All right, well, good afternoon, everybody, good morning. So I have no relevant disclosures as it relates to the content of this talk. But I will say that, like obscenity, we think we know sepsis when we see it, but it turns out it's really hard to define. And the currently used sepsis definitions that we have heard about already this morning were largely developed by consensus, were not data-driven, were not developed with a broad resource availability in mind. They can overlap with noninfectious diagnoses. They give the diagnosis of sepsis to patients who are not critically ill and are difficult to apply for all the purposes for which they're used. Here are the consensus definitions from 2005 that we've heard about. And again, we think about sepsis as defined by these definitions. And it applies to many, many patients on the regular ward who do not have life-threatening disease. We're more used to thinking about in the ICU patients with severe sepsis, septic shock, with organ dysfunction. And these definitions include a mixture of vital signs, laboratory abnormalities, and intervention-based criteria. And again, they were consensus-based and not data-driven. This is in contrast to the World Health Organization definitions, which are less dependent on testing and more dependent on interactions that can happen in any resource environment. We've heard about the sepsis-3 definitions as well, exclusively adult data, SOFA-driven, the low and middle-income countries were not specifically addressed. And the term sepsis does not differentiate local versus systemic effects. More on that soon. So we want to understand what the people wanted. So we drafted, revised, and disseminated an international survey that went out to 27 international societies, who then distributed it to their members in multiple different languages, with items that asked about demographics, resource availability, current practice for sepsis diagnosis, viewpoints around the usefulness of the current sepsis definitions, what the new sepsis definitions should mean to those people, and what, in fact, the word sepsis should mean. So as you can see, the survey went all over the planet, with the most heavy representation in North, South America, and Asia. There were 2,835 analyzable responses across the spectrum of high, upper-middle, and low and low-middle-income countries. The work settings ranged from academic and metropolitan to non-academic and non-metropolitan, although the majority came from academic metropolitan hospitals. The majority of the practitioners worked in the PICU, however, the ED and the regular hospital wards were represented robustly as well. The preponderance of the responses were from physicians, though the survey was indeed multidisciplinary. So we asked these sites, or individuals, what resources did they have available to them for testing and interventions? So around testing, 80% of respondents or more indicated that they had basic laboratory testing available to them, and similarly, 80% of respondents included that they were able to give fluid boluses, they were able to use vasoactives, and provide some level of ventilatory support. When we asked the respondents, what do you use to recognize children with sepsis, using the word sepsis rather generically, they indicated that vital signs, derangements in the host immune response, and infection-related testing were key in their decision-making. When we asked them, how do they recognize septic shock, measures and markers of organ dysfunction, most notably inotropic support led the way. So we then asked them, which of these patients would you say has sepsis? A child with local infection without organ dysfunction who's home, someone with pneumonia who's on room air, someone who might be hospitalized with local infection, but without severe organ dysfunction, a child on the ward with an oxygen requirement, for example. What about a patient who has life-threatening infection, but their only organ dysfunction is at the site of that infection? So let's say a patient with pneumonia who has an invasive mechanical ventilatory requirement, but whose other organs are working fine. What about the patient who has life-threatening infection, but with organ dysfunction that is remote from the site of infection? So again, a patient with pneumonia and a ventilator requirement who also has acute kidney injury. And then life-threatening infection with cardiovascular dysfunction, where you throw, say, a vasoactive requirement in the mix. Well, clearly our respondents felt that local infection without organ dysfunction wasn't sepsis. But when it came to life-threatening infection, there was not unanimity around what we should call sepsis. But the answer is somewhere in there, in the population of patients who has life-threatening organ dysfunction. Now the use cases for sepsis definitions include recognition. Does this patient have sepsis? Or early recognition. Is this patient developing sepsis? Correct disease classification. Is this or was this sepsis, or is it something else? What is the patient's risk for adverse outcomes or prognostication? Benchmarking. How good are we in taking care of this patient or population of patients compared to other practitioners? Who's getting sepsis? When are they getting it? Are they getting it with our epidemiology uses? And then, of course, understanding the biology of the disease, developing new therapies and testing them in studies. So we asked the respondents to evaluate how the consensus definitions, the sepsis 3 definitions, the WHO definition, worked in their hands. And they felt that recognition and early recognition was pretty good, but that the usefulness of these diseases, of these criteria, fell off using other use cases. But the fourth line in each of those figures is what they want the new definitions to do. They want them to be great and broadly useful across all use cases. So ask a question, you get the answer, right? So the limitation of this survey is the response rate was unknown. The denominator is unknown because we didn't know to whom the surveys were sent by the individual societies. There was representation from low and middle income countries, but still relatively underrepresented compared to upper, middle, and high income countries. These are perceptions, not actual practice. These are clinicians' views. And they did not specifically address neonates. So the conclusion here is that the international community of clinicians who care for kids with life-threatening infection, they do have limits on their availability of diagnostic and therapeutic resources, but vital sign measurement and basic laboratory testing are frequently available. The community feels that current sepsis definitions, they're inadequate for use across the spectrum of need, including recognition, quality benchmarking, and research. But they want a set of definitions that does it all. And it is not unanimous what the term sepsis should mean within our community, but there's strong belief that it should apply to life-threatening disease. So we turn to data for definitions which are pediatric-specific and broadly applicable across resource settings. So again, thanks to the group, particularly Luke, Lauren, Pierre, and Lauren for an outstanding bit of work around the survey.

sepsis

definitions

clinicians

organ dysfunction

pediatric