Effect of Urine pH, Storage Time, and Temperature on Stability of Catecholamines, Cortisol, and Creatinine

^a author for correspondence: fax 81-44-865-8116, e-mail miki{at}nih.go.jp

In studies of stress in humans and animals, urinary excretion of catecholamines and cortisol is widely used as a stress index (1)(2). In most studies that have measured urinary catecholamines, hydrochloric acid or some antioxidant has been added to the urine samples immediately after voiding. This is probably because classic methodology demands preservation of urine specimens to prevent catecholamine degradation (3)(4)(5). On the other hand, urine is not usually acidified when glucocorticoids are to be analyzed (2)(6). Because of the difference in urine treatment required for studies of catecholamines and glucocorticoids, it has been necessary to prepare two kinds of urine samples in parallel, i.e., acidified and unpreserved. Clearly it would be preferable if a single urine treatment could be used for analysis of both stress-related hormones. However, to our knowledge, little is known about the stability of these compounds in urine kept under different conditions. In the present study, we examined changes in catecholamine and cortisol concentrations in human urine samples stored at various pH values for different periods, using HPLC and fluorometry. The stability of urinary creatinine was also investigated, because creatinine excretion is commonly used to estimate the exact timing of urine collection.

In the present experiments, urinary free catecholamines were assayed using a modification of a method described previously (7). Briefly, 0.5 or 1 mL of urine was adjusted to pH 8.4–8.6 with 1 mol/L NaOH after addition of 100 pmol of isoproterenol as an internal standard and 1 mL of 0.1 mol/L Na₂EDTA (pH 8.6); the urine was then adsorbed to 150 mg of alumina packed in a glass column (6 mm i.d.). The alumina was washed twice with 5 mL of water and eluted with 1 mL of 0.25 mol/L acetic acid. The eluate was kept at -20 °C until HPLC analysis, unless otherwise stated. A small portion (100 µL) of the acidic extract was injected onto an HPLC column (Senshupak SCX-0201N 200 x 4 mm, Senshu Kagaku), using an autosampler, and developed with a mobile phase of 0.1 mol/L phosphate buffer (pH 3.5) containing 100 mL/L acetonitrile. Catecholamines in the eluate were derivatized by a postcolumn reaction, using the trihydroxyindole method, and detected by a spectrofluorometer with excitation and emission wavelengths of 420 and 520 nm, respectively. Absolute recovery from the alumina was 70–80%, but the amount of the amine was corrected using an internal standard. The CV of the calculated values was <6% (n = 10). The detection limit (signal-to-noise ratio = 5) was 0.05 pmol for adrenaline in the present HPLC analysis.

For analysis of free cortisol (8), 11-deoxycortisol or tetrahydrocorticosterone was added to urine as an internal standard, and the urine was extracted with dichloromethane. The organic layer was washed with 0.1 mol/L sodium hydroxide and water and then evaporated to dryness. The residue was dissolved in 100 mL/L acetonitrile and injected onto an HPLC column (Capcell-pak C8, 250 x 4.6 mm, Shiseido). The column was eluted with 300 mL/L acetonitrile, and the effluent was mixed with sulfuric acid to produce a fluorescent derivative. The fluorescence intensity was recorded continuously using a spectrofluorometer with excitation and emission wavelengths of 465 and 530 nm, respectively. Urine creatinine concentrations were measured using the Jaffe reaction, by the AutoAnalyzer method (9).

Each sample was assayed once after initial thawing, and measurements were performed in duplicate. Urine catecholamine and cortisol concentrations are expressed as nmol/L, and creatinine as g/L of original urine, unless otherwise stated. Authentic noradrenaline, adrenaline, dopamine, and cortisol were obtained from Sigma Chemical Co., and the HPLC system used was purchased from Hitachi Co. Ltd. (L-6000 series). Because the CVs in cortisol and creatinine measurements were similar to those in catecholamine analysis, we used values within 6% of the initial baseline, in principle, as the criterion for stability in the present study.

In a preliminary experiment, the stability of catecholamines in unpreserved urine during storage at -80 °C was examined. Urine samples obtained from four healthy human volunteers were divided into two portions, one of which was acidified with 6 mol/L hydrochloric acid (1.5 mL/h of urine collection). The acidified (pH 1.7–2.3) and unpreserved (pH 5.4–6.8) urine samples were poured separately into small plastic tubes. After storage in a freezer (-80 °C) for periods of up to 4 months, each urine sample was thawed under tap water, and the concentrations of catecholamines, cortisol, and creatinine were determined as described above.

During storage at -80 °C for 4 months, catecholamine concentrations were within 6% of the baseline concentrations on day 0 for both acidified and unpreserved urine samples, indicating that catecholamines in acidified and unpreserved urine are stable at -80 °C for at least 4 months.

The effect of urine pH on the stability of these hormones and creatinine was examined next. Two urine specimens were divided into six portions, and the pH was adjusted to 0.5, 1, 3, 5, 7, and 10, respectively, using hydrochloric acid or sodium hydroxide. Samples were stored in a freezer (-20 °C), in a refrigerator (~4 °C), or at room temperature (15–20°C) for up to 4 weeks. These urine samples were transferred to the -80 °C freezer after their allotted storage time so that all samples could be analyzed together at the conclusion of the experiment.

On day 0, the concentrations of catecholamines, cortisol, and creatinine were essentially the same among the different pH values, except for pH 10. Therefore, the ratio of the concentration to the average of the concentrations at pH 0.5, 1, 3, and 7 on day 0 was calculated to compare the stability of these substances among different storage conditions in two urine samples. When urine pH was within the range 0.5–7 and the storage temperature was -20 °C for 4 weeks, the concentrations of catecholamines did not change significantly (P >0.1) during storage, and the values were within 6% of the average. This was also the case for storage at pH 0.5–3 at -4 °C for 1 week. Changes in the ratios for these amines in urine samples stored under other conditions are listed in Table 1 . No changes in the concentrations of catecholamines were observed in samples at pH 7 stored at 4 °C for 3 days, but adrenaline concentrations seemed to decrease slightly after 1 day of storage at room temperature. Marked decreases in amine concentrations were recognized at pH 10, even when the specimens were stored at -20 °C. It was noteworthy that at pH 0.5 or 1, amine concentrations tended to increase during 1 week of storage at room temperature. Urinary free cortisol and creatinine concentrations did not change significantly (P >0.1) when urine was stored at 4 °C for 1 week at various pH values, except pH 10.

The results of this study concerning the stability of catecholamines in urine at pH 7 were similar to findings reported previously by others (10)(11), although the previous reports did not mention urine pH. Giles and Meggiorini (10) measured adrenaline, noradrenaline, and dopamine in unpreserved samples of pooled human urine stored at -30 °C for different times and found no significant correlation between storage time and catecholamine concentration. Boomsma et al. (11) measured the concentrations of three catecholamines in pooled, unpreserved urine stored at various temperatures and reported that the concentrations measured after 1 month at -20 °C were unchanged from their original values.

When the results of these studies are taken together, it is recommended that, for analysis of catecholamines, cortisol, and creatinine, urine samples–whether they are preserved or not–should be frozen as soon as possible. If the urine pH is between 3 and 7, storage at 4 °C for 1–2 days after urine collection might be permissible before freezing.

To test the effectiveness of these storage conditions, an additional 24 urine samples (pH 5.4–8.2) were investigated. Each sample was subdivided into two portions, one of which was acidified with hydrochloric acid (pH 1.7–3.6). After a 10-h storage period at 4 °C, both the acidified and unpreserved urine samples were transferred to the -80 °C freezer, and analysis was performed 2 months later.

The data indicated that the concentrations of catecholamines, cortisol, and creatinine in unpreserved urine samples were almost the same as those in the respective acidified urine samples. Regression lines and coefficients of correlation (r) were as follows: y = 1.00x - 0.8 (r = 0.995) for noradrenaline; y = 1.01x - 1.6 (r = 0.998) for adrenaline; y = 1.03x - 17.2 (r = 0.996) for dopamine; y = 0.87x 4.4 (r = 0.996) for cortisol; and y = 0.96x 3.1 (r = 0.999) for creatinine, where y is the concentration in unpreserved urine, and x is the concentration in acidified urine. The ratios of the concentrations in unpreserved urine samples to those in acidified samples showed no significant correlation with urine pH, as shown in Fig. 1 . It is assumed that urine pH has no substantial effect on the stability of these substances, at least within this pH (5.4–8.2) range.

View larger version (23K): [in this window] [in a new window]   Figure 1. Relation between urine pH and the ratios of catecholamines and creatinine in untreated vs acidified urine samples.

Thus, the present results confirm that unpreserved urine samples can be used for the measurement of catecholamines, cortisol, and creatinine. These sampling and storage conditions are convenient for large-scale field studies, because it is not necessary for subjects to carry out time-consuming, awkward urine treatments using toxic chemical reagents, which would otherwise be necessary immediately after voiding.

Footnotes

National Institute of Industrial Health, 21-1, Nagao 6-chome, Tama-ku, Kawasaki 214, Japan

References

1. Mason JW. Emotion as reflected in patterns of endocrine integration. Levi L eds. Emotion 1975:143-181 Raven Press New York. .
2. Fibiger W, Christensen F, Singer G, Kaufmann H. Mental and physical components of sawmill operatives' workload. Ergonomics 1986;29:363-375. [Medline] [Order article via Infotrieve]
3. von Euler US, Lishajko F. Improved technique for the fluorometric estimation of catecholamines. Acta Physiol Scand 1961;51:348-355.
4. Anton AH, Sayer DF. A study of the factors affecting the aluminum oxide-trihydroxyindole procedure for the analysis of catecholamines. J Pharmacol Exp Ther 1962;138:360-375. [Abstract/Free Full Text]
5. Shoup RE, Kissinger PT, Goldstein DS. Rapid liquid chromatographic methods for assay of norepinephrine, epinephrine, and dopamine in biological fluids and tissues. Ziegler MG Lake CR eds. Norepinephrine 1984:38-46 Williams & Wilkins Baltimore. .
6. Murphy BEP. Clinical evaluation of urinary cortisol determinations by competitive protein-binding radioassay. J Clin Endocrinol 1968;28:343-348. [ISI][Medline] [Order article via Infotrieve]
7. Sudo A. Measurement of adrenaline in the rat brain by high-performance liquid chromatography with fluorometric detection. Ind Health 1990;20:151-156.
8. Sudo A. Analysis of corticosterone in rat urine by high-performance liquid chromatography and fluorometry using postcolumn reaction with sulfuric acid. J Chromatogr 1990;528:453-458. [ISI][Medline] [Order article via Infotrieve]
9. Technicon Instrument Corporation. Technicon AutoAnalyzer methodology. Chauncy, NY: Technicon Instrument Corp., 1963:N-11a (creatinine)..
10. Giles HG, Meggiorini S. Stability of catecholamine in urine. Clin Chem 1983;29:595.[Free Full Text]
11. Boomsma F, Alberts G, van Eijk L, Veld AJM, Schalekamp MADH. Optimal collection and storage conditions for catecholamine measurements in human plasma and urine. Clin Chem 1993;39:2503-2508. [Abstract]