University Laboratory of Physiology, Parks Road, Oxford OX1 3PT, UK *Corresponding author
Accepted for publication: July 27, 2001
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Br J Anaesth 2001; 87: 897904
Keywords: complications, pulmonary oedema; acidbase equilibrium, active sodium transport; complications, hypoxia; rabbit
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Alveolar sodium transport can be inhibited both in vitro and in vivo by the delivery to the alveolar epithelium of amiloride, an inhibitor of apical sodium channels, and phloridzin, an inhibitor of apical sodiumglucose cotransport.2 The reduction of this transport using ouabain, an inhibitor of the basolateral sodiumpotassium ATPase (Na,K-ATPase) that drives the process, is limited to in vitro preparations, because of its systemic effects.5 Although it has proved possible to study the fatal pulmonary effects of elimination of sodium channels in genetically modified mice,6 selective inhibition of the sodium transport mechanism over the whole alveolar surface in adult animals still ventilated with gas has not proved feasible. In this study, we have utilized gentle complete liquid filling of the apnoeic left lung of the anaesthetized rabbit to deliver amiloride and phloridzin to the apical surface of the alveolar epithelium, and thereby assess the contribution of sodium transport to liquid flux across the healthy epithelium in vivo. This model has been validated previously.7
With reference to epithelial cell culture studies, Bärtsch8 has recently suggested that insufficiency of active alveolar sodium transport may contribute to the aetiology of high altitude pulmonary oedema (HAPO), because of the possible susceptibility of a transport mechanism requiring ATP to the effects of hypoxia. Suzuki and colleagues9 have shown that exposure of rats to a hypoxic environment of 10% oxygen for 48 h impaired transalveolar liquid transport, but that this impairment was related to a decrease in Na,K-ATPase activity, not a reduction in the availability of ATP. Conversely, Sznajder and colleagues10 found that exposure of rats to a hyperoxic environment of 85% oxygen for 7 days led both to an upregulation of Na+ channels and to an increase in Na,K-ATPase activity. In experiments on in situ perfused lungs from sheep aged 6 weeks to 6 months, Junor and colleagues found no correlation between the rate of lung liquid absorption and the oxygen partial pressure (PO2) in the lungs in the approximate range 666 kPa during acute experiments lasting approximately 1 h.11
Alveolar PO2 in climbers breathing air at the summit of Everest has been estimated to be 4.7 kPa, with the associated mixed venous PO2 being approximately 2.8 kPa.12 It remains uncertain whether acute exposure of the alveolar epithelium to moderate levels of hypoxia impairs liquid absorption associated with active sodium transport.
In this study we compared absorption over 3.5 h at lung PO2 values of approximately 6.5 and approximately 4.5 kPa to compare the effects of mild and moderate acute hypoxia on liquid movements, with and without inhibition of sodium transport. This enabled us to examine the hypothesis that active epithelial sodium transport is required to prevent net alveolar liquid secretion, and to examine whether transepithelial liquid flows differ between mild and moderate levels of hypoxia.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Anaesthesia, monitoring, and surgical preparation
All procedures were licensed by the UK Home Office. Surgery was performed under halothane anaesthesia following premedication with Hypnorm (fentanyl 0.315 mg ml1, fluanisone 10 mg ml1; Janssen, UK) 0.3 ml kg1 body weight i.m. Maintenance of anaesthesia during the 6 h study period was achieved by Hypnorm infusion (0.51.0 ml h1) at which time neuromuscular block was achieved with alcuronium 0.51.0 mg h1. Adequacy of anaesthesia was assessed by observing heart rate, arterial pressure, pupil size, and the presence or absence of tears. Tracheostomy was performed and a 4.0 or 4.5 mm internal diameter tracheal tube was introduced into the trachea. Animals breathed oxygen with 12% halothane during surgery, and were then ventilated with FIO2 1 or 0.3 without halothane during the study period. Monitoring included arterial pressure and blood gases (via a carotid artery catheter), ECG and heart rate, limb pulse oximetry, urine output, and rectal temperature. A 4-French gauge cannula was introduced via the left jugular vein into the pulmonary artery for monitoring pulmonary artery pressure (PAP) and sampling mixed venous blood. Cardiac output (Qt) was measured after 6 h of liquid filling of the left lung using the Fick principle for oxygen; for this purpose apnoeic oxygen uptake measurements for the whole animal were made at the same time at which arterial and mixed venous blood samples were obtained for estimation of blood oxygen concentrations.13 Pulmonary shunt fraction was estimated from the same blood gas measurements, using the standard alveolar gas equation to estimate alveolar PO2.
Hartmanns solution (Na+ 131 mM, K+ 5 mM, Ca2+ 2 mM, Cl 111 mM, lactate 29 mM) was infused via an ear vein at 5 ml kg1 h1. We have previously shown that this animal preparation is stable, in so far as that cardiac output remains constant over 6 h.14
Ventilation of the right lung
After neuromuscular block, both lungs were ventilated using a Bear Cub infant ventilator (Bear Medical Systems, Riverside, CA, USA). The ventilator setting was initially: gas flow 8 litre min1, peak pressure 2 kPa, positive end-expiratory pressure 0.3 kPa, I:E ratio 0.5, FIO2 1, rate 1420 min1. Throughout the experiment, including the 6 h study period, during which only the right lung was ventilated, the ventilator rate was adjusted to maintain arterial carbon dioxide partial pressure (PaCO2) between 4.7 and 6.0 kPa, with the other ventilator variables unchanged. Mean airway pressure remained at 0.60.7 kPa. FIO2 was held at 1 in all animals until 150 min after instillation of liquid into the left lung. At this time, in 12 animals FIO2 was switched to 0.3; in the other 12 animals FIO2 remained 1 throughout.
Bronchial isolation and liquid filling of the left lung
A balloon-tipped catheter was introduced into the left main bronchus no further than 1 cm from the carina. Correct positioning was achieved by observing chest movements, auscultation, blood gas sampling, and measurement of oxygen flow through the catheter in relation to the measured oxygen uptake of the whole animal. Confirmation of positioning was seen at post-mortem.
Using this catheter, we were able to estimate what fraction of the animals total lung tissue was present in the left lung (VLL). We did this by holding the animal apnoeic for a few minutes and measuring the separate flows of oxygen during apnoea into the left and right lungs. We assumed that the relative fractions of tissue in the two lungs would be the same as the relative flows of oxygen into the two lungs during apnoea. The purpose in measuring VLL was that the left and right lungs of the rabbit are of different sizes, and interpretation of transepithelial liquid flows, and of lung blood flows, are most usefully made in relation to the proportion of lung tissue studied.
Using the bronchial catheter, the left lung was filled over 12 min with a solution of glucose 10 mM and blue Dextran 1 mM that was made isotonic with the animals plasma using sodium chloride. Plasma and instillate osmolarity were measured by freezing point depression. Amiloride and phloridzin were added in 12 animals (see above). The volume of instillate injected equalled the volume of the left lung at a pressure of 0.3 kPa measured by helium dilution, and the rate of injection was chosen to equal the rate of oxygen uptake measured for the lung, in order to achieve complete liquid filling of the left lung without allowing either atelectasis or overdistention of the lung to occur. After filling the lung with liquid, the bronchial catheter was connected to a reservoir of instillate, the surface of which was kept at a height of 3 cm above the manubrium for 6 h to maintain a constant distending pressure of 0.3 kPa. The reservoir was marked to allow measurements of volumes as small as 0.01 ml; time was measured to the nearest second. Pleural leakage and atelectasis were excluded at post-mortem by inspection of the lung and pleural cavity to confirm that blue Dextran had been homogeneously distributed throughout, and contained within, the lung. Leakage of instillate past the bronchial catheter was excluded at post mortem by confirming that blue Dextran was retained within the study lung.
The volume of liquid that had flowed into, or out of, the left lung was measured every few minutes over 6 h. A steady flow was reached in all animals after approximately 100 min. It is assumed that the non-steady flow during the initial 100 min resulted from changes in lung volume consistent with the known difference in lung compliance between gas-filled and liquid-filled lungs.15 The results we present are the liquid uptakes over the period 200360 min, which we take to be direct measurements of transepithelial liquid flux.
Statistical analysis
All data are presented as mean (SEM). One-way analysis of variance (ANOVA) was used to assess whether measurements in the four groups of animals could be regarded as from the same normal distribution. Students t-tests were used to compare between-group means. Significance was regarded as present at P <0.05.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Initial filling of the left lung with liquid was associated with a rapid decrease in PaO2 from approximately 55 kPa during oxygenation of both lungs to approximately 15 kPa, followed by a gradual increase until approximately t=150 min. During the period 200360 min, all physiological measurements changed little, and consequently we report means for this whole period of measurements taken at 15 or 30 min intervals. Table 1 shows arterial and mixed venous blood gases, PAP, Qt, and shunt fraction in the four groups of six subjects. Also shown for each group is the fractional tissue volume (VLL) associated with the left lung as estimated by the fractional apnoeic oxygen uptake measured for the left lung in comparison with the whole animal. Cardiac output measurements were taken only at the end of the period 200360 min. P values in Table 1, derived from ANOVA, confirm that PaO2, PO2, and shunt fraction could not be regarded as from one normally distributed population.
|
In Table 2 are presented the liquid uptakes from the alveoli for the period 200360 min for each animal. A positive liquid uptake represents a movement of liquid out of the alveoli, that is a net absorption by the alveolar or airway epithelium; a negative liquid uptake represents secretion by the epithelium. The mean (SEM) values for the four groups are presented graphically in Figure 1 in relation to PO2. In control animals there was a net uptake across the alveolar epithelium of approximately 4 µl min1 (kg body weight)1. This uptake was independent of FIO2, and hence P
O2. In the presence of inhibitors of sodium transport, the epithelium became secretory with a net uptake of approximately 0.6 µl min1 (kg body weight)1. In the presence of inhibitors of sodium transport, there was no significant difference between the magnitude of the liquid secretion between low and high oxygen groups.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Condition of the animals within this study
Data from Table 1 suggest that animals were maintained in a state of good physiological health despite the 6 h duration of the study period and preceding preparation time, both under continuous general anaesthesia. Arterial pH, PaCO2, and cardiac output are close to reference values for unanaesthetized animals.16 17 Arterial oxygen partial pressures did not decrease to values that are associated with severe tissue hypoxia and acidosis. Previous studies using this model have observed a stable preparation over 6 h.13 14
Maintenance of alveolar dryness during health
The extent to which the adult lung requires sodium transport to maintain alveoli relatively free from liquid remains to be established.18 A method has yet to be found of selectively switching off this function of the alveolar epithelium in order to assess its effects on lung liquid balance. Early evidence that sodium transport might contribute to recovery from pulmonary oedema came in 1990 from Matthay and Weiner-Kronish19, who measured protein concentrations in oedema fluid aspirated from the lungs of patients before, and after, 12 h of intensive therapy. In those patients showing a clinical improvement, the protein concentration increased significantly by 47%, whereas in those patients showing no improvement the protein content changed little. It is likely that active sodium transport was rendered ineffective in the patients who were unable to concentrate protein in oedema fluid, possibly because an increase in alveolar solute permeability led to a loss of the normal balance of Starling forces across the alveolar epithelium.
In the rat lung, the experiment of Basset and colleagues 2 produced a result that suggested that loss of sodium transport leads to alveolar flooding. In the presence of amiloride and phloridzin in alveolar instillate in isolated-perfused lung, they measured a small, but statistically insignificant secretion into the alveoli, suggesting that the lung would continually fill with liquid derived from perfusate unless sodium transport reversed the direction of liquid flux. In the experiment of Vejlstrup and colleagues7, using a similar preparation to that used in this study, it was observed that alveolar flooding was induced by inhibition of active sodium transport unless alveolar hydrostatic pressure was elevated above 0.4 kPa, a figure commonly used as a mean airway pressure during ventilation of patients with positive end-expiratory pressure.
Unfortunately, the evidence alluded to above, and in the present study, is limited to lungs that are partially or completely filled with liquid before the measurement of absorption or secretion of liquid across the epithelium lining the airway tree. As there continues to be much uncertainty regarding the disposition of liquid within the alveoli of the healthy lung,20 it is difficult to make predictions for the initiation and intensification of pulmonary oedema in a normal gas-filled lung from studies in which fluxes and pressures have been obtained in the presence of large volumes of liquid. It remains uncertain, for example, whether any liquid lying within a healthy alveolus is bounded by a surface that is concave or convex towards the gas in that alveolus. Consequently, it remains unclear whether surface tension at any liquidgas interface generates a pressure within such liquid that is lower or higher than atmospheric, lower values tending to exacerbate the formation of oedema, and higher values tending to lead to its resolution. Nevertheless, the observation in this study that even an airway hydrostatic pressure of 0.3 kPa in the lung with inhibitors of sodium transport is insufficient to prevent continuing secretion of liquid into the airway, suggests that the healthy air-filled lungs with an airway pressure close to zero are susceptible to the initiation of oedema if sodium transport is removed.
Contribution of airway chloride secretion to transepithelial liquid flows
The upper airways are normally lined with a film of mucous lying on top of a watery sol that bathes the airway cilia, the maintenance of which appears to be a balance between chloride secretion and sodium absorption; this balance is disturbed in cystic fibrosis.21 In our study, measurements of absorption from, and secretion into, the airway tree are necessarily the sum of fluxes in the upper and lower airways. It is, therefore, likely that a component of the secretion we observe after inhibiting sodium uptake is the movement of liquid in response to chloride secretion. However, the marked variation of this secretion with alveolar hydrostatic pressure seen in the experiments of Vejlstrup and colleagues7 suggests that the passive movement of liquid under the influence of Starling forces plays a large role in determining the precise magnitude of the secretion that results after inhibiting sodium transport. Widdicombe and Widdicombe21 estimate the magnitude of the airway secretion in humans to be 14 ml day1, that is 0.14 µl min1 (kg body weight)1 for a 70 kg adult. This is about one-fifth of the secretory flows measured in this study following inhibition of sodium transport (0.51 and 0.76 µl min1 (kg body weight)1, Table 2). We know of no more precise data with the help of which it would be possible to compare the relative magnitudes of chloride secretion over the relatively small area of the upper airway with alveolar secretion as a result of passive Starling forces across the huge area of the alveolar-capillary membrane.
Contribution of sodium absorption to transepithelial liquid flows
The component of liquid flux that we attribute to active sodium transport is the difference between the two near horizontal lines in Figure 1. It equals 45 µl min1 (kg body weight)1. This figure should be regarded as a lower limit, as inhibition of sodium transport may have been incomplete at the doses of amiloride and phloridzin used. In rat lungs, cross-perfused with blood from a donor rat, Basset and colleagues2 measured a liquid uptake attributable to active transport of approximately 0.17 µl s1. The rats used weighed 0.3 kg, which gives a rate of uptake for two lungs of 33 µl min1 (kg body weight)1. Many studies have involved instillation of quite small volumes of liquid into gas-ventilated lungs;3 consequently, measured fluxes have referred to uncertain fractions of the alveolar surface. However, measurements in several species are broadly similar to those obtained in this study, and suggest that in humans22 sodium transport may generate a liquid uptake of the order of 1 litre day1.
Contribution of pulmonary vascular tone to lung liquid balance
Pulmonary capillary pressure is thought to influence liquid flow across the capillary endothelium and, according to the degree of buffering and lymphatic drainage provided by the interstitium, to influence liquid flow across the alveolar epithelium as well.1 The observation that transepithelial liquid flows, in both the presence and absence of inhibitors of sodium transport, remained unchanged as PO2 was lowered (Fig. 1) suggests that pulmonary capillary pressure is not altered within the range of P
O2 studied.
The most commonly accepted explanation for the occurrence of oedema during hypoxic exposure, such as that occurring at high altitude, is that pulmonary microvascular pressure increases and that this increase leads to alveolar flooding. The increase is envisaged as either within fairly small areas of lung as a result of local failure of arterial hypoxic vasoconstriction that permits transmission of a high PAP to some capillaries, or more widely throughout the lung as a result of the presence of venous hypoxic constriction.23 Evidence for constriction of pulmonary veins during hypoxia has come from in situ dog lung perfusion studies by Welling and colleagues24 and from myograph studies on rat pulmonary veins by Zhao and colleagues.25 Conversely, there is evidence for the rabbit that hypoxia leads to a decrease in pulmonary capillary pressure. In the isolated rabbit lung, Sanchez de Léon and colleagues found that hypoxic pulmonary vasoconstriction resulted in a reduction in liquid filtration rate from the vasculature during constant flow perfusion in a forward direction, and an increase in filtration rate during perfusion in a reverse direction from vein to artery.26 Such a finding suggests that there is not only arterial hypoxic constriction, but that there is a degree of venous hypoxic dilatation, as a constant flow across a constant or increased venous constriction would lead, respectively, to no change or to an increase in capillary pressure.
We have previously shown14 in this animal model both that there is a slow component of hypoxic pulmonary vasoconstriction that continues to intensify after 150 min, and that during the time period 200360 min blood flow diversion from the hypoxic left lung does not conform to the widely accepted model of hypoxic constriction of Marshall and colleagues that was derived from short experiments on rats and dogs lasting approximately 10 min.13 27 In brief, blood flow to the liquid-filled left lung is substantially (approximately 50%) higher when FIO2 to the right lung is 0.3 than when it is 1 (Table 1), whilst the model of Marshall and colleagues predicts that it should be slightly (approximately 7%) smaller.13 In view of the absence of a significant difference in PAP between these groups, our data suggest that lowering PO2 from 6.5 to 4.5 kPa in the one lung results in a large decrease in vascular resistance in that lung, with no change in capillary pressure. Whether this response is entirely a result of the local effect of the lung PO2, or is at least in part because of pulmonary innervation or humoral factors in the intact animal remains unclear, although arterial hypoxaemia appears to be insufficient to activate the chemoreflex vasodilatation in the lung observed by Levitzky.28
Our experiments were performed with a lung PO2 close to the P50 of 5.3 kPa for the hypoxic response in the rabbit in short experiments performed by Weissmann and colleagues.29 In claiming this, we are assuming that the PO2 of the apnoeic immobile lung tissue (and hence the alveolar PO2) was determined primarily by the PO2 of the blood arriving with a relatively high flow in the pulmonary artery (PO2) rather than the systemic arterial PO2 of blood arriving with a much lower flow via the bronchial circulation. Noting the result of Weissman and colleagues, we anticipated that the rate of change of pulmonary vascular resistance in the study lung, with respect to changes in P
O2, would be greatest at values of P
O2 around 5.3 kPa, and thus be most likely to induce a change in liquid flux across the pulmonary epithelium. Indeed, a measured hydraulic conductance of the epithelium in this animal model7 of 15.7 µl min1 (kg body weight)1 kPa 1 suggests that changes of capillary pressure as small as 0.1 kPa might be readily be detected by our measurements of liquid flux with an SEM of 0.151.10 µl min1 (kg body weight)1 (Table 2). It seems likely that the capillary pressure changed by less than 0.1 kPa during the change in P
O2 of 2 kPa induced in the experiments reported here.
In summary, we have observed a vasodilatation in the study lung on lowering PO2 from 6.5 to 4.5 kPa that has permitted a large increase in blood flow, but that has probably induced little or no change in the hydrostatic pressure across the alveolar-capillary structures, including the pulmonary interstitium. In future studies of this type, interpretation of the relationship between pressure and flow across the alveolar-capillary structures would be aided by obtaining histology of the lung to understand the extent to which interstitial oedema and lymphatic drainage might contribute to a difference between flows across the pulmonary capillary endothelium and the alveolar epithelium.
Independence of liquid flow associated with active sodium transport from PO2
Given that active sodium transport may be necessary to maintain the fluid balance of the healthy lung, the hypothesis of Bärtsch8 that inhibition of this process by hypoxia might lead to pulmonary oedema clearly needs to be considered. At the simplest level, it is conceivable that a process requiring ATP as fuel might fail at low levels of PO2. It has recently become apparent, however, that a wide range of both short-term and long-term physiological responses are regulated by hypoxia via the transcriptional regulator hypoxia-inducible factor 1.30 It seems possible that a reduction in alveolar liquid clearance by hypoxia lasting 48 h,9 and increase in alveolar liquid clearance by hyperoxia lasting 7 days,10 may be regulated by a system of this type. Experiments on ion transport in fetal rat distal lung epithelial cells by Ramminger and colleagues31 have shown time-dependent effects of raising PO2 from 23 to 100 mm Hg on different components of the sodium transport system. At 24 h an increase in sodium pump activity was measured, but 48 h was required to induce an increased abundance of the sodium channel protein -ENaC. An increase in sodium pump activity was measured at 3 h, but only became significant beyond 6 h.31 The experiments reported here suggest that 3 h are not sufficient for a change from mild (6.5 kPa) to moderate (4.5 kPa) alveolar hypoxia to induce a decrease in active sodium transport across the in vivo rabbit alveolar epithelium.
![]() |
Conclusion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 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: 32545[Abstract]
3 Matthay MA, Folkesson HG, Verkman AS. Salt and water transport across alveolar and distal airway epithelia in the adult lung. Am J Physiol 1996; 270: L487L503
4 Verghese GM, Ware LB, Matthay BA, Matthay MA. Alveolar epithelial fluid transport and the resolution of clinically severe hydrostatic pulmonary edema. J Appl Physiol 1999; 87: 130112
5 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: 21124
6 Hummler E, Barker P, Gatzy J, et al. Early death due to defective neonatal lung liquid clearance in alpha-ENaC-deficient mice. Nature Genet 1996; 12: 3258[ISI][Medline]
7 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: L4827
8 Bärtsch P. High altitude pulmonary edema. Med Sci Sports Exer 1999; 31: S237[ISI][Medline]
9 Suzuki S, Noda M, Sugita M, Ono S, Koike K, Fujimura S. Impairment of transalveolar fluid transport and lung Na+-K+ -ATPase function by hypoxia in rats. J Appl Physiol 1999; 87: 9628
10 Sznajder JI, Olivera WG, Ridge KM, Rutschman DH. Mechanism of lung liquid clearance during hyperoxia in isolated rat lungs. Am J Respir Crit Care Med 1995; 151: 151925[Abstract]
11 Junor RWJ, Benjamin AR, Alexandrou D, Walters DV. The independence of lung liquid absorption in postnatal sheep on pulmonary blood flow, blood gases or perfusion pressure. Exp Physiol 2001; 86: 3918[Abstract]
12 West JB, Wagner PD. Predicted gas exchange on the summit of Mt. Everest. Respir Physiol 1980; 42: 116[ISI][Medline]
13 ONeill M, Vejlstrup NG, Nagyova B, Dorrington KL. Dependence of pulmonary venous admixture on inspired oxygen fraction and time during regional hypoxia in the rabbit. Br J Anaesth 1995; 75: 6039
14 Vejlstrup NG, ONeill M, Nagyova B, Dorrington KL. Time course of hypoxic pulmonary vasoconstriction: a rabbit model of regional hypoxia. Am J Respir Crit Care Med 1997; 155: 21621[Abstract]
15 Von Neergaard K. Neue Auffassungen über einen Grundbegriff der Atemmechanik. Z Ges Exp Med 1929; 66: 37394
16 Weisbroth SH, Flatt RE, Kraus AL, eds. The Biology of the Laboratory Rabbit. New York: Academic Press, 1974; 57
17 Mitruka BM, Rawnsley HM. Clinical Biochemical and Hematological Reference Values in Normal Experimental Animals and Normal Humans, 2nd Edn, New York: Masson Publishing Inc. 1982; 56
18 Dorrington KL, Boyd CAR. Active transport in the alveolar epithelium of the adult lung; vestigial or vital? Respir Physiol 1995; 100: 17783[ISI][Medline]
19 Matthay MA, Wiener-Kronish JP. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am Rev Respir Dis 1990; 142: 12507[ISI][Medline]
20 Dorrington KL, Young JD. Development of the concept of a liquid pulmonary alveolar lining layer. Br J Anaesth 2001; 86: 6147
21 Widdicombe JH, Widdicombe JG. Regulation of human airway surface liquid. Respir Physiol 1995; 99: 312[ISI][Medline]
22 Sakuma T, Okaniwa G, Nakada T, Nishimura T, Fujimura S, Matthay MA. Alveolar fluid clearance in the resected human lung. Am J Respir Crit Care Med 1994; 150: 30510[Abstract]
23 Nunn JF. Nunns Applied Respiratory Physiology, 4th Edn. Oxford: Butterworth-Heinemann Ltd, 1993; 343
24 Welling K-LK, Sanches R, Ravn JB, Larsen B, Amtorp O. Effect of prolonged alveolar hypoxia on pulmonary arterial pressure and segmental vascular resistance. J Appl Physiol 1993; 75: 1194200[Abstract]
25 Zhao Y, Packer CS, Rhoades RA. Pulmonary vein contracts in response to hypoxia. Am J Physiol 1993; 265: L8792
26 Sanchez de Léon R, Orchard CH, Chakrabati MK, Sykes MK. Effect of hypoxia on fluid filtration rate during forward and reverse perfusion of isolated rabbit lungs. Cardiovasc Res 1982; 16: 7115[ISI][Medline]
27 Marshall BE, Marshall C, Frasch F, Hanson CW. Role of hypoxic pulmonary vasoconstriction in pulmonary gas exchange and blood flow distribution. 1. Physiologic concepts. Intensive Care Med 1994; 20: 2917[ISI][Medline]
28 Levitzky MG, Newell JC, Dutton RE. Effect of chemoreceptor denervation on the pulmonary vascular response to atelectasis. Respir Physiol 1978; 35: 4351[ISI][Medline]
29 Weissmann N, Gimminger F, Walmrath D, Seeger W. Hypoxic vasoconstriction in buffer-perfused rabbit lungs. Respir Physiol 1995; 100: 15969[ISI][Medline]
30 Semenza GL. HIF-1: mediator of physiological and patho physiological responses to hypoxia. J Appl Physiol 2000; 88: 147480
31 Ramminger SJ, Baines DL, Olver RE, Wilson SM. The effects of P O2 upon transepithelial ion transport in fetal rat distal lung epithelial cells. J Physiol 2000; 524: 53947