Cardiac index is a measure of how much blood your heart pumps per minute, adjusted for your body size. It takes cardiac output, the total volume of blood ejected by the heart each minute, and divides it by your body surface area, yielding a number in liters per minute per square meter. That single number helps clinicians compare heart performance across people of very different builds, which raw cardiac output alone cannot do. The value sits at the center of decisions in critical care, heart failure management, and surgery, but it has blind spots that even experienced clinicians can underestimate.
How Cardiac Index Differs from Cardiac Output
Cardiac output is straightforward: it is the volume of blood your heart pushes out every minute, typically somewhere around 4 to 8 liters in a resting adult. The trouble is that a cardiac output of 4 L/min might be perfectly adequate for a small person and dangerously low for a large one. Cardiac index solves this by dividing cardiac output by body surface area, a calculated estimate of your total skin area based on your height and weight. The result is a size-adjusted number that lets doctors compare a 50-kilogram woman to a 100-kilogram man on a more level playing field.
In healthy resting adults, cardiac index generally falls between roughly 2.5 and 4.0 L/min/m². One large imaging study using cardiac MRI found mean values of about 3.2 in men and 3.1 in women, with no meaningful difference between sexes.1Journal of Cardiovascular Magnetic Resonance. Cardiac output and cardiac index measured with cardiovascular magnetic resonance in healthy subjects, elite athletes and patients with congestive heart failure A cardiac index at or below 2.2 L/min/m² is widely used as the threshold for “low,” a signal that the heart may not be meeting the body’s needs.
How It Is Measured
There is no single way to measure cardiac index, and the method chosen depends on the clinical situation. In intensive care and cardiac catheterization labs, the most common approach is thermodilution via a pulmonary artery catheter (often called a Swan-Ganz catheter). A small bolus of cold saline is injected into the bloodstream, and a sensor downstream measures how quickly the temperature changes, which reveals how fast blood is flowing. The Fick method is the traditional gold standard: it calculates cardiac output from how much oxygen the body consumes and the difference in oxygen content between arterial and venous blood.2PubMed Central. Comparison of Accuracy of Estimation of Cardiac Output by Thermodilution Versus the Fick Method Using Measured Oxygen Uptake
Both methods have quirks. A study comparing thermodilution to direct Fick measurements in over 300 patients found that thermodilution tended to underestimate cardiac output at rest and overestimate it during peak exercise, with wide limits of agreement especially under exertion.3PubMed Central. Cardiac Output During Exercise: Thermodilution Versus Direct Fick This matters because decisions in the cath lab or ICU hinge on whether the number crosses key thresholds. If the method itself introduces a few tenths of a liter of error in either direction, a borderline patient could be classified differently depending on which technique was used.
Less invasive options exist. Echocardiography can estimate cardiac output by measuring blood velocity through the heart’s outflow tract with Doppler ultrasound. Whole-body bio-impedance devices send a small electrical current through the body and infer blood flow from changes in resistance. One study found good correlation between bio-impedance and echocardiographic readings, though echocardiography tended to read higher at higher workloads, likely because flow acceleration near the heart’s outflow tract can inflate Doppler estimates.4PubMed. Non-invasive measurement of cardiac output by whole-body bio-impedance during dobutamine stress echocardiography: clinical implications in patients with left ventricular dysfunction and ischaemia Cardiac MRI can also quantify cardiac output non-invasively, offering high accuracy, though it is expensive and not available at the bedside.1Journal of Cardiovascular Magnetic Resonance. Cardiac output and cardiac index measured with cardiovascular magnetic resonance in healthy subjects, elite athletes and patients with congestive heart failure
Why It Matters in Heart Failure and Cardiogenic Shock
Cardiac index is one of the cornerstones of heart failure classification. The Forrester hemodynamic subsets, a framework used since the 1970s, sort patients into four groups based on two variables: whether the cardiac index is low (the heart is not pumping enough) and whether filling pressures are high (fluid is backing up). These categories guide treatment: a patient with low cardiac index and high filling pressure needs a very different drug strategy than one with adequate pumping but congestion alone.5PubMed Central. Central venous pressure and pulmonary capillary wedge pressure: fresh clinical perspectives from a new model of discordant and concordant heart failure
In cardiogenic shock, where the heart abruptly fails to deliver enough blood, cardiac index becomes a minute-to-minute tracking tool. Pulmonary artery catheters measure it alongside pressures in the heart’s chambers and oxygen saturation in the mixed venous blood, giving clinicians a real-time hemodynamic profile.6PubMed Central. Management of cardiogenic shock: a narrative review Over 90% of cardiogenic shock patients receive at least one drug to boost blood pressure or heart contractility, and the cardiac index helps determine whether those drugs are working.7PubMed Central. State of Shock: Contemporary Vasopressor and Inotrope Use in Cardiogenic Shock
The relationship between cardiac index and survival is not perfectly linear, though. One study of 40 patients in cardiogenic shock found that 35% died with a persistently low cardiac index (below 2.2 L/min/m²), but 45% died even after their cardiac index had normalized above 2.2.8CHEST. Hemodynamic Profile of Cardiogenic Shock Normalizing the number on paper does not guarantee that tissues are actually receiving adequate oxygen, a limitation worth understanding in its own right.
Predicting Long-Term Outcomes
Beyond the immediate crisis, cardiac index carries prognostic weight over months and years. A study of patients with symptomatic heart failure found that low cardiac index was an independent predictor of long-term cardiac events, and that combining cardiac index with right atrial pressure created a clear risk stratification: patients with both a low index and elevated right-sided pressure fared the worst.9PubMed Central. Cardiac index predicts long-term outcomes in patients with heart failure This kind of stratification helps clinicians decide who needs aggressive intervention, such as a ventricular assist device or transplant evaluation, versus who can be managed with medication adjustments.
That said, cardiac index is just one piece of the puzzle. Some prognostic models for heart failure transplant evaluation have found that non-invasive variables like exercise capacity and clinical history performed as well as or better than catheterization-derived hemodynamics, including cardiac index, for predicting survival.10PubMed. Development and prospective validation of a clinical index to predict survival in ambulatory patients referred for cardiac transplant evaluation Cardiac index adds value, but it is not a standalone crystal ball.
How Cardiac Index Changes with Age
Your heart does not pump at the same rate throughout your life. The MRI study mentioned earlier found a gradual decline in cardiac index with age in healthy adults: about 8 mL/min/m² per year, dropping from an average of roughly 3.3 in people in their twenties to about 3.0 in those over 60.1Journal of Cardiovascular Magnetic Resonance. Cardiac output and cardiac index measured with cardiovascular magnetic resonance in healthy subjects, elite athletes and patients with congestive heart failure A scoping review focused on older adults confirmed that healthy people over 60 typically show a mean cardiac index between 2.1 and 3.2, with some studies reporting a yearly decline and others not.11PubMed. The normal cardiac index in older healthy individuals: a scoping review
In children, the decline is steeper. One imaging study tracking both cerebral and cardiac blood flow from infancy to adulthood found that cardiac index dropped at about 0.11 L/min/m² per year in children under 18, then slowed to roughly 0.02 L/min/m² per year in adults.12PubMed Central. Imaging Age‐Related Changes of Normal Cerebral and Cardiac Blood Flow in Children and Adults Aged 7 Months to 61 Years This makes intuitive sense: young children have high metabolic rates relative to their size, and their hearts pump proportionally more blood. As the body grows and metabolism per kilogram settles down, so does the cardiac index.
These age-specific norms matter clinically. Applying a single “normal” cutoff of 2.5 L/min/m² to an 80-year-old might flag a perfectly healthy person as having borderline low output, while missing a genuinely struggling 25-year-old whose index has dropped from 3.4 to 2.7.
Special Populations Where the Number Can Mislead
The formula for body surface area was developed decades ago using data from people of average builds. When a patient falls at the extremes, the math can distort clinical interpretation. In people with severe obesity, widely used body-surface-area formulas can inflate the denominator so much that the cardiac index appears artificially low even when raw cardiac output is perfectly adequate. A mathematical analysis warned that in these patients, cardiac and stroke indices can be misleading, and cautioned against using standard index thresholds to trigger treatment.13Journal of Cardiothoracic and Vascular Anesthesia. Effects of Body Surface Area-Indexed Calculations in the Morbidly Obese: A Mathematical Analysis For a clinician in the ICU, this means that a cardiac index of 2.0 in a patient with a BMI of 55 may not carry the same alarm as the same number in someone with a BMI of 25.
Young children pose a different challenge. The relationship between body weight and body surface area is not linear in small bodies. One study of hemodynamic indexing in young children found that indexing cardiac output to body weight produced more stable values across weight groups than the traditional body-surface-area approach, which introduced a systematic skew in the smallest patients.14PubMed Central. Indexing haemodynamic variables in young children Pediatric intensivists are often aware of this, but it remains an area where standard adult cutoffs can trip up decision-making if applied uncritically.
High Cardiac Index Is Not Always Good News
Most discussions of cardiac index focus on when it is too low. But heart failure can also occur in a high-output state, defined as a resting cardiac output above 8 L/min or a cardiac index above 3.9 L/min/m². Conditions like severe anemia, hyperthyroidism, large arteriovenous fistulas (sometimes created for dialysis access), liver cirrhosis, and beriberi can drive the heart to pump far more than usual. Despite that hyperdynamic state, the cardiac output still falls short of the body’s abnormally elevated demands, and the patient develops the same congestion and fluid retention seen in classic low-output heart failure.15OA Text. High output heart failure: A review of clinical status – epidemiology, pathophysiology, diagnosis, prognosis and clinical management If a clinician sees a cardiac index of 4.5 and assumes the heart is fine, they may miss a syndrome that requires treating the underlying driver rather than the heart itself.
The Microcirculation Gap
One of the most consequential limitations of cardiac index is what it cannot see. The number tells you how much blood is leaving the heart, but not whether that blood is reaching the tiny capillary beds where oxygen actually gets delivered to cells. In critically ill patients, especially those with sepsis, the microcirculation can be profoundly disturbed even when large-vessel flow looks adequate. Blood may shunt past capillary networks, leaving pockets of tissue starved of oxygen despite a reassuringly normal cardiac index.
Research on microcirculatory dysfunction in sepsis has shown that improvements in large-vessel hemodynamics correlate only weakly with improvements at the capillary level.16PubMed. Microcirculatory dysfunction in sepsis: pathophysiology, clinical monitoring, and potential therapies One review put it bluntly: despite adequate maintenance of systemic oxygen delivery, morbidity and mortality in septic patients remain high, because the assumption that tissue oxygenation can be preserved simply by maintaining blood supply only holds when capillaries are uniformly perfused, and in sepsis, they are not.17PubMed Central. Microcirculatory dysfunction and tissue oxygenation in critical illness Organ damage can progress even when cardiac index, blood pressure, and oxygen saturation all appear normal.18PubMed Central. Effects of impaired microvascular flow regulation on metabolism-perfusion matching and organ function
This gap is one reason modern critical care is moving toward multimodal monitoring. Cardiac index tells you part of the story. Lactate levels, venous oxygen saturation, capillary refill time, and emerging microcirculation imaging tools each add context. No single number captures whether a patient’s organs are truly getting what they need.
Cardiac Index in the Operating Room
During and after cardiac surgery, cardiac index helps guide one of the trickiest moments in an operation: weaning the patient off the heart-lung bypass machine that has been doing the heart’s job. A patient who entered surgery with a low cardiac index (below 2.5 L/min/m²) is more likely to struggle coming off bypass and may need drugs to support heart contractility. One study found that the inotrope milrinone effectively improved cardiac index and reduced vascular resistance in patients with low pre-bypass values, though many also needed a second drug to maintain blood pressure, illustrating how these medications involve constant tradeoffs.19PubMed. Hemodynamic effects of milrinone during weaning from cardiopulmonary bypass: comparison of patients with a low and high prebypass cardiac index
In patients who entered surgery with adequate function, the same drug sometimes pushed the cardiac index above the upper limit of normal and drove vascular resistance below safe levels, a reminder that pharmacological support calibrated to a patient’s baseline hemodynamics works better than a one-size-fits-all protocol. Cardiac index measured before, during, and after surgery helps the anesthesiologist titrate support to what that particular heart actually needs.
What Athletes Can Teach Us About Cardiac Reserve
Resting cardiac index does a poor job of distinguishing an elite athlete’s heart from an average one. A study using real-time cardiac MRI during exercise found that athletes and healthy non-athletes had essentially identical cardiac indices at rest. The difference appeared only during exercise: at peak effort, athletes reached a median cardiac index of about 12.2 L/min/m², compared with roughly 8.9 in healthy non-athletes.20Journal of Cardiovascular Magnetic Resonance. Assessing exercise cardiac reserve using real-time cardiovascular magnetic resonance The exercise cardiac index was so discriminating that it separated the two groups with near-perfect accuracy, performing as well as VO₂ max, the standard measure of aerobic fitness.
This has clinical implications beyond sports. When doctors suspect a patient’s symptoms, such as unexplained shortness of breath, might stem from the heart’s inability to increase its output with exertion, measuring cardiac index only at rest can miss the problem entirely. Exercise testing with hemodynamic measurement can expose a limited cardiac reserve that a resting snapshot would never reveal.
Machine Learning and the Future of Measurement
One of the persistent barriers to using cardiac index more widely is that the most accurate methods are invasive or expensive. Pulmonary artery catheters carry small but real risks of infection and blood-vessel injury; cardiac MRI requires specialized equipment and time. Researchers are now exploring whether machine learning can estimate cardiac output from something as simple as a peripheral arterial waveform, the squiggly pressure trace you get from a sensor on the wrist or neck. A recent study using simulated data from thousands of virtual patients found that convolutional neural networks could estimate cardiac output from radial, temporal, and carotid artery waveforms with correlations above 0.93, even when the pressure signal was normalized or had noise added.21Scientific Reports. Machine learning-enabled estimation of cardiac output from peripheral waveforms is independent of blood pressure measurement location in an in silico population The work is still in the simulated-data stage, not yet validated in real patients, but if it holds up, it could eventually allow continuous, non-invasive cardiac index monitoring with a simple wrist sensor. That would change the accessibility of the measurement dramatically, shifting it from an ICU and cath-lab tool to something feasible in emergency departments, general wards, or even outpatient settings.