Cardiac Output Monitoring in Anesthesia and Critical Care

Principles of CO Measurement

Fick’s Principle

The Fick principle determines CO by comparing the body’s total oxygen consumption (VO₂) with the difference between arterial and mixed venous oxygen content (CvO₂).^1,2 The relationship is expressed as:

To calculate CO, the equation is rearranged as:

This reflects the law of conservation of mass: under steady-state conditions, the oxygen entering the body must equal the amount consumed plus the amount remaining in venous blood.^1,2

Components of the Equation

• CO: Volume of blood pumped by the heart per minute, expressed in L/min.^1,2
• VO₂: Whole-body oxygen use per minute, measured by comparing inspired and expired gases or estimated with formulas based on age, sex, height, and weight.^3,4
  • Arterial oxygen content (CaO₂): Determined by hemoglobin concentration, arterial oxygen saturation, and a small amount of oxygen dissolved in plasma. It is expressed using the formula:^1,3

• CvO₂: Oxygen content of blood sampled from the pulmonary artery, after systemic venous return has fully mixed.^1,3

Direct vs. Indirect Fick

• Direct Fick (Gold Standard): Measures VO₂ directly by collecting and measuring all inhaled and exhaled gases over a set period. Requires a closed respiratory circuit and a pulmonary artery catheter (PAC) for mixed venous sampling. While highly accurate, it is an invasive procedure and is usually reserved for research or critical cases that require precise measurements.¹
• Indirect Fick: Estimates VO₂ using demographic-based formulas (age, sex, height, weight) rather than direct gas collection. This approach is much simpler and more feasible at the bedside, but it introduces estimation error.¹

Clinical Relevance

• Fick’s method is considered the reference standard for determining CO and is often used as the benchmark against which other monitoring techniques are validated.^1,2 It is particularly valuable in settings where precise CO data are required, such as research studies, exercise physiology testing, and cases where other CO monitoring methods give conflicting results. Because the equation can be rearranged, it can also be useful for calculating other physiologic variables when the CO is already known.²

Advantages^1,2

• Highly accurate when VO₂ is directly measured
• Based on fundamental physiology (conservation of mass)
• Can provide additional calculated variables if CO is known

Limitations¹

• Requires invasive blood sampling (arterial line and PAC)
• Fick is not a continuous monitor; it gives a single measurement.
• Assumes steady-state VO₂ and stable hemoglobin levels

Thermodilution

• Although the direct Fick method remains the reference standard for measuring CO in research settings, the PAC thermodilution technique is considered the clinical gold standard.^5,6 The technique uses a balloon-tipped PAC that “floats” into the pulmonary artery, where it can measure pressures, sample mixed venous blood, and record temperature changes.³

Bolus Thermodilution

• A known quantity of chilled solution is injected into the right atrium or superior vena cava, and the thermistor at the catheter tip measures the downstream change in blood temperature.^3,6
• Interpretation: A larger temperature change suggests slower blood flow (lower CO), whereas a smaller change indicates faster flow (higher CO). This is because, in a slower flow, the colder fluid remains around the probe for a longer period, resulting in a larger temperature drop.
• Respiratory Averaging: Because right ventricular output varies with the respiratory cycle, multiple injections at different phases are averaged to improve accuracy.⁶

Limitations⁶

• Variations from normal forward blood flow can distort temperature-time curves.
• Tricuspid or pulmonary regurgitation causes backward flow and prolonged mixing of the cold injectate, producing a lower amplitude but prolonged decrease in temperature and therefore resulting in inconsistent and often underestimated calculations
• Intracardiac shunts allow premature mixing of the injectate, altering the temperature-time curve, which, depending on the direction of the shunt, can over or underestimate CO.
• Hypothermia or rapid temperature shifts alter baseline blood temperature and reduce measurement accuracy.
• Very low CO may paradoxically overestimate flow by broadening the temperature-time curve.

Risks⁵

• Arrhythmias can occur during injection, particularly with cold injectate.
• Complications may be seen in patients with:
  • Endocarditis
  • Intracardiac masses
  • Coagulopathy or thrombocytopenia

Continuous Thermodilution

• Continuous thermodilution uses a PAC with a heating filament that pulses small amounts of heat into the blood at set intervals. The thermistor senses the resulting temperature fluctuations, and the monitor averages them over 30–60 seconds to display a near-continuous CO trend.
  • Advantage: Allows ongoing CO assessment without repeated boluses
  • Limitation: The averaging introduces a time lag and may not accurately represent rapid hemodynamic changes in real time.
• Some sources report that continuous thermodilution “compares favorably” to bolus methods.⁵ In contrast, others note the delay in responsiveness and lack of proven outcome benefit as reasons it is considered comparable rather than superior.⁶

Transpulmonary Thermodilutions

• This method differs from PAC-based thermodilution by injecting cold saline into a large central vein but measuring the temperature change downstream in a large systemic artery (usually the femoral artery).³

Advantages⁶

• Avoids PAC placement → less invasive
• Averages over several respiratory cycles → reduces variability
• Transpulmonary systems can also calculate:
  • Global end-diastolic volume: direct preload marker
  • Extravascular lung water: helps distinguish cardiogenic vs. noncardiogenic pulmonary edema when compared to intrathoracic blood volume

Limitations⁶

• Subject to the same error sources as PAC thermodilution:
  • Intracardiac shunts
  • Regurgitant lesions
• Cannot directly measure pulmonary artery pressures
• Cannot obtain mixed venous blood samples

Dye Dilution

Principle

• Traditional dye dilution techniques estimate CO by injecting a known quantity of a dye, commonly indocyanine green, into a central venous catheter and detecting its concentration downstream in the arterial circulation. The changing dye concentration is plotted over time to produce a dilution curve, from which the CO is calculated.³
• Although once widely used, dye dilution has largely been replaced by newer methods due to practical limitations, including indicator recirculation, the potential for toxic dye accumulation, and the availability of more accurate, less cumbersome techniques, including thermodilution and lithium dilution. It is now cited primarily as a historical reference method, against which newer modalities are validated.^3,5

Lithium Dilution (LiDCO System)

Principle and Technique

• The lithium dilution method (LiDCO) applies the same underlying indicator dilution principle, using lithium chloride instead of dye. A small bolus of lithium is injected via either a central or peripheral vein, and an ion-selective arterial sensor measures the lithium concentration over time to generate a washout curve.^3,6
• For improved precision, several initial injections are often averaged during calibration. The system then combines this calibration data with pulse power analysis of the arterial pressure waveform to provide beat-to-beat estimates of CO and stroke volume (SV).^4,6
• Once calibrated, no additional lithium injections are needed unless the patient experiences significant hemodynamic changes.^4,6

Advantages^3,6

• Less invasive than PAC thermodilution
• CO values correlate well with PAC-derived measurements
• Can use peripheral venous access if central access is unavailable
• After calibration, it allows real-time beat-to-beat monitoring, providing real-time feedback.

Limitations^3,6

• Indicator recirculation or accumulation can distort the dilution curve
• Repeated injections may lead to lithium accumulation or toxicity
• Requires both arterial and venous access (still semi-invasive)
• Contraindicated in:
  • Patients taking lithium therapy
  • First-trimester pregnancy
  • High doses of neuromuscular blockers interfere with the lithium sensor.

Clinical Relevance

• The lithium dilution technique is especially useful in perioperative and critical care settings when precise CO data are needed but a PAC is not feasible. After initial calibration, the LiDCO system provides continuous, beat-to-beat hemodynamic data derived from the arterial waveform.⁴

Minimally Invasive and Noninvasive Modalities

Doppler Methods

Principle

• Doppler CO monitoring is based on the Doppler effect, in which the frequency of sound waves changes when either the source or the observer is in motion. An ultrasound probe emits waves that reflect off red blood cells; the change in the returning frequency corresponds to the velocity and direction of blood flow.³

Esophageal Doppler Technique

• A flexible ultrasound probe is positioned in the esophagus, oriented parallel to the descending thoracic aorta. The probe measures the velocity of blood cells moving through the descending aorta. Accuracy is highest when the beam is aligned with blood flow (near 0° angle) and decreases as the angle approaches 90°.^3,6

CO Calculation

• The monitor estimates the aortic cross-sectional area using nomograms derived from patient variables such as age, height, and weight. CO is then calculated as:
  • Flow = area × velocity, representing blood flow through the descending thoracic aorta.
  • To estimate total CO, the device applies a correction based on the assumption that roughly 70% of total CO passes through the descending aorta.^3,6,7
• Because this calculation relies on several assumptions, particularly in estimating the aortic area based on patient demographics, even small inaccuracies can result in significant errors.⁷

Advantages^3,6

• Minimally invasive and provides real-time, beat-to-beat data
• Correlates reasonably well with thermodilution-derived CO
• Useful for tracking trends in SV and guiding fluid management in perioperative or critically ill patients

Limitations^6,7

• Accuracy decreases when the aortic cross-sectional area or angle of insonation is not estimated correctly.
• Accuracy decreases in states with altered flow distribution (e.g., pregnancy, aortic cross-clamping) as the blood flow passing through the descending aorta deviates from assumptions in the calculation.
• Aortic diameter is dynamic and varies throughout the cardiac cycle, but algorithms assume a fixed value.
• Probe movement or malposition can distort signals and lead to inaccuracies.
• Not well tolerated in awake patients

Clinical Relevance

• Esophageal Doppler techniques demonstrate good correlation with PAC thermodilution for analyzing trends in CO, but they are less reliable for measuring absolute CO values. They can be valuable for dynamic assessment of SV and fluid responsiveness. However, their clinical role has decreased as newer noninvasive systems have become more widely adopted.^3,6

Impedance Cardiography

Principle

• Thoracic bioimpedance measures changes in electrical resistance across the chest to estimate CO. Electrodes placed on the thorax deliver a small, high-frequency, low-amplitude electrical current. Because blood conducts electricity better than surrounding tissues, fluctuations in thoracic impedance correspond to variations in blood volume during the cardiac cycle. The monitor analyzes these changes to calculate SV and, subsequently, CO.^3,6

Technique and Variants

• In the traditional bioimpedance method, surface electrodes are placed on the chest and neck. The system continuously measures impedance changes produced by cyclic blood flow in the great vessels, allowing real-time CO monitoring.³
• To improve accuracy, the bioreactance technique was developed. Like bioimpedance, it applies a high-frequency current through the thorax but additionally measures the phase shift of the electrical signal, not just the amplitude change. This phase information improves measurement precision compared to conventional bioimpedance.⁶

Advantages^6,7

• Completely noninvasive; requires only surface electrodes
• Provides continuous, real-time CO data
• Simple setup suitable for bedside or perioperative use

Limitations^6,7

• Accuracy is limited, especially in critically ill or hemodynamically unstable patients.
• Motion artifacts, electrical interference, and poor electrode contact can introduce error.
• Irregular cardiac rhythms (e.g., atrial fibrillation) disrupt measurements.
• Thoracic fluid (e.g., pulmonary edema, pleural effusions, pericardial tamponade) and obesity degrade signal quality.
• Even with bioreactance, results show wide variability compared to gold-standard techniques, such as thermodilution

Clinical Relevance

• Although bioimpedance and bioreactance systems offer a completely noninvasive means of estimating CO, their accuracy remains inconsistent in complex clinical settings. These methods are best suited for trend monitoring rather than absolute CO quantification. Despite continued refinements, they are not yet considered reliable replacements for established invasive modalities such as thermodilution or lithium dilution.^3,6,7

Pulse-Based Estimations

Arterial Pulse Waveform Analysis

Principle

• Arterial pulse waveform analysis estimates SV and CO by integrating the systolic portion of the arterial pressure waveform. The area under the curve represents the volume of blood ejected per beat, and CO is calculated as SV multiplied by heart rate. Because this calculation depends on vascular impedance (reflecting the resistance and compliance of the arterial system), each device applies proprietary algorithms to account for it. Calibrated systems determine impedance using an independent reference method (such as thermodilution or lithium dilution) and then use waveform analysis to follow trends over time. Uncalibrated systems, such as FloTrac (Edwards Lifesciences, Irvine, CA), estimate impedance based on patient demographics and hemodynamic variables, and then rely on waveform analysis to monitor ongoing changes.^3,5,6

Technique

• A standard arterial line, most commonly in the radial artery, is connected to a specialized pressure transducer. The monitor continuously analyzes the waveform and averages data over multiple cardiac cycles (typically 10–20) to smooth out beat-to-beat variability. Calibrated systems require periodic recalibration when vascular tone or compliance changes significantly to maintain accuracy.^3,6

Physiologic Considerations

• Arterial pressure is influenced by both the volume ejected by the heart and the elasticity of the vessels. Real arteries are not rigid conduits but rather elastic reservoirs that expand during systole and recoil during diastole. Because the elasticity or compliance of the arteries cannot be measured directly, it is estimated from patient characteristics such as age, sex, height, and weight, and incorporated into the device’s algorithms in order to calculate blood flow from the pressure waveform.^5,6

Derived Parameters

• Most systems also provide indices such as SV variation (SVV) and pulse pressure variation (PPV), which predict fluid responsiveness in mechanically ventilated patients. Some include both systolic and diastolic phases to calculate cardiac power, an estimate of the total mechanical energy generated per cardiac cycle.^3,6

Noninvasive Pulse Contour Techniques

• The same analytical principles can be applied without an arterial catheter using volume clamp plethysmography (finger cuff technique). In this method, a finger cuff dynamically adjusts pressure to counteract arterial pulsations, while infrared photoplethysmography measures changes in blood volume. The pressure required to maintain constant blood volume in the finger mirrors the arterial pressure, allowing the system to reconstruct the waveform noninvasively. Another technique, applanation tonometry, uses a pressure sensor to flatten the radial artery and capture its waveform in a similar way. Both methods can provide continuous CO estimates but are susceptible to signal distortion from motion, low perfusion, or vascular stiffness.^5,6

Advantages^3,5,6

• Uses an existing arterial line; less invasive than pulmonary artery catheterization
• Provides continuous CO and SV trend data in real time
• Generates dynamic indices (SVV, PPV) for guiding fluid administration
• Plethysmography and tonometry permit completely noninvasive monitoring.

Limitations^3,5,6

• Waveform damping: improper damping calibration can distort the arterial waveform, leading to over- or underestimation of SV
• Both calibrated and uncalibrated systems are less accurate during periods of unstable hemodynamics; calibrated systems require repeated recalibration, whereas uncalibrated models rely on assumptions that become unreliable under such conditions, such as sepsis or vasopressor therapy.
• Conditions that disrupt the normal pressure-flow relationship, such as arrhythmias, the use of intra-aortic balloon pumps, or aortic regurgitation, render the results unreliable.

Clinical Relevance

• Pulse contour analysis provides a practical balance between invasiveness and data quality. It is especially useful in perioperative or ICU settings where continuous CO trends and dynamic indices guide fluid and vasopressor management. However, because accuracy depends heavily on waveform quality and stable vascular tone, this method is best suited for trend monitoring rather than absolute CO measurement.^3,5,6

SV and Fluid Responsiveness

Principle

• SV represents the volume of blood ejected by the left ventricle with each heartbeat, and CO is the product of SV and heart rate. According to the Frank–Starling mechanism, SV rises with increasing preload until the myocardium reaches its optimal stretch point. The steep portion of the Frank-Starling curve represents preload-responsive states, where small increases in RAP yield large increases in CO, while the flat portion reflects limited preload reserve, where further fluid loading produces minimal benefit.^3,5

Rationale for Dynamic Indices

• Traditional static preload markers, such as central venous pressure (CVP) or pulmonary capillary wedge pressure (PCWP), do not correlate well with true volume status because they fail to account for changes in ventricular compliance or intrathoracic pressure. Dynamic indices, derived from heart–lung interactions during positive pressure ventilation, offer a more reliable assessment of fluid responsiveness.^5,6

Physiologic Basis

• When a patient is on mechanical ventilation, during inspiration, the rise in intrathoracic pressure temporarily reduces venous return and right ventricular output while simultaneously decreasing left ventricular afterload, promoting a brief increase in left ventricular SV. After several beats, the reduction in right ventricular filling propagates to the left side, lowering left ventricular preload and output during expiration. These cyclical variations in SV form the physiologic foundation of systolic pressure variation (SPV), PPV, and SVV. In essence, by observing how SV changes with these cyclic preload shifts, clinicians can assess ventricular preload responsiveness, allowing them to identify where the patient lies on the Frank–Starling curve.^5,6

Dynamic Parameters

Each of these indices reflects a different aspect of preload dependence.

• SPV: Measures the total respiratory swing in systolic arterial pressure. A widened SPV, especially with a large “down” component, suggests hypovolemia and high preload reserve.^5,6
• PPV: Represents the percentage change in pulse pressure (systolic–diastolic) over a single respiratory cycle. PPV >13% generally indicates fluid responsiveness, whereas <9% suggests preload independence.^3,6
• SVV: Directly quantifies the change in ejected SV over the respiratory cycle. An SVV above 10–13% usually predicts a positive volume response.^3,5
• SPV was the first of these dynamic indices to be described, measuring direct systolic pressure swings during the respiratory cycle. It provides a simple visual marker of preload responsiveness when more sophisticated analysis is unavailable. However, PPV and SVV tend to be more quantitative and reproducible, which helps reduce observer variability and improves precision. PPV is most practical at the bedside because it requires only an arterial line waveform, whereas SVV is obtained from advanced hemodynamic monitors that calculate actual SV per beat. Both PPV and SVV typically change in parallel, since pulse pressure normally tracks SV. However, they may diverge in cases when vascular tone is altered, for example, during vasopressor use or marked changes in vessel compliance, because PPV reflects both flow and vascular stiffness, while SVV reflects flow alone.^3,6

Interpretation and Limitations

• A PPV or SVV greater than 13% generally indicates that the patient is fluid responsive, whereas values below 9% suggest preload independence and that additional volume is unlikely to augment CO. However, values between 9–13% fall into a “gray zone,” where fluid responsiveness is uncertain and should be interpreted in the context of other physiologic indicators such as urine output, tissue perfusion, and lactate trends.⁶
• Several physiologic and technical factors can also limit reliability:
  • Arrhythmias or spontaneous breathing efforts: disrupt the regular relationship between intrathoracic pressure and venous return
  • Low tidal volume or high positive end-expiratory pressure ventilation: reduces the magnitude of respiratory-induced preload variation, which can mask responsiveness.
  • Opening the thoracic cavity during surgery or altered lung compliance: changes the transmission of airway pressure to the vasculature
  • Use of vasopressors or severe vasoconstriction: alters arterial tone, causing PPV and SVV to diverge and measurements to become unreliable
  • Peripheral vascular disease or stiffness: exaggerates pressure changes without corresponding SVV
• Because these parameters depend on controlled positive-pressure ventilation and regular cardiac rhythm, PPV and SVV are not reliable in spontaneously breathing patients or those ventilated with low tidal volumes, which are commonly used settings for critically ill patients.^3,5,6
• Please see the OA summary on fluid responsiveness for more details. Link

Noninvasive Dynamic Monitoring

• Plethysmographic Variability Index (PVI) applies the same basic idea as pulse contour analysis but does so noninvasively using the pulse oximeter waveform. It looks at how much the infrared light signal from the pulse oximeter rises and falls with each breath during positive pressure ventilation. These cyclic changes reflect how SV and pulse pressure vary with respiration, providing an indirect estimate of fluid responsiveness.
• PVI is appealing because it requires only a standard pulse oximeter sensor; however, its accuracy can be affected by factors that alter the signal, such as motion, cold extremities, poor perfusion, or bright ambient light. In unstable or vasoconstricted patients, its correlation with invasive PPV and SVV measurements declines.^3,6

Clinical Use and Limitations

• Dynamic indices are particularly helpful in mechanically ventilated, hemodynamically stable patients where fluid responsiveness is uncertain and excessive fluid administration carries risk. They’re often used to inform fluid therapy in perioperative and ICU settings. However, their usefulness declines in spontaneously breathing patients, those with arrhythmias, or those receiving lung-protective ventilation strategies, where the physiologic assumptions underlying respiratory variation no longer hold. In these cases, clinicians should rely on a broader hemodynamic assessment to guide management.^3,5,6