Loading summary
Transcript29 lines
- [00:00]
A
Welcome back to Run the List, a medical education podcast in internal medicine. As a quick disclaimer, this podcast is made for educational and informational purposes only and should not be understood as medical advice under any circumstances. Before we get to the show, a quick word on the sponsors for today's episode.
- [00:16]
B
This episode is brought to you by Open Evidence. Open Evidence is a really incredible resource for people in medicine. It's an AI powered medical information platform that can help you answer clinical questions, provide high quality literature and and so much more. You can ask questions like what are the classic imaging findings for gout? Or tell me about the landmark trials in lupus nephritis. They recently partnered with the New England Journal of Medicine so they have access to their text, figures and tables within Open Evidence. It's free and unlimited for healthcare professionals, so I highly encourage you to go check it out. Go to openevidence.com to learn more. Today we'll be talking about septic shock, a very important topic. We have with us Dr. Vikramjit Mukherjee. He's the chief of Critical care at Bellevue Hospital and also the medical Director of the Special Pathogens program at Bellevue as well. Dr. Mukherjee, thank you so much for joining us here today.
- [01:18]
A
Thank you. Thank you for having me.
- [01:19]
B
Let's get into our case. Today we have a 55 year old gentleman who has a history of type 1 diabetes and AML, who's recently been on chemo and he's coming in with fevers, chills and dysuria. He's febrile to 102. He's tachycardic and his blood pressure is 85 over 50. He's answering your questions, but he's pretty pale and somnolent. And on review of his labs, you see that his white count is 20. He has neutrophilic predominance and his UA shows leucastrase and nitrites. They also did a VBG which shows a lactate of 3 and his blood culture and urine culture are pending at this time. So, Dr. Mukherjee, this is not sounding great for our patient. What are you thinking at this point when you see him?
- [02:02]
A
Thanks, Emily. You know, as many of us can recognize by now, this patient is screaming of someone who has an infection and a very severe infection at that. He has a lactate, which we'll talk about. He's hypotensive, he's febrile, he's at high risk because of the recent chemo. So this is something that we see in the ICUs all the time. It Seems like someone's heading towards or already is in severe sepsis, septic shock. And I just want to bring out that this is bread and butter medical critical care. In the US there are 1.7 million cases or recognized cases of sepsis and septic shock every year, leading to around 400,000 deaths. So two points on that. You know, one is this is something that you will see at the bedside, in the er, in the hospital wards, in the ICU all the time. And what we hope to cover in the next few minutes is how we can best improve this patient's outcomes. But also that because it is such a common disease, even the smallest increment in therapies and countermeasures that can improve outcomes has huge population benefits. So I'm glad we're talking about this. But at this point we are looking at a patient who is likely in septic shock and needs some urgent attention.
- [03:07]
B
You mentioned septic shock a couple of times. I know terminology can get kind of confusing for people, both patients, their families, but also even people in the medical field. Can you kind of define exactly what septic shock means?
- [03:19]
A
Absolutely. Great question, Emily. So let's start from some of these acronyms that we all love and are horribly confusing. So we'll start with sirs. Sirs, which is an acronym for Systemic Inflammatory Response Syndrome. Sirs. Basically you have to have two or more of the following four. Fever or hypothermia. Remember that many of your older patients will not spike a fever. They might come in with low temperatures, paradoxically, tachycardia, tachypnea, or an elevated white cell count with or without bendemia. When you see two or more of these four things, you should be thinking of something that's a little bit alright. And yes, it could be infection. But also keep in the back of your mind that there are some non infectious causes of sirs. Stuff like pancreatitis can cause sirs. Stuff like burns can cause sirs. So this is a blanket done that just basically tells you that someone's sick. When you see this, you should be looking for a source of infection. And we'll come to the common sources of infection in a second. But if you find a source of infection along with two or more of the SIRS criteria, you call it sepsis. If you have hypotension with that, you call it severe sepsis. And if you have hypertension that's not recovering with fluids, you have a patient in septic shock. And the reason it's important to bring up these different categories, if you may. Is that higher? In this sequence, a patient with septic shock is sicker and needs more attention than a patient with severe sepsis, who's more sick and needs more attention than a patient with just sepsis. So it helps you triage and risk stratify where this patient should land up in your hospital.
- [04:51]
B
Perfect. Thank you for taking us through that. So in our patient, they do appear to be in septic shock, given the persistent hypotension. We haven't talked about whether he's gotten fluids, but we'll assume that he has and lactate elevation in his case. We know that the UA is positive. The source is pretty clear. But in general, what are some of the more common causes or sources of septic shock?
- [05:14]
A
Good question, Emily. So, you know, at the bedside you should be looking for, the top four causes are the two L's and the two B's. So, L for lungs, pneumonia, L for lines. Person comes into your hospital from a nursing home with a PICC line or a central line or a dialysis catheter. Those things can get infected very commonly. So look for the lines, make sure you have a good physical examination, and then the two Bs. B for bowel, infections in the gut, ischemia or infections, C diff being a very big common one, and then bladder, especially in this case, look for UTIs and so on. So if you look for the two Ls and the two Bs lungs, lyins, bladder, and bowel, you'll catch around 80% of your causes of centric shock. If those don't give you a sign, look for something that's a bit more rare. You could have meningitis, uncommon, but something to look for. You could have an abscess hiding somewhere, an empyema around your lungs. But again, those are not as common. A first good step would be to look for the lungs, lines, bladder, and bowel. And that's a good starting point to look where the infection is.
- [06:18]
B
Is it possible to have a negative blood culture and still have septic shock?
- [06:22]
A
Absolutely. Very commonly, around 25 to 33% of your patients will have culture negative septic shock. And you will just never find a source of infection or blood cultures that will give you a bug and that lends itself to pneumonias, to bacteremias, and so on. Many reasons why, you know, while we trust our blood culture platforms a lot, they're not the most sensitive. And then again, some patients might have gotten antibiotics before they went into septic shock, reducing the yield. So, yes, just because you don't have culture positive sputum or culture positive blood doesn't mean the patient doesn't have sepsis. Pretty common.
- [06:56]
B
Okay, so in this patient, Dr. Mukherjee, what are you worried about?
- [06:59]
A
Great question, Ali. So, you know, this is again, bread and butter, critical care. But there are some red flags which tell us that if you don't pay attention to this patient quickly, things can go south pretty quick. So just going from the history, we know that this gentleman has AML with recent chemo, so his immune status itself is compromised. So a, he's at high risk for sepsis, but also his route to recovery is going to be longer. He is tachycardic and he is hypotensive. So I just want to pass on a comment of the hypotension. It takes a lot to drop your blood pressures. Your inherent protective mechanisms make you normal tensive for a fair amount before you become hypertensive. So just recognizing that he is already at an 85 over 50 says that you have a little bit of catching up to do because he's present a little late in the game. That said, just a quick marker on blood pressure. Blood pressures which reflect macrohemodynamics, don't always reflect what's going on in the tissue level. So take your blood pressure with a grain of salt. If you have a patient who's mapping 55, 60, but is warm, well perfused, maintaining well, being well, has a normal lactate, take that blood pressure with a bit of comfort rather than a blood pressure with a map of 60 and a patient who's obtunded and cold, clammy, oliguric, hyperlectaemic. So don't take your blood pressure just in isolation, but take it in the clinical context of everything else that's going on. And honestly, at the bedside, when you see a patient who's mapping at 55 or 60 but is otherwise looking good, you know, talking to you warm, peeing out a lot without diuresis, you have some time to figure it out. The last thing I would mention is the lactate of 3 on the Venus is worrisome. Anytime you see hyperlectemia, be very worried. Your lactate takes a long while to start peaking. So just like the blood pressure, you have a lot of reserve, which protects you before the lactate starts treatment. So when you see a lactate of 3a, it's a marker of worsening 28 immortality, but also that this patient is very sick and needs urgent attention.
- [09:03]
B
Thank you. So, just to kind of summarize Things that you're looking for, both on exam and on labs, are evidence of lack of end organ perfusion. So. So that can manifest as cool extremities, it can manifest as altered mental status, decreased urine output and then also a lactate elevation. You're also looking for Aki, elevated LFTs, those sorts of things. And of course, as we saw in this patient, typically you're going to see a leukocytosis with a left shift, though not always, especially in some of our immunocompromised patients. How are we thinking about treating this patient?
- [09:36]
A
So I think we've established the point by now that this person needs attention. And over the last 20 years, there's been rapid advances on how to treat patients presenting like this, which has completely changed the paradigm of treating septic shock. So just a bit of a historical perspective. Back in the 1990s, even early 2000s, patients with septic shock used to be relegated to a corner of the ed and someone used to come and check in on them once in a while. And the mortality, even in the late 90s, early 2000s, was close to 45, 50%. A lot of advances and none of this is new stuff. It's basically sticking to the basics. And it's been shown that just sticking to the basics, and we'll come to the basics in a bit, improves your mortality significantly. To where? At Bellevue, Our septic shock, severe sepsis mortality is down to 15, 16%. And for a disease that hits 1.7 million cases in the US every year, that change in mortality saves hundreds of thousands of lives if done in a. In a protocolized way. So in the first three hours, it's pretty standard of care now to do the following things. Give fluids. The Basic minimum is 30cc of crystalloids, but you have to give these patients fluids. They are very fluid, responsive for a multitude of reasons. They have beenodilation, they have capillary leak, so they're usually dry. So please be pretty aggressive with your fluids. The initial bolus again should be 30 CCK of crystalloids. Culture them quickly, but also early. Appropriate antibiotics is key. We'll go into that a little bit because it's important to notice the two adjectives. Early every hour delay in antibiotics leads to a mortality increase by around 8%. Just imagine six hours delay, you have increased mortality by 50%. Those hours pass by quickly and appropriate being the other defining feature. Your choice of antibiotics will differ whether this patient is coming in from a nursing home and has exposure to MDROs. In the past or the patient's coming in from the community and might just have a regular community acquired pneumonia. So you have to be a little bit thoughtful about your antibiotics, but early appropriate antibiotics will save a lot of lives. And then the only other thing is check a lactate. It'll give you a bit of an idea of how sick this patient is and how well your therapies are working. But in the first three hours, fluids, cultures, antibiotics and a quid lactate check will drop your mortality significantly.
- [11:53]
B
I just want to make a quick note and give us kind of an example of what 30 cc's per kilogram actually might look like in a patient. If we have a patient who's 100kg, that's 30 times 103 liters in the first three hours, which you know, might be more than a lot of us are used to giving on the, of course, on the regular medicine floor and maybe more than we might feel comfortable giving in. Some patients who have typical contraindications to receiving a lot of fluids, for example, patients who have CHF or end stage renal disease, do you still have the same recommendations for fluid administration you do?
- [12:26]
A
And great question, Emily, because that's where we see the fallouts happening often. But absolutely so in patients who have end stage renal disease and patients who have non decompensated heart failure. So they're not flashing right now, but they have an EF of 10%, you still give the 30ccs and that still improves your 28 day mortality. So just because a patient has an AF of 5%, 10%, don't get shy about giving the 30ccs. It does save lives. Of course, if a patient is in flash pulmonary edema or EF5% in your bi basilar crackles and the patient's struggling to breathe because of left ventricular failure, that's a different paradigm. But in stable heart failure patients, Even if the EF is super low, 30cc stabilize.
- [13:04]
B
What if our patient's not responding to fluids? What's our next step?
- [13:08]
A
So many patients will respond and hopefully they'll go on with their merry ways. But a significant fraction will then end up in your ICU with fluid refractory hypertension and that becomes a bit more difficult to manage. But things that will help there is that once they are up in your ICU or in your er still you should reassess whether this patient needs more fluids. You know, 30ccs is a great starting point, but some patients might be at home for a week with sepsis and present to you late in the game and they haven't had good PO intake for that week. They have been having a lot of insensible losses, so they might need more than 30ccs. Your job is to see if that is the case and see if you need to be on fluids more. If they are hypotensive and they're not fluid responsive, think about pressors early to get their perfusion going. We want to recheck a lactate. If the lactate was 5 in the initial three hours, but it's still 5 after your fluid bolus, then you're in a bit of trouble. Something else might be brewing. And then the last thing is, if you have a source that needs to be controlled, you have a big empyema. You've got to control that with the chest tube. If you have an abscess in the belly, having a surgeon or IR folks train that, because if you don't have source control, antibiotics won't be the win. You have to get rid of the pus collection, the fluid collection somehow.
- [14:20]
B
On the topic of pressors, this can be something that is very intimidating, especially for people starting out in residency. Can you kind of give us a little bit of a primer on pressors and septic shock? Where do we start? Where do you go if someone's not responding to low dose pressors? And then how do you think about peripheral versus central pressors?
- [14:39]
A
Yeah, absolutely. Thank you. So, you know, we'll keep it simple. And I think almost in all cases, levofed or norepinephrine is gonna be your go to pressor of choice. There are some places where you'll reach for nucinephrine or vaso or dobutamine, but in most other cases, levofed as a first choice will be the right choice. Sometimes you'll see that, you know, you don't have time to wait for the fluid bolus to act as you administer pressors. If the patient is deteriorating in front of you, you might wanna be reaching for pressors along with giving the fluid boluses so early pressors are key. You might not have a central line that sometimes takes time to administer pressors. So you might have to start pressors peripherally as you get a central line in. There is pretty good safety data for using peripheral pressors, but a lot of this is contingent on how well equipped your ICU is. The danger behind peripheral pressors is that it's a pretty irritant drug and extravasation into soft tissue can cause a lot of skin necrosis. So you want to avoid that, but if you have a good, and by good I mean 16 gauge, 18 gauge peripheral IV that's placed anti cube and above and your nurses can do Q1 hour checks on those pressors. You can go up to 2530 of Levo in that setting. But again, those are important qualifiers if you don't have the nursing strength to do Q1 hour checks on your IV. If you don't have a good IV, you shouldn't be using pressors through a peripheral Binky 22 gauge. It has to be a solid proximal IV. But in that setting there's a pretty good safety data behind using peripheral pressors.
- [16:12]
B
I remember taking care of some patients in the ICU who despite being on pressors were not improving, were persistently hypotensive in that scenario and in others we'll talk about stress dose steroids. And I remember being a little bit confused in patients who had septic shock because traditionally we're taught that steroids and infection are not a great match. How do you kind of think about when to use stress dose steroids?
- [16:34]
A
Great question, Emily. So there's been a, over the last 20 years again, and I'm dating myself here, there's been a slew of data looking at steroids for septic shock. And just summarizing those trials is that if you have a patient on two pressors, so you have Levo and Vaso or Levoine, you should be thinking of using corticosteroids. The safety data is pretty good. They usually improve outcomes. The latest data shows that you should be using a combination of hydrocortisone and flutrocortisone to improve your mortality. It's pretty clear that it helps you come off pressors quickly. It may or may not improve your overall ICU outcome, but there is very little downside to it and maybe a signal of benefit by improving time in shock coming off. But do use hydrocort plus strutracort together. Most of the positive trials that have shown improvement in outcomes use the combination.
- [17:25]
B
Of hydrocort, fluorochrote and the rationale there is to try and mimic the body's natural physiologic response that they're not able to mount while they're sick. Is that right?
- [17:35]
A
Absolutely.
- [17:35]
B
Great. So back to our case. We have our 55 year old gentleman started on pressors and over the next 24 hours is starting to do much, much better. He's on broad spectrum antibiotics. His blood pressure finally starts to stabilize. We're weaning our pressers, We've controlled the source and we are narrowing our antibiotics. He's transferred to the floor and he's thanking you for your care. Dr. Mukherjee, before we wrap up this episode, do you have any final pearls or take home points for our listeners?
- [18:03]
A
Well, glad this patient did well. Emily, thanks for that wrap. I guess just a couple of things, you know, One is most of the stuff that we spoke about doesn't need an expensive medication and doesn't need a lot of thinking out of the box. But if you're able to do what we spoke about in a systematic, protocolized way, you'll see that so many lives can be saved by this. The one vulnerability of our US Healthcare systems is that in a patient presenting with septic shock to the ed, it's pretty easy. The patient's coming to you with a fever and a white cone and so on and saying that he or she doesn't feel well. Time zero is pretty easy to identify and the clock starts ticking, you can do the right thing. It's not as easy in the inpatient wards to do that because patients get seen maybe two or three times a day and between being seen they could easily develop sepsis. And time zero to first intervention has a longer lag. The data for ED sepsis mortality is 19%. The data for hospital acquired sepsis is mortality of 31 35%. So we have a lot of catching up to do on the inpatient side of things and in the ICUs. But for our learners who work in hospital med surge wards in the ICUs keep your index of suspicion up for badness that comes your way and have a load threshold for escalating to your residents, fellows, attendings when you suspect something's going south. But that's it. Thank you for having me. Hope this was useful.
- [19:21]
B
This was an incredible episode. Thank you so much for your expertise and hope to have you again soon.
- [19:38]
A
Some content from this episode was generated with the assistance of artificial intelligence.