Chapter 6. Diffusion of Gases and Interpretation of Pulmonary Function Tests

The reader understands the diffusion of gases in the lung.

• Defines diffusion, and distinguishes it from “bulk flow.”
• States Fick’s law for diffusion.
• Distinguishes between perfusion limitation and diffusion limitation of gas transfer in the lung.
• Describes the diffusion of oxygen from the alveoli into the blood.
• Describes the diffusion of carbon dioxide from the blood to the alveoli.
• Defines the diffusing capacity and discusses its measurement.
• Interprets standard pulmonary function test (PFT) data.

Diffusion of a gas occurs when there is a net movement of molecules from an area in which that particular gas exerts a high partial pressure to an area in which it exerts a lower partial pressure. Movement of a gas by diffusion is therefore different from the movement of gases through the conducting airways, which occurs by “bulk flow” (mass movement or convection). During bulk flow, gas movement results from differences in total pressure, and molecules of different gases move together along the total pressure gradient. During diffusion, different gases move according to their own individual partial pressure gradients. Gas transfer during diffusion occurs by random molecular movement. It is therefore dependent on temperature because molecular movement increases at higher temperatures. Gases move in both directions during diffusion, but the area of higher partial pressure, because of its greater number of molecules per unit volume, has proportionately more random “departures.” Thus, the net movement of gas is dependent on the partial pressure difference between the 2 areas. In a static situation, diffusion continues until no partial pressure differences exist for any gases in the 2 areas; in the lungs, oxygen and carbon dioxide continuously enter and leave the alveoli, and so such an equilibrium does not take place.

Oxygen is brought into the alveoli by bulk flow through the conducting airways. When air flows through the conducting airways during inspiration, the linear velocity of the bulk flow decreases as the air approaches the alveoli. This is because the total cross-sectional area increases dramatically in the distal portions of the tracheobronchial tree, as was seen in Figure 1–2. The linear velocity of bulk flow through a tube is equal to the flow divided by the cross sectional area:

Linear velocity (cm/s) = Flow (cm3/s) ÷ Cross sectional area (cm2)

By the time the air reaches the alveoli, bulk flow probably ceases, and further gas movement occurs by diffusion. Oxygen then moves through the gas phase in the alveoli according to its own partial pressure gradient. The distance from the alveolar duct to the alveolar-capillary interface is usually less than 1 mm. Diffusion in the alveolar gas phase is believed to be greatly assisted by the pulsations of the heart and blood flow, which are transmitted to the alveoli and increase molecular motion.

Oxygen then diffuses through the alveolar-capillary interface. It must first, therefore, move from the gas phase to the liquid phase, according to Henry’s law, which states that the amount of a gas absorbed by a liquid with which it does not combine chemically is directly proportional to the partial pressure of the gas to which the liquid is exposed and the solubility of the gas in the liquid. Oxygen must dissolve in and diffuse through the thin layer of pulmonary surfactant, the alveolar epithelium, the interstitium, and the capillary endothelium, as was shown in Figure 1–5 (step 2, near the arrow). It must then diffuse through the plasma (step 3), where some remains dissolved and the majority enters the erythrocyte and combines with hemoglobin (step 4). The blood then carries the oxygen out of the lung by bulk flow and distributes it to the other tissues of the body, as shown in Figure 1–1. At the tissues, oxygen diffuses from the erythrocyte through the plasma, capillary endothelium, interstitium, tissue cell membrane, and cell interior and into the mitochondrial membrane. The process is almost entirely reversed for carbon dioxide, as shown in Figure 1–1.

The factors that determine the rate of diffusion of gas through the alveolar-capillary barrier are described by Fick’s law for diffusion, shown here in a simplified form:

where

• A = surface area of the barrier available for diffusion
• D = diffusion coefficient, or diffusivity, of the particular gas in the barrier
• T = thickness of barrier of the diffusion distance
• P1 − P2 = partial pressure difference of the gas across the barrier

The surface area of the blood-gas barrier is believed to be at least 70 m2 in a healthy average-sized adult at rest. That is, about 70 m2 of the potential surface area is both ventilated and perfused at rest. If more capillaries are recruited, as in exercise, the surface area available for diffusion increases; if venous return falls, for example, because of hemorrhage, or if alveolar pressure is raised by positive-pressure ventilation, then capillaries may be derecruited and the surface area available for diffusion may decrease.

The thickness of the alveolar-capillary diffusion barrier is only about 0.2 to 0.5 μm. This barrier thickness can increase in interstitial fibrosis or interstitial edema, thus interfering with diffusion. Diffusion probably increases at higher lung volumes because as alveoli are stretched, the diffusion distance decreases slightly (and also because small airways subject to closure may be open at higher lung volumes).

The diffusivity, or diffusion constant, for a gas is directly proportional to the solubility of the gas in the diffusion barrier and is inversely proportional to the square root of the molecular weight (MW) of the gas:

The relationship between solubility and diffusion through the barrier has already been discussed. The diffusivity is inversely proportional to the square root of the MW of the gas because different gases with equal numbers of molecules in equal volumes have the same molecular energy if they are at the same temperature. Therefore, light molecules travel faster, have more frequent collisions, and diffuse more rapidly. Thus, Graham’s law states that the relative rates of diffusion of 2 gases are inversely proportional to the square roots of their MWs, if all else is equal.

For the 2 gases of greatest interest in the lung,

Because the relative diffusion rates in the gas phase are inversely proportional to the ratio of the square roots of their MWs,

That is, because oxygen is less dense than carbon dioxide, it should diffuse 1.2 times as fast as carbon dioxide (which it does as it moves through the alveoli). In the alveolar-capillary barrier, however, the relative solubilities of oxygen and carbon dioxide must also be considered. The solubility of carbon dioxide in the liquid phase is about 24 times that of oxygen, and so carbon dioxide diffuses about 0.85 × 24, or about 20 times, more rapidly through the alveolar-capillary barrier than does oxygen. For this reason, patients develop problems in oxygen diffusion through the alveolar-capillary barrier before carbon dioxide retention due to diffusion impairment occurs.

The factors that limit the movement of a gas through the alveolar-capillary barrier, as described by Fick’s law for diffusion, can be divided into 3 components: the diffusion coefficient, the surface area and thickness of the alveolar-capillary membrane, and the partial pressure difference across the barrier for each particular gas. The diffusion coefficient, as discussed in the previous section, is dependent on the physical properties of the gases and the alveolar-capillary membrane. The surface area and thickness of the membrane are physical properties of the barrier, but they can be altered by changes in the pulmonary capillary blood volume, the cardiac output, or the pulmonary artery pressure; by changes in lung volume; or by diseases such as fibrosis or emphysema. The partial pressure difference of a gas (across the barrier) is the final major determinant of its rate of diffusion. The partial pressure of a gas in the mixed venous blood and in the pulmonary capillaries is just as important a factor as its alveolar partial pressure in determining its rate of diffusion. This will be demonstrated in the next section.

Diffusion Limitation

An erythrocyte and its attendant plasma spend an average of about 0.75 to 1.2 seconds inside the pulmonary capillaries at resting cardiac outputs. This time can be estimated by dividing the pulmonary capillary blood volume by the pulmonary blood flow (expressed in milliliters per second). Some erythrocytes may take less time to traverse the pulmonary capillaries; others may take longer. Figure 6–1 shows schematically the calculated change with time in the partial pressures in the blood of 3 gases: oxygen, carbon monoxide, and nitrous oxide. These are shown in comparison to the alveolar partial pressures for each gas, as indicated by the dotted line. This alveolar partial pressure is different for each of the 3 gases, and it depends on its concentration in the inspired gas mixture and on how rapidly it is removed by the pulmonary capillary blood. The schematic is drawn as though all 3 gases were administered simultaneously, but this is not necessarily the case. Consider each gas as though it were acting independently of the others.

The partial pressure of carbon monoxide in the pulmonary capillary blood rises very slowly compared with that of the other 2 gases in the figure. (Obviously, a low inspired concentration of carbon monoxide must be used for a very short time in such an experiment.) Nevertheless, if the content of carbon monoxide (in milliliters of carbon monoxide per milliliter of blood) were measured simultaneously, it would be rising very rapidly. The reason for this rapid rise is that carbon monoxide combines chemically with the hemoglobin in the erythrocytes. Indeed, the affinity of carbon monoxide for hemoglobin is about 210 times that of oxygen for hemoglobin. The carbon monoxide that is chemically combined with hemoglobin does not contribute to the partial pressure of carbon monoxide in the blood because it is no longer physically dissolved in it.

Perfusion Limitation

The partial pressure of nitrous oxide in the pulmonary capillary blood equilibrates very rapidly with the partial pressure of nitrous oxide in the alveolus because nitrous oxide moves through the alveolar-capillary barrier very easily and because it does not combine chemically with the hemoglobin in the erythrocytes. After only about 0.1 of a second of exposure of the pulmonary capillary blood to the alveolar nitrous oxide, the partial pressure difference across the alveolar-capillary barrier has been abolished.

As can be seen in Figure 6–1, the time course for oxygen transfer falls between those for carbon monoxide and nitrous oxide. The partial pressure of oxygen rises fairly rapidly (note that it starts at the

Figure 6–2A shows similar graphs of calculated changes in the partial pressure of oxygen in the blood as it moves through a pulmonary capillary. The alveolar

The effect of a low alveolar partial pressure of oxygen on oxygen transfer from the alveolus to the capillary is seen in Figure 6–2B. The low alveolar

The time course of carbon dioxide transfer from the pulmonary capillary blood to the alveolus is shown in Figure 6–3. In a normal person with a mixed venous partial pressure of carbon dioxide of 45 mm Hg and an alveolar partial pressure of carbon dioxide of 40 mm Hg, an equilibrium is reached in a little more than 0.25 of a second, or about the same time as that for oxygen. This may seem surprising, considering that the diffusivity of carbon dioxide is about 20 times that of oxygen, but the partial pressure gradient is normally only about 5 mm Hg for carbon dioxide, whereas it is about 60 mm Hg for oxygen. Carbon dioxide transfer is therefore normally perfusion-limited, although it may be diffusion-limited in a person with an abnormal alveolar-capillary barrier, as shown in the figure.

It is often useful to determine the diffusion characteristics of a patient’s lungs during their assessment in the pulmonary function laboratory. It may be particularly important to determine whether an apparent impairment in diffusion is a result of perfusion limitation or diffusion limitation.

The diffusing capacity (or transfer factor) is the rate at which oxygen or carbon monoxide is absorbed from the alveolar gas into the pulmonary capillaries (in milliliters per minute) per unit of partial pressure difference (in millimeters of mercury). The diffusing capacity of the lung (for gas x), DLx, is therefore equal to the uptake of gas x,

This is really just a rearrangement of the Fick equation given at the beginning of this chapter. The terms for area, diffusivity, and thickness have been combined into DLx and the equation has been rearranged:

The mean partial pressure of oxygen or carbon monoxide is, as already discussed, affected by their chemical reactions with hemoglobin, as well as by their transfer through the alveolar-capillary barrier. For this reason, the diffusing capacity of the lung is determined by both the diffusing capacity of the membrane (both the alveolar-capillary membrane and the plasma membrane of the erythrocyte), DM, and the reaction with hemoglobin, expressed as θ × Vc, where θ is the volume of gas in milliliters per minute taken up by the erythrocytes in 1 mL of blood per millimeter of mercury partial pressure gradient between the plasma and the erythrocyte and Vc is the capillary blood volume in milliliters. (The units of θ × Vc are therefore mL/min/mm Hg.) The diffusing capacity of the lung, DL, can be shown to be related to DM and θ × Vc as follows:

DA, or diffusion through the alveolus, is normally very rapid and usually can be disregarded; however, in conditions such as alveolar pulmonary edema or pneumonia it may be a major problem.

Carbon monoxide is most frequently used in determinations of the diffusing capacity because the mean pulmonary capillary partial pressure of carbon monoxide is virtually zero when nonlethal alveolar partial pressures of carbon monoxide are used:

Several different methods are used clinically to measure the carbon monoxide diffusing capacity and involve both single-breath and steady-state techniques, sometimes during exercise. The DLCO is decreased in diseases associated with interstitial or alveolar fibrosis, such as idiopathic pulmonary fibrosis, sarcoidosis, scleroderma, and asbestosis, or with conditions causing interstitial or alveolar pulmonary edema, as indicated in Table 6–1. It is also decreased in conditions causing a decrease in the surface area available for diffusion, such as emphysema, tumors, a low cardiac output, or a low pulmonary capillary blood volume, as well as in conditions leading to ventilation-perfusion mismatch, which effectively decreases the surface area available for diffusion.

We have now covered most of the basic principles of gas exchange between the ambient air and the pulmonary capillary blood. We have also discussed most of the standard pulmonary function tests. Figure 6–4 is a “decision tree” to use to help interpret pulmonary function test data as they relate to restrictive and obstructive respiratory diseases. Results of each pulmonary function test must be compared to those predicted for a normal healthy person of comparable height (and weight), age, sex, and race. Because most of the tests require the subject to understand what he or she must do, as well as cooperate and give a good effort while performing the test, the person administering the test must judge whether the test is valid. Note that although pulmonary function tests are discussed separately in this book, in current practice, a single machine or workstation can be used to perform several of the tests.

Restrictive diseases include all of those that result in difficulty inspiring, such as those causing decreased lung compliance, for example, fibrosis or atelectasis; or those causing decreased chest wall compliance, for example, kyphoscoliosis or obesity. Disorders that impede lung expansion, such as pneumothorax or hemothorax, or those that weaken the inspiratory muscles, result in patterns similar to those of restrictive diseases. Lung volumes and capacities will be lower than predicted (see Figure 3–3), lung and/or chest wall compliance curves will be shifted to the right (see Figure 2–7), and the FEV1/FVC will be normal or even slightly greater than predicted. The carbon monoxide diffusing capacity can be low (fibrosis) or close to normal (kyphoscoliosis).

Obstructive diseases can be divided into 3 groups. Fixed obstructions, such as foreign bodies or tumors in large airways, increase airways resistance during both inspiration and expiration (see Figure 2–25A). Variable inspiratory obstructions, which are associated with extrathoracic upper airway problems such as obstructive sleep apnea or a paralyzed vocal cord, affect inspiration more than expiration (see Figure 2–25B). The FEV1/FVC is usually normal. Variable expiratory obstructions (see Figure 2–25C), which are associated with intrathoracic airway problems, result in a low FEV1/FVC and a low peak expiratory flow (PEF). They include asthma, chronic bronchitis, and emphysema. A patient may have 2 or all 3 of these disorders together. If 1 of the 3 diseases predominates, it can be distinguished by the interpretation of the patient’s pulmonary function tests. Asthma is an episodic disease diagnosed by an improvement in the FEV1/FVC after a bronchodilator, and/or a worsening of the FEV1/FVC with bronchoprovocation. Chronic bronchitis, which is characterized by copious sputum production, is associated with a normal carbon monoxide diffusing capacity, high residual volumes and functional residual capacities, low arterial

How would each of the following conditions or circumstances be expected to affect the diffusing capacity (DL) of the lungs? Explain your answers.

a. Changing from the supine to the upright position

b. Exercise

c. Valsalva maneuver

d. Anemia

e. Low cardiac output due to blood loss

f. Diffuse interstitial fibrosis of the lungs

g. Emphysema

Explanations of how each condition or circumstance affects the diffusing capacity of lungs follow:

a. Changing from the supine to the upright position slightly decreases the diffusing capacity by decreasing the venous return because of pooling of blood in the extremities and abdomen. The decreased venous return decreases the central blood volume and may slightly decrease the right ventricular output, resulting in derecruitment of pulmonary capillaries and decreased surface area for diffusion.

b. Exercise increases the diffusing capacity by increasing the cardiac output. This recruits previously unperfused capillaries, increasing the surface area available for diffusion. Oxygen transfer across the alveolar-capillary barrier will also increase because at high cardiac outputs the linear velocity of the blood moving through the pulmonary capillaries increases and there is less perfusion limitation of oxygen transfer.

c. A Valsalva maneuver (an expiratory effort against a closed glottis) greatly decreases the pulmonary capillary blood volume and therefore decreases the diffusing capacity.

d. Anemia decreases the diffusing capacity by decreasing the hemoglobin available to chemically combine with oxygen. The partial pressure of oxygen in the plasma in the pulmonary capillaries therefore equilibrates more rapidly with the alveolar

e. A low cardiac output because of blood loss decreases the diffusing capacity by decreasing the venous return and the central blood volume. Pulmonary capillary blood volume decreases, resulting in derecruitment and decreased surface area for diffusion.

f. Diffuse interstitial fibrosis of the lungs thickens the alveolar-capillary barrier (as does interstitial edema), resulting in decreased diffusion of gases across the alveolar-capillary barrier, in accordance with Fick's law.

g. Emphysema destroys the alveolar interstitium and blood vessels, decreasing the surface area for diffusion.

A 65-year-old man came to a family physician’s office because of shortness of breath, which he says he first noticed during strenuous physical activity approximately 2 years ago. The dyspnea became progressively worse and now affects him even during mild activity. He has a dry cough that seems to be increasing in frequency. Physical examination revealed an higher than normal respiratory rate with low tidal volumes, clubbing of his fingers, and “crackles” that sound like Velcro being pulled apart that could be heard with a stethoscope from the bases of both lungs during inspiration. The family physician referred the patient to a pulmonologist.

Pulmonary function tests showed a restrictive pattern of lung volumes and capacities like the middle part of Figure 3–3. The functional residual capacity (FRC), total lung capacity (TLC), and residual volume (RV) were all below the predicted values for the patient, as were his forced expiratory volume in the first second (FEV1) and forced vital capacity (FVC). However, his FEV1/FVC was slightly greater than predicted. His diffusing capacity for carbon monoxide (DLCO) was only about one third of predicted. Pulse oximetry revealed lower than predicted arterial oxygen saturation (SaO2). Both the SaO2 and DLCO worsened with exercise.

The chest radiograph and a high-resolution CT scan showed patchy subpleural reticular opacities. A lung biopsy showed a heterogeneous pattern with alternating areas of normal lung and interstitial inflammation and fibrosis. The patient had no history of treatment with or use of drugs associated with pulmonary fibrosis, environmental or occupational exposure to agents known to cause fibrosis, or previous lung infection. There was also no evidence of any systemic disease associated with pulmonary fibrosis, such as sarcoidosis or scleroderma.

The pulmonologist makes a preliminary diagnosis of interstitial lung disease, based on the patient’s history, physical examination, chest X-ray and CT scan, and restrictive pattern of pulmonary function tests (Figure 6–4). Obstructive lung disease is ruled out by the slightly increased FEV1/FVC. The diagnosis of idiopathic pulmonary fibrosis (IPF) is made by ruling out all known causes of pulmonary fibrosis, such as those mentioned in the previous paragraph.

Pulmonary fibrosis causes low lung compliance and increased inspiratory work of breathing, which leads to the lower than predicted lung volumes and capacities, low tidal volume and high resting respiratory rate, and low FEV1 and FVC. The lower than predicted DLCO and SaO2 result from impaired diffusion through the alveolar-capillary barrier and some degree of ventilation-perfusion mismatching, probably caused by the patchy pattern of the pathology. The dry cough and crackles apparently are caused by the fibrotic tissue itself. Clubbing of fingers is common in many diseases, including some that cause chronic hypoxemia like IPF. The mechanism by which these diseases cause clubbing is not known, although growth factors released from the lungs are likely involved.

There are no truly effective treatments for IPF. Corticosteroids and other immunosuppressive drugs are used to try to decrease the fibrotic process, but are not entirely effective. The prognosis is poor, with a mean survival of only 2 to 3 years.