Urinary tract infections in nursing homes are caused by resistant flora

February 2007

An abstract presented at 46th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy raised a number of important issues concerning urinary tract infections caused by quinolone-resistant Escherichia coli in patients in long-term care facilities.

In the abstract, the researchers reported that in their health care organization, susceptibility to ciprofloxacin of E. coli originating from urinary tract was only 43% in nursing homes, compared with 70% in acute care settings. This finding should not be surprising, especially if hospital strains were not restricted to nosocomial urinary tract infections.

Although increased antimicrobial resistance would be expected for both hospital-acquired and nursing home-acquired infections, isolates from the hospital microbiology laboratory would include both community-acquired (clinic patients and new admissions) and hospital-acquired isolates.

Nursing home residents

Long-term–care nursing home residents represent a relatively stable group of patients who are replaced primarily when they die. They are frequently treated in emergency departments, hospital outpatient facilities and as inpatients, thereafter returning to the nursing home in ping-pong–like fashion. They are frequently credited with bringing resistant organisms into the hospital where, it often turns out, they may have acquired the organism in the first place. Examples are methicillin-resistant

Staphylococcus aureus, vancomycin-resistant enterococci and, the subject for discussion here, multidrug-resistant gram-negative bacilli.

Most community-acquired urinary tract infections are caused by E. coli that are susceptible to the fluoroquinolones, though the susceptibility has been decreasing. Hospital- and nursing home-acquired urinary tract infections are much less likely to be caused by E. coli.

Although E. coli have been the single most common urinary isolate reported in some nursing home studies, most isolates found in nursing homes consist of other Enterobacteriaceae,

Pseudomonas, enterococci and staphylococci. Nursing home-acquired E. coli (and other urinary tract organisms) are far more likely than community-acquired E. coli to be resistant to multiple antimicrobial agents, including fluoroquinolones.

Since hospital urinary isolates come from a mixture of community-acquired, hospital-acquired and nursing home-acquired infections, it is apparent that there will be less resistance than among isolates that are hospital-acquired or nursing home-acquired. The higher the proportion of community-acquired infection, the less resistance would be expected.

To determine where the resistant E. coli in nursing home patients are first acquired, it would be necessary to prospectively study stool and urine cultures from newly admitted residents. It would also be necessary to track cultures and susceptibilities over a period of time, including before and after hospital admissions or other contacts with health care-associated facilities. To my knowledge, such a study has not been done.

Prevalent gram-negatives

The most important message from the abstract is the high prevalence of resistant gram-negative bacilli in nursing homes, and the implied admonition to not assume susceptibility to quinolones in this population. In my nursing home, for example, a state Veteran’s Affairs home with about 80% males and very little use of indwelling catheters, ciprofloxacin is only effective against about 40% of all urinary pathogens isolated from residents with symptomatic urinary tract infections. Asymptomatic bacteriuria is not sought or treated.

The oral drugs cephalexin, trimethoprim/sulfamethoxazole and amoxicillin/clavulanic acid, each cover about 60% of urinary pathogens. Even restricting the analysis to E. coli, these three agents are active against a higher proportion of organisms than ciprofloxacin. Our low susceptibility rate to quinolones is present despite the lack of empiric use of quinolones for urinary tract infection for years. However, as in many nursing homes, a respiratory fluoroquinolone (i.e. levofloxacin) is our drug of choice for empiric treatment of pneumonia.

We have also found that patients frequently return from the hospital or urologist’s office having received ciprofloxacin (often inappropriately) for treatment or prevention of urinary tract infection. Therefore there is widespread exposure to fluoroquinolones.

It is important to remember that most of the urinary isolates in nursing homes are not E. coli and that conclusions regarding empiric therapy for treatment of urinary tract infections should not be made from susceptibilities patterns for only E. coli. Rather, it is important for each nursing home to track the susceptibility patterns of all urinary tract isolates within its own facility. It should also be noted that recent exposure to an antibiotic in a patient is a strong predictor of resistance to that antibiotic.

Perhaps the most important message to stress is that overuse of antibiotics should be avoided. There is no indication for treatment of asymptomatic bacteriuria or pyuria with or without a catheter. This is true for patients with diabetes, as well as those without diabetes. Furthermore, patients with no urinary tract symptoms and no catheter who develop fever are unlikely to have the fever as a result of a urinary tract infection. Asymptomatic bacteriuria has a prevalence of at least 10% in men and 30% in women in nursing homes. Thus, even when bacteriuria is subsequently found, it is likely to be asymptomatic bacteriuria in association with an unrelated cause of the fever.

For more information:

• Donald Kaye, MD is a Professor of Medicine at Drexel University College of Medicine in Philadelphia, Medical Director and Staff Physician at Southeastern Veterans Center in Spring City, Penn. and a Member of the Infectious Disease News Editorial Advisory Board.
• Kwan L, Malyuk R, Romney M. High rates of quinolone resistant E. coli causing urinary tract infections in long-term care: A reservoir of risk for acute care? Poster K-175. Presented at: 46th annual Interscience Conference on Antimicrobial Agents and Chemotherapy. Sept. 27-30, 2006. San Francisco.