Comparison with the NICE guideline and existing literature
It has been shown that childhood UTI's are often not recognised by general practitioners [
5]. Under-diagnosis of UTI in children is thought to be responsible for a significant number of patients developing end-stage renal failure as a consequence of acquired renal scarring [
6‐
8]. In the past, children with a confirmed UTI were thoroughly investigated so that any underlying predisposing cause was established. The new NICE guidelines place much less emphasis on advanced investigation. Their main thrust is to ensure that all children with UTI are correctly diagnosed and appropriately treated. The guidelines suggest this approach may be more effective in preventing acquired renal scarring. Thus one of the main objectives of the new NICE guideline is to encourage general practitioners to consider the diagnosis of UTI at an early stage when assessing a sick child. It is most encouraging to note that all of the doctors in this study agreed that it was important to consider the diagnosis of UTI in all children with unexplained fever.
This survey revealed that doctors accurately identified common symptoms and signs of UTI. Several studies have reported on symptoms and signs in children presenting with UTI to a hospital setting [
9‐
19]. Two studies have looked specifically at children presenting to a general practice [
20,
21]. In a preverbal child, fever is consistently the most common symptom. Verbal children, like adults, most commonly present with dysuria and frequency. Other common symptoms include abdominal pain, loin tenderness, vomiting and poor feeding. The majority of doctors in this study identified the appropriate common symptoms and signs (Table
1).
The NICE guideline places an emphasis on recording the presence of risk factors for UTI and serious underlying pathology. In this study, when doctors were given a list of potential risk factors, they were able to accurately identify 87% of the relevant risk factors for UTI. This demonstrates a high level of awareness amongst doctors of the relevant risk factors for UTI and underlying pathology. When compared with existing similar literature, doctors in this study demonstrated a superior knowledge of predisposing factors for UTI [
22]. One risk factor which was insufficiently recognized was constipation, with 17% of respondents indicating that this was not a risk factor for UTI.
Collection of an appropriate urine sample is an important component of the accurate diagnosis of urinary tract infection in children. It remains a challenging process, especially in children who are not toilet trained. Jadresic
et al showed that the more general practitioners send urine samples from children, the higher the diagnostic rate in that practice [
23]. The NICE guideline suggests that general practitioners advise parents to collect a clean catch sample where possible. This recommendation is mainly based on a systematic review which identified five studies that compared the diagnostic accuracy of clean catch urine samples with that of urine samples obtained by supra-pubic aspirate (SPA) [
24]. In general, the diagnostic accuracy of the clean catch samples was comparable to that of the SPA samples. In this study, most doctors said that they would advise parents to use a urine collection bag. It is possible that this preference is in part driven by parental choice. In one study that examined parental preferences for collecting a urine sample at home from an infant, the majority of parents found collection of a clean catch urine to be time-consuming and often messy [
25]. No respondent in this study indicated that they would use a urine collection pad to collect a urine sample from an infant. The NICE guideline reports insufficient evidence to recommend a preference for the use of pads or bags. It is noted, however, that pads are considerably less expensive, and, based on cost considerations, their use is recommended in the guideline. It has been shown also that parents find pads easier to use and more convenient than urine collection bags [
25].
The NICE guidelines recommend the use of dipstick testing only in the case of children over the age of three years. The diagnostic accuracy of leukocyte esterase and nitrite dipsticks is much lower in younger children [
26]. There is a considerable risk of missing a proportion of cases of acute UTI in infants and children younger than three years when using dipstick testing, as frequent bladder emptying leads to a lack of urinary nitrite [
27]. It is most interesting that 81% of doctors in this study indicated that they would use a urinary dipstick to help diagnose a UTI in two-year-old children. This may represent a situation where it is not feasible to give practical effect to evidence based guidelines, because there is no other test for UTI that provides immediate results unless doctors can carry out or access microscopy. This is acknowledged in the guidelines, with a suggestion that a urinary dipstick could be used for a relatively well child under the age of three years, with non-specific symptoms, provided the test is backed up by non-urgent microscopy.
This study demonstrates a clear variation in practice, amongst doctors working in primary care, with regard to investigation and specialist referral of children with UTI. In general the NICE guidelines recommend against detailed investigation and specialist referral of children with their first diagnosis of UTI, who respond well to treatment within 48 hours and have no atypical features. According to NICE, gender is no longer a major factor in influencing the decision to refer or investigate children with UTI. However, this study reveals that practice is gender dependent, with most respondents having a lower threshold for investigating and referring a boy.
In this study, co-amoxiclav was the antibiotic most commonly prescribed, by general practitioners for the 'blind' treatment of UTI in children. This is consistent with the results of a large Dutch family practice cohort study [
28]. Three randomized control trials which compared the effectiveness of different oral antibiotics in lower UTI in children reported no significant difference between treatments [
29‐
31]. NICE do not recommend a specific antibiotic for 'blind' treatment, but instead suggest that the choice should be based upon locally developed multidisciplinary guidance. It is suggested that an antibiotic with low resistance patterns, such as a cephalosporin or co-amoxiclav, should be used when treating an upper urinary tract infection with oral antibiotics. It is also suggested that trimethoprim, nitrofurantoin, cephalosporin or amoxicillin may be suitable for the treatment of lower urinary tract infection (cystitis). At a local level, the microbiology department servicing the study region recommends co-amoxiclav as a first line treatment for uncomplicated lower urinary tract infection. This recommendation arises from the increasing resistance to
E. Coli in this region, as well as increasing trimethoprim resistance rates which are currently over 20% in this laboratory catchment area.
There was a wide variation in the number of days of antibiotic treatment which doctors prescribe for a lower UTI in a six-year-old child. A Cochrane review which included 10 randomized control trials comparing short (2-4 days) with standard (7-14 days) duration of oral antibiotic is quoted in the NICE guideline. There was no significant difference between the two groups to justify the longer duration of therapy [
32]. NICE clearly recommends treatment with oral antibiotics for three days for children aged over three months with lower UTI. It also recommends that parents and carers should be advised to bring the child back to the general practitioner for reassessment if the child is still unwell after 24-48 hours.
The wide variation in the practice of general practitioners reported in this study is in keeping with the findings of similar studies in other countries [
28,
33]. However, it is important to note that this study was carried out 10 months after the publication of a significant clinical guideline on the subject. Thus, it was anticipated that this study might have demonstrated a more consistent approach by general practitioners to UTI in children. However, only minority of respondents indicated that they had accessed clinical guidelines on the subject.