Published in the August 2016 issue of Today’s Hospitalist
CONSIDER THIS familiar scenario: It’s your first day on service, and one of your patients is an 87-year-old woman who lives in a nursing home and was admitted from the ED 40 hours ago for altered mental status and presumed UTI. Antibiotics were prescribed, and the patient is resting comfortably. After a quick review of her labs and vitals and a brief exam, you continue all medications and move on.
What’s wrong with this picture? A close look at her labs indicates that the urine culture done after the ED’s urinalysis did find urogenital flora—but it didn’t actually grow any uropathogens indicating a UTI. The antibiotics should have been discontinued, but weren’t.
According to a study published in the March issue of the Journal of Hospital Medicine, the inappropriate use of antibiotics for suspected UTI—whether ordered in the ED or continued on the floors—is pervasive. In a retrospective chart review of more than 1,100 patients admitted through the ED to Baltimore’s Johns Hopkins Bayview Medical Center, almost half had a urinalysis done before admission. Among those patients, one-third had positive findings of pyuria, leukocyte esterase or nitrites, and among those with positive findings, the vast majority began antibiotic treatment for a UTI in the ED. However, more than half those patients (58%) had no indication for antibiotics such as fever or urinary symptom or sign.
“We should give up the idea that all bacteria in the urine are bad and should be treated.”
Following admission, most patients started on antibiotics in the ED continued taking them, even though continuation was inappropriate for two-thirds, based on culture results and absence of symptoms. Among patients receiving antibiotics who had no indication for UTI, 25% had an admitting diagnosis of altered mental status.
Because no guidelines exist for treating UTIs in the ED, study researchers used criteria from the Infectious Diseases Society of America, the CDC and the Society for Healthcare Epidemiology of America to determine whether antibiotics were appropriate. Antibiotics were deemed appropriate if the patient had a positive urinalysis and one of the following symptoms: fever; a UTI sign or symptom such as frequency, urgency, dysuria, or suprapubic or costovertebral angle tenderness or pain; altered mental status in the presence of a chronic urinary catheter; or one of two indications for treating asymptomatic bacteriuria: pregnancy or a planned urologic procedure.
Today’s Hospitalist spoke with study co-author Robin McKenzie, MD, an associate professor of medicine who heads Bayview’s antimicrobial stewardship program, about the findings.
What contributes to this inappropriate prescribing?
The big problem is that asymptomatic bacteriuria is common, particularly among the frail elderly, who have rates as high as 40% to 50%. Many providers don’t realize that bacteria can exist in the urine asymptomatically and not need to be treated. In fact, several studies suggest that asymptomatic bacteriuria may be protective and that eradicating these harmless bacteria can actually lead to symptomatic infection. We should give up the idea that all bacteria in the urine are bad and should be treated. And because asymptomatic bacteriuria is often accompanied by pyuria, the presence of pyuria is not a reason to administer antibiotics.
Another issue: The ED is focused on getting a diagnosis and a disposition. Many EDs now have standing order sets for certain labs, and many include a urinalysis.
Having lab information upfront enables ED providers to complete the evaluation more quickly. But having unnecessary information often leads to inappropriate diagnostic procedures and treatment. For instance, seeing pyuria on a urinalysis may lead to an unnecessary urine culture and inappropriate treatment. Many of the urinanalyses in our study were obtained via order sets, and we have re-evaluated those order sets and taken urinalysis out of many of them.
Frail, elderly individuals often have asymptomatic pyuria and bacteriuria, and they are at high risk for becoming delirious. When they become delirious, pyuria is often found. In this common situation it may be convenient to diagnose a UTI, but it is dangerous because other diagnoses may be missed. While there is a fear of withholding needed antibiotics, there is less consideration of the harm of unnecessary antibiotics—harm to the patient, the institution and the community.
Why are antibiotics continued if there’s no evidence of a UTI?
Probably several things are going on, but the main one is that hospitalists and housestaff aren’t re-evaluating the indication for antibiotics. The inpatient team generally has the advantage of urine culture results 24 hours later. But in many cases when the culture is negative, antibiotics are continued, and providers just go along with the plan started in the ED.
It’s hard to go into a patient’s room and say, “I know we told you that you had a UTI. But it turns out, you don’t, and the antibiotics we gave you aren’t doing you any good and could do you some harm.” Or hospitalists or housestaff might not question another provider’s diagnosis, or they might mistake the patient’s coincidental improvement as evidence of the antibiotic’s efficacy.
How should physicians try to reduce inappropriate antibiotics?
ED providers should know that asymptomatic bacteriuria is common and, in most instances, should not be treated.
And doctors should keep in mind that many things can cause altered mental status. For patients who have no urinary sign or symptom or fever, tachycardia, hypotension, or leukocytosis, then a UTI probably is not causing the altered mental status. Instead, they should consider medications, volume status or electrolyte changes.
If the patient is very stable except for a change in mental status, it may be appropriate to follow carefully without antibiotics. On the other hand, if a patient looks sick and there is no other obvious cause, I certainly agree it’s better to start antibiotics and then reassess.
After admission, it is very important to look at culture results and re-evaluate the patient. At 24 hours post-admission, hospitalists should reassess the need for continuing antibiotics started in the ED.
Would it help to track inappropriate prescribing?
Yes, but to evaluate the appropriateness of prescribing, you have to review individual records, and many hospitals do not have the resources to do that. You could initiate educational interventions to stress the frequency of asymptomatic bacteriuria and the importance of not treating it, in most cases.
The key message is to re-evaluate patients after admission. If the antibiotics aren’t needed, hospitalists can go to patients and say, “We have good news. You don’t have a UTI, and we don’t have to continue those antibiotics.”
Bonnie Darves is a freelance health care writer based in Seattle.