Only one-third of patients who go to the emergency department with suspected urinary tract infection (UTI) actually have evidence of this infection, but almost all patients were treated with antibiotics, without no use, contributing to the development of antimicrobial resistance, according to the study done by Dr Laura Shallcross, University College London, UK and colleagues, presented at European Congress of Clinical Microbiology & Infectious Diseases (ECCMID) in Amsterdam, Netherlands.
Concern over delaying antibiotic treatment for severe infection means that clinicians have a low threshold for initiating antibiotics in the emergency department (ED) for patients with suspected urinary tract infection (UTI) syndromes. Although a non-infectious cause is established for many of these cases, antibiotics are often continued unnecessarily, which drives the emergence of antimicrobial resistance (AMR).
‘Antibiotic usage among patients entering the emergency department with suspected UTI should be reviewed properly, to reduce the incidence of inappropriate antibiotic usage and thereby development of resistance.’
In this study, the authors estimated the frequency of over-diagnosis of UTI syndromes in the ED, in order to estimate the potential to reduce antibiotic prescribing by stopping antibiotics early for patients with no evidence of bacterial infection.
The authors undertook a cohort study in a large teaching hospital (the Queen Elizabeth Hospital, Birmingham, UK), using electronic health records (EHRs) from patients with suspected UTI syndromes who attended the ED. Individuals who had a sample submitted for microbiological culture of urine in the ED were eligible for inclusion.
The research team randomly selected a subset of 1000 patients (700 admitted to hospital) and described the clinical and demographic characteristics of this population. They then compared diagnoses made by the ED physician to clinical diagnosis based on urinary symptoms and microbiological outcomes and international classification of disease (ICD-10) diagnostic codes. Finally, they estimated how often antibiotics were stopped at or shortly after (<72 hours) admission to hospital.
A total of 943 patients were eligible for the study. Of these, 289 patients had an ED diagnosis of UTI syndromes including 56 cases of pyelonephritis, 42 cases of urosepsis and 191 cases of lower UTI.
Treatment with empirical antibiotics was recorded for 173 (91%) of patients with an ED diagnosis of lower UTI, but only 63 of these cases (36.4%) had clinical evidence of UTI. ICD-10 diagnostic codes were available for 83 patients with lower UTI who had been admitted to hospital. Of these, more than 40% (34/83) had a primary diagnostic code for a non-infectious condition, suggesting antibiotic treatment was not required.
The authors say: "A focus on antibiotic review in patients with an emergency department diagnosis of suspected urinary tract infections could support reductions in inappropriate antibiotic prescribing in secondary care, and help reduce the impact of unnecessary prescribing on the development of antibiotic resistance."
They add that it is difficult to generate accurate estimates of overprescribing. "Probably the most robust estimates are derived national survey data in the USA* which includes emergency department and outpatient settings," explains Dr Shallcross. "These estimated that around 30% of antibiotics that are prescribed in these settings are inappropriate. More recently, a study from Australia suggested that at least one third of emergency department prescribing is inappropriate.**" Our estimates of inappropriate prescribing of between 60-70% are higher because we have measured how often an emergency department diagnosis of UTI is supported by clinical evidence."
Going forward, the authors say that antimicrobial stewardship initiatives such as Start Smart then Focus in England recommend that all antibiotic prescriptions should be reviewed, taking account of the patient's clinical progression and microbiological results. Dr Shallcross concludes: "Our study highlights the potential impact on total antibiotic prescribing -- and antibiotic resistance -- that could be achieved by widespread adoption of such initiatives."