60 C. (b) Samples stored at 3 20 C: lane 1=fresh samples; lane
Ê Ê
2=incubation at 37 C; lane 3=incubation at 60 C. A signi®cant decrease in urinary uric acid concentrations was
observed in the stored samples, irrespective of the storage and incubation temperatures.

protein, glucose and creatinine, as well as urine
osmolarity, were signi®cantly higher in group A
than in group B (5´586 0´35 mmol/L versus
3´086 0´15 mmol/L, P< 0´01; 0´356 0´02 mol/L
versus 0´286 0´01 mol/L, P< 0´05; 746 7 mg/L
versus 526 2 mg/L, P< 0´01; 0´536 0´09 mmol/L
versus 0´286 0´02 mmol/L, P< 0´01; 13´6
6 0´83 mmol/L versus 10´66 0´58 mmol/L,
P< 0´01; 819´26 36´3 mOsm/kg H2O versus
700´46 25´8 mOsm/kg H2O, respectively, P< 0´05).
Urine pH was signi®cantly lower in group A than in
group B (5´46 0´1 versus 5´86 0´1, P< 0´05).
Urinary concentrations of sodium, potassium,
chloride, calcium, phosphate and magnesium did
not differ between the two groups (data not shown).
To further clarify which factor(s) contributed most
strongly to the decrease in urinary uric acid
concentration after storage, a stepwise regression
analysis was performed. Consequently, urinary uric
acid concentration immediately after voiding, urine
osmolarity and pH were found to make the
strongest contributors to a decrease in urinary

TABLE 1. Biochemicalparametersbetween groups with marked (< 0´59 mmol/L; group A) and slight (< 0´59 mmol/L;
group B) decreases in urinary uric acid concentration after storage

Group A (n=20) Group B (n=66) P value

Uric acid in fresh urine (mmol/L) 5´586 0´35 3´086 0´15 < 0´01
Urine pH 5´46 0´1 5´86 0´1 < 0´05
Urea nitrogen (mol/L) 0´356 0´02 0´286 0´01 < 0´05
Protein (mg/L) 746 7 526 2 < 0´01
Glucose (mmol/L) 0´536 0´09 0´286 0´02 < 0´01
Osmolarity (mOsm/kg H2O) 819´26 36´3 700´46 25´8 < 0´05
Creatinine (mmol/L) 13´66 0´83 10´66 0´58 < 0´01

Ann Clin Biochem 2000: 37

358 Yamakita et al.
FIGURE 4. Needle-like crystals (98% uric acid) were observed in the cases which showed a marked decrease in
urinary uric acid concentration after storage.
uric acid concentrations after preservation (see
Table 2).
Effect of dilution on urinary uric acid
concentration before and after preservation
To overcome this problem, urine samples from
subjects (n=27), including those showing a
marked decrease (n=6) in urinary uric acid
concentration after storage, were diluted 10-fold
before preservation. As a result, measurements
of the urinary concentration of uric acid were
not signi®cantly different before and after
storage (see Fig. 5).
DISCUSSION
Uric acid is an end-product of purine metabolism
in humans. High serum and/or urinary concen-
trations of uric acid are known as hyperuricae-
mia and/or hyperuricosuria, which lead to gout
and nephrolithiasis. Idiopathic hyperuricaemia
TABLE 2. Stepwise regression analysis of a decrease in urinary uric acid concentration on selected independent
variables in 86 subjects
Partial regression Standard regression Partial correlation Partial
Variable coef®cient
Uric acid in fresh urine (mmol/L) 0´93543
pH 3 9´06877
Osmolarity (mOsm/kg H2 O) 3 0´0722667
Ann Clin Biochem 2000: 37
comprises overproduction hyperuricaemia and
underexcretion hyperuricaemia. Uric acid
production is usually estimated by 24-h urinary
uric acid excretion, and its normal value is
generally considered to be 400±900 mg/day on a
purine-nonrestricted diet and 300±600 mg/day
on a purine-restricted diet. Uric acid excretion is
estimated by a uric acid clearance or uric acid
clearance to creatinine clearance ratio (fractional
uric acid clearance). Therefore, accurate mea-
surement of serum and urinary concentrations of
uric acid is essential.
Uric acid exists in urine as undissociated uric
acid and monovalent urate anion forms. Uric
acid is a divalent weak acid and its solubility in
urine depends on urine pH. The more urine pH
rises, the more dissociation progresses.4 More
than 70% and approximately 90% of uric acid
exist as a monovalent urate form at pH 6´0 and
at pH 7´0, respectively.5 The effect of storage and
pH on urinary uric acid concentration was
coef®cient coef®cient F value P
0´837062 0´837062 126´405 < 0´01
3 0´179419 3 0´32221 6´25619 < 0´05
3 0´411452 3 0´593919 29´4287 < 0´01
 Effect of storage on urinary uric acid concentrations
FIGURE 5. Effect of storage on uric acid concentration
in 10-fold diluted urine samples. Lane 1=fresh samples;
Ê Ê
lane 2=samples stored at 4 C and warmed at 37 C; lane
Ê
3=samples stored at 3 20 C and warmed at 37 C. No Ê
signi®cant difference was observed in urinary uric acid
concentrations before and after storage when diluted 10-
fold, irrespective of the storage conditions. NS=not
signi®cant.
investigated using arti®cial urine,6 and the
solubility of uric acid was found to be markedly
decreased at pH < 5´0. In fact, the solubility of
uric acid in urine decreases markedly at pH
values less than 5´75, the pK value at which 50%
of the uric acid is present in the undissociated
form.7 However, to the best of our knowledge,
the effect of storage on urinary uric acid
concentration has not been previously investi-
gated. The present study showed that urinary
uric acid concentration was decreased signi®-
cantly after urine storage, but this phenomenon
was not observed in every case, probably
because of the substances inhibitory to uric acid
crystalization present in some urine samples.8
Moreover, subjects with high urinary uric acid
concentration showed a high urinary urea excre-
tion, high osmolarity and low pH as well,
probably because these factors tend to be
associated with the consumption of a diet rich
in animal ¯esh and represent the end-products of
the nucleoprotein metabolism. As expected, the
underlying cause of this decrease in urinary uric
acid concentration is considered to be a high
urinary uric acid concentration and low urine pH
before storage, since uric acid stone formation
depends on a high level of urinary uric acid
concentration and low pH.9 Further, urine
osmolarity is suggested to be related to a decrease
in uric acid concentration, although it remains
undetermined if any substance relates to the
359
decrease. Uric acid, supersaturatedin the original
urine, may precipitate as micro- or macrocrystals
after storage, probably because uric acid is hardly
redissolved once crystalization occurs.
Our data suggest that misinterpretation of
uric acid excretion and uric acid clearance is
likely to occur in gout patients, who are
frequently associated with aciduria and high
urinary uric acid concentration. Uric acid over-
production may be underestimated, while uric
acid underexcretion may be overestimated using
stored urine samples, which leads to a mis-
interpretation of overproduction and under-
excretion of uric acid. Accurate measurement
of uric acid concentration in stored urine
samples is clinically very important. To measure
urinary uric acid concentration, we recommend
®rst to dilute urine samples 10-fold with distilled
water and then store them. This procedure
enabled us to determine urinary uric acid
concentration from stored samples accurately.
Since this procedure is simple and easy, it may
become a routine in clinical laboratories deter-
mining urinary uric acid concentration with
stored samples.
REFERENCES
1 Palella TD, Fox IH. Hyperuricemia and gout. In:
Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The
Metabolic Basis of Inherited Disease, 6th edn. New
York: McGraw-Hill, 1989; 1: 965
2 Decaux G, Dumont I, Naeije N, Mols P, Melot C,
Mockel J. High uric acid and urea clearance in
cirrhosis secondary to increased ``effective vascular
volume’’. Am J Med 1982; 73: 328±34
3 Kabasakalian P, Kalliney S, Westcott A. Determi-
nation of uric acid in serum, with use of uricase and
a tribromophenol-aminoantipyrine chromogen. Clin
Chem 1973; 19: 522±4
4 Shimizu T, Nishikawa M, Matsushige H. The
solubility of uric acid and monosodium urate in
urine. Adv Exp Med Biol 1989; 253A: 215±8
5 Wilcox WR, Khalaf A, Weinberger A, Kippen I,
Klinenburg JR. Solubility of uric acid and mono-
sodium urate. Med Biol Eng 1972; 10: 522±31
6 Rodgers AL, Wandt MAE. In¯uence of ageing, pH
and various additives on crystal formation in
arti®cial urine. Scanning Microsc 1991; 5: 697±706
7 Fox IH and Kelley WN. Uric acid and gout. In:
Seldin DW and Giebisch G, eds. The Kidney.
Physiology and Pathophysiology. New York: Raven
Press, 1985: 1753
8 Gardner GL, Doremus RH. Crystal growth
inhibitors in human urine. Invest Urol 1978; 15:
478±85
9 Halabe A, Sperling O. Uric acid nephrolithiasis.
Miner Electrolyte Metab 1994; 20: 424±31
Accepted for publication 8 December 1999
Ann Clin Biochem 2000: 37