Editorial III

Development of the concept of a liquid pulmonary alveolar lining layer

K.L. Dorrington and J.D. Young

A remarkable phenomenon has occurred in recent years in pulmonary physiology that may tell us something about the survival of orthodoxy despite ever more aggressive peer review. We refer to the widespread persistence of the notion that the pulmonary alveoli are normally lined with liquid. This notion persists in the scientific community in the face of a simple law of physics that shows it to be impossible. The law, appreciable by schoolchildren who blow soap bubbles, is that the only stable liquid–gas interface has the shape of a sphere, unless gravity upsets things badly. As a pulmonary alveolus is not a spherical structure, its epithelium cannot support a continuous thin lining with liquid that follows the non-spherical contour of its surface. It requires some explanation, therefore, why both the specialist medical literature, and school and university biology and physiology texts, are dominated by the view that the ‘alveolar surface is lined with liquid, which forms an air–liquid interface with alveolar gas’.1 To physicists this is inconceivable.2

The story reads like a detective case. In 1929, von Neergaard introduced the idea that surface forces contribute substantially to a tendency for the lungs to collapse, as inflation of excised lungs with liquid (which abolishes any surface facing gas) requires much lower pressures than inflation with gas.3 In 1957, Mead and colleagues took things further, performing not only inflations but also deflations of excised lungs.4 They observed that, for a given lung volume, pressure during inflation with gas was higher than during deflation, a phenomenon called pressure– volume hysteresis. The hysteresis largely disappeared during inflation with liquid.

The discovery by Brown and colleagues in 1959, that extracts from lung display a surface tension–area hysteresis on a Langmuir trough that looks a bit like the pressure– volume hysteresis seen in excised lungs,5 set the scene for an assumption that the alveoli themselves constitute a trough of liquid, albeit one with a complicated geometry. This assumption has pervaded the literature on lung mechanics for over 40 yr, and established itself firmly. Respiratory physiology texts speak with one voice about the liquid lining of the lung, assuming, or explicitly stating, that the whole alveolar surface is covered with a liquid.1 612

Early on it was appreciated that there was a problem with the assumption that liquid coats the alveolar epithelium. Even if one conceded, for the sake of simplicity, that individual alveoli could be regarded as spherical structures, then the connection of alveoli of different sizes presented a problem, unless the liquid that lines one alveolus was made to stop at an unwetted patch where the alveolus was connected to any neighbouring alveolus. The reason for this is explicitly presented in many textbooks of respiratory physiology.6 7 9 12 If two model spherical alveoli of different radii of curvature (r) are connected together, and an attempt is made to hold a liquid lining in both inflated by the same pressure (p), this proves to be impossible unless the surface tension (T) differs in the two alveoli. Laplace’s equation p=2T/r must be obeyed in each alveolus. You cannot change just r whilst T and p stay constant. The texts use this proof to support a separate idea. Studies of surfactant spread on a liquid surface have shown that T tends to increase with the area of the surface. Thus, conveniently, we can imagine that within a single alveolus, T can increase with r. The texts emphasize that, if T can increase with r, then p (~T/r) can be the same for both alveoli. What the authors do not draw attention to is that T can differ in neighbouring alveoli only if their liquid linings are not connected by a common surface.

This appears to have been clear in 1961 to Clements and colleagues in their seminal paper13 on ‘pulmonary surface tension and alveolar stability’ where a model of stability of the alveoli required the assumption that ‘the individual units were taken to act independently of one another’. In other words, different alveoli were permitted to have different surface tensions and different pressures across their liquid surfaces, and therefore, not be interconnected by a continuous lining of liquid. Yet this paper is remembered not for this vital aspect of its discussion section, but for its recognition of the possible physiological importance of the observation alluded to above, that lung extracts spread over a trough of liquid display a surface tension that increases with the area to which the surface is expanded. This idea spread, and with it, inappropriately, seems to have spread the postulated liquid lining layer, from one ‘independent’ model spherical alveolus to the whole network of millions, in violation of the authors’ apparent original intentions.

All seemed settled for a while. ‘The alveoli have concave surfaces lined with liquid.’10 The high surface tension of water in contact with air (70 mN m–1) was reduced to something manageable by the presence on the water–air surface of surfactant,14 the absence of which was responsible for collapse, oedema, and the generation of the distinctive histological finding of hyaline membranes in premature infants dying of what we now call infant respiratory distress syndrome.15 Dissident scientific voices16 17 were infrequent and unheeded by the writers of textbooks. Even before university, generations of school advanced biology students continued to be primed think that ‘the walls of the alveoli are covered with a thin layer of liquid’,18 not least because it is widely maintained without justification that ‘the respiratory surface must be moist, because gases cannot diffuse across it unless they are dissolved in fluid’.19

The literature that supports a liquid lining for the lung by considering the stability of neighbouring alveoli of different sizes is the strongest argument against it. The reason is that, if neighbouring alveoli of different radii of curvature cannot be joined by a shared liquid surface, then neither can neighbouring segments of a single alveolus. A stable film of liquid cannot connect two or more regions of an alveolar surface of different radii of curvature whilst maintaining the shape of those surfaces. The argument from Laplace’s law shows that a continuous liquid surface within a complex polygonal structure is unstable. This conclusion is consistent with studies of lung histology under electron microscopy in the presence of normal20 and excess21 volumes of liquid in the alveoli, in which liquid collections always exhibit a surface that forms part of a sphere, and are separated from each other by segments of epithelium on the surface of which liquid cannot be demonstrated. The assumption made recently by Bastacky and colleagues,22 that a continuous lining layer of the alveolus revealed by low-temperature scanning electron microscopy consists of ‘liquid’, is not a consequence of their observations; the layer would be unstable if it were a liquid layer, for the reasons given above. It is beginning to look as if particular care is needed in the use of the word ‘liquid’, if the nature of the alveolar surface is to be clarified with precision.

The most vociferous opponent of alveolar wetness has made two fundamental errors in his otherwise challenging critique of established dogma.23 The first is the claim that a hydrophobic surface is required to prevent water from spreading within a polygonal structure. In fact, according to our earlier reasoning, even a hydrophilic surface can only sustain a continuous liquid lining if a sufficient volume of liquid is present on that surface to immerse irregularities beneath a spherical liquid–gas interface. This phenomenon is demonstrated in the micrographs of protein-rich pulmonary oedema published by Bachofen and colleagues.21 Hills’ second error, uncorrected even in a publication well after extensive new findings in the field,24 is the claim that passive forces alone account for the absorption of liquid from the corners of alveoli. His model of the ‘corner pump’, in which a liquid globule presents a convex surface to alveolar gas, and is emptied into lung interstitium by surface tension raising the pressure within the globule above that of alveolar gas and interstitial liquid, contrasts strikingly with the model that arises from the work of Basset and colleagues.25 26 In this model of adult lung, type II epithelial pneumocytes (the ones that secrete surfactant on to the alveolar surface) utilize fuel (ATP) actively to pump sodium, and with it glucose and water, from any collections of water that bathe the apical surface of the cells, into the interstitium.27 Matthay and colleagues have extensively characterized this process of liquid absorption in animals and humans,28 and there is some evidence that without active sodium transport fatal alveolar flooding would occur.29 Interestingly, however, the presence of aquaporins (water transport channels) in alveolar epithelium makes no measurable contribution to recovery from a wide variety of forms of lung injury associated with alveolar oedema.30

If the detectives on the case appear to be reluctant to make use of this fascinating new evidence of vigorous transport activity down in the alveoli, they appear even more reluctant to incorporate the most remarkable claims emanating from the laboratory of Scarpelli,31 who offers a whole ‘new anatomy’ of the alveolar surface. On the basis of studies that are claimed to preserve surface structures destroyed by the customary methods of histological preparation, Scarpelli has proposed that the alveolar ducts are normally filled with foam surfactant, within which each alveolus can be identified as a separate and complete bubble. Thus, astonishingly, the gas within each alveolus is not free to flow by convection, at least not continuously throughout the whole of the respiratory cycle as has been normally supposed. In this model, the entrances to alveoli are spanned by extremely thin bilayers of surfactant that are presumed usually to contain no ‘hypophase’ liquid, and consequently take the form of ‘Newtonian black films’. Scarpelli’s model interestingly emphasizes the movement of ‘hypophase liquid . . . from regions of larger radii to those of smaller radii of curvature’ (Scarpelli,31 page 499), and thus indirectly, that much of the alveolar surface, and many of the black films spanning the alveolar entrances, must be devoid of liquid for the reasons discussed earlier in this article. On reading the almost gladiatorial contest between Hills and Scarpelli recently convened in the editorial pages of the Journal of Applied Physiology,32 one would be forgiven for missing the substantial similarities between their models of the alveolus with regard to the necessary absence of liquid from much of the epithelial surface.

A picture clearly emerges that needs to be projected into the less specialized basic scientific and clinical literature. The alveolar surface carries very little liquid, in the usual sense of something that readily flows in response to a pressure gradient. Where liquid has been demonstrated, it lies in isolated collections that are commonly partially submerging type II pneumocytes. These cells appear to be as busy actively emptying the puddles of liquid that bathe them as they are in reducing the surface tension of these puddles by secreting surfactant into them. The larger areas of alveolar surface that are normally unwetted sustain a tension that is imparted on them by neighbouring wetted regions. Wetted and unwetted areas pull gently at each other like competing teams in a tug of war at what the physicists call a ‘triple point’. Thus, the alveolar surface can be regarded as having a tension that von Neergaard was able to abolish by inflating the lung with a liquid, even though that surface is not normally entirely wetted. The type II pneumocyte polices alveolar tension over the wider unwetted alveolar epithelium by keeping tension low in the wetted surface within its immediate proximity. Surfactant spreads over wet and unwetted areas alike to provide Bastacky and colleagues’ ‘thin and continuous’ lining layer22 so strikingly revealed by their low temperature microscopy. The jury remains undecided about whether Scarpelli’s foam31 is a feature more of neonatal lung than adult lung.

A fuller understanding of alveolar mechanics will require more careful attention to the use of terms like wet, moist, dry, liquid, and solid when applied to this environment. More consideration needs to be given to the important observation of Bangham and colleagues that modest compression of surfactant on water (as in expiration) can generate a solid film that impedes movement within the surface,33 has the potential structurally to resist alveolar collapse, and, on re-expansion (as in inspiration), may break down into a myriad of planar regions of alternate solid and liquid reminiscent of Arctic pack-ice. The observation of hysteresis in the lung in the 1950s was an early warning that a time-dependent viscous behaviour of the alveolar surface would need to be modelled. Virtually no appreciation of the large literature on viscoelasticity has been brought to bear on understanding the role of surfactant in the alveolar epithelial lining. Characteristically, biological materials display time-dependent behaviour that can be thought of as showing features of both solids and liquids.34 The presence of such behaviour in a surface film undergoing repeated stretching may lead to gradients in surface tension and consequently to slow flows in the film that would not otherwise occur.35 Moreover, repeated stretching of the alveolar epithelium occurs not only with the cycle of breathing; the passage of every red blood cell, as it squeezes through a pulmonary capillary, generates a rapidly alternating curvature of the epithelial surface that has been predicted to assist absorption of alveolar liquid because of the way this deformation of the shape of the surface interacts with the hysteresis of surfactant.36 We can anticipate that a fuller understanding of the alveolar surface will incorporate a much more dynamic model of surfactant activity; the concept of a trough of water that has no time-dependent properties is no longer sustainable. Walters, in reviewing the role of surfactant in transepithelial movement of liquid, warns: "the movements of water molecules, ions and other solutes in such a thin complex environment are unlikely to obey the laws governing their mobility in more conventional spaces and terms such as ‘dry’ and ‘wet’ applied to this microlayer would require philosophical definition".37 If we can lay to rest the glaring widespread misconception that a layer of liquid, in the sense in which that word is commonly used, can exist on the whole of the normal alveolar surface, we shall be in a better position to make progress. The phenomenal success of the clinical use of surfactant replacement therapy is now well ahead of our understanding of how it works.38 Furthermore, as intensivists increasingly interest themselves in filling the lung with liquid in order to ventilate it less traumatically, it will be helpful to be clear about when it is wet and when it is dry.39

K. L. Dorrington

University Laboratory of Physiology

Parks Road

Oxford OX1 3PT

UK

J. D. Young

Nuffield Department of Anaesthetics

Radcliffe Infirmary

Woodstock Road

Oxford OX2 6HE

UK

References

1 Leff AR, Schumacher, PT. Respiratory Physiology. Philadephia: Saunders, 1993

2 Hunter RJ. Introduction to Modern Colloid Science. Oxford: Oxford University Press, 1993

3 Von Neergaard K. Neue Auffassungen über einen Grundbegriff der Atemmechanik. Z Ges Exp Med 1929; 66: 373–94

4 Mead J, Whittenberger JL, Radford EP. Surface tension as a factor in pulmonary volume–pressure hysteresis. J Appl Physiol 1957; 10: 191–6[Abstract/Free Full Text]

5 Brown ES, Johnson RP, Clements JA. Pulmonary surface tension. J Appl Physiol 1959; 14: 717–20[ISI]

6 Mines AH. Respiratory Physiology, 3rd edn. New York: Raven Press, 1993

7 West JB. Respiratory Physiology: the Essentials, 6th edn. Baltimore: Williams & Wilkins, 2000

8 Prange HD. Respiratory Physiology. Understanding Gas Exchange. New York: Chapman & Hall, 1996

9 Hlastala MP, Berger AJ. Physiology of Respiration. Oxford: Oxford University Press, 1996

10 Widdicombe J, Davies A. Respiratory Physiology, 2nd edn. London: Edward Arnold, 1991

11 Staub NC. Basic Respiratory Physiology. New York: Churchill Livingstone, 1991

12 Lumb AB. Nunn’s Applied Respiratory Physiology, 5th edn. Oxford: Butterworth Heinemann, 2000

13 Clements JA, Hulstead RF, Johnson RP, Gribetz I. Pulmonary surface tension and alveolar stability. J Appl Physiol 1961; 16: 444–50[ISI]

14 Brown ES. Isolation and assay of dipalmityl lecithin in lung extracts. Am J Physiol 1964; 207: 402–6[ISI]

15 Avery ME, Mead J. Surface properties in relation to atelectasis and hyaline membrane disease. Am J Dis Child 1959; 97: 517–22[ISI]

16 Brooks RE. Lung surfactant: an alternative hypothesis. Am Rev Resp Dis 1971; 104: 585–6[ISI][Medline]

17 Hills BA. The Biology of Lung Surfactant. Cambridge: Cambridge University Press, 1981

18 Jones M, Jones G. Advanced Biology. Cambridge: Cambridge University Press, 1997

19 Starr C, McMillan B. Human Biology, 3rd edn. London: Wadsworth, 1999

20 Gil J, Weibel ER. Improvements in demonstration of lining layer of lung alveoli by electron microscopy. Respir Physiol 1970; 8: 13–26[ISI]

21 Bachofen H, Schürch S, Michel RP, Weibel ER. Experimental hydrostatic pulmonary edema in rabbit lungs. Am Rev Respir Dis 1993; 147: 989–96[ISI][Medline]

22 Bastacky J, Lee CYC, Goerke J, et al. Alveolar lining layer is thin and continuous: low-temperature scanning microscopy of rat lung. J Appl Physiol 1995; 79: 1615–28[Abstract/Free Full Text]

23 Hills BE. The role of lung surfactant. Br J Anaesth 1990; 65: 13–29[ISI][Medline]

24 Hills BE. An alternative view of the role(s) of surfactant and the alveolar model. J Appl Physiol 1999; 87: 1567–83[Abstract/Free Full Text]

25 Basset G, Crone C, Saumon G. Significance of active ion transport in transalveolar water absorption: a study on isolated rat lung. J Physiol 1987; 384: 311–24[Abstract]

26 Basset G, Crone C, Saumon G. Fluid absorption by rat lung in situ: pathways for sodium entry in the luminal membrane of alveolar epithelium. J Physiol 1987; 384: 325–45[Abstract]

27 Dorrington KL, Boyd CAR. Active transport in the alveolar epithelium of the adult lung:vestigial or vital? Respir Physiol 1995; 100: 177–83[ISI][Medline]

28 Matthay MA, Flori HR, Conner ER, Ware LB. Alveolar epithelial fluid transport: basic mechanisms and clinical relevance. Proc Assoc Am Physicians 1998; 110: 496–505[ISI][Medline]

29 Vejlstrup NG, Boyd CAR, Dorrington KL. Effect of lung inflation on active and passive liquid clearance from in vivo rabbit lung. Am J Physiol 1994; 267: L482–7[Abstract/Free Full Text]

30 Song Y, Fukuda N, Bai C, Ma T, Matthay MA, Verkman AS. Role of aquaporins in alveolar fluid clearance in neonatal and adult lung, and in oedema formation following acute lung injury: studies in transgenic aquaporin null mice. J Physiol 2000; 525: 771–9[Abstract/Free Full Text]

31 Scarpelli EM. The alveolar surface network: a new anatomy and its physiological significance. Anat Rec 1998; 251: 491–527[ISI][Medline]

32 Scarpelli EM, Hills BA. Opposing views on the alveolar surface, alveolar models, and the role of surfactant. J Appl Physiol 2000; 89: 408–12[Free Full Text]

33 Bangham AD, Morley CJ, Phillips MC. The physical properties of an effective lung surfactant. Biochim Biophys Acta 1979; 573: 552–6[ISI][Medline]

34 Dorrington KL. The theory of viscoelasticity in biomaterials. In: Vincent JFV, Currey JD, eds. The Mechanical Properties of Biological Materials. Cambridge: Cambridge University Press, 1980; 289–314

35 Epinosa FF, Kamm RD. Thin layer flows due to surface tension gradients over a membrane undergoing nonuniform periodic strain. Ann Biomed Eng 1997; 25: 913–25[ISI][Medline]

36 Hills BA. Rapidly alternating curvature (‘oil canning’) as a mechanism preventing alveloar edema. J Appl Physiol 1992; 72: 2108–11[Abstract/Free Full Text]

37 Walters D. The role of pulmonary surfactant in transepithelial movement of liquid. In: Robertson B, Van Golde LMG, Batenburg JJ, eds. Pulmonary Surfactant: from Molecular Biology to Clinical Practice. Amsterdam: Elsevier Science Publishers, 1992

38 Gibson AT. Surfactant and the neonatal lung. Br J Hosp Med 1997; 58: 381–97[Medline]

39 Hirschl RB, Conrad S, Kaiser R, et al. Partial liquid ventilation in adult patients with ARDS: a multicenter phase I–II trial. Adult PLV Study Group. Ann Surg 1998; 228: 692–700[ISI][Medline]