51. Taking the Piss: Urine Culture, Unravelled

The title says it all. We're going back to the lab for this discussion. Join Jame and Callum as they have an in depth discussion about the nature of urine culture.

Prep notes for this episode available here. 

Chapters

• 0:00 Understanding Urine Culture Basics
• 10:38 Types of Urine Samples
• 22:08 Urine Culture and Chromogenic Agar
• 31:45 Quantifying Bacteria in Urine Samples
• 45:37 Understanding Urine Cultures and Antimicrobial Stewardship

Transcript

Jame: 0:00

Hi everyone, welcome to the Idiots podcast. That's infectious disease insight of two specialists. I'm Jame, that's Callum, and we're going to tell you everything you need to know about infectious disease. Callum, how are you doing.

Callum: 0:22

I am doing great, jame. I have been looking forward to this episode for a while. I think we both have. It's very important And I was thinking about it The other day. I was out. I was at an art exhibition. It was very enjoyable, and I was standing there looking at the paintings and I thought to myself Callum, right now, you're in culture, you're in culture. And then my head made that pun in the exhibition and I was like you're in culture. So that was the segue that was going for. Thank you.

Jame: 1:04

All right, no need to put a lampshade on it. Okay, so today we're talking urine culture, that's right. This is a follow up to our episode - are you laughing at your own pun?

Callum: 1:16

It's so bad. It is really it's just, I was just thinking there but if you come into this, it's worse than normal Like they're just getting worse.

Jame: 1:24

I don't know what to do.

Callum: 1:26

What you need to do, listeners, is listen to all the episodes and write the puns and then and then email it. I know I was just thinking myself there that this is probably a topic that is more applicable to a wider range of people working clinically. Is this a work up to another pun?

Jame: 1:44

So, no, so we might have people listening to this episode who have never listened before and they may have already been like no that's true, i don't know, but yeah, this episode is going to be on you're in culture and the basics of the test, how to use it correctly, and as part of an ongoing, very loosely connected series of the episodes about the thorny issue of urinary tract infection. So, callum, why don't we just sort of jump in? So when should you be considering taking urine for culture? That's a great question.

Callum: 2:23

I think the big picture answer is where you think someone has a urinary tract infection or it's likely that they do. You can narrow it. It's maybe easier to talk about when you shouldn't send to urine culture than when you should.

Jame: 2:38

Well, when shouldn't you? What are the common clinical scenarios where you should not send urine for culture?

Callum: 2:42

So someone's urine infections are extremely common, in fact I think I'm right in saying that they are the most common bacterial infection, particularly in women, just because the difference in anatomy, you ur ethra is much shorter, so it's much easier for the bugs to get up the urethra, and start causing problems. And I guess we were recommended by Nice not to send urine culture in women who don't have a complicated UTI. So essentially they've just got some mild urinary symptoms. There's no signs of a more severe infection. They don't have any risk factors for resistance, it's not recurrent, they aren't pregnant, they're not the catheter. In these situations they're very likely to have a bog standard E coli UTI or something else very common that will be sensitive And you can just give them antibiotics And that's nice and easy for everybody.

Callum: 3:43

And I guess the other end of the spectrum where you shouldn't send urine culture is where you don't think they have an infection And that might seem obvious. But I guess the classic example is people that have a catheter in place And they have something, so like the catheter is blocked or their urine is a bit smelly or they're just a bit muddled, and I think the reflex because urine samples are so easy to collect, although correcting them correctly is maybe not that easy, and it's a very simple test to send. It's a very commonly sent test And so I guess the other end of it is you know, before you send it think, does this? do I actually think this hasn't this pressure has an infection, and if it comes back with a result, you know, do I think I want to give them antibiotics, because it's really less about, in my mind, urine culture is less about making the diagnosis of a UTI in most of the situations, and it's more about culturing a bug so that you know what antibiotic to give them.

Jame: 4:56

Well, it's like what you said in the previous episode the you diagnose a UTI based on on symptoms And people take a urine culture And if they grow something, automatically assume that that thing is causing a UTI. But really what you've proven is that your urine culture has that bug in. You know certain, certain quantities in the urine And you know really what you're identifying is bacteria urea And, like you say with the catheterized patients or patients with abnormal anatomy, they can have bacteria urea and the bacteria are just sitting there, not necessarily causing infection. So you can go to all the trouble of identifying the bug and getting an anti biogram and knowing what you could treat the bug with. But the urine culture is not so certain, necessarily diagnosing the UTI. You should be doing that on the basis of symptoms. Where are these bugs coming from? Urine is normally a sterile fluid and usually it's formed in the kidneys, disaines through the ureter infoblader and then goes out through the urethra. It's like a waterfall. It's a one-way system. How are the bugs getting?

Callum: 6:14

in Yes, before I answer that, i think there's a lot of discussion going on with our improved molecular testing about what is sterile, because maybe the urine isn't sterile. I think that's a far too complicated question for right now.

Jame: 6:32

Yeah, i'm going to chat for another day For any lawyer listeners who are not, infection trainees and microbiologists. in particular, you can treat urine as if it is effectively sterile.

Callum: 6:42

Effectively sterile is probably a good way to put it.

Jame: 6:45

But bear in mind that there are probably some not neuropathogens but, eukomensals that are living there normally and not doing any harm. That don't really feature in. Where are they coming from?

Callum: 7:00

They're coming from? well, i guess they're just coming from the outside, in What's on your outside in that area? Well, your perineum and all your bowel bugs are coming forward. When we talk about samples from basically the waist down, you always consider that as if it's contaminated with fecal organisms Any wounds on the leg you'll commonly see bowel bugs there. When we say bowel bugs over and above anything else, we're looking at E coli. E coli is public enemy number one when it comes to urine retract infections. The other organisms that you're going to see are other enterobacteriallies and then staphylococcus saprophyticus and then things like B t, hemalidic strep, sometimes staph orius, sometimes other things. We'll cover that in a bit more detail when we talk about the treatment aspects.

Jame: 7:57

But yeah, it's basically bowel bugs are coming in Because the bugs that are living in your gut view, the genital area and the skin round about there and the glands penis and the vagina is just a fancy sort of gut that's in a different location. It's warm, it's dark, it's wet, it's relatively anaerobic. They'll just colonize it and then they'll try and crawl up the urethra on the first chance they get, and that's why you're dominated by your intrabacteriallies, even though, if you actually look at the microbiome of the vagina and the glands penis is not dominated by E coli, it's not dominated by C lepsiallies and pseudomonas and all the other things that you find in urine samples.

Callum: 8:52

So I guess when we're talking, you know what we want to focus on in this episode really is the actual urine culture itself and maybe just give our insights, I guess, into what we've learned training in the lab that we didn't know as clinicians before. that could be helpful for people because I think it's a bit of a mystery a lot of the time what happens in the lab and how that relates to the patient. And I guess when we talk about any lab test, there's three sort of phases that you can talk about, and this goes for any lab test, but I think it's maybe worth us talking about urine culture in these steps. So there's a pre analytical phase, which is basically everything that happens before you do the actual tests on the sample. There's analytical phase, which is the nitty gritty of what happens in a lab from the urine sort of arriving and being taken out of the pot. And then there's a post analytical phase, which is all of the sort of how we report that culture results, how that is interpreted and you know what goes out and how people receive that.

Callum: 9:58

And I think those three phases is useful to think about it like that, because it's easy to think that you know these are black box tests where you know, essentially just put the urine sample into the lab and then something happens and the result comes out and that's concrete. There's so many steps where things can go wrong or be you know, particularly at the beginning. So maybe we should start with the beginning, then we'll move on to the middle and we'll finish at the end, which is a really sort of it's a pretty classic way of doing it.

Jame: 10:33

It's inspiring stuff, Callum. So let's start with the pre analytical stuff. So so what kind of urine sample, common urine sample types do we have? Like, what can people take?

Callum: 10:47

Yeah. So the sort of standard urine sample that we do is a mid stream urine And that's essentially. You get someone start peeing, they discard the immediate first part, the first voidage urine, but then, without interrupting the flow, you then collect around about 10 mils of urine. You put that into a sample and send it off to the lab. So that's MSU, as everybody here about. Yeah, so that's the commonest. Is that the gold standard? It's not the gold standard, it's the recommended routine collection method. The gold standard or at least historically in papers, is something called a suprapubic aspirate or essentially aseptically. You put our needle with a syringe, insert that into the bladder and aspirate some urine back.

Jame: 11:37

Obviously. yeah, too invasive, I think, And difficult to do routinely. They do it sometimes in kids, don't they, i believe?

Callum: 11:45

so It's very hard to collect urine samples in an aseptic manner from small children, and so you're sometimes stuck with potentially quite significant results, because every infection in that age group is significant and you're not sure, and so you might do this to have a more definitive answer.

Jame: 12:07

Yeah, And what are the other commonly used?

Callum: 12:12

Yeah, so clean catch urine is essentially an MSU, but you don't discard the beginning and the end. That's pretty common as well. I don't know often if people really get the difference, And so when we get urine samples in the lab and I'm also not sure that I look at a clean catch urine or a midstream urine differently when reporting it- Well, i think, in reality, a lot of MSUs are actually clean catches of urine, exactly, and that's to do with training of the people that are acquiring the sample and you know, like the What's the time the patients collect the sample And lack thereof, or the patient doesn't know.

Callum: 12:53

It's pretty tricky Yeah.

Jame: 12:56

And yeah so. But the clean catch is. Why would you want a clean catch urine? You wouldn't necessarily want it. You would accept it if you were trying to get urine off a kid. You know So, like the mum or the dad are sat there with a kid, the kid's stripped down, the nappy's off And they've got a little sort of pot hovering delicately over this kid's penis, desperate for this kid to pee, and the kid's crying and wailing because it's ill, because it's got a urine retraction infection. You will accept any urine at that point, because the alternative is to do a super poop gastroint, which is not fun either. So that's the sort of situation where clean catch urine is desired.

Callum: 13:40

But why do we want a midstream urine rather than a clean catch?

Jame: 13:44

Well, because urinary tract may be sterile, but as you get near the urethra you're getting less and less sterile, and so the chances that your urine is going to be colonized by skin commensals like coagulase negative staphs goes up and up and up, and so what you don't want is for your urine culture to be overrun by those bugs or by whatever else happened to be in the urethra and meatus at the time that you started peeing. So you want the midstream sample. Is all those things have been kind of peed out with the first void and then you get the sterile middle bit, and then why not the bit at the end? Well, because they might stop peeing before you acquire your samples. So the midstream is the best.

Callum: 14:35

And I guess that is in contrast to things like if we're taking urine samples for gonococcus and chlamydia because they cause a urethritis. They're sort of the infection. Yeah, you want the first bit of urine. You want the first bit of urine because you're most likely to find it And also the way that you look for those is a very selective sort of culture or use a PCR to specific. You don't worry about the commensal.

Jame: 15:00

That's true, whereas the commensals will just overrun your urine sample. No matter how concentrated your bugs are in your urine, they'll be swamped by what is in the urethra.

Jame: 15:13

And then the last sample is catheter specimen of urine And that's exactly what it sounds like. The patient's catheterized, you get urine, you are. We'll cover this in a future episode, but you should be aware that within about 72 hours, almost all catheterized patients have their urinary tract colonized with stuff, so you will grow something. The issue there is to realise that you have to interpret the resultant context.

Callum: 15:45

Yeah, i think the key with the catheter specimen urine is to only send it when you think the patient's got catheterous urinary tract infection, and that's difficult. But there are some things that I think are commonly thought of as signs of catheter infection and some things And a lot of those have no correlation whether the patient has an infection or not, like dark smell of urine. Well, that's spoilers.

Jame: 16:12

Maybe that's a teaser for the CSU episode the catheter associated urine retract infection. So those are the kinds of samples that they send. Usually they send them in what just like a plain white top to plastic And a universal sort of sterile container is pretty standard.

Callum: 16:30

That's okay if you're in the hospital, i guess from the pre-analytical phase. Part of it is how the specimen was collected, part of it is the volume, part of it is the way in which it's transported to the lab. So the container, and if you're taking more than six hours, say, to get to the lab and the sample is not refrigerated, then if there are organisms in that sample or even low numbers of contaminants, then that gives them the opportunity to sort of overgrow and can affect the validity of your sample. So you'll commonly say that if there's going to be delays in the sample getting to the lab, then they're refrigerating it. Or if that's not possible, then often boric acid is used, which is in our center, a red top tube, but it'll be slightly different And you need to take the exact amount of urine that the bottle asked for And then that concentration of boric acid will sort of stabilize the organisms in the urine but shouldn't kill them, although it does affect some organisms.

Jame: 17:34

Well, I can kill them a wee bit, but it's a trade-off really, isn't it? So, yes, the red top tubes, I think that's fairly universal. I've seen it in lots of different places. But yeah, that might be why. So say, if you're sending any urine culture from primary care, that might be you might want to particularly know that your transport time is quite a long while. You might be using the red top tube instead of the white top tube, which is basically what we would use in hospital, because there's usually not very much of a delay. Anything more to say about pre-analytical Kelly?

Callum: 18:13

No, i don't think so. I think the key really is just to consider that we can look at a test and get the result at the end, but if you don't know how that was collected and what type of sample it is, how it was transported, all these factors, and then really the validity of the test itself is in question from the beginning. If it was taken in a non-clean way at the very beginning, then your result is meaningless. It's sort of like in computing you talk about garbage in, garbage out. Whenever you do programming, if you put bad data in, you're going to get bad analysis of that. That's equally true for any diagnostic test.

Jame: 18:53

Once it gets to the lab, the analytical phase starts. What then happens?

Callum: 19:01

That's where the magic happens In some centres, and so there's talked about the standards for microbiological investigation for urine. is urine microscopy might take place as a screening test, looking for, essentially, signs of yeah, pyuria, that's it. You might do urine microscopy looking for things like pyuria, so white blood cells in the urine That is shown to correlate with batch urea. Sometimes you can use that. That's quite a resource-intensive method and, to be honest, i don't know about your centre. our centre would just go straight to the urine culture step.

Jame: 19:46

No, we don't do it either. But I know there are some centres in the UK that do it and have worked in places abroad where what they would do is they would, before they analysed the urine, they would take a sample and put it through the flow cytometer. If it had a white count of more than a set cutoff I can't remember what it was they would go ahead and process it. If it was less than that, they would say well, you're not significantly infected, so we're not going to process the sample any longer. So it was used as a sparing manoeuvre to reduce the throughput. Because if you don't have an automated machine like a keystrah, the processing of urine cultures is fairly intensive And it's like you say, it's a really commonly taken test. So it's a significant proportion of your lab staff's time. So if you can cut down on that by doing a quick 20-minute flow cytometry pre-prepared assay and then discarding the ones that are not significantly pyuric, you can save a lot of time.

Callum: 20:54

But why do we not do that now? Oh well, callum, i can answer that for you. So I guess with any sample in the lab, it's all about turnaround time. You want to get your time from getting the sample to putting out the result as quick as possible so that your results are meaningful to the person managing the patient. That's the key, isn't it? That's what it's all boils down to. And the urine cultures that's probably the one that we look at the most. Such a common investigation. You've got that patient. They're waiting, they're uncomfortable. You want to get them that result quick. Just adding in this extra step in your process adds up complexity. It's something else that can go wrong. Flow cytometers are fairly complex machines. It can go wrong. If it breaks down, then what do you do? And I think ultimately you're coming back to saying do you really rely on that or do you just go and do the urine culture, which is the more definitive test anyway? And I think that's the direction of travel.

Jame: 21:50

Well, I think now that we've got chromogenic eGaRS which allow you to identify the organisms quite rapidly, Well, that sounds interesting. Well, to that end, let's talk about what happens when we get the sample to the lab.

Callum: 22:05

Okay, What happens?

Jame: 22:06

when we get the sample to the lab. So we'll have the sample and next up we'll have a EGAR plate that we want to inoculate on it. So we will take a one microlitre sample from the urine, after giving it a shake to make sure that it's mixed appropriately, and then we will inoculate that one microlitre sample onto the EGAR plate And we'll spread it around using fancy techniques known to the lab biomedical scientists and try and incubate that plate for one to two days, but actually usually at 24 hours you'll be able to read all of these plates, the EGAR that we're using. It varies from site to site, but usually what people are using now is yeah kind of Just before we move on to the plates.

Callum: 23:06

I'll probably just cut that bit out and add in. So it's important to note Jill James saying that we use one microlitre. That's not. all labs will use different amounts, but it's important to know what volume of urine you're using at the beginning, because we do something called semi quantitative culture And essentially by knowing how much urine you've plated out in the beginning, when you get a number of colonies at the end, you can do some maths and essentially work out and report how many colony forming units there were per mil, which is a sort of definitely semi quantitative. It's not an exact science but it's some sort of measure of concentration onto organisms.

Jame: 23:50

Yeah, that's true. That's true. Yeah, let's talk about that next. But yeah, let's first talk about the chromogenic agar.

Callum: 24:02

Yeah, so basically impregnated into the agar are substrates And those substrates can be used by some of the bacteria And each bacteria will have different enzymes to break down different substrates.

Callum: 24:19

And what they've done is they've attached to the substrate a color pigment. So when, say, your E coli picks up galactose, then it will break it down with beta-galactosidase And the color that's attached to that will make the E coli colonies appear pink or red. So it's really clever because you can glance with a very, very high degree of accuracy which is sort of validated, say that is an E coli, And it's so accurate that you know, say, on a wooden swab and you had an organism you would need to culture and then use some sort of identification. And even with the Malditoff being so quick, that adds an extra step and complexity and things where it can get wrong, Whereas with this the value is the speed at which you can say you can glance at the plate or a picture of the plate sometimes and say that's E coli and then put it for sense And that's taking seconds rather than this sort of coming off and all of that.

Jame: 25:22

Yeah, that's it.

Callum: 25:24

So you're able to get to the identity so much faster than even with Malditoff, like say, and the other thing that I think is worth pointing out is that you know so before we're doing this, you should put it on sort of like somewhat selective agars would often be the way of doing it, like you know, like a little bit of a cled or like Moconchi or something like that, because most of the things you're looking for are intramurales And organisms can like, even if it's the same bug, can look sometimes definitely like colonial variants, and so you end up doing this thing like oh, is it group mixers or not? You know what's the identification of these organisms And it you know, even though it seems like you know, you just culture the organism and then you get sensitivities or you do a gram. You know there's so much extra steps, the process becomes really complex. So if the chromogenic agar, you're really simplifying your process down to the bare minimum And then your fruit is much higher.

Callum: 26:22

So you know, the plate might be quite a bit more expensive than just a standard agar.

Jame: 26:27

It's worth it. Well, it's pricey, but it is per plate. but it is sort of smushing together colonial morphology and you know biochemistry, you know, to get the diagnosis faster. You know it's a bit like doing a mini API, but just with like two or three different things.

Callum: 26:43

So yeah, So so Indra Kokai are blue. other coliforms So like things that are intramurales but similar to E coli, are sort of a darker blue purple. So they look quite distinct that when you see them on the plate they look very different. And the reason that's kind of similar is that Indra Kokai have beta glucosidase and coliforms also have that, but they also have beta galactosidase, which the E coli has. So I guess red and blue equals purple.

Jame: 27:15

And in other things like darker blue at least. Yeah, proteus.

Callum: 27:18

Morganella Providentia. They all look sort of brown And then there's some that, just you know, look the normal things like pseudomonas. It's got a pretty classic appearance, Like any biomedical scientist will be able to see pseudomonas on a plate and say that's pseudomonas, And this is obvious.

Jame: 27:32

But they're not doing any of this reaction stuff.

Callum: 27:34

They're just appearing in their natural colours right, yeah, and I think Staphylococcus is the same. It's just a sort of normal colour. Staph saprophyticus in that plate has a colour change. But you know, we could talk all day about how good these things are. I think they're brilliant, but probably there's an interest people as much as it interests us?

Jame: 28:01

Well, maybe not, but yeah. So they are sort of incubated for one to two days. In fact, some labs just do it for one day, and if you don't go anything a day one, they just chuck the plate. It depends on your individual setup, but yeah, they'll. Then if you've got a positive culture Callum, what can you then do? Can you just take a sample from those coloured colonies and run it directly through antibiotics susceptibility testing, or do you need to re-culture them? Well, blood agar plate.

Callum: 28:41

Well, very conveniently, you can just take them straight from that chromogenic agar and put them into your plate. So sometimes there's certain agars that like you know, some selective agars for example, where because of what's in the agar then you can't put it straight into the sensitivity machine. Just because of the way it works It will be interfered with. But that's another advantage to the system is that you can move straight. You know, you just take those colonies. You need to have a sufficient volume to make up, you know a dilution, So you suspend them in it in some sort of sterile liquid, And then you suspend them in sterile water and then put them into your sensitivities in that way, in a standardized way, And then you get your antimicrobial susceptibility is usually in some sort of automated way, And that's really the process that you go through in this analytical phase.

Callum: 29:38

And I think in a way in my mind, the analytical phase is probably the bit where things go the most straightforward. The most straightforward or like there's the least opportunity for things to go wrong. Like I'm not going to, like you know, things do go wrong in the lab And that's why we have things like internal and external, external quality assurance, And there's so much but there's so much focus on the lab tests and it is very regimented and ordered, Everything's on a standard operating procedure that there's less of that uncertainty which you have in the pre analytical and post analytical phases. There's a lot of certainty in what should happen. I think the bit where the uncertainty comes in for me in the analytical phase part of urine culturing is how we interpret mixed growth And it's maybe worth touching on that because I'm sure anybody that's seen a urine culture result will have seen a fair share of mixed growth likely contamination.

Callum: 30:46

So maybe we should just talk about what the guidelines are for that and how we report it. Yeah, go for it. So essentially there are, at least the perceived wisdom is that when you have a urinary tract infection you're going to have one organism causing that. And so what we expect and what's easy is, when we get a urine culture we grow one bug is E coli It's you know, we do the sensitivities and report it out. So that's very easy to interpret and report.

Callum: 31:18

What becomes more difficult is that quite often you will have more than one organism that is grown, and that is particularly true in those urine samples we were talking about earlier on where it's not as cleanly collected For example, inverted commas, clean cats urine or catfiders, specimens or urine, these sort of things And in those situations you're more likely to see more than one bug. So what does that suggest? So it could be a problem with the pre analytical phase. It could be that you know we didn't clean the area properly. You know some of the skin commensals gotten. It could also be the transportation time was too long and some organisms over grew. It could be some contamination when you know the bottle was being put together. Or maybe the bottle wasn't properly sterile when it was put into it. That's quite unusual. So how do you report that? Because I guess you know, even if it's mixed growth, is there still a significant pathogen in. There is a predominant pathogen And it comes down. So the SMI they lay out you know what to do in different organism counts. So if you've got heavy growth, essentially it's it. I'm not going to go through the details of it but if you've got a predominant growth of one organism, even if there's other things growing, then you essentially work that one up And you report it as possible UTI If it's a sort of clean catch urine or probable colonization if it's a CSU, if there's one, if there's more than one organism, and If you've got more than three or more organisms, then generally just report that as faulty collection or transport.

Callum: 33:10

There's a contaminant there. So yeah, the reporting is slightly different depending on the type of sample that is sent, which is talking about pre-analytics. Maybe we didn't talk about one thing. Was that something that's really important in the lab is that the sample is missing. To get it right is calling the sample what it was. So if it's a CSU, saying it's a CSU rather than MSU, and that goes wrong all the time. And the other thing is putting in some clinical details. Because if we know what the clinical details are, so for example, if the clinical details say pylonephritis, then the organisms, you know part types are differently than if it just says dot dot dot or you know smelly urine. You know, you get the point like that that that clinical correlation of the sample is really important but I think really underappreciated how important it is.

Jame: 34:16

So you get a sample and you're growing something, say it's a monomycrobial sample, so you're pretty sure this is the cause of organism of it. And what would you let's talk a little second for our cutoff, for significant growth and no significant growth. So sometimes this gets pushed out into the report. So, like in Nadoch South, the level of the, or the number of organisms is reported as number of colony forming units per mil, and colony forming units is just a, you know, viable organisms that could, you know, replicate and form a colony if given the opportunity. And you know I mentioned there that we take a one microliter sample. Well, that's one thousandth of a milliliter, so 10 to the five or a hundred thousand colony forming units per mil. For a one microliter sample, that should produce 100 colony forming units And that's, you know, countable on a on an EGAR plate.

Callum: 35:32

Nobody's sitting and counting 100 colonies. Basically the colony is just one little blob of bacteria on a plate, or fungus or something else.

Jame: 35:41

So let me ask you then, callum, why 10 to the five, why not 10 to the four, or 10 to the three.

Callum: 35:46

Yeah, it's arbitrary, isn't it? Like we, there is a need when we report tests for cutoff values, And I think the the general understanding outside the lab is tests are positive or negative, And that's the way that we report them And that's kind of. We kind of need to draw the line somewhere, don't you? But the reality is on any test. So you're in, culture is the same. But you know something like, say, a PCR for COVID? Oh no, I've talked about COVID on the podcast. We never talk about COVID, And you know, there is a point that you determine, which is arbitrary, where you say this is positive and this is negative, And there'll be some negatives that are just below the positive line. There'll be some positives just Well.

Jame: 36:30

I mean, it's not arbitrary, It has been chosen with some care, but usually it's with reference to a normal range for a bunch of tests.

Callum: 36:38

Maybe arbitrary is the wrong word, but you choose a cutoff value for positive or negative and tests based on trying to optimize your sensitivity and specificity. So essentially, the lower your cutoff value is So say, in your example, the lower we called positive, the more sensitive it would be, the more likely we are to pick up those sort of low level positives. But the less specific the test would be, so the more likely we are to call something a urinary tract infection or call something a significant growth when it is gone.

Jame: 37:10

Yeah. So really, you know, because we're in the lab and we're sort of one step back from the patient, we're not really diagnosing UTI, like we said before, we're diagnosing bacteria. But what do you call a significant bacteria? So a bit of history time now. So the 10 to the five is based on something called the CAS criterion And this was a physician, edward CAS, who did loads and loads of work on urinary cultures in the 50s And he found that a bunch of people that were, you know, when people died they got an autopsy much more rapidly there And he was finding a lot of people done diagnosed pyelonephritis and sort of renal tract damage And he kind of was worried about, you know, something of a silent epidemic of people with undiagnosed urinary tract sepsis in particular.

Jame: 38:03

And mind, this was in the first decade that antibiotics were really being used in anger And so before that there was, you know, there was basically no cure for these infections. And he kind of got the idea that maybe I'll start quantifying bacteria in the urine as a way of diagnosing pyelonephritis. So if the bacteria is above a certain value, pyelonephritis will be more likely. And he kind of based that on sort of epidemiological studies that he did and tried to tell between contamination and infection, because of course he knew contamination was going to be a problem. And the sort of cutoffs that he chose was based on E coli, just fair enough, because it's the most commonly isolated organism which multiply quite rapidly They're doubling time is about every 20 minutes in ideal conditions. And so he then, you know, tried to.

Jame: 39:00

He published a bunch of papers sort of saying that for kind of making the association between pyelonephritis and higher urine CFU counts, and the cutoff that he chose was 10 to the five, or 100,000 or more colony forming units per per mil. Now he could have chosen, you know, 10 to the four, which 10,000 or more, or 10 to the three, which is 1000. But he, you know, the association was kind of less right there, and that's. This is going back to what you were saying, kamal, about it being semi quantitative, because it's sort of not a. It's not qualitative, like you know, a PCR with a set cutoff values, either positive or negative, and it's not quantitative, like you know. Something like a CRP for example there's. You could measure the colony forming units quantitatively, but then you have to apply a sort of cutoff at a certain level. And he chose 10 to the five, because that was where the association started to be really common in his studies.

Callum: 40:13

Okay, so that's it sorted. Should we move on? No, let's not move on, callum. Oh, what do you mean? 10? I've heard 10 to the five is the cutoff that we should use, and that's fine.

Jame: 40:24

Well, no, because you mentioned this in your analysis paralysis that when you were looking for sensitivity testing, that 10 to the four was the better kind of breakpoint for for dino's and UTI, particularly with reference to them. But people have kind of come back to this in kind of later years because of course all this was done in the 50s with 50s technology. People came back later and sort of decided and had another look at the data and sort of said, well, at what point or should be using now to diagnose sort of cystitis and applying modern statistical techniques to the data that they've got, because the 10th to the 5th cutoff was so we don't miss pylonephritis, an equal pylonephritis at that, it was sort of thought that maybe the 10th to the 5th was maybe a little bit too high and that for diagnosing our everyday UTIs maybe a lower cutoff would be more appropriate.

Callum: 41:29

Yeah, i wonder if there's something there about access in the UK anyway to healthcare is free and relatively easy, you know, not gonna say it's easy.

Callum: 41:41

So people present with much lower burden of disease And also, you know, i guess historically in the 50s you know a lot more TB around and other reasons why people might have pylonephritis. Essentially what people have done is they've gone to groups of young women essentially that had symptoms of cystitis, so they were symptomatic and they did paired urine and in-out catheter samples And the idea was they were trying to correlate what they were doing, what was happening in that sort of urine. The midstream urine culture and the sort of golds were slightly more sterile sample of putting the catheter into the bladder and taking a sample that way And essentially, as I said, are they wrong. The lower the organism count that you take as significant, the higher your sensitivity but the worse your specificity would be. So they went all the way down to greater than equal 10 colony forming units per mil, which if you're only taking one microliter is you know you might even get that on some of your samples.

Jame: 42:49

So that's you know how rare is that.

Callum: 42:52

This is a bit of a chance if you get even one colony from that. And there they were finding a 99% sensitivity for picking up. You know people that they were calling a true positive but their specificity was only 86%. So there was quite a few people that they were saying you know, we think you've got a urine retract infection when you know maybe a more reliable test was saying no.

Jame: 43:15

I mean, I don't know about the count, because 86% specificity is actually pretty good in most other circles, But it's, as you say, the same as in the spec they've got. There's a trade-off And usually that trade-off leads to the choosing of a breakpoint, usually with the rock curve being they're being utilized at some point. Yeah.

Callum: 43:35

Whereas if they went up to greater than equal to 10 to the five so the sort of more traditional one then they were looking at only 60% sensitivity, but the specificity was very high at 99%. Yeah, now, i think the way we approach it is essentially, you know, if it's greater than 10 to the five and it's one organism, yes, that's a UTI. If it's somewhere in the range of 10 to the three to 10 to the four, you're saying it's a possible UTI. You need to correlate it to the symptoms. You know small numbers And in lower than that, you probably are gonna pick it up with only taking one microliter.

Callum: 44:06

So do you really need to worry about it? Which is kind of a bit of an odds, in a way, to a way that we approach such fellow cultures, like if you go back to our back to, we go back to back to back to Remia, is it too many backs? And then we were talking a lot about how the volume of blood was really important for your sensitivity, and so then you left thinking, well, if we really want to pick up people of a UTI, would we not just culture more urine So you can concentrate down things and do pellets and that sort of thing?

Jame: 44:36

Well, you could do it, or you could just in lower your cutoff. So I mean, if you use 10 to the three on this paper, which we'll include in the show notes, your sense was 88%, which is pretty impressive, and your specificity was 95%, which is also very impressive. And then if we just move over to the positive predictive values and just take the 10 to the one, 10 to the three and 10 to the five cutoffs, so the positive predictive values were 92, 96 and 99% for 10 to the one, three and five, and then the negative predictive values were 99, 84 and 62% respectively. So obviously that middle value, the 10 to the three, would give you a positive predictive value of 96% and negative predictive value of 84, which seems to me to be the optimal cutoff there. But that's two orders of magnitude less than our presumed, our current level of 10 to the five.

Callum: 45:37

Yeah, we do culture down lower, but I guess it's worth saying as well that it's not as simple as if you use the same patient and you took three different urine samples at different times of the day whilst they have an infection, you might get different organism counts And there's lots of things that have been shown to affect the organism counter. you're gonna get in urine infection, so things like how much they've been drinking. So if you got someone with a UTI and you say drink lots of urine and it might help it clear quicker, then you probably expect them to have lower. it makes sense. you're diluting out the urine, so we're gonna pick up less urine. So actually the you're less likely to get a bug maybe, which is kind of strange to think about it that way. Also, given someone antibiotics just like an any other infection, and you're much more likely to have a negative culture.

Jame: 46:28

Yeah, and because the antibiotics that we use concentrate into urine, and that's what we use them. Even a relatively innocuous thing like Traumatoprim will functionally sterilize the urine retract and you won't be able to grow anything out of it.

Callum: 46:41

Yeah, particularly that urine sample sitting for a long time and then sort of bathing in an antibiotic soup. Yes, Yeah, trying to get that sample before your antibiotics is really important. It's not really much value of someone's on systemic antimicrobials to taking a urine sample. You see that happening, sometimes inpatient. They're on Tazacin or something and they spike a fever so they get a septic screen. So you take blood cultures, which is sensible Urine cultures. I'm not sure how much value that's got if they don't have urine retract symptoms.

Jame: 47:15

Yeah, this is going off topic, a sidebar if you will, but people might think well, what's the difference between blood and urine? The difference is that your antibiotic will be concentrated several times in the urine relative to the volume in the blood, and so if you're taking blood cultures, you might argue that the Tazacin would suppress growth of sensitive organisms. fine, but you might be taking it for insensitive or resistant organisms, but in the urine you get concentrations of the antibiotic that are so much higher than that of plasma that even organisms with required resistance to the antibiotic the patient's on may have their growth inhibited just because there's so much of the antibiotic that they can't overcome it. That's very much a tangent. So we've talked about the analytical phase. I think pretty much to death. I don't think we've got anything else to say on it, do?

Callum: 48:15

we Oh, there's so much more. We could talk about this for a whole day.

Jame: 48:17

Jane, there's this podcast already eats, so much longer than I thought it was going to be.

Callum: 48:23

We just need to move on. Ha ha, ha, ha ha ha ha ha So what about?

Jame: 48:28

is there anything else that we need to mention? Can we move on to post-analytical?

Callum: 48:32

No, I think we can move on.

Jame: 48:34

So the post-analytical phase. so the lab has done their duty. We've got the results, We've got the sensitivities to the bug. What?

Callum: 48:46

now. So now, essentially, there'll be a sort of quality check on that sample. So the machine that does the susceptibilities, there'll be some automatic checking to make sure, like you know, does this pattern of antimicrobial susceptibility fit with what we know this organism should usually have? So it's sort of like a basic sense checking that goes on. And then that urine sample will go to, depending on your center, it might be automatically authorized out with something called like rules-based authorization. So sort of simple samples will be under that pattern. So say it's an E coli urine retract infection and it's fully sensitive. Bang out, it goes, it's authorized, it's done.

Callum: 49:28

And you're getting into more complex samples. or you know niche patient groups like, say, a neonate or something, and then that will go to somebody with sort of the training to interpret and report that. so, whether that's sort of advanced biomedical scientists or clinical scientists or microbiologists or someone, then they will essentially have a think about the clinical situation, what's on the patient's notes, is there any allergies, what's the resistance pattern, And then authorize out the reporting of certain antimicrobial susceptibilities. And that part of it is really around antimicrobial stewardship. So we know that that is an effective way to sort of guide clinicians to choose antibiotics that we want to, we want people to use, essentially. So that's to do with the I think we've talked about this before that the sort of who criteria for different antimicrobials in terms of what the three categories again, jim.

Jame: 50:30

You mean the aware access, watch and reserve. So we should be using access whenever we can watch if we need to and reserve as an absolute last resort. And you know what counts as an access lawyer listeners, stuff like Tremeth, prim Night of your Antoine, amoxicillin, kotor Amox is all is in there, gentelmice, and isn't there as an IV option. You know like you can cure just most UTI, certainly in the UK, with those agents on their own, and then for watch, or the Kephalosporins, including Kefal X and which is a shame it should really be access, to be honest with you And Quinlones are in there as well, and then the kind of hoity toity stuff is reserve.

Callum: 51:19

Hoity toity. I've heard that in a while.

Jame: 51:22

Well, we're talking about all that in a future episode.

Callum: 51:25

All that hoity toity.

Jame: 51:28

But yeah, the post-analytic phase sort of blames sort of into the advice that we're given the clinical teams as well. So it's sort of like kind of molds into treatment of UTIs and how you want to do that and the culture that you want to instill in your hospital or local area, because they'll be going out to GPs too about how you want UTIs to be managed on your patch. But that is a story for another time.

Callum: 51:58

It's worth saying that the post-analytical phase of a result doesn't end in the lab, so that is sort of coming into the people reading the result, and that's, i think, another thing that's worth talking about. So all this has happened and the sample has been taken in the clinical picture that exists in a way, that has been taken, it's processed in a set way and it's reviewed and reported, and then the result lands on someone's desk And that person is often not the person who asked for the urinary sample to be sent or took it spoke to the patient knowing that they'll really have the insight into the lab side. And so you're left with this result which is pretty black and white And even, i think, when it's reported in the terms of possible UTI, i don't know how many people would read into that and understand why that's being said. I certainly didn't understand that before I came to the lab, and it was a complete mystery.

Callum: 52:59

What ends up happening is, you see, okay, there's an organism growing, i will treat it, and I think it's maybe just about walking that back a little bit and saying a number of things. So, if your patient compos mentis and can describe their symptoms, do they actually still ill, Or do they need an antibiotic, because a lot of UTIs will resolve on their own. So there's a whole step there after the urine samples received, and I think it's easy to think that that's simple, but there is. There's a fair bit of complexity in how we actually manage your retract infection, but we've rattled on long enough about this urine culture And I hope that you maybe have learned something new, whether you work in the lab or outside of the lab, or you're not working at all. Maybe you retired or on holiday All right now just waffling.

Jame: 54:02

All right questions, comments, suggestions. Why don't you send them into idiotspodcastingcom? Have a five star review in your pocket, Callum and I would love to have it. Please drop it in your podcast player of choice. We tweet at idiots underscore pod, and if you want to support the show directly, you may now do so. There's a link in the description, but until next time, i'm Jane.

Callum: 54:22

I'm Callum, i've just this minute thought of the pun for the next episode.

Jame: 54:29

Something to look forward to. See you then, bye. Thanks for watching.