Lung Mechanics and Surfactant

Fill an excised lung with saline instead of air and it takes far less pressure to inflate it to the same volume. The tissue is the same both times; the saline removes the air-liquid surface lining every alveolus. Gas exchange and breathing gave the first look at surfactant and compliance; here they get numbers.

The lung follows the chest because pleural pressure is negative

The visceral pleura covers each lung and the parietal pleura lines the chest wall, with a thin film of fluid between. The stretched lung recoils inward; the chest wall springs outward. At functional residual capacity (FRC), the volume left after a quiet expiration, the two recoils balance, and because they pull the pleural layers apart, pressure in the film is below atmospheric. Picture two springs hooked together, one stretched and one compressed, resting where their forces cancel.

Three pressures, each relative to atmospheric and in cmH2O (1 mmHg = 1.36 cmH2O): pleural pressure in the film, alveolar pressure in the air spaces, and transpulmonary pressure, alveolar minus pleural, the pressure across the lung that holds it open.

In a healthy adult at rest, pleural pressure is about -5 cmH2O at FRC (about -3.7 mmHg). As the diaphragm contracts it falls to about -8 cmH2O, pulling alveolar pressure about 1 cmH2O below atmospheric while air flows in. In expiration alveolar pressure rises about 1 cmH2O above atmospheric and pleural pressure returns to -5. When flow stops, alveolar pressure is zero. So pleural pressure is not a constant “-4 mmHg”: it swings about 3 cmH2O each quiet breath, is higher (less negative) in dependent regions, and rises above atmospheric in forced expiration. An esophageal balloon estimates it near the esophagus. The inspiratory fall also aids venous return (cardiac output and venous return). If air enters the pleural space, a pneumothorax, the lung recoils toward collapse.

Compliance is the change in volume per change in distending pressure, measured at zero flow so that only elastic forces act; elastance is its inverse. Lung and chest wall are in series, so their elastances add: 1/Crs = 1/CL + 1/Ccw, where Crs is the whole respiratory system’s compliance.

Worked problem, part 1: compliance and elastic work. Resting values above, tidal volume 0.50 L, zero flow at FRC and end-inspiration, a straight pressure-volume line, lung only.

1. Alveolar pressure is 0 at both points, so transpulmonary pressure is 0 - (-5) = 5 cmH2O at FRC and 0 - (-8) = 8 cmH2O at end-inspiration.
2. Lung compliance = 0.50 L / 3 cmH2O = 0.17 L per cmH2O, near the textbook adult figure of about 0.2.
3. Elastic work of the added stretch, above the 5 cmH2O baseline that chest-wall recoil balances at FRC, is the triangle under the line: 0.5 x 3 cmH2O x 0.50 L = 0.75 cmH2O x L. Since 1 cmH2O x L = 98.07 Pa x 0.001 m3 = 0.098 J, that is about 0.074 J per breath.
4. Halve compliance to 0.083 L per cmH2O: the same breath needs 6.0 cmH2O and 0.5 x 6.0 x 0.50 = 1.5 cmH2O x L, about 0.15 J, double.

Spirometry measures what moves, not what stays

A spirometer records volume against time. Four volumes do not overlap: tidal volume, inspiratory reserve volume (the extra that can be inhaled after a tidal breath), expiratory reserve volume (the extra that can be exhaled) and residual volume (RV, what stays after a maximal expiration). Capacities are sums: inspiratory capacity (tidal plus inspiratory reserve), FRC (expiratory reserve plus RV), vital capacity (all but RV) and total lung capacity (TLC, all four).

A spirometer sees only gas moving through the mouth, so it cannot measure RV, FRC or TLC. These need helium dilution (rebreathe helium in a closed circuit and solve C1 x V1 = C2 x (V1 + V lung)), nitrogen washout (collect the nitrogen washed out by pure oxygen) or body plethysmography (Boyle’s law in a sealed box). In obstructive disease the first two reach only gas that communicates with the airways and can underestimate RV, FRC and TLC.

One open textbook gives a tidal volume of about 500 mL and a TLC of about 6,000 mL in adult men and 4,200 mL in adult women: rough averages. Volumes vary with height, age and sex, so one person’s value is judged against reference equations from healthy populations, which give a lower limit of normal.

In the forced test, from TLC, blow out as hard and fast as possible: the total is the forced vital capacity (FVC), and the volume in the first second is FEV1.

Surface tension supplies much of the lung’s recoil

In 1929 Kurt von Neergaard compared pressure-volume curves of an isolated pig lung filled with air and with a gum solution that removed the air-liquid surface, and concluded that surface tension, not tissue elasticity, accounts for most of the lung’s recoil. Mead’s laboratory later found the same in excised lungs: an air-filled lung recoils more than a saline-filled one.

The static curves also show hysteresis: at the same pressure the lung holds more volume on deflation than on inflation. It comes mainly from the air-liquid interface, both through the surfactant film and through collapsed airspaces and small airways reopening, with a smaller share from the tissue, and it is larger for breaths starting near RV.

An air-water surface acts like a stretched sheet; its surface tension, T, is force per length, in mN per m (equal to dynes per cm). Laplace’s relation gives the pressure that holds a curved surface open:

P = 2T / r (one spherical air-liquid surface of radius r)

A soap bubble has two surfaces and needs 4T/r; an alveolus has one. The sphere simplifies polyhedral alveoli.

Worked problem, part 2: how hard is it to keep an alveolus open? Illustrative values; one spherical surface, static, 1 cmH2O = 98.07 Pa.

1. r = 100 micrometres = 1.0 x 10^-4 m. No surfactant, T = 0.070 N per m: P = 2 x 0.070 / 1.0 x 10^-4 = 1,400 Pa = 14.3 cmH2O.
2. With surfactant, T = 0.020 N per m: P = 400 Pa = 4.1 cmH2O, 3.5 times smaller. Compare the quiet swings of about 1 cmH2O alveolar and 3 cmH2O pleural.
3. Halve r to 50 micrometres at 0.070 N per m: P = 2,800 Pa = 28.6 cmH2O. At equal tension the small alveolus would empty into the large one.
4. Let tension fall with area (made-up): at 50 micrometres, T = 0.005 N per m, so P = 2 x 0.005 / 5.0 x 10^-5 = 200 Pa = 2.0 cmH2O, below the large alveolus’s 400 Pa. The small one no longer empties.

Answer: about 14 cmH2O without surfactant, about 4 with it; tension that falls steeply enough with area removes the small-into-large problem.

Small-alveolus radius 50 µm Large-alveolus radius 100 µm

Lining

No surfactant (70 mN per m) Fixed tension (20 mN per m) Tension falls with area (made-up rule)

In this sketch, the status appears here.

In this sketch: each alveolus is one free spherical air-liquid surface, static, so P = 2T/r (the lesson’s relation for one surface), with 1 cmH₂O = 98.07 Pa. The values 70 and 20 mN per m are the lesson’s illustrative tensions. The rule T = 20 mN per m × (r / 100 µm)², is made up for the sketch; it gives the lesson’s made-up 5 mN per m at 50 µm. The radius ranges are chosen for the sketch, and the small alveolus is never set larger than the large one (the other slider moves along).

Real alveoli are tethered polyhedra: neighbours share walls, so one alveolus cannot shrink alone like a free bubble. The spheres are a simplification. Drawing: circle size is to scale with radius, and a thicker outline means higher tension.

Keep this: at equal tension the smaller alveolus has the higher pressure and would empty into the larger one; whether a film prevents that depends on how steeply tension changes with area.

The small-into-large emptying, or “Laplace instability”, is a result of theory and of films in vitro; whether a film prevents it depends on how steeply tension changes with area (the dilatational modulus). Real alveoli are also tethered: in Wilson and Bachofen’s structural model, alveolar walls are platforms for surface tension, and neighbours share walls, so one alveolus cannot shrink alone like a free bubble.

Surfactant lowers tension most when the alveolus is small

Richard Pattle in England, from foam that stayed stable for hours, and John Clements in Maryland, with films from rat, cat and dog lungs on a modified Wilhelmy balance, each reported evidence of a low-tension lining, Pattle in 1955 and Clements in 1957 (Clements cited Pattle and measured 10 to 15 dynes per cm in lung bubbles, not under 1). Clements saw tension fall from about 46 to about 10 dynes per cm as the film was compressed. He called it a surface-active material; it later became known as pulmonary surfactant.

Surfactant is about 90 percent lipid, mainly phospholipid, and 10 percent protein, secreted by alveolar type 2 cells, which store it in lamellar bodies. Its main lipid is dipalmitoylphosphatidylcholine (DPPC; see lipids). Phospholipid alone adsorbs too slowly for a breathing lung; surfactant protein B (SP-B) speeds adsorption. Compressed at end-expiration, the film brings tension close to 0 mN per m; expanded in inspiration, tension rises. By making tension depend on area, surfactant prevents end-expiratory collapse of distal airspaces and lowers the work of breathing.

The classic experiment. Hyaline membrane disease (HMD) is the lung pathology of respiratory distress syndrome in newborns, most often preterm ones. Mary Ellen Avery saw Clements’s balance in December 1957. In their 1959 paper, she and Jere Mead put extracts of lungs from infants who had died, with or without HMD, on a surface balance and recorded tension as the film was compressed. If HMD lungs lacked the lining, their extracts should not reach low tension. In the paper’s table the mean lowest tension reached on compression was about 30 dynes per cm in extracts from nine infants with HMD and about 8 in extracts from infants over about 1,200 g who died of other causes; extracts from smaller infants who died of other causes also stayed high, about 26, so immaturity and disease were entangled. The authors concluded that HMD is associated with absence or late appearance of a substance that lets the lung surface reach low tension at low volume. That ruled in an association. Post-mortem, injured lungs left open whether the deficiency followed from the disease or from death.

Evidence on cause came later. Nebulized DPPC alone showed no clear benefit in 1960s trials; in the early 1970s Enhorning and Robertson found that preterm rabbits given natural surfactant survived longer than expected; and in 1980 Fujiwara and colleagues reported improved oxygenation in ten preterm infants given a modified bovine surfactant, with no untreated controls. Liggins and Howie’s 1972 trial showed that antenatal glucocorticoid treatment can speed fetal lung maturation (pregnancy, birth and lactation).

The modern experiment. If surfactant needs correct packaging as well as its proteins, a packaging defect should cause deficiency even at term. After an SP-B mutation in infants (1994) and SP-B knockout mice (1995) tied the protein to survival, Shulenin, Nogee, Whitsett, Dean and colleagues in 2004 sequenced ABCA3, an ATP-binding cassette transporter gene, in 21 infants born at or near term with severe neonatal surfactant deficiency of unknown cause. Mutations turned up in 16 (76 percent), and the authors reported abnormal lamellar bodies in lung tissue from four. A transporter needed for normal lamellar bodies is needed for surfactant function in humans, even near term. The study does not show what ABCA3 carries, how each mutation acts, or that preterm respiratory distress is genetic. Later, in adult mice whose SP-B could be switched off, high surface tension injured the alveolar epithelium before edema formed.

Airways warm, clean and resist the air

Inspired air reaches the alveoli at 37 degrees C, saturated with water vapour (47 mmHg), which ventilation and perfusion subtracts from barometric pressure in the alveolar gas equation. A separate rule, Henry’s law, says dissolved gas is proportional to partial pressure. Nasal hairs trap large particles; mucus from goblet cells and glands catches smaller ones, and cilia sweep it toward the pharynx on a watery layer set by epithelial ion transport, including CFTR (membrane transport). Alveolar macrophages take up most particles that reach the alveoli. A panting dog uses its airway surfaces for evaporative heat loss (thermoregulation).

Airway resistance is the pressure difference per unit flow. Poiseuille’s fourth-power rule assumes steady laminar flow in a rigid tube (blood and vessels), and airways are deformable. Small airways are so numerous, in parallel, that in health only a small share of airway resistance lies in them: classic measurements put all airways under 2 mm in diameter at roughly 10 to 20 percent. Most sits in larger airways.

In forced expiration, positive pleural pressure squeezes the deformable airways as well as the alveoli, so beyond some effort flow stops rising: dynamic compression. Breathing costs elastic work (part 1) and resistive work, and the central drive picks a tidal volume and rate that tend to minimise the total (control of breathing; for respiratory muscles competing with the legs for blood, exercise physiology).

An obstructive pattern lowers FEV1 more than FVC, so FEV1/FVC falls. A restrictive pattern lowers TLC; FEV1 and FVC fall together and the ratio is normal or high. Both are judged against reference limits for age, sex and height, and a low FVC alone does not prove restriction: it goes with a low TLC less than half the time.

Worked problem, part 3: two made-up spirograms with made-up lower limits. Volunteer 1: FEV1 1.5 L, FVC 3.0 L, ratio 0.50, below a limit of 0.68: obstructive. Volunteer 2: FEV1 2.2 L, FVC 2.6 L, ratio 0.85, within limits, but plethysmographic TLC 3.9 L, below a limit of 4.6 L: restrictive, which the low FVC alone would not have shown.

In emphysema, destroyed alveolar walls lower elastic recoil, so compliance rises. The resting balance with the chest wall moves to a larger volume, so FRC rises; airways close earlier in expiration, so RV rises; and with less recoil opposing the inspiratory muscles, TLC usually rises. Without radial traction, small airways compress earlier in forced expiration, so FEV1 falls more than FVC and the ratio falls. Fibrosis runs the other way: compliance and TLC fall, stiff tissue holds airways open (ratio normal or high), and the elastic work of a breath of a given size rises as in part 1, so breathing shifts to small, fast breaths.

Design problem

You have excised lungs from adult rabbits, a syringe pump and a pressure transducer. Show how much of the lung’s recoil comes from surface tension, and test what depleting surfactant does.

One way to do it

Inflate each degassed lung with air in steps, holding each until flow stops, to a fixed volume and back: a static curve with hysteresis. Refill it with saline and repeat. Predict a saline curve far to the left with little hysteresis; the pressure gap between curves at one volume estimates the surface-tension share of recoil.

Controls: the same volume history before each curve, randomized air-saline order in a second set, constant temperature, lung weight before and after (edema).

Lavage a third set to deplete surfactant: predict an air curve shifted right, with higher opening pressure and more hysteresis, and a saline curve that barely moves. Add back natural surfactant: partial recovery. Conditional SP-B-deficient mouse lungs would give a genetic version.

Limits: no chest wall, saline may alter tissue, one species and age. This is von Neergaard’s logic, and that of Mead’s laboratory.