An article by Tanya Babich and colleagues from this week’s Clinical Infectious Diseases describes a study performed Israel looking at outcomes of patients with catheter associated urinary tract infection (CAUTI). This is always a challenging topic because often CAUTIs are reported to national organizations (such as NHSN) for patients that may have asymptomatic bacteria or have an infection at another site. Often, clinically a patient has a positive urine culture may not have a urinary tract infection, especially those with indwelling catheters. Another challenging issue is that there is a high rate of resistance in specimens from patients that have chronic indwelling urinary catheters.
This aim of this study was to look at initial choice of empiric antibiotics and outcomes based on the susceptibility of uropathogens. They only included patients >18 years who had a urinary catheter for at least 7 days, had 10 white blood cells per mL and the uterine and met SIRS criteria with a positive urine culture for typical uropathogens. This was a cohort study and as such the patients were not randomized to any intervention, but were treated as per usual care. Importantly, the investigators did perform a power calculation based on a suspected mortality rate of 30% at 30 days and met their power number.
Important findings from the study included:
- Of the 315 consecutive patients (mean age79), 63% had an indwelling catheter for greater than 30 days, and 40% lived in a long-term care facility
- More than half the patients had received antibiotics within a month prior to the CAUTI
- 55% of the pathogens isolated were multidrug resistant gram-negative rods
- 30.8% of patients died at 30 days, the median survival was 82 days
- 38% of patient had died by 1 year, and 83% by 4 years
Key findings regarding the appropriateness of empiric therapy:
- Overall, only 49% patients received appropriate empirical treatment (antibiotics active against the isolate recovered)
- The receipt of appropriate empiric antibiotics did not increase the 30 day mortality on multivariate logistic regression
- In patients who were propensity-score-matched there was no difference between those who got appropriate antibiotics initially, with regard to morality
- Presence of heart failure, malignancy, increased SOFA score, and increased age did increase 30 day morality by multivariate logistic regression
This study suggests that in patients with catheter associated urinary tract infections, the appropriateness of empiric (initial) therapy does not affect overall outcomes. This is a surprising finding, but given the overall advanced age and illness of the cohort, a signal from appropriate antibiotics may be harder to detect. Another area of possible bias is that the patients may of had undetected alternate causes of there are SIRS, and simply had asymptomatic bacteruria, which would lead to an underpowering of the study.
In addition, many antibiotics are tested for blood concentration susceptibility , and some achieve higher concentrations in the bladder, so some of the antibiotics may have been at a effective concentration at the site of infection, thus leading to a classification of “inappropriate” in patients where enough drug was present to treat the infection.
The authors conclude that in patients who are stable with suspected CAUTI, monitoring without antibiotics may be reasonable approach. This is certainly an area for possible antimicrobial stewardship, but in many cases the patients are too ill to withhold antimicrobial therapy. As noted by their outcome data, these patients who are mostly elderly and debilitated with co-morbidities have a poor prognosis in the short and long-term. As such there may be some cases where this data will support withholding antibiotics but it will not be broadly applicable without data from prospective intervention.
-JAV
Reference:
ID PittStop 
Leave a comment