Can magnitude and days with fever predict urinary tract infection in infants with fever without source?

Introduction: Fever continues to be the most frequent cause of care in the pediatric population. It is frequently the use of invasive and unnecessary tests which results in discomfort to the patients. The present study aims to describe the epidemiology of urinary tract infection (UTI) in children with fever without source aged 0–36 months and to evaluate if there are differences in the magnitude and days with fever in patients with and without urinary tract infection and fever without focus. Methods We included children aged 0–36 months presenting with fever without source in our emergency unit. Demographic and clinical characteristics, investigations, and management procedures were recorded at the time of inclusion. Potential predictors of SBI-UTI were compared between patients with and without SBI-UTI.


Introduction
Fever continues to be the most frequent cause of care in the pediatric population (1). Fever may be the only nding during clinical evaluation in infants younger than 3 years and can be an initial manifestation of potentially serious infections such as meningitis, pneumonia, urinary tract infection (2,3). Fever without source (FWS) constituting a frequent cause of concern for both doctors and parents (4,5).
Despite reductions in serious bacterial infections (SBI) rates following the introduction of speci c vaccination against Haemophilus in uenzae type b and Streptococcus pneumoniae, other germs such as E. coli, S. pneumoniae serotypes not covered by these vaccines such as Neisseria meningitidis, Staphylococcus aureus, Salmonella spp, and Enterococcus have become more important (6-9). There is considerable controversy in the management of these patients with FWS. It frequently the use of invasive and unnecessary tests which results in discomfort to the patients and higher cost to the health system (10) (11). Most SBI is due to urinary tract infection (UTI) (5). The screening strategy for children with FWS should be adapted to more accurately predict when children with FWS may have UTI. The present study aims to describe the epidemiology of urinary tract infection (UTI) in children with fever without source aged 0-36 months and to evaluate if there are differences in the magnitude and days with fever in patients with and without urinary tract infection and fever without focus.

Patients And Methods
We conducted a retrospective study in a tertiary care hospital in Rionegro, Colombia. Children aged 0-36 months with FWS, who were evaluated in our emergency department (ED) between January 2015 to December 2017, were eligible for inclusion in this study. FWS was de ned as a temperature of ≥ 38 8C (less than 10 days of duration to avoid fever of unknown origin with different etiologies to FWS, including noninfectious cause) in the emergency room with no identi ed infectious source during a physical examination at the hospital (without associated respiratory symptoms including rhinorrhea or nasal congestion, diarrhea process, acute otitis media or other ndings on physical examination that allows identifying the source of the fever). All patients with an acquired or congenital immunode ciency or other major chronic condition (oncologic illness, chronic renal failure, transplant patient, sickles cell disease, presence of a mechanical device such as indwelling catheter, ventricle-peritoneal shunt, auditory prostheses, or an invasive diagnostic or therapeutic procedure in the previous 10 days) were excluded due to the different pathogens implicated in the case of FWS and the different management needed in these cases. Patients with ongoing antibiotic treatment or di culties communicating due to language problems were also excluded. We considered serious bacterial infection (SBI) those infants nally diagnosed with bacteremia, urinary tract infection, bacterial meningitis, or pneumonia. In our hospital were adapted since 2010 the clinical guideline from the National Institute for Health and Clinical Excellence (NICE) of how to assess and initially manage feverish illness in children aged under 5 years (12). The decision for inpatient versus outpatient management and antibiotic treatment was supported by clinical and laboratory ndings at the discretion of the treating physician on a patient-by-patient basis. Of each patient, we collected the following data: age, sex, personal history, complete vaccination for Haemophilus in uenzae type b (3 doses at 2,4, and 6 months of age) and Streptococcus pneumoniae (2 doses at 4 and 12 months of age), duration of fever, associated symptoms, temperature, previous consultation in the ED, appearance on arrival, physical examination, supplementary tests, isolated microorganism, nal diagnosis, and disposition of the patient. In all patients were mandatory to obtain white blood cell counts (WBC), absolute neutrophil count (ANC), serum C-reactive protein (CRP), and Urine dipstick. Urinary tract infection (UTI) was de ned as the presence of ≥ 10 4 colony-forming units per milliliter of a single uropathogen, cultured from a urine specimen obtained by bladder catheterization. Other tests (blood culture, urine culture, chest radiograph, cerebrospinal uid examination) were obtained at the discretion of the physician in charge.
Clinical and socio-demographic variables information were compared between groups (SBI and non-SBI) with Student's t-tests or Kruskal-Wallis tests for continuous variables and with Chi2 tests or Fisher's exact tests for categorical variables. Medians are reported with interquartile ranges (IQRs). All tests were two-tailed, and a value of P ≤0.05 was considered statistically signi cant. Odds ratios (OR) and 95% con dence intervals (CI) were calculated for each of the possible predictor variables of SBI using logistic regression models, adjusting for potential confounders. We carried out the statistical analysis using STATA 17.0. Patients with UTI were signi cantly younger than non-UTI. A greater proportion of patients in the UTI group were less than 6 months (44% vs. 15%, p<0.002). A higher proportion of patients in the UTI group had recurrent urinary tract infections (18% vs. 11%) differences were not statistically signi cant (p=0.32). A higher proportion of patients in the UTI group had not good general appearance in admission (36% vs. 22%) but this difference was not statistically signi cant (p=0.10). 36% of the patients in the UTI group had a temperature >39°C respect to 23% of the patients with the same temperature in a non-UTI group. A higher proportion of patients in the UTI group were classi ed as high risk (15% vs 9%) using the Tra c light system for identifying risk of serious illness (NICE) to a non-UTI group. but this difference was not statistically signi cant (p=0.67). There were no differences statistically signi cantly between the group's days with fever, white blood cell count, CPR levels, Table 2.

Discussion
We describe the epidemiology of SBI and UTI in a population of highly vaccinated infants with FWS. Of 137 infants 40 (29%) had SBI and of them 32 (80%) had UTI. The prevalence of SBI in our study was higher than in previous studies (13). In this Colombian prospective study with 101 patients less than 36 months of age the prevalence of SBI was 25%. These discrepancies could be explained by not all patients who underwent hemogram and urinalysis plus urine gram (91 and 64%) in this study; and this could lower the level of detection, especially of UTI. The predominance of UTI are consistent with recent studies conducted in the post-pneumococcal vaccine era in which there was an increase in the rate of UTI and decreasing in the rate of others diagnosis (14) In infants with FWS, always some uncertainty is still encountered because the symptoms are unspeci c and sometimes do little to elucidate whether or not the patient's condition is the result of a urinary tract infection (15). Patients with UTI were signi cantly younger than non-UTI. It was the only variable that distinguished between patients with and without UTI. Interestingly, no single biological marker, such as CRP or WBC, or another clinical variable such as the magnitude of the fever or its duration or general appearance in admission could e ciently identify patients with UTI. This reinforces the need to screen for UTI in all patients with FWS given the frequency of UTI and the poor association with signs, symptoms, or other biomarkers. Most infants with UTI (72%) had positive leukocyte esterase and nitrite in the urinalysis, as well as positive gram staining for germs (64%). In the present study, these parameters were more useful in discriminating patients with or without SBI than CPR or WBC or clinical characteristics. The germ most frequently isolated in the urine cultures was Escherichia coli, con rming it as the most frequent cause of urinary tract infection, a nding similar to that found in other reports (16).
Our study has certain limitations. The sample size was small, which limited to some extent the multivariate analyses and the performance of a predictive model; although this was not the objective of this study from the beginning. Additionally, the retrospective nature of the secondary source information may generate information biases; however, all the information from the patients was available at least to guarantee the completeness of the information presented here.
In conclusion, patients with FWS under 3 years of age, urinary tract infection is the main cause of SBI. A high degree of clinical suspicion should be maintained since neither symptoms, signs, CBC or CPR distinguish children with or without UTI in this population. Future studies should further validate, in a different setting and in a larger population, our results.

Declarations
Ethics approval and consent to participate: The protocol was approved by Review Boards of Clinica Somer and Uniremington University in accordance with the local regulations in clinical research. Informed consent was waived by Review Boards of Clinica Somer and Uniremington University, due to the retrospective nature of this research.

Consent for publication: Not applicable
Availability of data and materials: The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Competing interests: none
Funding: Own fundings