EMCrit Podcast 88 – Oxygen Physiology with Daniel Davis


  • Latency of Pulse Oximetry Signal with use of Digitial Probes Associated with Inappropriate Extubation (J Emerg Med 2012;42(4):424)
  • Latency and loss of pulse oximetry signal with the use of digital probes during prehospital rapid-sequence intubation. (Prehosp Emerg Care. 2011 Jan-Mar;15(1):18-22.)
  • Rate of decline in oxygen saturation at various pulse oximetry values with prehospital rapid sequence intubation. (Prehosp Emerg Care. 2008 Jan-Mar;12(1):46-51.)

Dan Davis at his best:

Did you like this episode? Then tweet the hell out of it…

Now on to the Podcast…

You finished the 'cast,
Now get CME credit

Already an EMCrit CME Subscriber?
Click Here to Get CME Credit for the Episode

Not a subcriber yet? Why the heck not?
By subscribing, you can...

  • Get CME hours
  • Support the show
  • Write it off on your taxes or get reimbursed by your department

Sign Up Today!


Subscribe Now

If you enjoyed this post, you will almost certainly enjoy our others. Subscribe to our email list to keep informed on all of the ED Critical Care goodness.

This Post was by , MD, published 3 years ago. We never spam; we hate spammers! Spammers probably work for the Joint Commission.


  1. Mike Jasumback says

    Finally! This needs to be widely publicized. Dan showed me the video a few years ago when we had him up for a Journal Club on post intubation managment. BE AFRAID of pulse ox. UNDERSTAND what it tells you. UNDERSTAND what the number mean and WHEN it meant that!
    How many people have we harmed by not understanding this.


    Mike Jasumback

  2. Jeff says

    Another excellent podcast Scott. Very useful information from Dr. Dan that has a lot of value in the pre-hospital setting where I work. Although I enjoy listening to all your podcasts, how can you ever get too much airway info??


  3. says

    Great stuff, gentlemen. The pulse oximeter is the world’s first actual time machine. It transports you 30-90 seconds into the past. When you bail on an attempt at 90%, don’t be surprised as you’re bagging effectively to see a nadir of 75% before it bounces back. So bail earlier.

  4. Minh Le Cong says

    thanks to you and Davis for this podcast!
    dont worry about Thumper..not everyone has to deal with Irukandji syndrome..but everyone has to deal with airway and ventilation in resuscitation!

    whilst not being perfect solutions to this issue, I did ponder that NODESAT a la Levitan, DSI a la you, and airway ultrasound a la Mike & Matt…might address some of the challenges cited in pulse ox latency and emergency airway mgt.
    for prehospital folks, NODESAT is very achievable with planning, and may minimise this very issue.whilst the debate may rage over whether to teach VL vs DL vs both to our residents, paramedics , nurses etc, we should all be adopting NODESAT as de rigeur
    which is the ideal RSI drug?
    Oxygen of course!
    one of my concerns about focus on VL intubations is it may detract from an overall airway strategy of safety, with neglect of preox because operators will rely upon a quick successful orotracheal intubation using their VL.

    • says

      Just taking your comments from a slightly different perspective. Most EMS providers can aim their vitals monitor at the eyes of intubator, this is usually not possible in the ED–a problem. Wouldn’t it be great if we could get the pulse ox number and ETCO2 waveform to appear on the video laryngoscope screen.

      • Minh Le Cong says

        actually thats very perceptive point
        most ED resus bay monitors Are positioned behind the intubator normally! all the wall suction and gas pipes are also behind.
        this is why I usually assign a dedicated staff member to monitor pulse ox and call it out regularly during RSI.
        because you can lose your situational awareness when the environment is designed for you to lose it!

        Du Canto had an idea to use HUD type goggles to display the monitor and VL feed simultaneously.
        interesting, but suspect only he could make it work for him!

          • says

            ah yes, monitors are cheap but where to place them is hard. can you send us a photo of your set-up. after making my comment yesterday, I plan to mount a stand-alone pulse-ox on our VL cart.

          • minh lecong says

            actually you gave me an idea spawned from aviation and an anaesthetic colleague

            in aviation there are terrain avoidance warning systems, typically, with an audible alarm announcing the height from ground and eventually an alarm voice instructing emergency flight maneuvers, like “pull up, pull up”

            what if the pulse ox had an option of the audible beep alarm, that you could set it to announce the pulse ox reading every 5 seconds or so..it would still go beep beep with the pulse but a soothing voice would announce the reading regularly, then at a critical level, it might announce “pull out, pull out” or “oxygenate patient”
            “cut the neck”

      • DocXology says

        We at least have an additional slave monitor mounted in front of the intubator in our resus cubicle. Something to consider when designing a
        new ED.

  5. says

    Great stuff — I especially liked the point made in the “squeezer” portion, because many people don’t understand the importance of releasing the bag (although the UPitt CCM guys have managed to use this phenomenon to create a kind of BVM “PEEP” effect — see first minute of http://www.ccmpitt.com/education/airway_course/videos/video2_2012.html).

    A few other random thoughts:

    1. Some of us have been using pulse ox (standard clip type) on ears forever. Awkward but it works and I did always imagine it’s somewhat more central than the finger. Thoughts?

    2. Many students seem puzzled by the fact that we “resat” faster than we “desat” on the pulse ox — i.e. they see how quickly the numbers rise after appropriate intervention, and thus assume that the reverse should be similarly brisk. Understanding the role of pulmonary reserve is important here.

    3. A related but separate topic would be photoplethysmograph lag — how long it takes a contraction of the heart to be reflected on the pulse ox waveform. This “Pulse Transit Time” is usually very short, under a quarter second… short enough that I’ve tried to measure BP using a manual cuff proximally and viewing the distal pleth waveform (tricky).

    4. Even less related but still interesting: bedside capillary glucometer lag. Stick a finger, how current is that measurement? The data says not very current in really sick and shocky people. I like a venous sample instead; even if it’s consistently 5-10 mg/dL off due to point-of-care glucometers being calibrated for cap blood, it’s far more “current” and reliable in acute patients. (Same reason the packaging says not to use a forearm whenever you’re in doubt; it’s even slower than the finger.)

    • says

      Somehow I missed these amazing comments. Bags with these duckbill valves are dangerous and horribly wrought. Problem with using them for CPAP is that unless you have the pressure gauge, no idea what CPAP the patient is getting and if the bagger likes to hold a little pressure between breaths, the patient can never exhale.

      • says

        Yep, it’s definitely a somewhat ghetto technique for creating PEEP, and I think quite skill-intensive — especially because it really requires a slow squeeze, fast release (to avoid impeding exhalation), then quick pop at the very end of expiry to “catch” the pressure. But you gotta be watching like a hawk to get the rhythm and know where the end is; too soon and they don’t exhale fully (and if you then bag in the same volume, you start stacking breaths), too late and you reach 0 pressure. More of a way to teach mastery with the device, I expect.

        I mainly find it interesting because in a prehospital setting at the BLS level (where I have yet to see a PEEP valve), it might be the only theoretical method for creating PEEP and hence oxygenating some of these profoundly shunted patients. (Except for partially taping over the exhalation port, which is probably an even worse idea.)

  6. says


    Most monitors allow a pulse-ox tone and if you choose tone modulation will vary the sound based on pulse ox decline. This is standard in ORs.

  7. minh lecong says

    I dont find the pulse ox tone decline to be sufficient a stimulus for most operators.
    thats why aviation systems employ a computer generated voice and have actual numerical call outs of approach altitude . its still not perfect but I find myself emulating this in RSI situations by assigning a role of the pulse ox steward.

    • DocXology says

      Tell everyone in the resus room to shut up when you are intubating. Actually you could apply that to most codes. I am told that the trauma director at The Alfred speaks softer and softer, the more critical circumstances get. Maybe its a prompt that there needs to more concentration and less noise.

  8. Seth Trueger @mdaware says

    Davis’s discussion of tidal volume & minute ventilation video seems to put hemodynamics at odds with lung protection. Should we be using 750mL Vt or 6-8mL/kg PBW?

  9. says

    I like the idea of the voice announcer for the pulse oximeter. It definitely goes some way to helping one regain situational awareness. Just thinking out loud, but id imagine there is not a standard port for all pulse oximeter probes. Could an iPhone app be used to watch the display using the camera, and then announce values at critical points…? Would this be useful in the ED/prehospital setting or would this just be an expensive toy which gets lost in the detritus of a resus?

    • says

      have tried this on myself and seems to work well for intubated patients. Forehead or tragus seems more realistic for patients you are going to intubate.

  10. says

    Great podcast! The first time I stumbled on this I was an R1 playing with on oximeter while waiting for my admission to come up, and trying to hold my breath and desaturate myself. I learned I had to exhale to as close to residual volume as I could and then gut it out, but even then, as I gasped for air, my sat would sit at about 97-98%. But wait a few more seconds, and watch it drop down – the “best” I managed was the mid 70’s – then come back up fairly quickly over 5-10 sec.

    Keep spreading the good word!


    ps disclaimer don’t try this at home!

  11. David Hersey says


    I think that having a goal EtCO2 and SPO2 is a great idea in most patients. However, it was my understanding that EtCO2 in patients with diffusion disorders (i.e. ARDS) or with increased dead spaces was problematic?

    I see many ICU patient’s whose ETCO2 is lower than their PaCO2.

    Thank you,
    David Hersey

  12. says


    Not sure where we are disconnecting. I think you are looking at the phrase in reverse. Maybe this would be helpful. P-40E-38=normal; P-40E-20=happens all the time in sick patients like the ARDS patient or patient with poor CO; P-100E-22=see this in COPD folks; P-20,E40=this is what you will not see. PaCO2 will be at least as high as ETCO2. Now some annoying sticklers will write in that they once saw a PaCO2 that was slightly lower than a simultaneous ETCO2 and while this is physiologically possible, esp in severe asthma. The difference is so small in these cases that it is not worth bothering about.

  13. Clint Kalan says

    Hey all,
    Just wondering if anyone is aware of any literature regarding safe oxygenation practices in the light of Dr. Davis’ discussion for those of us at high altitudes (7-10k feet or ~2130-3050 meters). Up here we consider an Sp02 of 95% fairly routine and normal. However, we also see patients with Hemoglobins of 17 and Hct of >50% and don’t blink. Any guesses on whether the relative polycythemia is going to give us some extra time or are we stuck having to be extremely expedient in our attempts at intubation. The 93% number is especially troublesome for my colleagues who work in the tourist towns of Colorado, where their patients have just stepped off the plane. On the other hand, with good preoxygenation (aided by waveform capnography), might we expect more reserve for us thick blooded mountain folks.

      • DocXology says


        The net improvement in total oxygen stores is only modest because the larger improvement in blood content is proportionately small compared to the effect of de-nitrogenation which contributes to >90% of the change i.e. it’s not going to buy you significantly more time.

        The more interesting question is whether mountain dwellers can tolerate greater degrees of hypoxia for longer periods of time without detriment. Factors such as ^2,3-DPG and better oxygen extraction may be factors to consider.

        From first principles in your example (Sao2 95%, paO2 80mmHg, Hb 17 g/DL

        Oxygen stores (room air – polycythaemic mountain dweller)

        FRC: 15% (venous O2 tension) x 3000ml = 450ml
        Hb stores: (1.34 x Hb x SaO2 + 0.023 x pO2) = 1.34 x 17 x 95% + 0.023 x 80mmHg = 23.48 ml/100ml = 1174 ml/5000ml
        Dissolved + Tissue bound (e.g. CPK) ~ 250ml

        TOTAL = 1874 ml

        Oxygen stores (100% O2 – polycythaemic mountain dweller)

        FRC: 100% x 3000ml = 3000ml
        Hb stores: 29.68ml/100ml = 1484 ml/5000ml
        Dissolved + Tissue bound (e.g. CPK) ~ 300 ml

        TOTAL = 4784 ml

        Oxygen stores (room air – Hb 15, sea dweller)

        FRC: 15% x 3000ml = 450 ml
        Hb stores: Hb stores: 22.4ml/100ml = 1120 ml/5000ml
        Dissolved + Tissue bound (e.g. CPK) ~ 250 ml

        TOTAL = 1820

        Oxygen stores (100% O2 – Hb 15, sea dweller)

        FRC: 100% x 3000ml = 3000ml
        Hb stores: 27ml/100ml = 1350 ml/5000ml
        Dissolved + Tissue bound (e.g. CPK) ~ 300 ml

        TOTAL = 4650ml

        • Clint Kalan says

          Makes perfect sense. I wonder if there’s any real difference here between type I and II reap failure up here. There’s some anecdotal stories of guys just consciously “breathing faster” to compensate for the fact that their body hasn’t accumulated enough CO2 to trigger their resp drive to keep up with oxygenation on climbs. Thanks for the resp phys numbers walk through!

          • DocXology says

            I am not sure what you are trying to say.

            At high altitude (and low FiO2), arterial hyperaemia becomes the dominant stimulus for alveolar ventilation.
            This results in a permanent reduction in paCO2 (compared to sea-level values)

            For those with lung disease, they will have a widened alveolar-arterial oxygen gradient. They will experience a greater reduction in paO2 for a given FiO2 when exposed to altitude. This may result in a greater ventilatory response to hypoxaemia and consequently they appear ‘to breathe faster’ compared to normal individuals. However, this is only an attempt at compensating for their poor oxygenation in the first place.

            The bottomline is you don’t want to be person with lung disease at altitude. And even 100% preoxygenated, they are going to experience lower oxygen tension levels, haemoglobin saturation and total oxygen content.

            If you need to brush up respiratory physiology and pathophysiology, I would recommend:

            Nunn’s Applied Respiratory Physiology (a core text for anaesthetist’s in this country).

  14. DocXology says

    The concept of ‘lag’ versus ‘lead’ monitors was emphasised to me by an anaesthesia professor I worked under. Both in the Australian Anaesthesia and ICU examination, equipment is a specific exam subject. One could argue that it should be the same for Emergency Physicians (particularly those who do a lot of resus/pre-hospital work). There is also some good stuff published by Prof Bill Runciman in the AIMS ( Australian incident monitoring system) study which looked at intra-operative incidents. The authors tried to find certain patterns and derive logical algorithms and mnemonics to deal with different scenarios e.g. acute desaturation or hypotension, loss of CO2 waveform, low/high pressure ventilation alarm. Not all of them are applicable to pre-hospital or EM work (e.g. seizures from hyponatraemia from glycine toxicity during TURP) but a lot of the principles are very useful in a crisis situation.


      • DocXology says

        The gist of it is there are four separate procedures depending on the urgency level – ‘SCARE’

        Scan = routine, regular systematic monitoring of all parameter, equipment and situation (the ‘idle’ situation)
        Check = provokes more active responses to new events
        Alert/Ready = when there is a major change in the situation
        Emergency = when the patient is peri-arrest/arrest

        The Scan procedure includes COVER ABCDA A mnemonic and a swift check.

        Within the ABCD mnemonic are further sub-procedures.

        The author reinforces the idea of regular:

        1) situational awareness and constant vigilance (before problems occur)
        2) defined actions in response to events according to their severity
        3) systematic review of possibilities when the problem is not immediately obvious (avoiding Eureka syndrome)

  15. Garrett, ER Tech says

    Hate to be the comment thread resuscitationist here, but here goes-
    Re: comments about a HUD or monitor visible from the HOB- this would be a great implementation of Google Glass, maybe a HUD that includes selected monitor parameters w/ a timer (also good for timing RSI meds)?
    Something tells me there’s some money to be made here…..


Speak Your Mind (Along with your name, job, and affiliation)