Access to POEMs and Essential Evidence Plus will no longer be included in CMA membership as of Dec. 1, 2023.
Clinical Question
Do older persons who reside in care facilities fare better if they receive antibiotics for asymptomatic bacteriuria?
Bottom line
The existing data, limited to a few small older studies at moderate to high risk of bias, show that antibiotics given to adults who develop asymptomatic bacteriuria while residing in a care facility will “clean the urine” but do nothing for clinically important outcomes. And they will cause adverse events. (LOE = 2a-)
Overuse alert: This POEM aligns with the Choosing Wisely Canada campaign called Using Antibiotics Wisely in Long-Term Care, which provides practice change recommendations on reducing unnecessary antibiotic use for asymptomatic bacteriuria.
Reference
Study design: Meta-analysis (other)
Funding: Self-funded or unfunded
Setting: Nursing home/extended care facility
Synopsis
These authors searched several databases and registries to identify randomized trials and observational studies that compared antibiotics versus no antibiotics to treat older persons who developed asymptomatic bacteriuria while residing in a care facility. The authors also performed supplementary searches to identify unpublished studies. Ultimately, they included 9 studies with 1391 participants: 5 were “traditional” randomized trials; 2 were quasi-randomized trials; 2 were open-label trials. Curiously (and frustratingly), one quasi-experimental study did not report the intervention. Although both quasi-experimental studies reported using a placebo as the comparator, the rest used no treatment as the comparator. It's interesting that this topic no longer appears to be of interest to researchers as 8 of the included studies were published between 1983 and 1998 and one was published in 2007. The interventions included trimethoprim, trimethoprim-sulfamethoxazole, cephalosporins, fluoroquinolones, and aminoglycosides. Overall, the studies were at moderate to high risk of bias, largely a function of poor reporting. Overall, bacteriological cure occurred more frequently in the participants treated with antibiotics (54.6% vs 19.4%; number needed to treat = 3; 95% CI 3 - 4), but the data were highly heterogeneous (I2 = 81%). Four studies (n = 317) reported that approximately 20% of the participants developed symptoms of a urinary tract infection regardless of intervention, but the data were also quite varied (I2 = 67%). Additionally, in 4 studies (n = 303), adverse events occurred in 0.8% of control patients and 6.5% of those receiving antibiotics (number needed to treat to harm = 18; 10 - 74; no heterogeneity). Seven of the studies reported no differences in mortality, but the follow-up periods ranged from 6 months to 9 years. Only 2 small studies reported complications, so the data are not terribly robust.
Reviewer
Henry C. Barry, MD, MS
Professor
Michigan State University
East Lansing, MI
Comments
Antibiotics for asymptomatic bacteruria in institutionalized
I find it challenging to operationalize the definition of "asymptomatic" since a principle of geriatric medicine is that patient presentation of common ailments can be subtle and atypical. I assume a UTI qualifies as a "common ailment". This is makes it easy to diagnose a "symptomatic" UTI because an elderly patient is "less perky" today than usual.
Urinary bacteria in the elderly
Not an enlightening study.
Asymptomatic bcteriuria
Confirms what I already knew.
bactériurie asymptomatique
Le défi, en particulier avec les patients atteints de troubles cognitifs, est de différentier une bactériurie asymptomatique d'une infection urinaire, afin d'éviter l'utilisation non judicieuse des antibiotiques, mais aussi afin de traiter les réelles infections qui entraînent des impacts cliniques.