Fick Cardiac Output Calculator

Cardiac output from oxygen consumption and the arteriovenous oxygen content difference — the reference method, and the one whose weak link is the oxygen consumption rather than the equation.

Last updated 8 Oct 2026

Fick cardiac output

VO₂ ÷ a-v O₂ difference

Oxygen consumption VO₂ (mL/min)Measured if you have a metabolic cart. If it is assumed instead, this is the number that decides the answer: the flat assumption is 125 mL/min/m² at rest and 110 mL/min/m² from age 70, so a 1.9 m² adult gets 238 mL/min. See the formula block for the LaFarge–Miettinen equations and for how far assumed VO₂ actually misses.

Haemoglobin (g/dL)Haemoglobin appears in both content terms, so it cancels almost exactly out of the content DIFFERENCE. Anaemia barely moves the Fick output at a fixed pair of saturations — and devastates oxygen delivery, which is a different quantity.

Arterial oxygen saturation SaO₂ (%)From a co-oximeter on an arterial sample, not from the pulse oximeter.

Arterial pO₂ (mmHg)Only the dissolved oxygen term depends on this, and it is small: about 0.29 mL/dL of roughly 17. It is kept because it does not cancel out of the arteriovenous difference, where it contributes about 3%.

Mixed venous oxygen saturation SvO₂ (%)A TRUE mixed venous sample, drawn from the pulmonary artery. A central venous saturation from a superior vena caval line is not the same measurement: it misses inferior caval and coronary sinus return entirely, and substituting it puts an unquantified error straight into the denominator of this equation.

Mixed venous pO₂ (mmHg)From the same pulmonary arterial sample.

5.14L/minExample

VO₂ 250 mL/min, Hb 13 g/dL, SaO₂ 97%, PaO₂ 95 mmHg, SvO₂ 70%, PvO₂ 40 mmHg

The equation is sound; the VO₂ usually is not

CO (L/min) = VO₂ (mL/min) ÷ [ (CaO₂ − CvO₂) × 10 ]
CaO₂ = 1.34 × Hb × SaO₂/100 + 0.003 × PaO₂
CvO₂ = 1.34 × Hb × SvO₂/100 + 0.003 × PvO₂

VO₂: oxygen consumption in mL/min. This is the weak link and the reason the page exists. The equation is an identity — in a steady state the oxygen leaving the lungs must equal the oxygen the tissues take up — so the output is only as good as the consumption figure put into it
× 10: converts the content difference from mL O₂ per dL to mL O₂ per litre, so that it matches a cardiac output in litres per minute
1.34: Hüfner’s constant, the millilitres of oxygen one gram of fully saturated haemoglobin binds. The same figure the site’s arterial oxygen content calculator uses; 1.36 and 1.39 are also published and the choice shifts content by under 4%. It appears in both content terms, so it largely cancels out of the difference
assumed VO₂, flat: 125 mL/min/m² at rest, falling to 110 mL/min/m² from age 70. Multiply by body surface area. For a 1.9 m² adult that is 238 mL/min
assumed VO₂, LaFarge and Miettinen: the age- and rate-dependent form, reproduced in the DICOM oxygen consumption equation table. Men: VO₂ = BSA × (138.1 − 11.49 × ln(age) + 0.378 × HR). Women: VO₂ = BSA × (138.1 − 17.04 × ln(age) + 0.378 × HR). For a 1.9 m² man of 65 at 80 beats per minute that is 1.9 × (138.1 − 47.96 + 30.24) = 229 mL/min
how far assumed VO₂ misses: in 155 patients of mean age 75, cardiac index by estimated Fick was out by 20% or more against thermodilution in 40.6% of patients using LaFarge–Miettinen, 26.5% using Dehmer and 36.1% using Bergstra, with limits of agreement as wide as −1.38 to +0.53 L/min/m². The authors concluded that estimated Fick cannot replace thermodilution in an individual patient. An estimated-Fick output is a calculation, not a measurement
mixed venous means the pulmonary artery: the venous sample must be drawn distal to the point where superior caval, inferior caval and coronary sinus blood have mixed, which is the pulmonary artery. A central venous saturation from a superior caval line is a different measurement with a different normal value, and the direction of the discrepancy is not fixed
the dissolved term: 0.003 mL O₂ per dL per mmHg. It contributes under 2% of arterial content and about 3% of the arteriovenous DIFFERENCE, because the arterial and venous tensions are far apart. Dropping it inflates cardiac output by roughly that much

Worked example

VO₂ 250 mL/min, Hb 13 g/dL, SaO₂ 97%, PaO₂ 95 mmHg, SvO₂ 70%, PvO₂ 40 mmHg
CaO₂ = (1.34 × 13 × 0.97) + (0.003 × 95) = 16.897 + 0.285 = 17.18 mL O₂/dL
CvO₂ = (1.34 × 13 × 0.70) + (0.003 × 40) = 12.194 + 0.120 = 12.31 mL O₂/dL
Arteriovenous difference = 17.18 − 12.31 = 4.87 mL O₂/dL, which is 48.7 mL O₂ per litre of blood
250 ÷ 48.7 = 5.14 L/min
Now assume the VO₂ instead. At 125 mL/min/m² a 1.9 m² adult is given 237.5 mL/min, and the output becomes 237.5 ÷ 48.68 = 4.88 L/min — 5% lower, from a number nobody measured
Halve the haemoglobin to 6.5 g/dL at the same saturations: the content difference falls to 2.52 mL/dL and the output becomes 9.93 L/min. That is not a Fick artefact — with half the carrying capacity and the same extraction fraction, the same consumption genuinely requires twice the flow
Narrow the saturation gap to 97% and 92%: the difference collapses to 1.04 mL/dL and the output becomes 24.13 L/min. The arithmetic is unchanged and the answer is meaningless — which is the warning about small arteriovenous differences, not a reason to trust the figure

The three routes to a VO₂, and what each one costs

Being more elaborate did not make the estimate better: in Kresoja and colleagues’ 155 patients the age-and-rate equation missed more often than the flat multiple of body surface area. The lesson is not to prefer one equation but to stop calling an estimated-Fick output a measurement.
Route	What it gives	Error it carries
Measured, metabolic cart	The patient’s actual oxygen consumption	The reference. Needs the equipment, a stable patient and no leak around the circuit
Flat assumption, 125 mL/min/m²	238 mL/min at 1.9 m²; 110 mL/min/m² from age 70	Error of 20% or more in 26.5% of 155 elderly patients using the Dehmer form (Kresoja et al., 2019)
LaFarge and Miettinen, by age, sex and heart rate	BSA × (138.1 − 11.49 × ln age + 0.378 × HR) in men; the age coefficient is 17.04 in women	Error of 20% or more in 40.6% of the same cohort — the worst of the three equations tested, despite using the most inputs

Reference ranges, and the two cards that print them

These are population figures from reference cards, not limits. Note that the cards’ own ranges are mutually consistent only because they share a lineage — where two independent sources exist, as for vascular resistance, they disagree substantially.
Figure	Range	Source
Cardiac output	4 to 8 L/min	Edwards Lifesciences reference card; identical on the WVU SICU card
Cardiac index	2.5 to 4.0 L/min/m²	Both cards, and the figure to read instead of cardiac output whenever body size is unusual
Mixed venous saturation, SvO₂	about 70 to 75%	The resting extraction fraction of roughly a quarter, as set out on this site’s oxygen extraction ratio page
Arteriovenous O₂ difference	about 4 to 5 mL O₂/dL at rest	Follows arithmetically from a normal content of about 17 mL/dL and an extraction fraction near a quarter

Why the Fick method is the reference and still gets the wrong answer

The Fick principle is almost an identity. In a steady state the oxygen the lungs add to the blood must equal the oxygen the tissues remove from it, so flow equals consumption divided by the arteriovenous oxygen content difference. There is no model in it, no calibration constant and no assumption about the shape of a waveform, which is why it has been the comparator against which every other way of measuring cardiac output is judged for a century.

The trouble is the numerator. Measuring oxygen consumption needs a metabolic cart and a tight circuit, and most catheter laboratories do not have one, so consumption is assumed from body surface area — 125 mL per minute per square metre, or a regression equation that adds age, sex and heart rate. Those assumptions were derived in cohorts that look nothing like a ventilated, sedated or elderly patient, and the error goes straight through the division. In 155 patients of mean age 75, cardiac index by estimated Fick was out by a fifth or more against thermodilution in a quarter to two fifths of patients depending on the equation used, and the elaborate equation did worse than the flat multiple.

The denominator has its own trap, and it is not arithmetic. The venous sample must be genuinely mixed, drawn from the pulmonary artery after superior caval, inferior caval and coronary sinus blood have come together. A saturation from a central venous line in the superior vena cava misses the inferior caval contribution altogether, and the gap is neither small nor reliably in one direction — substituting it adds an unknown offset rather than noise. The other trap is a narrow arteriovenous difference: dividing by a small number amplifies the error in it, so in a high-output septic patient with a venous saturation in the high eighties the Fick output is arithmetically exact and clinically worthless.

Two things are worth noticing in the arithmetic itself. Haemoglobin appears in both content terms and cancels almost completely out of the difference, so anaemia barely moves the Fick output — while halving oxygen delivery, which is the quantity the tissues care about. And the dissolved oxygen term, negligible in arterial content, contributes about 3% of the arteriovenous difference because the two tensions are far apart; dropping it inflates the output by roughly that much. Every figure here is derived from other measurements, so it carries their errors as well as its own. A derived index is never more reliable than the least reliable number that went into it. A derived haemodynamic number is read alongside the patient — the history, the perfusion, the lactate, the trend across serial measurements — and never instead of them. It supports a clinician’s judgement rather than replacing it.

Frequently asked questions

What is the Fick equation for cardiac output?

Cardiac output in L/min equals oxygen consumption in mL/min divided by the arteriovenous oxygen content difference in mL/dL times ten. The contents are calculated as 1.34 × Hb × saturation/100 plus 0.003 × pO₂, arterially and from a mixed venous sample. At a VO₂ of 250 mL/min and a content difference of 4.87 mL/dL the output is 5.14 L/min.

How accurate is estimated Fick cardiac output?

Not accurate enough to treat as a measurement in an individual patient. In 155 patients of mean age 75, cardiac index by estimated Fick differed from thermodilution by 20% or more in 40.6% of patients using the LaFarge–Miettinen VO₂ equation, 26.5% using Dehmer and 36.1% using Bergstra, with limits of agreement up to −1.38 to +0.53 L/min/m². The authors concluded that estimated Fick cannot replace thermodilution in elderly patients.

Can I use a central venous saturation instead of a mixed venous one?

Not in this equation without saying so. A true mixed venous saturation comes from the pulmonary artery, where superior caval, inferior caval and coronary sinus blood have mixed. A superior caval ScvO₂ misses the inferior caval return entirely, the discrepancy is not a fixed offset, and it lands directly in the denominator. ScvO₂ is a useful trend monitor in its own right — see this site’s oxygen extraction ratio page — but substituting it silently turns a reference method into an estimate.

What assumed VO₂ should I use?

Whichever one you are prepared to name on the report. The flat figure is 125 mL/min/m² at rest, falling to 110 mL/min/m² from age 70. The LaFarge–Miettinen form adds age, sex and heart rate: BSA × (138.1 − 11.49 × ln age + 0.378 × HR) in men, with 17.04 as the age coefficient in women. There is no best choice — the elaborate equation performed worse than the flat multiple in the one head-to-head cohort cited here.

Why does anaemia hardly change the Fick cardiac output?

Because haemoglobin multiplies both the arterial and the venous content term, so it very nearly cancels out of the difference. At fixed saturations, halving the haemoglobin halves the content difference and the calculated output doubles — which is the correct physiological answer, not an artefact: with half the carrying capacity the same oxygen consumption needs twice the flow. What anaemia does change directly is oxygen delivery, where haemoglobin does not cancel.

Related calculators

References

Fick Cardiac Output. In: StatPearls. NCBI Bookshelf NBK606091. Gives the relation as “Cardiac output = Tissue oxygen consumption/(Arterial oxygen content − Venous oxygen content)”, the assumed consumption of “125 mL/min/m2 at rest” and “110 mL/min/m2” from age 70, and that “mixed venous pulmonary artery oxygen content is used to measure venous blood oxygen content”.
Kresoja K-P, Faragli A, Abawi D, et al. Thermodilution vs estimated Fick cardiac output measurement in an elderly cohort of patients: a single-centre experience. PLOS ONE. 2019;14(12):e0226561. 155 patients, mean age 75.1 ± 6.8 years; limits of agreement as wide as −1.38 to 0.53 L/min/m²; “a rate of error ≥20% occurred with the equations by LaF, De or Be in 40.6%, 26.5% and 36.1% of patients”; “CI-eFM cannot replace CI-TD in elderly patients”.
DICOM Content Mapping Resource, CID 3664 Oxygen Consumption Equation/Table. Reproduces LaFarge–Miettinen as “VO2male = BSA (138.1 − 11.49 * loge(age) + 0.378 * HRf)” and “VO2female = BSA (138.1 − 17.04 * loge(age) + 0.378 * HRf)”. The 1970 paper itself was not reachable from here; this is a reproduction of its coefficients.
Edwards Lifesciences. Normal Hemodynamic Parameters and Lab Values (EU master reference card). Cardiac output 4–8 L/min, cardiac index 2.5–4 L/min/m², stroke volume 60–100 mL/beat, stroke volume index 33–47 mL/m²/beat, SVR 800–1200 and SVRI 1970–2390 dyn·s·cm⁻⁵ (·m²), PVR 100–250 dyn·s·cm⁻⁵, MAP 70–105 mmHg, CVP 2–6 mmHg, PAWP 6–12 mmHg; prints MAP as “[SBP + (2 x DBP)]/3” and SVR as “MAP-RAP x 80/CO”.
Leach RM, Treacher DF. The pulmonary physician in critical care 2: oxygen delivery and consumption in the critically ill. Thorax. 2002;57(2):170–7. The source this site’s oxygen content and oxygen delivery pages already use for Hüfner’s constant and the delivery relation.

Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/