A large prospective study has explored the diagnostic features of urinary tract infections in young children and produced a two-step rule for decision making
Undiagnosed urinary tract infections (UTIs) in young children lead not only to pain and other symptoms, but also possibly to long-term health problems. Half the children with UTIs presenting to primary care in the UK are not diagnosed at first presentation. The prospective Diagnosis of Urinary Tract infection in Young children (DUTY) study recruited 7,163 systemically unwell children presenting to primary care. Urine samples were collected for 6,390 of them and a wide range of symptoms and signs recorded to identify which, if any, could predict UTI. This article briefly describes the study, which has been reported in full elsewhere, and highlights useful clinical rules that primary care doctors and nurses can use to discriminate between UTIs and other causes of constitutional illness, helping them decide when to take a urine sample and when to prescribe antibiotics.
Citation: Harman K, Rumsby K (2017) Diagnosing urinary tract infections in children: research-based rules to aid decision making in primary care. Nursing Times [online]; 113: 5; 42-45.
Authors: Kim Harman and Kate Rumsby are clinical trial managers, Primary Care and Population Sciences, Faculty of Medicine, University of Southampton.
- This article has been double-blind peer reviewed
- Scroll down to read the article or download a print-friendly PDF here
Half of the children with urinary tract infections (UTIs) who present to primary care in the UK are not diagnosed at their first presentation (Coulthard et al, 2003). The team behind the Diagnosis of Urinary Tract infection in Young children (DUTY) study (Hay et al, 2016) thinks there are three possible explanations for this:
- The symptoms and signs of UTIs are very often non-specific, especially in young children;
- There is a lack of evidence about the children in whom primary care practitioners should suspect a UTI;
- Obtaining urine samples that are uncontaminated can be both challenging and time consuming, particularly in children who are very young.
DUTY was a large prospective cohort study designed to investigate the diagnostic features of UTIs in children below the age of five presenting to primary care. The aim was to develop a two-step clinical rule:
- Step 1 – using symptoms and signs, select in which children who are acutely unwell urine should be collected;
- Step 2 – once urine has been obtained, add dipstick test results to the clinical picture to guide pragmatic antibiotic treatment.
Accurate and timely diagnosis of UTIs in children is important to alleviate pain and symptoms in the short term (Butler et al, 2015), as well as to prevent any possible long-term consequences such as:
- Renal scarring;
- Impaired renal growth;
- Recurrent pyelonephritis;
- Impaired glomerular function;
- End-stage renal disease;
- Pre-eclampsia (Coulthard et al, 2014; Subcommittee on Urinary Tract Infection et al, 2011; National Institute for Health and Care Excellence, 2007; Farnham et al, 2005; Jacobson et al, 1989).
Accurately diagnosing UTIs is also important to prescribe appropriate antibiotic treatment and avoid investigations that can be burdensome, unpleasant and expensive (NICE, 2007).
Most of the long-term consequences of UTIs mentioned above are rarely seen in primary care and have only been identified in children whose illness has necessitated secondary care intervention. Similarly, UTI management guidelines, which emphasise the importance of fever as a symptom indicative of UTI (Royal Children’s Hospital Melbourne, 2015; Ammenti et al 2012; Subcommittee on Urinary Tract Infection et al, 2011), are largely informed by studies conducted in emergency departments and are often not applicable to primary care (Gorelick et al, 2003; Gorelick and Shaw, 2000; Shaw et al, 1998).
The DUTY study recruited acutely unwell children having a first contact with the NHS for this episode and presenting to any NHS primary care site or children’s accident and emergency department in the UK. The first contact could be with a doctor or a nurse working in general practice, at a walk-in centre or at a children’s emergency department. Children were eligible if they presented with any acute illness that had lasted for <28 days and was associated with:
- At least one constitutional symptom or sign identified by NICE as a marker of potential UTI – that is, fever, vomiting, lethargy/malaise, irritability, poor feeding and failure to thrive;
- At least one urinary symptom identified by NICE as a marker of potential UTI – that is, abdominal pain, jaundice (only for children under three months of age), haematuria, offensive urine, cloudy urine, loin pain, frequency, apparent pain on passing urine, and changes to continence (NICE, 2007).
As a result, any children who are constitutionally unwell – even if presenting with an apparently obvious cause for their symptoms (such as acute otitis media or bronchiolitis) – were included as long as they filled the above criteria.
Children were excluded if:
- They were constitutionally well and had no urinary symptoms (for example, if they presented with conditions such as acute conjunctivitis);
- They had a neurogenic or surgically reconstructed bladder or used a urinary catheter;
- Trauma was the main presenting problem;
- They had taken antibiotics in the previous week.
Data and urine collection
After explaining the study to, and obtaining written consent from, the child’s parent or legal guardian, and before obtaining urine, 107 data points were recorded on a standardised case report form. Parent-reported items included the child’s medical history and symptoms. Clinician-reported items, obtained from a full clinical examination, included the clinician’s impression of global illness severity (scored from 0 to 10, 0 being child completely well and 10 being child extremely unwell) and abdominal tenderness. Clinicians then recorded their clinical diagnosis and their level of certainty about that diagnosis, before and after receiving the urine dipstick test results.
Guidelines universally recommend urine sampling for microbiological confirmation of a UTI. In Europe, this is by ‘clean catch’, whereby the urine is collected into a sterile container rather than from a pad. This was our preferred urine collection method.
For toilet-trained children, we used a sterile bowl that the parent could hold under the child or put in a potty. For non-toilet-trained children, the parent cleaned the nappy (diaper) area using water alone and sat the child on their knee with the bowl placed under the child’s perineum. If it was not possible to obtain a sample at the site, the parent was given equipment and advice to take the sample at home. When clean catch was not feasible, we used Newcastle nappy pads; these are sterile pads placed inside the nappy and have been recommended by NICE (2007).
Urine testing and data analysis
Urine samples were tested at the site using Siemens/Bayer Multistix 8SG dipsticks and then forwarded to the usual NHS laboratory. Whenever sufficient urine had been obtained, a further sample was posted in a boric acid tube to the research laboratory at Cardiff University.
We examined the frequency of symptom and sign categories, blind to their associations with urine culture results, and merged the least frequent categories before analysis. We used logistic regression to estimate associations of data points with urine culture positivity.
Of 14,724 children assessed for eligibility, 7,561 (51.4%) were ineligible or not enrolled (for reasons such as having left before invitation, lack of or invalid consent, language barrier and poor data quality), leaving 7,163 (48.6%) recruited children with valid consent and index test data. Urine samples were obtained for 6,390 (89.2%) of them. Among these 6,390 children, reference standard results (both NHS laboratory results and research laboratory results) were available for 5,017 children: 2,740 (54.6%) used the clean catch method and 2,277 (45.4%) used the Newcastle nappy pad method.
The DUTY findings indicate that half the laboratory-confirmed UTIs had not been suspected at first contact in primary care. Nappy pad samples were six times more likely to show contamination than clean catch samples; because of this contamination rate, the former produced less reliable results than the latter and so the two data sets were analysed separately.
Clean catch urine samples
Among the 2,740 children in whom urine samples had been obtained using the clean catch method, 60 (2.2%) met our research laboratory definition of UTI (Box 1):
- n = 50 (83.3%) – Escherichia coli;
- n = 5 (8.3%) – Proteus species;
- n = 3 (5.0%) – Klebsiella species;
- n = 1 ( 1.7%) – Morganella morganii;
- n = 1 (1.7%) – Citrobacter farmeri.
Box 1. Definition of UTI
This study used a microbiological definition of UTI:
- ≥105 CFU/ml of a single uropathogen (‘pure growth’)
- ≥105 CFU/ml of a uropathogen with ≥3 log10 (1000-fold) difference between the growth of this and the next species (‘predominant growth’)
We defined uropathogens as members of the Enterobacteriaceae family.
CFU = colony-forming unit. UTI = urinary tract infection.
Of the 2,740 clean catch samples, 1,949 (71.1%) were from the older children in the study (aged 3-4 years) and a laboratory-confirmed UTI was identified in 38 (1.9%) of them. The remaining 791 clean catch samples (28.9%) had been obtained from children aged <3 years:
- n = 612 aged 24-35 months;
- n = 91 aged 12-23 months;
- n = 88 aged <12 months.
A UTI was identified in 16, two and four of the children in these age groups, respectively.
A clinical diagnosis of UTI before urine dipstick testing had been made in 168 (6.1%) children out of 2,740, of whom 28 (16.7%) were then found to be UTI positive. Clinical diagnosis achieved 46.6% sensitivity (true positive – child has the disease and test is positive), and 94.7% specificity (true negative – child does not have the disease and test is negative). This indication of the potential number of children missed is supported by the literature (Zorc et al, 2005; Coulthard et al, 2003).
The parent-reported index tests associated with a UTI were pain/crying while passing urine, smelly urine, previous UTI and, notably, the absence of severe cough. For pain/crying while passing urine and smelly urine, there was a graded association with increasing symptom severity – that is, the higher these index items were scored, the more severe the UTI proved to be.
Clinician-reported index tests associated with a UTI were higher global severity, which was strongly correlated with likelihood of UTI, as well as abdominal tenderness and absence of ear problems. Neither of these last two items reached statistical significance so they were not included in the clinical rule.
Analysis of the dipstick test results showed that the presence of nitrites was strongly associated with UTI and leucocytes were less strongly associated with UTI; the presence of haemolysed blood was only weakly associated with UTI.
Nappy pad urine samples
In the nappy pad group, reference standard results were available for 2,277 children:
- n = 68 aged ≥36 months;
- n = 353 aged 24-35 months;
- n = 884 aged 12-23 months;
- n = 972 aged <12 months.
Laboratory-confirmed UTI was found in none, seven, seven and 16 children in these age groups, respectively. In total, therefore, 30 (1.3%) of the 2,277 children were found to have a UTI.
Parent-reported index tests independently associated with UTI were smelly urine, darker urine, female sex and the absence of a nappy rash. There was evidence of graded association for the first two items.
After dipstick testing, the presence of leukocytes and nitrites (but not of blood) was found to be independently associated with UTI. No clinical examination findings were predictive of UTI.
Guidance for practice
Our two-step clinical rule can be used in primary care to:
- Support the identification of young children in whom non-invasive urine sampling is indicated;
- Guide antibiotic treatment.
Lessons for clinical practice that can be derived from the results of the DUTY study are summarised in Table 1 and further described below.
At Step 1, symptoms and signs that should guide the decision to collect urine include:
- Pain or crying while passing urine;
- Smelly urine;
- Previous UTI.
Abdominal tenderness and absence of ear problems (the two clinician-reported index tests that were associated with a UTI but did not reach statistical significance – and hence have been excluded from Table 1) can also be taken into account.
At Step 2, symptoms and signs plus dipstick test results that should guide the decision to prescribe antibiotics are:
- Pain/crying while passing urine;
- Smelly urine;
- Previous UTI;
- Absence of severe cough;
- Severe illness present;
- Dipstick test leukocytes-positive;
- Dipstick test nitrites-positive;
- Dipstick test blood-positive.
The presence or absence of fever is not a diagnostic indicator for UTIs in children aged <5 years in primary care and so should not be used to determine the likelihood of UTI in this population and setting.
Precisely how the above items are used is likely to depend on clinicians’ preferences. Some may wish to use them as risk factors to inform clinical judgement, while others may prefer to adopt a checklist approach and use the DUTY study points-based rules, which focus on factors that have predictive value rather than those with poor diagnostic utility, such as fever.
As a contaminated sample can mask a true infection, the clean catch method should be prioritised over the nappy pad method, especially when the child has nappy rash. However, when nappy pads are the only viable option, the addition of dipstick testing significantly improves diagnostic accuracy. Box 2 outlines points to note when assessing for UTI in children.
Box 2. Assessing for UTI: points to note
- The DUTY clean catch criteria are for children from whom a clean catch urine sample can be ontained
- UTI is defined as ≥105 CFU/ml of a single or predominant uropathogen cultured from a clean catch urine specimen
- Use the clinical characteristics listed in Table 1 to decide whether a clean catch urine sample should be collected and whether antibiotics should be given
- Be aware that it is not clear which of the following antibiotic treatment strategies is most cost-effective: immediate presumptive treatment of all sampled children; immediate dipstick-guided treatment; or laboratory-guided (delayed) treatment
- Consider advising parents of all children (clean catch urine-sampled, nappy pad urine-sampled and non-urine sampled) to seek medical advice if their child gets worse
- The DUTY clean catch clinical prediction guideline is designed to supplement, not replace, clinical judgement
CFU = colony-forming unit. DUTY = Diagnosis of Urinary Tract Infection in Young Children. UTI = urinary tract infection.
Source: Adapted from Hay et al (2016)
- The full DUTY study report has been published by Hay et al (2016). For more information click here.
- Urinary tract infections are difficult to diagnose in young children in primary care, notably because symptoms and signs are non-specific
- Whenever possible, urine samples in children should be collected using the ‘clean catch’ method
- Recent research has pinned down factors (symptoms, signs and dipstick test results) that can help decide which children should have a urine test and receive antibiotics
- The presence or absence of fever is not a diagnostic indicator for UTIs in young children in primary care
- Dipstick testing of urine improves the targeting of antibiotic treatment
Ammenti A et al (2012) Febrile urinary tract infections in young children: recommendations for the diagnosis, treatment and follow-up. Acta Paediatrica; 101: 5, 451-457.
Butler CC et al (2015) Childhood urinary tract infection in primary care: a prospective observational study of prevalence, diagnosis, treatment, and recovery. British Journal of General Practice; 65: 633, e217-e223.
Coulthard MG et al (2014) Does prompt treatment of urinary tract infection in preschool children prevent renal scarring: mixed retrospective and prospective audits. Archives of Disease in Childhood; 99: 4, 342-347.
Coulthard MG et al (2003) A nurse led education and direct access service for the management of urinary tract infections in children: prospective controlled trial. British Medical Journal; 327: 7416, 656.
Farnham SB et al (2005) Pediatric urological causes of hypertension. The Journal of Urology; 173: 3, 697-704.
Gorelick MH et al (2003) Validation of a decision rule identifying febrile young girls at high risk for urinary tract infection. Pediatric Emergency Care; 19: 3, 162-164.
Gorelick MH, Shaw KN (2000) Clinical decision rule to identify febrile young girls at risk for urinary tract infection. Archives of Pediatrics and Adolescent Medicine; 154: 4, 386-390.
Hay AD et al (2016) The Diagnosis of Urinary Tract infection in Young children (DUTY): a diagnostic prospective observational study to derive and validate a clinical algorithm for the diagnosis of urinary tract infection in children presenting to primary care with an acute illness. Health Technology Assessment; 20: 51: 1-294.
Jacobson SH et al (1989) Development of hypertension and uraemia after pyelonephritis in childhood: 27 year follow up. British Medical Journal; 299: 6701, 703-706.
National Institute for Health and Care Excellence (2007) Urinary Tract Infection in Under 16s: Diagnosis and Management.
Royal Children’s Hospital Melbourne (2015) Urinary Tract Infection.
Shaw KN et al (1998) Prevalence of urinary tract infection in febrile young children in the emergency department. Pediatrics; 102: 2, e16.
Subcommittee on Urinary Tract Infection et al (2011) Urinary tract infection: clinical practice guideline for the diagnosis and management of the initial UTI in febrile infants and children 2 to 24 months. Pediatrics; 128: 3, 595-610.
Zorc JJ et al (2005) Clinical and demographic factors associated with urinary tract infection in young febrile infants. Pediatrics; 116: 3, 644-648.