We have described national epidemiology and resistance rates in paediatric UTIs in Norway and used corresponding data from adults as a reference. E. coli was the predominant microbe in urine samples from children, with low resistance rates (< 10 %) for nitrofurantoin, mecillinam, cephalexin and amoxicillin-clavulanic acid.
The higher proportion of urinary isolates from infants compared to older children reflects the higher incidence of upper UTIs in this age group (1). The predominance of E. coli in urinary isolates from children is well known (1, 3), but a high incidence of other bacteria is also reported internationally (16). The higher proportion of Enterococcus spp. and Proteus spp. among preschool children in our study has also been observed in the United States (8).
Use of ciprofloxacin in children in Norway is very low (17, 18), but overuse among adults is viewed as problematic (19). Lower ciprofloxacin resistance rates in E. coli in children compared to adults have also been observed internationally (11–13) and are likely to be linked to the lower usage in children (9).
We found greater resistance of E. coli to both trimethoprim and trimethoprim-sulfamethoxazole in children compared to adults, which also corresponds with the findings in a US study (13). In the Norwegian Directorate of Health's guidelines for antibiotic use, trimethoprim and trimethoprim-sulfamethoxazole are among the empirical first choices for children with lower and upper UTIs respectively (10). Consumer statistics from the Norwegian Prescription Database show that trimethoprim and trimethoprim-sulfamethoxazole have been frequently used among children ≤ 5 years in recent years (17). We can assume that a significant proportion of the prescriptions were for treatment of a UTI. We can also assume that a significant proportion were prescribed empirically on the basis of the guidelines. In general, however, empirical antibiotics are not recommended for the treatment of UTIs if the E. coli resistance rate exceeds 20 % (6).
Our data support current recommendations that pivmecillinam and amoxicillin-clavulanic acid should be among the first choices for empirical treatment of upper UTIs in children (2, 10). We found lower resistance rates among both E. coli and Klebsiella spp. to pivmecillinam than to amoxicillin-clavulanic acid. Furthermore, E. faecalis is sensitive to amoxicillin-clavulanic acid, and has inherent resistance to pivmecillinam. Despite the lack of good clinical evidence for the use of pivmecillinam in upper UTIs, it is our view that the available literature, low resistance rates, favourable ecological profile and good clinical experience provide an adequate foundation for recommending this in Norway (20–22). Pivmecillinam is not available as an oral solution. The tablets can be crushed and mixed with food such as jam etc., but crushed tablets have an unpleasant taste. Permission was granted to market amoxicillin-clavulanic acid as an oral solution in Norway in 2019, and this is well established as a treatment for upper UTIs in children (23). However, there are concerns about high resistance rates in several countries (6, 7), and increasing use could also lead to a further development of resistance in Norway (24).
At Haukeland University Hospital and the University Hospital of North Norway, we found low cephalexin resistance rates in E. coli urinary isolates from children. These are not national data, but they nevertheless provide a good indication of the likely level of resistance nationally. Cephalexin is available as an oral solution and is recommended in other high-income countries as the first choice for upper UTIs in children (25, 26). A large observational study showed good clinical efficacy in UTIs in children (27). However, consideration should also be given to the possible selection of extended spectrum beta-lactamases in cases of excessive use (28). Overall, it is our view that cephalexin should be considered on an equal footing with amoxicillin-clavulanic acid in the empirical guidelines for upper UTIs, also to avoid excessive use of amoxicillin-clavulanic acid.
Nitrofurantoin is not for use in upper UTIs due to poor tissue penetration, but is a good first choice for lower UTIs. In 2009, a Norwegian literature review called for approval of nitrofurantoin in drop form (22), but it is still only available as an unpleasant-tasting water-soluble tablet.
Intravenous gentamicin and ampicillin are recommended for clinically weak patients, and there should be a liberal threshold for their use in children younger than 3–6 months (2). The gentamicin resistance rate was < 3 % for all bacteria included except enterococci. This confirms that gentamicin and ampicillin represent a potent empirical regimen for Norwegian children.
One of the strengths of this study is that using national data from a reliable register gave us a basis for assessing the level of resistance in children in relation to the current guidelines for choosing antibiotics in the treatment of UTIs in children in Norway.
The lack of clinical data is a weakness of the study. Randomised controlled clinical trials investigating the efficacy of relevant peroral antibiotics for the treatment of upper UTIs would be useful. Different data registration periods meant that we had to estimate the distribution of the microbes, but we consider this to be sufficiently precise for our purposes. Some UTIs are likely to have been treated without a urine culture sample. Our data are therefore probably not representative of all clinical UTIs during the registration periods.
Some of the urine samples in the NORM register are likely to be collection bag samples. This increases the risk of contamination, but the register's strict inclusion criteria mean that the isolates in our material are most likely uropathogenic. Resistance data for cephalexin are only available from two hospitals, but data on cephalexin and E. coli in urine samples will be included in future NORM reporting.