Practice variation in urine collection methods among pre-toilet trained children with suspected urinary tract infection: a systematic review

Abstract

Background

Urinary tract infections (UTIs) are a common cause of acute illness among infants and young children. There are numerous methods for collecting urine in children who are not toilet trained. This review examined practice variation in the urine collection methods for diagnosing UTI in non-toilet-trained children.

Methods

A systematic review was completed by searching MEDLINE (Ovid), Embase (Ovid), CENTRAL (Ovid), PsycInfo (Ovid), CINAHL (EBSCO), and JBI (Ovid) from January 1, 2000 until October 9, 2021 and updated on May 24, 2023. Studies were included if they were conducted in an acute care facility, examined pre-toilet trained children, and compared one urine collection method with another for relevant health care outcomes (such as length of stay in an ED, or re-visits or readmissions to the ED) or provider satisfaction. Two independent reviewers screened the identified articles independently, and those included in the final analysis were assessed for quality and bias using the Newcastle-Ottawa Scale.

Results

Overall, 2535 articles were reviewed and 8 studies with a total of 728 children were included in the final analysis. Seven studies investigated the primary outcome of interest, practice variation in urine collection methods to diagnose a UTI. The seven studies that investigated novel methods of urine collection concluded that there were improved health care outcomes compared to conventional methods. Novel methods include emerging methods that are not captured yet captured in clinical practice guidelines including the use of ultrasound guidance to aid existing techniques. Three studies which investigated healthcare provider satisfaction found preference to novel methods of urine collection.

Conclusions

There is significant practice variation in the urine collection methods within and between countries. Further research is needed to better examine practice variation among clinicians and adherence to national organizations and societies guidelines. PROSPERO registration number CRD42021267754.

Background

Urinary tract infections (UTIs) are a common cause of acute illness among infants and young children, with an estimated prevalence of around 7% in children under two years of age [1]. Many of these children present to the Emergency Department (ED) acutely unwell with an unexplained fever and non-specific clinical symptoms.

In cases of unexplained fever in a young child a diagnostic workup will include assessing for a UTI. To diagnose a UTI, a urine specimen must be collected from the child [2]. For children who are not toilet trained there is a high risk of contamination [3]. Thus, it is important to choose an appropriate urine collection method. Possible urine collection methods include bagged urine sample, clean catch specimens, pad sampling, catheterization, and suprapubic aspiration [4]. Catheterization and suprapubic aspiration are invasive methods but yield the lowest contamination rates. Bagged urine collection and clean catch specimens and pad sampling are non-invasive and thus are less painful for children but have higher rates of contamination [4]. Recently, there has been the emergence of new techniques aimed to improve urine collection effectiveness and patient experience. Methods include the Quick-Wee method, in which gauze soaked in cold saline is placed over the suprapubic area to stimulate voiding, and the bladder stimulation technique which involves gently tapping over the suprapubic area, followed by a lumbar massage, and repeating these manoeuvres until the child voids [5, 6]. An emerging invasive technique is adding point-of-care ultrasound to visualize the bladder before attempting catheterization [7].

Given the numerous methods available to collect a urine sample, it is speculated that there is a wide variation among emergency physicians about the appropriate urine collection method. This variation is also found in the recommendations made by various national health organizations and societies [2, 8]. This study aims to examine the presence of practice variation in the urine collection methods for diagnosing UTI in non-toilet-trained children and its effects on healthcare outcomes and utilization. The secondary objectives are [1] to characterize practice variation in urine collection methods among health care providers and [2] to determine practice compliance and healthcare providers’ satisfaction of urine collection methods with the local clinical practice guidelines.

Methods

The systematic review was reported in accordance with the PRISMA 2020 statement [9], and the protocol was registered with PROSPERO in August 2021 (registration number CRD42021267754).

Search strategy and study selection

A systematic search of the literature was completed to identify potentially relevant studies. An experienced health sciences librarian (M.L.) designed and executed the search strategy, using a combination of subject terms and keywords that were later translated for each database. The MEDLINE search was peer-reviewed by an independent health sciences librarian as per the PRESS guidelines [10]. Searches were performed in MEDLINE (Ovid), Embase (Ovid), CENTRAL (Ovid), PsycInfo (Ovid), CINAHL (EBSCO), JBI (Ovid), and Google Scholar from January 1, 2000 until October 9, 2021 and an updated search strategy was performed on May 24, 2023. The search was limited to 2000 onwards to ensure that any results or findings were relevant to current guidelines. The searches were designed to be broad, and no restrictions were used. Identified studies were first deduplicated using the Ovid de-duplication and then were deduplicated in EndNote X20 before being uploaded to Covidence. Clinical trial registry searches were conducted in ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform for any ongoing or upcoming trials. Our Ovid multi-file search strategy is available in Appendix A and all search strategies are available at https://0-doi-org.brum.beds.ac.uk/10.34990/FK2/IZ3M32.

We included studies if they met all of the following criteria: (1) conducted in an acute care facility that care for children, including EDs or Urgent Care Centres, (2) used at least one urine collection method to diagnose UTI, (3) compared one urine collection method with another urine collection method, (4) included one of the relevant outcomes, such as, healthcare outcomes (including length of stay in an ED, re-visits or readmissions to the ED), or healthcare utilization (such as ambulance transfers, interfacility/inter-ED transfers, potentially avoidable transfers), (5) observational studies, including cohort and cross-sectional studies, or controlled-clinical study, and (6) were published in the English language. Studies were excluded if the outcome was not relevant, did not include pre-toilet trained patients (defined as age ≤ 3 years), or if the setting was outside of an acute care facility.

The articles identified in the literature search were first screened by title and abstracts using Covidence systematic review software for inclusion in the systematic review by independent reviewers (LMW, CT, MA, VKWL, BO) [11]. The independent reviewers then reviewed the full-length manuscripts for inclusion in the final analysis. Disagreements during screening were resolved by discussion between reviewers or in consultation with a third reviewer (AA).

Data extraction

The data from the included studies was extracted by two independent reviewers (LMW, CT, BO). Reviewers used a customized data extraction tool to identify key characteristics of the articles, including information on study design, objectives, population, intervention, outcomes, and conclusion details. The tool was used to pilot test five studies after which it was adopted for the entire included studies. A third reviewer (AA) examined the data to ensure accuracy and identify any errors when appropriate.

Risk of bias assessment

The included articles were assessed for quality and bias using the Newcastle-Ottawa Scale (NOS) [12], a validated critical appraisal checklist for nonrandomised observational studies. A modified version of the NOS for cohort studies was used to assess the cross-sectional studies. We substituted the term “cohort” with “sample” in the selection domain. We removed questions 2 and 3 concerning follow-up and introduced a question that evaluates the statistical tests conducted in the outcome domain. The NOS rates articles on a star system in order to evaluate the selection of study groups, comparability of groups, and ascertainment of exposure or outcome of interest [12]. Two reviewers (LMW, CT, BO) independently completed the risk of bias assessment, and disagreements were resolved by a third reviewer (AA).

Data analysis and synthesis

Data were collected and managed using Excel and Covidence [11, 13]. Individual article characteristics were summarized and presented in tabular form, and the results were compared based on the primary and secondary objectives.

Results

Search results

The search and study screening were conducted initially in October 2021. The initial systematic search of the databases identified 3400 articles, and the updated search identified an additional 328 articles. After duplicates articles were excluded, 2535 titles and abstracts were reviewed, 65 full text articles screened, and 8 studies were included in the final analysis. Full details are presented in Fig. 1, the PRISMA diagram.

Fig. 1

PRISMA flow diagrams of articles identified on initial screening, and updated in October 2022 and included in the final analysis

Characteristics of included studies

Characteristics of studies included in the analysis are presented in Table 1. Of the 8 included studies, 2 were conducted in the United States [6, 14], 1 in Turkey [7], 1 in Israel [15], and 4 in Australia [5, 16,17,18]. All studies were completed in EDs. The American studies by Baumann et al., and Ravichandran et al. compared ultrasound guided catheterization [14], and bladder stimulation technique to conventional catheterization [6, 7], respectively. The study by Akca Caglar et al. was conducted in Turkey and compared point-of-care ultrasound-guided catheterization to conventional catheterization [7]. Kozer et al. investigated transurethral catheterization to suprapubic aspiration in Israel [15]. Finally, the Australian studies by Ho et al., Kaufmann et al. (2017), Kaufman et al. (2019) and Lennon et al. compared a variety of urine collection methods. Ho et al., compared urine collection pads to clean catch urine specimens [16]. Kaufman et al. (2017) compared the Quick-wee method to clean catch urine specimens [5], whereas Kaufman et al. (2019) compared the Quick-wee method to clean catch urine, urine collection bags, catheterization, and suprapubic aspiration [17]. Lennon et al. compared point-of-care ultrasound to assess bladder volume and stimulate the micturition reflex to traditional clean catch urine collection [18].

Table 1 Study characteristics

Overall, five studies investigated the use of catheterization for urine collection as compared to other urine collection methods. Two studies sought to explore the role of ultrasound-guided catheterization as a means of urine collection method. Four studies sought to investigate novel non-invasive urine collection methods as compared to conventional techniques. Of these studies, three focused solely on non-invasive collection methods, and they were all conducted in Australia.

Outcomes

Primary outcome: practice variation and healthcare outcomes and utilization

Eight studies were included in the final analysis. Of the studies that examined health care outcomes and utilization, no studies looked at outcomes of interest as defined in the methods section.

In total all eight studies investigated the outcome of interest of practice variation in urine collection method [5,6,7, 14,15,16,17]. Full details are presented in Table 2. Every included study adhered to the local clinical guidelines cited in the paper. Of these studies, five cited the American Academy of Pediatrics [2] as the clinical practice guidelines, which recommends invasive techniques including bladder catheterization or suprapubic aspiration as first line collection method [2]. Two studies referenced the UK NICE Guidelines [8], which recommends clean catch urine and non-invasive methods as first line collection method [8]. One study referenced American and British recommendations that were not the American Academy of Pediatrics, or the UK NICE Guidelines, respectively [19, 20]. They also cited Australian clinical practice guidelines [21, 22]. The final study did not report a clinical practice guideline but was conducted in the United Kingdom. The study listed clean catch urine as their recommended method which adheres to the UK NICE guidelines.

Table 2 Included studies and the clinical practice guidelines on urine collection methods

Seven studies included the primary outcome of interest, practice variation in urine collection methods and its effects on healthcare outcomes and utilization [5,6,7, 15,16,17,18]. Table 3 incudes the outcomes of interest. The study by Akca Calgar et al. compared the success rate of point-of-care ultrasound-guided catheterization versus conventional catheterization. The study found a statistically significant difference (p = 0.03) between the success rates of ultrasound-guided catheterization (93%) and conventional catheterization (78%) [7]. The study did not find a significant difference between success rates stratified by patient sex.

Table 3 Primary outcome: healthcare outcomes and utilization

Four studies compared clean catch urine samples to alternative urine collection methods. Ho et al. compared the time needed to collect a urine sample between the clean catch urine method to the pad sampling technique. The study found that the pad sampling was statistically significantly faster than clean catch urine collection (30 min [10-1135] vs. 107.5 min [30–330]; p < 0.002) [16].

Kaufman et al. (2017) investigated the voiding time and success rate of clean catch urine sampling to the Quick-Wee method. The study found that the Quick-Wee method decreased the five-minute voiding time (mean difference 19.0, 95% CI 11–28) and increased the success rate of urine collection (mean difference 21.0, 95% CI 13–29) [5]. It adjusted for age and sex.

Kaufman et al. (2019) compared clean catch urine collection to four other methods – urine bags, the Quick-Wee method, catheterization, and suprapubic aspiration. The study investigated the cost-effectiveness of urine collection methods and measured the time to collect a urine sample, and the success rate of each method as part of its cost-effectiveness study. Catheterization was found to be the most cost-effective (£25.98), followed by suprapubic aspiration (£37.80), voiding simulation (£41.32), clean catch (£52.84), and urine bag (£92.60). Its model estimated that the Quick-Wee was the quickest voiding time (5 min), followed by suprapubic aspiration (8 min ± 4 min), catheterization (12 min ± 7 min), clean catch (31 min ± 42 min), and finally urine bag (85 min ± 67 min). The highest success rate was urine bag (96% ± 48%), followed by catheterization (90% ± 47%), clean catch urine (64% ± 45%), suprapubic aspiration (44% ± 22%), and finally Quick-Wee (30% ± 47%) [17].

Lennon et al. examined the time to obtain a clean catch urine sample. The study compared using point-of-care ultrasound to measure bladder volume and stimulate the micturition reflex prior to initiating urine collection to standard clean catch urine collection methods. The study found that ultrasound assistance had a statistically significant reduction in the mean (52 min ± 42 min) and median (40 min, IQR 52 min) time to collection compared to standard clean catch urine practices (mean 82 min ± 90 min; median 55 min, IQR 81 min; p = 0.038) [18].

Kozer et al. measured the difference in neonatal pain and duration of cry between transurethral catheterization and suprapubic aspiration. Pain was rated using two independent measures, one by parents and nurses, the other by investigators on the research team. Suprapubic aspiration was found to be more painful (mean difference 2.5, 95% CI 1.4–3.7; 19.6 95% CI 7.4–31.8) and have a longer duration of cry than catheterization (13.2 Sect. 95% CI -4.3 to 30.7 s) [15]. The study adjusted for age and use of analgesia.

The final study by Ravichandran et al. compared the voiding time within 300 s between bladder stimulation to conventional catheterization. There was a statistically significant difference between bladder stimulation (median 73 s, IQR 125 s), with 38% success rate and catheterization (median 9.5 s, IQR 17) with success rate (77%). The study adjusted for a number of factors [6].

Secondary outcome: practice variation in urine collection method

The review did not identify any papers that assessed physician compliance to the local guidelines. However, we used authors recommendations on urine collection method highlighted in Table 2 as an indirect marker of physician adherence. All eight studies included this marker, and this was discussed as part of the primary outcome of interest.

Secondary outcome: healthcare professional satisfaction

Three studies were included in the final secondary outcome, practice compliance and satisfaction of healthcare provider with various urine collection methods (Table 4) [5, 6, 14]. Baumann et al. investigated caregiver, nurse, and physician satisfaction using standardized, seven-point Likert scale questionnaires comparing ultrasound-guided catheterization and conventional catheterization. The study found that both caregivers and healthcare providers had greater satisfaction with ultrasound-guided catheterization (nurses: 3.0, 95% CI 2.5–3.5; physicians: 4.3, 95% CI 3.7–4.9) compared to conventional catheterization (nurses: 5.5, 95% CI 5.1–6.0; physicians: 5.7, 95% CI 5.2–6.1) and would prefer this modality with future urine collection attempts [14].

Table 4 Secondary outcome: healthcare professional satisfaction

Kaufman et al. (2017) looked at parental and clinical satisfaction between the Quick-Wee method and clean catch urine using a five-point Likert scale survey. The Quick-Wee method was preferred over clean catch urine by both parents and clinicians (mean difference 1.0, 95% CI 0.6–1.4) [5].

The final study by Ravichandran et al. investigated provider satisfaction between the bladder stimulation method and catheterization. The study used a five-point Likert scale to elicit provider’s perspectives on the two collection methods. It found that clinicians felt that compared to catheterization, the bladder stimulation technique was effective (p < 0.001), easy to perform (p < 0.001), well tolerated by patients (p < 0.001), and had high parental satisfaction (p = 0.002) [6].

Risk of bias across studies

The NOS [12] was used to evaluate the included studies. The results of the assessment are presented in Table 5. All 8 studies were rated as having a low risk of bias in the areas of representativeness of exposed sample, definition of controls, method of ascertainment for cases and controls, comparability bias, and follow-up bias. Four of the eight studies were deemed high risk for selection bias relating to the ascertainment of the exposure. One study was deemed high risk for its non-response rate. Finally, two studies were evaluated as high risk for assessment bias.

Table 5 Secondary outcome: healthcare professional satisfaction

Discussion

The results of this systematic review have shown that there was variation in the practice of urine collection methods within and between countries. When examining the adherence to the recommendations of national pediatric associations and societies all eight studies adhered to the guidelines cited, however, only Baumann et al., Lennon et al., and Ravichandran et al. cited the recommended guidelines of the country the study was conducted in [6, 14, 18]. Of those that referenced guidelines from other countries, only two acknowledged the preferred collection method of their home country. Ho et al. stated that a recent Australasian study found that clean catch urine was the preferred method across 13 study sites, which is in keeping with the conclusions in their paper [16, 23]. Lennon et al. cited two Australian clinical practice guidelines. The first was by Kidney Health Australia, which recommends clean catch urine, midstream urine, or catheterization as standard urine collection methods [22]. The other reference was a 2005, now rescinded, policy from the New South Wales government in Australia, which recommended clean catch urine samples in cases of children presenting with fevers of unknown origin [21]. The remaining four studies did not reference their own national guidelines, but instead referenced the guideline that fits their research question.

While all eight studies used methods in their studies that adhered to the local guidelines, the authors conclusions regarding a recommended collection method did not match the current guidelines in seven of the studies. The recommendation suggested by Kozer et al. to use catheterization as the preferred urine collection methods aligned with the American Academy of Pediatrics [15]. Of the remaining seven studies, six of them aligned with the clinical practice guidelines on whether invasive or non-invasive methods were preferred. However, the recommended means of urine collection was different than what their society recommended. The remaining study, by Ravichanran et al. concluded that bladder stimulation technique is preferred over suprapubic aspiration in infants admitted to the NICU, which differs from the American Academy of Pediatrics recommendation of performing catheterization or suprapubic aspiration on pre-toilet trained children [6].

The results of this systematic review suggest that there are improved healthcare outcomes and utilization when comparing novel urine collection methods to those recommended by national pediatric health organizations and societies. Seven of the studies included in the primary outcome compared first line urine collection techniques to novel methods and reported improved health outcomes using novel methods [5,6,7, 15,16,17,18]. The studies reported that urine bags and clean catch urine collection was slower and less cost-effective than novel methods such as the Quick-Wee method or bladder stimulation technique, and to invasive methods including catheterization and suprapubic aspiration. Only one study compared a non-invasive method (bladder stimulation) to an invasive method (catheterization), and while catheterization had a statistically significantly higher success rate and voiding time, caregivers and heath care providers indicated they were more satisfied with the non-invasive method [6]. The findings suggest that incorporating new evidence and skills such as ultrasound guided catheterization into existing guidelines can improve both the success rate of urine collection and improve patient outcomes [5,6,7, 15,16,17,18].

Finally, our review investigated healthcare provider satisfaction with urine collection methods. Three studies compared novel techniques and methods to first line recommendations. Clinical satisfaction was higher with the novel method compared to the current practice. Kaufman et al. collected clinical satisfaction using a Likert scale but did not explore reasons for preferring the Quick-Wee method over clean catch urine [5]. Baumann et al. and Ravichandran et al. explored a number of reasons why health care providers preferred their chosen methods, including aspects of the procedure itself as well as patient-related factors [6, 14].

Overall, there are several strengths and limitations to this study. First, the research team included an experienced health science librarian who developed the search strategy. This resulted in a rigorous search process. Limitations of the study included that there were relatively few studies identified during the systematic review despite urine collection being a common procedure in young children. Of those studies those identified, there was a wide range of outcomes. This meant that we were unable to perform a meta-analysis and investigate the strength of association within these studies, limiting out ability to draw strong conclusions about current urine collection method practices.

Future research is needed to continue to explore urine collection methods for pre-toilet trained children presenting to the EDs with signs and symptoms of a UTI. This review identified that there was limited research into the actual practice variations among clinicians. Future work is needed to determine if physicians adhere to the guidelines recommended by national pediatric societies and organizations. The novel methods identified have the possibility to help inform clinical practice guidelines and allow for improved outcomes in collecting sterile urine samples. However, randomized controlled trials are needed to confirm or refute if the identified novel urine collection methods are associated with better health care outcomes.

Conclusions

This study demonstrates that there is currently significant practice variation in the urine collection methods choice within and between countries. This study highlights the importance of future research needed to better examine practice variation among clinicians and adherence to national organizations and societies guidelines. Novel methods have increasing utility in practice but are not yet integrated into standard of care. The results identify that additional research can help identify methods that have positive clinical and patient outcomes and integrate them into the guidelines of pediatric societies and organizations.