What FiO2 on a ventilator actually sets, and how fast to wean it

FiO2 on a ventilator is the fraction of inspired oxygen the machine delivers, settable anywhere from 21% to 100% on a full-range machine. Not every transport ventilator is one: in a bench study by Savary and colleagues, the turbine transport ventilators ran 21% to 100%, while the pneumatic ones offered either two fixed steps, 70% or 100%, or a 40% to 100% Venturi range. The number that matters isn’t what gets dialed in at intubation. It’s how fast that number comes back down. This page covers what a given FiO2 value usually represents, how to wean it without chasing the saturation monitor, and which SpO2 target you’re actually weaning toward, since more than one legitimate target exists and the protocol in front of you decides which applies. It’s one dial. The initial ventilator settings walkthrough runs the whole sequence around it.

What is FiO2 on a ventilator?

Fraction of inspired oxygen: the share of the delivered gas that’s oxygen. Room air runs 21%. A ventilator blends that up to 100%. Written as a decimal (0.50) or a percentage (50%), it’s the same number two ways, which causes most of the confusion. FiO2 is what you set. On its own it says nothing about how the patient is actually doing with it.

What does 50 mean on a ventilator?

An FiO2 of 0.50: half of what the ventilator is delivering is oxygen. Some screens show it as 50%, others as 0.50; it’s the same setting either way. It sits next to tidal volume, rate, and PEEP as one of the few numbers you set directly on the vent.

How do you calculate FiO2 from liters per minute?

A ventilator sets FiO2 directly and never needs this estimate; it answers a different question, what a nasal cannula flow is roughly delivering before a patient reaches the vent. Each added liter adds roughly 4%, so 2 L/min runs about 28% and 6 L/min about 44%. The conventional table only runs 1 to 6 L/min on a standard low-flow cannula, not HFNC and not a ventilator circuit, and the same chapter says the estimate is least accurate below 5 L/min. The StatPearls chapter behind it carries an Archived banner, and measured studies find a smaller real effect, closer to 2.5% per liter, because the patient’s own breathing pattern changes how much of that flow actually gets inhaled.

What is a normal FiO2 on a ventilator?

There isn’t one. 21% to 100% is the whole settable range on a full-range ventilator, and where a patient sits on it says more about their lungs than about a single normal value. The table below states what each commonly cited value actually represents, and who’s willing to source it.

Three of these four values carry a reference chapter, StatPearls twice and the Merck Manual once, and the 21% floor comes from a peer-reviewed bench study of ventilators. Normal ventilator values, parameter by parameter covers the rest of the vent screen the same way.

Why FiO2 starts at 100% after intubation

Most patients default to 100% FiO2 immediately after intubation, then get titrated down, per StatPearls’ EMS ventilator chapter. That default exists because the apneic period just spent whatever reserve the patient had, and nobody is fine-tuning oxygen while confirming tube placement. The same chapter is just as direct about why it doesn’t stay there: prolonged exposure can run into hyperoxemia, which may carry its own long-term harm. The RSI sequence that put the tube in is what actually decided how much reserve was left by the time the vent took over.

Is 100% oxygen bad for a ventilated patient?

Not for the few minutes around intubation, which is exactly what it’s for. Sustained high FiO2 is the separate problem, and the anchors in Hochberg’s AJRCCM review are mostly animal: laboratory animals died of progressive respiratory failure above 0.7 sustained for three to six days, and preclinical work found biological changes below 0.6 in three hours or less. The human data are shorter and softer. Healthy volunteers breathing 100% oxygen developed cough, substernal discomfort and a drop in vital capacity between hours 12 and 16, which did not happen at 50%. The mechanism is reactive oxygen species. A related but distinct problem, absorption atelectasis, comes from nitrogen washout: oxygen replaces the nitrogen splinting alveoli open, so some collapse. The 0.6 line is Merck’s long-standing convention, not a measured cliff.

How to wean FiO2 without chasing the number

Come down to the lowest FiO2 that holds the SpO2 target, in steps, rechecking between them, instead of tracking the saturation number minute to minute.

1. Confirm the saturation is actually stable and the waveform is real before touching anything.
2. Step the FiO2 down.
3. Give it time to show on the monitor.
4. Stop where the target holds.

Two things worth saying plainly. A saturation that won’t hold on a falling FiO2 is a finding to work up, not a dial to spin back up and forget: the oxygen goes back up while the reason gets found, and the reason is as often the tube, the circuit, secretions, a pneumothorax, perfusion, or the probe as it is the lung; the DOPES walkthrough runs that order. And if it does get spun back to 100%, the crew that did it should say so at handoff, along with what it was on before.

Do you drop PEEP or FiO2 first?

Neither drops first by rule. When the goal is a lower FiO2, PEEP is the dial that gets you there: raising or holding it recruits alveolar surface, so oxygenation improves without more oxygen, which is literally how a unit buys room to bring FiO2 down, per the Merck Manual. Dropping PEEP while the FiO2 is still high hands back the recruited lung that was holding the saturation. The ARDS Network card walks the two down together on a stepped ladder, with its tables marked as not required, rather than moving either alone; the ARDSNet protocol settings card covers that table. What PEEP does on a ventilator covers the other half of this pair.

Do you wean FiO2 on a short transport?

No published source this page’s research found answers that question for the transport window. NAEMSP’s position statement says prehospital ventilation should be disease-specific and mirror in-hospital best practices, with no oxygen-titration bullet of its own, and StatPearls’ prehospital chapter says to titrate FiO2 quickly to the minimum that holds the target. So a stable saturation on a short run is not a reason to park at 100%, and it is not a reason to chase the number in the back of a moving truck either. What travels well: a stable saturation, a real waveform, a documented FiO2, and a handoff that includes the trend.

Which SpO2 target are you weaning to?

Two different targets exist for two different clinical pictures, and your protocol decides which one governs, not this page. The BTS oxygen guideline sets 94-98% for acutely ill adults not at risk of hypercapnic respiratory failure. The ARDS Network protocol card sets a lower goal, 88-95% or a PaO2 of 55-80 mmHg, for mechanically ventilated patients on its lung-protective protocol specifically, not as a general oxygen-therapy target. Patients at risk of hypercapnic respiratory failure carry their own separate, lower target, set by guideline and local protocol; COPD ventilator settings and hypercapnic risk covers that group.

FiO2 is what you set. The P/F ratio is how it scores.

FiO2 alone tells you nothing about the patient, only what you’re delivering. What the patient is doing with it comes from the saturation and the blood gas, and the P/F ratio turns that gas into one number: PaO2 in mmHg divided by FiO2 as a decimal. A PaO2 of 80 mmHg on an FiO2 of 0.50 gives 160. The Berlin definition classifies ARDS severity by that ratio, measured on at least 5 cm H2O of PEEP: mild at 201-300, moderate at 101-200, severe at 100 or below.

The free vent settings calculator runs tidal volume and rate math next to this same oxygen picture, one screen instead of six.

Sources

1. Kuhl EA, Perera TB. Prehospital Mechanical Ventilation. StatPearls; 2024. NBK537072. https://www.ncbi.nlm.nih.gov/books/NBK537072/
2. O’Driscoll BR, Howard LS, Earis J, Mak V. BTS Guideline for Oxygen Use in Adults in Healthcare and Emergency Settings. Thorax. 2017;72(Suppl 1):ii1-ii90. https://www.brit-thoracic.org.uk/clinical-resources/guidelines/emergency-oxygen/
3. Semler MW, Casey JD, Lloyd BD, et al. Oxygen-Saturation Targets for Critically Ill Adults Receiving Mechanical Ventilation (PILOT). N Engl J Med. 2022;387(19):1759-1769. https://www.nejm.org/doi/full/10.1056/NEJMoa2208415
4. Patel BK. Overview of Mechanical Ventilation. Merck Manual Professional Edition; 2026.
5. Hochberg CH, Semler MW, Brower RG. Oxygen Toxicity in Critically Ill Adults. Am J Respir Crit Care Med. 2021;204(6):632-641.
6. NIH NHLBI ARDS Clinical Network. Mechanical Ventilation Protocol Summary (protocol card).
7. ARDS Definition Task Force, Ranieri VM, et al. Acute Respiratory Distress Syndrome: The Berlin Definition. JAMA. 2012;307(23):2526-2533.
8. Fuentes S, Chowdhury YS. Fraction of Inspired Oxygen. StatPearls; 2022. NBK560867 (archived).
9. Saeed F, Lasrado S. Extubation. StatPearls; 2023. NBK539804.
10. Baez AA, Qasim Z, Wilcox S, et al. Prehospital Mechanical Ventilation: An NAEMSP Position Statement and Resource Document. Prehosp Emerg Care. 2022;26(sup1):88-95.
11. Mackle D, Bellomo R, Bailey M, et al. Conservative Oxygen Therapy during Mechanical Ventilation in the ICU (ICU-ROX). N Engl J Med. 2020;382(11):989-998.
12. Schjorring OL, Klitgaard TL, Perner A, et al. Lower or Higher Oxygenation Targets for Acute Hypoxemic Respiratory Failure (HOT-ICU). N Engl J Med. 2021;384(14):1301-1311.
13. Savary D, Lesimple A, Beloncle F, et al. Reliability and Limits of Transport-Ventilators to Safely Ventilate Severe Patients in Special Surge Situations. Ann Intensive Care. 2020;10:166. https://pmc.ncbi.nlm.nih.gov/articles/PMC7724620/