Urinary Tract Infection in Young Infants: Practice Patterns in Evaluation and Treatment

Using the electronic health record, we identified infants aged ≤60 days with UTI who were hospitalized at a tertiary care children’s hospital from January 1, 2013 through May 7, 2018. The study was approved by the institutional review board; informed consent was waived for a retrospective study. We used encounter diagnosis codes to identify all patients 0 to 60 days old who had fever or UTI. Of these, we only included patients who were hospitalized, had a urine culture obtained, and had a final discharge diagnosis of UTI. Laboratory and/or microbiologic evidence of UTI was not required for inclusion. Exclusion criteria included treatment of a separate source of infection requiring antibiotic therapy (eg, bacterial meningitis), chronic severe systemic disease (including neuromuscular disorder, congenital or acquired immunodeficiency, malignancy, quadriplegia or paraplegia, hydrocephalus or neurologic malformations, or organ transplant), infants in whom urinary tract foreign body exposure might have introduced infection (indwelling genitourinary devices, genitourinary tract instrumentation, or surgical intervention within the previous 7 days), or a diagnosis of hospital-acquired UTI.⁶  Data were included from the first admission and not subsequent admissions for readmitted patients.

Variables were extracted from the participant’s chart by reviewing documentation in notes from the admission history and physical and the emergency medicine provider’s documentation (Supplemental Fig 1). Four pediatric hospitalists performed the chart reviews. Questions about data extraction were settled by group consensus; interrater reliability was not assessed. For the purposes of analysis, an undocumented subjective variable in the history or physical examination was considered a negative response. Vital signs were recorded on the basis of the first data documented. Antibiotic therapy was abstracted from the medication administration record.

Demographic and clinical variables were described by using counts (proportions) and medians (interquartile range [IQRs]). Primary outcome was duration of parenteral therapy. Secondary outcomes were length of stay (LOS) and readmission rate. LOS was determined by subtracting the time of the first vital sign from the time of discharge order placement in hours. Duration of parenteral therapy was calculated by subtracting the time of administration of last dose of intravenous antibiotic from the time of the first dose in hours. Readmission rate was calculated on the basis of any admissions within 30 days of the initial admission that had UTI as a clinical diagnosis.

Associations between demographic and clinical characteristics and outcomes were evaluated. Age was categorized as ≤28 days and 29 to 60 days; duration of fever was categorized as ≥24 and <24 hours; bacteremia was categorized as either present or absent. A χ² test or Fisher exact test was used for categorical variables, and the Mann–Whitney test was used for continuous variables. We used a generalized linear model to explore whether duration of antibiotic therapy and LOS were associated with age, duration of fever, irritability, lethargy, and presence of bacteremia because these variables were significantly associated with 1 or both outcomes in the univariable analysis.

One hundred ninety-three subjects were included in this study (Supplemental Fig 2). Sixty-nine (35.8%) infants were ≤28 days. Participants were 54.4% male; 59.6% were white, 18.7% were African American, and 21.7% were other or unspecified race. The median duration of parenteral antibiotic therapy was 59 (IQR: 43–114) hours. The median duration of total antibiotic therapy was 13 (IQR: 10–14) days. The median LOS was 71 (IQR: 57–127) hours (Table 1).

Univariable analysis did not reveal a statistically significant relationship between age and either duration of fever or presence of bacteremia (Supplemental Tables 4 and 5). Younger patients (age ≤28 days) and patients with bacteremia had longer durations of parenteral antibiotics and LOS (P < .0001 for all associations) (Table 2). Lethargy on initial examination was associated with LOS and duration of antibiotics. Irritability and duration of fever were associated with LOS only. Prematurity, other signs of ill appearance on initial examination (poor capillary refill or poor tone), and abnormal imaging (including renal ultrasound and/or voiding cystourethrogram) were not associated with either LOS or duration of parenteral therapy.

In multivariable analysis, age, presence of bacteremia, irritability, and lethargy, but not duration of fever, were independently associated with the duration of parenteral antibiotic therapy and LOS (Table 3). Patients ≤ 28 days had longer duration of parenteral antibiotic (mean duration was 1.65 times that of older patients, P < .0001) and longer LOS (mean duration was 1.60 times that for older patients, P < .0001). Clinical findings of bacteremia, lethargy, and irritability on initial examination were also found to be independently predictive of both duration of parenteral therapy and LOS.

Nine infants (4.6%) were readmitted within 30 days, and 5 (2.6%) of these readmissions were for UTI recurrence. Four of the readmissions for recurrent UTI were due to the same organism causing the initial admission, 3 had concomitant bacteremia on initial admission, and 3 had abnormal genitourinary anatomy. There was no association found between the duration of parenteral antibiotics in readmitted and non-readmitted infants (estimated ratio of means: 1.23, 95% confidence interval: 0.68–2.22).

In this single-center retrospective analysis of UTI management in young infants, age ≤28 days, presence of bacteremia, irritability, or lethargy on initial examination, and fever duration ≥24 hours were all associated with longer parenteral antibiotic therapy and/or LOS, whereas urinary tract abnormalities and prematurity were not. Although there was variability, the duration of parenteral therapy and LOS was generally short (median 59 hours and <3 days, respectively). This is consistent with recommendations published for UTI treatment in infants aged <1 month because it is approximately the amount of time it would take for blood and cerebrospinal fluid cultures to result.⁷ 

The presence of irritability or lethargy on initial examination is suggestive of a more ill infant, which would reasonably lead to a longer duration of parenteral therapy and therefore LOS. However, we also found that age group, independent of other clinical considerations, led practitioners to prolong parenteral therapy. There is no evidence that a longer duration of parenteral antibiotic therapy is needed for treatment of UTI in infants ≤28 days of age. For older infant populations, shorter durations of parenteral antibiotics are not associated with treatment failure.^4,8  Authors of a retrospective study of infants age ≤60 days admitted for UTI showed that the use of parenteral antibiotics for ≥4 days decreased from 50% to 19% from 2005 to 2015, with no impact on readmission rate.⁹  Researchers of a multicenter retrospective study of infants age ≤60 days with bacteremic UTIs also demonstrated no advantage to long-course parenteral antibiotics (>7 days) with respect to recurrence or readmission.¹⁰  The importance of bacteremia in the setting of UTI is debated; researchers in multiple studies have suggested that the clinical symptoms and outcomes are not significantly different in bacteremic UTIs.^11,12  Duration of fever, although often used to approximate a response to therapy, has not been shown to be predictive of any difference in clinical course in young children with UTI.¹³ 

There are several limitations to our study. This was a single-center study at an academic institution, and our sample size was small. Some of the variables were dependent on documentation practices and therefore possibly inaccurate. Readmissions could have occurred at hospitals outside of our system. It is possible that infants 29 to 60 days with UTI could have been managed as outpatients, although we anticipate this would be a small number because the typical practice in our hospital is admission of all infants ≤60 days requiring antibiotics according to institutional clinical practice guidelines. Because we were most interested in the practices of providers after diagnosing a UTI, we did not limit our cohort with regard to urinalysis or microbiologic data, and it is possible that some of the infants did not have a UTI.

Without clear guidelines for management of hospitalized young infants with UTIs, age remains a driving factor in the duration of parenteral antibiotic treatment and LOS in our hospital. The risks associated with prolonging hospitalization need to be weighed with the risk of transitioning to oral antibiotics earlier. With our findings, we provide further support for relatively short durations of parenteral intravenous therapy in young infants with UTI.

We thank Stacey Hay and Kim Gajewski for database design and build, Brett Bordini for assistance with revisions of the article, and Alyssa Stephany and Kelsey Porada for assistance in preparation of the article.