Urinary tract infections (UTIs) remain a significant cause of serious bacterial infections in children [1,2]. In addition, infections of the urinary tract can result in chronic kidney disease and hypertension [3•,4]. Thus, prompt diagnosis and treatment are critical in preventing the pathologic sequelae of UTIs.
Diagnosing UTIs has been the focus of many studies over the past 60 years. The classic study by Kass  defined the difference between asymptomatic bacteriuria and UTI for adult women with a cut-off of 105 organisms per ml of urine. Over the following decade, Pryles and others [6–10] had studied many aspects of diagnosing UTI in infants and advocated the technique of suprapubic aspiration of the bladder for obtaining urine for culture. While clarifying much of our understanding of UTIs in children, these studies continued to point out the complexities of assessing children with suspected UTI as well as the problems associated with obtaining urine without contamination for culture .
The review will discuss results of recent studies as well as review older studies that remain relevant that will aid the clinician in determining the need for urine culture and how to obtain the urine for culture. This will include utilizing technologic advances in the analysis of urine that will allow clinicians to quickly eliminate the need for urine culture. More importantly, the prevalence of UTI in various populations will be reviewed, which will also aid in the decision to culture the urine and possibly treat empirically while the culture results are pending. Thus, a complete history and physical examination of the patient are just as critical in the evaluation of the patient with suspected UTI as the laboratory tests that are ordered.
Definition of urinary tract infection
Culture of the urine remains the gold standard for diagnosing and treating UTIs [4,11•]. Urine is usually sterile; thus any bacteria growing in it should be considered an infection. Unfortunately, oftentimes the culture will grow a bacterium that is obviously a contaminant, either from the skin or from other parts of the genital tract. Thus, most investigators define a UTI as the presence of organisms in the urine combined with signs or symptoms of UTI in the patient [4,11•].
In addition, numerous studies indicate that many people have asymptomatic bacteriuria which is distinguished from a true UTI. This was firmly established by Kass  in his landmark study published in 1956. He found that, if the urine culture obtained by clean catch had fewer than 104 colony-forming units per ml, the patient did not have urinary symptoms. When the colony count was greater than 105 colony-forming units per ml, the likelihood of symptoms was much greater. This study then led to the definition of a UTI as being a urine culture revealing greater than 105 colony-forming units per ml in an adult with urinary symptoms. Whereas this study was performed in adult women, it has become applied to all age groups.
Pryles  reviewed the existing pediatric data in 1960 and came to similar conclusions regarding the definition of UTI in children. His conclusions remain just as applicable today as they did almost 50 years ago. He stated that urine cultures with fewer than 103 colony-forming units per ml were almost always contamination, those with between 104 and 105 colony-forming units per ml were suspicious and should be repeated, and those with more than 105 colony-forming units per ml were indicative of infection.
The technique of obtaining urine for culture will affect the rate of contamination and will thus influence the interpretation of the culture result. Many of the initial studies on UTIs utilized urine that was obtained by clean catch. This is certainly the easiest way to obtain urine since it requires almost no preparation or special equipment. Unfortunately, there are theoretical and practical concerns that the urine could become contaminated with bacteria from the skin or genital area and thus not represent what is in the urinary tract. In addition, collection by this means is almost impossible in infants who are not toilet trained.
A number of studies have made direct comparisons of the results of cultures of urine obtained by both clean catch and catheterization. An early study by Pryles and Steg  indicated that, for girls between the ages of 2 and 12 years, clean voided specimens had similar rates of contamination to those obtained by catheterization. The first reported study in children by Masters  had similar results and even suggested that urine specimens obtained by catheterization might have higher rates of contamination than clean voided specimens.
A more recent study in younger patients (under 2 years of age) showed that culture results from clean voided samples and catheterized samples were similar for the boys who were studied but showed less correlation for the girls . The patients in this study did not have UTIs and were being studied only to compare the rates of false-positive cultures. However, it was interesting to note that they had high rates of contamination using both methods of urine collection.
The technique of suprapubic aspiration of the urine was introduced in 1959 by Pryles et al.. A follow-up study 10 years later continued to show that this technique was well tolerated and was very useful in diagnosing UTIs, especially in patients who had low colony counts from urine cultures performed by other techniques . It is interesting to note that many review articles continue to mention suprapubic aspiration of the urine as the preferred method of obtaining urine for culture, yet most recent studies have compared all other techniques except for suprapubic aspiration [2,14•].
Conflicting results were found when comparing clean voided specimens and catheterized specimens with those obtained by suprapubic aspiration of the urine . These investigators found the results of clean voided urine and catheterized urine samples to have such a high rate of contamination that they were not acceptable. This calls into question the need for training of personnel to correctly perform the methods for obtaining urine for culture.
The American Academy of Pediatrics (AAP) published their Practice Parameter for the diagnosis and treatment of UTIs in infants and young children in 1999 . If an infant was found to be ill enough to begin empiric antibiotic treatment, they strongly recommended that the urine culture be obtained by suprapubic aspiration or by catheterization. The false-positive rate of urine cultures obtained by bag collection was found to be too high to be clinically acceptable. The Academy reiterated the fact that the suprapubic aspiration of urine for culture is considered the ‘gold standard’ .
A more recent review of the literature outlined the WHO guidelines for obtaining urine for culture [17••]. These recommendations are very similar to those of the AAP. The WHO guidelines state that, for infants who are unwell enough to consider empiric treatment, the recommendation is to obtain urine either by suprapubic aspiration or by catheterization. The high false-positive rate of bagged urine specimens precludes it from being a useful technique in this setting. In older children who can void on command, the use of clean catch, especially the midstream urine, was found to be an acceptable technique for obtaining urine for culture [17••]. This review also regarded the suprapubic aspiration technique as the ‘gold standard’. Complications from this technique were found to be minimal, but the main problems with its usefulness were inadequate training and parental anxiety [17••].
Urine analysis: microscopic examination
Regardless of the technique of sampling the urine for culture, the problem remains regarding the length of time needed to perform the actual culture. In most cases, preliminary results of the culture are not available for 24 h and conclusive results with susceptibilities of the organisms are not available for 48–72 h. Because of the risk of renal damage from an untreated UTI, many physicians would prefer to treat the infant with antibiotics while the culture results are pending. Thus, there is a need for a more rapid determination of the probability of the presence of a UTI to help guide the clinician in the decision to treat empirically.
Observations from more than 50 years ago indicated that pyuria was found in almost all urines that had elevated colony counts and were interpreted as being infected . This was confirmed in a study that evaluated children with urine obtained by catheterization . A more recent study by the same group comparing pyuria and urine culture results obtained by catheterization showed that 10 leukocytes/mm3 was very good at distinguishing those urines that were sterile from those that had high colony counts indicative of infection .
An ‘enhanced’ urinalysis consists of the number of leukocytes/mm3 in a Neubauer hemocytometer and the presence of any bacteria in a gram-stained smear . This has been evaluated in a number of studies to determine its usefulness in predicting the presence or absence of UTI [20,21]. A more recent meta-analysis that compared a number of different methods for screening urine samples for possible infection pointed out that, although the enhanced urinalysis might be useful, it has only been reported in the above two studies and that these were performed in infants under 2 years of age . Thus, although the enhanced urinalysis might eventually prove beneficial, it requires further investigation.
These data have been reviewed again in a meta-analysis by other investigators, who came to similar conclusions . The presence of greater than 10 leukocytes/mm3 together with the presence of bacteria on the gram stain was highly predictive of a UTI.
More recent studies have examined the use of automated cell counters to determine the number of leukocytes as well as the quantity of bacteria in the uncentrifuged urine [24•,25]. Again, these tests were helpful in predicting the presence of UTI but have limitation in that they require specialized equipment as well as trained personnel for the analysis.
Urine analysis: dipstick analysis
The urinalysis also incorporates the dipstick analysis for the presence of leukocyte esterase and nitrites. These tests have a number of advantages over microscopic analysis, which include their ease of use, rapid result time, and the requirement for very little specialized equipment . The leukocyte esterase test will detect the presence of this enzyme that is released from white blood cells that are present in the urine, which makes this a surrogate for the microscopic analysis of pyuria. Nitrites are produced by the bacterial conversion of nitrates that are normally found in the urine. Most of the gram-negative rods that are found in urinary infections are capable of producing nitrites using this reaction. However, the clinician must realize that many gram-positive cocci will not produce nitrites and would then yield a false-negative result for the presence of nitrites .
There have been numerous studies performed to evaluate the ability of these tests to help determine which patients will need further evaluation for UTI. A meta-analysis has been performed to evaluate these studies which covered the years from 1975 to 1998 . At the time, the gram stain was found to be very sensitive and specific for determining which urine cultures would be positive. The authors also found that, if both the leukocyte esterase and nitrites were positive, the sensitivity and specificity were very close to those of the gram stain . Since this meta-analysis was published, another group has extended this analysis and came to similar conclusions .
Additional studies have confirmed that a combination of the presence of leukocyte esterase and nitrites is highly predictive of a positive urine culture . Results of these screening tests need to be interpreted in light of the patient, since at least one study showed that the dipstick urinalysis would have missed a significant number of urine infections . The specificity of this combination has also been shown to be good, so that if both are negative the likelihood of a UTI is very low.
Prevalence of urinary tract infection
The interpretation of any lab test has to take into account the clinical findings of the patient at the time the test is performed. Significant questions to answer about the patient include the presence or absence of fever, urinary symptoms, and past history of urine infections or urologic malformations. These factors can greatly affect the probability of the patient having a UTI and have been the focus of recent studies.
A recent meta-analysis searched all publications on the topic of UTI in children from 1966 to 2005 [28••]. This study found the overall prevalence of UTI in children evaluated for fever to be about 7% but it varied greatly depending on age, sex, and circumcision status. This study has verified that UTI remains the most common serious bacterial infection in children. The prevalence was found to be higher in white infants (8%) than black infants (4.7%). The highest prevalence of UTI was found in uncircumcised male infants less than 3 months of age and females less than 12 months of age. Interestingly, in older children with urinary symptoms, the prevalence was only 7.8% in patients whether or not they had fever [28••].
These data on prevalence of UTI can be applied to clinical decision making when evaluating patients for suspected UTI. If the suspicion of UTI was very high, most physicians would elect to treat a patient with empiric antibiotics while urine culture results are pending. However, if the suspicion of UTI was very low, the physician might wait for 1–2 days for the culture result prior to initiating antibiotics. The problem is in knowing what the odds are that the patient has an actual UTI. Thus, knowing the prevalence of UTI in the various patient populations can help in this decision-making process. A recent study reviewed the presenting signs and symptoms of a child who presented with a fever to determine which signs and symptoms are most indicative of a UTI. They offered a rational approach to evaluation and treatment on the basis of the presentation of the patient [29••].
Because UTIs remain a significant cause of serious bacterial infections in children, diagnosing and treating these infections promptly are of paramount importance to the clinician. Whereas the work-up can be somewhat complicated, recent studies have focused on understanding the patient population and assessing the likelihood of the patient having a UTI. Then an appropriate work-up can be implemented that will reduce the chances of false-positive and negative results that can lead to unnecessary follow-up studies.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (p. 276).
1 Hsiao AL, Chen L, Baker MD. Incidence and predictors of serious bacterial infections among 57- to 180-day-old infants. Pediatrics 2006; 117:1695–1701.
2 Chang SL, Shortliffe LD. Pediatric urinary tract infections. Pediatr Clin North Am 2006; 53:379–400, vi.
3• Vachvanichsanong P. Urinary tract infection: one lingering effect of childhood kidney diseases: review of the literature. J Nephrol 2007; 20:21–28. Recent review of the literature that emphasizes the long-term effects of UTIs in children.
4 Zorc JJ, Kiddoo DA, Shaw KN. Diagnosis and management of pediatric urinary tract infections. Clin Microbiol Rev 2005; 18:417–422.
5 Kass EH. Asymptomatic infections of the urinary tract. Trans Assoc Am Physicians 1956; 69:56–64.
6 Pryles CV. The diagnosis of urinary tract infection. Pediatrics 1960; 26:441–451.
7 Pryles CV, Atkin MD, Morse TS, Welch KJ. Comparative bacteriologic study of urine obtained from children by percutaneous suprapubic aspiration of the bladder and by catheter. Pediatrics 1959; 24:983–991.
8 Pryles CV, Steg NL. Specimens of urine obtained from young girls by catheter versus voiding; a comparative study of bacterial cultures, gram stains and bacterial counts in paired specimens. Pediatrics 1959; 23:441–452.
9 Pryles CV, Luders D, Alkan MK. A comparative study of bacterial cultures and colony counts in paired specimens of urine obtained by catheter versus voiding from normal infants and infants with urinary tract infection. Pediatrics 1961; 27:17–28.
10 Saccharow L, Pryles CV. Further experience with the use of percutaneous suprapubic aspiration of the urinary bladder. Bacteriologic studies in 654 infants and children. Pediatrics 1969; 43:1018–1024.
11• Prajapati BS, Prajapati RB, Patel PS. Advances in management of urinary tract infections. Indian J Pediatr 2008; 75:809–814. Excellent review of the diagnosis and management of UTIs in children.
12 Masters PL. Urinary changes in infections of the urinary tract in childhood. Guys Hosp Rep 1953; 102:76–85.
13 Lau AY, Wong SN, Yip KT, et al
. A comparative study on bacterial cultures of urine samples obtained by clean-void technique versus urethral catheterization. Acta Paediatr 2007; 96:432–436.
14• Bauer R, Kogan BA. New developments in the diagnosis and management of pediatric UTIs. Urol Clin North Am 2008; 35:47–58. Excellent review of UTI in children from the urologic perspective.
15 Hardy JD, Furnell PM, Brumfitt W. Comparison of sterile bag, clean catch and suprapubic aspiration in the diagnosis of urinary infection in early childhood. Br J Urol 1976; 48:279–283.
16 Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Pediatrics 1999; 103:843–852.
17•• Long E, Vince J. Evidence behind the WHO guidelines: hospital care for children: what are appropriate methods of urine collection in UTI? J Trop Pediatr 2007; 53:221–224. Review of the evidence for our practice of collecting urine for culture and analysis for evaluation of UTI. The WHO has specific guidelines for how this should be approached in patient care.
18 Hoberman A, Wald ER, Reynolds EA, et al
. Is urine culture
necessary to rule out urinary tract infection in young febrile children? Pediatr Infect Dis J 1996; 15:304–309.
19 Hoberman A, Wald ER, Reynolds EA, et al
. Pyuria and bacteriuria in urine specimens obtained by catheter from young children with fever. J Pediatr 1994; 124:513–519.
20 Hoberman A, Wald ER, Penchansky L, et al
. Enhanced urinalysis as a screening test for urinary tract infection. Pediatrics 1993; 91:1196–1199.
21 Shaw KN, McGowan KL, Gorelick MH, Schwartz JS. Screening for urinary tract infection in infants in the emergency department: which test is best? Pediatrics 1998; 101:E1.
22 Gorelick MH, Shaw KN. Screening tests for urinary tract infection in children: a meta-analysis. Pediatrics 1999; 104:e54.
23 Huicho L, Campos-Sanchez M, Alamo C. Metaanalysis of urine screening tests for determining the risk of urinary tract infection in children. Pediatr Infect Dis J 2002; 21:1–11, 88.
24• dos Santos JC, Weber LP, Perez LR. Evaluation of urinalysis parameters to predict urinary-tract infection. Braz J Infect Dis 2007; 11:479–481. This study shows the promising use of automated cell counting devices in screening urine for predicting the presence of UTI.
25 Whiting P, Westwood M, Bojke L, et al
. Clinical effectiveness and cost-effectiveness of tests for the diagnosis and investigation of urinary tract infection in children: a systematic review and economic model. Health Technol Assess 2006; 10:iii–xiii, 1.
26 Whiting P, Westwood M, Watt I, et al
. Rapid tests and urine sampling techniques for the diagnosis of urinary tract infection (UTI) in children under five years: a systematic review. BMC Pediatr 2005; 5:4.
27 Novak R, Powell K, Christopher N. Optimal diagnostic testing for urinary tract infection in young children. Pediatr Dev Pathol 2004; 7:226–230.
28•• Shaikh N, Morone NE, Bost JE, Farrell MH. Prevalence of urinary tract infection in childhood: a meta-analysis. Pediatr Infect Dis J 2008; 27:302–308. The approach by these authors was to reexamine the prevalence of UTI in various patient populations. This will then allow the clinician to focus on those patients who have the highest likelihood of having a UTI.
29•• Shaikh N, Morone NE, Lopez J, et al
. Does this child have a urinary tract infection? J Am Med Assoc 2007; 298:2895–2904. The authors apply their findings on the prevalence of UTI to specific patient populations. They have generated algorithms that are excellent guides to the evaluation of patients for potential UTI. The approach of evaluating patients on the basis of prior probabilities of the presence of UTI and working through how pertinent clinical data alter those probabilities is the best way to approach any clinical problem.