Urinary tract infection (UTI) is defined as detection of microbial growth in a sample of urine obtained through sterile methods from an individual with symptoms compatible with infection.1–5 Isolation of bacteria in urine culture in the absence of symptoms is defined as asymptomatic bacteriuria and does not require treatment.1–5

Based on the clinical presentation and the results of diagnostic tests, UTI can be further classified into acute pyelonephritis (upper UTI), or cystitis (lower UTI).1–5

A case of UTI is considered atypical and carries a higher risk of complications if any of the following apply: persistent fever of more of 48hours’ duration after initiation of appropriate antibiotherapy, development of sepsis, isolation of organism other than non-extended spectrum β-lactamase (ESBL)-producing Escherichia coli, acute kidney injury and/or detection of an abdominal or vesical mass.1,3

We defined recurrent UTI as 2 or more episodes of upper UTI, 1 episode of upper UTI plus 1 episode of lower UTI, or 3 or more episodes of lower UTI in one year.1,3,4

Urinary tract infections are among the most frequent infections in the paediatric age group and can be particularly severe in infants aged less than 3 months.1–6 Age and sex are key factors associated with the incidence of UTI, with a higher incidence in males up to 6 months of age and in females from 1 year of life.1–7

After an initial episode of UTI, the time interval used to differentiate between relapse and reinfection in case of recurrence is of 2 weeks.8 The risk factors for recurrent UTI are: obstructive uropathy, bowel or bladder dysfunction, sexual activity in adolescents and use of indwelling urinary catheters.1–8

The different epidemiological studies conducted in the paediatric population of Spain have shown that E. coli is the most frequent aetiological agent,7,9–11 with a prevalence ranging from 60% to 80%. Previous use of antibiotics or a history of urinary system abnormalities increase the probability of infection by other microorganisms,9–11 such as Proteus mirabilis (6%–10%) and Klebsiella pneumoniae (3%–5%). Fewer than 2% of cases are caused by other enterobacteria: Klebsiella oxytoca, Enterobacter cloacae, Citrobacter spp., Serratia marcescens and Morganella morganii. The most frequently involved gram-positive bacteria are Enterococcus species in infants aged less than 3 months and children with renal or urologic diseases, and Staphylococcus saprophyticus in female adolescents presenting with uncomplicated UTI.9–11

Age is a determining factor, for the shorter the age of the patient, the more nonspecific the symptoms.1–5,12

In infants and young children that are not toilet-trained, the key finding is fever without a source.1–5,12 The identification of a source of the fever does not exclude UTI, but reduces its likelihood.1,4 In the absence of fever, the level of suspicion of UTI should be low.1–5,12 The presence of other symptoms, such as prolonged jaundice or irritability in infants, food refusal, vomiting or weight faltering are not specific for UTI and could result from many other disease processes.1–5,12 For this reason, a urine culture should only be requested when other, more frequent diseases characteristic of childhood have been ruled out. A positive result of urine culture in an infant or non-toilet-trained young child in whom the suspicion of UTI is low may only be indicative of asymptomatic bacteriuria, which does not require treatment or additional diagnostic tests.1–5

In toilet-trained children, classic urinary symptoms gain more importance.1–5 Painful or frequent urination, a sense of incomplete bladder voiding, urinary urgency and/or suprapubic pain in the absence of fever are suggestive of lower UTI, although they, too, are nonspecific, as they can develop in other uropathies, such as voiding disorders or renal lithiasis.1–5,12 If blood is present in the urine, the most frequent cause is UTI, although in this case further tests should be performed to rule out other possible causes.1–5,12 The association of fever, pain in the region of the renal fossa, general malaise and/or chills are suggestive of pyelonephritis.1–5,12

The objective of antibiotherapy is to achieve symptom relief, prevent sepsis and reduce the risk of complications.3–5 It must be initiated early after collection of adequate specimens for culture.3–5 Early treatment is particularly important in patients with febrile UTI, toxic appearance, immunodeficiencies or known renal or urinary system anomalies.3–5

Although there are regional variations, more than 60% of E. coli isolates in Spain are resistant to amoxicillin or ampicillin, and 20%–40% to cotrimoxazol,9,28 so we recommend against the use of any of these antibiotics for empirical treatment of UTI.

In recent years, the prevalence of resistance to first-generation cephalosporins and amoxicillin–clavulanic acid has been increasing in Spain, reaching percentages of more than 15% in some areas, although there is wide variability between regions.9,28 Therefore, these should also not be used as first-line empirical therapy. The antibiotics that continue to demonstrate significant antimicrobial activity are second- and third-generation cephalosporins, fosfomycin and aminoglycosides.9,28 Due to the limited use of quinolones in paediatrics, E. coli strains isolated in children tend to be more sensitive to this group of antibiotics compared to strains isolated in adults,9,28 although the proportion resistant to ciprofloxacin can be as high as 15%–20%. It is important to keep in mind the intrinsic resistance of Enterococcus faecalis to cephalosporins and aminoglycosides, so we recommend the addition of ampicillin in patients aged less than 3 months or if gram-positive cocci are detected.3

Another alarming phenomenon is the progressive increase in the prevalence of ESBL-producing bacteria isolated from community samples, especially of E. coli and Klebsiella species, which are resistant to most beta-lactam antibiotics with the exception of the carbapenems.9,28

Other emerging mechanisms of drug resistance, such as the production of AmpC beta-lactamases or carbapenemases, have been found mainly in hospital settings, so we will not address them in the present document.

At present, follow-up in specialty care and performance of a renal sonogram are only recommended in infants aged less than 6 months with a first episode of UTI or patients with atypical or recurrent UTI.1,4,26 In children with abnormal sonographic findings or atypical or recurrent UTI, we recommend the additional performance of a voiding cystourethrogram or a contrast-enhanced ultrasound of the bladder, especially if the patient is aged less than 6 months.1,4 In cases of atypical or recurrent UTI, a follow-up renal scintigraphy is recommended 4–6 months after the episode, especially in children aged less than 3 years.1,4 In any case, the performance of imaging tests after a UTI and the need for follow-up in specialty care services remain controversial, and therefore decisions regarding these aspects must be made on a case-by-case basis.1,4,30

The prognosis of a first UTI that responds well to treatment in the first 48h is excellent. The risk factors associated with renal scarring are: atypical UTI, recurrent UTI, obstructive uropathy and delay of 48–72h or greater in the initiation of appropriate antibiotherapy.1,4

The measures recommended for prevention of new episodes of UTI include: avoiding poor voiding habits (such as voluntary urine retention) and inadequate fluid intake and correction of constipation and bladder or bowel disorders. There is no evidence supporting any other measures, such as frequent diaper changes or the consumption of probiotics or cranberry juice. There is also insufficient evidence to support circumcision, although it could be considered in boys with recurrent UTIs.1,4

The use of chemoprophylaxis has been decreasing. It has not been proven to achieve a reduction in UTI recurrence or renal scarring, whereas it is proven that the selective pressure exerted by antimicrobial use is associated with an increased risk of infection by multidrug-resistance pathogens. Chemoprophylaxis should only be contemplated on a case-to-case basis in patients with obstructive uropathies, selecting the antibiotic with the narrowest-possible spectrum (trimethoprim or cotrimoxazole) and administering only 25% of the effective dose as a single dose at night.1,4

We assessed the quality of the evidence and determined the strength of the recommendations applying the evidence grading system for clinical practice guidelines of the Infectious Diseases Society of America and the United States Public Health Service, summarised in Table 6.

A positive urine culture is not proof of UTI. In the absence of symptoms, isolation of bacteria in urine culture is known as asymptomatic bacteriuria, a condition that does not require treatment. Quality of evidence: II. Strength of recommendation in favour: B.

In infants and toddlers who are not toilet trained, the key symptom for suspicion of UTI is fever. In toilet-trained children, the classic urinary symptoms become more important. Quality of evidence: I. Strength of recommendation in favour: A.

In every case, diagnosis requires confirmation by a urine culture of a sample collected before initiation of antibiotherapy, which allows subsequent targeted treatment based on the results of antimicrobial susceptibility testing. Quality of evidence: I. Strength of recommendation in favour: A.

In toilet-trained children, the urine sample used for culture should be a midstream urine specimen. Quality of evidence: I. Strength of recommendation in favour: A.

In infants and children who are not toilet trained, the specimen for urine culture should be obtained by urinary catheter; suprapubic aspiration is also an option in newborns and young infants. Quality of evidence: II. Strength of recommendation in favour: B.

Samples collected with a urine bag should not be submitted for culture. Quality of evidence: I. Strength of recommendation against: A.

In patients that do not meet the criteria for hospital admission, there is no evidence of significant differences in clinical outcomes or the development of sequelae based on the exclusive use of oral antibiotherapy versus short-term intravenous antibiotherapy followed by oral treatment. Quality of evidence: II. Strength of recommendation in favour: B.

The selection of antibiotics for empirical treatment should be based on the local susceptibility pattern. Quality of evidence: I. Strength of recommendation in favour: A.

The recommended empirical oral antibiotherapy in cases of cystitis is a second-generation cephalosporin in children aged less than 6 years and fosfomycin trometamol in older children. Quality of evidence: III. Strength of recommendation in favour: B.

We recommend the use of third-generation cephalosporins for empirical treatment of patients with pyelonephritis that do not require hospital admission. In patients with pyelonephritis admitted to hospital, we recommend treatment with aminoglycosides. Ampicillin should be added in patients aged less than 3 months. Quality of evidence: III. Strength of recommendation in favour: B.

A follow-up urine culture is not necessary in patients that respond well to treatment. Quality of evidence: I. Strength of recommendation in favour: A.

Roi Piñeiro Pérez coordinated the writing of the entire document and was directly involved in writing the following sections: Introduction, Clinical Features, Follow-up and Conclusions, in addition to performing a final revision of the entire document.

María José Cilleruelo Ortega coordinated the writing of the entire document and was directly involved in writing the following sections: Introduction, EpidemiologyandAetiology, Diagnosis and Conclusions, in addition to performing a final revision of the entire document.

Josefa Ares Álvarez was directly involved in writing the following sections: Diagnosis, Follow-up and Conclusions, in addition to performing a final revision of the entire document.

Fernando Baquero-Artigao was directly involved in writing the following sections: Diagnosis, Treatment and Conclusions, in addition to performing a final revision of the entire document.

Juan Carlos Silva was directly involved in writing the following sections: EpidemiologyandAetiology, Treatment and Conclusions, in addition to performing a final revision of the entire document.

Roberto Velasco Zúñiga was directly involved in writing the following sections: Clinical Features, Treatment and Conclusions, in addition to performing a final revision of the entire document was directly involved in writing the following sections:

Leticia Martínez Campos was directly involved in writing the following sections: Diagnosis, Treatment and Conclusions, in addition to performing a final revision of the entire document.

Begoña Carazo Gallego was directly involved in writing the following sections: Epidemiology andAetiology, Clinical Features and Conclusions, in addition to performing a final revision of the entire document.

Antonio José Conejo Fernández was directly involved in writing the following sections: Treatment, Follow-up and Conclusions, in addition to performing a final revision of the entire document.

Cristina Calvo coordinated the writing of the entire document and was directly involved in writing the following sections: Introduction, Diagnosis and Conclusions, in addition to performing a final revision of the entire document.

Santiago Alfayate Miguélez, Alicia Berghezan Suárez, César García Vera, Juan José García García, Marisa Herreros Fernández and Carlos Rodrigo Gonzalo de Liria reviewed the final manuscript.

The authors have no conflicts of interest to declare.