Oxygen Flow Rate and FiO2: Understand the Relationship!

Oxygen, we all need it! We do not need a lot of it under normal circumstances, with 0.21 being the fraction of inspired oxygen (FiO2) of room air. FiO2 is defined as the concentration of oxygen that a person inhales.  The air that we inhale on a day to day basis is made up of 21% of oxygen, 78% of nitrogen and 1% of trace elements such as argon, carbon dioxide, neon, helium and methane. For the purposes of this article, fractions and percentages will be used interchangeably for ease of explanation. 

Sometimes, 21% of oxygen may not be enough to maintain adequate oxygen saturations. In these situations, supplemental oxygen can be administered via various oxygen delivery devices ranging from nasal prongs to invasive ventilation. This allows the concentration of oxygen to be increased, potentially increasing the FiO2 to 100%.


In settings outside of critical care areas, FiO2 has historically not received much attention. But things are changing! In standard hospital settings these days, there is an increasing use of humidified high flow oxygen therapy that requires an understanding of the relationship between oxygen flow rate and FiO2. In most clinical areas that require an FiO2 to be documented, you will be able to find a table that outlines an approximate correlation between oxygen flow rate and FiO2, similar to the table below:


It is all well and good to memorise that for every 1 L/min increase of oxygen flow rate, the FiO2 increases by 4%. But it would be better to understand WHY the FiO2 increases in those specific increments! So let’s discuss that…

flowmeter-medical-oxygen-15-lpm_CIG515800_exlargeMy first question for you is this: what is the FiO2 of the air you are breathing right now?

If you said 21%, excellent!

My second question for you is this: what is the FiO2 of the oxygen being delivered through the oxygen flow meter as soon as you turn it on?

And this is where people start saying “it depends on the oxygen flow rate”. Despite this being true when we are discussing the FiO2 that the person is inhaling, that is not actually the question that I asked.

Therefore, my third question for you is this: does the oxygen flow rate really change the FiO2 of the PURE oxygen that is being delivered through the oxygen flow meter?

The answer is NO! The oxygen flow meter is connected to either a bottle of oxygen or a medical wall supply of oxygen. This oxygen is PURE, it is 100% oxygen! Therefore, anything that comes out of that oxygen flow meter has an FiO2 of 100%. Consider the following:

If I have the oxygen flow rate set at 1 L/min, I will have 1 L/min of 100% oxygen…

If I have the oxygen flow rate set at 5 L/min, I will have 5 L/min of 100% oxygen…

If I have the oxygen flow rate set at 10 L/min, I will have 10 L/min of 100% oxygen…

If I have the oxygen flow rate set at 15 L/min, I will have 15 L/min of………………….?

If you said 100% oxygen, excellent!

So my fourth question for you is this: why does the table above show different FiO2 values corresponding with these oxygen flow rates that we have just said is always 100% because it is pure oxygen?


This is the point that people start scratching their heads, shrugging their shoulders and backing away slowly while avoiding eye contact with me. Hang in there! The lightbulb will go off very shortly!

The answer to this question comes down to the flow requirements of the patient! What do I mean by that? You are currently breathing air in and out of your lungs while you are reading this blog, hopefully with enough interest to share it with your friends and colleagues after you finish reading it *wink wink*. The air that you are breathing has to get from point A (the atmosphere) to point B (your lungs). If a car was trying to get from point A to point B, it can only do this if you press the accelerator to achieve a certain speed. The faster the speed, the faster you get from point A to point B. The same principle applies to how we breathe, but we refer to this speed as our peak inspiratory flow.

Our normal peak inspiratory flow tends to range between 20 – 30 L/min. Our respiratory muscles are comfortable and do not tire when we breathe at a normal respiratory rate with this peak inspiratory flow. Now consider what your breathing does when you go for a run; or if you are allergic to running like me, imagine what your breathing does! Asides from your respiratory rate increasing, you start sucking in for more air. You are trying to get the air from point A to point B faster, which means that your peak inspiratory flow requirement has increased. The same goes for a person that is “struggling to breathe” or has an “increased work of breathing”, they have a high peak inspiratory flow requirement.

So back to patient flow requirements and FiO2…

If you are breathing in normally at a peak inspiratory flow rate of 30 L/min at room air with an FiO2 of 21%, you can easily calculate the average FiO2 you are breathing in an almost redundant formula:

30 x 21 = 630%
630 ÷ 30 = 21%

Now consider you are receiving 10 L/min of oxygen via a face mask at an FiO2 of 100%. You still have a normal peak inspiratory flow rate of 30 L/min, but 10 L/min if being blown in your face via the face mask. Therefore, you still need another 20 L/min to meet your inspiratory flow requirements. Where are you going to get this from? You are going to suck it in from the surrounding atmosphere with an FiO2 of 21%. So let’s apply the same formula as before:

(10 x 100) + (20 x 21) = 1420%
1420 ÷ 30 = 47%

However, if you had an increased peak inspiratory flow rate of 50 L/min but were still only receiving 10 L/min of oxygen via a face mask at an FiO2 of 100%:

(10 x 100) + (40 x 21) = 1840%
1840 ÷ 50 = 37%

Or a decreased peak inspiratory flow rate of 20 L/min while receiving 10 L/min of oxygen via a face mask at an FiO2 of 100%:

(10 x 100) + (10 x 21) = 1210%
1210 ÷ 20 = 60%

In the above examples, nothing changed with the oxygen flow rate being delivered to the patient. The only thing that has changed was the patient’s inspiratory flow demand and how much that “diluted” the pure oxygen being delivered with the FiO2 of 21% found in room air. If the flow rate being delivered to the patient is greater than their peak inspiratory flow rate, they have no reason to have to suck in atmospheric air and “dilute” the pure oxygen. Consider sticking your head out the car window as you are driving at the maximum legal speed. All that air that is blown in your face makes it a lot easier to breathe, it reduces the effort required to suck in the air. f7223692c3a9a623e370f25174e83915So if you were breathing with a normal peak inspiratory flow rate of 30 L/min but were receiving ≥ 30 L/min of pure oxygen via a high flow oxygen delivery device, you do not need to suck in any more air from the surrounding atmosphere and would therefore be receiving an FiO2 of 100%.

Unless the flow rate being delivered to the patient is more than their peak inspiratory flow demand, it is impossible to know what the patient’s exact FiO2 is because you do not know their exact peak inspiratory flow. The tables utilised to outline a relationship between oxygen flow rate and FiO2 are based on mere estimations of normal peak inspiratory flow rate, ranging between 20 – 30 L/min.

So let’s take this one step further and discuss the practical application of understanding oxygen flow rate and FiO2. As discussed in the blog post titled Respiratory Failure: Type 1 or Type 2, you can have a patient that has a problem with oxygenation or a patient that has a problem with ventilation. If your patient has a problem with oxygenation, they require a higher FiO2 to aid with this. In most settings, this is achieved by turning up the oxygen flow rate in order to subsequently increase the FiO2. If your patient has a problem with ventilation, they require a higher flow rate to aid with this. If we are aiming to set a flow rate higher than their inspiratory flow demand, it is not ideal to use just pure oxygen and deliver an FiO2 of 100% to someone that may not even have an oxygenation problem. They may only require an FiO2 of 21% with a higher flow rate, that can be achieved with a high flow air meter. Or the patient may require something in between these two extremes, which can be achieved with a dual flow adaptor that utilises both an oxygen and an air meter.


For example, 15 L/min of oxygen at an FiO2 of 100% and 15 L/min of air at an FiO2 of 21% to give a total of 30 L/min of flow at a diluted FiO2 of 60%. Or perhaps 15 L/min of oxygen at an FiO2 of 100% and 30 L/min of air at an FiO2 of 21% to give a total of 45 L/min of flow at a diluted FiO2 of 47%. The world is your oyster! Devices such as the AIRVO 2, do all of the above calculations for you. All you need to do is dial up how much total flow you want to set for your patient, and increase the oxygen flow meter to achieve the desired FiO2 to maintain adequate oxygen saturations.

So next time you are looking after that asthmatic patient who is sucking in air like their lives depend on it (pardon the nursing humour), consider making their breathing easier by giving them some additional flow! Imagine how much easier it would be for them to breath in if instead of having to make all of the effort to suck in the air, they had some of that air blown into their face? And next time you are looking after that patient with suboptimal oxygen saturations, do what we always do and turn up the oxygen! 

Also remember the following:

  1. If your patient has a problem with oxygenation, they need more FiO2
  2. If your patient has a problem with ventilation, they need more flow
  3. If you patient has a problem with oxygenation AND ventilation, they need more FiO2 AND flow


University of Colorado Denver. (n.d). Rules on oxygen therapy. Retrieved from: http://www.ucdenver.edu/academics/colleges/medicalschool/departments/medicine/intmed/imrp/CURRICULUM/Documents/Oxygenation%20and%20oxygen%20therapy.pdf

If you liked this post, why not subscribe to Blogging for your Noggin? Be one of the first to know when new content becomes available! I promise there will be no annoying spam! 

Enter your email address to follow this blog and receive notifications of new posts by email.



38 thoughts on “Oxygen Flow Rate and FiO2: Understand the Relationship!

  1. I’m thinking about this for use with mild hyperbaric oxygen therapy (1.3atm pressure). I have seen on other sites that a person at rest will breath 7.5LPM. So then would a person at rest in a mhbot chamber be receiving 100% FiO2 if the oxygen concentrator is set to 8.5LPM?

    Thank you!

    Liked by 1 person

  2. Loved the article as I was looking for some information regarding this process. I am interested from a Medical Gas design standpoint. I am looking to revise some of my design criteria when sizing medical air or oxygen lines that will feed a ventilator. The issue comes into play when I have to account for a ventilator. I need to know will they be doing invasive or noninvasive ventilation, what flow rates will they be doing (15 lpm max?), will they be using oxygen connection on the wall or medical air, or does the unit use the medical air compressor in the unit and connect oxygen. I am looking for some guidelines to use when designing a medical gas system which to support these flow rates. Thank you.


    • Hi Jared,

      I understand that FiO2 is technically a fraction, therefore 1.0 when talking about pure oxygen. For ease of explanation, I have discussed this in terms of oxygen concentration (but having people understand the relationship to FiO2). I have amended the introduction to discuss this interchangeable use of fractions to percentages. Thank you for taking the time to comment.

      Kind regards,




    Liked by 1 person

  4. It’s nice. Can u tell me the approximate FiO2 thru ETT on a patient with Ambu bagging and not on ventilator and we are giving 15lit/min thru bag.

    Liked by 1 person

    • Hi Muhammad. Sorry about the late response. If you are bagging your intubated patient, you have a closed circuit delivering an FiO2 of 1.0 (100% concentration) to the patient. Remember, any flow of oxygen coming from the wall is pure oxygen with FiO2 of 1.0 (100% concentration). If we are not diluting it with room air, which we aren’t with a patient that is intubated with an ETT, they will be receiving an FiO2 of 1.0 (100% concentration).

      Hope this helps!

      Liked by 1 person

      • You have provided a brilliant article on the flow rate of oxygen. I was searching the net for something meaningful on it and couldn’t find, until I came across your blog. Many thanks for sharing…

        Liked by 1 person

    • BiPAP is different in terms of FiO2 because you can dial this up in the machine. The total flow is also different as the machine will deliver the required flow to meet the peak inspiratory demands of the patient. You can see the flow being delivered on the flow graphics of your BiPAP machine. Because the peak inspiratory flow is matched and you have a relatively closed system with the BiPAP mask, the FiO2 is accurate for what is programmed in – in that the patient does not dilute it further by sucking in air from the atmosphere. Your IPAP determines the peak inspiratory pressure and thereby augments tidal volume and aids work of breathing, while your EPAP splints the alveoli open at end expiration to aid in oxygenation.

      Hope this helps!


    • I think we should keep in mind the purpose of flow and that is to reduced the work of breathing by satisfying the inspiratory demand increasing the fio2 is not all ways necessary to improve perfusion.


      • Hi OldSchool,

        Definitely true, as mentioned in the comment reply above. We always need to be considering ventilation requirements and oxygenation requirements as two separate issues that work synergistically together.

        Kind regards,



  5. Thanks a lot for the answer! It was very helpful 🙂
    I mean ipratropium inhaler first and then salbutamol inhaler. Thanks for letting me know.
    I think the idea of saline nebs is that they turn mucous thiner and that will help to mobilize them, isn’t it?
    Have you done any post/document comentting differences between CPAP and BIPAP?
    Thank you very much.


    Liked by 1 person

    • I believe the use of saline nebs has originated from paediatric management of bronchiectasis. They use hypertonic saline which works on the principle of osmosis to draw fluid into the lungs, which loosens mucous and makes it easier to clear. However, 0.9% normal saline does not have this hypertonic effect. It is similar to pouring water over oil, it just slides straight over the mucous.

      I have replied to another comment in the post titled: The Importance of Counting Respiratory Rate with some information related to CPAP and BiPAP. I think the differences between CPAP and BiPAP and practical applications of use would be an interesting blog post; so watch this space. If you haven’t already, sign up with your email address to the blog and you’ll be one of the first to know when new blog posts become available 🙂


  6. So, does it mean that if a person has a high respiratory rate, that’s related with an increased peak inspiratory flow, if we use a venturi mask of 40% or a venturi mask with another FiO2 that doesn’t mean that the actually FiO2 that the person is taking is 40% because obviously he/she’s got an increased peak inspiratory flow, right? Is that the reason why DPOC patients may have low O2 sat’s with 3l/min but have good O2 Sat’s with a BIPAP without O2? Thanks a lot, Joana

    Liked by 1 person

    • Hi Joana! Thank you for your question!

      You are absolutely right in your statement about the Venturi devices in relation to peak inspiratory flow. Venturi devices tend to use the Bernoulli principle to entrain air to provide the patient with approximately 40 L/min of flow. If this is more than the patient’s inspiratory flow demand, then the FiO2 that is associated with how you have dialled up the device will in fact be accurate. However, if the peak inspiratory flow demand of the patient is greater…the FiO2 can no longer be considered accurate and would probably be less than what we have “dialled” up due to the dilution effect.

      In terms of improvement in saturations on BiPAP, you need to remember that we are dealing with a closed system (if you have a good seal with your BiPAP mask) and an open system (with the venturi device or low flow nasal prongs/mask). Therefore, you are right in your thinking that the 3L/min would be substantially diluted if your patient has a high peak inspiratory demand (think 3L/min of 100% and 57L/min of 21% as an example). Furthermore, the ability to provide PEEP in order to splint open the alveoli to aid in gas exchange is drastically better with BiPAP compared to venturi/high flow devices. You are also dealing with pressure support/IPAP that will take the peak inspiratory pressure in the lungs to the set value above PEEP, thereby further improving the oxygenation potential of the alveoli stretched open. The closed system with BiPAP also reduces the work of breathing, that reduces metabolic demand, that reduces oxygen consumption which in turn will stabilise/improve your oxygen saturations.

      The final difference in saturations between dry oxygen at 3L/min compared to BiPAP could be attributed to the heated/humidified system. In order for oxygen to be able to diffuse across the alveoli, it needs to be fully saturated which usually occurs at a temperature of 37 degrees. Despite our bodies normally being able to achieve this, a heated/humidified system almost gives this process a jump start…if that makes sense?

      I hope this has answered your question! Please feel free to ask any other questions you may have!


      • Thank you very much for your comment! It was a very insteresting answer.
        Now I understand better why the effect of a nebulizer with ipatropium and/or salbutamol can make a huge difference in a person with exacerbation of asthma or DPOC. Tha’s because the low O2 sat’s it’s related with ventilation and not oxygenation right? So opening their airways will improve ventilation a lot. Would you be able to explain why nebulizers act and their differences against inhaler devices? Also can you explain me why ipratropium has to be administered first than salbutamol (in inhaler devices)? It can be an idea for a next post maybe. 🙂 In other day, in order to reduce hiperkalemia in a patient without any venous access, I was asked to administer 10UI of rapid isulin + 2 formulas of glucose, resin and 5mg of salbutamol nebulizer. I would like to understand why that will decrease the potassium levels in the blood.
        I hope it’s not too much questions. If you answer just one i’ll be already happy 🙂

        Thanks a lot,


        Liked by 1 person

      • Great questions Joana 🙂 Let me do my best to answer them for you!

        The relationship between low oxygen saturations and ventilation problems such as asthma or COAD is a little bit more complex than opening up the airways to help ventilation. It’s all to do with how much space there is for oxygen within the alveoli if there is too much carbon dioxide taking up space there instead. The following article that I have recently shared on my facebook page explains this concept really well: http://www.prehospitalpush.com/2016/02/14/low-spo2-o2-problem-or-co2-problem/

        The improved ability to exhale without as much resistance following bronchodilator therapy helps clear more CO2. This in turn allows more room within the alveoli for oxygen to fill and thereby diffuse across into the bloodstream.

        In terms of nebulizers versus inhaler devices, there have actually been a lot of studies comparing the two. The research based on children with asthma actually shows a better effect with an inhaler (with a spacer attached) compared to a nebuliser. In the adult population, the evidence shows there is no difference between the two in terms of outcomes. But for some reason, clinicians feel like nebulisers work better. Maybe the same way we incorrectly believe that normal saline nebulisers actually independently mobilise mucous? It is not actually reflected in the evidence, but sometimes we see something working and we feel that it is actually better.

        I’ll have to look up more evidence on your question regarding ipratropium needing to be given before salbutamol; I’ve always been use to given them simultaneously for more efficacy. The only reason I can think of right now would be onset of action; ipratropium takes around 15 minutes to work while salbutamol takes less than 5 minutes. Therefore, using ipratropium first would mean you give it more time to work in the body while the salbutamol works at a faster rate…but that is only a guess…

        In terms of hyperkalemia, insulin and salbutamol can be utilised as temporary measures to reduce the levels in the blood stream. Remember, we have A LOT more potassium in our intracellular compartments than in our extracellular compartments. It is the extracellular compartment that we can measure via a blood test and what causes issues within our body if the potassium level is too high/low. The aim of both insulin and salbutamol is to push the potassium from the extracellular space back into the intracellular space. Insulin does this in a similar manner to how it carries glucose from the extracellular space into the cells; it allows potassium to bind to it and acts as a transport vehicle back into the cells. Of course, insulin will also drop the blood glucose level of a patient simultaneously. The glucose that was given acted to counteract this effect and maintain a normal blood glucose level. Salbutamol is a beta 2 agonist that activates the sodium-potassium pump, thereby increasing the swap over of potassium from the extracellular space with sodium within the intracellular space. Resin should be one of the last resorts according to current research, as it can be potentially harmful to the patient with unproven efficacy. But the theory is that it binds to potassium within the colon for excretion. Because potassium is bound to the resin, it cannot be reabsorbed back into the blood stream.

        *Phew* I can now take a breath haha! I hope this answers the questions that you have asked and made it a little clearer for you to understand 🙂


    • Great question!

      If you have access to a machine that can provide Bilevel Positive Airway Pressure (BiPAP), absolutely! Where Continuous Positive Airway Pressure (CPAP) only provides positive end expiratory pressure (PEEP) that has more of an effect on splinting the alveoli open in order to improve the oxygenation capacity, BiPAP has PEEP as well as pressure support. This may be known to some as expiratory positive airway pressure (EPAP) and inspiratory positive airway pressure (IPAP), respectively.

      The pressure support that you get with BiPAP can be titrated for what the patient requires, more pressure support if the patient has a higher inspiratory flow demand. I would normally recommend starting with a pressure support of 5 to get the patient use to it, then up to 10 as stock standard. You may need a higher pressure support for a patient that is really gasping for air.

      I would then educate the patient that “breathing in is going to become a lot easier now, but they have to focus on breathing out”. Otherwise, you start to get the blowfish effect in that not all the air that they breathe in comes out. Also the higher your pressure support gets, the more it is going to push the mask off your patient resulting in an air leak…so just be aware of that as you are titrating up to meet the inspiratory flow demand of your patient. And then you can set the FiO2 on the machine to whatever is required to meet the oxygenation demands of your patient.

      A handy trick that I have used in the past to hold a patient over while setting up the BiPAP machine is an air viva. If they can, I get the patient to hold it themselves with the mask to their face and tell them to press the air viva every time they need to take a breath. It substantially increases flow delivery to the patient to ease the ventilatory demand, and they can press it as softly or as hard as they like! Plus, you can use it not connected to any oxygen which will be an FiO2 of 21%, or you can attach it to oxygen and titrate to whatever the patient’s oxygen saturations require.

      I hope this helps with your question! Please feel free to ask any other questions you may have!


Do you have a question? Or a suggestion for a new topic?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s