107. Staphylococcus aureus bacteraemia (SAB) - interpreting and applying the literature

What exciting times! Clinical trials into the management of SAB! SNAP publication on the horizon! 

But how do we interpret and apply the results of these trials to our patients? This week Jame and Callum are joined by Dr Clark Russell, Clinical Lecturer in Infectious Diseases to discuss:

1. The difficulty in interpreting current clinical trials in SAB.
2. The emerging concept of "low risk" SAB and how to define this.
3. The heterogeneity of SAB and how this might be exploited.

Notes for this episode here: https://idiots.notion.site/107-SAB-update-1316a1ea09d8800ba701ca7ebc8d4093 

Previous episodes for the basics of Staphylococcus aureus and SABATO & SNAP trials:

1. It starts with Staph

65. SNAP trial protocol

72. SABATO trial & 73. SABATaddendum

Transcript

Clark & Callum: 0:18

Hello Clark. Hi Callum. Hi. we're delighted to be joined today by a special guest speaker, Dr. Clark Russell. Dr. Russell is a clinical lecturer and specialty trainee and infectious diseases and medical microbiology in Edinburgh, who has a research interest in Staphylococcus aureus. Yeah. Thank you very much for the invite. It's good to be here. I, work with Clark. but I also heard, um, give a talk recently at the Scottish Antimicrobial Prescribing Group meeting. and if you're wondering who they are, I. We did a podcast episode on that with Andrew Seton and team, so I can go back and have a listen, all about them. But Clark gave a really, excellent, talk on some recent research and how to interpret it on Staphylococcus in particular, bacteraemia. And, so I invited him to come and share some of his insights, So I guess before we dive into Staphylococcus Aureus bacteremia and where we are now, I thought I would just maybe signpost listeners to where you can hear a bit more of the basics about Staphylococcus Aureus, on our podcast, So, episode one. It starts with staff. So right back at the very beginning, Jane and I took a very lab focused look at Staphylococcus. So if you haven't listened to that, that might be worth listening to you before this, if you're not so familiar with this organism, and also episode 65, briefly covers the SNAP trial more than that later. And then 72 and 73 cover the saboteur trial, two trials, one ongoing, the other published about STA caucus or bia. So that's my shameless self plug. Or maybe I could say unplugging J So we're gonna talk about Staal Caius bacteremia, and how do we interpret the recent research that's available, which is excellent. to have, but how do we put that into clinical practice and help use it to, to actually manage patients in real life? so Clark, over to you. What's your sort of opening thoughts on that? Quite broad question. Okay, thank you. So we know that staph ous bacteria is a very heterogeneous disease, which clearly poses challenges for people doing clinical trials. In 2022, which was pre-snap, we systematically identified all currently published clinical trials. Have a look at how they address this. And the first thing to say is that a lot of people have done a lot of great work at trying to improve outcomes for people with SAB over the years. And what we found was that there was quite a lot of. Variation between the different trials and some quite important patient characteristics. So that includes things like patient comorbidities, the source of bacteremia, whether or not they had a complicated bacteremia or not. And fundamentally you could see that these different clinical trials are almost studying different diseases. Mm-hmm. and I think the best bit of evidence that, that this might be a problem in how we interpret the findings from the trials comes from looking at the, 90 day mortality. And we were able to compare, first of all, the 90 day mortality in the control arm of the trials. We then predicted real world patients that would've been eligible for these trials and worked out their 90 day mortality. We then looked at 90 day mortality in unselected real world patients with sab, and we found a step-by-step increase, so the control arm of the trials at the lowest mortality, potentially eligible real world patients, higher mortality, and then higher mortality still in unselected real world patients. So not only are the different clinical trials studying. Slightly different variants of this disease, but overall, they are including patients who differ, in their likelihood of death and therefore probably other important characteristics from unselected patients in the real world. So I think that's a challenge. And then another related point that I think relates to the evidence behind some of the alternative backbone agents, daptomycin, and also Ketta Birol. and I guess you guys might have some views about the correct way to pronounce that drug actually. but,

Jame: 5:09

I dunno what you're talking about.

Clark & Callum: 5:11

very nice clinical trials both published in the New England Journal of Medicine. That, compared to these newer agents with current standards of care, both included quite high proportions of people with Mr. RSA bacteremia and therefore how that evidence applies to settings where 90% plus of SAB is caused by MSSA is again, uncertain.

Jame: 5:34

Can I ask you something about that? I get this, criticism a lot. I'm thinking about particularly to the SABOTEUR trial, when people were, the sort of setup of the trial was such that, if you had MSSA battery, you got different agents compared to if you had MRSA battery immune. so both arms could get Clindamycin, but the MSSA arm got cold room and the MRSA arm got lin lid and. That was used when it finally got published to criticize it to say, oh, well, if it's, it's MRA, you can't necessarily apply to MSSA, surely the underlying basis for that is MRSA is just the presence of, PPP two A on the, The staph aureus. So setting to one side the idea that they travel with other resistance mechanisms and people can look at the SPAR 2024 report for details of what is common in the uk. I don't really understand why. If you've got another agent, which you know that the, thing is susceptible to why it makes any difference, if it's MSSA or MRA, why aren't we able to cross extrapolate between them?

Clark & Callum: 6:44

Yeah, that, that's a nice question. I think the point that I'm trying to get at is that the comparison of these newent anti-coal beams isn't known. Yeah, that's interesting. So I guess of when they come to market, they're advertising and, positioning themselves as anti MRSA drugs and com comparator drug as vancomycin, not necessarily bactam. that, is that Indeed, yeah, that, that's the point I'm trying to make. So I guess Clark you mentioned there about the different clinical trials that we have in staph or bacteremia and how, they include different patients and their mortality is different and that differs from real life So what are these trials that we're talking about? I mentioned Saboto and snap, where they're on and really explain what those are. What are the sort of, for you, the big trials and Staphylococcus aureus bacteraemia that we are looking at? Okay. So I think there are a mixture of trials. So I think we have slightly more historic trials of the different backbone agents, as we've already touched on, maybe more related to MRSA. And those were the studies that provided the evidence for daptomycin and, KEF Bpro. But now we have, an emerging interest in identifying patients at low risk, of adverse outcomes. And, generating evidence that these patients can be treated with, earlier oral switch. And then the other, of trial, which I hope we can come back to at some point are trials of combination or adjunctive treatment. which on the face of it have all been negative. but perhaps with some nuances and how they're interpreted, there can be more encouraging findings. Okay. And what we'll do is, we'll, as always, have a linked show notes, on notion, and now we'll put in all the studies that we're talking about. So if you haven't read them, go and do so, so, you mentioned, selecting certain patients which are low risk. So clearly that must be a universal, decision, is clearly defined and, very easy to tell me who the low risk patients on my ward are this week. Yep. Thanks for that question. So, in an ideal world that would obviously be the case, but that's not really the scenario that we have just now. And I would say that there are probably. Three main competing definitions for low risk sab. one is the definition used in the SABBA trial, which I know you've spoken about, previously. The other is the definition used to determine suitability for randomization to early oral switch at D seven in the SNAP trial. And the other is a definition proposed by very nice study by Hendrick Al, which I'm sure the citation will be provided for, aiming to identify patients that require a less intensive workup, in particular in terms of, cross-sectional imaging. these are the three definitions that I'm probably most interested in. and if you apply the definitions to a cohort of. Patients, the cohort of patients in Edinburgh, around one in five patients meet each of the, definitions, but unfortunately they're not the same patients. and if you quantify the agreement between the definitions using a statistic, the Cohen's Kappa statistic, that gives a result between zero, which is no agreement, one, which is perfect. comparing Sabato with Hendrix gives you a statistic of 0.2 snap with the Hendricks definition 0.4, and snap with Sabato with 0.3. So we have these definitions, but they identify similar proportion of patients, they don't actually agree with each other all that well.

Jame: 10:33

And can I ask, what are the differences? What are Hendricks looking at and calling low risk that Sabato is not and Snap is not, et cetera.

Clark & Callum: 10:42

It's all slightly messy because as you can imagine, the inclusions and exclusion criteria are quite, complicated. But briefly, Sabato, had some restrictions based on the presence or absence of. Prosthetic material, unless the source of bacteremia was skin or soft tissue or an IV catheter. they excluded people who had immunosuppression, liver disease or dialysis, again, unless they had a soft tissue or IV catheter source. And they excluded drug users and they excluded people with, pneumonia. The SNAP definition is probably the most permissive. and includes almost anybody with a skin or soft tissue source or an IV catheter source. but with some exclusions related to the presence of prosthetic material. and then the Hendricks definition, was restricted to people who had a hospital onset breia. So people with community-acquired Breia were not included. Mm-hmm. The features that the three definitions have in common are that you need to have negative follow-up blood cultures. You need to have defeveresced by day three. must not have any clinically apparent metastatic infection. and if the source of bacteremia as an IV catheter, it has to be removed within a stipulated timeframe, right? they're all quite restrictive, and I think for the nuances would encourage. Your listeners to look them up because that was really quite a potted summary of them. What I can do is I can take them from the papers and put them in the show notes so you don't have to, dear listener,'cause we know you're very busy. I mean, potentially the differences between these definitions does matter though, and it maybe depends a little bit on what. What type of low risk you're talking about or low risk of what? because again, when we identified patients in our cohort meeting, these different, definitions, were quite big differences in attributable mortality and also differences in rates of microbiologic failure. Ah, so not only are they identifying different patients, they are identifying patients who have different outcomes of disease. So there probably is a case to be made for. to come up with a consensus definition of low risk, to move forward with. Yeah, that would be really helpful.'cause I found it quite difficult to sometimes apply.

Jame: 13:11

I mean the lead author Ash will gladly admit to the fact that because Saboto was developed initially in 20 13, 20 14, sometime like that, there. the idea of having a, a liberal, definition of low risk, SAB would've probably meant that the trial, wouldn't have been approved ethically. So he was setting this trial up in a completely different environment to what we've got now.

Clark & Callum: 13:35

So I guess there's not been these big trails before recently to actually have defined that. So it's hardly surprising that they've all come to conclusion.'cause there was, it just, there wasn't any way of knowing, is that fair to say? Yeah, totally. And what we're seeing absolutely must not be interpreted as any kind of criticism about the individual studies. and I suppose it should really just be viewed as iterative progress

Jame: 13:57

we wouldn't have gotten Snap without saboteur or maybe we would've, but maybe it would've taken a bit longer, et cetera, et cetera, and ditto poet for all the trials that came before it.

Clark & Callum: 14:06

precisely. Okay. So we will have an image in the notes which, summarizes the overlap, Or lack of, of these uncomplicated patients. Yeah. That's really interesting. Well, I guess the other thing we could mention on this note is that when we. pooled out of our database, the patients that would potentially have been eligible for sabato, just based on the inclusion and exclusions. First of all, we found that although 4% of screened patients were included in Sabato, potentially 15% of the of these real world patients would've been eligible. Yeah. And in this case, although the outcomes were actually very, very similar, which was excellent news, I. there were some differences, in that the potentially eligible real world patients had more comorbidity and also perhaps more significantly were more likely to have an unknown source of their breia, compared to the patients in the trial. So again, probably they're just as these findings get applied in practice, there probably is a need for ongoing prospective observational research to ensure that the. Outcomes of applying it in the real world match those that are expected from the trial as people have done very nicely with, the poet study. There's the, obviously the North American study of it, of oral transitional therapy in real life. and also there's the poetry observational study as well, but I think the same investigators as the original trial excellently named. So Clark, you've talked to us about low risk SAB or whatever that means, and the various definitions. What about complicated sab? So just, I always find this in urinary tract infections, such a headache when we're talking about uncomplicated u uti i and complicated uti i, and these definitions often are, maybe that's a bit more established, but in staph aureus, what would, what do the studies tell us about complicated SAB and how does that translate into the real world? Yeah, so there is, a definition of complicated SAB provided, in the idsa, MRSA BIA management, guideline, and includes various patient and microbiological factors, actors. However, when you map the presence or absence of these different factors onto individual patients, you find that actually they don't tend to overlap all that much, so. Having prosthetic material in situ, being febrile at day three, having positive follow up blood cultures, having metastatic complications, they don't really overlap perfectly. different people are all meeting this definition of complicated sab, but for different reasons. And then if you overlay onto that, people who die again, often the people who die don't have any of these. Potential predictors of complicated sab. Mm-hmm. And I think the issue here is that, after you've become bact remic with staph aureus, there are probably at least two different trajectories. One is heading towards, unfortunately, death due to infection, and the other is towards having metastatic complications of infection and they don't tend to occur in the same people. there could obviously be a risk of immortal time bias and that you, if you die before your metastatic complication is recognized, then in studies like we've done, you will be wrongly classified as not having had metastatic infection. But even after taking steps to correct for that, it still seems to be the case. And part of the reason seems to be that the risk factors for dying and the risk factors for metastatic infection. Again, even after taking steps to try and avoid immortal time bias issues, are a bit different.,

Jame: 17:56

I suppose it's a shame that nobody has, summarized it. In some sort of, leading paper Clark. Isn't that right?

Clark & Callum: 18:02

Yeah, so there, appears to be distinct risk factors for, death and for metastatic infection, even after taking steps to avoid immortal time bias problems, and comorbidity are the major risk factors for death. Whereas, actually increasing age and increase in comorbidity don't really increase your risk of having metastatic complications. factors that are associated with metastatic infection include, community onset of bacteremia and bacteremia from an unknown portal. things you might expect like injection drug use, some things which are probably more of a consequence, like persistent breia. but. Probably overall, the determinants of metastatic SAB are much less well-defined than the determinants of death due to sab. We've got graphics from Clark and co-author papers, in the show notes. and the first has got, You've got a eular diagram, which is the type of Venn diagram where the circles are proportional to the number of patients, which I've just learned. And I edited out Clark explaining that Mike Clark, looking at the components of complicated sap. and then there's also a. Plot of odds ratio, on the logit scale against p value for determinants of both, fatal SAB and metastatic sab, which is quite useful. Took me a while to wrap my head around those images. but there's obviously notes next to them, again, that could support, what Clark has just told us there. so yeah, I'm just looking at that and I guess., Parts of it don't surprise me in terms of what you've said. we know from recent literature that community acquired SAB is much higher risk of mesta metastatic infection, but, there's some, is that, is this eosinophils thing of interest or is it, do you know, there's probably some things that just, if you measuring so many different variables, you might find some yes and no. so eosinophils are included on it for a reason. I think there, because I think it is potentially interesting. I guess there's two explanations. If you're sick, your bone marrow changes what it's turning out. So maybe eos Nia is just a marker of being sick, and certainly there's evidence for that for other diseases. There is a bit of evidence, including from animal models of staph aureus, pneumonia, that th two responses are, beneficial actually in bacterial clearance, which is not necessarily what you would expect. So there it is included there as a little nod to that, as being something that's potentially interesting. I think what a point that I was trying to make but didn't make well is that for fatal sab, it's easy to identify things which are probably causal in your risk of death. So being older, having less physiologic reserve, having comorbidities. Mm-hmm. Meaning that your organ function at baseline is worse metastatic sab, although, you know, we've drawn a diagram like that with lots of statistically significant correlations on it. There's probably only one thing though, which is causal and that's injection drug use nothing else, I should say, is likely to be causal in terms of why you've got metastatic infection. So I think that the determinants of why that happens to some people, I. it's still not that well described. So I guess what you're saying is that we can look at the clinical parameters and, other investigations that might suggest to us. Our patient is likely to have a metastatics app, but as you say, those are, they're consequences that we're picking up the pointers towards it. So they're more the egg than the chicken. Yeah, precisely. I think causality, apart from very specific scenarios like injection drug use, is not that well understood. Hmm, that's really interesting. I'm gonna, I'm gonna take some time. We put the DOIs for all these papers in the, in this notes, I know I've read through them. I think now I'm gonna read them again because, I'm getting a bit more insight into them. Jane, do you have any questions at this point?

Jame: 21:54

So Clark, you've, identified some clinical sub phenotypes of, SAB from your BR cohort and from another couple of, cohorts. Do you wanna talk about them?

Clark & Callum: 22:05

Yeah, absolutely.

Jame: 22:06

I.

Clark & Callum: 22:06

I may just, give a shout out to my colleague Micah Vez, who, was absolutely deterministic in doing this work. and also the collaborators that shared their data with us. So we, yeah, we tried to. Ask can the heterogeneity that we've spoken about in SAB be rationalized in a data-driven or unbiased way? we used a clustering technique called latent class analysis to try and answer that question. And started out by looking at our Edinburgh cohort and found that we could identify five different sub phenotypes. And we use the term sub phenotype'cause these are groups of patients with shared traits within a disease. hence sub phenotypes, but I suppose lacking biologic or mechanistic data to make them disease endotypes. So we picked sub phenotypes, and we then, thanks to nice collaborations, we're able to reproduce. sub phenotypes, first of all in the UK arrest trial, which I suspect will be, familiar to listeners and then also in a smaller Spanish trial, called the CFO trial. And we thought that was important to have some geographic diversity. and in work that's not published, we can also say that we've been able to reproduce them again very nicely in another European cohort. Hmm. So the five sub phenotypes we found. First of all, SAB associated with older age and comorbidity, this sub phenotype was associated with one of the highest, rates of 90 day mortality. We identified a sub phenotype associated with nosocomial acquisition with IV catheters, A source occurring in younger people without comorbidity. So perhaps people who are in hospital for some other reason. and. Acquired it that way, and that was associated with the lowest mortality and also the lowest rates of microbiologic failure. Hmm. A sub phenotype associated with community acquisition, unknown portal of entry, high CRP and metastatic complications, associated with the worst microbiologic outcomes. So the highest rates of persistent and recurrent breia. SAB associated with chronic kidney disease, often in people receiving hemodialysis, often with an IV catheter as the source associated with quite high comorbidity scores using Charleston Comorbidity Index and also associated with high mortality. then staph Aureus BIA associated with injection drug use. Which was associated with very high rates of metastatic infection, including endocarditis, but, with the lowest mortality. perhaps because these were the youngest patients without other comorbidities aside, sometimes from, from liver disease, which was presumably hepatitis C. So we thought that this provided quite helpful framework to, to try and rationalize the heterogeneity that we've been talking about. and as a lot of people rightly pointed out, if you asked an experienced infection doctor to, to just drop some subgroups of sab, they would've probably come up with something quite similar to this. So it's not that novel. And does it matter? Well, since we had included the arrest trial in our study, we were able to then a post hoc analysis of the outcomes of adjunctive rifampicin stratified by sub phenotype. And that analysis would suggest that this might be a helpful way of classifying patients.'cause we saw two things which are potentially of importance. The first is that being randomized to receive Rifampin, if you belonged to the sub phenotype of hospital acquired IV catheter related SAB with very good outcomes at baseline, being randomized to Rifampin actually increased your risk of 90 day or 84 day mortality. Which perhaps tells us that in some patients less is more. and again, it speaks to the concept of low risk sab and perhaps some patients don't benefit from extra treatments and probably benefit from less intense, or less, invasive treatment approaches. Then in the sub phenotype associated with community acquisition and metastatic infection, who had the worst microbiologic outcomes? randomization to receive rifampicin reduced the rate of microbiologic failure. Now, that was a signal that was seen in the trial overall. But the number needed to treat or number needed to prevent one failure, in this sub phenotype was about half of that, of what it was in the, in the original trial. So suggesting that the signal for that probably came from this subgroup of patients, again, we've got an image of different sub phenotypes, A to E, they're titled. and then the 84 day mortality and composite microbiological outcome. Why is it 84 day of interest rather than 90? I think that it was because arrest used 84 days and therefore we. For Edinburgh, we could have used any day,'cause we had, had all the data. But for comparison, yeah, so for it was to compare between the three cohorts. I can't remember which was the limiting factor. One thing that strikes me is the sub phenotype B. So that's our, younger patients capture associated outcome mobility. mortality in the, of those who had injunctive, rifampin is so much higher. To me that seems a bit strange. Like I don't really see people that are getting adjunctive, rifampin, why might that be? Like, it just seems very odd. It's a good question. and obviously you. It could be just statistical noise. you know, this is a postoc analysis, but assuming that it's true mm-hmm. I think the, the rationale that we put forward in the discussion of our paper is that perhaps Rifampin is just complicating treatment of whatever other condition the person was in hospital with, whether that's drug, drug interactions, or drug toxicity. And I suppose if you are receiving an intervention that causes harm when, because of the nature of your disease there is, 0% chance of benefit from it, then you may see a, an outcome like that. Yeah. It can be very difficult though, interactions with rifampicin and sometimes it feels like we've been, pushing to use it and it, definitely we can cause harm first through no harm, I guess we're told. I suppose potentially we're seeing that the patients in this sub phenotype, and we did look at this actually, they overlap quite a lot with the various definitions of low risk s that were available at the time. If we're seeing that these patients are low risk and could probably actually be managed with a week of IV and then a week of oral antibiotics. Giving them actually more intense IV treatment and then an additional drug with a lot of side effects. Mm-hmm. And risk of toxicity on top of that, is perhaps just going to cause problems where there's actually no likelihood of conferring benefits. That was our hypothesis, I suppose, to explain that finding. The other question that sometimes comes into my mind when we talk about low risk sab, and this might be, taboo in the infection community to say this, but staph Aureus does live on the skin and a percentage of people, and I have seen patients, I can think of two, where. The patient has been incredibly well. They've got a staph orange bacteremia identified, but it doesn't really make clinical sense. you know, no fever, no clear source, and we've called a contaminant and the patients have no treatment and gotten better. do those sort of patients, do you think, make it into trial and can that potentially skew things a little bit? Because I think there probably is a group of patients where there's a contaminants rather than a true, pseudo emia we might call it. any thoughts on that? I think many trials have an inclusion criteria along the lines of evidence of disease. Right. I think I couldn't put a number to that. That reassures me,

Jame: 30:24

I agree with Clark.

Clark & Callum: 30:26

Okay. So, just to summarize what we've heard from, Dr. Clark Russell there, in a rapid rundown of the recent research and some, questions and thoughts about how to interpret it. So, we heard about the difficulty interpreting the current clinical trials in SAB due to the inter trial differences in recruited patients, the often low eligibility of the real world patients in cohorts in Edinburgh and others. the fact that evidence, often relates to MRSA since a lot of the more historic trials have been performed in the us. And the difference in the source of the emia and mortality of trial versus real world patients. Then we heard about the emerging concept of low risk sab, so how that's variously defined between trials such as Sabato, snap and Hendrick etl. The fact that around one in five real world patients meets, one of these definitions of low risk, but there's limited. and agreement between the definitions. and as such, they identify different patient groups with differences in outcomes. So there's a real need for a consensus about what low risks are really is. And then finally, how the heterogeneity in SAP can be exploited, by looking at sub phenotypes in particular, the potential differences of the rifampicin effect. such as a higher mortality in the low risk category and better microbiologic outcomes in community acquired. and potentially, looking at things like stratified trial recruitment, might inform future, clinical trial. recruitment, and I guess even before it gets to that I think if you read the sub phenotypes paper, certainly I think that helps me in my head. Sometimes you're presented with a staff always battery immune clinical practice and it can be quite hard to make. I. Make out, am I worried about this patient? What sort of level of investigation, what sort of timing of oral switch might I consider? So this can potentially help by saying, actually, this person fits into that sub phenotype. So I'm maybe gonna be a bit more conservative with my management. I'm a bit more intensive with my workup and investigation. That's personally what I might take from this. So Clark, any thoughts or statements for listeners? yeah. So, first of all, thank you for, having me on the podcast. I think, uh, heterogeneity is a common theme in sab. I think it complicates, the application of some of the emerging evidence that we've spoken about to real world patients. I think that low risk SAB exists, but there are multiple competing and unfortunately, currently distinct definitions that need to be rationalized Early. Oral switch is clearly an appropriate option for some patients, but we maybe need a more precise definition to allow these patients to be reproducibly identified in real life practice. Probably we should prospectively study the real world outcomes of early oral switch. When that becomes a more mainstream practice, it's likely that of the many adjunctive therapies that have been studied and rejected in sab, there are differential treatment effects. And it may be that potentially beneficial therapies have been falsely rejected because of current trial designs. That's because the investigation of SAB as a single syndrome might be overly simplistic and in some cases might be inappropriate leading to falsely rejecting beneficial treatments. We've shown that you can reproducibly identify clinically relevant sub phenotypes amongst patients with sab, and that using that type of approach for patient stratification could identify some of these, differential treatment effects. In the example that we've. Is with adjunctive Rifampin. Well, I guess all that remains to say is thank you for coming on the podcast and for, all the work that you've been doing, with colleagues to, to give us some answers because, I'm not a researcher, so I'm just very glad that there's lots of people working hard to give us the answers, so that we can do the best thing.

Jame: 34:35

Yeah. Thanks a lot.

Clark & Callum: 34:36

thanks for the opportunity to talk about it all. It's been really good. don't think I'm a natural so thanks for, yeah, no, thanks for keeping the show on the road. It's a growth mindset thing, so I don't think Jane and I are natural podcasters either. We just, we just got it wrong enough that we stopped worrying about it.