Cardiac output (Fick principle)

Clinical background

Cardiac output (CO) is central to right heart catheterisation for the classification of pulmonary hypertension, the assessment of heart failure and prioritisation for transplantation and mechanical circulatory support. Thermodilution is the most commonly used method at the bedside, but it presupposes that a bolus indicator passes through the pulmonary artery without being diluted by regurgitant flow across the tricuspid valve. In significant tricuspid regurgitation, very low flow or intracardiac shunting, thermodilution has been considered unreliable, and the Fick principle is then the alternative used.

The Fick principle in its direct form, in which VO₂ is measured with a metabolic cart, is the accepted gold standard for determining CO. In clinical practice, direct measurement of VO₂ is often unavailable and an assumed (estimated) VO₂ is therefore used. It is precisely this assumption that is the method's weakest point: when VO₂ is estimated incorrectly, CO becomes incorrect, and since VO₂ sits in the numerator, the estimation error is amplified in the result.

Calculating cardiac output (Fick principle)

The calculator computes CO as:

$$\text{CO}\ (\text{L/min}) = \frac{125 \times \text{BSA}\ (\text{m}^2)}{1{.}36 \times \text{Hb}\ (\text{g/dL}) \times 10 \times (S_{aO_2} - S_{vO_2})}$$

where $S_{aO_2}$ is the arterial and $S_{vO_2}$ the mixed venous O₂ saturation, expressed as fractions (0–1; entered as percentages in the calculator and converted internally). The constant 1.36 is the oxygen-carrying capacity of haemoglobin (mL O₂ per g Hb) and the factor 10 converts g/dL to g/L. VO₂ is estimated as 125 mL/min/m² × body surface area. Cardiac index (CI) is obtained as CO/BSA (L/min/m²).

The derivation of estimated VO₂ belongs to LaFarge and Miettinen (1970), based on patients undergoing cardiac catheterisation [1]. The three commonly used estimation formulae for VO₂ (LaFarge–Miettinen, Bergstra, Dehmer) were all derived in patients undergoing cardiac catheterisation, largely children and young adults with congenital heart disease, 30 to 60 years ago [2]. The simplified constant of 125 mL/min/m² is a crude approximation of resting VO₂ in a stable adult patient without acute illness.

Interpretation in practice

The normal resting cardiac index is 2.5–4.0 L/min/m². The table below gives interpretation bands and clinical action.

In pulmonary hypertension, a low CI places the patient in a higher risk category and influences the choice of vasodilator and consideration of transplantation. In heart failure, decisions on transplantation and mechanical support are guided by CI combined with PCWP.

A traditional indication for the Fick method has been significant tricuspid regurgitation, in which thermodilution has been considered to give too low a value. A 2024 systematic review challenges this: across four studies with a total of 602 patients (258 with TR), the correlation between thermodilution and direct Fick was r = 0.90 in moderate to severe TR and r = 0.86 in mild or no TR, that is, not significantly different [3]. Heterogeneity was, however, high (I² 71 and 68% respectively) and the risk of bias greater in older studies, so the finding should be confirmed in a modern cohort.

Validation and performance

The assumed VO₂ is the dominant source of error. Palanques-Tost et al. evaluated three established eFick formulae against thermodilution as the reference in a modern intensive care cohort [2]. In a postoperative cardiothoracic ICU, the mean error (MAPE) was 30% and the coefficient of determination R² was negative (−1.5), meaning that the model performed worse than simply guessing the mean. The error arose because the static VO₂ estimate could not track the dynamic physiology of critically ill patients. Their CORE model, which incorporated dynamic physiological variables, achieved a MAPE of 14% and R² of 0.58.

Narang et al. found in a clinical cohort of 535 patients undergoing right heart catheterisation that measured VO₂ differed significantly from all three estimation formulae, with greater deviation in severe obesity (BMI >40) [4]. Chase et al. reported in a cohort of patients with heart failure and reduced ejection fraction that all formulae produced substantial error and misclassification of haemodynamic categories, including cardiogenic shock and hypoperfusion [5]. In summary, eFick is most reliable in stable adult patients without acute illness and without severe obesity — that is, the population resembling the derivation cohort.

Limitations

The assumed VO₂ does not apply in:

• Critical illness. VO₂ in intensive care is dynamic and is influenced by sympathetic tone, vasopressors and sedation, which violates the steady-state condition the Fick principle presupposes [2,3].
• Severe obesity (BMI >40). The estimation formulae overestimate VO₂, which gives too low a CO [4].
• Extremes of age. The formulae were derived in children and young adults with congenital heart disease [2].
• Intracardiac shunting. A shunt requires separate sampling from the superior and inferior vena cava rather than a mixed venous sample from the pulmonary artery; otherwise the arteriovenous difference is miscalculated [3].

Common errors:

• Preferring eFick over thermodilution solely because of tricuspid regurgitation. The systematic review shows that thermodilution is not necessarily unreliable in TR [3].
• Relying on a single value in unstable physiology. CO in critical illness is not static; repeated thermodilution measurements may be more informative [2,3].
• Assuming that 125 mL/min/m² applies to all patients. In fever, sepsis and hypermetabolic states VO₂ is raised; in sedation and hypothermia it is reduced.

When clinically important decisions rest on CO and thermodilution is unreliable, VO₂ should be measured directly rather than estimated [2,4].

References

1. LaFarge CG, Miettinen OS. The estimation of oxygen consumption. Cardiovasc Res. 1970;4(1):23–30. PMID: 5416840
2. Palanques-Tost E, et al. Cardiac output estimation in the intensive care unit. JACC: Advances. 2025. PMID: 40286350
3. Abualsaud R, et al. Time to calm the Fick down? A systematic review and meta-analysis of thermodilution compared to direct Fick in tricuspid regurgitation. CJC Open. 2024. PMID: 39525819
4. Narang N, et al. Inaccuracy of estimated resting oxygen uptake in the clinical setting. Circulation. 2014;129(2):203–10. PMID: 24077170
5. Chase PJ, et al. Comparison of estimations versus measured oxygen consumption at rest in patients with heart failure and reduced ejection fraction who underwent right-sided heart catheterization. Am J Cardiol. 2015;116(11):1724–30. PMID: 26443561