It might be reasonably assumed that the elastic properties of the lungs are due to the yellow elastin fibres of the lung parenchyma, the fibre type that gives most other organs their elasticity. In fact, only about half the elastic recoil of the lungs comes from the elastin fibres in the alveolar walls, bronchioles and capillaries. Also present in the lungs are collagen fibres, which are less easily stretched and limit overexpansion of the lung. The elastin fibres act in a rather complicated way to provide elasticity. The fibres are kinked and bent round each other, and during inspiration unfold and rearrange in a manner that has been likened to the straightening of the fibres of a nylon stocking when it is put on (Fig. 3.6).

In the lung disease emphysema the elastin fibres are degraded and therefore do not recoil so strongly. The lung is easier to inflate; it is more compliant.

In the disease lung fibrosis there is an increase in the elastin and collagen of the interstitial tissues and the lung recoils more strongly, i.e. it is less compliant.

So, about half the elastic recoil of the lungs comes from the elastic properties of their tissues, just as there is recoil in an inflated rubber balloon. The remaining half of the elastic recoil of the lungs comes from their unique structure of millions of tiny bubble-like alveoli, lined with liquid and connected to the atmosphere by a series of tubes (the bronchial tree). The importance of this structure was demonstrated by von Neergaard, in 1929, when he showed that an isolated lung completely filled with water is about twice as easy to inflate as one filled with air. The cause of this change in ease of inflation lies in the removal of the air–liquid surface that lines the millions of spherical bubbles surrounded by lung tissue by filling the air space with water to form one single small air–liquid interface somewhere in the trachea (Fig. 3.7).

To get an idea of what is happening, imagine an enormous bottle with a very narrow neck. When filled with air it has a large internal surface area exposed to the air; when filled with water right up to its neck this surface area is greatly reduced. To complicate matters, the internal surface of the alveoli is curved, which has a significance we will deal with later.

Not only does filling a lung with fluid make it easier to inflate, it also abolishes the hysteresis seen in the normal lung. Hysteresis (Greek, hysterion, to lag behind) means that inflation of the lung follows a different pressure/volume relationship from deflation (Fig. 3.8). It requires a greater pressure to reach a particular lung volume when you are inflating it than to hold it at that volume when you are deflating it. Something is ‘propping the lung alveoli open’. These peculiar changes are due to the nature of bubbles and the properties of the liquid that makes up the liquid lining of the lung. We need to consider the nature of the surface of liquids, the nature of bubbles and the very special properties of the liquid lining of the alveoli to understand static and dynamic compliance of the lungs.

That liquids form a clear boundary between themselves and the air above them could almost be a definition of a liquid. That this boundary is under tension or stress is clearly seen in the liquids we come across in everyday life: the surface of a cup of coffee, if touched lightly with a spoon, seems to leap up to the spoon. This is the effect of surface tension (Fig. 3.9).

The ‘skin’ or surface of a liquid exists because at the surface there is an imbalance of the forces acting on the molecules at the surface, as explained in the Appendix (p. 153).

Just as mechanical and chemical systems move to a state of minimal energy (maximum entropy), so surfaces seek minimal energy and hence minimal area (this is why water droplets are spheres, the shape that has minimal surface for a given mass). The tendency to reduce in surface area produces tension in a liquid surface which can be measured by a surface balance (Fig. 3.10). In this, a bar of metal dips into the liquid and is exposed to the same forces as act on molecules at the surface of liquids, or the coffee spoon. These forces can be measured by a sensitive transducer. The total force will depend on the surface tension and the length of the bar, and the units of surface tension are therefore N m⁻¹. The surface balance was modified to produce the Wilhelmi Balance by the ingenious addition of a movable barrier that can compress the surface of the liquid in the balance trough. Under these circumstances the depth of the liquid alters as the barrier is moved, but this doesn’t matter as it is the air–liquid interface that is important. If pure water is placed in a Wilhelmi Balance, moving the barrier to and fro, expanding and contracting the surface, has no effect on the measured surface tension. If, however, a phospholipid of the type found lining the alveoli of the lungs is placed on the surface of the liquid it spreads out to form a layer between water and air, and the surface tension changes as the barrier advances, forcing the phospholipid molecules closer together, or retreats allowing them to separate. The tension reaches a minimum when all the phospholipid molecules are neatly packed as a single layer on the surface. When the surface area is reduced further the molecules pile up on each other and the tension begins to rise again. It is possible to measure the size of the molecules of the phospholipid from the area of the surface at which the surface tension is minimal.

Anyone who has indulged in the congenial occupation of blowing soap bubbles using a child’s pipe or wire loop will have demonstrated most of the basic physical principles underlying elastic recoil of the lung due to its liquid lining.

They will have noticed that once a complete sphere has formed the bubble is stable. While there is a hole in the bubble, however, through the pipe or wire loop, if you stop blowing the bubble immediately collapses, returning to a flat layer of soap stretched across the pipe or loop. The molecular basis of this effect is explained in the Appendix (p. 156).

The same forces that produce this collapse are at work in the liquid lining the alveoli of the lung. The fact that the outer surface of the ‘bubble’ is in contact with lung tissue does not alter the tendency of this bubble to collapse, driving air out through the ‘hole’ which is its connection with the bronchial tree.

The stable, completely spherical, bubble you blew with your wire loop was held inflated by an excess of internal pressure. The relationship between this excess pressure, the surface tension of the liquid of the bubble and the radius of the bubble is described by the Laplace Relationship

where P is the excess pressure (Pa), T is the surface tension (N m⁻¹) of the liquid making up the bubble, and R is the radius (m) of the bubble. The constant 4 appears in this equation because a bubble has two surfaces exposed to the air. For alveoli whose outer surface is in contact with the lung tissue it becomes:

Human alveoli are about 0.1 mm in diameter; if they were lined with interstitial fluid the pressure required to hold them open (the excess pressure inside them above the surrounding intrapleural pressure) would be 3 kPa. This is more than twice the pressure found in normal individuals. The liquid lining of the lungs must therefore be very different from interstitial fluid, with a significantly lower surface tension.

From the Laplace Relationship we can see, somewhat surprisingly, that the pressure in a small bubble can be expected to be greater than the pressure in a large one. This might lead us to anticipate problems in the lungs, as there exists a whole variety of sizes of alveoli, those at the top being larger than those at the bottom (see Chapter 5). Under these circumstances an unstable situation might be expected to arise, with small alveoli (containing higher pressure) emptying into large ones (containing lower pressure) (Fig. 3.11A). The nature of the liquid lining of the lungs provides an ingenious solution to this problem.

The liquid lining of the lungs can be extracted by washing them out with saline (bronchial lavage). The washed-out liquid can then be investigated in a Wilhelmi Balance, where it shows some interesting and useful properties. Adding the extract to water in the balance reduces the surface tension from 70 to 40 × 10⁻³ N m⁻¹. Surface tension falls even further when the surface is compressed, reaching a minimum below 10 × 10⁻² N m⁻¹ This effect is due to surfactant secreted by type II cells of the alveolar epithelium (Fig. 3.12).

The surfactant spreads over the inner surface of the alveoli and into the bronchioles. It is made up of dipalmitoyl phosphatidylcholine: the structure of this and the way it arranges itself at an air–liquid interface is shown in Figure 3.13.

The straight structure of these molecules enables them to pack more closely during expiration than would other shapes. This packing and unpacking during inspiration and expiration causes surface tension to decrease and increase in a manner that produces the characteristic hysteresis of the lungs. The change in surface tension also provides the solution to the problem posed earlier of the alveoli of different sizes containing different pressures, and the tendency for the small to empty into the large (Fig. 3.11A). If the surface tension in the large alveolus is sufficiently greater than the tension in the small the effect of the larger radius (R) in the Laplace Relationship

will be matched or even overpowered by changes in tension (T).

And this is what happens in the lung. The surface tension changes to match the radius, so that all alveoli contain about the same air pressure and the small do not tend to empty into the large (Fig. 3.11B).

The air pressure inside an alveolus not only resists the effects of surface tension causing the alveolus to collapse, but also resists exudation of fluid from pulmonary capillaries into the alveolar space. If surface tension is reduced at any particular air pressure within an alveolus more of that air pressure will be available to resist exudation and prevent pulmonary oedema.

The presence of surfactant is clearly important to normal lung function. It:

This important substance appears rather late in human embryological development, at about 30 weeks, and fetuses above that age which are not producing surfactant are stimulated to do so by the administration of corticosteroids.

Even in an isolated lung the relationship between pressure and volume is complicated by the interaction of the surface-active forces already dealt with and the elastic tissue of the lung. One of the most important aspects of this is the tendency for alveoli to close at low lung volumes despite the assistance from surfactant to stay open.

Because of the pressure of overlying tissue the alveoli in the lower parts of the lung are always smaller than those at the top (except at total lung capacity), and therefore show a greater tendency to collapse. The lung volume at which the airways in the lower part of the lung begin to close (Range 1) is known as closing volume. This has considerable significance in life, as regions of the lung closed off from the atmosphere are functionally useless. In young people closing volume is less than functional residual capacity (FRC) and all is well, but by the average age of 66 closing volume equals FRC, with the unfortunate consequence of increased airway closure and reduced ventilation of the lower lung.

In respiratory physiology compliance is defined as the change in volume produced by a change in pressure across the wall of the structure being investigated: the lungs alone, the lungs and chest wall, or the chest wall alone (which is very rarely measured).

For the lungs alone, in life at least, the appropriate pressure measurement is from the alveoli to the intrapleural space. Change in volume can easily be measured using a spirometer, and as we are interested in measuring change in pressure we can use the change in pressure in the oesophagus within the chest as an indication of the change in pressure of the intrapleural space, which is difficult to measure in subjects or patients. Intraoesophagal pressure is usually measured by introducing a small air-filled balloon attached to a pressure transducer by a catheter into the oesophagus via the nose. Because intrapleural pressure varies with the position in the chest it is usual to standardize the position of measurement by pushing the balloon in 30–35 cm from the tip of the nose.

Measurement of total compliance (lungs + chest wall) is much easier because in this case the appropriate pressure measurement is from alveoli (or mouth pressure under static conditions) to atmosphere.

Neither pressure nor volume has the dimensions of time, and therefore these measurements can be made under static conditions where the subject breathes in slightly, relaxes his respiratory muscles while measuring mouth pressure with a transducer; he then breathes in a known volume and again relaxes against the transducer. The slope of the pressure–volume graph so formed is his total static compliance (Fig. 3.14).

The slope of the line that makes up the loop in Figure 3.15 is the compliance of the lungs, which in this case have been taken out of the body. This loop describes the condition of the lungs from total collapse to maximum volume.

In normal breathing there are about 3.0 L of air left in the lungs at the end of expiration (FRC), and we seldom inhale to total lung capacity (TLC). Most quiet breathing takes place within the shaded area of Figure 3.15, which shows less hysteresis than the total collapse to total volume manoeuvre, but still shows that compliance reduces due to airway collapse when residual volume (RV) is approached.

The lung compliance of an average healthy young male is about 2 L kPa⁻¹. That of a female is slightly greater, even taking into account the fact that lung compliance depends on lung size, and therefore body size. This effect is taken into account by measuring specific lung compliance (sC_L), which is compliance divided by maximal lung volume.

The nature of specific compliance can be visualized if you consider the inflation of one or more balloons by a pump that provides as much air as you like, but only to a maximum pressure of 1 kPa above atmospheric.

If you connect a single balloon to the pump and it inflates 2 L the balloon has a compliance (volume increase/pressure) of 2 L kPa⁻¹ (Fig. 3.16A).

If two balloons are connected to the pump they will both be subjected to a pressure of 1 kPa and increase in volume by 2 L each (4 litres in all), which gives the sys-tem a compliance (volume increase/pressure) of 4 L kPa⁻¹ (Fig. 3.16B).

The ‘static conditions’ referred to above and in Figure 3.14 consist of the lung neither inhaling nor exhaling while pressure is being measured. These no-flow conditions only occur at two points during normal breathing: at the peak of inspiration and at the trough of expiration (Fig. 3.17).

This fact can be used to measure compliance while the subject is breathing – dynamic compliance – by measuring intrapleural pressure and lung volume at the same time (Fig. 3.17A). Alternatively, these two variables can be displayed as a loop (Fig. 3.17B), and in this case the angle of the long axis of the loop (volume/pressure) represents dynamic compliance. In this case the area of the loop represents the work of breathing.

In healthy lungs the ratio between dynamic compliance and static compliance is fairly constant at all frequencies of breathing. In obstructive lung diseases such as asthma the obstructed areas of the lung do not have sufficient time to fill or empty, a condition which worsens at high frequencies of breathing, resulting in a fall in the ratio as breathing frequency increases.

The term chest wall compliance (C_W) is frequently used to describe what should be called thoracic cage compliance, because the diaphragm and the abdominal contents pressing on it represent an important element of this component of respiratory mechanics which, unlike the lungs, contains no element of surface tension.

At the end of expiration the lungs do not collapse totally because the thoracic cage is holding them out in a slightly expanded condition. This means that the thorax is slightly pulled in. Because of this the elasticity of the thorax initially helps inspiration. The thorax reaches its neutral position at about two-thirds vital capacity, and after that the direction of its elastic forces is in favour of expiration (Fig. 3.18).

Perhaps surprisingly, considering their very different structures, the compliance of the thoracic cage is about the same as that of the lungs (2 litres kPa⁻¹). This truly elastic measurement is very difficult to make as it can only be properly measured when the respiratory muscles are totally relaxed – a very difficult thing for the conscious subject to do and best obtained in the anaesthetized patient.

Just like in the lungs, thoracic cage compliance is influenced by disease, and perhaps even more than the lungs by posture and position. Ossification of the costal cartilages, and scars resulting from burns of the chest, reduce compliance. The diaphragm passively transmits intra-abdominal pressure from obesity, venous congestion and pregnancy, and by this mechanism can reduce the static compliance of the total respiratory system by up to 60% as a result of changes in posture and changes from the supine to the prone position.

Because the lungs fit inside the thorax, rather like the innertube inside a tyre, they must be treated as elements in parallel rather than in series when their properties are added together (Fig. 3.19).

We can see that the pressure gradient for both the lungs and chest wall is from the intrapleural space to the atmosphere (or very nearly atmosphere, in the case of the lung alveoli). They are therefore in parallel with each other in terms of pressure gradients. In adding lung and chest wall compliance to give total compliance we must therefore use the relationship appropriate for parallel structures:

Because the compliance of the lungs and that of the chest wall are about the same (2 L kPa⁻¹), an artificial ventilator needs to apply twice the normal change in intrapleural pressure to the air in the lungs of a paralysed patient to produce the normal volume change.

In asthma there is no change in compliance: the pressure–volume relationship of the lungs is moved bodily upward without any change in slope.

Bangham, A. D. Lung surfactant: how it does and does not work. Lung. 1987; 165:17.

Cotes, J. E. Lung Function: Assessment and Application in Medicine, fifth ed. Oxford: Blackwell Science; 1993.

De Troyer, A. The respiratory muscles. In Crystal R. G., West J. B., Barnes P. J., Weibel E. R., eds. : The Lung: Scientific Foundations, second ed., New York: Raven Press, 1997.

Haddad, D. F., Greene, S. A., Olver, R. E. Core Paediatrics and Child Health. Edinburgh: Churchill Livingstone; 2000.

Macklem, P. T. Respiratory mechanics. Ann. Rev. Physiol.. 1978; 40:157.

Rahn, H., Otis, A. B., Chadwick, L. E., Fenn, W. O. The pressure/volume diagram of the thorax and lung. Am. J. Physiol.. 1946; 146:161–178.