Pediatric Urinary Tract Infections

Diagnosis and Treatment

Maria Bitsori; Emmanouil Galanakis


Expert Rev Anti Infect Ther. 2012;10(10):1153-1164. 

In This Article

Setting the Index of Suspicion

UTI Burden in Childhood

UTI is one of the most common diagnoses among adults who seek medical advice, but in children, it is far behind respiratory tract infections and gastroenteritis.[9,14] Higher incidence of UTI in adults may be explained by predisposing factors such as sexual activity but may equally point to UTI underdiagnosis in children. UTI burden has been calculated by several estimates, including the annual incidence or the cumulative incidence for various age groups. However, accuracy has been hampered by differences in data sources, diagnostic criteria and clinical entities included in the term UTI. Estimates based on population studies from Sweden,[2,15] general practice reports from the UK[16,101] and hospital data from Canada[17] are summarized in Table 1. The overall conclusion of these figures is that one out of ten girls and one out of 30 boys will have a UTI by the age of 16 years, with a predominance of boys during neonatal period and early infancy.[101]

Which Children to Test for UTI

Estimates are not directly helpful in deciding which child to test for UTI. Routine testing is costly, time consuming and hampered by practical difficulties;[8,9] therefore, it is important to identify the children at high risk. A study on the clinician's threshold to obtain a urine culture showed that only 10% of clinicians believed that a urine culture was indicated if the probability of UTI was <1%, whereas 80–90% would obtain a culture if the probability of disease was 3–5% and all would do so if the probability exceeded 5%.[18] A recent meta-analysis concluded in a probability of 7–8% for sick children <2 years, which was highest for girls throughout infancy and uncircumcised boys <3 months (Table 1).[19] Accordingly, a urine culture should always be considered in the investigation of these groups.[19] The 2011 AAP guidelines seem to give special attention to the contribution of probability in the decision to test a child for possible UTI in an effort to avoid overdiagnosis, whereas NICE guidelines seem to focus mainly on underdiagnosis.

The Uncertainty of Clinical Presentation

The symptoms of frequency, urgency and loin pain that are highly suggestive of a UTI in an adult are often absent in or cannot be articulated by the younger children. Even when present, their accuracy seems to be limited: the prevalence of UTI was 7.8% for children aged 2–19 years[19] with suggestive symptoms as compared with 50% for female adults.[20] Clinicians are well aware that the younger the child, the more diverse and less specific can be the presentation of a UTI. Fever, poor feeding, vomiting, irritability, failure to thrive, smelly urine and crying on passing urine are the usual presentations in infancy.[21,101] Recently,it has been found that toilet-trained toddlers may start wetting again or complain of abdominal pain and older children might complain of dysuria or loin pain.[22] Nevertheless, older children might present with general symptoms, and dysuria in childhood might well have a different cause than UTI, such as local irritants, hypercalciuria, vulvovaginitis or dehydration.[22] Abdominal and back pain and new-onset urinary incontinence are symptoms and signs that increase the likelihood of a UTI in older children; their absence, however, is not sufficient enough to exclude the diagnosis (Table 2).[23] A critical review of the existing data concluded that no individual symptom/sign or any combination of them were sufficient enough to identify children with a UTI,[9] although some of them were more suggestive of a UTI than others for children 0–24 months (Table 2).[23,24] Combination of symptoms and signs seems to work better, especially to rule out a UTI for this age group.[9,23]

Setting the Index of Suspicion

On the basis of a combination of symptoms and signs that could set threshold probabilities of UTI at at least 1% and at least 2%, the 2011 AAP guideline suggested only clinical observation without testing for febrile young children with a probability of UTI less than 2%.[12] However, a substantial rate of underdiagnosis on the basis of physician's judgment was found in the largest study with 15,871 participants aged 0–5 years on the accuracy of clinical symptoms to predict a serious bacterial infection, including UTI.[24] The authors proposed a diagnostic model based on 27 clinical signs and symptoms, which could facilitate decision making in emergency departments,[24] although the feasibility of this scoring in busy primary care settings or emergency departments might be debated. Hopefully, DUTY, a well-designed study currently running in the UK, will achieve its aims and provide a simple and reliable clinical rule for the identification of UTI in acutely unwell children.[103]