So, I'm Thomas Zagmané, and I work in the UK, so there is a bias right there, because I work in a slightly different environment. These are my disclosures, none of which is relevant for the talk. And I think the first question is which bundle, because we have heard about that, you know, there is the three-hour bundle, there is the SEP1 bundle, the one-hour bundle, and, you know, I've got no idea what bundle I'm talking about. In the UK, we've got the infamous, slightly infamous, SEP6 bundle, which was sort of developed by an advocacy group without any data behind it, and that has been revised recently. So, which bundle we are talking about, that is the first question, and then once you have decided the bundle, then I will ask that who is septic, and, you know, which patients we are treating with these bundles. And I go, again, it is biased, and I just want to make sure you hear the word, it is biased, because I'm going to show you data from the developed countries. And this is data from the U.S., Tim Buckman four years ago produced a series of papers on the Medicare and Medicaid benefit recipients to see what is happening with sepsis and what is happening with the mortality. And as you can see, the number of patients are going up. The mortality is maybe slightly going down, but it's still very, very high. They have repeated the study a year later, gathering some new data. And then you can see that the majority of the patients and the increase in the number of patients are actually the non-severe category where you don't know what is the organism which is causing the problem. The septic shock patients are relatively stable in numbers. So when you think about the bundle and the very different elements of that bundle, do they apply to your 24-year-old who has got a rash and a deteriorating GCS and a progressive hypertension? You give antibiotics, fluid, and vasopressors, that is entirely appropriate. But if you are septic, whatever sepsis means, and you are on the ICU with a ventilator-associated pneumonia, do you need to do the same thing? Or do you need to do the same thing to your 87-year-old nursing home resident with advanced heart failure, COPD, et cetera? And you give the same bundle to everybody. When you look at the data, and this is on the left-hand side or on your right-hand side, this is our data, which we will present in the RSDs later, looking at the Welsh population, you can see that, again, it is the non-severe cases which are giving us the bulk of the patients. Now, age, again, it does mean a big difference that who is that patient that you are treating. And from the U.S., Chanru Rhee and colleagues showed us that probably 70% of the patients that you are treating with sepsis, whatever sepsis is, are over 60. So they are not in the young category. And again, from our data, it is very clear that the older category, and we looked at over 85-year-olds where the incidence of sepsis diagnosis was going up, is higher. But also, in those patients who had got severe comorbidities, you can see that the incidence is going up much faster than in the other groups. This is data, again, from our little point of prevalence studies in Wales, and I just highlighted a couple of lines. We do have a lot of patients who are relatively in the middle-older age category and with a lot of comorbidities and a lot of heart failure, a lot of COPD that is very prevalent. So maybe have a think about whether you want to use the same hammer to the very different nails in there. And I'm going to talk a little bit about bundle elements. And I know this is not in the SEP1 bundle as a highlighted one, but it is in our sepsis 6 bundle to give oxygen. Giving oxygen is good, isn't it? It's very, very clear that to give oxygen to critically ill patients is good. Well, maybe not so. And then if you go to the thought leader sessions, which is coming up with the late breakers, then you will hear from Mark Peters about the results of the OxyPICU study, which does seem to suggest that actually limiting the oxygen exposure in critically ill children, most of them who were septic, is associated with better outcomes, and that's a large RCT. There are two major other studies going on. One is the MEGAROX sepsis, which is done in Australia, New Zealand, and in Canada, which is looking at the oxygen exposure, especially in septic patients. And then the other one, the UKROX, which is done in the UK, again, looking at limiting the oxygen exposure, are very large studies. We are talking about, when we combine it, over 25,000 patients randomized into these trials. They will report in probably two years' time, and maybe we should have a think whether we recommend a very liberal use of oxygen until we've got some evidence. Now the other very clear thing, which we all know, or at least we have been taught, is that we give early antibiotics, and that's, you know, we have to give the early antibiotics. Now we can talk about what is early. The only randomized control trial data is the Fantasia trial from the Netherlands, where the paramedics gave a broad-spectrum antibiotic to patients who they screened septic and in a randomized fashion, and the results were absolutely negative. Now the authors have reanalyzed that data very recently, and what they have found is that this very early antibiotic, you know, before hospital, you can't really go closer to time zero than that. Before hospital administration of antibiotics was probably associated with better outcomes in the young patients, in the below 40 population. It was associated with a potentially worse outcome in the elderly population. Now we've got young people with sepsis who are really in a life-threatening situation at this level, and this is the number of patients we treat who are in the other category. So maybe have a think about that. And again, this is some UK data to show that it is really difficult to change the clinician's culture. So if you start an antibiotic, then it will be really, really difficult and sticky to stop it. On your left, there is a graph from the AHRQ study, which looked at in a step-by-step randomized fashion to try to limit the antibiotic exposure in the hospitals in the UK. They had a hard stop at day three of antibiotic administration in the intervention arm, whereas compared it to the usual care. And then you can see that there was no difference in the mortality when they looked at this outcome in the 30-day mortality. But they have been able to reduce the broad-spectrum antibiotic exposure quite significantly. So maybe we are not producing so many motor-resistant drugs. And then on the right side of the slide is just to show that what is happening. This is, again, our study when we looked at what is happening in real life. Patients have azosine deficiency. So PIPTA's deficiency is something that we have to treat or we are treating, but only in about 64% of the patients who have been screened being septic. So what is going on? Are we giving early antibiotics, very broad spectrum, but we are not giving it to everybody or to the right patient? So I do have quite a bit of concern about that. I'm not going to labor the fluid. The last 20 years of critical care and RCTs told us that less is more. So I don't think that the 30cc is particularly a good idea. And again, just to show that if you do have significant hypotension and you do need vasopressors, if you treat your patient to a lower goal, a lower target of mean arterial pressure, you can actually have the same or even better outcomes. This is from the 65 study from the UK ICUs, which was published a couple of years ago. So Doleen has showed us the IMPRESS study. Are you impressed with 19% of compliance? I am, because that is quite good. In Wales, with a much simpler bundle, we could only manage 14%. And then you can see on this sunburst plot that then the rest is the clinicians just chose what they want. And as I said in the beginning, if you've got a bundle which really works, and then the clinicians will only apply it to 1 in 5, 1 in 6, 1 in 7 patients, then that is really the worst of all. And why I say I showed you biased data? Because we are talking about a global problem. And yes, we've got data from the developed countries. But in the less fortunate countries, and especially in the low-income countries, the sepsis patients and the sepsis mortality is very, very different. That is from the IMPRESS study that we have already discussed. And then just to show that the evidence that we are basing our global sepsis guidelines on are actually coming only from the developed countries. So are we trying to force something which doesn't seem to be working, to be honest, in the developed countries on the other side of the world? And are we doing more harm than good? So I think in summary, I do think that the sepsis bundle elements, or in fact any kind of bundle elements, should be rigorously tested in our cities before we bundle them together. Because that's how we will get better patient outcomes. Thank you. Excellent.

sepsis management

treatment biases

bundle protocols

patient demographics

clinical practices