Collecting the golden water: Quality assessment on approach of diagnosing urinary tract infections in 0 to 36 months old children

Abstract

Objective

The study aimed to assess current practices of a community hospital for collection of urine sample when diagnosis of urinary tract infection (UTI) is suspected in children aged 0 to 36 months old.

Methods

An analysis of paediatric patients aged 0 to 36 months old was performed in two separate audits to assess the quality of urine sampling. The first, retrospective analysis comprised of urine collections techniques in a community hospital for diagnosis of UTI followed by an education intervention in which the hospital staff was briefed regarding the Canadian Paediatric Society (CPS) position statement for diagnosis and management of UTI. CPS recommendations were transposed using PowerPoint presentations, reminders at unit huddles, and other educational forums. Second audit was a prospective analysis which was conducted 6 months after the educations intervention.

Results

Bagged sampling had higher sensitivity and lower specificity due to sample contamination, versus transurethral bladder catheterization and suprapubic aspiration. The first audit showed that while 66% of culture-positive urine sampling was performed via the bagging, only 26% those positive cultures were repeated before treatment. In the second audit, after educational intervention, 33% of culture-positive urine collection was done via the bagging method and repeat testing was done in 83% of positive results on a bagged sample before initiating treatment. The false-positive rate for the diagnosis of UTIs in the first and second audit was 65.7 and 60%, respectively.

Conclusion

Our study recognizes the flaws in community hospital practices in the diagnosis of UTI in children and validates the significance of educational intervention in improving health care.

Urinary tract infection (UTI) is a common diagnosis in paediatric age groups, especially in the first few years of life. Preverbal children pose a diagnostic challenge as they are unable to verbalize the symptoms of UTIs. In a meta-analysis, 7% of children aged 2 to 24 months who initially presented with no source of infection were diagnosed with a UTI (1). The primary diagnosis in 8% of children aged 2 to 19 years who presented with a fever was a UTI (1). Diagnosis of UTI in females aged 0 to 3 months, 3 to 6 months, 6 to 12 months, and greater than 12 months was 7.5, 5.7, 8.3, and 2.1%, respectively (1). Early recognition of UTI is crucial for treatment of infection, prevention of urosepsis, and subsequent renal pathologies such as hypertension (2) and renal scarring (3). In a systematic review conducted by Shaikh et al., 57% of children with UTIs showed changes of acute pyelonephritis on an acute-phase Dimercaptosuccinic Acid (DMSA) scan (4) after an initial UTI, and 15% displayed evidence of renal scarring at the follow-up DMSA scan (4). In another audit, prompt treatment of UTI resulted in significant reduction in renal scarring (5).

It has always been a challenge to collect reliable urine samples from nontoilet-trained children under 36 months without using invasive procedures. During our literature search, it was evident that the preferences of the sample collection technique vary among countries, health care workers, and hospitals. Bagged urine samples are frequently used in this age group for the reasons mentioned below, but have a high risk of contamination and thus a high false-positive rate (6). National Institute for Health and Care Excellence (NICE) guidelines from the UK, recommends using noninvasive methods of urine sample collection. Parents are asked to sit in a waiting area with a cup to catch the golden water in a nappy-free child spontaneously. Despite this being more effective than other noninvasive methods, in one study, the contamination rate reported as high as 26% for clean catch urine samples (7). In order to minimize the waiting time for a clean catch sample, other methods of clean catch urine sampling were introduced recently. In a randomized controlled trial conducted in Australia, the Quick-Wee method was used, in which a gauze soaked in cold saline was rubbed on the suprapubic area of the child in a circular pattern for 5 minutes or less if the clean catch sample was obtained, without any significant difference in contamination rates with standard clean catch sample (8). Another randomized control trial done in Europe, they applied simultaneous use of bladder stimulation and lumbar paravertebral massage as an effective method of collecting clean catch urine (9). All of these methods are associated with false-positive results. Thus, the Canadian Paediatric Society and the American Academy of Paediatrics reserve the use of noninvasive sampling for screening purposes only. For confirmatory diagnosis and continuation of antibiotic treatment, invasive methods like transurethral bladder catheterization (TUBC) or suprapubic aspiration (SPA) are recommended (10–12).

In Canada, the Canadian Paediatric Society’s (CPS) recommendations should be considered a standard of care (12,13). CPS first published its position statement regarding the diagnosis and management of UTI in children between 2 and 36 months of age in 2014 and reaffirmed it in 2017. The statement affirms that any toxic appearing child aged less than 3 years with a fever (39.0°C oral/38.5°C rectal) of unknown origin should have a urine sample obtained for urinalysis and culture. Unless the test is completely normal, urine samples should be sent for culture via a TUBC or SPA. In cases of a possible UTI, when antibiotics are started, reassessment of the diagnoses should be made after the availability of results. Antibiotics should be stopped if UTI seems unlikely (12,13).

TUBC and SPA are the only reliable methods for urine culture in nontoilet-trained children. Bagged urine samples should be restricted to screening purpose only. While a negative bag culture result has a high negative predictive value, a positive result is not predictable due to its susceptibility to contamination and requires repeat testing of the sample obtained via TUBC or SPA.

With the above context, we decided to perform an audit to assess our local hospital adherence to CPS recommendations. After identification of aberrant practices for urine sample collection in the first audit, an educational intervention was done followed by the second audit to close the loop.

METHODS

This study was performed to assess the quality of urine sample collection in paediatric patients in a community setting and to compare it to CPS recommendation. Approval from the Ethical review board of Scarborough Health Network, Canada was obtained before the study was initiated.

Eligibility criteria

Male or female paediatric patients up to 36 months old tested for suspected or proven UTI in the Emergency Department, Paediatric ward or neonatal intensive care unit (NICU).

First audit

Data were extracted from the electronic medical records of the hospital considering the eligibility criteria as defined above. All urine culture reports were reviewed that has been sent prior to the initiation of antibiotics. The method of urine collection was recorded for all the positive urine culture specimen. Colony count of ≥5 × 10⁴ colony-forming unit (CFU)/mL for catheter sample and ≥10⁵ CFU/mL for bagged urine sample and midstream urine (MSU) were considered as positive urine culture for UTI. Additionally, it was determined whether urine cultures were repeated in case of a positive bagged sample. The chart review was performed in July 2015 through December 2015.

Educational intervention

Findings from the first audit led us to conduct an educational intervention, which highlighted the flaws in the current hospital practices and steps to rectify the problems. Results of the first audit were shared by one of the authors with paediatricians and nurses at the departmental rounds, via PowerPoint presentation. CPS guidelines were part of this presentation alongside the proposed changes to calibrate hospital practices toward Canadian standards with respect to the diagnosis of UTI. To further bolster the awareness, and mend for missed attendance, these proposed changes were e-mailed separately to all paediatricians, emergency department physicians and nurses, paediatric and NICU staff. The proposed changes encouraged the use of TUBC or SPA as the primary method of urine collection when UTI was suspected. This informnation was periodically dispersed to hospital staff, including clinicians and nurses over a span of 5 months, using various approaches including person to person contact, group discussions/huddles, and hospital rounds.

Second audit

After the educational intervention, prospective analysis of the second audit was performed from July through December 2016, considering the same eligibility criteria. Data were reviewed and analyzed after 6 months.

Measures

Percentages of the different urine collection methods were calculated. The true/false positive and negative rates were calculated for the UTI culture results collected in the first and second audits. Initial positive bagged urine culture with a negative catheter specimen on repeat culture was considered as false positive, whereas the positive catheter culture on repeat was counted as true positive.

RESULTS

Table 1 demonstrates demographic details for the first and second audit.

Table 1.

Demographics of the patients sampled for the first and second cycle of the audit

	First cycle of audit	Second cycle of audit
Sex
Male	67 (69%)	17 (40%)
Female	30 (31%)	25 (60%)
Location
Neonatal Intensive Care Unit	18 (19%)	14 (33%)
Emergency Department	42 (43%)	17 (40%)
In-patient Paediatrics Ward	37 (38%)	11 (26%)
Total	97	42

In the first audit, out of 97 patients, 36% tested positive for UTI by urine culture. Of the positive results, only 31% underwent urine collection via TUBC, whereas 66% of the urine collection was performed via the bagging method and for 3% source was not identified. Repeat testing was done in only 26% of perineal bag samples after a positive culture. Out of those repeated, 33% tested positive for UTI (true positive). Therefore, approximately 67% of the repeats were negative (false positives). Figure 1 illustrates the breakdown of urine testing for susceptible UTIs in the paediatric population during the first audit. In the second cycle of the audit, out of 42 patients, 43% tested culture positive for UTI. Among the positive results, in–out catheterization was the primary method of urine collection. Sixty-seven per cent (67%) underwent urine collection via TUBC, while 33% of the urine samples were collected using the perineal bagging method. Furthermore, 83% of the patients who underwent UTI screening via the bagging method were given a repeat test via the TUBC technique. The repeat test rendered a 60% false-positive rate for UTI screening using the perineal bagged sample method. Figure 2 illustrates the breakdown of urine testing for susceptible UTIs in the paediatric population during the first audit. True positive percent for the diagnosis of UTIs was 33 and 40% for the first and second audit, respectively.

Figure 1.

Breakdown of urine testing methods for susceptible urinary tract infections (UTIs) in the paediatric population (<3 years of age) during the first audit, July–December 2015.

Figure 2.

Breakdown of urine testing for susceptible urinary tract infections (UTIs) in the paediatric population (<3 years of age) during the second audit, July–December 2016.

Table 2 demonstrates the comparison between the first and second audit.

Table 2.

Comparison of the results obtained from the first and second cycle of the audit

	First cycle of audit	Second cycle of audit
Total number of positive urine culture	35	18
Perineal bagging method	23 (65.7%)	6 (33.3%)
Catheterization method	11 (31.4%)	12 (66.6%)
Repeat testing of positive culture bagged sample	6 (26%)	5 (83.3%)
False-positive UTI*	4 (66.6%)	3 (60 %)
True positive UTI**	2 (33.3%)	2 (40%))

*False-positive results on repeat testing of the positive cultured bagged sample via catheter technique.

**True positive results on repeat testing of the positive cultured bagged sample via catheter technique.

DISCUSSION

This audit was done to assess the quality and accuracy of diagnosis of UTI in paediatric patients between the age of 0 to 36 months, in a community hospital setting. Since our study was primarily focused on the technique of urine sample collection, we included all infants younger than 36 months of age, including newborns. We compared the prevalent practice of urine sample collection in our hospital and correlated it with contemporary CPS position statement on the diagnosis and management of UTI in infants and children. In the first part of the study, we found perineal bagging as the most widely employed method for collection of a urine sample. Finding from our first audit led us to carry out an educational intervention as described above.

We conducted a second audit 6 months after the completion of our first audit in order to determine the effect of our educational intervention on the prevalent practice of urine sample collection. Findings from our second audit were encouraging and affirming of the educational intervention. As evident from the results retrieved from the first and second audit, we found that the use of the perineal bagging method had reduced to 32.4% in comparison to the first audit (66 versus 33%). Similarly, an increment of 35% in repeat testing for the culture-positive perineal bag sample was observed in the second audit (31 versus 67%).

There are four accepted methods of urine sample collection in nontoilet-trained children less than 3 years of age. These include perineal bagging; clean catch (MSU), TUBC, and SPA. Despite its limitations, perineal bagging is still used widely due to the fact that its fast, noninvasive, and a convenient method for urine sample collection for nontoilet-trained patients. The limitation of perineal bagging relates to its lack of reliability (specificity of only 70%) for the accurate diagnosis of UTI. This is mostly due to a high rate of contamination (up to 63%) (14). In a study by Selekman et al., parental refusal and difficulty with catheterization were the main reasons why the urine sample was collected by a perineal bag (15). A positive bagged sample culture is unreliable and requires repeat testing by TUBC or SPA as recommended by CPS and evident from our first and second audit with a false-positive rate of 67 and 60%, respectively. Midstream is another noninvasive and convenient but a time-consuming method of urine sample collection. The author has work experience in the UK, where there is a practice of collecting MSU sample for screening and diagnosis of UTI in children.

This was the first study of its kind which compared the techniques of urine sample collection for diagnosis of UTI in children with established guidelines. The study also underscores the significance of ongoing education and its impact on health care outcomes. There were some limitations to the study. There was a discrepancy in sample size between the audits. Similar sample size may have caused variations in the results of the audits. The number of charts reviewed was too small in both audits to make any meaningful conclusion. A similar but larger study is needed to validate our findings. It is also important to consider that the CPS guidelines used here as the benchmark are defined for children between the ages of 2 to 36 months. Granting our primary aim was to assess the technique of urine sample collection, this study used these guidelines for children between the ages of 0 to 36 months. The workup of febrile infants younger than 3 months old includes evaluation of UTI as part of the septic workup. Albeit symptomatology and diagnostic criteria for less than 3 months, old children differ but the preferable urine sampling techniques are the same for both age groups. It is conceivable that the educational intervention may have led to more judicious assessment, urine sampling and hence a smaller sample size on the second audit. To our dismay, despite educational intervention, we were not able to fully alter the long-standing practice of bagging for urine sample collection.

There is no denying that adherence to the standard of care practices improve outcome. Our study results endorsed the CPS recommendations for urine sample collection methods under 36 months old. In order to improve and implement standard practices, we removed the urine bags from our emergency unit and also omitted the option of bagging method from the hospital electronic medical record. Additionally, we want to incorporate aspects of feedback from our clinical staff to determine the current limitations towards adhering to the CPS guidelines. Input from staff should be explored to identify any potential barriers in utilizing urinary catheters. Also, future training sessions regarding the guidelines for the management of UTIs may be of significant benefit.

We understand that this audit was confined to catching the golden water only. Future studies will be planned to assess the quality of diagnostic measures and treatment options for UTI as well.

Funding: There are no funders to report for this submission.

Potential Conflicts of Interest: All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Institution where study conducted: Scarborough Health Network, Toronto, ON.