Medical Physiology, 3rd Edition

Static Properties of the Lung

The balance between the outward elastic recoil of the chest wall and the inward elastic recoil of the lungs generates a subatmospheric intrapleural pressure

The interaction between the lungs and the thoracic cage determines VL. The lungs have a tendency to collapse because of their elastic recoil, a static property represented by the inwardly directed arrows in Figure 27-1A. The chest wall also has an elastic recoil. However, this elastic recoil tends to pull the thoracic cage outward (see Fig. 27-1B). The stage is thus set for an interaction between the lungs and the chest wall: at equilibrium, the inward elastic recoil of the lungs exactly balances the outward elastic recoil of the chest wall (see Fig. 27-1C). This interaction between lungs and chest wall does not occur by direct attachment but via the intrapleural space between the visceral and parietal pleurae (see p. 597). This space is filled with a small amount of pleural fluid and is extremely thin (5 to 35 µm). Because the lungs and chest wall pull away from each other on opposite sides of the intrapleural space, the intrapleural pressure (PIP) is less than barometric pressure (PB); that is, the intrapleural space is a relative vacuum. Although the designation PIP implies that we are referring exclusively to the intrapleural space, this description is not entirely accurate. Indeed, PIP is probably similar to the pressure in several other regions of the chest cavity in addition to the intrapleural space:

1. The virtual space between the chest wall or diaphragm and the parietal pleura

2. The virtual space between the lung and the visceral pleura

3. The interstitial space that surrounds all pulmonary airways

4. Around the heart and vessels

5. Around and—to the extent that smooth-muscle tone can be neglected—inside the esophagus


FIGURE 27-1 Opposing elastic recoils of the lungs and chest wall.

It is helpful to think of PIP as the intrathoracic pressure—the pressure everywhere in the thorax except in the lumens of blood vessels, lymphatics, or airways. imageN27-1


Measurement of PIP

Contributed by Walter Boron

Measuring PIP is intrinsically difficult because the space between the visceral and parietal pleurae is very thin (5 to 35 µm). The approaches include use of a pleural needle, pleural catheter, esophageal balloon, pleural balloon, or rib capsule (embedded in a rib) in direct contact with pleural fluid.

Each of these methods reports a vertical gradient in pleural pressure, on the order of 0.5 cm H2O/cm in head-up dogs. This pressure gradient drives a downward viscous flow of pleural fluid, presumably along the flat surfaces of the ribs. According to a model, recirculation of pleural fluid would be achieved by an upward flow of fluid along the margins of adjacent lobes of the lungs (here the fluid-filled space is larger, which leads to a reduced resistance), energized by movements of the beating heart and ventilating lungs. Finally, a transverse flow of pleural fluid from these margins back to the flat portions of the ribs would complete the circuit.


Lai-Fook SJ. Pleural mechanics and fluid exchange. Physiol Rev. 2004;84:385–410.

The vacuum is not uniform throughout the intrapleural space. When the subject is upright, the vacuum is greatest (i.e., PIP is least) near the apex of the lungs and progressively falls along the longitudinal axis to its lowest value near the bases of the lungs (Fig. 27-2). If a subject whose lungs are ~30 cm tall has finished a quiet expiration, and if PB is 760 mm Hg, PIP is ~753 mm Hg near the apices of the lungs and ~758 mm Hg near the bases. The PIP gradient is about what one would expect, given the density of the lungs. Note that PIP is subatmospheric throughout the chest cavity. Because respiratory physiologists historically measured these small pressures with water manometers rather than with less sensitive mercury manometers, it has become customary to express PIP in centimeters of H2O relative to a PB of 0 cm H2O. Thus, PIP is about −10 cm H2O at the apex and −2.5 cm H2O at the base of the lungs.


FIGURE 27-2 Intrapleural pressures. The values are those after a quiet expiration (i.e., FRC).

The reasons for the apex-to-base PIP gradient are gravity and posture. When an individual stands vertically on the surface of the earth, gravity pulls the lungs downward and away from the apex of the thoracic cage. This force creates a greater vacuum (i.e., a lower PIP) at the apex. Gravity also pushes the bases of the lungs into the thoracic cavity, reducing the vacuum there. Standing on one's head would invert these relationships. Lying on one's side would create a PIP gradient along a frontal-horizontal axis (i.e., from side to side), although the PIP gradient would be much smaller because the side-to-side dimension of the thorax (and therefore the gradient created by the weight of the lungs) is less than the longitudinal dimension. In outer space, the PIP gradient would vanish. Thus, the local PIP depends on the position within the gravitational field.

For most of the remainder of this book, we ignore the PIP gradient and refer to an average PIP of about −5 cm H2O after a quiet expiration (see Fig. 27-2).

Contraction of the diaphragm and selected intercostal muscles increases the volume of the thorax, producing an inspiration

We have seen that the opposing elastic recoils of the lungs and chest wall create a negative PIP that keeps the lungs expanded. Any change in the balance between these elastic recoils will cause VL to change as well. For example, imagine a healthy person with a functional residual capacity (FRC) of 3 L and a PIP of −5 cm H2O. If that person now develops pulmonary fibrosis, which increases the elastic recoil of the lungs, FRC would decrease because a PIP of −5 cm H2O would no longer be adequate to keep the resting VL at 3 L. Moreover, as the lungs shrink, PIP would become more negative, causing chest volume to decrease as well. Under normal circumstances, the key elastic recoil is the one we control: the elastic recoil of the chest wall, which we change moment to moment by modulating the tension of the muscles of respiration.

The muscles of inspiration expand the chest, increasing the elastic recoil of the chest wall and making PIP more negative. Despite the PIP gradient from the apex to the base of the lungs when no air is flowing at FRC (see Fig. 27-2), the ΔPIP during inspiration is similar throughout the thoracic cavity. Responding to this enhanced intrathoracic vacuum, the lungs expand passively. The increase in VL is virtually the same as the increase in thoracic volume. The muscles that produce a quiet inspiration are called the primary muscles of inspiration and include the diaphragm and many intercostal muscles.

The most important component of the increase in chest volume is the rise in the chest cavity's rostral-caudal diameter, a result of the action of the diaphragm. Stimulated by the phrenic nerves (derived from cervical roots C3 to C5), the diaphragm contracts and moves downward into the abdomen ~1 cm during quiet ventilation.

The external and internal intercostal muscles, innervated by segmental spinal nerves, span the space between adjacent ribs. The action of each such muscle depends partly on its orientation but especially—because of the shape of the rib cage—on its position along the rostral-caudal axis and around the dorsal-ventral circumference of the rib cage. Thus, not all external intercostals are inspiratory, and not all internal intercostals are expiratory. Inspiratory neurons preferentially stimulate the most rostral and dorsal external intercostals and the parasternal internal intercostals, both of which have inspiratory mechanical advantages. imageN27-2 The contraction of these muscles has two consequences (Fig. 27-3A). First, the rib cage and the tissues between the ribs stiffen and are therefore better able to withstand the increasingly negative PIP. Second, thoracic volume increases as (a) ribs 2 through 10 rotate upward and outward, increasing the transverse diameter (bucket-handle effect; see Fig. 27-3B) and (b) the upper ribs rotate the sternum upward and outward, increasing the anterior-posterior diameter (water pump–handle effect).


FIGURE 27-3 Actions of major respiratory muscles. External intercostal muscles slope obliquely between the ribs, mostly forward and downward. Internal intercostal muscles also slope obliquely between the ribs, but mostly backward and downward.


Actions of External and Internal Intercostal Muscles

Contributed by Walter Boron

Contrary to conventional wisdom, recent work shows that not all external intercostal muscles are inspiratory and that not all internal intercostal muscles are expiratory. For an exhaustive analysis of this subject, consult the review below.


De Troyer A, Kirkwood PA, Wilson TA. Respiratory action of the intercostal muscles. Physiol Rev. 2005;85:717–756.

During a forced inspiration, the accessory (or secondarymuscles of inspiration also come into play:

1. Scalenes. These muscles lift the first two ribs.

2. Sternocleidomastoids. These muscles lift the sternum outward, contributing to the water pump–handle effect.

3. Neck and back muscles. These elevate the pectoral girdle (increasing the cross-sectional area of the thorax) and extend the back (increasing the rostral-caudal length).

4. Upper respiratory tract muscles. The actions of these muscles decrease airway resistance.

Relaxation of the muscles of inspiration produces a quiet expiration

During a quiet inspiration, normal lungs store enough energy in their elastic recoil to fuel a quiet expiration, just as stretching of a rubber band stores enough energy to fuel the return to initial length. Thus, a quiet expiration is normally passive, accomplished simply by relaxation of the muscles of inspiration. Thus, there are no primary muscles of expiration.

Expiration is not always entirely passive. One example is a forced expiration in an individual with normal airway resistance. Another is even a quiet expiration of a person with a disease that increases airway resistance (e.g., asthma, chronic bronchitis, emphysema). In either case, the accessory muscles of expiration help make PIP more positive:

1. Abdominal muscles (internal and external oblique, rectoabdominal, and transverse abdominal muscles). Contraction of these muscles (see Fig. 27-3C) increases intra-abdominal pressure and forces the diaphragm upward into the chest cavity, decreasing the rostral-caudal diameter of the thorax and increasing PIP.

2. Intercostals. The most ventral and caudal external intercostals, the most dorsal and lateral of the caudal internal intercostals, and an intercostal-like muscle called the triangularis sterni all have an expiratory mechanical advantage. Expiratory neurons selectively stimulate these muscles so as to reduce both the anterior-posterior and the transverse diameters of the thorax. These actions are particularly important for coughing.

3. Neck and back muscles. Lowering of the pectoral girdle reduces the cross-sectional area of the thorax, whereas flexion of the trunk reduces the rostral-caudal diameter.

During a forced inspiration, the accessory muscles of inspiration use their energy mainly to increase VL (rather than to overcome resistance to airflow); the lungs store this extra energy in their elastic recoil. During a forced expiration, the accessory muscles of expiration use their energy mainly to overcome the resistance to airflow, as discussed below.

An increase of the static compliance makes it easier to inflate the lungs

Imagine that a person experiences a puncture wound to the chest cavity, so that air enters the thorax from the atmosphere, raising PIP to the same level as PB. This condition is called a pneumothorax (from the Greek pneuma [air]). With no vacuum to counter their elastic recoil, alveoli will collapse—a condition known as atelectasis. The upper part of Figure 27-4A illustrates an extreme hypothetical case in which pressure is atmospheric throughout the thorax. Even though the lungs are collapsed, VL is not zero because proximal airways collapse before smaller ones farther downstream, trapping air. The resulting minimal air volume is ~10% of total lung capacity (TLC), typically ~500 mL.


FIGURE 27-4 Collapse and reinflation of the lungs. In A, we assume that PIP rises to PB, so that PTP falls to zero, collapsing the lungs.

We now wish to re-expand the collapsed lungs to their original volume (i.e., FRC). What are the forces at work during such a re-expansion? The force responsible for distending an airway is the transmural pressure (PTM)—the radial pressure difference across an airway wall at any point along the tracheobronchial tree:



PAW is the pressure inside the airway, and PIP is the pressure in the interstitial space surrounding the airway. A special case of PTM is the transmural pressure across the alveolar wall—transpulmonary pressure (PTP):



PA is alveolar pressure. When the glottis is open and no air is flowing, the lungs are under static conditions, and PA must be 0 cm H2O:



Thus, with the glottis open under static conditions, the pressure that inflates the alveoli (i.e., PTP) is simply the negative of PIP. We can re-expand the lungs to FRC by any combination of an increase in PA and a decrease in PIP, as long as PTP ends up at 5 cm H2O (see Fig. 27-4A, lower panels). Thus, it makes no difference whether we increase PA from 0 to +5 cm H2O with PIP fixed at zero (the principle behind positive-pressure ventilation in an intensive care unit) or whether we decrease PIP from 0 to –5 cm H2O with PA fixed at zero (the principle behind physiological ventilation). In both cases, VL increases by the same amount.

A clinician would treat the pneumothorax by inserting a chest tube through the wound into the thoracic cavity and gradually pumping out the intrathoracic air. The clinician might also insert a tube through the mouth and into the upper trachea (to ensure a patent airway), use a mechanical ventilator (to ensure gas exchange), and sedate the patient (to prevent the patient from fighting the ventilator). Between the inspiratory cycles of the ventilator, the lungs are under static conditions and VL depends only on PTP—that is, the difference between PA (which is set by the ventilator) and PIP. As we remove air from the thorax, PIPbecomes more negative and the alveoli re-expand. We can characterize the elastic (or static) properties of the lungs by plotting VL versus PTP as VL increases (see Fig. 27-4B, purple curve). How do we obtain the necessary data? In principle, we could determine VL by using Equation 26-4. We could read off PA (needed to compute PTP) directly from the ventilator. Finally, we could in principle measure PIP by using a pressure transducer at the tip of the chest tube. Most important, we must take our readings between inspiratory cycles of the ventilator—under static conditions.

During the reinflation, measured under static conditions, we can divide the effect on VL into four stages, starting at the left end of the purple curve in Figure 27-4B:

Step 1: Stable VL. In the lowest range of PIP values, making PIP more negative has little or no effect on VL. For example, decreasing PIP from 0 to –1 cm H2O (i.e., increasing PTP from 0 to +1 cm H2O), we record no change in VL. Why? As discussed below, it is very difficult—because of the surface tension created by the air-water interface—to pop open an airway that is completely collapsed. Until PTP is large enough to overcome the collapsing effects of surface tension, a decrease in PIP has no effect on VL.

Step 2: Opening of airways. Decreasing PIP beyond about –8 cm H2O produces VL increases that are at first small, reflecting the popping open of proximal airways with the greatest compliance. Further decreasing PIP produces larger increases in VL, reflecting the expansion of already-open airways as well as recruitment of others.

Step 3: Linear expansion of open airways. After all the airways are already open, making PIP increasingly more negative inflates all airways further, causing VL to increase in a roughly linear fashion.

Step 4: Limit of airway inflation. As VL approaches TLC, decreases in PIP produce ever smaller increases in VL, which reflects decreased airway and chest-wall compliance and the limits of muscle strength.

What would happen if, having inflated the lungs to TLC, we allowed PIP to increase to 0 cm H2O once again? Obviously, the VL would decrease. However, the lungs follow a different path during deflation (see Fig. 27-4B, red curve), creating a PIP-VL loop. The difference between the inflation and the deflation paths—hysteresis—exists because a greater PTP is required to open a previously closed airway, owing to a deficit of surfactant at the air-water interface, than to keep an open airway from closing, due to the abundance of surfactant. We will discuss surfactant in the next section. The horizontal dashed line in Figure 27-4B shows that inflating previously collapsed lungs to FRC requires a PIP of –12 cm H2O (purple point), whereas maintaining previously inflated lungs at FRC requires a PIP slightly less negative than –5 cm H2O (red point). During normal ventilation, the lungs exhibit much less hysteresis, and the green PIP-VL loop in Figure 27-4B lies close to the red deflation limb of our original loop. The changes in VL in Figure 27-4B reflect mainly changes in the volume of alveoli, with a small contribution from conducting airways.

We will now focus on just the red curve in Figure 27-4B, a portion of which is the middle curve in Figure 27-5. Here, PTP is +5 cm H2O when VL is at FRC. As the subject makes a normal inspiration with a tidal volume (VT) of 500 mL, PTP increases (i.e., PIP decreases) by 2.5 cm H2O. The ratio of ΔVL to ΔPTP (i.e., the slope of the PTP-VL curve) is the compliance, a measure of the distensibility of the lungs. In our example,




FIGURE 27-5 Static pressure-volume curves for healthy and diseased lungs.

Because we made this measurement under conditions of zero airflow, C is the static compliance. Static compliance, like VL, is mainly a property of the alveoli. The elastance of the lungs, which is a measure of their elastic recoil, is the reciprocal of the compliance (E = 1/C). Lungs with a high compliance have a low elastic recoil, and vice versa.

Figure 27-5 also shows representative PTP-VL relationships for lungs of patients with pulmonary fibrosis (bottom curve) and emphysema (top curve). In pulmonary fibrosis, the disease process causes deposition of fibrous tissue, so that the lung is stiff and difficult to inflate. Patients with restrictive lung disease, by definition, have a decreased C (i.e., a decreased slope of the VL-PTP relationship in Fig. 27-5) at a given VL. The same ΔPTP that produces a 500-mL VL increase in normal lungs produces a substantially smaller VL increase in fibrotic lungs. In other words, static compliance (ΔVL/ΔPTP) is much less, or elastic recoil is much greater.

In emphysema, the situation is reversed. The disease process, a common consequence of cigarette smoking, destroys pulmonary tissue and makes the lungs floppy. An important part of the disease process is the destruction of the extracellular matrix, including elastin, by elastase released from macrophages. Normal mice that are exposed to cigarette smoke develop emphysema rapidly, whereas the disease does not develop in “smoker” mice lacking the macrophage elastase gene. The same increase in PTP that produces a 500-mL VL increase in normal lungs produces a substantially larger VL increase in lungs with emphysema. In other words, static compliance is much greater (i.e., much less elastic recoil).

Because it requires work to inflate the lungs against their elastic recoil, one might think that a little emphysema might be a good thing. Although it is true that patients with emphysema exert less effort to inflate their lungs, the cigarette smoker pays a terrible price for this small advantage. The destruction of pulmonary architecture also makes emphysematous airways more prone to collapse during expiration, which drastically increases airway resistance.

Two additional points are worth noting. First, compliance (i.e., slope of the PTP-VL curve) decreases as VL increases from FRC to TLC (see Fig. 27-5). Second, the PTP-VL curve is the amalgam of pressure-volume relationships of all alveoli. Different alveoli have different PTP-VL curves and may experience different intrapleural pressures, depending on their position within a gravitational field (see Fig. 27-2). This inhomogeneity of static parameters contributes to regional differences in ventilation (see pp. 687–689Box 27-1).

Box 27-1

Restrictive Pulmonary Disease

Two major categories of pulmonary disease—restrictive and obstructive—can severely reduce total ventilation; that is, the amount of air entering and leaving the lungs per unit of time. We discuss obstructive diseases (which affect the resistance of the conducting airways) in Box 27-2.

Pulmonologists use the term restrictive lung disease in an inclusive sense to refer to any disorder that reduces FRC, vital capacity, or TLC (see Fig. 26-8B) and thereby makes the lungs difficult to inflate. Pure restrictive disease does not affect airway resistance. Restrictive disease can target the lung parenchyma or three extrapulmonary structures, as outlined in the next four paragraphs.

Lung Parenchyma

Restrictive diseases of the lung parenchyma decrease the static compliance of the lung—mainly a property of the alveoli. To overcome increased elastic recoil, the patient must make extra effort to inhale. The patient compensates by making rapid but shallow inspirations. In newborns, an example is infant respiratory distress syndrome, caused by a deficiency in surfactant. Pulmonary edema is a buildup of fluid in the interstitial space between the alveolar and capillary walls and, eventually, the alveolar space. Interstitial inflammation of a variety of etiologies (e.g., infection, drugs, environmental exposure) can lead to the deposition of fibrous tissue and a group of diseases called diffuse interstitial pulmonary fibrosis.


A buildup in the intrapleural space of either air (pneumothorax) or fluid (pleural effusion) can restrict the expansion of a vast number of alveoli.

Chest Wall

Rigidity of the chest wall makes it difficult to increase thoracic volume even if the neuromuscular system (see next) can generate normal forces. Ankylosing spondylitis is an inflammatory disorder of the axial skeleton that may reduce the bucket-handle rotation of the ribs during quiet inspirations and the flexion and extension of the trunk during forced inspirations and expirations. In kyphoscoliosis (angulation and rotation of the spine), deformation of the vertebrae and ribs may reduce ventilation. In both conditions, impairment of coughing predisposes to lung infections.

Neuromuscular System

The central nervous system may fail to stimulate the respiratory muscles adequately, or the muscles may fail to respond appropriately to stimulation. In polio, the virus occasionally attacks respiratory control centers in the brainstem (see p. 702). Amyotrophic lateral sclerosis (ALS or Lou Gehrig disease) leads to the destruction of premotor and motor neurons, including those to the muscles of respiration (see p. 700). Indeed, dyspnea on exertion is a common early symptom of ALS. Certain drug overdoses (e.g., barbiturate poisoning) may temporarily inhibit respiratory control centers in the brainstem. In the absence of supportive therapy (i.e., mechanical ventilation), the respiratory failure can be fatal. The pain that accompanies surgery or other injuries to the chest can also severely limit the ability to ventilate. Local paralysis of intercostal muscles allows the enhanced intrathoracic vacuum to suck in intercostal tissues during inspiration. This paradoxical movement reduces the efficiency of inspiration. Paradoxical movement may also occur with broken ribs, a condition known as flail chest.

Surface tension at the air-water interface of the airways accounts for most of the elastic recoil of the lungs

What is the basis of the elastic recoil that determines the static compliance of the lungs? The elasticity of pulmonary cells and the extracellular matrix (e.g., elastin and collagen), what we might think of as the “anatomical” component of elastic recoil, generally accounts for a small part. The basis of most of the recoil was suggested in 1929 by von Neergaard, who excised lungs from cats and inflated them by applying positive pressure to the trachea under two conditions. When he filled the lungs with air, the PTP-VL curve looked similar to the one we have seen before (Fig. 27-6A, blue curve). However, when he degassed the airways and reinflated them with saline, he found that (1) the PTP-VL relationship (see Fig. 27-6A, orange curve) exhibited far less hysteresis, and (2) the static compliance was substantially greater (i.e., much less pressure was required to inflate the lungs). These changes occurred because the saline-filled lungs lacked the air-water interface that generated surface tension in the air-filled lungs. It is this surface tension that is responsible for a large fraction of the lung's elastic recoil.


FIGURE 27-6 Effect of surface tension on the lung.

Surface tension is a measure of the force acting to pull a liquid's surface molecules together at an air-liquid interface (see Fig. 27-6B). Water molecules in the bulk liquid phase are equally attracted to surrounding water molecules in all directions, so that the net force acting on these “deep” water molecules is zero. However, water molecules at the surface are equally attracted to others in all directions but “up,” where no molecules are available to pull surface water molecules toward the air phase. Thus, a net force pulls surface molecules away from the air-water interface toward the bulk water phase.

We can think of the surface water molecules as beads connected by an elastic band. The force that pulls a water molecule down into the bulk also creates a tension between the molecules that remain at the surface, in a direction that is parallel to the surface. If we try to overcome this tension and stretch the air-water interface (see Fig. 27-6C), thus increasing its area, we must apply force (image) to bring water molecules from the bulk liquid (a low-energy state) to the surface (a high-energy state). If the body of water on which we tug has a length of l, then the surface tension (T) is



For a simple air-water interface at 37°C, the surface tension is ~70 dynes/cm.

A drop of water falling through the air tends to form into a sphere because this shape has the smallest surface area and thus the lowest energy. Put differently, when the drop is spherical, it is impossible for any additional water molecules to leave the surface.

In the reverse scenario, a spherical air bubble surrounded by water (see Fig. 27-6D), unbalanced forces acting on surface water molecules cause them to dive into the bulk, which decreases the surface area and creates tension in the plane of the air-water interface. This surface tension acts like a belt tightening around one's waist. It tends to decrease the volume of compressible gas inside the bubble and increases its pressure. At equilibrium, the tendency of increased pressure to expand the gas bubble balances the tendency of surface tension to collapse it. The Laplace equation describes this equilibrium: imageN27-3


Laplace's Law for a Sphere

Contributed by Emile Boulpaep, Walter Boron

Imagine a soap bubble that is perfectly spherical. At equilibrium, the tendency of the transmural pressure (P) to expand the bubble exactly balances the tendency of the surface tension (T) to collapse the bubble. An infinitesimally small change in the radius of the sphere (dr) would produce infinitesimally small changes in both the area of the sphere (dA) and the volume of the sphere (dV). The infinitesimally small amount of pressure × volume work done in expanding the volume would equal the tension × area work done in expanding the surface area:


(NE 27-1)

Because the volume of a sphere is (4/3)πr3,


(NE 27-2)

Similarly, the area of a sphere is 4πr2. Thus,


(NE 27-3)

Substituting the expressions for dV (see Equation NE 27-2) and dA (see Equation NE 27-3) into Equation NE 27-1, we have


(NE 27-4)

Rearranging and solving for P, we have


(NE 27-5)

In our example of a soap bubble—and also in the hypothetical example of a perfectly spherical alveolus—T is a constant, r is the independent variable, and P is the dependent variable. Thus, a sphere with a smaller radius must have a larger transmural pressure to remain in equilibrium.

The article by Prange discusses the limitations in applying Laplace's law to alveoli. As noted in the text on page 613, the alveoli are neither perfect spheres nor uniform spheres. Moreover, their interiors are interconnected, and individual alveoli often share walls with their adjacent neighbors (principle of interdependence).


Prange HD. Laplace's law and the alveolus: A misconception of anatomy and a misapplication of physics. Adv Physiol Educ. 2003;27:34–40.



P is the dependent variable, the surface tension T is a constant for a particular interface, and the bubble radius r is the independent variable. Therefore, the smaller the bubble's radius, the greater the pressure needed to keep the bubble inflated. See pages 455–457 for a description of how Laplace's treatment applies to blood vessels.

Our bubble-in-water analysis is important for the lung because a thin layer of water covers the inner surface of the alveolus. Just as surface tension at the air-water interface of our gas bubble causes the bubble to constrict, it also causes alveoli and other airways to constrict, contributing greatly to elastic recoil.

The analogy between air bubbles and alveoli breaks down somewhat because an alveolus only approximates a part of a sphere. A second complicating factor is that not all alveoli are the same size; some may have a diameter that is three or four times larger than that of others. Third, alveoli are interconnected.

Figure 27-6E shows what would happen if two imaginary air bubbles in water were connected by a tube with a valve that allows us to make or break the connection between the bubbles. For both, assume that the surface tension T is 70 dynes/cm. The valve is initially closed. The first bubble has a radius of 0.010 cm. The second is only half as wide. At equilibrium, the pressure required to keep the smaller bubble inflated is twice that necessary to keep the larger bubble inflated (see calculations in Fig. 27-6E). If we now open the valve between the two bubbles, air will flow from the smaller bubble to the larger bubble. To make matters worse for the smaller bubble, the smaller it becomes, the greater is the pressure needed to stabilize its shrinking radius. Because its pressure is less than required, air continues to flow out of the smaller bubble until it implodes completely.

In principle, the lung faces a similar problem. Smaller alveoli tend to collapse into larger ones. As we shall see, pulmonary surfactant minimizes this collapsing tendency by lowering surface tension. However, even without surfactant, the collapse of small alveoli could proceed only so far because each alveolus is tethered to adjacent alveoli, which help hold it open—the principle of interdependence.

Why would it matter if many smaller alveoli collapsed into a few larger alveoli? Such a collapse would reduce the total alveolar surface area available for diffusion of O2 and CO2 (see p. 661). Thus, from a teleological point of view, it is important for the lung to keep the alveoli as uniformly inflated as possible.

Pulmonary surfactant is a mixture of lipids—mainly dipalmitoylphosphatidylcholine—and apoproteins

As noted above, surface tension accounts for most of the elastic recoil in normal lungs. However, if it were not for pulmonary surfactant, total elastic recoil would be even higher, and the lungs would be far more difficult to inflate. During quiet breathing, surfactant reduces surface tension to ~25 dynes/cm or less, far below the value of 70 dynes/cm that exists at a pure air-water interface.

The term surfactant means a surface-active agent. Because surfactants have both a hydrophilic region (strongly attracted to water) and a hydrophobic region (strongly repelled by water), they localize to the surface of an air-water interface. An example of a synthetic surfactant is dishwashing detergent. As a younger student, you may have done a simple experiment in which you filled a small-diameter cup with water and carefully floated a thin sewing needle—lengthwise—on the surface. The needle, like an insect that walks on water, is supported by surface tension, which pulls in the plane of the air-water interface. When you add a drop of liquid detergent to the surface of the water, the needle instantly sinks. Why? The detergent greatly reduces the surface tension.

Detergent molecules orient themselves so that their hydrophilic heads point toward (and interact with) the most superficial water molecules, whereas the hydrophobic tails point toward the air (Fig. 27-7). The hydrophilic surfactant heads pull strongly upward on the most superficial water molecules, greatly reducing the net force on these surface water molecules and minimizing their tendency to dive into the bulk water. What prevents surfactant at the air-surfactant interface from diving into the bulk water? The hydrophobic tails exert a counterforce, pulling the surfactant upward toward the air. The situation is not unlike that of a fishing line with a bobber at one end and a sinker at the other: as long as the bobber is sufficiently buoyant, it remains at the water's surface. Thus, unlike surface water molecules, which are subjected to a large net force pulling them into the bulk, surfactant experiences a much smaller net force. The greater the surface density of surfactant molecules at the air-water interface (i.e., the smaller the surface occupied by water molecules), the smaller the surface tension.


FIGURE 27-7 Effect of a surface-active agent on surface tension.

Pulmonary surfactant is a complex mixture of lipids and proteins. Type II alveolar cells (see p. 599), cuboidal epithelial cells that coexist with the much thinner type I cells, synthesize and secrete pulmonary surfactant. Clara cells in the respiratory bronchioles manufacture at least some components of pulmonary surfactant. Lipids make up ~90% of surfactant and are responsible for the surface-active properties. About half of the lipid is dipalmitoylphosphatidylcholine (DPPC). Figure 27-8 shows the synthesis and structure of DPPC (also known as dipalmitoyl lecithin), which contains two fully saturated 16-carbon fatty-acid chains (i.e., palmitates). The second most common lipids in pulmonary surfactant are phosphatidylcholine molecules with unsaturated fatty-acid chains. Compared with cell membranes, phosphatidylglycerol (~11% of lipid) is overrepresented in surfactant.


FIGURE 27-8 Synthesis of DPPC. CDP, cytosine diphosphate; CMP, cytosine monophosphate; CTP, cytosine triphosphate; CoA, coenzyme A; Pi, inorganic phosphate; PPi, inorganic pyrophosphate. imageN27-12


DPPC Synthesis near Birth

Contributed by Ervin Jones, Walter Boron

Dipalmitoylphosphatidylcholine (DPPC)—also known as dipalmitoyl lecithin—is the major component of pulmonary surfactant. As outlined in Figure 27-8, the condensation of diacylglycerol with cytosine diphosphate choline ultimately leads to the production of DPPC. In the fetus, shortly before birth, an increase in plasma cortisol levels upregulates several enzymes that are important for the synthesis of DPPC, including fatty-acid synthase and phosphocholine transferase.

Proteins account for the remaining ~10% of pulmonary surfactant. Plasma proteins (mainly albumin) and secretory immunoglobulin A make up about half of the protein, and four apoproteins (SP-A, SP-B, SP-C, and SP-D) make up the rest. SP-A and SP-D are water soluble and have collagen-like domains (Table 27-1). Both contribute to “innate immunity” by acting as opsonins to coat bacteria and viruses, thereby promoting phagocytosis by macrophages resident in the alveoli. In addition, SP-A imageN27-4 may be important for feedback control that limits surfactant secretion. The two hydrophobic apoproteins, SP-B and SP-C, are intrinsic membrane proteins that greatly increase the rate at which surfactant enters the air-water interface and then spreads as a surface film. The hereditary absence of SP-B leads to respiratory distress that is fatal unless the newborn receives a lung transplant.

TABLE 27-1

Surfactant Apoproteins






Innate immunity
Formation of tubular myelin



Speeds formation of monolayer
Formation of tubular myelin



Speeds formation of monolayer



Innate immunity
Metabolism of surfactant?


Surfactant Apoprotein SP-A

Contributed by Walter Boron


EFIGURE 27-1 Representation of apoprotein SP-A.

The lipid components of pulmonary surfactant enter type II cells from the bloodstream (Fig. 27-9A). Type II cells use the secretory pathway (see pp. 34–35) to synthesize the four apoproteins, all of which undergo substantial post-translational modification. The final assembly of surfactant occurs in lamellar bodies, which are ~1 µm in diameter and consist of concentric layers of lipid and protein (see Fig. 27-9B). Some of the material in these lamellar bodies represents newly synthesized components, and some of it represents recycled surfactant components retrieved from the alveolar surface. Each hour, the normal lung secretes into the alveolar space ~10% of the material present in the lamellar bodies.


FIGURE 27-9 Generation of pulmonary surfactant. In B, the structures with concentric layers are lamellar bodies, which are continuous with tubular myelin (grid-like structures). (B, Courtesy of Dr. M. C. Williams, University of California, San Diego.)

The secretion of pulmonary surfactant occurs by constitutive exocytosis (see p. 35). In the fetus, both synthesis and secretion are quite low until immediately before birth, when a surge in maternal glucocorticoid levels triggers these processes (see p. 1156). Infants born prematurely may thus lack sufficient levels of surfactant and may develop infant respiratory distress syndrome (IRDS; see p. 1156). In postnatal life, several stimuli enhance the surfactant secretion, including hyperinflation of the lungs (e.g., sighing and yawning), exercise, and pharmacological agents (e.g., β-adrenergic agonists, Ca2+ionophores).

After its secretion into the thin layer of water that covers the alveolar epithelium, freed from the physical constraints of confinement to a lamellar body, pulmonary surfactant undergoes major structural changes. In this aqueous layer, surfactant takes on the form of a meshwork known as tubular myelin (see Fig. 27-9B), which is rich in surfactant apoproteins. It is not clear whether surfactant normally passes through the tubular-myelin state before forming a surface film at the air-water interface. However, tubular myelin is not required; SP-A knockout mice lack tubular myelin but have a normal surface film.

Two mechanisms remove components of pulmonary surfactant from the surface of alveoli. Alveolar macrophages degrade some of the surfactant. Type II cells take up the rest, and either recycle or destroy it.

Pulmonary surfactant reduces surface tension and increases compliance

The pulmonary surfactant present at the alveolar air-water interface has three major effects.

First, because surfactant reduces surface tension, it increases compliance, making it far easier to inflate the lungs. If surfactant suddenly disappeared from the lungs, mimicking the situation in IRDS, total elastic recoil would increase (i.e., compliance would decrease) twofold or more, causing small airways to collapse partially. The situation would be similar to that described by the fibrosis curve in Figure 27-5. Because the compliance of the lungs is far lower than normal, an infant with IRDS—compared with a normal infant—must produce far larger changes in PTP (or PIP) to achieve the same increase in VL. Therefore, infants with low surfactant levels must expend tremendous effort to contract their inspiratory muscles and expand the lungs.

Second, by reducing surface tension, surfactant minimizes fluid accumulation in the alveolus. In the absence of surfactant, the large surface tension of the liquid layer between the air and the alveolar type I cells would cause the “air bubble” to collapse, drawing fluid into the alveolar space from the interstitium. The net effect would be to increase the thickness of the liquid layer and thereby impair gas diffusion. With normal levels of surfactant, the surface tension of the water layer is low, and the tendency to draw fluid from the interstitium to the alveolar space is balanced by the negative interstitial hydrostatic pressure (i.e., PIP), which favors fluid movement from the alveolar space into the interstitium.

Third, surfactant helps keep alveolar size relatively uniform during the respiratory cycle. Imagine that we start—after a quiet expiration—with two alveoli having the same radius (e.g., 100 µm) and the same surface density of surfactant (to yield a surface tension of 20 dynes/cm), as indicated by the inner dashed circles of the two alveoli in Figure 27-10. However, either the conducting airway leading to the lower alveolus has a higher resistance or the lower alveolus itself has more fibrous tissue. Either way, the lower alveolus inflates more slowly during inspiration and—at any time—has a smaller volume than the upper one. This size difference has two negative consequences: (1) The total surface area of the two alveoli is less than if they had inflated equally, which impairs gas diffusion. (2) Because the final volume increase of the upper alveolus may be greater than that of the lower one, its ventilation may be greater. Such unevenness of ventilation impairs effective gas exchange (see p. 693).


FIGURE 27-10 Braking action of surfactant on inflation.

Fortunately, surfactant helps alveoli dynamically adjust their rates of inflation and deflation, so that ventilation is more uniform among alveoli. During rapid inflation, the alveolar surface expands more rapidly than additional surfactant can reach the surface from a surfactant pool beneath the surface. Thus, surfactant on the surface is thought to break up like a flow of ice on the sea, with open areas of pure water between clusters of surfactant. With more exposed water at the surface, surface tension increases. Surface tension may double during inspiration, compared with the resting value at FRC. This effect would be exaggerated in rapidly expanding alveoli, which would develop a higher surface tension more quickly than slowly expanding alveoli. This higher surface tension produces a greater elastic recoil that opposes further expansion. Thus, the dilution of surfactant tends to put more of a brake on rapidly expanding airways, slowing their expansion to more nearly match that of alveoli that tend to inflate more slowly. imageN27-5


Role of Surfactant in the “Static” Pressure-Volume Loop

Contributed by Emile Boulpaep, Walter Boron

The classic static pressure-volume loop exhibits considerable hysteresis (see large loop formed by purple and red curves in Fig. 27-4). This hysteresis is due in large part to the movement of surfactant into the air-water interface as the lung inflates from RV to TLC, and to the movement of surfactant out of the air-water interface as the lung deflates from TLC to RV.

During normal, quiet breathing, the pressure-volume loop (see small green loop in Fig. 27-4B) lies near the deflation limb (red in Fig. 27-4B) of the “classic” loop. The quiet-breathing loop differs from the classic loop in three respects. First, the green quiet-breathing loop covers a much smaller range of pressures and volumes. Second, the green quiet-breathing loop exhibits very little hysteresis because relatively less surfactant moves into the air-water interface during inspiration and out of the air-water interface during expiration at these small VT values. Third, at comparable lung volumes, the average compliance in the quiet-breathing loop (i.e., the slope of a line drawn between the two ends of the loop) is less than the static compliance of the classic loop's deflation limb. Again, this difference arises because during quiet breathing, less surfactant is present at the air-water interface, causing the lungs to be stiffer.

During quiet breathing, the pool of surfactant available to move into and out of the air-water interface with each VT gradually decreases as the result of the catabolism of surfactant (see Fig. 27-9). As a result, the lungs gradually stiffen. In other words, the fall in compliance makes the lungs more difficult to inflate. In fact, some alveoli collapse, a condition termed atelectasis. This collapse is particularly prevalent in dependent regions of the lungs (e.g., the base of the lung in an upright individual), which reflects the relatively less negative PIP near the base of the lung (see Fig 27-2) and thus the smaller degree of inflation. Fortunately, stretch receptors in the lung sense this atelectasis and trigger a large inspiratory effort to TLC—a sigh or a yawn (see Box 32-4). Sighing and, even more so, yawning are powerful stimuli for trafficking surfactant to the air-water interface, increasing compliance and reversing the atelectasis.

The opposite appears to happen during expiration, when the surface area of rapidly contracting alveoli falls more rapidly than surfactant can dive back down into the subsurface pool. The compression of surfactant causes surface tension to fall precipitously. Surface tension during expiration may fall to half the resting value at FRC. The more rapidly an alveolus shrinks, the more quickly its surface tension falls, the lower is its elastic recoil, and the greater is its tendency to re-expand. This action puts a brake on rapidly contracting alveoli, slowing their rate of shrinkage to more closely match that of slowly contracting alveoli. These changes in surfactant contribute to the small amount of hysteresis in a PIP-VL loop during quiet breathing (green loop in Fig. 27-4B).