I was really privileged to give a talk on upper GI/gastrointestinal emergencies last year at St Emlyn’s LIVE. You can read more on the overwhelming impostor syndrome I felt standing there not only with, but also in front of and talking to, such a wealth of experience in emergency medicine and critical care, echoed in Nat’s post from a few years ago. You can watch the talk below or listen to the podcast on our iTunes channel. This blog is designed to give you the background behind the talk.
However, this post isn’t about impostor syndrome, and we’ve got no time to worry about that anyway. The bat phone has just gone off. It’s a red standby.
33 year old, vomiting blood, blood pressure 70 systolic. 6 minutes.
That patient is on the way to your emergency department. They are bleeding out, haemodynamically unstable, cold, shocked, in trouble.
The concept behind my talk was that if they’d been hit by a car – if they were a “major trauma”, then they would be taken to a specialist centre, greeted on arrival by a team leader and at least a dozen people focused solely on keeping them alive.
However, we don’t have a major upper GI bleed team in my hospital, and I suspect you don’t in yours either. These bleeding GI patients may benefit from all the experience and skill that a bleeding trauma patient gets, however we often try to manage them in house, to start with at least.
When they arrive, they will undoubtedly be the sickest patient in the department. The sickest patient in the hospital. What would happen if we adopted a major trauma approach? Who would we have in our team, and why?
We love trying to keep our airway skills in house, but really for this patient we need a senior anaesthetist or two, together with an experienced ODP, ready to manage what is undoubtedly going to be a difficult intubation if it’s needed.
There is evidence that intubating these patients isn’t great, and they are more likely to develop cardiopulmonary complications as a result, which isn’t too surprising given how unwell they are in the first place. However sometimes they just need a tube. If this does happen you need to be prepared.
Over at LITFL there are some fantastic tips on how we can modify the RSI approach to this patient, but in summary there are four key components.
- Lines – your patient will crash. get lines in and start vasopressors
- Position – head up, but make sure the bed tips head down in case of vomiting
- Laryngoscopy – use video, but with a device such as C-MAC or McGrath that allows for direct laryngoscopy in case the view is obscured
- Suction – you’ll need one person focused solely on keeping that view clear, and if you are set up for the SALAD technique, this is the time to use it
After definitive treatment, this patient is going to end up in critical care, or not at all, so get them down and involved early. They’re also great at coordinating things over the phone to allow you to continue to lead the resuscitation efforts.
As emergency physicians we should be good at getting lines in, but in this team-based approach it’s important to delegate tasks too. An arterial line will help you out here – instant blood pressure, instant bloods, and super handy if the patient needs intubation or vasopressors. One tip from me is to have a transducer set hung and ready to go at the start of the shift, as if you need one in a hurry it’s often a rate-limiting step.
It doesn’t have to be a doctor running the resus, and our nurses have a huge range of other skills too. Talking to the family, managing the airway, and getting access are just some of the things they can do.
They’re also key in making sure patients get any drugs that they need in a timely fashion. But what do they need? We know PPIs have shown no mortality benefit.1 Tranexamic acid is a drug we probably will give because of good evidence in trauma, but at the moment there is no good evidence in GI bleeding. The HALT-IT trial2 was supposed to conclude in 2017, but has been extended (with an increased recruitment) and only stopped recruiting on the 21st June 2019. What does this extension mean? We can only speculate at the moment but hopefully we’ll get some answers soon.
For suspected variceal bleeds, terlipressin reduces mortality by a third3 and antibiotics should be given as well, as 20% of cirrhotic patients with these type of bleeds will develop a bacterial infection within 48 hours.4
Bleeding patients are bad, but the body can also cope with blood loss, up to a point. We also know that transfusion comes with risks, so we need to try to give our patients exactly the right amount of blood products, not too much or too little. This is difficult.
We have to look at the full clinical picture, but the evidence is there that in general a restrictive strategy is best.5 With our arterial line in we can get up to the minute haemoglobins from our gas machines. Don’t forget that these values are less reliable in patients with ongoing bleeding.
NICE guidelines lay out their suggestions for the use of various blood products based on FBC and clotting. If your department has access to TEG or ROTEM then this can also be used to guide transfusion.6
Don’t forget to check what meds your patient is taking as well, and discuss with haematology any of the newer anticoagulant drugs, as they might be able to help with reversal strategies.
This is ultimately the definitive treatment and where you want your patient to end up. We can keep throwing blood products down a cannula, but unless we put a plug in the bleeding, there’s not much point…
Make sure you know whether it’s the surgeons or the medics who run your GI bleed rota, so you know who to call at 3am when this patient arrives in your department.
Don’t forget as well, that in the patient with variceal bleeding, a Sengstaken-Blakemore or Minnesota tube can be a lifesaver. You need to know not only how to insert one of these, but also where they are kept in your department. In my talk I referenced Scott Weingart’s blogpost over at EMCrit, as well as an awesome video that Jess Mason from EM:RAP created to demonstrate exactly what to do.
Endoscopy can fail, but luckily there is another option. Interventional radiologists may be few and far between, but it’s increasing as a specialty and some of the stuff they can do is magic. If you have access to IR in your hospital, don’t forget them for patients who are bleeding. They may have some answers for you.
This paper from the US gives a good overview of what radiologists can do for your patient, together with some cool pictures.7
So, those are some of the people you might want on your team. But what about you? What skills are you going to bring?
If we’re going to take a trauma team approach, we need a team leader. Someone with the communication and leadership skills needed to co-ordinate this group of people in the resuscitation of the sickest patient in the hospital.
You’ll also have knowledge around this topic, which you can impart on those team members in order to help them perform their role to the max.
You will know how to optimise the airway, and how to modify the RSI to try to make intubation as easy as possible. What lines to insert, and when. You will be able to make decisions on which drugs and blood products to give, based not only on the clinical picture in front of you, but also the evidence-base behind. You will also be forward planning where this patient is going to end up, having activated your upper GI bleed team and pathway. You’ll also know if IR can act as a backup, in case endoscopy is delayed. You will know how to insert a Sengstaken-Blakemore tube, having watched Jess’ video a million times, and carried out some hospital wide simulation to practise this too.
We shouldn’t be doing this alone.
Think, if this patient was about to arrive in your emergency department, would you be able to offer them the best possible management, first time?
- 1.Sreedharan A, Martin J, Leontiadis GI, et al. Proton pump inhibitor treatment initiated prior to endoscopic diagnosis in upper gastrointestinal bleeding. Cochrane Database of Systematic Reviews. July 2010. doi:10.1002/14651858.cd005415.pub3
- 2.Roberts I, Coats T, Edwards P, et al. HALT-IT – tranexamic acid for the treatment of gastrointestinal bleeding: study protocol for a randomised controlled trial. Trials. November 2014. doi:10.1186/1745-6215-15-450
- 3.Ioannou GN, Doust J, Rockey DC. Terlipressin for acute esophageal variceal hemorrhage. Cochrane Database of Systematic Reviews. January 2003. doi:10.1002/14651858.cd002147
- 4.Lee YY. Role of prophylactic antibiotics in cirrhotic patients with variceal bleeding. WJG. 2014:1790. doi:10.3748/wjg.v20.i7.1790
- 5.Villanueva C, Colomo A, Bosch A, et al. Transfusion Strategies for Acute Upper Gastrointestinal Bleeding. N Engl J Med. January 2013:11-21. doi:10.1056/nejmoa1211801
- 6.Kumar M, Ahmad J, Maiwall R, et al. Thromboelastography‐Guided Blood Component Use in Patients With Cirrhosis With Nonvariceal Bleeding: A Randomized Controlled Trial. Hepatology. May 2019. doi:10.1002/hep.30794
- 7.Ramaswamy RS. Role of interventional radiology in the management of acute gastrointestinal bleeding. WJR. 2014:82. doi:10.4329/wjr.v6.i4.82