PEDIATRICS Vol. 119 No. 6 June 2007, pp. e1288-e1293 (doi:10.1542/peds.2006-2392)
ARTICLE |
To Clean or Not to Clean: Effect on Contamination Rates in Midstream Urine Collections in Toilet-Trained Children
a Division of Pediatric Emergency Medicine
c Departments of Pediatrics and Epidemiology and Biostatistics, Montreal Children's Hospital
b Montreal Children's Hospital Research Institute, McGill University, Montreal, Quebec, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. Urinary tract infection is one of the most common bacterial infections among children. Difficulty in specimen collection and interpretation of inadequately collected specimens may contribute to misdiagnosis of urinary tract infection. Our objective was to assess the effect of perineal/genital cleaning on bacterial contamination rates of midstream urine collections in toilet-trained children.
METHODS. We conducted a randomized trial in toilet-trained children who presented to a tertiary care pediatric emergency department between November 1, 2004, and October 1, 2005. All toilet-trained children who were between the ages of 2 and 18 years and had a midstream urine sample requested were eligible. Those whose parents consented were cluster-randomized by week to either cleaning or not cleaning the perineum with soap. The risk for a contaminated urine culture (defined as growth of <108 colony-forming units per liter [<105 colony-forming units per milliliter] of a single organism or a mix of 
2 organisms) and the risk for a positive urinalysis (defined as a positive leukocyte esterase and/or nitrites on dipstick or 5 white blood cells per high-powered field on a standard microscopic examination) were analyzed by intention to treat.
RESULTS. A total of 350 children were enrolled. The rate of contamination in the cleaning group was 14 (7.8%) of 179 vs 41 (23.9%) of 171 in the noncleaning group. Children who were randomly assigned to cleaning were less likely to have a positive urinalysis (37 of 179 [20.6%]) than those in the noncleaning group (63 of 171 [36.8%]).
CONCLUSIONS. Urine contamination rates are higher in midstream urine that is collected from toilet-trained children when obtained without perineal/genital cleaning. Cleaning may reduce the risk for returning for repeat cultures and for receiving unnecessary antibiotic treatment and investigations.
Key Words: urinary tract infection • midstream clean catch • contamination rates
Abbreviations: UTI—urinary tract infection • ED—emergency department • CFU—colony-forming unit(s) • RR—relative risk • OR—odds ratio • CI—confidence interval • GLIMMIX—generalized linear matrix model
Urinary tract infection (UTI) is one of the most common bacterial infections among children.1,2 UTIs are an important cause of chronic morbidity in children; long-term complications include hypertension and reduced renal function secondary to renal scarring.3 The accurate diagnosis of a UTI is necessary to ensure appropriate therapy for infected children and to avoid unnecessary therapy and prevent hospital admission for additional evaluation in noninfected children.4,5 A urine culture result is considered the gold standard for diagnosis of a UTI, but difficulty in specimen collection and interpretation of inadequately collected specimens may contribute to its misdiagnosis in children.
Few studies have addressed the method of collecting a midstream urine specimen in outpatient children. The "clean-catch" midstream void technique has been the method of choice in adults since the late 1950s.6,7 The clean-catch midstream technique, however, is time-consuming to explain, frequently performed incorrectly, and associated with increased costs. Several studies in adults have reported no difference in contamination rates between midstream clean-catch and midstream non–clean-catch urine samples.8–13 Authors of these studies hypothesized that cleaning does not lower contamination rates because the urethra is irritated during cleaning. In toilet-trained children, only 3 studies have evaluated the effect of cleaning on bacterial contamination rates in midstream specimens. All 3 studies showed no difference in contamination rates,14–16 but none was randomized; therefore, their results may be confounded by differences in those cleaned versus uncleaned. In this article we describe the first randomized trial, to our knowledge, to address this question.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Design and Patients
We conducted a randomized trial in toilet-trained children who presented to the emergency department (ED) of a tertiary care pediatric center between November 1, 2004, and October 1, 2005. All toilet-trained children who were between the ages of 2 and 18 years and had a midstream urine sample requested by a physician or triage nurse for any medical reason were eligible for the study. Children were excluded when they were not toilet-trained or their parents believed that they would be unable to comply with the collection technique because of developmental delay. Informed consent was obtained by either the research assistant who worked in the ED or the physician who requested the urine sample. The hospital research ethics board approved the study.
At the beginning of each week throughout the study period, the method for collecting a urine specimen for all children who presented to the ED during that week was randomly assigned by the study investigators. A card stating either "cleaning" or "noncleaning" was randomly chosen from a hat by the study investigator on each Monday morning at 8 AM. Multiple posters were then displayed to inform the ED staff as to whether it was a cleaning or noncleaning week. In addition, during the cleaning week, the soap was given to the children and their parents, and during a noncleaning week, the soap was not given to the children and their parents. Allocation to either the cleaning group or the noncleaning group was the same for all children who presented during that week. The nurses and physicians were not blinded to the collection technique. After informed consent was obtained, verbal and written instructions for obtaining the urine specimen were given to the child and parent(s).
During the cleaning week, the child was given liquid soap (Isoderm; Lalema, Montreal, Quebec, Canada), several gauze pads, a sterile urine-collection container, and an instruction sheet. The child and/or the parent was instructed to spread the labia (for girls) or retract the foreskin (for noncircumcised boys), to clean the urethral meatus and perineum with gauze and liquid soap twice (wiping from front to rear in girls), to urinate into the toilet, and, midway through urination, to collect the urine into the sterile container. During the noncleaning week, the child was given a sterile urine-collection container and an instruction sheet. The child and/or the parent was instructed to spread the labia (for girls) or retract the foreskin (for noncircumcised boys), to urinate into the toilet, and, midway through urination, to collect the urine into the sterile container.
A questionnaire was administered with the urine-collection instructions to the parent of each participating child to document the age, gender, circumcision status (for boys), antibiotic use in the previous 2 weeks, and previous renal problems (other than a previous UTI). Recent antibiotic use was documented because bacterial colony counts may fall below the characteristic range for an infection when an antimicrobial agent is present in the urine. To assess compliance with the collection instructions, we also asked whether the child cleaned with soap before collecting the urine sample. We did not collect data on the presence or absence of urinary symptoms. Information that was recorded from the urinalysis included the presence of nitrites or leukocyte esterase in fresh urine and the presence or absence of bacteria and the number of white blood cells per high-powered field on microscopy of spun urine.
Laboratory Methods and Definitions
On the basis of previous studies,17 the standard of practice at our institution is to perform the urine dipstick on all urine specimens. When the dipstick indicates a positive finding for any component (including leukocyte esterase, nitrites, ketones, protein, glucose, or blood), a microscopic urinalysis is performed. A positive urinalysis is defined as a positive leukocyte esterase and/or nitrites on dipstick or 5 white blood cells per high-powered field on a standard microscopic examination done after centrifuging the urine.17 All samples with a positive urinalysis were sent for culture. Given that false-negative urinalysis results in toilet-trained children are rare, urine samples are not sent for culture when the urinalysis is negative.
Urine specimens were sent to the microbiology laboratory in sterile containers. Standard quantitative culture was performed by laboratory technologists. Urine was plated onto MacConkey agar plates and sheep-blood agar plates with a 0.01-mL calibrated loop. Cultures were incubated for 48 hours at 35°C and examined daily for growth. Microbiologists at the laboratory were blinded to the randomization week and hence to the cleaning or noncleaning status of each child's specimen. A positive urine culture was defined as growth of a single pathogenic organism with 108 colony-forming units (CFU)/L (105 CFU/mL) of urine. Urinary tract pathogens included Escherichia coli, Proteus, Klebsiella, Enterococcus, Enterobacter, and Pseudomonas. Nonpathogenic organisms included Lactobacillus, Corynebacterium, and coagulase-negative staphylococci.18 A contaminated urine culture was defined as a culture growing a single organism with <108 CFU/L (<105 CFU/mL) or a mix of 2 or more organisms.13,19 A negative culture was defined as no growth. For assessment of whether the contamination rate was affected by the time of day when the sample was plated in the microbiology laboratory, the time of collection of the urine sample was recorded.
Statistical Analysis
We estimated the number of children needed to detect a clinically important difference of 20%, based on an 
value of .05 and a power of 80%, as 182 in each group. This estimation was based on a predicted contamination rate of 20% in urine that was obtained after cleaning. Relative risks (RR) and their 95% confidence interval (CI) for contamination and for a positive urinalysis were calculated using SPSS 11.0 (SPSS, Chicago, IL). Analysis was based on intention to treat. To test for a difference in the culture result between the cleaning and noncleaning groups among samples with a positive urinalysis, we calculated a 
2 test statistic. Because the intervention was randomized by week rather than by individual patient, a generalized linear mixed model (GLIMMIX) was also used to account for potential within-week clustering of outcome results and to adjust for potential baseline imbalances in gender, age, and history of renal problems. The interaction terms for treatment by gender and age were also included in the GLIMMIX model.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
During the study period, 350 children who met the eligibility criteria were enrolled (Fig 1). The majority were girls (211 [60%]). Sixty-five percent of the boys enrolled were noncircumcised. Preschool children (aged 2–5) composed 45% of the study sample. Table 1 displays the baseline characteristics of all participating children and shows that the 2 intervention groups were similar in these characteristics.
|
|
Reported adherence to the urine-collection instructions was excellent: >90% of parents stated that their child followed the instructions given. The parents of 10 children in the noncleaning group indicated on the questionnaire that they cleaned their child with soap before giving the urine sample, and 15 parents of children in the cleaning group stated that they did not clean with soap despite being instructed to do so.
The overall prevalence of UTI in the study population was 7%, with the most common pathogen, Escherichia coli, growing in 19 (83%) of 23 of the positive cultures. Two other pathogens isolated were Enterococcus (n = 1) and Proteus (n = 3) species. Ninety-six percent of the positive cultures were seen in girls, with a mean age of 8 years. The majority of contaminated specimens grew a mix of 2 or more species (n = 34). The remaining contaminated cultures grew either 1 organism <107 CFU/L (<104 CFU/mL; n = 15) or nonpathogenic organisms, including Corynebacterium (n = 3) and coagulase-negative Staphylococcus epidermis (n = 3). Of the 55 contaminated urine specimens, 21 were collected during the day and 34 were collected during the evening. None of the specimens that were collected overnight was contaminated. The contamination rate in the cleaning group was 14 (7.8%) of 179 compared with 41 (23.9%) of 171 in the noncleaning group (RR: 0.37; 95% CI: 0.19–0.57; Table 2). Urine specimens that were obtained with cleaning were less likely to have a positive urinalysis (37 of 179 [20.6%]) than those that were obtained in the noncleaning group (63 of 171 [36.8%]) (RR for positive urinalysis: 0.56; 95% CI: 0.40–0.79; Table 2). In the GLIMMIX model used to account for potential clustering within weeks, the effect was similar; the OR for contamination in the cleaning group was 0.27 (95% CI: 0.14–0.51), and the OR for a positive urinalysis in the cleaning group was 0.45 (95% CI: 0.28–0.72). In the cleaning group, the predictive value of a positive urinalysis for UTI was 15 (40.5%) of 37, compared with 8 (12.7%) of 63 in the noncleaning group (P = .001). Sensitivity, specificity, and negative predictive value could not be calculated, because only specimens with positive urinalyses were sent for culture. When a positive urine culture was defined as 107 CFU/L (104 CFU/mL) instead of 108 CFU/L (105 CFU/mL), only 1 urine sample that previously was classified as contaminated would be reclassified as positive and the contamination rates were almost identical: 14 (7.8%) of 179 in the cleaning group versus 40 (23.4%) of 171 in the noncleaning group (RR: 0.38; 95% CI: 0.20–0.59).
|
We also assessed the treatment effect after stratification by gender, age, and circumcision status. In the age group 2 to 5 years, 15.3% in the noncleaning group were contaminated versus 10.3% in the cleaning group. In the age group 6 to 18 years, 28.6% of samples in the noncleaning group were contaminated versus 6.3% in the cleaning group. The interaction between age and treatment was statistically significant (P = .04), whereas the interaction between gender and treatment was not (P = .15). Only 4 boys had contaminated urine specimens, 3 of whom were uncircumcised (1 in the cleaning group and 2 in the noncleaning group), with no circumcised boys in the cleaning group, thereby preventing testing for interaction by circumcision status.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
In this cluster-randomized trial, we observed that children who were randomly assigned to receive genital/perineal cleaning with soap before urine collection had lower rates of contaminated urine specimens and of positive urinalyses. Moreover, the predictive value of a positive urinalysis for UTI was significantly higher in the cleaning group. We are aware of only 3 previous studies that compared methods of obtaining a midstream urine sample in children, none of which found cleaning to reduce the contamination rate.14–16 The definition of a contaminated urine sample used in those studies was similar to that used in our study, but none used randomized allocation, their sample sizes were smaller, and the collection techniques were not standardized.
The results of our study and other studies that assessed rates of contamination depend on the definition of a contaminated urine specimen. Since the 1950s, the traditional <108 CFU/L (<105 CFU/mL) cutoff point has been used by clinicians to denote a positive culture.20 The prevailing view is that a mixed culture in an uncomplicated outpatient likely indicates contamination. Low levels (<107 CFU/L [<104 CFU/mL]) of organisms that commonly are found on the skin and external genitalia are considered to be contaminants. After reviewing the literature,13,18,20–24 we defined a positive urine culture as growth of 108 CFU/L (105 CFU/mL) of urine from a clean-catch midstream specimen. Our overall results were very similar when we defined a positive culture as 107 CFU/L (104 CFU/mL): all but 1 of the contaminated specimens were attributable either to mixed specimens or to nonpathogenic organisms with <107 CFU/L (<104 CFU/mL). In previous studies in adults, contamination rates for clean-catch midstream urine specimens have ranged from 0% to 32%.25–27
In exploratory posthoc analyses, we observed a significantly larger difference in contamination rates according to treatment among children who were 6 to 18 years than among those who were 2 to 5 years. This difference in effect of cleaning may be attributable to poorer local hygiene and/or more careful cleaning in the older children.
Because urine samples with a positive urinalysis were sent for culture and because we observed more positive urinalyses among the noncleaning group, a higher percentage of samples from the noncleaning group were sent for culture; this has implications for clinical practice, because clinicians prefer to identify children with a sufficiently increased likelihood of UTI to justify presumptive antibiotic treatment while awaiting the results of the urine culture, which may not be available for 24 to 48 hours. Management decisions are therefore often based on the results of the urinalysis and can include treating the child immediately with antibiotics, waiting until the culture result is known, or obtaining a repeat collection of the urine sample.
It has been hypothesized that daytime urine (fluid diuresis) and frequent daytime voiding can result in lower colony counts. Conversely, overnight dwelling of urine in the bladder may lead to an increased concentration of bacteria. However, time of collection did not affect the contamination rates in our study.
Although written instructions were given to the child and the parent, it is difficult to verify that either cleaning or midstream collection was accurately performed. We attempted to assess the collection procedure by having the child and/or the parent complete a questionnaire to assess compliance with the collection instructions. Our procedure reflects what occurs in clinical practice, because neither the physician who requests the urine nor the nurse who gives the instructions is involved in observing or performing the collection of the urine. In any case, any unreported noncompliance with the allocated procedure should have reduced differences in contamination rates between the 2 study groups.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Urine contamination and positive urinalysis rates were lower in midstream urine that was collected from toilet-trained children who were randomly assigned to receive perineal/genital cleaning. We also observed a higher positive predictive value for the urinalysis in those children. Although our study was limited to children who were seen in an ED setting, our results should be generalizable to other clinical settings. It is important to make an accurate diagnosis of a UTI in children to provide appropriate treatment and evaluation for these children. However, a collection method that results in higher contamination rates is likely to lead to an increased proportion of children's having to return for repeat cultures and/or potentially receiving unnecessary treatment and investigations. Our results therefore strongly suggest that toilet-trained children should have their perineum cleaned with soap before collecting a midstream urine specimen.
|
ACKNOWLEDGMENTS |
|---|
This research was supported by a grant from the Montreal Children's Hospital Research Institute.
Duncan Lejtenyi, Aileen Frew, and the nurses of the Montreal Children's Hospital Emergency Department assisted in recruiting children into the trial. Sebastian Dube provided statistical assistance. Dr Kramer is a Senior Investigator of the Canadian Institutes of Health Research.
|
FOOTNOTES |
|---|
Accepted Nov 21, 2006.
Address correspondence to Suzanne Vaillancourt, MD, Montreal Children's Hospital, Place Tourelles, 2222 Blvd Rene Levesque Ouest, First Floor, Room T131-C, Montreal, Quebec, Canada H3H 1P3. E-mail: suzanne.vaillancourt{at}mail.mcgill.ca
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Hoberman A, Chao HP, Keller DM, Hickey R, Davis HW, Ellis D. Prevalence of urinary tract infection in febrile infants. J Pediatr. 1993;123 :17 –23[CrossRef][Web of Science][Medline]
- Siegel SR, Siegel B, Sokoloff BZ, Kanter MH. Urinary infection in infants and preschool children: five-year follow-up.
Am J Dis Child. 1980;134
:369
–372
[Abstract/Free Full Text] - Downs SM. Technical report: urinary tract infections in febrile infants and young children. Pediatrics 1999;103(4) . Available at: www.pediatrics.org/cgi/content/full/103/4/e54
- McKerrow W, Vidson-Lamb N, Jones PF. Urinary tract infection in children. Br Med J (Clin Res Ed). 1984;289 :299 –303[Medline]
- Al-Orifi F, McGillivray D, Tange S, Kramer MS. Urine culture from bag specimens in young children: are the risks too high? J Pediatr. 2000;137 :221 –226[CrossRef][Web of Science][Medline]
- Beeson PB. The case against the catheter. Am J Med. 1958;24 :1 –3[CrossRef][Web of Science][Medline]
- Boshell BR, Sanford JP. A screening method for the evaluation of urinary tract infections in female patients without catheterization.
Ann Intern Med. 1958;48
:1040
–1045
[Abstract/Free Full Text] - Leisure MK, Dudley SM, Donowitz LG. Does a clean-catch urine sample reduce bacterial contamination?
N Engl J Med. 1993;328
:289
–290
[Free Full Text] - Baerheim A, Digranes A, Hunskaar S. Evaluation of urine sampling technique: bacterial contamination of samples from women students. Br J Gen Pract. 1992;42 :241 –243[Web of Science][Medline]
- Holliday G, Strike PW, Masterton RG. Perineal cleansing and midstream urine specimens in ambulatory women. J Hosp Infect. 1991;18 :71 –75[Web of Science][Medline]
- Lifshitz E, Kramer L. Outpatient urine culture: does collection technique matter?
Arch Intern Med. 2000;160
:2537
–2540
[Abstract/Free Full Text] - Lipsky BA, Inui TS, Plorde JJ, Berger RE. Is the clean-catch midstream void procedure necessary for obtaining urine culture specimens from men? Am J Med. 1984;76 :257 –262[CrossRef][Web of Science][Medline]
- Schlager TA, Smith DE, Donowitz LG. Perineal cleansing does not reduce contamination of urine samples from pregnant adolescents. Pediatr Infect Dis J. 1995;14 :909 –911[Web of Science][Medline]
- Lohr JA, Donowitz LG, Dudley SM. Bacterial contamination rates for non-clean-catch and clean-catch midstream urine collections in boys. J Pediatr. 1986;109 :659 –660[CrossRef][Web of Science][Medline]
- Lohr JA, Donowitz LG, Dudley SM. Bacterial contamination rates in voided urine collections in girls. J Pediatr. 1989;114 :91 –93[CrossRef][Web of Science][Medline]
- Saez-Llorens X, Umana MA, Odio CM, Lohr JA. Bacterial contamination rates for non-clean-catch and clean-catch midstream urine collections in uncircumcised boys. J Pediatr. 1989;114 :93 –95[CrossRef][Web of Science][Medline]
- Hoberman A, Wald ER. Urinary tract infections in young febrile children. Pediatr Infect Dis J. 1997;16 :11 –17[Medline]
- Shaw KN, McGowan KL. Evaluation of a rapid screening filter test for urinary tract infection in children. Pediatr Infect Dis J. 1997;16 :283 –287[CrossRef][Web of Science][Medline]
- American Academy of Pediatrics, Committee on Quality Improvement, Subcommittee on Urinary Tract Infection. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children [published corrections appear in Pediatrics. 2000;105:141;Pediatrics. 1999;103:1052; and Pediatrics. 1999;104:118].
Pediatrics. 1999;103
:843
–852
[Abstract/Free Full Text] - Kass EH. Bacteriuria and the diagnosis of infections of the urinary tract: with observations on the use of methionine as a urinary antiseptic. Arch Intern Med. 1957;100 :709 –714[Web of Science][Medline]
- Shaw KN, Hexter D, McGowan KL, Schwartz JS. Clinical evaluation of a rapid screening test for urinary tract infections in children. J Pediatr. 1991;118 :733 –736[CrossRef][Web of Science][Medline]
- Ramage IJ, Chapman JP, Hollman AS, Elabassi M, McColl JH, Beattie TJ. Accuracy of clean-catch urine collection in infancy. J Pediatr. 1999;135 :765 –767[CrossRef][Web of Science][Medline]
- Bulloch B, Bausher JC, Pomerantz WJ, Connors JM, Mahabee-Gittens M, Dowd MD. Can urine clarity exclude the diagnosis of urinary tract infection? Pediatrics. 2000;106(5) . Available at: www.pediatrics.org/cgi/content/full/106/5/e60
- Gorelick MH, Shaw KN. Screening tests for urinary tract infection in children: a meta-analysis. Pediatrics. 1999;104(5) . Available at: www.pediatrics.org/cgi/content/full/104/5/e54
- Lifshitz E, Kramer L. Outpatient urine culture: does collection technique matter?
Arch Intern Med. 2000;160
:2537
–2540
[Abstract/Free Full Text] - Aronson AS, Gustafson B, Svenningsen NW. Combined suprapubic aspiration and clean-voided urine examination in infants and children. Acta Paediatr Scand. 1973;62 :396 –400[Web of Science][Medline]
- Valenstein P, Meier F. Urine culture contamination: a College of American Pathologists Q-Probes study of contaminated urine cultures in 906 institutions. Arch Pathol Lab Med. 1998;122 :123 –129[Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  L. Winckworth and R. Ayat QUESTION 1 Arch. Dis. Child., September 1, 2009; 94(9): 736 - 739. [Full Text] [PDF]
|
|
|
|
 |
|
 Midstream Urine Samples in Children: Clean First, Then Collect Journal Watch Pediatrics and Adolescent Medicine, June 20, 2007; 2007(620): 2 - 2. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
