V & A in the ED – blood gases – St.Emlyn’s

arterial and venous blood gas

“Why are you doing an Arterial Blood Gas on that patient?”

I thought it was a decent enough question to put to one of our senior EM trainees in the resus room at StE’s. I, along with many clinicians I know, have dispensed with the default position of A>>V. You see, I believe that arterial blood gases are, in most circumstances, completely unnecessary. This post will try to explain why I feel this way and why you should really consider the reasons for this investigation.

Firstly, they really hurt! I would suggest to anyone that says they don’t, that they allow a nervous junior doctor, syringe in hand to aspirate blood from their radial artery. I suspect any keenness for inflicting this procedure on their own patient will rapidly diminish. It’s been a few years since the publication of the British Thoracic Society‘s excellent and practical emergency oxygen guideline, which explicitly states:

“Local anaesthesia should be used for all arterial blood gas

specimens except in emergencies or if the patient is
unconscious or anaesthetised” (grade B)

Firstly, I know we deal with a lot of emergencies, but finding the time to administer a small amount of lidocaine and wait before the blood gas is surely humane and in the vast majority of cases doesn’t delay treatment to any great detriment. How often does it happen? Not enough, IMNSHO.

Secondly, we have a very useable proxy in the form of a saturation monitor. I’m not suggesting that pulse oximetry is the be all and end all of respiratory monitoring, as it patently isn’t and has well known flaws including nail polish, skin pigmentation, poor cardiac output states, abnormal haemoglobinopathies etc., as this article from Critical Care attests. Unless you are applying higher oxygen rates (and I’d argue that unless you were using a venturi mask, which we rarely do in my ED, you have no idea what FiO2 you have got) then it probably doesn’t matter.

Thirdly, we don’t use the data that we get from blood gases properly, in the majority of occasions. I have asked numerous trainees of all levels about the alveolar gas equation and have in the vast majority of cases been met with blank looks. Now I am a geek, have an interest in critical care and have as one of my bookmarks this fantastic webpage but it seems to me that we forget about both the AGE (which, thankfully for StE’s, which is at sea-level, can be rationalised to 95 x FiO2% – PaCO2/R) and more importantly the Oxygen carriage equation in which the SaO2 is vastly more important than the PaO2. I’m not suggesting that you should be able to bore at length down the pub about the four types of hypoxia, and more specifcally hypoxaemic hypoxia, to do an ABG, but we do have a duty of care to our patients to use all the information properly: it’s akin to looking at an ECG, noting the AF, but failing to spot the inferior infarct.

Fourthly, we, as humans, recognise patterns. We look for trends in our blood gases, looking at the change in values over time. Got a COPD patient that’s obtunded? Do you really need an ABG to tell you that they need BiPAP and reduced oxygen concentration? What are you actually going to do with the exact partial pressures that your ABG will reveal? Stuff all, that’s what. You’ll bosh on the BiPAP, dial down the FiO2 from the positively harmful prehospital levels to achieve sats of 88-92% and watch the PvCO2 slowly improve, along with the patients clinical state. If you do need to know about PaO2, (thinking about shunt, fibrosis, V/Q mismatch) then stick in an arterial line, with local anaesthesia, and refer them to HDU if they are that sick. Alternatively do earlobe gases as suggested here and here. Or cap gases; they’re good enough for children, and, as we all know, adults are just big kids {probably more so in the case of Simon and myself!}. Personally, I’ll stick to the pH and bicarb levels of a venous gas, based on this from the original evangeliste for VBG in the ED, Anne-Maree Kelly.

We should look more at Professor Kelly’s work. Her page above details a chunk of work that has at its heart, common sense and practicality, which should be preeminent in EM research. She successfully debunked the need for ABGs in DKA and, as far as I can tell, was the first to publish the potential utility of VBG in respiratory failure.

Having said all of this, half an hour after the initial discussion with the trainee, I was doing what I had counselled against. However, this patient had a spurious “history of asthma”, sats of 89% on room air and looked clinically very well. Her ABG (taken with lidocaine) showed a significant shunt of 17 kPa that was used to make a very esoteric and eponymous diagnosis that I am keeping back for another blog post as it was straight out of MRCP part 2!

The bottom line of this rant, and I hope that it’s more educational and challenging than the grumpy, bald man firing off pot-shots at what irks him, is that we, as reflective clinicians, should be continuously questing for the reasons why we do what we do. If we’re doing an invasive procedure, we owe it to our patients, ourselves and our colleagues to use ALL the data properly, not just that that we know how to interpret. It also comes back to a common recognizance of mine, that the most important stuff I needed to learn at med school was hidden in the first couple of years when I was drinking beer and chasing girls; and physiology, biochemistry and anatomy were not seen as the vitally important subjects that they are, but an inconvenient distraction from my social life, not brilliantly taught poorly learned and not applied to my desire to heal the sick. It’s taken me the best part of two decades to work this out and I’m still playing catch up. Hopefully, I can stop this happening to you.



NB Ed – Virchester is not a million miles away from the Victoria and Albert museum in Londonshire. Well worth a visit if you find yourself in our vicinity someday and fancy visiting the world’s best collection of art and design (discuss).

6 thoughts on “V & A in the ED – blood gases – St.Emlyn’s”

  1. Quite agree. Sadly (on so many levels) the usual reason for about 80% (wild unsubstantiated guestimate) of ABGs is “because the Med Reg was going to insist on it before accepting the patient”. I dispair.

  2. Pingback: Blodgas | Kurs i Akutsjukvård

  3. Pingback: CEM conference Exeter 2014 - day 1 videos - St Emlyns

  4. Pingback: Self Experimentation in Medical Education - LA for ABGs - St Emlyns

  5. Pingback: The LITFL Review 078

  6. Pingback: The problem with ABGs • The BREACH

Thanks so much for following. Viva la #FOAMed

Scroll to Top