Latest / The SHEA Podcast / Culture of Urine Cultures: Limiting Inappropriate Testing & Treatment

Transcript

Culture of Urine Cultures: Limiting Inappropriate Testing & Treatment · The SHEA Podcast

• 0:09
  SPEAKER_02: Welcome to the Diagnostic Stewardship Podcast, brought to you by the Society for Healthcare Epidemiology of America, SHAY, promoting the prevention of healthcare associated infections and antibiotic resistance and seeking to advance the field of healthcare epidemiology and antibiotic stewardship.
• 0:25
  SPEAKER_02: SHAY is excited to launch the second episode of this podcast, The Culture of Urine Cultures, Limiting Inappropriate Testing and Treatment.
• 0:32
  SPEAKER_02: Availability of urine cultures has led to a culture of culturing, irrespective of your analysis being done or of its results.
• 0:39
  SPEAKER_02: Consequently, overtreatment of asymptomatic bacteriuria has become common, leaving patients exposed to antibiotics unnecessarily.
• 0:46
  SPEAKER_02: Additionally, in catheterized patients, this may be leading to an increase of labeling of catheter-associated urinary tract infections.
• 0:53
  SPEAKER_02: This podcast will focus on diagnostic stewardship strategies to better target patients for urine cultures when and if appropriate.
• 1:30
  SPEAKER_02: I'd like to get our conversation started with this question.
• 1:33
  SPEAKER_02: When you think about where the opportunities for diagnostic stewardship exist for urine cultures, what comes to mind and can you relate it to a patient case?
• 1:41
  SPEAKER_00: Thank you very much, Chrissy, for this question.
• 1:43
  SPEAKER_00: Urinalyses and urine cultures are often done in patients admitted to the hospital based on subjective findings, whether it's related to urine color, consistency, or part of a disease workup.
• 1:54
  SPEAKER_00: And it's often not related to infection in the case of urinalysis.
• 1:57
  SPEAKER_00: In addition, when we look at the prevalence of bacteriuria, it really varies by age, gender, and patients' clinical conditions.
• 2:04
  SPEAKER_00: So the vast majority of patients with bacteriuria are asymptomatic, and healthcare providers will administer antibiotics for these patients with positive cultures.
• 2:12
  SPEAKER_00: So the clinicians may erroneously equate bacteriuria with symptomatic urine tract infection.
• 2:19
  SPEAKER_00: The adverse consequences affect the patient who receives the antimicrobials inappropriately.
• 2:24
  SPEAKER_00: So that patient may develop the side effects related to the antibiotic in addition to the risk of clostrobioidase dipicil or of becoming colonized with a resistant organism.
• 2:35
  SPEAKER_00: We may also forget that misdiagnosing a patient as having UTI may mean that we missed the true diagnosis that the patient presented with.
• 2:43
  SPEAKER_00: Finally, I want to mention that the environment and community will also be affected with increased antimicrobial resistance and unnecessary utilization of resources.
• 2:52
  SPEAKER_00: So it's imperative for us to encourage the use of the most optimal tests to make the correct diagnosis.
• 3:00
  SPEAKER_01: Talking about overdiagnosis at ETI, I think I'd like to illustrate that with a case that I've seen that's representative of cases I've seen in so many different institutions through different projects trying to work with antibiotic and diagnostic stewardship.
• 3:14
  SPEAKER_01: And this is a classic scenario where it's an older patient that was nonverbal and really can't move much, requires total care with uh activities of daily living, who developed a fever.
• 3:26
  SPEAKER_01: And the urine uh which was running through a catheter was clearly cloudy and smelly, and so he had a urine culture scent, and the antibiotics were started for the urine.
• 3:36
  SPEAKER_01: Well, still had a fever the next day, but of course we're still waiting for the urine culture result.
• 3:41
  SPEAKER_01: So by the time it comes back, the organism is actually resistant to the antibiotics he's on, which is Cypro.
• 3:46
  SPEAKER_01: But he was changed to Cypro, and not only did he still have a fever at that point, his vital signs became unstable, and he went to the emergency room.
• 3:53
  SPEAKER_01: And at that point, a CAT scan was done, and he actually had necrotizing fasciitis that had blossomed from the edge of a sacral wound that he was being treated for with wound healing.
• 4:05
  SPEAKER_01: So overdiagnosing UTI and of someone whose symptoms were actually coming from a completely different infection led in that case to significant diagnostic delay.
• 4:14
  SPEAKER_01: We see that all the time as clinicians.
• 4:17
  SPEAKER_01: We feel comfortable because we are now covering the urine, but we're missing what the real problem is that the patient is there for.
• 4:24
  SPEAKER_02: It sounds like there are definitely concerns for the possibility of underdiagnosing or not diagnosing the reason why somebody may have presented, but do we think that there may be some underdiagnosis concerns as it relates to urinary tract infections?
• 4:36
  SPEAKER_01: Yes, let me take that from the other side.
• 4:38
  SPEAKER_01: Again, hindsight is 20-20 in all of these cases, but we know from our own work of looking at 10,000 urine cultures sent in different hospital settings that when patients have bacteria related to a urinary catheter, we do not currently know how to predict which of those patients will become bacteremic.
• 4:56
  SPEAKER_01: They can actually have asymptomatic catheter-associated bacteria and then present with bacteria and sepsis, or they can have catheter-associated UTI symptoms, mainly fever, and present with sepsis of urinary origin.
• 5:09
  SPEAKER_01: We don't have a way to mark right now when the catheter-associated bacteria progresses and becomes bacteremic and something to worry about.
• 5:16
  SPEAKER_01: So there's always that concern in the back of your mind as a clinician should I be treating this?
• 5:20
  SPEAKER_02: Right.
• 5:21
  SPEAKER_02: And with regards to any limitations that we have to our current diagnostic tests for UTI, can you think of any examples of those?
• 5:29
  SPEAKER_01: Well, three.
• 5:30
  SPEAKER_01: First of all, the current diagnostic test for UTI don't take into account symptoms.
• 5:33
  SPEAKER_01: They can only tell you if there's bacteria or pyuria in the urine or not.
• 5:37
  SPEAKER_01: That asymptomatic bacteria and symptomatic UTI both have bacteria and pyuria.
• 5:42
  SPEAKER_01: So the test cannot distinguish between those two.
• 5:46
  SPEAKER_01: Number one, no symptoms.
• 5:47
  SPEAKER_01: Number two, there's a delay in results.
• 5:49
  SPEAKER_01: You don't get the results of a urine culture back for two days, that's like an eternity with an inpatient, and you've already had to make your treatment decisions.
• 5:56
  SPEAKER_01: So once antibiotics are started, it's really hard to stop them.
• 5:59
  SPEAKER_01: And point three is interpretation of results.
• 6:02
  SPEAKER_01: When they come back and people see, oh, there's this highly resistant kleb CL in the urine, that's going to trigger antibiotic use even if the patient didn't have symptoms.
• 6:10
  SPEAKER_00: I fully agree with what you're saying, Barbara.
• 6:13
  SPEAKER_00: I think the pretest probability is the most important area that we need to focus on.
• 6:19
  SPEAKER_01: Let's use a case to illustrate the challenges with diagnostic testing for UTI.
• 6:23
  SPEAKER_01: A 72-year-old man with mild dementia and obesity was admitted to my team for acute onset of nausea, vomiting, unable to keep food down.
• 6:31
  SPEAKER_01: Because he complained of sevague abdominal pain, he had urine tests sent.
• 6:36
  SPEAKER_01: And the urine test showed a pyuria and hematuria on the urinalysis.
• 6:39
  SPEAKER_01: So a culture was sent, he was put on CIPRO, he was admitted to the team.
• 6:44
  SPEAKER_01: So that illustrates to me all the limitations in the diagnostic test because he had urine tests sent, but no symptoms of UTI.
• 6:51
  SPEAKER_01: We don't have the results for two more days, and then when the results did come back on those urine tests at 48 hours, he had absolutely no growth in his urine.
• 7:00
  SPEAKER_01: So the man had probably passed a kidney stone, which causes nausea vomiting, but heck he passed it two days ago at that point.
• 7:06
  SPEAKER_01: Uh and we didn't have the test results for two days.
• 7:09
  SPEAKER_01: And then to make it worse, since he'd already been started on CIPRO for UTI, when we went to go do his discharge planning, the CIPRO ends up in his discharge meds because he's already on it the day that we're trying to discharge him, even though we think he no longer needs it.
• 7:21
  SPEAKER_01: So the delay in the urine culture results not only resulted in a diagnostic delay of a man who really had a kidney stone, but resulted in being discharged with antibiotics he didn't need.
• 7:31
  unknown: Right.
• 7:31
  SPEAKER_02: And I think that that's a great example of what a lot of us face when we're out on the wards and when we're seeing these patients.
• 7:37
  SPEAKER_02: And I think that to sort of maybe overcome that gap in time that we have in getting some of these answers back, there's been this reflexive behavior, which means that people are actually going ahead and ordering a urine analysis and urine culture at the same time.
• 7:51
  SPEAKER_02: So, in that sort of setting, and with that in mind, what do you feel is the role of reflex urine culturing?
• 7:57
  SPEAKER_00: So my main reservation is related to how reflex testing is used.
• 8:02
  SPEAKER_00: I believe if we establish the basic diagnostic stool tip elements to urine testing, we'll achieve optimal results.
• 8:09
  SPEAKER_00: And I suggest not to simplify how we deal with diagnostic workup for UDI.
• 8:14
  SPEAKER_00: So we need to address the ordering, which is the pre-analytic phase, the analytic, which is how we collect and how we process the urine samples, and the post-analytic, how we report them to the provider that will be taking care of these patients.
• 8:27
  SPEAKER_00: And I do believe that the pretest probability is the most important factor before testing, whether using a urinalysis, then reflex to culture, or directly doing urine culture.
• 8:37
  SPEAKER_00: So let's give the example of obtaining blood cultures in a patient in the hospital.
• 8:42
  SPEAKER_00: You know, it is much less frequently done compared to urine cultures when you look at those that are asymptomatic.
• 8:50
  SPEAKER_00: So it's rare for us to do a blood culture on someone who does not have a fever, for example, or who's not, you know, having significant vital sign changes.
• 8:59
  SPEAKER_00: So there is certain signs and symptoms that we use before we order a blood culture.
• 9:05
  SPEAKER_00: It's not always perfect.
• 9:07
  SPEAKER_00: On the other hand, historically, urine testing, whether urinalysis or urine culture is frequently done based on subjective symptoms not related to infection.
• 9:16
  SPEAKER_00: So urinalysis may be obtained for multiple other reasons, including screening for renal disease or metabolic issues.
• 9:23
  SPEAKER_00: To use blindly the results of a urinalysis to decide whether or not to do a urine culture without taking into account the clinical findings that triggered the test will likely lead to inappropriately either overdiagnosing UTI or in some instances missing cases of symptomatic UTI.
• 9:41
  SPEAKER_00: So another important factor to also consider is the prevalence of bacteriuri in a patient population.
• 9:46
  SPEAKER_00: Contrary to blood, which is normally sterile, the urine may harbor organisms without leading to active infection of the host.
• 9:53
  SPEAKER_00: So the prevalence of bacteriuria varies from almost zero in adult young male population to up to 50% in a long-term facility.
• 10:03
  SPEAKER_00: Unless we clearly evaluate clinically the populations with high prevalence for asymptomatic bacteriuria.
• 10:09
  SPEAKER_00: By that I mean checking for signs and symptoms consistent with UTI will likely falsely mislabel a microbiologically normal finding for a UTI.
• 10:19
  SPEAKER_00: The second thing I want to mention is that when we address the importance of pretest probability, the other area to discuss is the value of a urinalysis and what would be the parameters to reflex to a urine culture.
• 10:32
  SPEAKER_00: So there's a recent survey of 52 hospitals or facilities that were part of the Shay research network.
• 10:39
  SPEAKER_00: They looked at whether they do reflex urine cultures or not.
• 10:43
  SPEAKER_00: And they had about half of them that were doing a reflex urine culture based on urine analysis results.
• 10:50
  SPEAKER_00: Out of 26 that used reflex urine analysis to urine culture, nine facilities had pyuria and predominantly white cells, more than 10 per high power field as the trigger for a culture.
• 11:01
  SPEAKER_00: On the other hand, the rest included the presence of nitrite or leukocyte esterase or presence of white cells.
• 11:08
  SPEAKER_00: So clearly we don't have a standard of what constitutes a positive urine analysis that leads to triggering a culture.
• 11:16
  SPEAKER_00: So with the high prevalence of bacteriuria in an elderly population or a diabetic population, let's say, it makes it difficult only to rely on a laboratory parameter to decide whether a culture should be done based on a urine analysis.
• 11:31
  SPEAKER_01: People are already doing a lot of reflex urine testing, and there are two ways to do it.
• 11:37
  SPEAKER_01: One is the forward-facing reflex testing, which is when a urine analysis is ordered, it immediately goes on to urine culture if it meets certain parameters, such as the white count is higher than a certain amount.
• 11:49
  SPEAKER_01: So that's what I would call forward-facing reflex testing.
• 11:53
  SPEAKER_01: The culture is automatically ordered if the urine analysis meets certain parameters.
• 11:57
  SPEAKER_01: What I'm actually a huge fan of, but I don't see as often, is the reverse reflex testing, which is if someone just orders a urine culture and there has not been a urinalysis, the urine is first sent for urinalysis, and if there isn't a white count to sort of meet criteria, the culture is canceled.
• 12:15
  SPEAKER_01: There was a nice study done by the CDC that showed that that reverse reflex urine testing, where you say if you have a culture without a urinalysis, you do a urinalysis first and then you can cancel the culture.
• 12:24
  SPEAKER_01: That showed a real efficacy at decreasing the amount of urine culture sent.
• 12:29
  SPEAKER_01: And since we know that so many of urine cultures are unnecessary, I think the reverse urine testing is sort of an appropriate way to consider using reflex testing.
• 12:38
  SPEAKER_01: I think the bottom line is we could be using information technology to support our antibiotic stewardship and our infection prevention efforts here.
• 12:46
  SPEAKER_01: We could be using information technology more effectively by implementing both the reflex testing and the reverse reflex testing.
• 12:54
  SPEAKER_02: Yes, I agree.
• 12:55
  SPEAKER_02: I think there's a lot of information technology that actually helps us to become a lot more efficient at managing the types of tests that we're sending.
• 13:02
  SPEAKER_02: And so to that end, do we think that there are any tools that are already available to help us improve on the diagnosis of urinary tract infections?
• 13:09
  SPEAKER_00: So we need to realize that incorporating an ordering mechanism into the electronic records without addressing how clinicians practice will not likely lead to marked improvements in practice.
• 13:21
  SPEAKER_00: Improving the culture of culturing should address both clinician culture.
• 13:25
  SPEAKER_00: By that I mean the practice patterns and what has been ingrained in their education and training, what they have been exposed to from their peers and local experts.
• 13:36
  SPEAKER_00: The other element would be related to how we use the laboratory to optimize testing.
• 13:40
  SPEAKER_00: The laboratory component is a supportive function, but not a substitute for clinician evaluation and decision making.
• 13:47
  SPEAKER_00: There's a recent paper in HP that reported an intervention to improve testing of urine with changes in order stats and inclusion of new urine cultured reflex tests in commonly used order stats.
• 14:00
  SPEAKER_00: Prior to the intervention, they had a urinealysis which would reflex to culture if the included protein, blood, nitride, luxide esterase, fossil results.
• 14:10
  SPEAKER_00: With the intervention, the urine analysis reflex to culture only if it were positive for either nitride or luxide esterase.
• 14:17
  SPEAKER_00: And they've seen about 45% reduction in urine cultures performed at that facility.
• 14:22
  SPEAKER_00: So we can reduce the amount of urine cultures done.
• 14:25
  SPEAKER_00: What we can learn from this example is that there's quite a bit of variation in how and when we test for UTI based on urine analysis.
• 14:33
  SPEAKER_00: So adjusting the parameters for reflexing, and I do like what Barbara has shared.
• 14:38
  SPEAKER_00: It's like if you have an order for urine culture, you can do UA.
• 14:42
  SPEAKER_00: I think it helps.
• 14:43
  SPEAKER_00: But going from UA to urine culture and just increasing a threshold to do the urine culturing, I see it as a non-discriminatory reduction in testing.
• 14:53
  SPEAKER_00: And you may be reducing those that are asymptomatic that should not be tested, but you may be also missing cases that may be symptomatic urine tract infections.
• 15:05
  SPEAKER_00: It's not based on the clinical findings of these patients.
• 15:09
  SPEAKER_00: It's just saying if you have this number, we will not culture it.
• 15:13
  SPEAKER_00: So not having the clinical parts integrated when you're dealing with an EMR testing algorithm, I think it's going to be tough to prove that it's going to improve the care.
• 15:32
  SPEAKER_01: And it's not a benign error.
• 15:34
  SPEAKER_01: I think we're all aware of that, but we have done studies where we looked actually at 4.5 million hospitalizations, of which there were 1.2 million urine cultures on admission, and created matched cohorts of people who did and did not get a urine culture.
• 15:47
  SPEAKER_01: The simple test of a urine culture not only led to increased antibiotic use during hospitalization, but increased length of stay.
• 15:55
  SPEAKER_01: And you can imagine how that scenario happens.
• 15:57
  SPEAKER_01: You know, it's the waiting on the culture results or trying to start IVE antibiotics or resistant organism found in the urine that maybe didn't need to be treated in the first place.
• 16:06
  SPEAKER_01: So a lot of our work has been about creating a decision aid that helps people decide between asymptomatic bacteria and UTI, and then using case-based feedback to teach people how to use this decision aid.
• 16:19
  SPEAKER_01: We have had success in decreasing unnecessary testing and treatment of asymptomatic bacteria at our institution, and we're currently studying whether we can implement this program successfully in multiple other VA hospitals.
• 16:32
  SPEAKER_02: Great.
• 16:33
  SPEAKER_02: So what do you think is the role of rapid diagnostic within the sphere of diagnosing of urinary tract infections?
• 16:40
  SPEAKER_01: I think having information faster at the point of care would be helpful because we're doing studies right now of urine cultures and people that present to clinics, both men and women, and in our hospital and in hair county clinics as well, all outpatients.
• 16:54
  SPEAKER_01: And about a third of both men and women that were diagnosed with UTI based on symptoms and based on maybe on a urine analysis and treated by their provider and sent out with antibiotics, turn out to have a negative culture.
• 17:05
  SPEAKER_01: So having rapid diagnostics that could tell you right then this patient doesn't have organisms in their urine that need treatment, I think that can help assist your antibiotic stewardship programs.
• 17:16
  SPEAKER_01: The problem though is we know for sure that if we start telling people more rapidly the organisms and details about them that are in the urine, there's gonna be a lot of treatment of asymptomatic bacteria.
• 17:28
  SPEAKER_01: So the benefit of the rapid diagnostics is faster information on people that actually have a negative culture.
• 17:33
  SPEAKER_01: The pitfall of the rapid diagnostics is they still don't take symptoms into account.
• 17:37
  SPEAKER_01: And by telling people faster what organisms are in the urine, we're gonna be treating probably a lot more asymptomatic bacteria.
• 17:43
  SPEAKER_00: So I do agree with Barbara with her comments about uh rapid diagnostic testing.
• 17:48
  SPEAKER_00: And you know, when we have a result right away, we may tend to just start an antibiotic then right away, versus having time to take a deep breath and say, oh, the patient is stable, and maybe we do not need to start that antibiotic right away.
• 18:04
  SPEAKER_00: I'm a little bit repetitive here because I think this is probably the key element with how we diagnose UTI, is until we have a test to distinguish patients that have colonization from others with active infection, which at this point is not at least commercially available, it will be important for all clinicians to know what is consistent with UTI sentence.
• 18:27
  SPEAKER_00: So, how do we mitigate the clinician's bias from, you know, for unnecessarily testing the urine?
• 18:34
  SPEAKER_00: And it starts with promoting a better understanding between the clinicians on when to test the urine for fossil infection.
• 18:41
  SPEAKER_00: So urine quality, especially in catheterized patients, because you can see the bag, the urine bag.
• 18:47
  SPEAKER_00: So the color, the smell, the sediments, the turbidity, all of those in catheterized patients do not constitute signs of infection.
• 18:55
  SPEAKER_00: So if the healthcare workers evaluating these patients with indwelling urine catheters can see the urine, and then they trigger that urinalysis based on these subjective findings that are not associated with UTI, we need to let them know about it and we need to give that feedback to them.
• 19:11
  SPEAKER_00: So changing that culture, not to do tests based on subjective findings.
• 19:17
  SPEAKER_00: The second thing is screening urine cultures.
• 19:19
  SPEAKER_00: Quite often, there's certain practices that are whether on admission or before a non-urologic surgery, there's a urine culture urinalysis screen.
• 19:30
  SPEAKER_00: Patients completely asymptomatic.
• 19:32
  SPEAKER_00: This will trigger further testing as far as urine cultures and potentially inappropriately giving antibiotics.
• 19:40
  SPEAKER_00: The third thing is standing orders for urinalysis or urine cultures without an appropriate indication.
• 19:45
  SPEAKER_00: Examples include pre-checked urinalysis or cultures and disease-specific order sets.
• 19:50
  SPEAKER_00: So if I'm treating a pneumonia or I'm treating PHF, should I have a urinalysis pre-checked there?
• 19:56
  SPEAKER_00: And what does it do if it's abnormal and very often it's perceived as abnormal finding in someone who's elderly?
• 20:04
  SPEAKER_00: The fourth thing is time culturing.
• 20:06
  SPEAKER_00: And this is where we should have some mindfulness in evaluating the site of infection before we do blood, urine, and urine culture without looking at the source.
• 20:16
  SPEAKER_00: And the fifth thing is avoiding urine cultures in asymptomatic patients.
• 20:20
  SPEAKER_00: And this includes those that have a urinalysis with puria and are asymptomatic.
• 20:26
  SPEAKER_00: Resisting that push that if I have some pyuria or or I have bacteriuria, I won't go to the next level and do a urine culture unless I'm suspecting a UTI.
• 20:38
  SPEAKER_00: And this hurts the most vulnerable population, which is the elderly.
• 20:42
  SPEAKER_00: The very elderly have very high prevalence of bacteriuria and some of the diabetics.
• 20:49
  SPEAKER_02: Thank you to our panelists, Dr.
• 20:51
  SPEAKER_02: Troutner and Dr.
• 20:52
  SPEAKER_02: Fakey, for sharing your perspectives, your work, and your experiences on diagnostic stewardship of urine culturing.
• 20:58
  SPEAKER_02: This has been a very informative discussion and has helped to highlight the potential uses for algorithms or electronic medical records to help guide ordering and possibly limiting unnecessary urine culturing.
• 21:08
  SPEAKER_02: Looking to expand your knowledge and diagnostic stewardship?
• 21:11
  SPEAKER_02: Then join us at the Sixth International Conference on Healthcare Associated Infections, Decennial 2020.
• 21:17
  SPEAKER_02: This conference will be held in Atlanta, Georgia from March 26th to March 30th and is co-hosted by Shea and the CDC.
• 21:25
  SPEAKER_02: Find out more and register at www.decennial2020.org.
• 21:30
  SPEAKER_02: This concludes the second episode of the Diagnostic Stewardship Podcast series.
• 21:34
  SPEAKER_02: Keep an eye out for our third episode, Procalcitonin Weighing the Benefits and the Costs.