Intrapulmonary Shunt and Physiological Dead Space

Shunt and dead space describe different failures to match blood flow with ventilation. Shunt emphasizes blood that leaves without adequate oxygenation. Dead space emphasizes ventilation that contributes little or nothing to carbon dioxide elimination. Both can coexist, and the measurement method determines exactly what a reported fraction means.

32.1 Shunt and Dead Space Are Opposite Ends of a Distribution

A lung region exchanges gas according to the ventilation and perfusion reaching it. At one extreme, blood flows through a region receiving no ventilation. At the other, gas enters a region receiving no blood flow. Between these extremes lies a continuum of ventilation–perfusion ratios, abbreviated V/Q.

True intrapulmonary shunt includes perfusion of nonventilated alveolar regions. Flooding or collapse can create this functional shunt even when blood follows the usual pulmonary vascular route. In contrast, an anatomical right-to-left pathway bypasses ventilated alveolar gas exchange through a vascular connection. Intracardiac right-to-left shunting is an additional possible source of bypass, but it is not an intrapulmonary lesion.

A low-V/Q region receives some ventilation relative to its blood flow. Its blood may leave incompletely oxygenated, but the region differs from a completely nonventilated shunt. Increasing inspired oxygen can raise its alveolar oxygen tension and improve oxygen transfer, depending on how little ventilation reaches it.

A high-V/Q region receives more ventilation than its perfusion can use effectively. Its alveolar carbon dioxide falls toward the inspired value as perfusion becomes very small. Complete absence of perfusion is the limiting case of alveolar dead space, while less extreme high-V/Q regions contribute to inefficient ventilation.

The lung therefore rarely consists of only normal units plus two completely abnormal compartments. Riley and Cournand's analysis of ideal alveolar air helped organize these relationships into a useful gas-exchange framework. Such compartment models summarize the resulting gas and blood mixture; they do not mean that each calculated compartment has a sharply defined anatomical boundary. 1

This distinction guides interpretation. Hypoxemia can reflect low V/Q, true shunt, low inspired oxygen, hypoventilation, or diffusion-related limitations in different combinations. An increased carbon dioxide requirement can reflect dead space, metabolic production, or inadequate total ventilation. The observed blood gas is the integrated result.

32.2 The Shunt Equation Is an Oxygen-Content Mixing Balance

The shunt equation compares the oxygen content expected from ideal gas-exchanging blood with the measured arterial content after mixing. Content is the amount of oxygen carried per volume of blood. It differs from oxygen tension, which describes the dissolved gas pressure and cannot be mixed linearly to obtain the arterial result.

Let Qs be shunted blood flow and Qt total pulmonary blood flow in the model. Cc′O₂ is ideal end-capillary oxygen content from the gas-exchanging compartment. CaO₂ is measured arterial oxygen content, and CvO₂ is mixed venous oxygen content entering the lung. All three contents must use the same units, commonly mL O₂/dL.

Oxygen content includes hemoglobin-bound and dissolved oxygen. In a common approximation, Hb is hemoglobin in g/dL, S is the corresponding oxygen saturation as a fraction, and PO₂ is oxygen tension in mm Hg:

The result is mL O₂/dL. The constants approximate hemoglobin oxygen capacity and dissolved oxygen solubility under conventional conditions. Measured dyshemoglobins and other departures from those assumptions require appropriate handling. The dissolved component becomes particularly relevant when estimating ideal end-capillary content at very high inspired oxygen.

The shunt calculation then applies a stable mixing model with one ideal oxygenating compartment:

The numerator is the arterial content deficit relative to ideal end-capillary blood. The denominator is the difference between that ideal content and incoming mixed venous content. The ratio identifies the fraction of venous-like blood needed to explain the measured arterial deficit under the model.

For an illustrative calculation, let Cc′O₂ be 20 mL O₂/dL, CaO₂ 17 mL O₂/dL, and CvO₂ 12 mL O₂/dL. The arterial deficit is 20 − 17 = 3. The full ideal-to-venous difference is 20 − 12 = 8. The ratio is 3 / 8 = 0.375, or 37.5%.

The mixing check makes the meaning explicit. If 62.5% of flow has content 20 and 37.5% has content 12, mixed content is 0.625 × 20 + 0.375 × 12 = 17 mL O₂/dL. Mixing oxygen pressures in the same way would be incorrect because the oxygen dissociation relationship is nonlinear.

In practice, ideal end-capillary content is calculated rather than sampled from every capillary. It depends on an estimate of ideal alveolar oxygen and assumed end-capillary equilibration. Arterial and mixed venous samples should describe the same physiological state. Central venous oxygen content is not automatically equivalent to true mixed venous content.

At ordinary inspired oxygen fractions, the result can include the effects of low V/Q and other departures from the ideal compartment. It is then often described as calculated venous admixture. Calling it an anatomical shunt fraction would imply more specificity than the calculation provides.

32.3 The Response to Oxygen Helps Explain the Mechanism

Inspired oxygen can reach a low-V/Q region because some gas still enters it. Raising inspired oxygen usually raises alveolar oxygen in that region and improves the content of blood leaving it. The improvement depends on the degree of ventilation, blood flow, diffusion conditions, and the available oxygen-content increase.

Oxygen cannot enter a completely nonventilated region through its ordinary airway route. Blood traversing that region therefore remains close to incoming venous blood despite an increased inspired oxygen concentration elsewhere. This is the basis for describing true shunt as relatively resistant to oxygen therapy.

Relatively resistant does not mean that the arterial oxygen tension can never rise. Oxygen content in ventilated regions can still increase, particularly through dissolved oxygen at high alveolar tensions. The final arterial tension also depends on the shunt fraction and incoming venous content. A large tension change and a large content change are not interchangeable observations.

As inspired oxygen approaches a very high fraction, low-V/Q regions that still communicate with the airway may become better oxygenated. This can make a calculated venous-admixture estimate more reflective of bypassed gas exchange. However, extreme low V/Q may still behave like shunt over the time sampled, and the physiological state may change during the assessment.

High oxygen exposure can alter absorption of alveolar gas and promote atelectasis in susceptible regions. Changes in vascular tone, ventilation distribution, and circulation can also change the state being measured. A response to oxygen is therefore physiological evidence, rather than a perfectly selective anatomical test.

The actual inspired fraction must also be known. A nominal oxygen setting with substantial entrainment, leaks, or variable support may not describe what reaches the airway. Adequate equilibration and stable measurements matter when comparing oxygen responses.

At the bedside, an oxygenation pattern can strengthen a shunt hypothesis without identifying its cause or location. Lung imaging and cardiac assessment may be needed to distinguish nonventilated perfused lung from a vascular or intracardiac pathway. The response does not by itself determine which pressure, positioning, or circulatory intervention will help.

32.4 Cardiac Output and Mixed Venous Oxygen Modify the Arterial Result

Shunted blood carries the oxygen content with which it entered the lung. If tissues extract more oxygen from each unit of blood, mixed venous content falls. Mixing that more deoxygenated blood into arterial blood produces a larger arterial deficit for the same shunt fraction.

The content balance can be rearranged to make this relationship visible:

Return to the illustrative fraction of 0.375 and ideal content of 20 mL O₂/dL. With venous content 12, arterial content was 17. If venous content falls to 10 while the other values remain fixed, arterial content becomes 0.625 × 20 + 0.375 × 10 = 16.25 mL O₂/dL. The fall does not require a new increase in the modeled shunt fraction.

Cardiac output affects this process through systemic oxygen delivery and extraction. If output falls while metabolic demand remains similar, tissues may extract more oxygen and lower mixed venous content. That can worsen arterial oxygenation when shunt is present. Changes in hemoglobin and metabolic demand also affect the balance.

Output can additionally change pulmonary perfusion distribution and the measured shunt fraction. In patients with acute respiratory failure, Dantzker and colleagues examined V/Q distributions using multiple inert gases during ventilatory changes. Reductions in measured shunt accompanied reductions in output under some conditions. Their study cautions against assuming that a smaller shunt estimate necessarily means more lung tissue has opened. 2

These influences can oppose one another. Lower output may reduce perfusion through some poorly ventilated regions while also lowering mixed venous oxygen content. The arterial result depends on both changes. A single saturation cannot show which effect dominated.

This is why an apparent pulmonary improvement must be assessed with systemic perfusion. A favorable oxygenation or shunt measurement accompanied by an important loss of forward flow may not represent a favorable overall state. Follow cardiac performance, hemoglobin, oxygen content, and organ perfusion alongside the lung measurements.

32.5 Dead Space Includes Conducting Airways, Apparatus, and Alveolar Inefficiency

Anatomical or airway dead space is the effective volume of conducting passages that do not exchange gas with pulmonary capillary blood. Gas in these passages is exhaled before much of the alveolar gas appears at the measurement site. The effective volume depends on the airway geometry and gas-mixing process rather than being one fixed number for every breath.

Fowler developed a single-breath method using expired nitrogen after oxygen inspiration to estimate this conducting-airway volume. The method illustrates why functional measurement and geometric airway size are related but not identical. Its result depends on the pattern of gas washout. 3

Apparatus dead space is additional rebreathed volume in equipment through which gas travels in both directions. Connectors, catheter mounts, and some heat-and-moisture exchangers can contribute. The relevant volume is the gas that can be rebreathed, rather than the entire volume of all tubing in a circuit with separate inspiratory and expiratory limbs.

An artificial airway also changes the anatomical pathway by bypassing portions of the upper airway. Equipment dead space must therefore be assessed in the actual configuration. A change that adds a modest volume can be important when tidal volume is small because it consumes a larger fraction of each breath.

Alveolar dead space describes ventilation that is poorly used because perfusion is absent or insufficient relative to ventilation. Pulmonary vascular obstruction, low pulmonary blood flow, and compression of capillaries in excessively expanded regions can contribute. A high dead-space measurement is not specific for embolism.

Physiological dead space in the Bohr framework combines the conducting-airway and alveolar contributions. Apparatus may also be included or excluded depending on where gas and volume are measured and how the system corrects for it. State the measurement boundary before comparing values from different configurations.

In a simple volume model, effective alveolar ventilation equals rate multiplied by tidal volume minus dead-space volume. This model helps explain why a fixed added equipment volume matters more at smaller tidal volumes. It does not make every clinical dead-space index a direct anatomical volume that can be subtracted without further assumptions.

32.6 Bohr's Equation Measures Dilution of Alveolar Carbon Dioxide

Expired gas contains carbon dioxide from gas-exchanging regions mixed with gas that contributed little or none. The resulting dilution can estimate how much of a breath was ineffective for carbon dioxide exchange. Bohr's equation expresses this balance.

Let VD be dead-space volume and VT expired tidal volume. PACO₂ is the mean alveolar carbon dioxide partial pressure used by the method. PĒCO₂ is the partial pressure corresponding to mixed expired gas collected across the complete breath. Use the same pressure units, commonly mm Hg, for both carbon dioxide values.

For an illustrative mean alveolar PCO₂ of 40 mm Hg and mixed expired PCO₂ of 24 mm Hg, the dilution difference is 40 − 24 = 16 mm Hg. The fraction is 16 / 40 = 0.40, or 40%. If the expired tidal volume is 500 mL under compatible measurement conditions, the model-equivalent dead-space volume is 0.40 × 500 = 200 mL.

The meaning follows the mixing balance. The expired breath carries carbon dioxide as though 60% of its volume had the specified alveolar concentration and 40% had no carbon dioxide. In a heterogeneous lung this is an equivalent-volume description; it does not show that exactly 200 mL occupies a single nonperfused compartment.

The familiar form assumes negligible inspired carbon dioxide. Rebreathing or an elevated inspired concentration requires the appropriate mass-balance correction. Gas volumes and concentrations must be measured on compatible temperature, pressure, and water-vapor conventions. A concentration expressed as a fraction is not interchangeable with a partial pressure without the appropriate conversion.

The difficult quantity is representative mean alveolar PCO₂. End-tidal PCO₂ samples the end of expiration and is not automatically its equivalent. In uneven lungs, different regions empty at different times and have different carbon dioxide values. The method used to estimate the alveolar reference therefore influences the calculated fraction.

Bohr's equation remains a physiological measurement of ventilatory dilution when its inputs and assumptions are satisfied. It should be distinguished from a calculation that replaces the alveolar reference with an arterial blood value.

32.7 Enghoff's Substitution Changes the Physiological Meaning

The Enghoff modification replaces mean alveolar PCO₂ with arterial PCO₂. Arterial sampling is practical and avoids directly measuring the alveolar reference. However, arterial blood integrates the effects of all perfused regions, including blood that bypassed well-ventilated gas exchange.

PaCO₂ is arterial carbon dioxide partial pressure. PĒCO₂ remains mixed expired pressure; it has not become end-tidal pressure. Both must describe the same physiological state and use the same pressure units.

Using the previous illustrative mixed expired value of 24 mm Hg, suppose arterial PCO₂ is 50 mm Hg while the mean alveolar value used for Bohr remains 40 mm Hg. Enghoff's numerator is 50 − 24 = 26, and its fraction is 26 / 50 = 0.52. The Bohr fraction was 0.40. The difference arises from changing the reference from alveolar gas to arterial blood, not from a demonstrated new volume of nonperfused lung.

Shunt and low-V/Q blood can raise arterial PCO₂ relative to the mean of the ventilated alveolar gas. Using that arterial value therefore incorporates effects beyond ventilatory dead space. The Enghoff result is better understood as a global index of gas-exchange inefficiency than as a direct measure of true dead-space volume.

Suarez-Sipmann and colleagues evaluated Enghoff-based shunt-correction approaches in surfactant-depleted pigs. Even proposed corrections continued to differ from Bohr measurements. The experiment supports preserving the distinction between the indices rather than assuming that a simple correction fully isolates dead space. It does not establish one correction method for every clinical setting. 4

The arithmetic difference between Enghoff and Bohr is also not a direct shunt fraction. Carbon dioxide content–pressure relationships, V/Q distribution, and the methods used to estimate alveolar gas influence it. A difference may provide mechanistic information without being numerically interchangeable with Qs/Qt.

Enghoff remains clinically useful, including for prognosis and trends. The requirement is to name it accurately and interpret the broader mixture of abnormalities it captures. Calling every increase worsening alveolar dead space loses that information.

32.8 Mixed Expired and End-Tidal Carbon Dioxide Answer Different Questions

Mixed expired carbon dioxide is a volume-weighted average of the whole exhaled breath. It includes early conducting-airway gas as well as later alveolar gas. End-tidal carbon dioxide is the value at the selected end of expiration. It does not include the same averaging and cannot replace mixed expired PCO₂ in either equation without changing the calculation.

Time-based capnography plots carbon dioxide against time. Volumetric capnography plots it against expired volume. Integrating carbon dioxide concentration across expired volume and dividing by total expired volume provides a mean expired fraction. Converting that fraction to partial pressure requires the applicable pressure and water-vapor convention.

The early capnogram contains gas with little carbon dioxide, followed by a transition and an alveolar phase. The alveolar phase often slopes upward rather than forming a perfectly flat plateau. Sequential emptying, regional V/Q differences, and continuing gas exchange contribute to its shape. Its final point need not equal the mean alveolar value needed by Bohr's method.

Tusman and colleagues validated a volumetric-capnography approach to estimate mean alveolar PCO₂ and Bohr dead space against a multiple-inert-gas-based reference in lung-lavaged pigs. Agreement supported that method under the studied conditions. It did not establish that a routine end-tidal display is an equivalent measurement or validate every device algorithm in every ICU patient. 5

Clinical comparison studies also matter. Kallet and colleagues compared volumetric capnography with metabolic-monitor measurements in patients with ARDS. Their findings supported agreement when gas collection and circuit effects were handled appropriately. This validates aspects of measurement, rather than proving that monitor-guided treatment improves outcomes. 6

Leaking circuits, gas-sampling delay, misalignment between flow and carbon dioxide signals, condensation, and changes in apparatus can alter the curve or its integral. Expired gas collected at the ventilator may also be diluted by circuit-related gas unless accounted for. A plausible curve does not eliminate the need to understand where and how it was measured.

32.9 The Arterial–End-Tidal Gap Is a Clue, Not a Complete Dead-Space Measurement

The arterial–end-tidal carbon dioxide gap is PaCO₂ minus PETCO₂. It compares a blood value with the final expired-gas sample. It does not contain the volume-weighted mixed expired value, so it is not the Bohr or Enghoff equation.

A widening gap can occur when substantial ventilation reaches poorly perfused regions, lowering expired carbon dioxide relative to arterial blood. Reduced pulmonary blood flow can produce a similar pattern. Uneven emptying, altered expiratory timing, shunt, and the choice of end-expiratory sample also affect the relationship.

Nunn and Hill measured dead-space components and the arterial–end-tidal difference in anesthetized patients. Their observations illustrate that the gap and measured dead-space quantities describe related but distinct features of gas exchange. Results from that selected population do not supply a fixed normal gap or universal conversion for every critically ill patient. 7

End-tidal PCO₂ can fall abruptly when pulmonary blood flow decreases, even if delivered minute ventilation has not changed. It can also fall when effective ventilation increases or the sampling signal is lost or diluted. The clinical meaning therefore depends on the waveform and concurrent arterial pressure, output, and ventilation.

A small gap does not prove uniform V/Q matching. Different regional abnormalities can offset one another in the end-expiratory sample. Under some conditions, the end-tidal value can even exceed arterial PCO₂, so a positive difference should not be imposed as an invariant rule.

Use a paired arterial sample when the relationship matters for a decision and the patient's state permits meaningful comparison. A blood gas from an earlier ventilator or circulatory state should not be subtracted from a new end-tidal reading as though the two were simultaneous.

Serial gaps can strengthen a mechanistic hypothesis when acquisition and conditions are comparable. They should not be used alone to diagnose pulmonary embolism, quantify an anatomical dead space, or determine that a ventilatory intervention improved the entire circulation.

32.10 PEEP Responses and Prognostic Associations Require Method-Specific Interpretation

PEEP can change shunt and ventilatory efficiency through several pathways. Recruitment can improve ventilation of perfused tissue and reduce venous admixture. Expansion of already aerated regions can instead increase high-V/Q ventilation by compressing vessels or changing blood-flow distribution. Changes in output can modify both expired carbon dioxide and mixed venous oxygen.

An Enghoff value may improve when shunt decreases even if true Bohr dead space changes little. Conversely, a rise can reflect worsening venous admixture rather than newly nonperfused alveoli. This is why following the index during a PEEP change requires the calculation method, arterial gas, expired curve, mechanics, and circulation together.

Doorduin and colleagues compared dead-space calculations in post-cardiac-surgery patients and patients with ARDS. Replacing alveolar with arterial PCO₂ and changing the method of mixed expired measurement produced clinically relevant differences, especially in ARDS. Their study demonstrates that values bearing the same VD/VT label can measure different combinations of physiology and technique. 8

Prognostic evidence remains useful despite those distinctions. Nuckton and colleagues prospectively measured the Enghoff-modified fraction early in ARDS in 179 patients. Higher values were associated with hospital mortality after adjustment. The measurement captured a clinically important gas-exchange abnormality, but the study did not test a strategy of normalizing it or establish that lowering the number itself improves survival. 9

A later observational analysis by Kallet and colleagues examined factors affecting the relationship between the fraction and ARDS outcome. Such work supports considering disease context and trajectory rather than assigning one index an identical meaning across all patients. Prognostic association remains different from identifying a causal treatment target. 10

The same limit applies to selecting PEEP from a minimum value. A minimum gas-exchange index identifies the best value of that index under the sampled conditions. It does not independently establish acceptable lung stress, regional aeration, right ventricular load, or systemic oxygen delivery. A favorable trajectory is stronger when those other observations agree.

32.11 Integrating Hypoxemia, Hypercapnia, and Expired Gas

Consider an illustrative patient with ARDS and worsening hypoxemia. After PEEP increases, arterial oxygenation improves, but end-tidal carbon dioxide falls and arterial pressure decreases. Arterial PCO₂ subsequently rises despite unchanged delivered minute ventilation.

Several mechanisms are possible. Useful recruitment may have improved oxygenation, while reduced pulmonary blood flow lowered carbon dioxide delivery to ventilated regions. Additional expansion may have impaired regional perfusion. A new rise in dead-space ventilation is possible, but the end-tidal change alone cannot establish it.

Begin by checking the signals and timing. Confirm the expired waveform, volume delivery, circuit configuration, leaks, and the relation of the arterial sample to the ventilatory change. A disconnected sampling line and a true fall in pulmonary perfusion can both produce a low displayed end-tidal value through very different mechanisms.

Then connect gas exchange to circulation and mechanics. Review output or its clinical surrogates, right ventricular behavior, plateau and driving pressures, and regional aeration when available. An arterial oxygenation gain does not establish that oxygen delivery improved if output fell importantly.

If expired-gas measurements are obtained, label what was calculated. An Enghoff fraction describes a combined gas-exchange inefficiency. A validated Bohr estimate adds more specific information about ventilatory dilution. Their difference can support an account of the response without becoming a direct shunt percentage.

Also consider metabolic carbon dioxide production and total effective ventilation. Fever, changing nutrition or activity, and altered respiratory effort can affect the balance. Hypercapnia at one moment is not a complete diagnosis of dead-space failure, particularly before carbon dioxide stores have equilibrated after a change.

The useful endpoint is a coherent account of why oxygenation and carbon dioxide elimination changed and whether systemic perfusion remains adequate. Shunt and dead-space concepts make that account more precise when the definitions, measurement methods, and time course remain explicit.

References

1. Riley RL, Cournand A. Ideal alveolar air and the analysis of ventilation-perfusion relationships in the lungs. J Appl Physiol. 1949;1(12):825–847. doi:10.1152/jappl.1949.1.12.825

2. Dantzker DR, Lynch JP, Weg JG. Depression of cardiac output is a mechanism of shunt reduction in the therapy of acute respiratory failure. Chest. 1980;77(5):636–642. doi:10.1378/chest.77.5.636

3. Fowler WS. Lung function studies. II. The respiratory dead space. Am J Physiol. 1948;154(3):405–416. doi:10.1152/ajplegacy.1948.154.3.405

4. Suarez-Sipmann F, Santos A, Böhm SH, Borges JB, Hedenstierna G, Tusman G. Corrections of Enghoff's dead space formula for shunt effects still overestimate Bohr's dead space. Respir Physiol Neurobiol. 2013;189(1):99–105. doi:10.1016/j.resp.2013.06.020

5. Tusman G, Suarez Sipmann F, Borges JB, Hedenstierna G, Böhm SH. Validation of Bohr dead space measured by volumetric capnography. Intensive Care Med. 2011;37(5):870–874. doi:10.1007/s00134-011-2164-x

6. Kallet RH, Daniel BM, Garcia O, Matthay MA. Accuracy of physiologic dead space measurements in patients with acute respiratory distress syndrome using volumetric capnography: comparison with the metabolic monitor method. Respir Care. 2005;50(4):462–467. PubMed

7. Nunn JF, Hill DW. Respiratory dead space and arterial to end-tidal carbon dioxide tension difference in anesthetized man. J Appl Physiol. 1960;15(3):383–389. doi:10.1152/jappl.1960.15.3.383

8. Doorduin J, Nollet JL, Vugts MPAJ, et al. Assessment of dead-space ventilation in patients with acute respiratory distress syndrome: a prospective observational study. Crit Care. 2016;20:121. doi:10.1186/s13054-016-1311-8

9. Nuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med. 2002;346(17):1281–1286. doi:10.1056/NEJMoa012835

10. Kallet RH, Zhuo H, Ho K, Lipnick MS, Gomez A, Matthay MA. Lung injury etiology and other factors influencing the relationship between dead-space fraction and mortality in ARDS. Respir Care. 2017;62(10):1241–1248. doi:10.4187/respcare.05589