»» Yeah, definitely a theme here today. So like I said, my name's Ina. And the abstract I'm going to talk to you guys about is use of focused cardiac ultrasound and how it's associated with higher SEP1 fluid compliance in our patients with septic shock. So just a brief about me, I'm a PGY-3 emergency medicine resident at the University of Michigan. I did my undergrad at UT in Austin. So we are from similar places and got my medical doctorate from UT San Antonio. And I'm currently applying for a fellowship in anesthesia critical care medicine. I have no disclosures. And the things that we're going to kind of talk about today, you guys got a great overview already of the surviving sepsis recommendations. We're going to touch on those briefly. We're going to talk about some previous literature that we have about using ultrasound in our acute resuscitation periods. And then we're going to look at the influence of actual ED providers using cardiac ultrasound and how this interplays with their septic shock resuscitation strategies. So this very infamous well-known SEP1 3-hour bundle that a lot of emergency physicians are kind of graded on, hospitals are graded on, includes a couple of different pieces that for the most part we can agree on, right. Like antibiotics for acutely ill and infected patients make sense. Measuring a lactate to look for microcirculatory dysfunction or organ malprofusion, that makes sense. And getting blood cultures to look for a causative organism makes sense. But by my previous colleague here, Clover Study, the class that's been out recently, there is a lot of concern and kind of looking at whether or not we should be as aggressive with fluids as we've been previously taught. And so there's a lot of controversy with this 30cc per kilo fluid bolus bundle. And like really where it comes from and especially from like the emergency department community is like we worry about causing harm, right. We see all these papers that show that positive fluid balance is not good for our patients and especially those patients with CHF, CKD who may not be able to tolerate volume loads. We worry about actually causing them harm without necessarily clearly having proof of efficacy. And our current guidelines represent that and reflect that. So in 2021 it went from recommendation to suggestion for this 30cc per kilo bolus. And starting last year CMS SEP1 measures actually allow providers to document why they did not give the full 30cc per kilo bolus without being penalized as being deemed non-compliant. And I think where a lot of this comes from is like we as providers and clinicians, we want to kind of tailor care for the patient in front of us. Like some people need that 30cc, but maybe some people don't. And again this is reflected in some of the guideline verbiage where they talk about using dynamic measures to assess for resuscitation and fluid responsiveness. And one of the things that they recommend is echocardiography, which here I'm going to call Focused Cardiac Ultrasound because it's provided by the bedside clinician. But the question is, what is the data for that? Do we know if it's actually helpful? And unsurprisingly it's kind of all over the place. So we do have some concerns that using ultrasound kind of early in the acute resuscitation period could be harmful. So this was a Mojira et al. article in Critical Care Explorations in 2019. And they essentially looked at emergency department patients that came in in shock and then they either went to the ICU or died. And they found that those patients that underwent ultrasound prior to important interventions like fluids, like vasopressors actually did worse. They had higher mortality. And the concern was that we are actually delaying important interventions because we are doing these ultrasounds. We have some evidence that it does nothing. So this is the SHOCK-DD trial. It was a randomized multicenter RCT published in Annals in 2018 that essentially randomized emergency department patients that presented in shock with over half of them being septic. They essentially got a version of a cardiac exam but also just looking at multiple different ultrasounds to try to identify cause. So they included cardiac and IVC ultrasound here. And they found no difference. So the patients that got all this ultrasound and really got looked at up front, there was no difference in mortality, in fluids, in ICU length of stay, in like vasopressor use, there was truly no difference. But clearly I believe that there is something behind this. And so not to be annihilist, there is some evidence of benefit too. So this was a single-center RCT, basically in the pediatric literature. So they did, in 2018 they took pediatric patients admitted to the PICU with septic shock and they randomized them to either getting serial echoes, so bedside cardiac ultrasound to assess kind of how their resuscitation was going versus like your usual standard of care. And they found that they actually had faster time to shock reversal. They had less fluids, more vasopressor use. And there was a trend towards better mortality outcomes. And so with all this kind of like mixed literature and the trend towards getting a little bit more restrictive with our fluids, the hypothesis that we came up with is that we thought that our emergency department patients that presented with septic shock, if they got a cardiac ultrasound within the first three hours, so in that three-hour bundle window, we thought they would actually get less IV fluids. The way our study was set up, so this was a single-center retrospective cohort study. We were at one academic tertiary care center and we used our quality assurance sepsis database to essentially pull out these patients. Our primary exposure was the use of focused cardiac ultrasound within the first three hours. And importantly here, this is something that's performed by the emergency physician most often is the one who is directly involved in their care. And our primary outcome was assessing for compliance with the 30 cc per kilo SEP1 fluid bolus measure. And we utilized a couple of different statistical things including multivariable logistic regression analysis. To take a look at how we got to our actual patient population, so we looked at all emergency department patient encounters from 2018 to 2021. So we included COVID. We made no effort to try to exclude those patients. And they gave us almost 300,000. Taking those patients, again, we used like our sepsis quality dashboard that uses like the JAMA redefinition and other kind of EMR-based things like fever, tachycardia, blood cultures, antibiotic administration to essentially get us a surveillance and diagnostic population that is considered to have sepsis. And so with that we had about 4,200 patients. Where we thought it would be the most interesting and the most important is in those with septic shock. So utilizing ICD-10 codes we pulled out about 1,800 patients that had septic shock. And from there we kind of pulled out those that were either transfers from another center or they had shock onset after they were already admitted. This is kind of something that, you know, we're a big referral center. We have problems with boarding from time to time. And in our minds those patients have like already started resuscitation and are very different from those that like hit your door in fluorid septic shock. And so that left us with about 1,000. Of those, around 300 actually got cardiac ultrasound within the first three hours leaving us with slightly over 700 that got it outside of that three-hour window. And again we picked that timeframe because it would make sense for that to influence you know the use of fluid in that period. If you got an ultrasound seven hours down the road I don't think it really did anything different about what you got acutely. And so because these groups were not matched there are some important differences that I want to point out to you guys. We see that our Focused Cardiac Ultrasound Group looks like it's older. And this was statistically significant. For comorbidities like CHF and CKD this was something based off of ICD-10 codes, so diagnoses that these patients already had. And there was a lot more patients that had CHF in the cardiac ultrasound group. And like this kind of makes sense, right. They come in, they're sick. We know their heart doesn't work super well so we want to take a peek and see what's going on before we start fluid loading them. This doesn't mean statistical significance. There may be a sign that the cardiac ultrasound group was sicker based off of not quite having a higher lactate or systolic shock index, but maybe a little bit of a signal being there. Importantly, and this is something to keep in mind, their sources of sepsis were not the same. And so you can see that the cardiac ultrasound group had a much higher rate of pulmonary sources of their sepsis in comparison to those that did not have cardiac ultrasound in the first three hours. And so looking at our primary outcome, again this is reaching that 30 cc per kilo fluid bolus metric within the first three hours. It was actually the exact opposite of what we thought. So we thought the cardiac ultrasound group, whether the provider saw something or had some concerns, they'd be less likely to give fluid. And it was the opposite. Like 44% of these patients actually met that fluid metric in comparison to those that did not get any cardiac ultrasound at 34%. And this difference was statistically significant. We did assess a few kind of standard secondary outcomes including mortality, vasopressor use and mechanical ventilation all within kind of the Day 1 period. And I'll include briefly that the actual fluid received was statistically significantly different. But I think most of us would agree that 250 mLs is probably not clinically different. And I think this more so reflects the weights of the groups. So again, looking at our secondary outcomes, the use of vasopressors was significantly higher in our patients that received the cardiac ultrasound. So about 70% of those patients actually required vasopressors within the first 24 hours of their care versus about 57% in those that did not receive ultrasound. Importantly, mechanical ventilation rates on Day 1 were not different. And so this is even more interesting to me because the cardiac ultrasound group had more pulmonary sources of their sepsis and got more fluid, but it didn't seem like they needed mechanical ventilation at higher rates. And there was no statistically significant difference of 28-day mortality when we're comparing the two. And so looking at those things it's like, okay, your groups weren't matched. Is this just because the cardiac ultrasound group was sicker? And so what we ended up doing is a multivariable analysis essentially trying to tease that out with the comorbidities and whether or not these patients got ultrasound. And so this is all odds ratios looking at getting that 30 cc per kilobolus. And the bottom part makes sense, right. If your patient has a really high lactate, it makes sense that they are more likely to get more fluid with an odds ratio of 1.7. If their systolic shock index is incredibly high, so their heart rate is like 150 and their blood pressure is 60 over 20, it makes sense that this has a very high odds ratio of actually meeting that fluid bolus recommendation. But this is all intrinsic to the patient, right. You can't really influence that. This is how sick they are whenever they come to you. But even with all these other things being held equal, right, their initial illness severity, whether or not they have CHF, CKD, things that make us worry about fluid tolerance, the odds ratio of reaching this metric was still significant at 1.7 if these patients got some sort of ultrasound within the first three hours, more specifically the cardiac ultrasound. And so in conclusion, we see that focused cardiac ultrasound is associated with improved compliance with our 30 cc per kilo sepal and fluid bolus suggestion. We see that our cardiac ultrasound patients are more likely to receive vasopressors in addition. But there does not appear to be an impact on 28-day mortality. There are some limitations to this study. This was all retrospective. This is a single center. This is just a reflection of our patient population and what we see and how we manage them. The skill set with ultrasound is pretty variable. I'm lucky and where I train at right now has a robust ultrasound curriculum for the residents. But even within that there's still variation as is like the attending comfort. So there's no standardized way that everyone is taught outside of a course for the first two years. And again, there is no standardized protocol. So I think most of us are comfortable with like the standard views of bedside echo. But it's not like there was a, oh, IVC is collapsible, give this. Or your EF is hyperdynamic, they need this. And so some of the next steps with this data, I'm sure as many of you guys are ultrasound aficionados, your first question is like, well, it's not just putting the probe on the chest, right? Like what did it show you? And that's what I'm working on. So give me a little bit of time. But essentially what we're going to look at is like ultrasound findings of fluid tolerance, finding that and seeing the interplay with how much volume was administered. And especially looking at this within our patients with CKD and CHF, those patients that we are the most worried about harming with aggressive fluid resuscitation. Thank you so much.

cardiac ultrasound

septic shock

fluid administration

fluid bolus

28-day mortality