Pediatric Trauma II: Solid Organ Injury

This podcast is an interactive discussion about solid organ injury between Dr. Todd Ponsky, Dr. Mark McCollum, and Dr. David Notrica.

Dr. David Notrica is the trauma medical director at Phoenix Children's Hospital and is associate professor of surgery at the Mayo Clinic College of Medicine and Associate Professor of Surgery at the University of Arizona College of Medicine in Phoenix.

Additional Content

Can you tell us a little bit more about what ATOMIC is and what you guys did about it?

Atomic started around 2010 and it's interesting because when we decided that we were going to do an algorithm for how to treat cells organ injury. We were really inspired by all of the research that has been done up until that point. So, while previously there was some evidence on how long you should keep patients in the hospital. No one had ever tried to do a prospective study and the evidence was starting to come out that maybe you didn't need to keep patients in the hospital as long. So, we developed the algorithm by looking to see what information was out there and then we started asking questions and those questions well ultimately became the great publication that we published in The Journal of Trauma and Acute Care Surgery. So, the studies that were being done were really showing that you did not keep patients in the hospital for a plus one and that a lot of limitations that we thought were important probably were not as important. I had amazing collaborators when we started the atomic at the time it started really with five hospitals and we expanded it to ten hospitals. And now we're continuing to expand the opportunities to do multicenter collaborative research. So, going back to your original question ATOMIC is a consortium of hospitals that do research together for trauma and really it was started out with people that knew each other and then expanded to more people who knew each other and started doing some really great research that ultimately led to numerous publications that kind of you know present evidence behind what we've been doing.


In patient with blunt solid organ injury, what is the evidence to support non-operative management based on hemodynamic status as opposed to a grade of injury?


We started to see from the work done by the folks in Arkansas and then the folks in the work that St. Peter did. That was the original decision to manage patients based on grade of injury rather than hemodynamic status. It was kind of a consensus opinion or expert opinion that's how it developed. Steve did that and he was really important at the time because for the first time, we were getting C.T grades on patients we didn't know what to do with it. But shortly after that decision was made Sam Smith and the folks in Arkansas said wait a minute, you can actually manage these patients based on their hemodynamic status, you don't need to know what the rate of injury was and that led to a paper very early on that said hey you divide it up into stable and unstable and that's how you manage them and then nobody seemed to pay attention to that paper which was a great paper and they followed up with another paper called Throwing up the great book if you know anything about trauma literature is important to have a great title. So that paper does have a great title. We've looked at it and realized that he was on to something. St. Peter then did a study where he managed patients based on hemodynamic status and then did a prospective study, and the Arkansas folks did another prospective study and we started to have some real scientific evidence that said: yes you can manage patients based on hemodynamic status rather than grade of injury. That's how we took that first leap, it was going from throwing out the great book effectively to let's look at the patient. If you think about it pediatric surgeons have always said it's let's look at the patient let's make the decisions based on our physical examination skills and C.T. scans are important but looking at the patients is still important even in 2018.


Can you describe some of the other factors that we would use to define hemodynamic stability?


Hypotension is late finding that the time kids are hypotensive is very late. And then you look for the evidence to support it, you can't find it. We all know those of us who take care of injured patients and know that that's true, but we couldn't find the literature to show where to support that. That haven’t been said we've had since that time since the original study, we actually went back and did do that research which did show that hemodynamic status is a late finding in those patients who need to be transfused. Almost half of patients who need to be transfused early are not hypotensive. So, that's half the answer to the question.


The other answer to the question about what to use in order to define who is stable or unstable was incredibly problematic because even though we know with a stable patient or unstable patient looks like when you go to define it becomes extremely difficult. So, some patients who have concurrent head injuries may not have tachycardia. If you're an adult patient you have hypotension and then you know that they're unstable, some pediatric patients are unstable and not hypotensive. Some patients are tachycardiac due to pain and are not actively bleeding and so ultimately, we had to abandon the terminology stable and unstable and we changed the terminology to patients who were bleeding or having bled recently.


I will go further and tell you that when a patient arrives to you and they are hypovolemic in pediatric patients arrives to you hypovolemic, the only thing that we know is that they have bled. You don't really know if they are still bleeding. You often don't know if there's still bleeding until you give them blood and they prove that they're still bleeding. That was a very difficult concept for adult trauma surgeons but for pediatric surgeons we all kind of knew in the back of our mind that a patient who comes in looking in shock may or may not still be bleeding because often they'll bleed a bit and stop. They'll be in shock, but they won't continue bleeding. Kids have an amazing ability to stop bleeding and far superior to adult patients.


Do you use that in your center? Are the vital signs indexes like the shock index or laboratory values like serum lactate in addition to physical exam findings?


The SIPA which is the Shock Index Pediatric Adjusted was first described from the folks in Denver Colorado. So, accurate at all they basically said you can use the shock that which is the heart rate divided by the blood pressure. But what the pediatric adjusted shock index does it takes the old adage which is that if your heart rate is higher than your blood pressure you're in trouble and then it looks at it for kids and says actually if your heart rate is 1.1 times your blood pressure then you're in trouble, and they kind of just made an adjustment for kids. It turns out what the shock index does for you it allows you to identify patients who are not currently in shock, and it misses very few patients who are in trouble. So, if you're shock index is not elevated there's a very low likelihood that your patient is experiencing hypovolemic shock and that's how we use shock and unfortunately have a fair number of patients who have elevated shock index who are not actively bleeding or are not hypovolemic. But the converse which is if shock index is not elevated, those patients are probably not actively bleeding.Just for reference shock index is defined as you mentioned is heart rate over systolic blood pressure and the cut offs that we've used in pediatric trauma is between ages 4 and 6. A shock index greater than 1.2 in ages 7 through 12 shock index greater than 1.0. And then in kids older than 13 they follow along the lines of the adult literature which suggests shock index of greater than 0.9 indicates instability.


From a crystalloid infusion standpoint, what are your thoughts as far as limiting crystalloid infusion and patients that you know are actively bleeding not crystalloid, but bleeding blood?


You get to start with the old adages if you're not bleeding crystalloid. The thing that we have figured out is that if you want your patient to survive giving them a tremendous amount of crystalloid is not a good way to do it. So, let me start with where we were at before we started ATOMIC publication which is ATLS saying that you effectively gave 60 ml/kg crystalloid before you went to blood and the literature was really showing that adult patients who got a lot of crystalloid did not do as well. You actually diluted out the benefit of 1:1:1 resuscitation. We started to see that from work from Jiff Holcom that your chances of managing patients not operatively went down if you gave a lot of crystalloid. So, there was pretty good secondary evidence to suggest that after you've given a patient 20 ml/kg blood, it is time to start blood transfusion that you want to manage them non-operatively and if you want to have a good outcome and the evidence to support that was really pretty good. Now, there was some evidence that children were a little bit more resistant to the negative effects of crystalloid than adults. But even in that paper which was another Denver paper the patients who got access crystalloid still had some adverse effects including beyond the ventilator longer. So really strong tangential evidence to support it and some great head on evidence on adults to support switching to blood transfusion early and that's how that got into the Atomic algorithm and now that we've been doing that and become very comfortable with it we are really happy with it. The other thing that that happened was one of the trials that was set up to look at massive transfusion, they were 1:1:1 transfusion found that when they gave blood early and often not as many people went in to require a massive transfusion. That I think speaks the world to the effectiveness because if you if you give this stuff early and don't give them a ton of crystalloid, they don't even go into that DIC and do not need that massive transfusion and that's actually where you want to be.


What about the complications of transfusion short term and long term and how do we know that that is now not going up because we're giving more blood?


We're starting to do some work on massive transfusion and I know Mark's doing some studies on massive transfusion in kids too. They are hard studies because they're not that many children that are massively transfused. So, when you have a relatively small population receiving a massive transfusion protocol and the complications of it are fairly rare it's going to be hard to know that this is completely safe. That having been said, we know from the adult series because they have a lot more patients with massive transfusion protocol is that it does seem to be safe. It does seem to be the right thing to do.


It seems like now real viscoelastic assays TEG and Rotem are getting a lot more traction so that we're directing component therapy specifically as opposed to shot gunning a transfusion. I would love your thoughts on that?


So, using the TEG and the Rotem, I think makes a lot of sense from the theoretical standpoint. In my institution we do that we actually use TEG and Rotem to guide the therapy I think we use TEG technically. It's much harder to find great evidence to say yes this is the right thing to do just because in every small problem that we had before. But like I said right now we're going in directed therapy and I think there's a randomized controlled trial comparing 1:1:1 or 1.1:1:1 versus TEG in children that will ultimately answer the question. I don't know that we have the answer right now. But those numbers are starting to come back and we're seeing more and more data suggests directed component therapy not only saves money but it saves product and more efficiently can resuscitate a patient than the shotgun approach of a ratio driven massive transfusion.


I would have to agree. And you know at some point I thought we were going to have whole blood more readily available so it didn't matter whether or not we had the components we were just going to get back with their bleeding but we still don't have that available in Phoenix and I'm hoping that adult centers can drive the desire and the need and the availability of full blood transfusion for trauma patients.


In a patient who has ongoing bleeding is he hemodynamically unstable and you're in the process of resuscitating them with blood products. What are your thoughts on angioembolization its safety and efficacy in pediatric of blunt liver and spleen injury?


Let me take the questions separately. So, from a safety standpoint I think we've proved safety. So, I think angioembolization is safe in kids. Efficacy is a much harder question because we know that a lot of patients who have contrast extravasation on CT will stop leaving even without angioembolization. What we saw in our failure patients is that none of the patients who got immobilized for splenic bleeding went on to fail, so that's good news. But a lot of the patients who had it angio-embolization for liver bleeding did ultimately go on to need a laparoscopy a wash out another procedure. A lot of those were converted from active bleeding to managing bile complications or managing a lot of blood in the abdomen.


So, there were good evidence to support that the angioembolization was helpful even in the hepatic injury patients and I think that the data is supportive of angioembolization for splenic injury and for hepatic injury. Not every patient who has contrast extravasation needs an angioembolization because so many of them will stop. So, you really need to look your patient and if you continue to bleed then that patient needs to have an angio-embolization. We captured that in the algorithm by saying look the criteria for these angioembolization is similar to the criteria for failure of non-operative management. If they don't stop if you're having to give blood, then take them to the angioembolization if they are stable enough. Worth noting, things that we have to do to keep adult patients alive may not benefit children. That is the hardest part about changing the paradigm for managing children is that they don’t behave like adults.


So once we have these patients resuscitated and stable in kids with solid organ injury should ICU admission be determined then by injury grade, hemodynamics or a combination of both?


Before I wrote the paper, I was absolutely certain that hemodynamic status was the only thing that mattered and that and that we could save a lot of ICU stays. When I actually did a literature search that was supported up until Grade 5. So, if you have a grade 1-4 injury you can definitely say all right well if they are hemodynamically stable or if they're not actively bleeding they can be managed on the floor safely. But for the grade 5 injuries, there was enough evidence to suggest that those patients ended up themselves based on the grade of injury more in ICU, I would recommend grade 5 go to the ICU. Grade 5 injuries that are hemodynamically stable it is a very small subset of patients who could potentially be managed on the floor of the grade 5 injury. So, it is not a big deal. Most grade 5 injuries will have bled significantly and would have gotten into ICU based on that, but those patients were grade 5 injuries need to be in the unit.


Do you have a threshold of volume of transfusion that would then indicate failure or is it a case by case?


So there's a good evidence that we finally have a threshold for failure and the threshold is 40 ml/kg of all blood products, and that data came to us from the military experience. They were the ones that showed that if you got transfused more than 40 ml/kg of blood products you are more likely to need operation and more likely to die. We'd all kind of felt that that 40 ml/kg was a cut off, in some of the expert panels they had said yes 40 ml/kg was a failure point. We didn't have great evidence going into it. But now that we've really started to say yes that's the failure point, it seems to hold up really well.


What are your thoughts as far as time frame for bedrest? How long and what are the parameters that help you decide?


I'll give you a little background on bed rest. I ran in to see a patient, I said you know I think we are requiring too much bed rest for these kids. He said “we never required bed rest” I said really. It turns out, bed rest wasn't really part of the initial requirement. Even on the APSA protocols, but it became a very important part of the culture of what we did for the kids. You're absolutely right, they were not resting in bed, they were jumping on the bed. St. Peter said yes they are anything but resting, and there is not much literature to support bedrest, matter of fact there's no literature to support bedrest. There's one study in adults where were they compared to adults who were getting at bedrests and those who were not, and there was no increased incidence of bleeding. The problem that we ran into is that almost every study that that we wanted to reference had used bed rest, even though there's no evidence to support it. My personal feeling is that I don't think bedrest is anything, we were able to eliminate it in our renal injury protocol because we had studied those patients and found it makes no difference for renal injury whether you're walking around going to the bathroom or whether you're on strict bed rest, and I don't think that makes a difference for the liver or spleen injury. I don't think that taking an injured organ and walking into the bathroom is going to suddenly cause it to bleed.

 In hemodynamically stable patients, do you use a timeframe for observation or are they able to be fast tracked and may be discharged within 24 hours?


St. Peter answered that for us in a prospective study and then I did a follow up on it, and actually turns out you don't need to have them in the hospital for a minimum period of time. Most of these kids come in in the afternoon or evening and they are hospitalized “overnight” and then sent home the next day. Trying to pin down how long that was a little bit problematic, and what we came up with was a protocol that says yes, we need to keep him in the hospital for 18 hours. If they hadn't bled 18 hours and they were very unlikely to bleed that was an incredibly powerful for us to limit the amount of time, they are in the hospital. We've further supported that by showing to the patients who were transfused and the patients who were failed who failed non-operative management all failed early, and that further supported the fact that you if patients hadn't failed that 18 hours, they're probably not going to fail.


Do you have criteria or a threshold, Hematocrit or hemoglobin point that indicates lab variability versus actual continued bleeding?


So, we checked serial hemoglobin so long that we saw hemoglobin go down in patients who are not bleeding. The reality is they go down for a variety of reasons: lab error, being drawn incorrectly, if they are drawn right around or upstream from an I.V. It goes down, and If we diluted out too much fluid they go down. Ultimately, the better criteria for whether a patient is bleeding is not serial hemoglobin, it's physical examination and vital signs. So, if your patient's heart rate is continuing to go up, you have a problem. If your patient is showing clinical signs of poor capillary refill or pallor or cold extremities. Those are the signs that let you know the patient is bleeding and there's good evidence to suggest that serial hemoglobins are not necessary that you don't need to follow a serial H&H on a stable patient that you should follow a physical examination and the vital signs and that will identify every patient who will need transfusions and every patient who is failing non-operative management. I think that underscores the importance of serial examination by a primary examiner as opposed to the ship mentality of a different resident every day. Although I don't think we'll ever get back to the work hours that you and I had in training. More importantly, half the patients who fail non-operative management do it because they develop peritonitis not because they continue to bleed.


So, whenever these kids are ready to go home, do you have a standard follow up regimen of a week, two weeks, four weeks. The rest issue I'll let you discuss as well, as far as time frame where they really stay off of activities until you see them back?


The time to follow up was interesting. What we found is that a lot of patients with low grade injuries never showed back up for follow. So, when you think of all the grade one grade two injuries how showed for their clinic appointment. We thought well we should find out something about these kids. It is such a low incidence of there being anything that we do for the kids at the follow up appointment. So, we started doing telephone follow up and what did for us is it actually let us know that the kids were OK and it also prevented them from missing even more school because they're already met missed some school. So, we did telephone follow up for low grade injuries and we bring back the patients for follow up with higher grade injuries. That haven't been said, we haven't found anything of any use in the follow up visits. I know as surgeons we love our follow up visits and we think that they're super important. But ultimately the patients who needed additional imaging or who needed additional therapy all presented back to the emergency room with pain or problems, and the ones that came to the office didn't provide much benefit to the patient.

Is there a type of injury a finding on imaging or a symptom that would then maybe move you to schedule additional imaging in follow up to avoid a complication like a pseudoaneurysm or AV fistula or something along those lines?


It came to figure out what patients were having complications or which ones needed follow up. We were an information free zone and so what we did was we looked to the literature to find all the complications of non-operative management that we could identify, and we converted those complications into symptoms and listed those in the discharge instructions. So what patients are having complications, come back for abdominal pain. They come back for respiratory problems. So, you have a splenic injury and they get a pleural effusion and they're having trouble breathing that brings it back to emergency room. So, we kind of just took that whole group of problems and put in to this is what you watch out for, and we put those in the ATOMIC discharge instructions and basically all of them direct the patient back to the emergency room except for jaundice which is fairly common if you have a hematoma and you do get jaundice even though they're not having a complication and we directed them to call the office even if you are jaundice because you're having a little bit of a bile leak or biloma. Most patients can at least be triaged with a phone call. So, they don't automatically have to go to the emergency room. Equally, you really don't want to miss a patient with a bile leak and keep them at home. But a lot of times you can determine whether they've got a bile leak or if they're just mobilizing the hematoma just by talking with a mother or talk with the patient.


So, then it even in a grade 4 or a grade 5 injury with an active extravasation there's no real utility van and scheduling a post or a follow up ultrasound or additional imaging?


My personal opinion is that most patients with spleen injuries don't need it. There's a possibility that patients with liver injury might benefit from having a follow up ultrasound. In studies where they did routine ultrasounds of every single patient they found a lot of pseudoaneurysms in fact they found so many pseudoaneurysms that it was so discordant with the number of patients who have delayed bleeding or delay complications that just having a pseudoaneurysm of itself does not mean that you're going to go on to have a delayed bleed or a delayed complication. But, we don't know what subset of those patients with the asymptomatic pseudoaneuryism really are going to go on to have problems. It's certainly less than 10% but it may be even less than that. My guess is based on the study that we did that it's going to be extremely rare that any patient who goes home from the hospital doing well would develop any finding that would potentially put them at risk for or a life-threatening bleed or anything of that nature.


For a patient who is stabilized to the point that he is ready for discharge, what criteria do you use then for timeframe of follow up?


We switched to a follow up that was really based back on grade. So, they have a low-grade injury, we did telephone a follow up on those patients. Patients with higher grade injuries we do ask them to come back for a follow up visit. That have been said, the patients who do follow up with those grave 3, 4, or 5 injuries, at no point do we find anything particularly useful or helpful about that follow up visit. The patients who have problems ended up coming to the emergency room either before or after the visit, and when you really break it down to what that follow up visit did for them other than having missed an extra day of school. It's hard to point to any benefit for that in person follow up. The nice thing about the telephone follow up for the low grade is that it allows us to know whether or not we were doing the right thing and so from a study standpoint it was great. But from a practical standpoint I don't know that you need to drag those patients back into the clinic.


Regarding activity restrictions I'm going to answer that in two ways. So, it turns out with the activity restrictions that you need there's no reason those children cannot be back at school. They may need to change class 5 minutes before the other kids don't get roughed up in the hallways. Kids can do a lot of bumping and contact sports in the hallways, but they definitely be back in school. They just can't be in gym class or physical education class. And then for how long that is the activity restriction. We didn't have any evidence for how long you needed activity restrictions that we defaulted back to the APSA recommendation which was grade plus two in weeks so grade 4 injury is six weeks. I don't know that there's science behind that. I'm pretty sure there's not science behind it but there is expert opinion behind it and it's some point we'll have to do some research and figure out whether they really need those activity restrictions or not to answer that question.


Review for the for the audience some of the points that you made.

·       There is an abundance of evidence that supports non-operative management based on hemodynamic status rather than grade of injury and that's been a long time coming.

·       The Hemodynamic stability, hypotension is a late finding in children that are in hemorrhagic shock and clinical exam findings like pallor, poor distal perfusion serum lactate maybe shock index as well as tachycardia all can play a more significant role than actual hypotension when defining the child in hemorrhagic shock at an earlier stage.

·       There really is a limitation to the amount of crystalloid that should be infused prior to consideration or actual transfusion of packed red cells and that's a 20ccs/kg of crystalloid. If a patient is actively bleeding, crystalloid is not the fluid they need.

·       Patients who have solid organ injury that do need transfusion whether we should be transfusing them packed red blood cells utilizing massive transfusion early with transfusion ratios of 1:1:1 or if we should be directing a blood component resuscitation based on real time viscoelastic assays and I think those certainly are coming around and are going to be the standard of care soon.

·       1:1:1 ratio is a ratio of blood products used in transfusion and broken up since we don't use whole blood because of cost and availability. We then break up the components of the blood into packed red cells, platelets and FFP and that would be the 1:1:1 ratio of those three components.

·       Angioembolization in blunt liver and spleen injury and although we know it works. The interesting part is in spleen injuries non-operative management works in almost all cases. So, really there's not a great added benefit for patients that need intervention. if they need intervention and they're unstable splenectomy in most cases is the way to go. It is useful however in patients with multiple solid organ injuries like a combined spleen liver or spleen kidney injury so that you can manage active bleeding in one or two of the organs not knowing specifically what your source of hypotension is.

·       Requiring ICU admission and that should be determined by clinical judgment including both consideration of grade specifically grade 5 injuries and any Hemodynamic instability.

·        Lack of need of specific bedrest requirements in these patients and early mobilization and limited activities would certainly be safe.

·       It has been fairly well established in the literature that a 40 CC/kg blood transfusion would indicate clinical failure of non-operative management of solid organ injury. Now that needs to be put in perspective given the patients hemodynamic status and where you are on the evaluation management curve. But 40 cc/kg of blood transfusion would indicate a failure in most cases.

·       Patients with blunt liver and spleen injury the time frame for hospitalization and we've decide that in most cases patients with hemodynamic instability and no continued blood loss can be safely discharged within 24 hours.

·       In the asymptomatic patients despite the grade of injury or early imaging findings those patients in most cases do not need additional imaging, although you have to take that case by case. But patients with significant complications related to blunt liver or spleen injury often present with symptoms.

Resources, I want to direct the listeners to the paper that most of this has been published in that David led authorship on. It's called Non-Operative Management of Blunt Liver and Spleen Injury in Children Evaluation of the Atomic Guideline Using Grade. This was published in the journal of Trauma Acute Care Surgery in 2015. David Notrica is the lead author of this paper and in the paper each one of these questions is addressed and answered very clearly. So, this is absolutely a paper that everyone should have printed in their office. This is one of those sentinel papers that you'll be referring to repeatedly mostly because of the second page of the paper that has what I've been referring to all the time. It takes you through all the different situations and how to manage these patients, and it is spectacular. These do hit on all of the radical changes that Marc and David just pointed out that have totally changed the way we've been taught to manage trauma patients; giving early blood, not necessarily having those strict criteria for discharge or activity. These things are very new and need to be understood by everyone because this is a big change in our field.


Final comment, even though I was the lead author of this paper the this was the work of a lot of people. The coauthors that you see on the paper all helped develop this and by the time that we had an algorithm that we were taking to study prospectively we were on the 11th version of this algorithm. We've learned a lot in the process about how to make it how to operationalize the algorithm. But we also learned something that kind of snuck in there around the 9th addition which is that if you have recurrent hypotension or if you fail to have a sustained response to that first blood transfusion, you've failed. That was critically important because we did have a patient who responded became stable went to the C.T. scanner and then became unstable again, and the decision was made to try and continue to manage the patient non-operatively. I don't think we had really considered that particular scenario when we first were developing the guidelines and we realized late yeah early recurrence hypotension is a failure. It's not like being in the ICU and your hemoglobin is drifting down you didn't make it to the ICU yet. That patient really needs to go to the operating room or at least go to the angioembolization very quickly.