Article Text

Download PDFPDF

Can oral antibiotics be used to treat urinary tract infections in infants aged 2–3 months?
  1. Daniel Cave
  1. General Paediatrics, Leeds Children's Hospital, Leeds, UK
  1. Correspondence to Dr Daniel Cave, General Paediatrics, Leeds Children's Hospital, Leeds, LS1 3EX, UK; dan.cave{at}

Statistics from

Clinical bottom line

  • Oral antibiotics are well absorbed in infants younger than 3 months (Grade B).

  • First-line oral antibiotic therapy is supported by several international consensus statements for well-looking infants older than 2 months (Grade B).

  • A period of observation, with measurement of procalcitonin, may be a safe strategy to identify young infants with urinary tract infection suitable for outpatient management (Grade C).


A 10-week-old boy was brought to Accident and Emergency due to poor feeding. On assessment, he was happy and alert and remained asymptomatic after a period of observation, breast feeding well. However, a routine clean-catch urine sample was obtained and revealed white cells on microscopy, confirmed on repeat testing. He was therefore cannulated and started on parenteral antibiotics as per national and trust guidelines. However, could the patient and family experience be improved by commencing oral antibiotics first line?

Structured clinical question

In asymptomatic infants aged 60–90 days (patient), are oral antibiotics (intervention) compared with parenteral antibiotics (comparison) equivalent (outcome) for initial treatment of urinary tract infection (UTI)?


A literature search was conducted using Medline, EMBASE and PubMed databases. Search terms were “urinary tract infection” AND “oral antibiotics” AND (“children” or “infant”). Papers were then reviewed for studies including infants under the age of 3 months. A total of 269 articles were found, of which 4 were included for detailed review. Papers were excluded from the main evidence summary (Table 1) if they did not use oral antibiotics as initial therapy.

Table 1

Main evidence summary


Asymptomatic UTI is a commonly encountered problem in young infants and causes uncertainty among clinicians.1 Urine collection for infants under 3 months is recommended for non-specific symptoms such as lethargy and poor feeding by national guidelines,2 increasing the diagnosis of incidental UTI. Prevention of renal scarring is a key target of UTI management, with its development increasing the risk of hypertension and renal failure later in life.3 Young infants are thought to be at higher risk of short-term and long-term complications after UTI, reflected in the tendency towards more aggressive management seen in clinical practice.2 Febrile neonates with UTI have a 11%–21% risk of bacteraemia.4 5 However, the incidence falls to 6%–13% at 30–60 days, and 2%–4% at 60–90 days.4 5 These figures may be overestimated, given data that infants who appear more unwell are more likely to have blood cultures taken.6 The risk of coexisting meningitis in infants aged 60–90 days is <0.1%.5 7

Current National Institute for Health and Care Excellence (NICE) guidelines for UTI in children recommend assessment by a paediatric specialist and parenteral antibiotics for any infant under 3 months.6 The recent update to the Fever in Under 5s guideline (published in 2019) advises intravenous antibiotics in febrile infants aged 1–3 months who appear unwell or with deranged white cell count.8 No recommendations address well afebrile infants with suspected (lower) UTI. The guidance for lower UTI in infants older than 3 months is for 3 days of oral antibiotics.8

A recently published Swiss consensus statement (Buettcher et al) provides updated guidance on the management of childhood UTI.9 This guideline agrees with many NICE recommendations for diagnosis and follow-up imaging. However, this statement also helps to clarify the advice for treatment of young infants. It states that for infants >60 days who are not clinically unwell or vomiting, with no pre-existing urological issues, oral antibiotics may be used first line.9 The suggested oral agents are broad spectrum, including co-amoxiclav and cephalosporins, although adjustment to narrow-spectrum antibiotics after culture results is also advised. This approach has already been adopted by guidelines in the USA10 and Canada,11 recommending oral antibiotics even for febrile infants.

There is limited evidence on the equivalence of oral antibiotics with parenteral antibiotics in infants under 3 months of age. Three randomised trials have compared broad-spectrum oral and intravenous antibiotics in young children. Hoberman et al 12 randomised 306 young children (47% aged <6 months) with a first febrile UTI to receive oral cefixime or intravenous cefotaxime (3 days) followed by oral cefixime for a total of 14 days. Children older than 8 weeks assigned to oral treatment were discharged after their first dose, or observed for up to 4 hours if vomiting prevented the administration of the initial dose. Resolution of fever was not affected by route of treatment (oral 25 hours vs intravenous 24 hours, p=0.76).12 Only three were too ill to be recruited. There was also no difference in prevalence or extent of renal scarring on dimercaptosuccinic acid (DMSA) scintigraphy at 6 months.1 In a post-hoc subanalysis, children with vesicoureteric reflux (VUR) given oral antibiotics had similar rates of persistent renal parenchymal damage on DMSA at 6 months (risk ratio (RR) 1.88, 95% CI 0.83 to 4.24). However, those with VUR grade 3–5 may be more likely to have renal damage if treated with oral antibiotics (RR 7.33, 95% CI 1.00 to 54.01).13 14 Most patients (61%) included in this study had proven acute pyelonephritis, and this result may therefore not apply to afebrile lower UTIs even in young infants.

Montini et al conducted a similar trial comparing oral co-amoxiclav with intravenous ceftriaxone followed by oral switch for presumed acute pyelonephritis.13 A wide age range was included, with around 190 infants under 6 months.2 All patients were admitted for an average of 5 days, and patients with anatomical abnormalities, dehydration or vomiting were excluded. However only eight patients (3.3%) required a switch from oral to intravenous treatment for side effects, and none changed treatment due to clinical deterioration. Baseline DMSA confirmed pyelonephritis in 55% of cases.13 No difference between the oral and intravenous treatment groups was detected in short-term or long-term outcomes, with DMSA performed again at 12 months.2 A more recent randomised trial by Bocquet et al using oral cefixime yielded very similar results.15 In this study, children were admitted for a minimum of 12 hours, and only one was changed to intravenous therapy for apparent sepsis. Outcomes were specifically reported for infants younger than 3 months, and no difference in renal scarring was found between treatment groups.15 However, this group was small with only 21 patients. Neither trial reported any serious illness or adverse treatment affects. A 2014 Cochrane review and meta-analysis collated evidence for the treatment of acute pyelonephritis in children more than 1 month old.14 In comparison with intravenous treatment (3 days) followed by oral switch (10 days), children treated with oral antibiotics only (10–14 days) had similar duration of fever (mean difference 2 hours, 95% CI −0.8 to +4.9), rates of persistent UTI after 72 hours (RR 1.1, 95% CI 0.07 to 17.4), and persistent kidney damage at 6–12 months (RR 0.82, 95% CI 0.59 to 1.12).14

A common myth is that young infants have limited enteral absorption of antibiotics. Despite delayed absorption of penicillins reported in some neonatal cohorts, serum levels reach therapeutic values with standard dosing regimes.16 17 The rate of absorption is likely similar to older children after the neonatal period.18 Cefalexin, commonly used to treat upper UTI, reaches maximum concentration after 3 hours in the newborn, compared with 2 hours in infants aged 3–6 months, and 1 hour in those older than 9 months.19 A study in Pakistan demonstrated that oral cefalexin is highly effective for treatment of periumbilical cellulitis in clinically well newborns, with a cure rate of 99.5%.20 Oral antibiotics are effective for a variety of indications in young infants,21 and are included in the British National Formulary for Children.22 When oral antibiotics are used after short-course parenteral treatment to treat UTI in young infants (<60 days), clinical cure rate is >97%.23 Compared with 14 days of intravenous therapy, oral switch after 3 days does not increase the risk of recurrent UTI or renal scarring.14 In a large retrospective study of infants with UTI (68% younger than 3 months), treatment failure was not associated with length of intravenous treatment.1

Another concern that may arise with oral antibiotic use is the antimicrobial resistance to these agents. A Swedish randomised trial by Mårild et al compared two oral antibiotics, ceftibuten and trimethoprim–sulfamethoxazole, for initial treatment of febrile UTI in children aged 1 month–12 years.24 Once daily ceftibuten resulted in a clinical cure rate of 93%, with 91% urinary bacterial elimination at follow-up and only 2% antimicrobial resistance.24 However, the numbers of young infants were low, and this may influence resistance patterns and consequently antibiotic choice. Salomonsson et al retrospectively reviewed the age-related patterns of urinary pathogens in Danish children with pyelonephritis.25 Children younger than 6 months were more likely to have a non-Escherichia coli UTI (23% vs 11%, p<0.001), and more likely to have other Gram-negative rods.25 This was thought to be due to the larger number of boys in this group. Resistance to co-amoxiclav was 10% compared with 6% in those older than 6 months.25 Resistance to cephalosporins was low and amoxicillin and trimethoprim showed high resistance in all ages.25 Awareness of local antimicrobial resistance patterns is important when considering first-line oral therapy.

Asymptomatic bacteriuria in young infants, detected by suprapubic aspiration screening, is associated with leucocyturia in up to 50% of cases.26 Most cases remain asymptomatic and spontaneously resolve; however, in one study, 4% of infants developed pyelonephritis shortly after screening.27 There is therefore a need to identify the infants with positive urine leucocytes who do not require invasive treatment. Several novel biomarkers have been used to stratify risk of serious illness for infants with UTIs. Procalcitonin (PCT) has been proven highly sensitive for detecting serious illness in febrile infants.28 29 In one study of 2047 febrile infants aged 7–91 days, PCT (cut-off value 0.3 ng/mL) was superior to C reactive protein for the prediction of invasive bacterial infection (area under the curve 0.91 vs 0.77, p=0.002).29 In infants aged 25–90 days with UTI, ‘well’ appearance combined with PCT <0.7 ng/mL conferred a very low risk for bacteraemia (negative predictive value 99.5%).28 ‘Well’ appearance was measured using the Paediatric Assessment Triangle, a useful adjunct that increases the objectivity of triage in children presenting to acute services.30

Oral antibiotics can be used as initial treatment for UTI in infants older than 60 days, and their use is supported by several national guidelines. Broad-spectrum oral antibiotics are likely to be well absorbed even in young infants, and resistance patterns are similar to older children. However, consideration of local resistance patterns is vital before translation to clinical practice. The data from three randomised trials have demonstrated non-inferiority compared with parenteral treatment for short-term and long-term outcomes after UTI. However, variable numbers of children under 3 months were included in each study. When presented with a well-looking infant with suspected UTI, clinicians should consider if oral antibiotics may be used to avoid cannulation and hospital admission. A holistic approach should be taken, considering the best interests of the child and family. The use of novel parameters such as PCT on initial assessment may reassure clinicians and support outpatient treatment.

Ethics statements

Patient consent for publication



  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles