Coronary Perfusion, Myocardial Oxygen Supply, and Demand

The heart must receive enough oxygen to support its own work while supplying the rest of the body. Myocardial ischemia develops when oxygen delivery cannot meet the muscle's needs. Restricted coronary flow, low blood oxygen content, unfavorable pressure conditions, and increased demand can contribute together.

13.1 Oxygen Delivery and Oxygen Consumption Are Different Quantities

Myocardial oxygen delivery is the amount of oxygen brought to the myocardium each minute. It depends on coronary blood flow and the oxygen content of that blood:

The units must match before multiplying. Suppose coronary flow is 250 mL of blood/min and arterial oxygen content is 18 mL of oxygen/dL of blood. A deciliter contains 100 mL, so first convert blood flow:

250 ÷ 100 = 2.5 dL of blood/min

Then calculate oxygen delivery:

2.5 × 18 = 45 mL of oxygen/min

These illustrative values describe oxygen arriving in the coronary circulation. They do not show how much oxygen the myocardium uses. Some oxygen remains in the blood leaving the tissue.

Myocardial oxygen consumption depends on how much is extracted. Subtract coronary venous oxygen content from arterial oxygen content to find the amount removed from each unit of blood. Multiplying that difference by the corresponding coronary flow gives an estimate of consumption. This is the Fick principle applied to the heart. The flow measurement and venous sample must represent the same myocardial territory.

The myocardium already extracts a substantial proportion of delivered oxygen at rest. It therefore has less room to increase extraction than many other organs. When demand rises, increasing coronary blood flow is a major means of providing more oxygen.

Arterial oxygen content depends mainly on hemoglobin concentration and saturation:

Hemoglobin is expressed in g/dL, saturation as a fraction, and oxygen tension in mm Hg. The result is in mL of oxygen/dL. The first term estimates oxygen bound to hemoglobin. The second estimates oxygen dissolved in plasma. The coefficients are clinical approximations.

Saturation describes the fraction of available hemoglobin binding sites carrying oxygen. It does not describe how much hemoglobin is present. A patient with anemia can therefore have a high saturation and low oxygen content. Adequate content also cannot compensate for severely restricted flow to a myocardial region. Evaluate both components of delivery.

13.2 Why Left Ventricular Perfusion Favors Diastole

Coronary blood flow depends on pressure within the vessels and forces compressing them from outside. During left ventricular systole, contracting muscle compresses vessels within the myocardial wall. This particularly restricts perfusion of the deeper layers.

During relaxation, that compression falls. Aortic pressure continues to drive coronary inflow, allowing much of left ventricular perfusion to occur during diastole. Flow does not stop everywhere during systole, and the pattern differs among myocardial territories.

A useful bedside approximation describes the pressure available to support left ventricular diastolic perfusion:

For illustration, let aortic diastolic pressure be 60 mm Hg and left ventricular end-diastolic pressure be 10 mm Hg:

60 − 10 = 50 mm Hg

Now let ventricular end-diastolic pressure rise to 25 mm Hg while aortic diastolic pressure remains 60 mm Hg:

60 − 25 = 35 mm Hg

The arterial pressure is unchanged, but the estimated pressure difference is smaller. Raising filling pressure can therefore make the perfusion environment less favorable even when the arterial monitor looks stable.

This subtraction is an explanatory model, not a regional flow measurement. Pressures vary throughout diastole. Tissue compression, coronary venous pressure, vascular resistance, and any pressure loss across a stenosis influence actual flow. One end-diastolic pressure difference cannot capture all of them.

Buckberg and colleagues produced subendocardial ischemia in dogs with normal coronary arteries by creating unfavorable hemodynamic conditions. Their experiments related the diastolic pressure and time available for perfusion to systolic pressure generation. They demonstrate that ischemia can occur without a fixed epicardial obstruction. 1

The normal right ventricle has a lower systolic pressure and less restriction of systolic perfusion. As its pressure and wall tension rise, systolic perfusion can become more difficult. The relevant pressure environment therefore depends on which ventricle and myocardial territory are being assessed.

13.3 What Determines Myocardial Oxygen Demand

The myocardium uses energy to activate contraction, generate force, shorten, handle calcium, relax, and maintain its cells. Cardiac output describes blood leaving the heart, not the energy used to achieve that output. A ventricle generating high pressure with little ejection can still consume substantial oxygen.

Pressure contributes to wall stress, the force acting within the myocardium relative to tissue area. Geometry modifies that stress. At a comparable pressure and wall thickness, a larger chamber radius requires greater wall stress. A dilated ventricle may therefore face a demanding mechanical task even without an unusually high arterial pressure.

Duration also matters. Sustaining tension for longer requires more energy than reaching a brief pressure peak under otherwise comparable conditions. Sarnoff and colleagues investigated these determinants using the tension–time index. The index relates the pressure generated during systole, integrated over time, to myocardial oxygen consumption. It captures aspects of pressure generation that peak pressure alone misses. 2

Greater contractile activation adds another requirement. Calcium must be released, moved, and returned to its stores. Inotropic stimulation may improve ejection while increasing energy used for activation. In canine hearts, Suga and colleagues found an additional oxygen requirement during enhanced contractile activation with epinephrine or calcium. This demonstrates a physiological tradeoff, not a uniform clinical response to every inotrope. 3

Heart rate determines how often these tasks are repeated. More contractions per minute can increase total oxygen consumption even if work per beat changes little. Fever, pain, agitation, and increased work of breathing may increase rate, adrenergic stimulation, or the circulation's required output.

The rate–pressure product multiplies heart rate by systolic arterial pressure. It can track aspects of demand in some settings. It does not include chamber geometry, regional wall stress, filling pressure, or all changes in activation. Treat it as a limited surrogate rather than a measurement of myocardial oxygen consumption in the ICU.

13.4 Tachycardia Can Increase Demand While Reducing Supply Time

Tachycardia increases the number of contractions performed each minute. At the same time, it shortens each cardiac cycle. The heart may need more oxygen while having less time for the diastolic perfusion that supplies much of the left ventricle.

Calculate total cycle length by dividing 60 seconds by the number of beats per minute:

At 60 beats/min: 60 ÷ 60 = 1 second per beat

At 120 beats/min: 60 ÷ 120 = 0.5 seconds per beat

This arithmetic describes the whole cycle. It does not establish how long systole or diastole lasts. They do not shorten in identical proportions, so the diastolic perfusion time cannot simply be assumed to halve.

The effect is particularly important when reserve is already limited. Low aortic diastolic pressure, high ventricular filling pressure, or a coronary stenosis can reduce the pressure supporting flow. Shortening diastole then restricts the remaining opportunity to deliver oxygen, while the higher rate raises demand.

Tachycardia also leaves less time for ventricular filling. If relaxation is slow, filling may be incomplete before the next contraction. Stroke volume can fall, limiting or eliminating the increase in cardiac output expected from the faster rate. Myocardial energy use can still rise.

Determine why the heart is fast before attempting to slow it. Compensatory sinus tachycardia during shock differs from a rapid atrial arrhythmia. Atrial fibrillation also removes coordinated atrial contraction and produces irregular filling intervals. These changes can impair filling beyond the effect of rate alone.

If output depends on the current rate because stroke volume is severely limited, abrupt rate reduction can reduce perfusion. Treat the driver and assess rhythm, filling, and ventricular reserve. Judge any rate intervention by its effect on forward flow and arterial pressure as well as the displayed heart rate.

13.5 Why the Subendocardium Is Vulnerable

The inner left ventricular layers experience strong compression during systole. Their perfusion therefore depends particularly on the pressure and time available during diastole. This makes the subendocardium vulnerable when demand increases or diastolic supply becomes restricted.

High left ventricular filling pressure can worsen this balance. It reduces the simple aortic-to-ventricular diastolic pressure difference and can accompany increased tissue compression. An acceptable mean arterial pressure does not exclude an unfavorable subendocardial perfusion environment. The arterial pressure pattern and ventricular filling pressure also need attention.

Hypertrophy creates another challenge. More myocardial mass requires more oxygen. If microvascular growth and vasodilator reserve do not keep pace, blood supply may be adequate at rest but insufficient during stress. A preserved ejection fraction does not establish normal coronary reserve in a hypertrophied ventricle.

Severe aortic stenosis combines several of these mechanisms. The ventricle generates a high systolic pressure to eject through the narrowed valve. Hypertrophy and impaired relaxation can accompany that pressure loading. Higher demand and elevated diastolic pressure can then threaten supply even without a new coronary thrombus.

Buckberg's experimental work illustrates how pressure generation and diastolic perfusion interact to disadvantage the inner layers. Its clinical lesson is the mechanism, rather than a single pressure–time ratio that should be imposed as a universal ischemia threshold. 1

A systemic stressor and a focal coronary lesion can also coexist. Anemia or tachycardia may reveal ischemia beyond a stenosis, while acute coronary pathology may develop during a systemic illness. New ischemic symptoms, electrocardiographic changes, or regional dysfunction require assessment even when a plausible systemic contributor has already been found.

13.6 Autoregulation, Flow Reserve, and Coronary Stenosis

Coronary resistance vessels adjust their caliber as perfusion pressure and myocardial needs change. If pressure falls within a range that the vessels can accommodate, dilation can help maintain flow. This pressure-buffering behavior is called autoregulation.

Mosher and colleagues altered coronary perfusion pressure while controlling cardiac performance in experimental preparations. Their work demonstrated a vascular response beyond the immediate passive effect of changing pressure. The range over which flow is maintained depends on demand and other conditions; it is not one fixed pressure interval shared by all patients. 4

As pressure falls, resistance vessels use some of their dilation reserve to preserve flow. Once they are near maximal dilation, further pressure reduction is more likely to reduce perfusion. Increased demand can also use reserve that would otherwise help the circulation tolerate a pressure fall.

An epicardial stenosis causes a pressure loss between the aorta and the distal circulation. Downstream dilation may preserve resting flow despite that loss. During stress, the vessels have less ability to dilate further and increase oxygen delivery. Tachycardia, anemia, or increased pressure work can then produce ischemia despite apparently adequate resting perfusion.

Coronary flow reserve compares maximal stimulated flow with resting flow. It reflects the epicardial vessels and the microcirculation together. The ratio can be low because maximal flow is restricted, resting flow is already high, or both. It is not simply a measurement of stenosis diameter.

Fractional flow reserve, or FFR, is a different measure. It uses coronary pressures during hyperemia to assess the functional effect of an epicardial lesion. Pijls and colleagues demonstrated its usefulness in patients with moderate coronary stenoses. Interpretation depends on the conditions and assumptions of the pressure measurement, including the hyperemic response. It is not a routine aortic pressure measurement or a complete test of microvascular health. 5

Coronary anatomy, resting flow, and the capacity to increase flow therefore answer different questions. A normal systemic pressure does not guarantee adequate pressure beyond a lesion. The absence of severe epicardial obstruction does not establish normal microvascular reserve.

13.7 Hypoxemia, Anemia, and Treatment Evidence

Hypoxemia and anemia reduce arterial oxygen content through different parts of the equation in Section 13.1. Hypoxemia reduces hemoglobin saturation. Anemia reduces the amount of hemoglobin available to carry oxygen. Either can threaten myocardial supply when coronary flow cannot increase enough to compensate.

Consider hemoglobin of 8 g/dL, saturation of 98%, and arterial oxygen tension of 100 mm Hg. Enter saturation as 0.98. The oxygen bound to hemoglobin is:

1.34 × 8 × 0.98 ≈ 10.51 mL of oxygen/dL

The dissolved component is:

0.003 × 100 = 0.30 mL of oxygen/dL

Adding the two gives:

10.51 + 0.30 ≈ 10.8 mL of oxygen/dL

Now use hemoglobin of 12 g/dL with the same saturation and oxygen tension:

(1.34 × 12 × 0.98) + 0.30 ≈ 16.1 mL of oxygen/dL

The pulse oximeter can show the same saturation while blood oxygen content differs substantially. The values are illustrative and do not specify a transfusion decision.

Correcting hypoxemia addresses a supply limitation. Raising an already adequate oxygen tension is different: the dissolved component ordinarily contributes much less than hemoglobin-bound oxygen. Additional oxygen cannot be assumed to replace missing hemoglobin or restore blood flow beyond an obstructed artery.

In DETO2X–AMI, Hofmann and colleagues randomized patients with suspected myocardial infarction and baseline saturation of at least 90% to supplemental oxygen or ambient air. Routine oxygen did not reduce one-year mortality. This finding concerns routine supplementation in the studied population; it does not argue against correcting hypoxemia. 6

A 2025 patient-level meta-analysis pooled four transfusion trials in patients with acute myocardial infarction and anemia. The primary 30-day death-or-infarction comparison remained uncertain. Restrictive treatment was associated with higher six-month mortality, a secondary outcome. These findings challenge the assumption that a restrictive strategy has the same effects across all clinical populations. 7

The 2025 AABB guideline suggests a liberal red cell transfusion strategy for hospitalized patients with acute myocardial infarction when hemoglobin is below 10 g/dL. The recommendation is conditional and based on low-certainty evidence. It requires clinical context and measures to reduce transfusion-related harm. It is not a universal threshold for sepsis, every troponin rise, or all ICU anemia. 10

Physiology explains why hemoglobin affects supply. The diagnosis, active ischemia, bleeding, congestion risk, patient preferences, and applicable evidence guide the treatment decision.

13.8 Septic Myocardial Dysfunction Is Not Always Global Ischemia

Sepsis can impair myocardial performance through altered cellular signaling, inflammatory effects, abnormal calcium handling, and changes in loading. A low ejection fraction during septic shock does not automatically mean that global coronary blood flow is inadequate.

Cunnion and colleagues measured coronary flow and myocardial metabolism in seven patients with septic shock. Flow was preserved, including in those with myocardial depression, and the metabolic observations argued against global ischemia as its explanation. This small physiological study challenged a universal global-ischemia mechanism. It did not exclude every form of myocardial ischemia during sepsis. 8

Regional supply can still be inadequate. Coronary disease, an embolus, severe hypotension, or microvascular abnormalities may affect territories differently. A whole-heart flow assessment can conceal regional differences. Similarly, an adequate global cardiac output does not establish adequate flow to every myocardial region.

Troponin may rise during sepsis because of ischemic or nonischemic myocardial injury. Its serial pattern and the assay's reference limit help identify injury. They do not identify the mechanism by themselves. A rise is neither proof of coronary thrombosis nor a reason to dismiss possible ischemia as an inevitable feature of infection.

Integrate the time course, prior coronary disease, electrocardiograms, ventricular imaging, rhythm, and current supply–demand disturbances. New regional abnormalities deserve attention, although their cause still requires interpretation. Global dysfunction and regional coronary disease may be present together.

The clinical assessment must allow both possibilities: sepsis can depress contraction without globally inadequate coronary flow, and a patient with sepsis can develop an acute coronary event. Infection identifies the setting, not the complete cardiac diagnosis.

13.9 Myocardial Injury and Infarction: Current Diagnostic Language

The 2026 Fifth Universal Definition distinguishes acute myocardial injury from myocardial infarction. Acute injury requires a troponin rise or fall with at least one value above the assay's sex-specific 99th-percentile upper reference limit. Infarction additionally requires evidence of acute ischemia. 9

The document replaces the numerical classification with primary, secondary, and procedure-related infarction. Primary infarction arises from acute coronary pathology. Secondary infarction results from another acute condition causing an oxygen supply–demand imbalance. Procedure-related infarction concerns complications of cardiac procedures. 9

A systemic stressor plus a troponin rise does not confirm secondary infarction. Where appropriate and feasible, confirmation uses coronary or cardiac imaging: obstructive coronary disease without acute coronary pathology, or a new or presumed new regional wall-motion abnormality or loss of viable myocardium consistent with ischemia. A likely clinical diagnosis may remain necessary when imaging is unsuitable. 9

For ICU practice, establish what is known before assigning a cause. A biomarker change may establish acute injury. The symptoms, electrocardiogram, and imaging help determine whether ischemia occurred and what caused it. Finding anemia or shock explains vulnerability but does not settle the diagnosis.

Serial information helps connect the findings to the event. Compare troponin values, electrocardiograms, and regional function over time. A single abnormal result provides less information about when injury occurred or how it relates to the current deterioration.

Sedation, ventilation, and altered consciousness may prevent symptom reporting. Lack of reported chest discomfort then provides little reassurance. Use the objective findings available and assess whether a coronary diagnosis would change urgent management. Correct reversible supply–demand disturbances while investigating the injury rather than allowing a physiological explanation to become an unverified diagnostic label.

13.10 Improving the Balance Without Creating a New Problem

An intervention can improve oxygen supply while increasing demand. Raising arterial pressure may improve the pressure available for coronary perfusion. It can also increase the systolic pressure that the ventricle must generate, increasing wall stress and oxygen requirements.

The balance depends on the starting problem. During severe hypotension, inadequate perfusion pressure may be the dominant limitation. If arterial pressure is already adequate but the ventricle is dilated and ejecting poorly, further arterial loading may reduce stroke volume and increase myocardial work. The same arterial pressure change can therefore have different consequences.

Additional filling has a similar tradeoff. It may increase stroke volume when the circulation is fluid responsive and needs more flow. If it mainly raises left ventricular diastolic pressure, it reduces the estimated coronary driving difference while increasing pulmonary congestion. Assess the flow response and the rise in filling pressure together.

Treatable causes of excessive demand include pain, fever, agitation, respiratory effort, and an unstable tachyarrhythmia. Correcting the cause may reduce myocardial work. Reducing rate or adrenergic activity without maintaining output, however, can lower arterial pressure and create a new supply problem.

An inotrope may support ejection while increasing activation-related oxygen use. Reducing excessive afterload may allow more ejection at a lower ventricular pressure. That benefit depends on preserving adequate systemic and coronary perfusion. Neither choice can be made from ejection fraction alone.

After each intervention, reassess the intended benefit and its consequences. Observe forward flow, arterial pressure, congestion, rhythm, electrocardiographic evolution, and organ perfusion. Improvement in one monitor value is insufficient if output or myocardial supply deteriorates elsewhere.

13.11 An Integrated Bedside Example

Consider an illustrative patient with septic shock and a heart rate of 130 beats/min. Hemoglobin is 8 g/dL and arterial saturation is 98%. Aortic diastolic pressure is 60 mm Hg. A valid assessment indicates left ventricular end-diastolic pressure near 25 mm Hg. New electrocardiographic abnormalities raise concern for ischemia.

Separate the contributing mechanisms. Low hemoglobin reduces oxygen content despite the high saturation. Tachycardia repeats the energetic task more often and shortens each cardiac cycle. Elevated filling pressure reduces the estimated diastolic driving difference:

60 − 25 = 35 mm Hg

This difference is an explanatory estimate, not a universal ischemic threshold. The findings together explain why supply may fail to meet demand. They do not prove the cause of the electrocardiographic changes or exclude an acute coronary lesion.

The pressure inputs also need scrutiny. PAWP should not be silently substituted for measured left ventricular end-diastolic pressure because the two can differ. Check what was measured, under which respiratory conditions, and at what time relative to treatment.

Review serial electrocardiograms and troponin, both ventricles, valve function, prior coronary disease, and the evolving shock state. Sepsis and anemia may contribute to injury while an acute coronary process remains possible. The diagnostic assessment determines which explanation is supported and which treatment evidence applies.

Address the identified limitations while preserving systemic perfusion. Further fluid deserves particular scrutiny because it may raise filling pressure without increasing useful flow. Consider pressure support, the driver of tachycardia, oxygenation, and anemia in the context of the diagnosis. A high saturation alone does not establish an adequate oxygen supply.

A meaningful response includes improved flow and perfusion, an acceptable pressure environment, and less excessive demand. Continue evaluating the myocardial injury as those disturbances improve. Physiological improvement and diagnostic confirmation are related tasks, and both may be necessary.

References

1. Buckberg GD, Fixler DE, Archie JP, Hoffman JI. Experimental subendocardial ischemia in dogs with normal coronary arteries. Circ Res. 1972;30(1):67–81. doi:10.1161/01.RES.30.1.67.

2. Sarnoff SJ, Braunwald E, Welch GH Jr, Case RB, Stainsby WN, Macruz R. Hemodynamic determinants of oxygen consumption of the heart with special reference to the tension-time index. Am J Physiol. 1958;192(1):148–156. doi:10.1152/ajplegacy.1957.192.1.148.

3. Suga H, Hisano R, Goto Y, Yamada O, Igarashi Y. Effect of positive inotropic agents on the relation between oxygen consumption and systolic pressure volume area in canine left ventricle. Circ Res. 1983;53(3):306–318. doi:10.1161/01.RES.53.3.306.

4. Mosher P, Ross J Jr, McFate PA, Shaw RF. Control of coronary blood flow by an autoregulatory mechanism. Circ Res. 1964;14:250–259. doi:10.1161/01.RES.14.3.250.

5. Pijls NHJ, de Bruyne B, Peels K, et al. Measurement of fractional flow reserve to assess the functional severity of coronary-artery stenoses. N Engl J Med. 1996;334(26):1703–1708. doi:10.1056/NEJM199606273342604.

6. Hofmann R, James SK, Jernberg T, et al.; DETO2X–SWEDEHEART Investigators. Oxygen therapy in suspected acute myocardial infarction. N Engl J Med. 2017;377(13):1240–1249. doi:10.1056/NEJMoa1706222.

7. Carson JL, Fergusson DA, Noveck H, et al. Restrictive versus liberal transfusion in myocardial infarction—a patient-level meta-analysis. NEJM Evid. 2025;4(2):EVIDoa2400223. doi:10.1056/EVIDoa2400223.

8. Cunnion RE, Schaer GL, Parker MM, Natanson C, Parrillo JE. The coronary circulation in human septic shock. Circulation. 1986;73(4):637–644. doi:10.1161/01.CIR.73.4.637.

9. Mills NL, Newby LK, Zaman S, et al.; ESC/ACC/AHA/WHF Universal Definition Document Advisory Group. Fifth universal definition of myocardial infarction (2026). J Am Coll Cardiol. 2026;88(11):1314–1368. doi:10.1016/j.jacc.2026.07.025.

10. Pagano MB, Stanworth SJ, Dennis J, et al. Red Cell Transfusion in Acute Myocardial Infarction: AABB International Clinical Practice Guidelines. Ann Intern Med. 2025;178(10):1469–1477. doi:10.7326/ANNALS-25-00706.