All right. So no departmental update today so we'll dive right into grand rounds. And we're very excited to have Dr. Louise Man with us today from the Department of Hematology and Oncology. Dr. Man is relatively new to the division, having joined faculty in 2017. 

She does have some roots here at UVA, going to undergrad here and then did her residency training at Eastern Virginia Medical School and then came over here for fellowship before we were fortunate to keep her on for faculty. And early in her career, she's really become a popular teacher amongst residents and fellows alike. She's been active in all the improvement within the division and has been an active clinical researcher as well, serving as an investigator for several trials here at UVA. 

And she has really developed an expertise in non-malignant hematologic abnormalities, specifically coagulopathies and other disorders of the like. Today she'll give us insight into that kind of complex field. I think it's something that comes up not only on the hem/onc wards but for all of us. We'll go through that in a case-based interactive manner. So with that background, I'll turn it over to Dr. Man. 

Thanks, Anthony. Can you guys hear me? Yes? OK. Thank you. Thanks, Anthony, for the introduction. So today, I'm going to talk about acquired inpatient disorders of coagulation. I would like to say that I have no relevant disclosures, but I have a seven-month-old. So when I say something, it's not always what I mean to say. So just bear with me, and I really mean for this to be interactive. So if someone has a question, it's totally fine to raise your hand and stop me just to clarify anything. 

So at the conclusion of the talk I would like for everyone here to understand the pathogenesis, the features of diagnostic testing for HIT, and management of HIT. And then the second part of the talk, I'm going to talk about causes, presentation, diagnostic testing, and management of DIC. My bias with preparing a talk like this-- I really felt like I should spend a little bit more time with HIT. The American Society of Hematology-- ASH-- published new guidelines in the fall of 2018, so there's just more updates with that. And just my bias, also. I feel like we can do a little bit more with HIT, and DIC we can talk about. But HIT I do want to spend most of the time on. 

So this is actually a case of a patient who was just seen by our hem consult service within the last month-- a lot of similarities to patients I'm sure you guys have seen in the hospital. And it's just sort of the foundation of the talk for HIT, as well as sort of considerations for DIC. 

So a 65-year-old male with hypertension, stage III chronic kidney disease, coronary artery disease, severe aortic stenosis is admitted for planned CABG and aortic valve replacement. Blood counts were normal on admission, but his post-operative course, unfortunately, is complicated by cardiogenic shock. He requires pressors for one day. He does develop afib with our RVR, even though he has no history of afib. On post-op day one, he is started on a heparin drip for that. And then he develops oliguric AKI on top of his stage III CKD, and he is requiring CRRT. And you're called because you're concerned about his platelets. 

So I'll just give you guys a moment, just to kind of briefly peruse the blood work on this gentleman. Of note, his counts are normal on admission. Actually, I lie. Hemoglobin 12.9-- technically below normal for male. Platelets 170, creatinine 1.7 on admission, INRS-- you can see a little bit of a creep up there. Fibrinogen is normal. LFTs, I'm telling you are normal. Now the question is, if you see this gentleman on post-op day two, what goes through your mind about the cause of low platelets? And is it any different than if you're called to see this guy on post-op day seven and you're concerned about his platelets? Are your differentials for post-op day two and post-op day seven different? 

And of course, everyone knows the title of the talk and my objective, so I do have to tell you he receives heparin intraoperatively, as well as on post op day one when he was on the heparin drip. And a question for this particular subject, since we're talking about thrombocytopenia, and we're going to focus on HIT, is what is his 4T score on post-op day two, versus what is the 4T score if you see him on post-op day seven? All right, I didn't realize my pointer was on. 

So is this HIT? Is this DIC? Is this when hem consult service says, oh, it's multifactorial? It's a little bit of everything. Just brush it off. What is this? So that's a case, just keep in mind that this was a gentleman who we saw. Having a face and the name to the patient lens helps kind of humanize, try to figure this out. And this gentlemen's course became very complicated. 

So what is HIT? Well, we all think of it as a heparin exposure, to start things off. Although I will tell you in a little minute that you don't have to have heparin exposure to have HIT. But typically heparin exposure, and you develop these multimolecular complexes that form between positively-charged platelet factor 4, expressed on the surfaces of platelets, and negatively-charged heparin. 

And then antibodies here form against these multimolecular complexes. And because these antibodies bind to these complexes, they bind on the Fc receptors of platelets, leading to the activation of the platelets. What is also known is that PF4 is also expressed on endothelial cells, so they also become activated. So this cascade leads to a highly prothrombotic state from HIT. 

I feel like when I was in medical school, everyone talked about type 1, type 2 HIT. When we talk about HIT now, we're really talking about type 2 HIT. Type 1 HIT was just the old nomenclature is heparin-associative thrombocytopenia, which is a different mechanism from this, but we're talking about type 2 HIT. You'll see HITT, which means with thrombosis, and then S just means for the syndrome. So HITT, HITTS, all the same, except for the type 1 HIT that I was mentioning briefly. 

Who gets HIT? So this is from the CHEST guidelines in 2012. Patients who are in a post-operative status in the hospital, they're at the highest risk for development of HIT. For patients getting heparin, either prophylactic or therapeutic doses, the incidence is 1% to 5%. And for patients who have had cardiac surgery, their incidence is 1% to 3%. For the rest of the people who are listed on this table, they can develop HIT. They're just at much lower risk. 

So as I just pointed out, the incidence does vary, depending on your patient population in the hospital. Thrombosis is seen in about 40% to 50% of patients. And let me just go back one slide. Low molecular weight heparin, there's a lower risk of development of HIT with low molecular weight heparin than with unfractionated heparin. And I don't think that this data on the incidence of HIT has really changed over the course of decades. This is sort of what we're seeing. 

Well, what does HIT look like? Well, HIT looks like having a positive platelet-activating antibodies and clinical symptoms. When people talk about HIT, people think that you have to have low platelets to have HIT. It's not true. You don't need to have low platelets. You just need a platelet decrease from your baseline, typically of greater than 50% and, as I discussed, plus or minus the presence of clot. And again, when people have a decreasing platelet count, it generally occurs 5 to 14 days after initiation of heparin. 

That's sort of the typical case of HIT that you see. One other type of case that we can see commonly is that someone has received heparin within the last 30 days. They go home from the hospital, they come back in for some other-- they get readmitted for whatever reason, they have a repeat exposure within 30 days, and then they have a precipitous rapid drop in their platelet count. It's rare, but you can develop delayed-onset HIT, where your platelets start to drop about two to three weeks after heparin exposure. And as I alluded to, there is such a phenomenon of spontaneous HIT, where if you develop this antibody, with or without the heparin exposure, you can develop autoimmune HIT, the same pathogenesis of platelet activation. 

So we talked about clots in these cases. About half of the patients who have HIT have thrombosis. Venous thrombosis is more frequently seen than arterial clots. When we see lines and clots in the setting of HIT, we think, oh, this is line-associative. Maybe it's not really due to HIT. It actually is. Sites of vessel injuries are actually where people are at higher risk of developing these clots. People can get skin necrosis from if they just got heparin subQ, they can get skin necrosis there, and they can also develop a systemic anaphylactoid reaction with an infusion of unfractionated heparin if they do have HIT. These are some of the other types of clots we can see with HIT. 

So this is something we sort of forget about and then come back to every single time someone calls us about thrombocytopenia. So the pre-test probability for looking at HIT is established with a 4T score that was developed in the mid 2000s. And we're all familiar with looking at the degree of your drop in platelets, the timing of your drop in platelets relative to your heparin exposure, thrombosis, or other causes for thrombocytopenia. I don't need to read this to you because you guys know it, and you guys can easily look this up. 

In real life, sometimes it's just not clear if someone is a 2, or 1, or a 0, and so there is a lot of interobserver variability with this 4T score. I mean, even on hem consult team, if we look at a case, we're thinking, oh, when do we use as a baseline? When do we use the time comparison? If it's not either the 2 or the 0, it's going to be a 1. And I'm trying to think of if there's any other common scenarios where we struggle with doing this. If someone had a new clot that was explained, and then they developed thrombocytopenia, do you count that clot? There's just some limitations of this test. 

And again, the pre-test probability breaks down the low, intermediate, high. Again, a high negative predictive value, so if you have a low 4T score, it's not likely to be HIT. But again, if it's a high probability, it still sometimes could be explained by other reasons. You have thrombocytopenia from being sick in intensive care, and so this has lower utility in an intensive care setting. So there are some limits with this 4T score, but from the mid 2000s until recently, that's the best thing that we had. 

So can we improve on a 4T score? So back in 2010, Adam Cuker and a collaborative group with multiple hospitals, they developed something called the HIT Expert Probability score, HEP score. And what they did was they took 50 consecutive patients who they were concerned had HIT, and they calculated both a HEP and a 4T score on these patients, using 26 experts. And they would pull three experts from that pool of 26, and then they would have everyone calculate the HEP and the 4T score. And then all patients who were looked at had both antibody and SRA testing. 

So if you thought the 4T score was a little cumbersome, this is going to be a little bit more. And specifically, I had this pulled up, and I know it doesn't project to the back too well. It's because it's a very, very involved scoring system. These are not new concepts. How much does the platelet count fall? What's the timing? What's the nature of the platelet count? The presence or absence of clots, skin necrosis, reactions. But what is interesting is that this scoring system has mitigating, subtracting of points, where you can take down the score if you know that the person has severe infections, severe DIC, they have other causes. You have mitigating scores that you can contextualize the thrombocytopenia. 

So what they did was they looked at the interobserver agreement for this HEP score. They looked at whether or not it saved them patients' exposure to direct thrombin inhibitors. What they found was that the HEP score had greater interobserver agreement, so from one expert to another, they were more likely to agree on the score. There was a little bit more concordance with HIT laboratory testing, 100% sensitive and 60% specific for determining the presence of HIT. 

And the biggest thing is that DTIs are very expensive. And when the patient is on heparin, and you have to switch them to a DTI, everyone has to change what they're doing, how they're managing the patient. If this in the setting of cardiac surgery, it gets very complicated to manage a DTI. There's volume issues with bivalirudin, but 41% reduction in the number of patients who were receiving DTIs, so it's cheaper if they can stay on heparin and don't need to come off a DTI. So this has potential implications for large health care systems and management of health care resources. 

Now unfortunately, even though this score looks great, it has not become quite widely used because it's been more cumbersome. So this is out there in benign hem world, but really not implemented in health care systems, large settings. So now if you're faced with a scenario where think the person has HIT, what do you do? Well, if you have a low probability pre-test score, you can continue with heparin, but we stress making sure that you reassess them in 24 to 48 hours. If you still have a nagging thought in your head, nah, maybe this is HIT. Maybe they can get HIT, even if we don't think it is now, then you ought to reassess them in 24 to 48 hours. 

If the person has either intermediate or high probability, this is empiric treatment. You need to stop heparin. You do need to add heparin to an allergy list, so they don't get additional heparin. It's important to note this. Two hours after the last heparin exposure is when this HIT antibody test should be drawn. You should not draw it within two hours of them getting it. It's a higher likelihood that it'll be a false positive. You should obtain ultrasounds of the lower and upper extremities, as people can have silent DVTs, so without swelling, without symptoms. 

This is new. The ASH guidelines in 2018 suggested that if someone has a line, you should do a Doppler of that upper extremity. But if they don't have a line and if they don't have symptoms, they suggest against doing Dopplers, but lowers do need to be evaluated, for sure. And then the recommendation is still, with CHEST, with ASH, you must start therapeutic anticoagulation. Don't do prophy. It must be therapeutic with an alternative anticoagulant, even if they don't have a clot because there is a high risk of thrombosis that people can get if they do have HIT, even after heparin is stopped. 

So we're all familiar with the HIT antibody test. The sensitivity is 100%. But again, you can have a positive antibody test and then negative confirmatory SRA testing. I think most people are aware that about three months ago-- oh, I don't what happened. Sorry, the battery on the mic went out. Can you guys hear me? OK. Should I stop talking? 

Yeah, we can just cut it off. 

Everyone can go home. Everyone can go home. 

So about three months ago, the antibody test went live at UVA in-house. So Monday to Friday and actually some weekends, we've been surprised, they do run a HIT antibody testing, and you will find out within same business day. If you run it late in the afternoon, then it won't be back until the next morning. But the HIT antibody test is now active in-house. 

If the HIT antibody is positive, there is a reflex send out test that is still done, and that's the SRA, the functional assay that shows the HIT antibodies activating the platelets after they're washed. And again, not to confuse anybody, that HIT reflex is a reflex test for the SRA. Does anyone have questions about the testing? Yes. 

Are there scenarios like that middling and that prequel range that you would [INAUDIBLE] maybe send to antibody testing but not start? 

Oh, that's such a good real life question. So the question is, is there a middle of the road option if you are somewhere between a 3 or a 4, you're not really sure. So the 4T score has its faults. We ran into that this week on hem consult service. Looking at 4T score has limitations. If you are intermediate, you should stop heparin and do therapeutic anticoagulation. 

The benefit of having this in-house test is that exposure to therapeutic anticoagulation for patients who are thrombocytopenic who don't have HIT is that you would be able to stop them off anticoagulation soon. And that's, I think, one of the benefits of the in-house antibody test. I think that's one place where we can prevent morbidity and mortality, either direction, bleeding or clotting issues. It's a good question. 

So when we talk about alternative anticoagulants in the acute setting, so in other words, someone has confirmed HIT, their platelets are still low, your options are argatroban, which is FDA approved, direct thrombin inhibitor, preferable for patients who have renal dysfunction but have normal liver function. Bivalirudin is technically off-label, so note that. That's off-label DTI for HIT. It's preferred for patients who both have hepatic and renal dysfunction. 

It's preferred for patients who need urgent cardiac surgery because the half-life is about 25 minutes, whereas the half-life of argatroban is 45 minutes, and neither are reversible. So your preference for patients who might need urgent surgeries is to use bivalirudin. 25 minutes half-life, assuming no renal dysfunction. Fondaparinux, which is an indirect 10A inhibitor, is off-label, but it has been used. And so in guidelines, it's still suggested that you can use fondaparinux. These are other anticoagulants that are not available to us, so I'm not going to discuss them. 

So one of the challenges with patients who have HIT is that a lot of times they're critically ill. They might have other contributing factors to low platelets, and what do you do if the platelets have not improved yet? What can you do with their anticoagulation? So for CHEST guidelines-- and the most recent CHEST guidelines were in 2012. There's no update to it. But the CHEST guidelines recommend against starting a vitamin K antagonist, i.e. warfarin, until platelets have substantially recovered and, in their recommendations, they say, if they're greater than 150. 

This guideline is to the point where if someone gave warfarin and your platelets are still low, they actually recommend reversing warfarin. And the rationale is that acute HIT is an intensely prothrombotic state. So as we all know, your protein C, your natural anticoagulant protein, gets knocked down before your clotting factors get knocked down. So while you're on warfarin, before you're therapeutic, you really have a high risk of clotting while you start it. And so to the point where if someone started it in the setting of HIT, with still having low platelets, you actually ought to reverse warfarin. Get them off warfarin, reverse it with vitamin K, and keep them on argatroban or bivalirudin. 

The 2018 ASH guidelines recommend the same thing, pretty much. Don't start warfarin. If going to use warfarin, don't start until the platelets are normal. So for UVA's HIT guidelines, we say avoid starting heparin until platelets are greater than 100, but preferably over 120. 

So how do you manage subacute HIT? So to clarify, subacute HIT means that the patient was recently diagnosed with HIT and now has platelet recovery, so they're sort of outside of the immediate danger zone. So for ASH 2018 guidelines, which is new, they actually suggest that you can treat patients with subacute HIT with direct oral anticoagulants, and that's Pradaxa, Xarelto, or Eliquis, rather than warfarin. For a long time, we had warfarin, and that was the only recommended anticoagulant to use for HIT, but now as of late 2018, this is suggested as an alternative and anticoagulant. 

And the UVA guidelines for HIT have not been updated to reflect this recommendation, but it will be updated summer. Some housekeeping issues, when people are recommended to take DOACs, it's the same usual things you have to consider. You have to look at cost, affordability, out-of-pocket costs for the patient, whether or not they have any contraindications, like renal dysfunction, and so on. 

Another thought, if someone is transitioning a patient to warfarin, remember that argatroban significantly prolongs your INR, and bival does to a certain extent too. So don't transition them to a goal INR 2 to 3. They're going to not be adequately anticoagulated. And there's guidelines for this transition in the UVA guidelines. 

How long does someone need anticoagulation for? Per ASH 2018 guidelines, if someone doesn't have a clot you can anticoagulate them until platelet recovery. So traditional practice was people got warfarin for one month. That was traditional practice. If someone has a clot, ASH specifically does not tell physicians how long they need to be anticoagulated for, but we usually say three to six months, just by practice. 

Are platelet transfusions OK? We get this question a lot. Patients are critically ill still. Their platelets are a 30. Should we transfuse them platelets? So in 2012 with the guidelines by CHEST, there's no evidence of risk, per se, but the evidence was too limited to support their safety. So CHEST decided they couldn't really support doing platelet transfusions. 

There was a database study in 2015 looking at patients who had HIT over the course of one year, who either did get platelet transfusions or didn't get platelet transfusions. And the ones who did had a slightly increased risk of arterial clots compared to those who did not get platelets. It doesn't seem that venous clots occur more frequently for those being transfused platelets. So ASH made a statement saying that they suggest against routine platelet transfusions, but to consider them if the patient needs some, really low platelets or bleeding complications. 

This question comes up quite a bit too. Is it OK for patients to get heparin again? So if people follow the titers of HIT antibodies, Dr. Warkentin, up in Canada, has done extensive studies looking at the duration of these HIT antibodies being present. And they're actually, on a median, undetectable at 40 to 100 days, so they're actually quite transient. There's no evidence that patients who have a prior history of HIT who get exposed to heparin, there is no evidence of an anamnesic response, but they develop this reaction again. Again, this is not talking about the 30 to 100 days. We're talking years down the line, there's no evidence of this response. 

So CHEST guidelines in 2012 says that for patients who have a history of HIT, whose heparin antibodies are now negative, you can consider short-term use of heparin over non-heparin anticoagulants. In the little details of the fine print of their guidelines, I think they said something like four days, limit it to four days. But ASH specifically says, limit this to intraoperative use only. The non-heparin anticoagulants are very difficult to manage intraoperatively for surgical teams, which is why their heparin protocols are just much more established. So that's why for surgical reasons, it's easier to use heparin, and so ASH does say it's OK to consider. 

And again, this is sort of saying the same thing. But talking about if platelets have recovered, but this is still a recent event, if you have an elective surgery, you really should try to put it off. If you can't put it off, you can maybe use intraoperative heparin, but with caution. There's no protocols on using intraoperative heparin with an antiplatelet agent to offset platelet activation. It's just by the practice of the person who's doing it at that point. There's no consensus protocols on this. Does anyone have questions so far? 

Can we prevent HIT? Can we do anything about this? It's not that many patients, but this sounds like a lot of trouble for patients if they do have HIT. So can we prevent it? So in Toronto, there are multiple hospital systems that undertook a one year intervention of trying to remove heparin from regular use. So tertiary care centers in, Toronto there's a hospital wide intervention to reduce the use, 2006. So what they did was they changed all of their orders to low molecular weight heparin. They may have restricted it to certain ordering physicians, certain services, and they changed all saline flushes to heparin flushes. Sorry, saline flushes replaced heparin flushes. 

So they did this intervention for one year, and they compared the post-intervention five year data to pre-intervention three year data. And they found that the cases of suspected HIT went down by 42%. Your positive HIT antibody assay was reduced by 63%. True HIT cases decreased by 79%, and HIT with thrombosis decreased by 91%. So this was the first study that ever demonstrated successful prevention of HIT. Can't believe they only saved the hospital $260,000, but-- that's US dollars, it's not Canadian dollars. 

So when I was reading this study, one of the things I thought that was interesting was it's very hard to be prescriptive and tell someone, you must use Lovenox over heparin. We all know if someone has renal dysfunction, if someone is in a fresh post-op state, those are the patients who we worry about bleeding, so we worry about giving them a longer-acting agent like Lovenox over a short-acting one like heparin. So real considerations on saying, it's not really easy to just say, change happened to Lovenox. Tell that to the surgeons. 

So this is just a snapshot of the UVA HIT guidelines. You can access it. I don't know if anyone's noticed, but there's a new blood drop icon on people's desktops. So you can click that, and that'll take you to the anticoag website. And then you'll find a lot of guidelines on anticoagulation, but specifically, this HIT guideline will be there. And again, it's going to be updated 2019 summer. Has anyone seen the blood drop icon? No one's noticed it? OK. Well, it's easy to get to now, thanks to Mary Stack, sitting there. 

All right, so back to our case. You guys remember the story. What are the causes of thrombocytopenia in post-op day 2? What's the 4T score on this day? So I'll spare you the mental math. This person had a 4T score of 1 when calculated on post-op day 2. Who would check ultrasounds on-- I just want a show of hands. Would anyone check ultrasounds on this patient on post-op day 2, 4T score of 1? I'm going to be honest, I wouldn't. So the 4T score was 1, and that does become relevant later. 

So they did check a HIT antibody, was checked anyways, and it was negative. So here's a question. The HIT antibody was checked on post-op day 2 and negative. Does this exclude HIT? Just show of hands. Yes? Excludes HIT? No, doesn't exclude HIT? The answer is yes. Remember, low probability 4T score does exclude HIT for right now. For right now. 

The problem is you guys already know what happens. The platelets keep going down, and it's post-op day 7. And just to give you a little bit of background on this gentleman, he eventually does come out of the ICU when he comes off CRRT. He's not infected. He's off of heparin. He looks better. He's starting to sit up in his bedside chair. Now you're scratching your head, the platelets are 45? Still going down. Coags look a little bit worse, but his fibrinogen is 214, so we don't think he has DIC. 

4T score here. They did a Doppler. He has a line-associated clot. Could this be HIT? You should all get this right. So unfortunately, he does have HIT. Which of the following is the most appropriate next step? It's a lot to read. I'll give you guys some time to read. 

All right, so who would do A? All right, who would do B? C? D? All right. So clearly, stop heparin. When you're 4T score is a 6, you should order the heparin antibody, drawn two hours after last heparin exposure. Start argatroban. Would not do warfarin yet because platelets are 45, so it hasn't recovered. So this is the acute HIT. 

OK, so fast forward. The person has HIT. They're on argatroban. On post-op day 9, the platelets have improved sluggishly to 80. No bleeding complications are noted. Which of the following is the most appropriate next step at this time? I'll let you guys read. All right, so who would do A? [INAUDIBLE]? Why wouldn't you do A? 

Platelets are not up yet. 

Yeah. Platelets still not up. What about B? Platelets still not up. C? Platelets not up. D? Yeah, don't need urgently to get them up, so E, continue argatroban for now. 

All right, post-op day 11, Dr. Mann. Platelets are improved to 151. He's on argatroban, his renal function has improved, he's off CRRT. He lives alone, he's compliant, but says, yeah, I get a little confused if someone tries to make too many changes to my medications. So what's the best thing to do for this gentleman? Hopefully, you don't have to read very far to figure out the right answer. 

So you can stop argatroban. You can transition to Eliquis. We'd avoid Xarelto, given his renal dysfunction, as he's got stage 3 CKD. You don't need to continue argatroban. Could you do warfarin? You could. Sorry, that's a alternative. Stop argatroban, start warfarin. Again, you have to bridge with warfarin. Yes. 

[INAUDIBLE] 

So I think if there is a suspicion of HIT, I would recommend go ahead and putting-- oh, yes. So the question is, can you wait for the HIT antibody to come back to put heparin on an allergy list, so the patient does not continue to receive heparin? If you have a high enough suspicion, I recommend early edition of heparin to the allergy list. The last thing you need is for someone who has HIT to get a flush of heparin. That's a good question. Yes? Yes? 

Someone had a case about three or four years ago, [INAUDIBLE] for checking SRA [INAUDIBLE] and then giving them heparin if it's negative. 

Yep. So the question is, if someone has a remote history of HIT, should you routinely check SRAs to see if heparin is safe? So based on guidelines and based on available information, I would. However, if the patient is fearful that you are going to murder them by giving them heparin, then probably not. There's no right answer for that question. But in reality, you can check SRA and can administer heparin, if needed. That's a good question. Does anyone have any questions or comments on that? 

Just on [INAUDIBLE] 

That's a good-- yeah, I didn't put it on the list of differential diagnoses, but that's certainly a consideration. In this particular case, the person had cardiogenic shock and was requiring pressors and so was hypotensive and had ATN. Or I should say-- the thought was that he had ATN. No schistocytes on his smear. That's certainly a consideration, at least initially. I would say that in his particular case, it turned out that it could have been multifactorial. It may have been with that. 

And that actually brings up a good point because if someone has-- this is sort of off topic for this talk, but there's questions about whether or not you should plasma exchange someone who's got refractory HIT. But that's a good point, Dr. Lobel. And the comment was consideration of HUS and atypical HUS in the differential for his thrombocytopenia. 

So I'm going to talk about DIC as the second part of my talk. So DIC is interesting because it is sort of a diagnosis, but it's sort of not. It's sort of what is a syndrome that happens as a result of something else. So you can say that someone has DIC or doesn't, but it's always secondary to an underlying condition. Whenever you pull it up in a textbook or UpToDate, there are a million potential causes of DIC. 

Essentially, this is the coagulation system going awry from excessive overactivation, to the point where all of your natural breaks in coagulation are lost. I knew I was going to regret this, but I do think it's important to understand normal hemostasis, in order to understand DIC. So I'm going to boldly do what nobody does. I'm going to talk about the coagulation cascade, right after lunch. 

So hemostasis is actually kind of simple, in the sense that it starts out with-- let me make sure I get my mouse. Hemostasis is relatively simple. What happens is you have a site of an injury, and you have temporary vasoconstriction at a site. You have platelets that are circulating, and they attach in a non-energy dependent fashion to a von Willebrand factor that's expressed on the injured endothelial surface. So you're platelets adhere, and that begins everything. 

Your platelets adhere, and when they adhere, they undergo shape change, they release their granules, they get more of their buddies to come and join them. And so they recruit themselves, and they aggregate, and they form this primary hemostasis plug. It's interesting to note that platelets are always drawn as circular, but when they're activated, they look like starfish. They look like cells that have multiple pseudopods, and they're long spindly arms. So platelets don't actually look like this once they're activated. But you get the point. The point is that platelets start clotting cascade, and they form a platelet plug. 

This is absolutely horrendous, but what happens in secondary hemostasis is that you have activation of different clotting factors that ultimately coalesce on the common clotting cascade, and that's factor 10 becoming activated to 10A. And then in the presence of 5A, you make thrombin. Now this is called the extrinsic pathway. This is called the intrinsic pathway, but in reality, your extrinsic pathway is the primary driver for hemostasis. That's really important to remember because remember, whenever there's injury, you have expression of tissue factor. It's like an explosion happening on this side, on the right side of the screen. That is more in line with physiology than factor XII starting everything. 

One of the things we always forget is we just talked about platelets. They form this nice little platelet plug, and they sit there. But what is important to remember is that this explosion of thrombin generation happens on phospholipid surfaces, and platelets provide roughly 75% of the phospholipid surfaces for this to happen on. So once you have the platelet plug, it actually serves as the substrate for all of this to occur. That makes sense. 

And again, once you have thrombin generated, you have fibrinogen becoming activated to fibrin. You crosslink it, and then you make your nice little firm clot. So that's what that picture was. So that's a picture of an activated platelet. You have the inactive platelet, and then you have activated platelet. Doesn't look like a nice little disk at all. 

The interesting thing happens when you form a clot. When you form a clot, the clot actually activates its own destruction. What do I mean by that? When you have thrombin generation, and you have activation of fibrinogen to fibrin, and you crosslink a clot, the clot actually pulls in plasminogen. Plasminogen is activated by TPA made by your own endothelial cells, activated by TPA to become plasmin. As soon as you have plasmin, plasmin breaks down the clot. 

So the interesting thing is, while you're laying down the clot, you're actually putting down the foundation to destroy it. So you have auto-destruction. And if you have excessive plasmin, it's bounded by alpha-2 antiplasmin. If you have any excess plasmin, it gets bound up. So for every clot that's formed, its own destruction is actually already at play. This is what happens in the endothelium. This is all physiologic. This happens normally. You secrete TPA, and you activate fibrinolysis, like I just showed you. 

The other thing is that you have thrombomodulin that's expressed on surfaces of endothelial cells. So as soon as you have thrombin generated, it's bound up by thrombomodulin, you activate protein C, you inactivate two clotting factors up here. You also have an endogenous expression of heparan, and once it's linked up with antithrombin, potently inactivates thrombin. So all of these counterregulatory measures normally occur on a continuous basis, so that you only form a clot where you need to. That is what happens endogenously. It's like what I just said, everything happens at the same time. 

But DIC doesn't happen all the time because everything is very tightly regulated. So how is DIC different than what I just described? The problem with DIC is you have excessive overactivation of coagulation. You cannot stop that from happening. So what happens is you make a ton of fibrin. You use up all your platelets or clotting factors. Fibrin gets in the way of platelets being activated. You have clotting, you have bleeding, the whole thing is just a complete mess. 

And because you made so much clot, you try to break it down, you have fibrin degradation products, it interferes with platelet function, fibrin crosslinking, and you have more bleeding. So that's how you can have bleeding and clotting at the same time. And we get this question a lot. How do you have bleeding and clotting at the same time? This is what happens. You can show them this slide, actually. 

So features of DIC, you can have major bleeding happening at the same time as thrombosis. Thankfully, there's very few cases where people die from major bleeding, but it still happens with DIC. And there is a lot of thrombosis, unfortunately, in our cancer and septic patients. This is a slide that just shows you there are multiple spectra of DIC. You have acute DIC, which is what you see in the hospitals, so that's the scope of our talk. 

But there is such a thing as chronic DIC. You can have DIC happen locally. It's very unusual, but there are cases of localized DIC. And again, manifestation, bleeding, and clotting. This is sort of a dizzying list of different causes of DIC. The things highlighted in blue are typically thrombotic in nature when they present, and then the ones in red are typically hemorrhagic. 

How do we diagnose DIC? Well, DIC is clinical, so good luck. Use your judgment. You can have bleeding and clotting at the same time, so you can't say, oh, the person is bleeding. It's not DIC. Oh, the person's clotting. It's not DIC. You can have either. You can have both. No single test to diagnose DIC. We typically see coags are prolonged, fibrinogen is low, that's the first tip off. Starting from health staff to really any stage of training and learning, we all know coags prolonged, fibrinogen low, platelets low, we start thinking about DIC. 

There are a lot of labs to qualify what's going on. I had a couple of slides here about thromboelastography. I won't go into the details of it, but essentially, you can also do thromboelastography to look at what is your clot formation looking like in real time. The problem with these tests, like I said, is that the test is good for the time that you drew it. So if you drew a fibrinogen one week ago, and the person's crashing, burning, their platelets are low, that does not exclude DIC. 

The other thing is that fibrinogen typically is an acute phase reaction, so it can still be elevated in DIC. So don't look at a high fibrinogen to reassure yourself that it's not DIC. Or don't look at that as the only way. What else looks like DIC? We see this quite a bit. Coagulopathy of severe chronic liver disease, a delusional coagulopathy, thrombotic storm, like if someone has catastrophic APS. There are a lot of things that can look like DIC. 

I wanted just to spend a couple of minutes talking about the coagulopathy of severe liver disease, just as a very brief aside. The interesting thing about how I just described coagulation becoming completely disregulated, a lot of those findings, you do see in chronic liver disease. You have coags that are prolonged, you have platelets that are low, fibrinogen that's low. You can have bleeding and clotting at the same time in chronic liver disease. So the problem is that it's hard to distinguish chronic liver disease in DIC, but you really have to take the contextual scenario and figure out if it's DIC or not. 

ISTH, which is the International Society of Thrombosis Hemostasis, they have a DIC scoring system looking at probability of DIC. So these are things that we all know, low platelets, high d-dimer, a prolonged PT, more than a certain number of seconds past your normal fibrinogen level. 

How do we treat DIC? This is why I wanted to spend most of the time talking about HIT because with DIC, the most important thing is to treat the underlying condition. That's the single best thing you can do for the patient. If there's underlying bleeding issues, if there's underlying clotting issues, there are certain things that we'll talk about in a minute to see if you can address those active issues, but again, the equilibrium is completely disregulated in DIC, so treating your underlying condition that got you there is going to be the most important thing. 

This is always tricky. If someone has a hemorrhagic presentation for DIC, their platelets are low, their coags are out, and their fibrinogen is low, how much blood product should you really be giving this patient? In general, we say to avoid routine platelet transfusions, unless the patient is bleeding. You can try to keep them over 30, but again, you want to address the issue that resulted in them having low platelets in the first place. 

You can give them cryo if their fibrinogen is extremely low and they're bleeding. I would be cautious about FFP replacement, so for a bag of FFP, that's 300 ml. You have 300 units of every single clotting factor. It's not a lot of clotting factor, so I would avoid using FFP to replace clotting factors in this scenario. 

Heparin, there's not a whole lot of direct evidence for the benefit of heparin, but conventional wisdom is if the person's clotting, you should try to anticoagulate them. And again, this doesn't come up anymore because we used to monitor PTT. And again, if someone has a prolonged PTT from DIC, PTT is not reliable means to monitor heparin, so we all use a hospital-wide anti-10A monitor for the most part anyways. So you can anticoagulate them with heparin. 

I'll just spend a few minutes talking about novel therapies for DIC. I had some information on here, so that people can access the slides and look at it. But long story short of it is there's purified concentrates of antithrombin. There's recombinant activated protein C, there's humanized thrombomodulin, there's a lot of means to treat DIC. nothing Has been really promising. So ATIII you can give thrombate. It's called thrombate in the concentrate. Not a whole lot of evidence for benefit on mortality, so it's generally not recommended. 

Activated protein C, the critical care folks would know better than myself that there was a study that showed that there was better mortality benefit for activated protein C infusions. Unfortunately, after a groundbreaking study came out, there was just some issues with irregularity in how the study was designed, and it seemed a little too good to be true. So when repeat studies were done a couple of years later, there was actually no reduction in mortality for patients with sepsis and DIC receiving activated protein C. The drug was actually withdrawn from the market in 2011. 

Can you use a DTI to anticoagulate instead of heparin? As far as we know, right now there's no clinical role for them quite yet. Antifibrinolytics, can you use antifibrinolytic agents, like Amicar and tranexamic acid, to stop excessive clotting or excessive bleeding? You can, but there is a risk of thrombosis with them, so the recommendation is give heparin and then give antifibrinolytic agent. So not a whole lot of people would do this. It would make them really nervous, rightfully so. 

The only thing I was going to highlight is that the ISTH, in 2013, did say that potentially down the road, you could do human thrombomodulin, as I just showed earlier here. Thrombomodulin binds up thrombin and activates protein C, so potentially there's a role for recombinant human thrombomodulin down the road. Not quite there yet for primetime. Treatment summary, treat the underlying cause if you can. DIC is just very, very difficult to get ahead of because you're already behind, by definition. But again, probably some early risk stratification of patients who might be at risk for DIC would be helpful. 

And I'm going to just close with one case. Of course, I'm in the division of hematology, so there's going to be a cell up on a slide. 40-year-old female presents with epistaxis, gingival bleeding, easy bruising, unusually heavy menses for one week. So she's got a normal total white count, but she's got a lot of promyelocytes on her dif, some metamyelocytes. Hemoglobin is 8, platelets are 35. You look at the microscope, and you're thinking, you know, I'm not a hematologist, but that looks like an Auer rod. 

PT is 17.1, PTT is 55.6, fibrinogen is 89. Which the following is the most appropriate single next step? I know everyone remembers this from residency days. All right, someone can just shout it out. They don't have to-- So yes, you should treat before you verify. I would start ATRA on this patient to get ahead of what's causing DIC happening in her case. So I think whenever this comes up hem consult service, thankfully not that often, the person's having not significant bleeding. You haven't had time to do a bone marrow biopsy. You haven't had time to do flow. Go ahead and start ATRA. It won't hurt them. 

Question two, which of the following is the most appropriate next step after you've already done that? Would you do a bone marrow biopsy? Would you give her platelets, give her FFP, or give cryo and consider platelet transfusion for her? Remember, epistaxis, gingival bleeding, heavy periods, hemoglobin of 8. You would probably give cryo and consider a platelet transfusion for her. Later on, you do a peripheral blood flow and bone marrow biopsy, showing that she does have AML. 

Just a brief slide to tie up this case. APML is associated with DIC. It's not the only type of acute leukemia that can be associated with DIC though. You can have acute monocytic leukemia. You can have acute lymphoblastic leukemia associated with DIC. Most patients come in with bleeding complications. In the early population-based studies, the mortality was 30% And so I think with early initiation of treatment, like ATRA, fewer people die from bleeding. And you also correct their coagulopathy very quickly, relatively quickly. So probably within several days of treatment the coagulopathy would correct. 

All right. I did have one more case, but I'm not going to do that. So do you guys have any questions? 

[APPLAUSE]