The stability of urine specimens submitted for culture remains a challenge for many laboratories because of delays in specimen transport. We evaluated the usefulness of BD Vacutainer Plus Urine C&S Preservative Tube in ensuring specimen stability.
Clinical urine specimens collected in sterile collection cups (n = 110) were plated onto sheep blood and MacConkey agar following standard laboratory procedures guidelines. Thereafter, specimens were divided into 3 storage conditions: nonpreservative, refrigerated; nonpreservative, room temperature (RT); BD Vacutainer Plus Urine C&S Preservative Tube, RT. For each sample type, additional cultures were set up at 2, 4, 24, and 48 hours.
Initially, 18 specimens had no growth, 32 showed mixed skin flora, and 60 yielded at least 1 uropathogen. Increased colony counts of uropathogens were observed for nonpreserved urine samples stored at RT; these changes were statistically significant. Minor differences between refrigerated urine samples and BD Vacutainer Plus Urine C&S Preservative Tube samples were seen but were not statistically significant.
The use of preservative-containing collection tubes is desirable to ensure specimen stability when prompt processing or refrigeration is not feasible.
Urinary tract infections (UTIs) are among the most commonly encountered infectious diseases, and urine specimens submitted for quantitative bacterial culture account for a significantly large volume of test requests in clinical microbiology laboratories.1–4 The gold standard for diagnosis of a UTI is the detection of a urinary pathogen via culture in freshly collected urine specimens.3,4 Typically, patients are asked to provide a “clean-catch,” midstream urine specimen. Other specimen types may include those obtained from a catheter (single, straight catheter, or Foley catheter) or from a suprapubic aspiration of the bladder. Considering the human anatomy of the urinary tract, it is not surprising that UTIs are among the most common bacterial infections.1–3 However, voided urine samples are frequently contaminated by organisms of the urethral, skin, genital, and/or fecal flora. The importance of appropriate specimen collection and transport to avoid contamination has been previously described.4–7 Usually, contaminant organisms are present in low numbers, ie, less than 104 colony-forming units (CFU)/mL, whereas uropathogens are present in significantly higher numbers, usually greater than 105 CFU/mL. However, over time, even contaminant microorganisms can grow to significantly high numbers when specimens are left at room temperature (>15°C). The effects of delayed urine culture as well as impact of various transport and storage conditions have been described.8–11
The current guidelines for urine collection, transport, and culture emphasize the need for using either transport tubes containing preservatives or the need for not exceeding the 2-hour interval from collection to processing.4 If preservative tubes are not used and a transport time of less than 2 hours may not be achievable, refrigeration of the urine specimen has been shown to also limit the overgrowth of organisms; however, it is unrealistic to expect that no urine specimen will spend more than 2 cumulative hours unrefrigerated in most settings.9 In the 2005 College of American Pathologists (CAP) Q-Probes study on urine culture contamination, the investigators found that only a small number of microbiology laboratories enforce the 2-hour cutoff rule for limiting transport time of urine specimens.12 Because most of the evidence on specimen integrity for urine is based on studies performed in the 1980s, and considering that in the CAP Q-Probes study most laboratory sites used refrigeration of urine specimen for preservation, we decided to investigate the usefulness of the BD Vacutainer Plus Urine C&S Preservative Tube (BDU; Becton Dickinson, Franklin Lakes, NJ) and compare its performance to refrigerated nonpreservative and room-temperature–exposed nonpreservative urine specimens. This study further aimed at assessing stability of both the specimen and CFU counts over time.
Materials and Methods
This research was approved by the institutional review board of the Johns Hopkins Medical Institutions (Baltimore, MD). Between December 2010 and August 2011, patient urine samples from 2 medical units and the emergency department at our tertiary care medical center were screened upon receipt in the microbiology laboratory using a manual dipstick method for urinalysis (UA) (Multistix-10-SG, Siemens Healthcare Diagnostics, Tarrytown, NY) for potential enrollment in this study. All urine specimens were collected and initially received in sterile collection cups, and specimens with a volume in excess of 8 to 10 mL were considered as potential candidate specimens. Demographic patient data and the time from specimen collection to receipt in the laboratory were recorded. Based on the UA results, specimens likely to yield positive cultures as well as negative urine samples were enrolled.
A total of 110 urine specimens were enrolled in this study. These specimens were collected using the sterile collection cup from the BD Vacutainer “Urine Complete Cup Kit” collection system (Becton Dickinson). Cultures were performed on each of these initial urine specimens using a 0.01-mL inoculation loop, using streak plate method on a trypticase soy (TSA II) with 5% sheep blood agar (SBA) and a MacConkey agar (Becton Dickinson, Sparks, MD). This culture was considered the T0 (reference) culture for all 3 subsequent specimen storage conditions. After setting up the T0 culture, urine was drawn from the BD collection cup into 1 preservative-free BD Vacutainer Urinalysis Plus Conical Urine Tube, No Additive, and 1 BDU. The remainder of the urine sample was left in the original urine collection cup and was stored at 4°C (refrigerated nonpreservative urine [R-NPU]). The other 2 specimen tubes, BDU and the BD Vacutainer Urinalysis Plus Conical Urine Tube, No Additive (room temperature nonpreservative urine [RT-NPU]) were stored at room temperature in the laboratory for the duration of the study. In addition to the initial culture (T0), cultures were performed at 2 (T1), 4 (T2), 24 (T3), and 48 (T4) hours. Furthermore, at each time point T1 to T4, urine from the BDU as well as the nonpreservative specimens (R-NPU and RT-NPU) was serially diluted at 1:100 and 1:10,000 using 5 mL of 0.9% sterile saline, and then cultures were performed on SBA using a 0.01-mL inoculation loop. All urine specimens were returned to their respective storage conditions between the time points T1 to T4 for culture. All cultures were performed according to standard laboratory procedures and guidelines.4,13 Inoculated agar plates were incubated for 20 to 24 hours at 35°C before they were assessed for growth and CFU counts determined. In plates with positive growth, the organism was identified using standard and approved methods in our clinical laboratory. Bacterial growth on agar plates was classified according to commonly accepted categorical assessment for urine cultures: (1) negative, (2) mixed skin flora (MSF) only; and (3) uropathogen, regardless of additional presence of MSF. The CFU counts were assessed and classified as follows: 0 CFU/mL, negative; greater than 0 to less than 104 CFU/mL, insignificant growth; 104 to 105 CFU/mL, positive for UTI if uropathogen present; and greater than 105 CFU/mL, positive for UTI. Statistical analysis using descriptive statistical methods was performed using STATA 11 software (StataCorp, College Station, TX).
A total of 110 urine specimens were included in this study. Seventy-four urine samples were obtained from female patients (67%) and 36 urine samples were from male patients (33%). Patients ranged in age from 17 to 95 years, with an average age of 57 years. The average time between the collection of the urine specimen and its arrival in the laboratory was 25 minutes (range, 5–70 minutes). At T0, 18 specimens had no growth of any organisms, 32 exhibited growth of only MSF, and 60 grew at least 1 potentially pathogenic organism alone or in combination with MSF or another pathogen. The following gram-negative organisms (total number of cultures with this organism) were recovered over the 48-hour study period: Escherichia coli (n = 32), Klebsiella pneumoniae (n = 15), Proteus mirabilis (n = 7), Enterobacter cloacae (n = 6), Pseudomonas aeruginosa (n = 2), Citrobacter freundii (n = 1), Klebsiella oxytoca (n = 1), Shigella species (n = 1), Raoultella ornithinolytica (n = 1), Gardnerella vaginalis (n = 1). In addition, 8 cultures were positive for yeast, including Candida albicans (n = 2), Candida parapsilosis (n = 2), Candida tropicalis (n = 2), and Candida species, not further specified (n = 2). Thirteen cultures were positive for Enterococcus species, and 10 cultures were positive for various gram-positive cocci, including Staphylococcus saprophyticus, other coagulase-negative staphylococci, group B streptococci, and viridans group streptococci. All organisms were identified using standard identification methods approved by the Clinical and Laboratory Standards Institute and the Food and Drug Administration, which are commonly used in clinical microbiology laboratories. The culture results by categorical analysis for all specimens over all time points are summarized in Table 1. Differences in the number of specimens with more than 105 CFU/mL pathogens (range, 33–36) between the refrigerated urine specimens and the BDU samples and between the time points T0 and T3 (24 h) were not statistically significant. Similarly, no statistically significant difference was seen in the number of cultures with more than or equal to 104 pathogens but less than 105 CFU/mL between T0 and T3 for the BDU and R-NPU samples. By contrast, the numbers of cultures having more than 105 CFU/mL pathogens in the nonpreservative urine specimen that was stored at room temperature (RT-NPU) changed significantly by 24 hours after collection with the number of specimens having more than 105 CFU/mL pathogens increasing from 35 to 70. During the same period (T0–T3), the number of specimens with 104 CFU/mL pathogens but less than 105 CFU/mL decreased from 10 to 4 for the RT-NPU storage condition. Extending the window of observation to a 48-hour period, the number of samples with pathogens at T0 (n = 60) increased to 81 in the RT-NPU group, 61 R-NPU samples, and 63 BDU samples. Although the changes for the R-NPU and BDU samples were not statistically significant, the changes for the RT-NPU samples were significant (P < .001) (Table 1).
Furthermore, we analyzed urine specimens that were negative for growth (n = 18) and/or showed MSF only (n = 32) at T0. These results are shown in Table 2 and Table 3. Again, a significant number of samples in the RT-NPU group changed from no-growth and/or MSF at T0 to presence of pathogens in various colony counts. However, changes in the R-NPU and BDU groups were not statistically significant.
Results for all cultures based on storage condition were further analyzed with regard to whether the culture result would be considered clinically significant, ie, leading to a complete organism identification and antimicrobial susceptibility testing Table 4. Culture results showing the presence of more than or equal to 104 CFU/mL of any potential uropathogen were considered clinically significant. Considering the results of all urine cultures and storage conditions at T0, clinically significant results were present in 45 specimens, whereas 65 specimens had no growth or had clinically insignificant growth or presence of MSF that would not have resulted in a complete urine culture workup. In this analysis, specimens stored at room temperature showed a significant change of culture results toward the presence of pathogens at any later point of culture (T1 – T4). These changes were statistically significant (P < .001). Minor trends observed for the other 2 specimen types (R-NPU and BDU) were not statistically significant.
Finally, the stability of culture results assessed as CFU (log10)/mL based on type of microorganism and type of storage condition over time for all urine specimens with significant numbers of a (potential) uropathogen at T0 were compared, as shown in Table 5. The mean value for the amount of organisms present in the BDU urine samples remained fairly constant over a 48-hour period, and the minor variation in CFU (log10)/mL were not statistically significant. Similarly, no statistically significant change was observed in the refrigerated (R-NPU) urine samples. In contrast, the amount of organisms, expressed as CFU (log10)/mL, identified in the RT-NPU samples increased significantly between T0 and T4 time points, and these changes were highly statistically significant (P < .001) for the gram-negative organisms (including E coli and K pneumoniae); less significant changes were observed for gram-positive organisms and yeast.
Urine has long been recognized as an excellent culture medium. Several studies during the 1970s and early 1980s investigated the effects of delayed transport and culture on semiquantitative urine culture results.7,9,10,14 In their report on the effects of delays in urine culture, Hindman et al8 described that delays greater than 2 hours are more likely to result in significant changes in colony counts of bacteria and therefore could lead to errors in diagnosis. Other studies investigated the clinical usefulness of urine collection tubes containing preservative media (eg, boric acid) for avoiding significant changes in bacterial growth because of delayed bacterial culture.7,9,10,14 The BDU with buffered boric acid as a preservative offers several advantages to clinical microbiology laboratories with regard to collection, transportation, processing, and storage of urine specimens.15 In general, the kit is easy to use, and tubes can be stored at room temperature with minimal requirements for space. According to the product information, the BDUs “are stable when stored between 4–25°C.” The package insert further specifies that the lyophilized urine maintenance formula can maintain the bacterial population in the urine specimen for a period of up to 48 hours at room temperature at levels comparable to urine specimens stored without additive under refrigeration for the same period.
Previous studies suggested that in some cases boric acid used in urine preservative tubes could lead to decreased colony counts below the threshold of 105 CFU/mL when samples were processed for culture more than 4 hours after specimen collection.16–18 Many of these previous studies were performed more than 20 years ago, and over the past decades, the approach to UTI diagnostics as well as the workflow in clinics, hospitals, and laboratories may have changed. We therefore conducted this study to re-evaluate the usefulness and performance of urine collection tubes containing boric acid as a preservative against urine samples stored at 4°C and/or room temperature. Within the design of our study, we took the results, experience, and unanswered questions from the previously conducted studies into account. We did not observe any significant contamination of urine specimens throughout the study process.
The results of our study support the findings of these previous studies that urine specimens are fairly stable at room temperature for up to 2 hours after collection; our results furthermore confirm that a urine preservative such as the buffered boric acid in the BDU has a beneficial effect on urine samples submitted for bacteriologic culture and examination. At the time of the initial reference culture (T0), 41% (n = 45) of all urine samples included in this study were positive (>104 CFU/mL) for a uropathogen. After storage of the BDU preservative urine samples for 48 hours (T4) at room temperature, cultures yielded a 44% positivity rate; this change in positivity rate was not statistically significant. During the same period, unpreserved but otherwise identical urine samples stored at room temperature changed from the initial 41% to 71%, resulting in an increase in “false positives,” and urine samples stored at 4°C decreased from 41% to 40%. There was no statistically significant difference in the results between T0 and T4 for refrigerated (R-NPU) or preserved (BDU) urine samples with regard to negative cultures or cultures with the presence of MSF. Lauer et al9 evaluated the effects of 2 different urine preservative fluids, including boric acid, over a 24-hour period; however, these investigators were left with 2 unanswered questions: (1) Does boric acid have a toxic effect for particular bacterial species, and (2) What is the effect of the preservative beyond the first 24-hour interval after specimen collection? Studies have suggested that boric acid may have a toxic effect on microorganisms present in urine samples stored for prolonged intervals (eg, ≥24 hours), resulting in an increase in negative culture results. However, we did not identify any significant increase in the number of negative cultures after storage for 24 or 48 hours for any type of urine sample (16% at T0, 17% for BDU at T4, and 18% for R-NPU at T4).9,10 Although Lum and Meers10 used a concentration of 20 g/L of boric acid for their experiments (0.5 g boric acid in tubes filled with 25 mL of urine) and Lauer et al9 used a mixture of boric acid, glycerol, and sodium formate in a glass tube filled with 4 to 6 mL of urine, the BDUs used in our study are plastic tubes with boric acid, sodium formate, and sodium borate, filled with 4 mL of urine. Considering the results of these studies, it appears plausible that pure boric acid is likely to have an inhibitory effect on the growth of bacterial micro-organisms, while preservative formulations using some form of buffered boric acid may not exert this inhibitory effect on bacterial growth.
Hubbard et al17 reported an overall 10% fewer positivity rate with BDU specimens compared with refrigerated urine specimens. Following a suggestion by Guenther and Washington16 that specimens yielding more than 104 CFU/mL should be considered equivalent to more than 105 CFU/mL when BDUs were used, Hubbard et al17 suggested that this approach could eliminate the problems associated with false-negative urine culture results when using the BDUs and prolonged storage of specimens (≥24 hours). However, our study results did not confirm these findings and results of categorical assessment of urine cultures as well as semiquantitative assessment with R-NPU samples. However, these studies have some differences that must be considered. Hubbard et al17 used glass tubes for urine collection, which contained a slightly different preservative, compared with the plastic tubes with buffered boric acid that were used in our study. Furthermore, considering potential differences in thresholds (>104 vs >105 CFU/mL) used by various laboratories for ascribing clinical significance, we chose a most conservative approach for our analysis, using a threshold of more than 104 CFU/mL. There was a statistically highly significant change in ascribing significance to culture results for urine specimens in the RT-NPU group, particularly for specimens processed more than 4 hours after sample collection. However, no statistically significant changes were observed for specimens in the BDU and R-NPU groups for all intervals throughout this study.
So far we focused on the qualitative assessment of urine culture results, which is important with regard to the presence and/or absence of uropathogens and the occurrence of changes in MSF and/or pathogens present in urine cultures. However, we also investigated whether the preservative in the BDUs may have an effect on the colony counts over time compared with corresponding RT-NPU and R-NPU urine samples. In their 1981 clinical study, Guenther and Washington16 reported a decline in colony counts in urine specimens using the BDUs when specimens were stored for 24 hours. Furthermore, these investigators proposed that 104 or more CFU/mL in urine preserved for 24 hours should be considered equivalent to 105 or more CFU/mL in freshly collected or refrigerated urine. Predicated on this assumption, there was at least 98% agreement between initial urine cultures yielding 105 or more CFU/mL and preservative urine cultures at 24 hours. However, of the specimens containing 104 to 105 CFU/mL in the initial culture, only one-third yielded less than 104 CFU/mL in urine specimens preserved for 24 hours and would then have been considered as “negative.” To our knowledge, only 2 studies have been published to date that investigated the stability of colony counts over time.16,19 In another study, Weinstein20 demonstrated a lack of toxicity of lyophilized buffered boric acid. Furthermore, he found no difference in quantitative urine culture results between specimens in preservative-containing urine collection tubes kept at room temperature and specimens in conventional sterile cups refrigerated during the holding period of 18 to 24 hours. Contrary to the results of Guenther and Washington16 but similar to those of Weinstein, we did not find a significant decline in colony counts in urine samples preserved in the BDUs compared with the initial urine culture results (Table 5). The minor changes in colony counts observed in both refrigerated and preserved urine specimens were not statistically significant. Furthermore, we extended the time for culture and colony count comparison to 48 hours after specimen collection and again did not see statistically significant changes in colony counts for either type of urine specimen. Based on our study results, it appears not necessary to exercise particular caution when interpreting the culture results and colony counts for urine samples preserved for 24 hours.
However, one must consider that our study was performed under the most ideal of conditions, that is, urine samples were rapidly transported to the laboratory (average transport time, 25 min), and samples in the R-NPU group were kept constantly at 4°C refrigeration, with the exception of the time for inoculating agar plates. All samples and cultures were processed under strict clean/aseptic conditions to avoid external contamination during the process. In the 2005 CAP Q-Probes study on urine culture contamination, the investigators found that only a small number of microbiology laboratories enforce the 2-hour cutoff rule for limiting transport time of urine specimens.12 The investigators found that 90% of the specimens in the surveyed laboratories were received within 9 hours. Although the majority of these laboratories processed urine specimens through a central receiving area, only 35% of these laboratories refrigerated the specimens before or during the processing. The data showed that even a greater number of sites, both adjacent and nonadjacent to a laboratory, do not consistently refrigerate urine specimens before transporting them to the laboratory. Numerous studies in the past have demonstrated that most urine specimens submitted for microbiologic analysis and culture are either negative or contain a mixed urogenital flora at low colony counts (<104 CFU/mL).5,8,9,12,16,17 As the results of these previous studies and our current study have demonstrated, colony counts and categorical assessment of urine specimens remain stable within the first 2 hours. Some of these earlier studies described differences in urine culture interpretation and colony counts between refrigerated urine samples and urine samples containing preservative media (such as boric acid). In the current study, however, we did not identify statistically significant differences between these types of urine sample preservation methods.
The results of the CAP Q-Probes study lead us to question the accuracy of the assumption that patients’ urine samples are continuously kept at 4°C (refrigeration) from the time of collection to the time of processing in the laboratory. It is therefore unlikely that urine specimens will be continuously refrigerated from the time of collection to the time of culture setup in routine clinical and laboratory settings. The results of our study demonstrate that bacterial growth in urine samples will significantly change even after a short period of exposure to ambient/room temperature during specimen transport. Considering these constraints of everyday clinical and laboratory practice, the use of some form of preservative, such as the buffered boric acid, as an additive to urine collection tubes is highly desirable and should perhaps be even mandatory for specimens sent from a distant specimen collection site. Most studies, including the present study, demonstrate that the majority of urine specimens sent for microbiologic examination and culture will be negative for growth, show some growth of mixed urogenital flora, or at best are positive for a uropathogen at significantly less than 104 CFU/mL, the latter often not being considered indicative of a UTI but rather being contaminated during the specimen collection process.1–3 If such nonpreserved urine specimens were to present, however, with higher colony counts because of delays in specimen transport and processing and if the false-positive rate could be in excess of 15% as suggested by the CAP Q-Probes study, reporting of these urine culture results would have significant implications on patient care, antibiotic use, and other economic health care–related cost.
The present study has some limitations. First, urine specimens were not processed for culture at time points between 4 and 24 hours after sample collection. Many of the changes in colony counts and categorical assessment occurred during this period, and it may have been helpful to better understand the timely sequence of these changes. Second, the overall number of urine specimens included is limited, and the organisms that cause UTIs less frequently, eg, P aeruginosa, were not included in this study in significant numbers to allow for a more detailed and/or organism-specific analysis. Finally, the comparison among various studies that investigated the usefulness of preservatives in urine collection systems is limited by the fact that the various preservatives used in these studies may be inherently different (eg, boric acid vs buffered boric acid) or that simply different types of collection tubes (glass vs plastic) were used.
In conclusion, the BDU is an effective approach for maintaining the qualitative as well as semiquantitative assessment of urine specimens for culture for up to 48 hours after specimen collection. The results of the preserved urine samples were equal and/or better than those using the approach of refrigeration. Prompt plating or even prompt and consistent refrigeration of urine specimens cannot be assured in most routine clinical and laboratory settings. Therefore the use of urine collection and transport tubes containing a preservative such as buffered boric acid is highly recommended and should perhaps be mandatory when specimens are sent from a distant collection site or routinely exceed a 2-hour specimen transport time.