Hello. I am Paul Auwaerter, for Medscape Infectious Diseases, speaking from the Division of Infectious Diseases at Johns Hopkins University School of Medicine. Today I want to talk about asymptomatic bacteriuria.
Many of us in the ID field know that if we use the term "asymptomatic bacteriuria," we are not supposed to give antibiotics to that patient. When this comes up on the wards and I am working with medical students or residents, the admonition "Don't do something; stand there" often comes into play. However, many physicians have itchy prescribing fingers or, when faced with test results that show bacteria in the urine, many feel the urge to treat regardless. Even though guidelines[1]strongly advise against treatment, information suggests that up to 80% of people with asymptomatic bacteriuria get antibiotics.[2] This represents antibiotic abuse.The guidelines for asymptomatic bacteriuria have been available since 2005.[1] These guidelines recommend treatment only in certain pregnant woman or those who may be having a genitourinary procedure with anticipated bleeding. This is a pretty small group. You can talk about education gaps and so on, but really the guidelines do not seem to be making a dent in this message.Albert Einstein defined insanity as doing the same thing time and again and expecting different results. Thus, I read with some interest a recent Clinical Infectious Diseases article by Leis and colleagues[3] at the University of Toronto. Their approach was to hide the data from noncatheterized patients' urine samples. They would not report the urine culture results on these patients but instead required physicians to call the microbiology lab to get the results. There certainly is a precedent for this. Many hospital laboratories now hide certain microbiology results -- an organism's susceptibility to some broader-spectrum antibiotics, for example -- from culture data reports, in hopes of limiting those antibiotics from being used inappropriately. Here at Johns Hopkins, for example, our laboratory does not report Western blot bands with its Lyme disease testing results. The report is a simple yes or no; you have a positive result or a negative result, so there is less tendency for misinterpretation.Leis and colleagues studied 415 people; only 2% of those people met criteria for urinary tract infection (UTI). They found that even with this baseline 2% rate of asymptomatic bacteriuria, 48% of the patients in their hospital were getting antibiotics. That declined to 12% with their pilot study, in which providers had to call the lab to procure culture results. That extra step led to a higher predictive value, that people would not take the time to call the microbiology lab, and therefore would not receive results representing asymptomatic bacteriuria and leading to treatment.Although this was a limited study, I found it very interesting. This was only a 16-week study, and although there was no evidence of missed UTIs, the study clearly needs to be expanded to fully assess negative effects of this approach to care. Patients could still receive empiric antibiotics if they were quite ill and were thought to have an active UTI. An accompanying editorial[4] uses the term "asymptomatic paternalism." This obviously refers to someone who knows better and is therefore trying to foster and change behaviors. This is similar to a kind of parenting. It could be called "saving ourselves from our best misguided intentions."Of interest, this paper did not mention the Choosing Wisely® campaign. Choosing Wisely suggests not ordering urine cultures at all,[5] which would be an even better step. Of course, that is another story. The American Geriatrics Society has said that not ordering urine tests or cultures on elderly patients without symptoms is one of their top 5 recommendations.We still have far to go, but I think these are the sorts of measures that are needed and are more interesting in terms of affecting behavior and strategies for our patients. Thanks for listening.
Many of us in the ID field know that if we use the term "asymptomatic bacteriuria," we are not supposed to give antibiotics to that patient. When this comes up on the wards and I am working with medical students or residents, the admonition "Don't do something; stand there" often comes into play. However, many physicians have itchy prescribing fingers or, when faced with test results that show bacteria in the urine, many feel the urge to treat regardless. Even though guidelines[1]strongly advise against treatment, information suggests that up to 80% of people with asymptomatic bacteriuria get antibiotics.[2] This represents antibiotic abuse.The guidelines for asymptomatic bacteriuria have been available since 2005.[1] These guidelines recommend treatment only in certain pregnant woman or those who may be having a genitourinary procedure with anticipated bleeding. This is a pretty small group. You can talk about education gaps and so on, but really the guidelines do not seem to be making a dent in this message.Albert Einstein defined insanity as doing the same thing time and again and expecting different results. Thus, I read with some interest a recent Clinical Infectious Diseases article by Leis and colleagues[3] at the University of Toronto. Their approach was to hide the data from noncatheterized patients' urine samples. They would not report the urine culture results on these patients but instead required physicians to call the microbiology lab to get the results. There certainly is a precedent for this. Many hospital laboratories now hide certain microbiology results -- an organism's susceptibility to some broader-spectrum antibiotics, for example -- from culture data reports, in hopes of limiting those antibiotics from being used inappropriately. Here at Johns Hopkins, for example, our laboratory does not report Western blot bands with its Lyme disease testing results. The report is a simple yes or no; you have a positive result or a negative result, so there is less tendency for misinterpretation.Leis and colleagues studied 415 people; only 2% of those people met criteria for urinary tract infection (UTI). They found that even with this baseline 2% rate of asymptomatic bacteriuria, 48% of the patients in their hospital were getting antibiotics. That declined to 12% with their pilot study, in which providers had to call the lab to procure culture results. That extra step led to a higher predictive value, that people would not take the time to call the microbiology lab, and therefore would not receive results representing asymptomatic bacteriuria and leading to treatment.Although this was a limited study, I found it very interesting. This was only a 16-week study, and although there was no evidence of missed UTIs, the study clearly needs to be expanded to fully assess negative effects of this approach to care. Patients could still receive empiric antibiotics if they were quite ill and were thought to have an active UTI. An accompanying editorial[4] uses the term "asymptomatic paternalism." This obviously refers to someone who knows better and is therefore trying to foster and change behaviors. This is similar to a kind of parenting. It could be called "saving ourselves from our best misguided intentions."Of interest, this paper did not mention the Choosing Wisely® campaign. Choosing Wisely suggests not ordering urine cultures at all,[5] which would be an even better step. Of course, that is another story. The American Geriatrics Society has said that not ordering urine tests or cultures on elderly patients without symptoms is one of their top 5 recommendations.We still have far to go, but I think these are the sorts of measures that are needed and are more interesting in terms of affecting behavior and strategies for our patients. Thanks for listening.