Ketamine sedation – what do like with your special K?

I must declare a “conflict of interest”.  I love ketamine, therefore anything you read from here on in is likely seriously biased.  Ketamine has had a big resurgence in popularity over the past 5 years in anaesthesia and ED practice.  There have been plenty of papers looking at its use for procedural sedation in the last few years. Last week I got into a minor Twitter-debate with @jvrbntz @keeweedoc and @PBSherren about current practice and evidence.

But when I look around my ED, and talk to colleagues in other places I find that there is a wide range of practice when it comes to using ketamine for procedural sedation.  So I thought I would open up a forum for discussion and link in some evidence for us all to consider and try to reach some consensus.

Here are the basic ways I see people using ketamine, and the drugs that get used with it in order to reduce side-effects or improve the sedation.

  1. Straight ketamine.   Do you like it IMI or IV (or maybe even intranasal?)
  2. Ketafol (K + propofol): What is the ratio to use? What dose? In one or 2 syringes?
  3. Antisialogogue ?  Atropine or glycopyrrolate
  4. Benzos – midazolam (or other) – before, with or after K?
  5. Antiemetics: ondansetron, droperidol, metoclopramide? Routine or as required?
  6. Monitoring – what do you use as a standard?
  7. Oxygen or no oxygen during sedation?
  8. Strategies for avoiding emergence phenomena – how do you do it?

Here is a link to the great, free-full text review of ketamine sedation from the Annals of Emergency Medicine, May 2011.  It is worth a read and might change your practice.  It has references to pretty much all the latest data out there on ketamine in the ED.

For the record – my current ketamine sedation routine (based on some evidence and a lot of guesswork and bias) is as  follows:

  • IV access – try to keep the kid as calm as possible, or allow hem to settle down before starting sedation
  • No IMI ketamine unless the IV is really tough or causing lots of distress.
  • Preload with ondansetron – NNT = 13, but not much downside to this other than cost.
  • No antisialogogues
  • Preference for Ketofol ( 1:1 ratio) for painful, brief procedures
  • Benzos – I use midazolam on a PRN basis if emergence is happening – just hang around as the sedation wears off and watch them closely.
  • Oximetry and continuous waveform capnography via nasal prongs
  • Oxygen via nasal prongs with a BVM on standby if they have a prolonged apnoea.
  • Keep the room quiet, turn off unnecessary lights and have mum nearby for the recovery phase

Ok, let me hear your current practice and idiosynchracies  – do you follow guidelines, evidence or just do what you find works well in your shop?


  1. minh le cong says

    nice one Casey.
    check out

    we reached the current world record for ketamine infusion dose in an unintubated sedated psychiatric patient with drug induced psychosis, recently.
    you would have been proud of the remote hospital team as they must have listened to our. podcast…sedated in resus room, full monitoring and special nursing for over 12 hrs!
    patient made full recovery, amnestic for whole event and sedation period and discharged by psychiatrist after aeromedical retrieval. adjunctive droperidol not so helpful but midazolam was.

  2. Matthieu G. says

    Disclaimer: seriously biased toward ketamine use after one year in the deep african bush without any resuscitation drug or equipment, not even oxygen.

    • preference for IV route, more easily titratable. IM occasionally, tried intranasaly for the first time recently.
    • -Ketofol: love it… I know the evidences show mostly no benefit, but love it anyway: amazing cardiovascular and respiratory stability, quick and calm awakening (most patients following command in less than 10 minutes after injection), almost no side effect. I use them in two separate syringes, both 10mg/ml. I first start with ketamine, titrating until the patient has a nystagmus and is slighty obtunded but still following command, then push the propofol. Depending on the procedure, I aim for a level of sedation that some would consider general anesthesia (no reaction to pain). The dose range depends on the patient and the procedure, but is usually in the range of 0,3-0,5mg/kg for ketamine and 0,5-0,8mg/kg for propofol. Thanks to our orthopods, we do a lot of prosthetic hip dislocation reduction in our ED. Those frail and elderly patients very often require no more than 20mg of each.
    • Never give any antisialogogue. Hypersiallorhea is seen almost exclusively in children and is mostly a cosmetic issue.
    • Used to always give a benzo with it, then moved to a PRN basis, then moved to ketofol.
    • Since I shifted to ketofol, never had and issue with vomiting, so no routine anti-emetic
    • Full monitoring is standard of care in my shop (but no non-invasive capnometry). All procedural sedation take place in the resus room with adequate equipment and drugs at hand. With ketamine alone for procedural sedation of a stable patient without major comorbidity, I would be happy with no monitoring at all.
    • Oxygen is also standard of care in my shop, but would non invasive capnometry be available I would be happy to use it on a PRN basis.
    • Prevention of emergence phenomena: pre-sedation suggestion to the patient (i.e choose the dream you would dream to have, like with your favorite porn star), minimal handling of the patient during emergence, minimal noise (alarm beep…) and lound talk around the patient. I prefer to keep the familly away (except for children) has patient tend to say (admit?) things during the emergence that they would never say in other circumstances. Anyway, since I moved to ketofol, true emergence reactions are just a vague and distant memory. Slight and transient euphoria at worst (or best?).


    • Great stuff Matthieu

      Oxygen can be used as a routine if you have waveform capnography as you are not relying on the sats probe to tell you when the patient is apnoeic / hypoventilating.
      If you have no CO2 monitor, then use oxygen with caution as it is easy to miss apnoea / hypoventilation until 2 – 3 minutes later when the sats start to drop.

      Personally the longest apnoea I have seen with ketofol is about 30 seconds, but I am sure it happens if you dose liberally.

  3. Not much different, but I wanted to ask – how do you do Co2 capnography on nasal prongs? Dedicated kit or do you jerry-rig something (I use a Hudson mask and usually bung a 14G catheter through the holes to connect ETCO2 monitor)

    Interested to knwo what kit you use so I can put in an order to blow the budget (again)

    • Tim
      We have nasal cannulae that act purely as sampling devices, as well as one kit that we use in endoscopy suite that you can use to deliver O2 at the same time.
      Will get back to you with the brand etc on the email (no endorsements here..)

  4. David Wong says

    Thanks for the post Casey. What’s your opinion or any of the others on fasting times for ketamine procedural sedation? 6 hours for solids (2 hrs for clear fluids etc) being the gold standard for GA’s, we are using a protocol that allows for 3 hours fasted for food but using ketamine +/- midaz only. Adding propofol into the mix could make fasting times trickier…

  5. Hi David
    I try to get 4 hours, though it depends on the context. If you have a kid who ate a hamburger then fractured the arm 10 mins later, they are never gonna be fasted.
    The dose of propofol you need is tiny, the actual “unresponsive” time can be a minute if you titrate carefully. So for short painful procedures I am a bit more relaxed.
    I read somewhere that > 4 hrs actually increased the risk of vomit / aspiration.
    Will have to dig that study up

    • Susan Lambe has a talk on CME download about fasting and procedural sedation. She concludes that there is very little evidence to support using the ASA Guidelines for procedural sedation in the ED. Our patients are less deeply sedated, we don’t as a rule manipulate their airway and we use different agents that are less emetic.

      For Kids there are 3 papers (Agrawal et al 2003; Bell et al 2007 out of Brisbane; can’t remember the third) Combined, there were over 3500 patients, vomiting was rare (about 160 in total) and no cases of aspiration. There was also no association between fasting and vomiting or even vomiting and aspiration.

      In adults, she could only find 2 relevant papers. There is a literature review of 25 procedural sedation reports with a total n = 4657 by Thorpe in 2010. Only 136 were fasted according to the ASA Guidelines. Overall, 17 people vomited (16 or whom were not fasted) and there were no reports of aspiration.

      The second paper was a multi-centre study by Taylor et al in 2011 out of Melbourne (Do Australians pig out and then rock up to emergency departments?). Of 2623 patients, 52% were not fasted by ASA Guidelines and only 34 vomited. Increased fasting time did not have a significant effect on vomiting. There was only 1 case of aspiration and that was in a fasted patient who did not vomit.

      So bearing in mind that people may be reluctant to report cases of aspiration, there is currently little evidence to support routinely fasting patients in the ED according to ASA Guidelines. Instead we should probably look at risks / benefits on a case by case basis – which I suspect many of us are already doing

  6. Ah, but for the kid who’s eaten a hamburger and got a broken arm, I’d be bunging in an ETT tube not faffing around with sedation.

    Wise old anaesthetist taught me ETT for all trauma, low threshold for RSI even if ‘fasted’

    • Tim
      Yes, sorry I was not clear. the kid with the hamburger, fracture, ED scenario – not one for ED sedation. Needs an ETT. Too much sleep to lose over wondering otherwise

  7. Dineo Moiloa says

    Just how common is ketamine-induced laryngospasm? Had a a bad case trying to LP a 3 yr old recently.

  8. Marty graves says

    Interesting debate. I’m interested in use of 5HT3 antagonists with ketamine. It’s thought that the sympathetic effects as well as the psychotropic ones are due to inhibition of mono amines, including 5HT. I haven’t found any studies however where ondansetron reduces emergence phenomenon. Any thoughts anyone? Is it because ondansetron is more Specific for the 5HT3 receptor?

Speak Your Mind