To the Editor—The prevalence of asymptomatic bacteriuria (ASB) varies widely based on the studied population. Currently, the prevalence of ASB in patients hospitalized in acute care institutions is unknown. Awareness of the prevalence of ASB in this setting would be useful in both medical decision making as well as public reporting of hospital-acquired urinary tract infections. In this prevalence study, 200 randomly selected patients admitted in April/May 2013 to a tertiary care academic center had urine samples collected for culture within 24 hours of being admitted. Data from the medical records were collected during these hospitalizations up to 30 days post-enrollment. The objective was to determine the prevalence of ASB. Of the 200 patients, 17 were found to have ASB for a prevalence of 8.5%.
Because infections acquired during a hospital stay are not always reimbursed by insurers, knowing what conditions were present on admission can be relevant from the hospital’s perspective. ASB, usually defined as 1 (in men) or 2 separate (in women) urine samples with microbial growth above a certain threshold in the absence of typical urinary tract symptoms, is such a condition. While the prevalence of ASB in patients hospitalized in acute care institutions is currently unknown, it has been determined in other populations and ranges from 1% to 5% in healthy premenopausal women to 100% in long-term catheterized patients. 1 ASB should not routinely be screened for; however, if it is first detected during the hospital stay after a catheter is placed or during a fever, it can easily be misinterpreted as healthcare-associated. 2 ASB is not a treatment indication (with few exceptions) but inappropriate antibiotic administration for ASB is common and associated with higher occurrence of antibiotic-resistant bacteria generating a major opportunity for antimicrobial stewardship. 3 , 4 Our objective was to determine the prevalence of ASB among patients admitted to an academic medical center.
We conducted a prevalence study from April 1 to May 31, 2013, and included 200 adult patients who were admitted to Barnes-Jewish Hospital, a 1,250-bed university-affiliated tertiary care center in St. Louis, Missouri, for a variety of reasons with the exception of a UTI diagnosis (or compatible symptoms). To apply inclusion and exclusion criteria, a convenience sample of 5–10 newly admitted patients were interviewed within 24 hours and asked for any urinary tract symptoms. Other exclusion criteria were fever ≥38°C of unknown etiology (because UTI could be part of the differential diagnosis) and patients unable to communicate their symptoms. After obtaining informed consent, a midstream clean-catch urine sample was collected in the same 24-hour time window and was evaluated for urinalysis using a dipstick test and routine culture. A positive urine culture was defined as a single urine sample with microbial growth of >105 colony-forming units of a single organism. 2 Data from the medical record were collected during the patient’s hospitalization, ending 30 days post-enrollment (if the patient was still admitted at that point). We considered a sequence of 200 enrolled patients in the order of their admission to the hospital (without prior sample size calculation). The results were not shared with the treating physicians. The Washington University Institutional Review Board approved the study.
Of the 200 included patients, 110 were women (55%). The mean age was 47.8 years (±16.5). Most patients were white (112; 56%) or African-American (83; 41.5%). The admitting service was general medicine in 139 patients (69.5%) and neurology in 19 patients (9.5%), with comparatively fewer patients admitted to surgical services. In addition, 41 patients (20.5%) carried a diagnosis of diabetes mellitus. None of the patients had a urinary catheter in place on the day of admission.
Of the 200 patients, 17 (8.5%) were found to have ASB; all 17 were women. Another 102 (51%) patients had positive urine cultures but with insignificant growth according to the definition set forth above. The retrieved organisms are shown in Table 1. Comparing patients with ASB versus no ASB, there were no differences in age or race. Both the proportion of patients admitted to the ICU during their stay (1 patient of 17 patients [5.9%] with ASB vs 8 patients of 183 patients [4.4%] without ASB; P=.7) and the overall length of hospital stay (3 days [range, 2–10] vs 3 days [range, 2–34]; P=.7) were similar. Only 1 of the 200 patients was diagnosed with a UTI over the course of hospitalization, and 1 fatality occurred among the cohort; both of these occurred in the non-ASB group. During their hospital stays, 2 patients had a urinary catheter, and 14 of the 200 were receiving antimicrobials on admission (all in the non-ASB group). No ASB patient received therapy, as the culture results were not disclosed to treating physicians.
NOTE. CFU, colony-forming units.
We found the prevalence of asymptomatic bacteriuria to be 8.5% in a general hospital population on the day of admission, with all affected patients being women and Enterobacteriaceae being the most common pathogen group. This rate is similar to data from other populations; however, to our knowledge, ASB prevalence has never been determined for acute care hospital admissions. 5 – 7 The significance of the 51% samples with growth in urine cultures below the threshold is unclear; none of them developed a symptomatic UTI while being admitted. When obtaining urine cultures in patients admitted to an acute care hospital, providers should be aware that approximately 1 in 10 may arrive with ASB.
Our study had several limitations. It was a single-center study, and enrolled patients were middle-aged adults admitted mostly to lower acuity wards under general medicine or neurology services. Thus, the findings are difficult to generalize. Regardless, when working up a possible infection over the hospital course that would qualify as hospital-acquired, the possibility of ASB that was present on admission should be given consideration.
Given that a substantial number of patients receive unnecessary antibiotics while hospitalized, these results serve as a reminder that the clinical picture must not be forgotten when interpreting laboratory findings. This is particularly relevant for positive urine cultures, a common justification for starting antibiotics irrespective of symptoms. 8 – 10 Raising the awareness of ASB and its potential misinterpretation as nosocomial bacteriuria can lead to lower antibiotic consumption and thus decrease the development of antimicrobial resistance.
Financial support: This work was supported by the National Institutes of Health (grant nos. KL2RR024994 and KL2TR000450 to J.M.), the Burroughs-Wellcome Fund Career Award for Medical Scientists (to J.P.H.), the National Institutes of Health (grant no. R01DK099534 to J.P.H.), the CDC Prevention Epicenters Program (grant no. 5U54CK000162 to J.M.). In addition, J.M. was supported by the Barnes-Jewish Hospital Patient Safety and Quality Fellowship Program.
Potential conflicts of interest: All authors report no conflicts of interest relevant to this article.