Although our study is a retrospective study, the population of the treatment groups was similar in terms of demographic and clinical characteristics of UTI. Also, in children ≥ 3 months, the treatment outcome between CTX and TZP showed similar results including the frequency of recurrence. However, in children < 3 months, a higher frequency of recurrent UTIs was observed in the TZP treatment group. Based on these results, we selected TZP as the initial empirical antibiotic for UTIs in children ≥ 3 months, unless there were adverse effects, whereas CTX was used for those < 3 months.
This study also highlights the importance of considering the age of onset when approaching the clinical characteristics of UTI. For UTI cases occurring after three months of age, clinicians can follow the standard approach for typical UTIs, including treatment and additional diagnostic plans as appropriate. However, if abnormal prenatal USG findings are present, a thorough evaluation may be necessary due to the potential risk of UTI recurrence caused by conditions such as Congenital Anomalies of the Kidney and Urinary Tract (CAKUT).
However, in infants younger than three months, a personalized approach for UTI is important, and it is crucial to differentiate UTIs accompanied by bacteremia, also known as urosepsis. Further, special attention should be paid to male infants, those with prolonged fever, those with confirmed bacteremia, and those with detection of ESBL − positive uropathogens. And age − related pharmacokinetic factors such as antibiotic metabolism and excretion, as well as variations of clinical response to the antibiotics based on age, may have influenced these differences in recurrence. Therefore, careful follow − up observation for post − treatment UTI recurrence is necessary in this age group.
In general, most drugs achieve high concentrations in the urinary tract, surpassing the minimum inhibitory concentration (MIC) of blood serum [13]. However, in addition to the above − mentioned age − related differences in clinical response to medications, in children < 3 months, who have weaker immune defenses mechanism against bacteria, bacteremia can occur more frequently. Further, there is a higher incidence of UTI recurrence with ESBL − positive strains. Therefore, when UTI occurs in children < 3 months, it is necessary to consider modifying antibiotic treatment based on antimicrobial susceptibility results, especially in cases where ESBL − positive bacteria are detected. Switching to drugs such as carbapenems may be appropriate for treating ESBL − positive infections in children < 3 months.
In contrast, for UTIs occurring in children ≥ 3 months, the presence of ESBL − positive bacteria was not associated with an increased risk of UTI recurrence, which is consistent with a previous report (14). Even if ESBL − positive uropathogens are identified, if the patient shows an improvement in fever, resolution of pyuria, and sterile urine on follow − up urine culture after empirical antibiotic treatment, there is no mandatory need to switch to carbapenems. TZP has gained attention as an alternative to carbapenems for the treatment of ESBL − producing UTI in adults [15, 16]. Most antibiotics achieve higher concentrations in the urinary tract than in the blood. Successful treatment of UTI is possible even with this concentration discrepancy between urine and blood, suggesting that consistent application of MIC is misguided. Based on our research findings, it is possible to maintain the initial intravenous use of TZP or switch to an oral alternative with good susceptibility results for ESBL − positive UTI in children ≥ 3 months. This approach is recommended for children ≥ 3 months who do not have immunodeficiencies. However, it is advisable to closely monitor for UTI recurrence.
At our institution, an antimicrobial stewardship program is conducted by pediatric infection disease specialists to ensure the judicious use of antibiotics. These usage strategies are reflected in the results of this study. For example, patients who received carbapenems had 100% urine culture − proven UTIs, and the proportion of ESBL − positive strains was significantly higher than that in the other treatment groups.
The hospitalization period differed among the groups, with patients receiving CTX or TZP being discharged after switching to oral antibiotics based on the antibiotic susceptibility results, whereas those with ESBL − positive strains remained hospitalized and received intravenous carbapenem antibiotics.
One strength of this study is that our institution is one of the facilities responsible for the primary care of pediatric patients on Jeju Island, and therefore has unique characteristics. Our institution is also well − equipped to provide specialized medical care for pediatric patients. In this facility, 18.3% of community − acquired UTIs were caused by ESBL − positive bacteria, a rate which exceeds the previously reported rate of 16.7% in the South Korean province of Jeolla [17]. Seo et al. reported a significant increase in ESBL − positive E. coli from 4.6% in 2008 − 2012 to 15.8% in 2013 − 2016 [18]. The collaborative relationship between pediatric infection disease specialists and nephrologists in our institution plays a crucial role in determining treatment strategies for pediatric UTI, which is one of the most common pediatric infection diseases. This collaboration not only allowed for a comprehensive approach to the management of pediatric UTIs, but also contributed to reducing antimicrobial resistance among pathogens within the healthcare facility. It is suggested that in areas with an increasing prevalence of ESBL − producing pathogens and a high incidence of related infection diseases, collaborative relationships and multidisciplinary teams within healthcare institutions, such as ours, should be taken into consideration when developing empirical treatment protocols and antibiotic management programs [5].
The limitations of our study include its retrospective cross − sectional design and its limited scope to a single institution, which may introduce selection bias. Additionally, the study did not provide clear mechanistic explanations for the differences in antibiotic efficacy based on age. These aspects highlight the need for future research to further investigate these issues.
In conclusion, for UTIs in children < 3 months, an individual patient approach is important, and special attention to antibiotic − resistant uropathogens is required. The risk of recurrent UTIs caused by ESBL − positive uropathogens was not significant in patients with UTIs aged ≥ 3 months. Further, TZP can be an alternative to CTX for febrile UTI in infants and children ≥ 3 months.