It's great to be here today to talk about something that I'm truly passionate about and has become a big part of my career as it has evolved. And it really comes down to the question is, do we or are we in a place where we can begin to manipulate the microbiome to improve outcomes in our patients? And I think it really largely gets back to the idea is we spend a lot of time as critical care physicians or physicians in general trying to eliminate bacteria or give antibiotics that destroy bacteria when the reality is, of course, we really should probably be preserving them. So these are what I would like to call my alignments of interest. I'd like to believe I'm not in conflict with the NIH, the DOD, or any of the industry sponsors I work with. I think, I hope, we're all in alignment in trying to improve outcomes for our patients and how they recover. So again, rather than eradicating bacteria, I think I would advocate that the majority of the bacteria that we live commensally with in our body, symbiotically with, are friends we didn't know we had, most of us, and friends we want to keep around or give back or put back when we are sick because we lose them, as I will show you. And so I think the day where we prescribe antibiotics as first-line therapy hopefully will become less and less. And the idea that we're going to give back bacteria and other microbiome components as first-line therapy is coming to reality, right? There's books about it. You can do this at home, although I discourage it. And in fact, there's prescriptions you can get. The kids at MIT will tell you that they have poop pills they'll sell you, and they will make you smarter. The FDA has a problem with that, but nonetheless, they're available. So we know that we can cure disease, and we do have meaningful therapy with microbiome interventions. Stool transplants for C. diff, as I will show you some compelling data, are undoubtedly successful and very compelling and now becoming standard of care interventions for refractory C. diff. And that brings the question is, should we be residing the lawn of the normal microbiome that we lose when we have critical illness, we receive antibiotics, we have COVID-19? Should we be giving it back as stool transplants, probiotics, or the magical poop pills? So why should you care about your microbes? Do they actually do things for you every day that might imply they actually have a therapeutic value? Well, have you ever wondered why you get bit by all the mosquitoes when you go into the woods and your wife or husband or friend doesn't? It's not because you're sweeter. It's because the bacteria that live on your skin attract or repel the mosquitoes themselves. So they play a pretty big role in disease and in mosquitoes. In fact, we are finding that perhaps even our choice of who we marry and or mate with may be due to the microbiota that live in our gut, if you're a fruit fly. Maybe true for humans as well, but we will see. But again, these have very fundamental effects in organisms, humans, and really all of life. And so how do they do all that, right? And that begs the question is, are we really more human or are we really more microbial, right? If you have 30 trillion human cells in your body right now sitting here, there's 38 trillion microbial cells. So you are more than 50% microbial as you sit here today. Perhaps you think it's your genes that make you human and define your response to illness. Well, you have about 20,000 human genes in your body now and you have about 20 million microbial genes. So you are 1% genetically human and about 99% microbial. So we're poised to begin to answer and use this information to perhaps treat illness and perhaps improve outcomes in our patients. And so what's in our patients' guts, and I will also add mouths, noses, skin, and can it help or hurt them? Because all the things we do in the ICU create a very nasty and robust effect on our microbes. The opiates we give cause virulence in our microbiome, phosphorus levels, a lot of the different drugs we give, the antibiotics obviously change our microbial composition and make them more aggressive. I actually trained with John over at the University of Chicago and he was the first person to sort of put this in my mind before he went on to be the great leader in this field that he is. And so the idea is can we restore balance that is lost when we become ill with things like probiotics or poop pills or stool transplants and re-sod that lawn. What's the science behind this? Well, you heard some excellent purvey of the science and I'll try to bring it to even a more straightforward thought. Basically when we become ill, we lose our healthy commensal bacteria. They get wiped out and our pathogens that live in our guts, because all of you have pseudomonas in your guts right now, whether you like it or not, probably more as an intensivist, but they realize everyone's left and they attack us. It's a little bit like when a hurricane hits New Orleans, everyone evacuates and the looters come and they bring their looter friends and the other bad bacteria come. And so the idea is by the loss of the people that live in the city or the gut, perhaps the bad actors appear. And so the idea is if you put the regular people back, maybe the looters will leave and they'll get overwhelmed by the people that are supposed to live there in the first place. And so the idea that commensal flora can be given back as probiotics or other intervention stool transplants causes the suppression of the virulence and the amount of the abnormal and pathogenic bacteria. Prebiotics are critical as well. We know we need to fuel the bacteria with prebiotics like fibers and other things, and that has direct immune effects. It affects searching fatty acids, which affects how our immune system responds. The lack of fiber actually leads to all kinds of very dysfunction and inflammation because our normal bacteria can't produce the stretching fatty acids they need. So what happens briefly to the microbiome and critical illness? So we had proposed a probiotic trial years back and the NIH said, we don't believe that one probiotic is going to treat everyone. Go back and tell us what's missing or what is lost that is a possible target in different patients. So we embarked on the ICU Microbiome Project with Rob Knight, who's a microbiome scientist, and we published the first paper looking at a large number of patients, 115 patients across four centers. Everyone was on a ventilator. We discovered you can't be on a ventilator in North America for 48 hours and not get an antibiotic. Not a single patient didn't get antibiotics in this group. And you can see, we compared that to this large population of American gut. This is Rob Knight's group of thousands of people who send them their poop to get their microbiome readout back. What they do that we don't know, but he's funded millions of dollars of science by people sending them their poop and their dog's poop and their friend's poop. And so the American gut normal people are in red. The ICU patients are in blue. They're clearly different. And in fact, key commensal organisms that help us maintain health, like Firmicutes, disappear. Those big white dots are the ICU patients. White means none. Red means lots. You can see the normal people in the small dots have a lot of this family that's important for health. And so again, that's a major change that happened within days. And in fact, normally the NIH microbiome project is the small dots. And you can see there's normally a separation in the constituency of our gut, our mouth, and our stool. And our skin, sorry. When we become critically ill, we lose that. We lose barrier function. In all those places, our stool, our mouth, and our skin begin to look the same. And that's one of the findings we found. And of course, our pathogens abound. These are the looters that come in. And you can see in the red there, both an emission and a discharge, there's an abundance of pathogens that in a healthy subset, the graph below, aren't present. So again, specific things are lost and gained. The pathogens are gained in large amount. They grow up very quickly and take over the guts. Often 95% of all bacterial species being made up by one species of pathogen, often Enterobacter. And we lose very healthy bacteria, like Fecalibacterium, which makes short-chain fatty acids, which is essential to immune function and health. And so that's rapidly depleted. So again, this is one of the findings we found. And of course, we lose diversity. And we think diversity, we hope, will continue to give us the thought that that equals health and the robustness and the diversity our gut needs to have for us to maintain health. So we also found that when this diversity is lost, patients stay longer and do more poorly. So longer lengths of stay were associated in the ICU with less diversity. And this is what this graph exhibits here. We did find targets. A lot of them are the same targets that inflammatory bowel disease researchers had already found. And so these are some of the targets we're exploring, Fecalibacterium in particular, as potential probiotic interventions. So we also found the loss of this diversity led to increased risk of death. And so we're continuing to analyze this data. So what's the clinical data? Because that's what you're here for. So we do have some robust data that interventions with probiotics and other things can improve health and treat disease. We know if you have antibiotic-associated diarrhea. We see a lot of that in the ICU. Using a probiotic clearly reduces the risk of this. This is a large paper, 11,000 patients, 40% reduction in antibiotic-associated diarrhea. Virtually all the different kinds of commensal probiotics work. They will outgrow the pathogens almost every time, given the right substrate. And so again, it doesn't seem to necessarily be a certain one, although there may be benefits to certain organisms, as we'll see. This led to studies of things like vanilla associated pneumonia. And this was one of the seminal trials. Dr. Andy Morrow did this as his K award from the NIH, 146 patients on mechanical ventilation to prevent VAP, vanilla-associated pneumonia. He gave lactobacillus GG to the mouth and to the stomach. That's important. Remember that. And he saw a 50% reduction in vanilla-associated pneumonia by doing this in this NIH-funded study, and significant reductions in C. diff as well. This then led to Deb Cook, and a group of us worked with her on the pilot trial for this, to doing a large randomized trial worldwide on lactobacillus only given to the gut, not now smeared in the mouth, looking at the effects on vanilla-associated pneumonia. And she did not see an effect in this population. So again, we learned when we do large trials, it's hard to see an effect in large patient numbers. But I think there's some differences in how she gave it. You can't just treat the gut. The mouth is very important to our microbiome and to what our lung sees. And so these are some of the learnings we had from this. The other thing, there was no effect on any of their outcomes, but they did see a small incidence of lactobacillus showing up in blood and or other sites. It didn't necessarily cause severe disease. Lactobacillus is not particularly a pathogen, but it can escape the gut, and it can end up in the blood and in other places. So there are some risks involved that have to be thought of when using probiotics. And so there is a meta-analysis now that has taken all the data into account, including this large study. And there still is a strong signal, which we had published before in our meta-analyses, of a benefit on villain-associated pneumonia on outcome, on mortality, or on villain-associated pneumonia occurrence, I'm sorry, using probiotics in general. And this is all different kinds of probiotics, not just LGG, which was studied in that trial. And actually, on a lot of the outcomes that were studied, probiotics continue to show a signal of benefit in the trials that have been done, including serious adverse events and other things like ICU length of stay. But we continue to need better trials and probably more mechanistic understanding of how these work. So low-certainty RCT evidence suggests there still is a benefit. I still use them at Duke in certain patients, and so I think there's still a potential benefit. This has been shown in other areas, like elective surgery and trauma as well. But again, we continue to do more trials, and we continue to get more data. What about prebiotics? That was one of the things I was asked to talk about. And on a bright note, we have new unpublished data that we only got a week ago. So these are the ones I don't encourage you to take pictures of. I encourage you to share on social media everything else I say and everything else I show. But these are new data that we haven't published. Don't share these. This is the idea that fibers are critical to the recovery of those normal bacteria in our guts. And if people receive antibiotics and they're on a fiber-free diet, like intramutrician, they have pathogens that are maintained. They don't recover their normal commensals that you see in green and blue there. And so the idea is if we gave back prebiotics in our intramutrician feeding, perhaps our patients would recover their commensals better and reduce their pathogen load. So this is Dr. Mara Surbanescu, who I tell you is the next superstar in the microbiome world. She's the only intensivist I know that is able to analyze her own microbiome data. She came to my lab from Hopkins, where she trained with Craig Coopersmith as a medical student at Emory. And so we've known Mara for a long time, and she's really outstanding. But this was a trial of trauma patients using a fiber-enriched formula versus a non-fiber-enriched formula in trauma patients who didn't have GI injury. And so this is new data. So again, the method is single-center RCT, mechanically ventilated patients with severe trauma without GI tract involvement, and that received a fiber formula with fructooligosaccharides, which is a common thing in some of our enteral feeds, or a non-fiber formula. And so ultimately, we collected swabs from the mouth and from the stool over 20 days. We did 16S sequencing. And we ended up with 10 fiber patients and 7 control patients, because it's very expensive to analyze lots of samples from this number. So you can see how many samples we had from each. And you can see, as you'd expect in all the patients, we saw a decrease in the diversity and richness. The Shannon Index talks about the diversity of the microbiome or the richness, and the number of species. We should have lots of species in our gut. And that drops quickly in the mouth and the gut when we're traumatized or critically ill. We've seen this before. And in fact, we saw significant increases in pathogens. Shigella, for instance, Escherichia shigella, and Staphylococcus in the mouth went up dramatically in the patients over the few days they were in the ICU over the first two weeks. The fiber did cause a difference in the beta diversity, or the differences between the organisms. So all we know is fiber was having an effect. And now I'm going to show you what that effect was. Unfortunately, the diversity went down in the fiber group. It was unexpected. And again, it's only 17 patients, but it was a lot of samples. And that was true in the mouth as well. So our diversity actually dropped with this fiber intervention. There are some reasons why Lactobacillus actually didn't go up as we'd hoped it would. But Shigella did. So it looks as though, and I'm going to show you more data for this, that this fiber is actually feeding the pathogens that are the looters living in the town with no one else around because they begin to take advantage of it. So the idea that we can give fiber alone probably isn't ideal. Now some of the beneficial species also use the fiber and went up as well. But it looks like it feeds the pathogens and the commensals. And so perhaps we need to be giving back symbiotics, which we'll come to. In the mouth, it was the same. So Neisseria is a helpful commensal that's been shown to be associated with reduced infections in other parts of the body. The fiber receiving individuals actually did not have as much of this commensal. And there was more of this pathogen Varanella, which is associated with other pathogenic outcomes and pathogenic species. And so it didn't do actually what we'd hoped it did. And then we did these network analyses, and I won't go into too much detail of this, because it's actually beyond my total understanding. This is what microbiome scientists do. But essentially what this showed is Shigella is interacting with pathogens and commensals to use the fiber to help themselves all grow up. And so it looks as though the fiber was promoting this relationship between the pathogens and the commensals to grow. And so in the end, in contrast to other populations where fiber has been shown to be beneficial, in ICU patients with trauma, it looks as though it may help to grow up the pathogens as much as it does the commensals. And so it may foster an interaction that allows this to happen. And so again, it may be feeding both, and so I think we have to understand how the bacteria interact in each individual disease state. And ultimately, I think we need to be giving back commensals if we're going to give fiber, probably if we want to get them to grow. So again, what about where we know probiotics have worked, or interventions with the microbiome have worked? C. diff is definitely one. Just briefly, there's a 64% reduction in your C. diff risk if you get a probiotic, especially if you're at high risk. This Cochrane analysis showed that at 70%, if you're at a high risk of C. diff, have received multiple days of antibiotics. So there's compelling data for probiotics and C. diff. And that was true in adults and children with different probiotic species and at different doses. So again, a fast-growing commensal will outgrow C. diff every time. It's a weak organism. And we know stool transplants are very effective. It has a cure rate of over 90% in our patients. And what's the effect? This is a brilliant study from the University of Minnesota that looked, these are the normal, what a normal microbiome looks like in volunteers. The NIH spent $200 million collecting this picture right here. And this is a group of patients with C. diff from the University of Minnesota and their ICUs in the stars. So you can see, they should be down in that fecal dark area, but they begin to lose, like we saw in the ICU project, they begin to lose that sort of diversity and that sort of dysbiosis occurs. This is the donor down there in red. That's what their microbiome looked like, like it should. And this is what happened to the patients who received it when they got that donor. Their microbiome was measured each day with fecal swabs, and it immediately returns to a normal fecal picture like the donor, and it stays there. All those lines are day after day after day. So it's a consistent effect. So really compelling data. Quickly a trial you have to know, as we think about infection and sepsis and probiotics, is this Nature paper. This is a large trial of lactobacillus with fiber, which I think is, if it's going to work, this is how it's going to work, in full-term infants in India, looking at respiratory sepsis and death in full-term healthy babies. And they looked at thousands of these young people, randomizing them to a lactobacillus plus fiber or a placebo or a control, and what they found was a massive reduction in pulmonary infection, sepsis, and death in these infants. And this is by far the preeminent microbial intervention for infection that's been in the literature. This is one you have to know. And it actually cited some of the research my lab did on the effects of the microbiome and lactobacillus on T helper cells and T suppressor cells. So it has a salivating effect on the lung, on our immunity. And that was one of the reasons they chose this particular probiotic. So again, there's other data for this. This large meta-analyses have shown this is true in adults as well. There's lots of randomized trials that show you all could be reducing your risk of respiratory tract infections if you were taking a probiotic. You can reduce the length of your illness by two days as well. And this is true in both adults and children, 32% risk of adult RTI, respiratory tract infections, if you take a probiotic. And so this is a 10,000 patient meta-analysis. We took this data when COVID started and thought, can we prevent COVID with probiotics? This is the last thing I'm going to tell you. This was just published actually about a month ago. So it's new data as well. This was a trial we ran during the first six months of COVID before the vaccine came out. And we looked at 182 participants, and we found them in our Duke database of people who had COVID. We called their families, said, would you be randomized? The whole family took a probiotic or a placebo, and it was all done over the internet and by mail. And so we randomized 182 participants. They got Lactobacillus GG, much like the critical care trial I told you about. And then we followed them for symptoms and COVID diagnosis remotely. And so we mailed them overnight these things, and we sent it to the whole family. This is what the trial looked like. We collected swabs. So we got microbiome data, and that also looked at compliance. So we had a number of different levels of analysis we did. There were patients that, of course, got COVID symptoms while we were enrolling them. There were patients that sort of were at different levels that never took the probiotic. And then there were people who actually were asymptomatic and started taking it. And I'm going to show you data for each briefly. So we reduced symptomatic COVID in people who took a probiotic in the family when exposed to someone who had it by 50% approximately across the board, 40 to 50%. So it was quite effective at reducing symptoms of COVID. And in fact, there was a trend towards reduced COVID diagnosis absolutely in the group that didn't have symptoms at the start of treatment. And in fact, there was a significant reduction over time of the symptoms they had from taking an LGG probiotic, like the kind you get in the store. And there was a statistically significant reduction in time to COVID diagnosis. So less people were diagnosed over time if they took this probiotic. We wanted to enroll more patients, I will tell you, but the vaccine coming out really slowed our enrollment because we didn't want to enroll vaccinated people. We looked at the fecal microbiome. We found that people taking the probiotic actually seemed to be taking it. They had a great enrichment of lactobacillus. And we did see significant changes in people with COVID's microbiome as well as people taking probiotics microbiome. So it was having effects hopefully in a favorable way because it reduced symptoms, reduced diagnosis. So again, via intention to treat analysis, we had about a 40 to 50% reduction depending on the group you looked at in symptoms and a significant reduction in the time to COVID diagnosis or prolongation, I suppose you would say, in patients who received it. And this was just published in the last couple of weeks. So again, we were encouraged that this had a positive potential effect. Not only people who weren't vaccinated, but we know probiotics like this can help the vaccine response and influenza and give us a more robust vaccine response. So our hope is it would actually help a vaccinated person as well. So I think there's a lot of promise for this. And as I close, what have we learned so far? We believe a healthy gut is a diverse gut, and all of us should strive for that. Unfortunately, critical illness in Western diet leads to a not very diverse gut, which is not ideal for outcomes. And we need to probably learn to correct this and be thoughtful about it. Because the world we live is in a microbial world, and we can change this world, we believe, by prescribing stool and bacteria and microbial components rather than killing them off with antibiotics. And I think that day is coming. So I hope we can learn to re-sod the lawn properly, continue to do more and more research in this area, restore balance to our patients, and make it so easy a child can do it. And with that, I'll thank Rob Knight for all his help in this. And please feel free to reach out by email if you'd like these slides or any of these papers, and share any of this information on Instagram or Twitter. Thank you.

microbiome

probiotics

antibiotic-associated diarrhea

C. difficile infections

microbial diversity

prebiotics