Urine is a heterogeneous solution containing numerous elements, some of which are labile and may change over time; for example, neutrophils and casts degrade rapidly in alkaline and hypotonic urine.1,2 Likewise, bilirubin and urobilinogen concentrations decrease with storage, especially when exposed to light.1 Stored urine also provides a favorable environment for bacterial growth. Bacterial contamination and bacterial overgrowth can occur when a sample is not handled properly or when it is left at room temperature before being analyzed.1 This could lead to decreased glucose concentrations and alkalization of urine because of the conversion of urea to ammonia.1 Also, it has been shown that the specific gravity of canine urine stored either on a diaper or in nonabsorbable litter increases over time.3 The authors of previous studies4–6 have recommended that urine samples be analyzed ≤ 30 minutes after collection or refrigerated and examined as soon as possible. Refrigerated samples should be warmed to room temperature before analysis.1,6–8 This can be achieved by leaving the sample on a counter at room temperature for ≥ 15 minutes or by holding the sample in hand.1,6,7 Another study9 concluded that urine samples should be evaluated ≤ 60 minutes after collection to minimize temperature and time-dependent effects on crystal formation in vitro. According to the Clinical and Laboratory Standards Institute guidelines,10 urine samples should be tested ≤ 2 hours after collection. Currently, if a delay is expected in analysis, it is recommended that urine samples be refrigerated to reduce cell degradation and bacterial growth.3,11 However, refrigeration has been reported to disintegrate some cells in urine.1 Also, refrigeration and increased storage time have been associated with a significant increase in crystal formation.9 Thus, it is necessary to focus on the preanalytic aspects of urinalysis and find strategies to maximize the stability of urine analytes to improve the reliability of the test results.
Studies12–14 in human medicine have shown that urinalysis tubes containing 0.4% chlorhexidine, 5.6% ethyl paraben, and 94% sodium propionate as preservatives satisfactorily preserve urine at room temperature until an analysis can be performed. Results of those studies12–14 indicated good agreement between chemical and microscopic variables of baseline urine samples and samples stored in the preservative-containing tubes for various intervals, with the exception of a few variables, particularly WBC counts at 24 hours and 72 hours and RBC counts at 72 hours after collection. To our knowledge, the use of preservative-containing tubes of this type and their effect on urinalysis results have not been evaluated in veterinary medicine. The purpose of the study reported here was to investigate and compare urinalysis results for canine urine samples stored in preservative-containing tubes at room temperature (20°C to 25°C [68°F to 77°F]) or refrigerated at 4°C (39.2°F) in plain glass tubes with results for the same samples analyzed immediately after collection (ie, reference samples). We hypothesized that urine physical, chemical, and microscopic variables for samples kept in preservative-containing tubes at room temperature or refrigerated in plain glass tubes would be highly correlated with the same variables for the reference samples.
Materials and Methods
Animals
Notices were placed throughout the Louisiana State University School of Veterinary Medicine to recruit veterinary teaching hospital staff and students to enroll their healthy dogs in the study. The selection criteria were that dogs had no clinical signs of systemic disease and no obvious clinical signs of urinary tract problems on the basis of owner observations. Informed consent was obtained from all owners prior to their dogs’ participation in the study.
A convenience sample of 20 apparently healthy dogs was used in the study. There were 9 neutered males, 2 sexually intact males, 8 spayed females, and 1 sexually intact female; the group comprised 5 Golden Retrievers, 5 mixed-breed dogs, 4 Labrador Retrievers, 2 American Staffordshire Terriers, and 1 each of the following: Great Pyrenees, Airedale Terrier, Weimaraner, and German Shepherd Dog. Median age of the dogs was 5 years (range, 1 to 15 years), and all dogs were receiving prophylaxis against heartworm infection and flea infestation; 9 dogs were receiving prophylaxis against ticks, and none were known to have tick infestations. No dogs were receiving medications expected to influence urinalysis results. The vaccination status of all dogs was current. The study was approved by the Clinical Protocol Committee and the Institutional Animal Care and Use Committee of the Louisiana State University School of Veterinary Medicine.
Urine sample collection and processing
Owners were instructed to bring their dogs to the clinic in the morning. An adequate volume of free-catch urine (> 50 mL) was collected in a clean, detergent-free containera to avoid potential interference with the biochemical analysis. Urine sample collection was performed immediately after the dog started urinating. The prepuce of male dogs was not cleaned prior to collection. The sexually intact female dog was not in estrus at the time of urine collection. Samples with macroscopic (gross) hematuria were excluded.
Each urine sample was aliquoted into 6 separate tubes (8 mL/tube). Three aliquots were placed in plain glass tubes,b and 3 aliquots were placed in preservative-containing tubes.c According to the manufacturer, the preservative-containing tubes stabilize urine for up to 72 hours at room temperature and do not interfere with chemical strip and sediment examination on various urine analyzers.15 All urine samples were submitted to the Clinical Pathology Laboratory of Louisiana State University School of Veterinary Medicine at predetermined time points and were analyzed ≤ 2 hours after submission.6,13,16,17
One of the 3 samples in plain glass tubes was submitted to the laboratory immediately after collection (0 hours); this was used as the reference sample for comparison with all other samples from the same animal. The other 2 aliquots in plain glass tubes were refrigerated at 4°C and were submitted for analysis 24 and 72 hours after collection. Refrigerated samples were warmed to room temperature (20°C to 25°C) by being left on a countertop for 15 minutes before submission to the laboratory.
The 3 samples in preservative-containing tubes were inverted 8 to 10 times to ensure proper mixing of the preservative with the urine according to the manufacturer's instructions. One of these samples was submitted to the laboratory immediately after collection (0 hours). The other 2 samples were kept at room temperature and protected from light until they were submitted for analysis 24 and 72 hours after collection.
Urinalysis
Laboratory technologists and clinical pathologists were blinded to sample identification, including the tube type and storage conditions, to minimize potential bias. To achieve blinding, urine samples in preservative-containing tubes were transferred to new plain glass tubes at the time of submission to the laboratory immediately prior to analysis. All tubes were labeled with alphanumeric codes. Samples of each type for each time point were submitted to the laboratory concurrently for analysis.
Physical variables—Urine color and clarity (turbidity) were manually noted before and after centrifugation for each urine sample. For analysis purposes, results for each variable were converted to ordinal scales as follows: color was assigned a scale from 1 to 3 (light yellow = 1, yellow = 2, and dark yellow = 3), and turbidity was assigned a scale from 1 to 4 (clear = 1, slightly hazy = 2, hazy = 3, and cloudy = 4).
Biochemical analysis—Urine pH as well as glucose, ketone, hemoglobin (indicator of the presence of blood), protein, and bilirubin concentrations were evaluated with commercially available urine dipsticks.d A dipstick was dipped in each urine sample and read with an analyzere as recommended by the manufacturer.18 If a urine sample tested positive for ketones, the presence of ketones was manually confirmed by use of a commercially available reagent tablet.f If a urine sample tested positive for bilirubin, the presence of bilirubin was confirmed by use of another commercially available reagent tablet.g For analysis purposes, results for each variable were converted to ordinal scales as follows: protein was assigned a scale from 1 to 4 (negative = 1, trace = 2, 1+ = 3, and 2+ = 4), glucose was assigned a scale from 1 to 6 (negative = 1, 100 mg/dL = 2, 250 mg/dL = 3, 500 mg/dL = 4, 1,000 mg/dL = 5, and ≥ 2,000 mg/dL = 6), ketones were assigned a scale from 1 to 6 (negative = 1, 5 mg/dL = 2, 15 mg/dL = 3, 40 mg/dL = 4, 80 mg/dL = 5, and 160 mg/dL = 6), bilirubin was assigned a scale from 1 to 5 (negative = 1, trace = 2, small amount = 3, moderate amount = 4, and large amount = 5), and hemoglobin was assigned a scale from 1 to 7 (negative = 1, nonhemolyzed trace = 2, nonhemolyzed moderate = 3, hemolyzed trace = 4, small amount = 5, moderate amount = 6, and large amount = 7).
Quality control testing of the dipstick analyzer and reagent tablets was performed once weekly. The diagnostic laboratory used a quarterly external quality assurance program provided by the Veterinary Laboratory Association. Urine specific gravity was analyzed with a veterinary refractometer.h Quality control testing of the refractometer was performed in accordance with the manufacturer's instructions and laboratory quality assurance procedures.
Microscopic analysis—A microscopic urine analysis was performed manually for each urine sample by a registered medical technologist in accordance with guidelines established by the American Society for Clinical Pathology. Urine sediment was prepared by centrifuging 7 mL of urine (standardized volume for the laboratory) in conical centrifugation tubes. Low-speed centrifugation (481 × g for 5 minutes) was used to minimize disruption of cellular elements of the urine sediment.19 The supernatant was removed by decanting, and the sediment was resuspended in the small remaining volume (approx 0.5 mL) of urine. Once suspended, a drop (approx 50 μL) was placed on a microscope slide with a coverslip. The sediment sample was evaluated with low- and high-power magnification. Initially, the entire sample was systematically scanned with the aid of the low-power (10X) objective lens to assess the quantity of sediment present and suitability of the preparation. The sediment was evaluated for the presence of WBCs, RBCs, bacteria, sperm, fat droplets, transitional cells, squamous cells, casts, and crystals. Ten fields of the wet-mount preparation were evaluated at each magnification. The numbers of these particles per view were recorded; the mean number of casts per low-power field and the mean number of remaining particle types per hpf were reported. Additional solubility testing to confirm identification of crystals was not performed. Samples were primarily analyzed by a laboratory technologist and reviewed by a clinical pathologist at the technologist's discretion according to standard practice. A cytocentrifugationi method was used according to the manufacturer's recommendations with Wright-Giemsa stainj used to highlight identification of cells, including bacteria, in the sediment at the discretion of the technologist or the clinical pathologist.17 For analysis purposes, ordinal scales were assigned to each variable as follows: WBCs and RBCs (negative = 1, rare = 2, occasional = 3, 0 to 5/hpf = 4, 5 to 10/hpf = 5, 10 to 20/hpf = 6, 20 to 50/hpf = 7, 50 to 100/hpf = 8, and > 100/hpf = 9), bacteria (negative = 1, rare = 2, moderate amount = 3, and many = 4), sperm (negative = 1, few = 2, moderate amount = 3, and many = 4), fat droplets (negative = 1, rare = 2, few = 3, several = 4, moderate amount = 5, and many = 6), transitional cells (negative = 1, rare = 2, few = 3, moderate amount = 4, and many = 5), squamous cells (negative = 1, rare = 2, few = 3, several = 4, moderate amount = 5, and many = 6), and crystals (negative = 1, rare = 2, few = 3, several = 4, moderate amount = 5, and many = 6). The scale for scoring of crystals was used regardless of type.
Statistical analysis
Statistical analysis was performed with a commercially available statistical software package.k The readings (except pH) from the urine dipstick and microscopic evaluation were assigned an ordinal scale for analysis. For each of the 2 storage conditions, associations between urinalysis results at 24 or 72 hours and results for the reference sample (time 0 sample without preservatives) for the same dog were evaluated by Spearman correlation analysis. Absolute correlation coefficients of 0.0 to 0.299, 0.3 to 0.499, 0.5 to 0.699, 0.7 to 0.899, and 0.9 to 1.0 were accepted as negligible, low, moderate, high, and very high correlation, respectively.20 Differences in distribution for 2 numeric variables (pH and specific gravity), compared with results for the reference sample, were assessed for each storage condition at each time point with the Wilcoxon signed rank test (pH) or paired t test (specific gravity). Assumptions of the t test (normality of residuals and homoscedasticity of residuals) were assessed by examining standardized residual and quantile plots. Significance was set at P < 0.05.
Results
Reference samples
Analysis of reference samples that were sent for analysis immediately after collection without preservative use (n = 20) revealed that color (light yellow [1], yellow [13], or dark yellow [6]) did not change after centrifugation. Turbidity was rated as clear (n = 8), slightly hazy (4), hazy (3), or cloudy (5) before centrifugation and clear (14), slightly hazy (2), hazy (3), or cloudy (1) after centrifugation.
In biochemical analyses, 20 of 20 samples tested negative for glucose and ketones, and 1 sample tested positive for hemoglobin (trace amount). Protein was detected in 13 of 20 samples (trace amount [n = 3], 1+ [8], or 2+ [2]), and bilirubin was detected in 11 (trace amount [1], small amount [8], or moderate amount [2]). The mean ± SD specific gravity and pH of urine specimens were 1.033 ± 0.015 and 7.1 ± 0.9, respectively.
On microscopic examination, WBCs were identified in 14 of 20 samples (rare [n = 7], 0 to 5/hpf [6], or 20 to 50/hpf [1]), RBCs were identified in 7 samples (rare [4] or 0 to 5/hpf [3]), and squamous cells were identified in 8 samples (rare [1], few [1], several [3] or moderate amount [3]). One sample contained rare transitional cells. Other findings included bacteria (n = 2; rare [1] or many [1]), sperm (2; moderate amount [1] or many [1]), fat droplets (15; few [9], several [2], moderate amount [2], or many [2]), and crystals (6; rare brushite and rare calcium phosphate [1]; rare calcium phosphate, rare brushite, and rare struvite [1]; many struvite [2]; several amorphous and many struvite [1]; or many amorphous and many struvite [1]). No casts were identified in any samples.
Correlation analysis
All correlations between reference samples and stored samples were significant except for RBC and fat droplet counts in samples refrigerated for 72 hours in plain glass tubes, fat droplet counts in samples stored at room temperature for 72 hours in preservative-containing tubes, and squamous cell counts in samples refrigerated for 24 hours in plain glass tubes (Table 1). Color and turbidity before and after centrifugation of samples refrigerated in plain glass tubes for 24 or 72 hours and of samples stored at room temperature in preservative-containing tubes for 0, 24, or 72 hours were each highly or very highly correlated with the respective findings for reference samples, except for color before and after centrifugation at 24 hours. Results for these 2 variables were moderately correlated with those for reference samples under both storage conditions.
Results of Spearman correlation tests between urinalysis results for canine urine samples stored with refrigeration (4°C) in plain glass tubes or at room temperature (20°C to 25°C) in preservative-containing tubes for up to 72 hours and reference samples from the same dogs.
| Variable | Plain glass tubes | Preservative-containing tubes | |||
|---|---|---|---|---|---|
| 24 hours | 72 hours | 0 hours | 24 hours | 72 hours | |
| Physical characteristics | |||||
| Color | 0.651* (0.293 to 0.849) | 0.723* (0.412 to 0.883) | 0.911* (0.786 to 0.965) | 0.651* (0.293 to 0.849) | 0.723* (0.412 to 0.883) |
| Color after centrifugation | 0.651* (0.293 to 0.849) | 0.723* (0.412 to 0.883) | 0.909* (0.776 to 0.965) | 0.651* (0.293 to 0.849) | 0.723* (0.412 to 0.883) |
| Turbidity | 0.960* (0.899 to 0.984) | 0.973* (0.93 to 0.989) | 1* (1 to 1) | 0.857* (0.668 to 0.942) | 0.870* (0.699 to 0.948) |
| Turbidity after centrifugation | 0.832* (0.607 to 0.933) | 0.714* (0.384 to 0.882) | 0.925* (0.811 to 0.971) | 0.769* (0.484 to 0.907) | 0.799* (0.529 to 0.922) |
| Biochemical characteristics | |||||
| Specific gravity | 0.994* (0.985 to 0.997) | 0.993* (0.982 to 0.997) | 0.982* (0.954 to 0.993) | 0.994* (0.984 to 0.998) | 0.995* (0.988 to 0.998) |
| pH | 0.923* (0.807 to 0.97) | 0.923* (0.807 to 0.970) | 0.921* (0.802 to 0.97) | 0.923* (0.812 to 0.969) | 0.921* (0.807 to 0.969) |
| Protein | 0.938* (0.843 to 0.976) | 0.829* (0.601 to 0.932) | 0.893* (0.738 to 0.958) | 0.806* (0.565 to 0.920) | 0.823* (0.598 to 0.927) |
| Glucose | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) |
| Ketones | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) |
| Bilirubin | 0.980* (0.947 to 0.992) | 0.959* (0.895 to 0.985) | 0.923* (0.807 to 0.97) | 0.910* (0.782 to 0.964) | 0.959* (0.898 to 0.984) |
| Hemoglobin | 0.781* (0.506 to 0.911) | 0.863* (0.673 to 0.946) | 0.714* (0.386 to 0.882) | 0.946* (0.866 to 0.979) | 0.946* (0.866 to 0.979) |
| Microscopic characteristics | |||||
| WBCs | 0.738* (0.427 to 0.893) | 0.820* (0.584 to 0.929) | 0.837* (0.618 to 0.936) | 0.687* (0.351 to 0.866) | 0.544* (0.134 to 0.795) |
| RBCs | 0.537* (0.110 to 0.797) | 0.323 (−0.153 to 0.678) | 0.857* (0.659 to 0.944) | 0.797* (0.548 to 0.916) | 0.474* (0.04 to 0.758) |
| Bacteria | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) |
| Sperm | 0.999* (0.997 to 1) | 0.999* (0.997 to 1) | 0.999* (0.997 to 1) | 0.999* (0.997 to 1) | 0.993* (0.982 to 0.997) |
| Fat droplets | 0.630* (0.247 to 0.843) | 0.315 (−0.162 to 0.673) | 0.651* (0.28 to 0.853) | 0.522* (0.103 to 0.783) | 0.279 (−0.002 to 0.642) |
| Transitional cells | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) | 1* (1 to 1) |
| Squamous cells | 0.399 (−0.001 to 0.715) | 0.687* (0.351 to 0.866) | 0.653* (0.297 to 0.85) | 0.658* (0.304 to 0.852) | 0.657* (0.302 to 0.852) |
| Crystals | 0.868* (0.684 to 0.948) | 0.747* (0.443 to 0.897) | 0.928* (0.82 to 0.972) | 0.992* (0.978 to 0.997) | 0.997* (0.992 to 0.999) |
Data are shown as rs (95% CI). Urine samples were obtained by free-catch collection from 20 apparently healthy dogs and aliquoted into 6 tubes (3 plain glass tubes, including the reference sample [submitted for analysis immediately after collection], and 3 preservative-containing tubes). Samples in preservative-containing tubes were transferred to plain glass tubes immediately prior to laboratory submission to enable testing in a blinded manner.
Correlation with the reference sample result was significantly (P < 0.05) different from 0.
Results for all biochemical variables were highly or very highly correlated between samples stored in plain glass tubes for 24 or 72 hours and the reference samples and between samples stored in preservative-containing tubes for 0, 24, or 72 hours and the reference samples (Table 1). Among microscopic characteristics, correlation with reference samples was very high for bacteria, sperm, transitional cells, and crystals under both storage conditions. For samples in preservative-containing tubes, correlation with reference sample results ranged from high (0 hours) to moderate (24 and 72 hours) for WBCs, from high (0 and 24 hours) to low (72 hours) for RBCs, and from moderate (0 and 24 hours) to no significant correlation (72 hours) for fat droplets; for squamous cells, this correlation was moderate at all time points. For samples in plain glass tubes, correlation with reference sample results was high for WBCs at 24 and 72 hours, but was moderate at 24 hours and nonsignificant at 72 hours for both RBCs and fat droplets. For squamous cells, this correlation was nonsignificant at 24 hours and moderate at 72 hours.
Additional analyses
For urine specific gravity, no significant differences were found in mean results for samples refrigerated in plain glass tubes at any time points, compared with results for reference samples (mean ± SD, 1.033 ± 0.0149). Significant differences in results for this variable were found for samples stored in preservative-containing tubes for 0 (1.034 ± 0.0148; P < 0.001), 24 (1.033 ± 0.0148; P = 0.005), and 72 (1.034 ± 0.0149; P = 0.001) hours, compared with reference sample results. There were no significant differences observed for urine pH in preservative-containing tubes at 24 or 72 hours or in plain glass tubes at 24 or 72 hours, compared with reference sample pH (median, 7; range, 6 to 8.5).
Discussion
Other studies21–23 have reported the effects of storage in sample tubes containing boric acid preservative on bacterial culture results for canine urine. However, to the authors’ knowledge, the present study was the first in which the effects of using sample tubes containing a combination of chlorhexidine, ethyl paraben, and sodium propionate for storage of canine urine samples for up to 72 hours have been reported. In the present study, we included evaluation of a 0-hour sample immediately after collection and mixing in the preservative-containing tube that was not included in previous human medical studies.12–14 Our aim for including this sample was to determine whether there was a direct effect of the preservative, independent of storage duration, on any of the urinalysis variables evaluated. Our data revealed that there was a high or very high degree of positive correlation between results for 0-hour samples placed in preservative-containing tubes and reference tube results. In addition, we found a high degree of positive correlation between reference tube sample results and results for samples stored by either method in regard to most urine physical, chemical, and microscopic variables, leading us to partially accept our hypothesis.
Urine turbidity is a physical variable that is usually gauged subjectively. Excess turbidity results from the presence of suspended particles in the urine. In our study, there was high or very high correlation with reference sample results for turbidity of urine samples stored by either method for up to 72 hours, indicating that preservative-containing tubes at room temperature and plain glass tubes that are refrigerated can provide a similar degree of stability in urine turbidity for this interval.
Our study showed that there was high or very high correlation of results for all biochemical variables of samples maintained in preservative-containing and plain glass tubes at all evaluated time points, compared with reference sample results. In human medicine, it has been shown that storage of urine for extended times results in an increase in pH.24 Lysis of particles (RBCs, WBCs, bacteria, squamous epithelial cells, and hyaline casts) may also occur in samples with higher pH values or low specific gravity.24 Data from a human study12 that evaluated the same type of preservative-containing tubes revealed that there was a significant change in urine pH after storage for 72 hours, but no significant change in specific gravity was observed over the same interval. For dogs and cats, effects of storage temperature and time on pH and specific gravity of urine samples have previously been evaluated.9 In that study,9 urine samples were analyzed ≤ 60 minutes after collection or after storage at room or refrigeration temperatures for 6 or 24 hours; the authors concluded that storage time and temperature did not have a significant effect on urine pH or specific gravity. We found very high correlation for urine pH and no significant differences in pH values with storage for up to 72 hours under both conditions when compared with reference samples. However, these comparisons should be interpreted with caution, as a pH meter was used in the previously described study9 and we used a dipstick to measure urine pH. The dipstick method has been shown to overestimate urine pH in dogs and underestimate urine pH in cats.25,26 Results of a recent study26 showed a significant difference between the mean pH values measured by dipstick (7.11 ± 0.12) and those measured by a pH meter (6.83 ± 0.12) in urine samples from dogs. Notably, pH meters provide excellent results, but they are not commonly used for routine urinalysis because they are expensive, they require regular calibration with test solutions, and additional user training is necessary.6,26 With the pH meter method, if freshly voided samples are not tested, samples must be collected in containers layered with paraffin oil, which can shorten the life of the electrode.27 On the other hand, the dipstick method is widely used by human and veterinary laboratories as it is fast, is cheaper to use, and requires much less user training; the test is technically less demanding and less time-consuming.6,26,27 In addition, dipsticks are single-use test strips that can measure a range of variables in addition to pH.
The small differences noted in urine specific gravity for samples in preservative-containing tubes, although statistically significant, did not appear to be clinically relevant for evaluation of apparently healthy dogs. However, this interpretation could vary if applied to dogs with clinical abnormalities such as renal disease.
A recent study28 was performed to evaluate the effects of common storage temperatures and container types on urine protein-to-creatinine ratios for 10 dogs with proteinuria. The storage conditions included refrigeration of urine samples for 12 hours; urinary protein concentration and protein-to-creatinine ratio were not affected. Data from our study provided additional support for this conclusion, with urinary protein concentrations under both storage conditions for up to 72 hours found to be highly correlated with results for reference samples. In our study, there were 2 dogs with 2+ proteinuria identified on analysis of the reference samples; one of these dogs had a urine specific gravity of 1.046, and the other dog had a urine specific gravity of 1.043. A repeated urinalysis or assessment of the urine protein-to-creatinine ratio was recommended for these dogs but was not pursued at the time. It was possible that these 2 dogs were proteinuric and thus would not have been qualified as healthy. However, neither dog had clinical signs of disease, and 1 dog had a follow-up urinalysis performed as part of participation in a blood donor program, at which time the results revealed 1+ proteinuria with urine specific gravity of 1.053. The owner of the other dog reported no systemic or urinary system–related clinical signs during a 2-year follow-up communication.
Bilirubin concentrations in urine can decrease over time if urine samples are left at room temperature.1 Our data suggested that correlation for urine bilirubin with the reference sample measurement remained very high for urine in preservative-containing tubes, just as it did for samples refrigerated in plain glass tubes for up to 72 hours. No samples tested positive for glucose or ketones in our study, and thus no conclusions could be drawn regarding the effects of the described storage conditions and times on these variables. For the same reason, the reagent tablet test for confirmation of ketones was not used in our study. However, we did conclude that the use of preservative-containing tubes did not lead to any false-positive results for urinary glucose and ketones. Future studies are needed to determine the effect of various storage conditions and times on the stability of these variables.
Urine sediments in our study were evaluated by microscopy, which is the gold standard for determining the presence of cells, crystals, and debris.29 Correlation with reference sample results was very high for bacteria, sperm, transitional cells, and crystals in samples stored under both conditions at all time points. Correlation of RBC concentrations with reference sample results was subjectively higher for samples in preservative-containing tubes than for those in plain glass tubes at 24 hours (high vs moderate) but was low for the former and nonsignificant for the latter at 72 hours, whereas correlation for WBC concentrations was subjectively lower (moderate) for samples in preservative-containing tubes at 24 and 72 hours than for those in plain glass tubes (high). Our findings related to WBCs and RBCs in urine samples stored in preservative-containing tubes were similar to those previously reported for samples from human patients.12 Changes in cell morphology could affect the accurate detection and identification of these cells. Overall, refrigerated plain glass tubes appeared to be better for keeping WBCs stable, whereas preservative-containing tubes seemed to be better for keeping RBCs and squamous cells stable in the present study.
Bacterial contamination is a common occurrence when a urine sample is left at room temperature before being analyzed.1 In research involving experimentally inoculated canine urine samples, it was shown that samples stored at room temperature for 24 hours before plate inoculation were diagnostically unreliable, with false-negative results for 1 of 25 (4%) samples and false-positive results for 13 (52%) samples.30 In the previously described study12 to evaluate the effects of storage in preservative-containing tubes on stability of human urine, bacteria were detected in 1 of 48 (2%) aliquots at 24 hours and 2 of 48 (4%) aliquots at 72 hours of storage at room temperature. In one of the aliquots stored for 72 hours, an increased pH, likely related to bacterial growth, was noted.12 In our study, 2 samples had bacteria detected in the reference sample, and results for bacteria in all samples stored under both study conditions were highly correlated with the reference sample results. There were no instances in which the reference sample was deemed negative for bacteria but the corresponding preservative-containing or plain glass tubes were identified as having bacteria at any time point with storage. The prevalence of subclinical bacteriuria in our study (2/20 [10%]) was similar to the reported prevalence of 8.9% in a recent study of healthy dogs.31
Ideally, urine samples collected in any container should be analyzed within a very short period of time for the presence of crystals. Storage of urine leads to formation of various types of crystals in vitro,9 dependent on time and temperature. Increased storage time (24 hours, compared with 6 hours) and decreased storage temperature (refrigeration, compared with room temperature) are shown to be significantly associated with formation of crystals in urine samples collected from dogs and cats.9,l For urinary crystals in the present study, correlation with reference sample results was subjectively lower for samples refrigerated in plain glass tubes than for those stored in preservative-containing tubes at 24 hours, and this value further decreased for samples in plain glass tubes at 72 hours. This was attributable to formation of crystals in 5 of 20 (25%) plain glass tubes (data not shown), consistent with previous studies.9,l Crystals that formed in vitro included rare struvite, magnesium ammonium phosphate, calcium oxalate dihydrate, calcium phosphate, and brushite.
In veterinary medicine, it is important to keep the cost of materials and diagnostic tests in mind. At the time our study was conducted, the cost of the preservative-containing tubes was $0.42 each, which was similar to the cost of the plain glass tubes ($0.50 each). The sizes of plain glass tubes needed may vary among laboratories, as the standardized urine volumes used for urinalysis can differ. For example, a clinical pathology laboratory at another university uses a standardized volume of 5 mL of urine for urinalysis, whereas our laboratory uses a standardized urine volume of 7 mL.32 However, laboratories frequently receive urine samples with smaller volumes than their preferred standardized volumes.32 For the preservative-containing tubes used in our study, the recommended range for urine volume collection is 7 to 8 mL to create a proper preservative-to-urine ratio. The amount of required urine volume for these tubes could potentially limit its use for samples from cats and small dogs. Currently, the authors are unaware of what effect on urinalysis results, if any, would be expected when urine volumes < 7 mL are collected and stored in the preservative-containing tubes. Future studies are needed to determine this.
There were several limitations to the study reported here. Evaluation of urinalysis variables at additional time points (eg, after 12 and 48 hours of storage) may have been beneficial for informational purposes and for comparison with future studies. We chose a 24-hour time point to mimic the time required for first-class postal delivery (sample parceled, left in a vehicle overnight, and examined after 24 hours) and a 72-hour time point to mimic a situation in which samples are shipped on a Friday and are not received and analyzed until the following Monday. Our study included a small number of dogs, and the number of samples that tested positive for certain variables (particularly sperm, bacteria, transitional cells, and hemoglobin) was very small; thus, the information obtained should be interpreted with caution. Another limitation was that all urine samples obtained were from apparently healthy dogs, which does not fully mimic the clinical situation in which urinalysis is relevant. However, it is important to determine whether products such as those used in the study influence results for urine from healthy animals before they are used with samples from patients undergoing evaluation for suspected disease processes. Overall, the study results suggested that the preservative-containing tubes used in our study may represent a viable option for storage of canine urine samples at room temperature prior to testing. When urinalysis is expected to be delayed, use of these tubes to store samples for 24 to 72 hours at room temperature can provide results for most urine variables that correlate highly with results for fresh urine samples. Future studies are needed to evaluate the effects of storage, including analysis of agreement between results for freshly collected samples and samples stored in preservative-containing tubes, on urinalysis results for dogs with suspected pathological conditions that affect the urinary system.
Acknowledgments
Supported by a Veterinary Clinical Sciences Competitive Research grant. Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing and publication of the manuscript.
The authors declare that there were no conflicts of interest.
Footnotes
8-ounce specimen cups, Leica Biosystems Inc, Buffalo Grove, Ill.
Vacutainer, Becton, Dickinson and Co, Franklin Lakes, NJ.
Vacutainer Plus Urinalysis Preservative Tube, Becton, Dickinson and Co, Franklin Lakes, NJ.
Multistix 10 SG Reagent Strips, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
Clinitek Status (+) analyzer, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
Acetest Reagent Tablets, Bayer Healthcare LLC, Whippany, NY.
Healthineers Ictotest Reagent Tablets, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
AO Scientific Instruments Division of Warmer-Lambert Technologies Inc, Keene, NH.
Shandon Cytospin, Thermo Fisher Scientific, Waltham, Mass.
Sigma-Aldrich, St Louis, Mo.
SPSS Statistics for Windows, version 24.0, IBM Corp, Armonk, NY.
Duderstadt JM, Weingand KW. Effects of overnight refrigeration on routine dog urinalysis (abstr), in Proceedings. 25th Annu Meet Am Soc Vet Clin Pathol 1990;20.
References
1. Brobst D. Urinalysis and associated laboratory procedures. Vet Clin North Am Small Anim Pract 1989;19:929–949.
2. Wiwanitkit V. Effect of storage time on dog urine. J Small Anim Pract 2010;51:53
3. Steinberg E, Drobatz K, Aronson L. The effect of substrate composition and storage time on urine specific gravity in dogs. J Small Anim Pract 2009;50:536–539
4. Fairburn AJ, Canfield PJ, Paris R. Simple urinalysis for the practitioner. Aust Vet Pract 1983;13:46–53
5. Osborne CA, Low DC, Finco DR. Canine and feline urology. Philadelphia: WB Saunders, 1972;39–61.
6. Osborne CA. Handbook of canine and feline urinalysis. St Louis: Ralston Purina Co, 1981;13–18.
7. VetScan UA10 urine test strips [package insert]. Union City, Calif: Abaxis, 2017.
8. Gunn-Christie RG, Flatland B, Friedrichs KR, et al. ASVCP quality assurance guidelines: control of preanalytical, analytical, and postanalytical factors for urinalysis, cytology, and clinical chemistry in veterinary laboratories. Vet Clin Pathol 2012;41:18–26.
9. Albasan H, Lulich JP, Osborne CA, et al. Effects of storage time and temperature on pH, specific gravity, and crystal formation in urine samples from dogs and cats. J Am Vet Med Assoc 2003;222:176–179.
10. Clinical and Laboratory Standards Institute. GP16-A3. Urinalysis; approved guideline—third edition. Wayne, Pa: Clinical and Laboratory Standards Institute, 2009;4–10.
11. Kierkegaard H, Feldt-Rasmussen U, H⊘rder M, et al. Falsely negative urinary leucocyte counts due to delayed examination. Scand J Clin Lab Invest 1980;40:259–261.
12. Ercan M, Akbulut ED, Abuşoglu S, et al. Stability of urine specimens stored with and without preservatives at room temperature and on ice prior to urinalysis. Clin Biochem 2015;48:919–922.
13. Ekşioglu MK, Madenci ÖÇ, Yücel N, et al. The effectiveness of BD Vacutainer Plus Urinalysis Preservative Tubes in preservation of urine for chemical strip analysis and particle counting. Biochem Med (Zagreb) 2016;26:224–232.
14. Avci E, Aybek H, Kangal Z, et al. The role of tubes with preservative in urinalysis of pregnant women. Med Sci (Turkey) 2018;7:610–612.
15. BD Vacutainer urinalysis preservative tube [package insert]. Franklin Lakes, NJ: Becton, Dickinson and Co, 2011.
16. Swenson CL, Boisvert AM, Kruger JM, et al. Evaluation of modified Wright-staining of urine sediment as a method for accurate detection of bacteriuria in dogs. J Am Vet Med Assoc 2004;224:1282–1289.
17. O'Neil E, Horney B, Burton S, et al. Comparison of wet-mount, Wright-Giemsa and Gram-stained urine sediment for predicting bacteriuria in dogs and cats. Can Vet J 2013;54:1061–1066.
18. Siemens. CLINITEK Status+ quick reference guide. Rev B. Camberly, England: Siemens AG, 2009
19. Reine NJ, Langston CE. Urinalysis interpretation: how to squeeze out the maximum information from a small sample. Clin Tech Small Anim Pract 2005;20:2–10.
20. Mukaka MM. Statistics corner: a guide to appropriate use of correlation coefficient in medical research. Malawi Med J 2012;24:69–71.
21. Allen TA, Jones RL, Purvance J. Microbiologic evaluation of canine urine: direct microscopic examination and preservation of specimen quality for culture. J Am Vet Med Assoc 1987;190:1289–1291.
22. Perrin J, Nicolet J. Influence of the transport on the outcome of the bacteriological analysis of dog urine comparison of three transport tubes. Zentralbl Veterinarmed B 1992;39:662–667.
23. Rowlands M, Blackwood L, Mas A, et al. The effect of boric acid on bacterial culture of canine and feline urine. J Small Anim Pract 2011;52:510–514.
24. Kouri T, Malminiemi O, Penders J, et al. Limits of preservation of samples for urine strip tests and particle counting. Clin Chem Lab Med 2008;46:703–713.
25. Heuter KJ, Buffington CA, Chew DJ. Agreement between two methods for measuring urine pH in cats and dogs. J Am Vet Med Assoc 1998;213:996–998.
26. Athanasiou LV, Katsoulos PD, Katsogiannou EG, et al. Comparison between the urine dipstick and the pH-meter to assess urine pH in sheep and dogs. Vet Clin Pathol 2018;47:284–288.
27. Kwong T, Robinson C, Spencer D, et al. Accuracy of urine pH testing in a regional metabolic renal clinic: is the dipstick accurate enough? Urolithiasis 2013;41:129–132.
28. Moyle PS, Specht A, Hill R. Effect of common storage temperatures and container types on urine protein : creatinine ratios in urine samples of proteinuric dogs. J Vet Intern Med 2018;32:1652–1658.
29. Langlois MR, Delanghe JR, Steyaert SR, et al. Automated flow cytometry compared with an automated dipstick reader for urinalysis. Clin Chem 1999;45:118–122.
30. Padilla J, Osborne CA, Ward GE. Effects of storage time and temperature on quantitative culture of canine urine. J Am Vet Med Assoc 1981;178:1077–1081.
31. Wan SY, Hartmann FA, Jooss MK, et al. Prevalence and clinical outcome of subclinical bacteriuria in female dogs. J Am Vet Med Assoc 2014;245:106–112.
32. Cornell University College of Veterinary Medicine eClinpath. The sample collection page. Available at: www.eclinpath.com/urinalysis/sample-collection/. Accessed Feb 5, 2020.
