Consequences of vascular flow on lung injury induced by mechanical ventilation.

To investigate whether the magnitude of blood flow contributes to ventilator-induced lung injury, 14 sets of isolated rabbit lungs were randomized for perfusion at either 300 (Group A: n = 7) or 900 ml/ min (Group B: n = 7) while ventilated with 30 cm H2O peak static pressure. Control lungs (Group C: n = 7) were ventilated with lower peak static pressure (15 cm H2O) and perfused at 500 ml/min. Weight gain, changes in the ultrafiltration coefficient (DeltaKf) and lung static compliance (CL), and extent of hemorrhage (scored by histology) were compared. Group B had a larger decrease in CL (-13 +/- 11%) than Groups A (2 +/- 6%) and C (5 +/- 5%) (p < 0.05). Group B had more hemorrhage and gained more weight (16.2 +/- 9.5 g) than Groups A (8.7 +/- 3.4 g) and C (1.6 +/- 1.0 g) (p < 0.05 for each pairwise comparison between groups). Finally, Kf (g . min-1 . cm H2O-1 . 100 g-1) increased the most in Group B (DeltaKf = 0.26 +/- 0. 20 versus 0.17 +/- 0.10 in Group A and 0.05 +/- 0.04 in Group C; p < 0.05 for B versus C). We conclude that the intensity of lung perfusion contributes to ventilator- induced lung injury in this model.

The effects of alcohol consumption on laboratory-induced panic and state anxiety.

BACKGROUND: This study tested whether alcohol consumption reduces anxiety and panic associated with a panic-challenge procedure. METHODS: Subjects with panic disorder were randomly assigned to consume either a moderate dose of alcohol or a nonalcoholic placebo. All subjects were told that they were drinking alcohol to control beverage expectancies. Following the beverage administration, subjects underwent a panic challenge (35% carbon dioxide) and a series of anxiety symptom assessments. RESULTS: Subjects who consumed alcohol reported significantly less state anxiety both before and after the challenge. In response to the challenge, subjects who consumed alcohol experienced significantly fewer panic attacks when applying liberal panic criteria; however, this effect only approached significance when applying conservative panic criteria. CONCLUSIONS: These findings suggest that alcohol acts acutely to reduce both panic and the anxiety surrounding panic, and they lend support to the view that drinking behavior among those with panic disorder is reinforced by this effect. We suggest that this process may contribute to the high rate at which alcohol-use disorders co-occur with panic disorder.

Induction of alveolar epithelial injury by phospholipase A2.

Severe damage to the alveolar type I epithelial cell is a characteristic morphological feature of lung injury due to numerous cases. It is postulated that excess phospholipase A2 (PLA2) activity might be responsible for these changes, as one of the naturally occurring products of this enzyme, lysophosphatidylcholine (lysoPC) has been shown to cause selective injury to the type I pneumonocyte when it is instilled into the lower air spaces of the lung. To further investigate this potential mechanism of type I epithelial cell toxicity, we have measured the epithelial permeability-surface area product (PS) for [14C]sucrose as well as whole-lung lysoPC content at several times after instilling PLA2 (Naja naja venom) into either the air spaces or the perfusate of an isolated hamster lung preparation. As a molar percentage of total phospholipids, the normal hamster lung contains approximately 1.5% lysoPC, and this value is not affected by fluid filling of the air spaces or perfusion of the excised lung for periods up to 90 min. When 0.15 U/ml PLA2 is instilled into the air spaces, lung lysoPC content increases to approximately 2.5% and there are barely detectable increases in [14C]sucrose PS. With air space PLA2 concentrations of 0.30 U/ml, lysoPC content increases to between 4 and 5%, [14C]sucrose PS increases by greater than a factor of 10, and flooding of the alveolar spaces occur. Ultrastructural studies of similarly treated lungs show widespread but selective damage to the type I epithelial cells. These same biochemical and functional changes are not seen when the same concentrations of PLA2 are added to the lung perfusate.(ABSTRACT TRUNCATED AT 250 WORDS)

Injurious effects of lysophosphatidylcholine on barrier properties of alveolar epithelium.

We studied the effects of lysophosphatidylcholine (lysoPC) on the barrier properties and the morphology of the alveolar-capillary membrane in isolated, fluid-filled hamster lungs continuously perfused. When instilled into the airspace at initial concentrations of 8-128 micrograms/ml, lysoPC causes dose-dependent increases in the permeability-surface area product of the alveolar epithelium for small (14C-sucrose, 342) and large (125I-neutral dextran, 70,000) solutes, with maximal values for each solute approximately 15 times control. Rapid whole-lung weight gains are caused by 128 micrograms lysoPC per milliliter, but each of the lower concentrations has no effect on net lung water balance. Electron-microscopic studies demonstrate that type I pneumonocytes are the lung cells most susceptible to lysoPC exposure, with cell swelling being the most prominent feature from low-dose exposure with more severe disruptive changes at the highest concentration tested. The effects of lysoPC are relatively specific, as several structurally related lipids have little or no effect at equivalent concentrations. Instillation of phospholipase A2 causes functional changes similar to those seen with lysoPC, presumably by generation of lysoPC from endogenous phospholipids. Studies employing a 14C-radiolabeled compound show that instilled lysoPC rapidly partitions into the lung lipid fraction where a major portion of the acyl group becomes incorporated into phosphatidylcholine. The amount of instilled lysoPC required to produce functional and morphological effects comprises only a few percent of total lung phospholipids. Since lysoPC is a normal component of lung phospholipids, severe lung dysfunction might result from minor abnormalities in the formation or degradation of this compound.

Effect of proteolytic enzymes on transepithelial solute transport.

The effects of proteases on air-space clearance (AC) of small ([14C]sucrose, 342 daltons) and large (125I-neutral dextran, 70,000 daltons) solutes were studied in isolated, fluid-filled hamster lungs that were perfused in a nonrecirculating system. When instilled into the air spaces, porcine pancreatic elastase (0.1-0.4 mg/ml) and bovine pancreatic trypsin (BPT) (0.5-2.0 mg/ml), but neither Clostridium histolyticum collagenase (5.0 mg/ml) nor phenylmethylsulfonyl fluoride-inactivated BPT caused large increases in the AC of both tracer molecules. BPT-induced solute clearance was further characterized functionally and morphologically. The functional characteristics of solute AC under steady-state conditions did not indicate that transepithelial transport was diffusion-limited. Inhibition by millimolar concentrations of Zn2+ and by lung cooling, along with electron microscopic studies employing horseradish peroxidase as a macromolecule tracer, were consistent with epithelial solute transport by a vesicular mechanism (transcytosis). Solute transport from the interstitial compartment to the lung exterior was shown to occur via two pathways. By unknown mechanisms BPT caused small amounts of water to flow through an incompletely identified, extravascular pathway. In BPT-exposed lungs efflux of 125I-dextran 70 occurred almost exclusively through this pathway, whereas [14C]sucrose was transported to the lung exterior partly through this same pathway and partly through the vasculature. The large differences in the diffusion coefficients of the two tracers may have accounted for these observed patterns of solute efflux from the lung. The possible significance of our findings to the pathogenesis of experimental emphysema are discussed.

Solute conductance of blood-gas barrier in hamsters exposed to hyperoxia.

Hamsters were exposed to greater than 95% O2 continuously for up to 5 days to determine longitudinal changes in the diffusive conductance of the alveolar epithelium and capillary endothelium as a result of hyperoxia. Permeability X surface area (PS, cm3/s X 10(-4)) was measured by isolated, perfused lung techniques. Alveolar epithelium PS for [14C]sucrose and 125I-bovine serum albumin (BSA) were determined at seven exposure times. Control PS (sucrose) and PS(BSA) averaged 1.00 and 0.022, respectively. Values were unchanged until 4.5 days, when significant increases in both, but especially PS(BSA), occurred. After 5 days, PS values were 4.69 and 0.691, respectively. Capillary endothelium PS for 125I-BSA and fluoresceinisothiocyanate dextran-150 (D-150) were measured at four exposure times. Control endothelium PS(BSA) and PS(D-150) averaged 0.232 and 0.048, respectively. These values were also unchanged after 4 days but increased to 0.440 and 0.131 after 5 days. Wet lung weight significantly increased after only 4 days. Hyperoxia thus increased both endothelium and epithelium PS, but epithelium changes were much greater. These functional changes do not occur for several days, occur simultaneously, and follow increases in lung wet weight.

Effects of elastase and cigarette smoke on alveolar epithelial permeability.

To determine whether instilled porcine pancreatic elastase (PPE) increases alveolar epithelial permeability, we measured alveolar epithelium permeability X surface area (PS) for [14C]sucrose and 125I-bovine serum albumin (125I-BSA) in isolated perfused lungs from hamsters previously exposed to PPE and/or cigarette smoke. Saline (0.5 ml) with 0, 5, or 20 units PPE was instilled intratracheally in anesthetized hamsters. Those exposed to smoke for 4-6 wk received 0 or 5 units; PS was measured 3 h later. Nonsmokers received 0, 5, or 20 units; PS was measured 3 h, 24 h, or 5 days later. Control PS values were (cm3/s X 10(-4), +/- SE) 0.84 +/- 0.11 for sucrose and 0.030 +/- 0.006 for BSA. Three and 24 h following 20 units PPE, (PS)sucrose was twice the control valve. (PS)BSA was four times control at 3 h but not significantly increased at 24 h. Five days after PPE both were back to control levels. Five units PPE or smoke exposure alone caused no PS changes. Smoke exposure and 5 units PPE caused (PS)sucrose to increase markedly (1.85 +/- 0.32); (PS)BSA was not significantly increased (0.076 +/- 0.026). Thus instilled PPE causes reversible increases in alveolar epithelial PS; cigarette smoking potentiates this effect.

D- and L-glucose transport across the pulmonary epithelium.

Previously we observed what appeared to be augmented D-glucose transport across the pulmonary epithelium. To investigate this phenomenon we placed fluid containing L-[3H]glucose and D-[U-14C]glucose in the alveoli of isolated Ringer-perfused lungs from 4-wk-old rabbits. The appearance of radioactivity in recirculating glucose-free perfusate was measured. 3H appearing in the perfusate was associated with L-glucose. 14C, however, was associated with three compounds, with approximate molecular weights of 180 (glucose), 300, and 560. The nonglucose species were not identified. This 14C movement was inhibited by phlorizin, but not phloretin, in the alveolar fluid. A similar pattern of 14C movement occurred when D-[U-14C]glucose was replaced with 2-deoxy-D-[U14C]-glucose, but not with methyl-alpha-D-[U-14C]glucopyranoside. The activation energy of the 14C metabolism-transport process was found to be 34 kcal/mol, and L-glucose transport showed an unusual temperature dependence, with maximum conductance at 15 degrees C. It appears that some D-glucose crosses the pulmonary epithelium as does L-glucose. However, most enters epithelial cells and is incorporated into larger molecules which enter the vascular but not the alveolar space.

Endothelial albumin permeability measured with a new technique in perfused rabbit lung.

This study describes two methods of measuring the albumin permeability-surface area product (PS) of the pulmonary microvascular endothelium in isolated, perfused rabbit lungs. Both involve direct sampling of lung tissue to measure how much labeled albumin has entered the extravascular space during a known exposure period. In one, the traditional "multiple-sample" method, extravascular albumin is measured for several different exposure times to determine the kinetics of transcapillary albumin movement, and PS is calculated from that data. We found single-exponential kinetics, with a half time of 77 min and a PS of 2.28 X 10(-3) cm3 . min-1 . g blood-free wet lung-1. The second, "single-sample" method, is a new modification that involves exposing the tissue to labeled albumin for only 3 min, followed by a 3-min vascular washout. Tissue is then analyzed for extravascular albumin and PS calculated assuming the interstitial albumin concentration is essentially zero. The PS obtained was 2.44 X 10(-3) cm3 . min-1 . g blood-free wet lung-1, not significantly different from the value obtained using the more accepted method. Possible sources of error in the new method are discussed, and we conclude that it is an easy, quick, reliable way of accurately measuring macromolecular PS in the lung and presumably other organs as well.

Morphometric estimates of diffusing capacity in lungs fixed under zone II and zone III conditions.

Comparative morphometric estimates of the diffusing capacity (DL) were made in rabbit lungs fixed by vascular perfusion under lower zone II and zone III conditions and in lungs fixed by instillation of fixatives into the airways. Owing to a reduction of both capillary volume and membrane diffusing capacity DL of zone II lungs (0.074 +/- 0.007 (SD) ml . sec-1 . mbar-1). was found to be lower by some 25% than DL of instillation-fixed lungs (0.102 +/- 0.012 (SD) ml . sec-1 . mbar-1). The average value of DL of air-filled zone III lungs, on the other hand, almost matched the DL of instillation-fixed lungs. However, DL is not equal in all regions but increases along the vertical axis of zone III lungs. Hence, the previous conclusion that morphometric estimates of DL in instillation-fixed lungs reflects a structural limit for O2 diffusion, which cannot be reached under physiologic conditions, must be revised.

Bleomycin-induced changes in pulmonary microvascular albumin permeability and extravascular albumin space.

Pulmonary microvascular permeability to serum albumin and the extravascular albumin space (EAS) were measured in rat lungs 5 days after intratracheal instillation of bleomycin. The albumin permeability-surface area product (PS) was measured using a new method: lungs were removed and perfused with Ringer's solution; they were then perfused for 3 min with Ringer's containing [125I]albumin, followed by 3 min with plain Ringer's to clear the vascular space. The PS was calculated from the 125I activity in perfusate and homogenized lung tissue. In separate experiments the EAS was measured using standard methods. Compared with control rats, the injected animals showed a slight, but significant, increase in PS, and a doubling of the EAS. In previous work, using other techniques, the EAS increase was interpreted as an increased PS. Our new method for PS measurement is easy and more accurate than those previously used, and shows that the acute pulmonary response to intratracheally administered bleomycin involves significant interstitial changes with little alteration in the microvascular endothelium.

Surfaces and volumes of alveolar tissue under zone II and zone III conditions.

The functional anatomy of alveolar septa has been studied in rabbit lungs fixed by vascular perfusion under middle zone II and zone III conditions at a constant lung inflation level of 60% of total lung capacity. Capillary volume increases down the vertical axis of the lungs in both conditions and is larger by about 30% in zone III than in zone II lungs. The concomitant changes of free alveolar, epithelial, and capillary surface areas are small. Capillary volume increases within the observed range of pressures neither by recruitment of collapsed capillaries nor by unfolding of pleats of alveolar septa and capillary walls, but rather by deformation and distension of tissue: scanning electron micrographs reveal a more conspicuous bulging of capillaries in zone III than in zone II lungs. Accordingly, the mechanical structure of alveolar septa appears to be largely consistent with a sheet-flow model.

Perfusion fixation of lungs for structure-function analysis: credits and limitations.

The quality of tissue preservation in lungs fixed by vascular perfusion has been reevaluated. Excised rabbit lungs inflated to 60% of total lung capacity were perfused (zone III conditions) with different but widely used fixatives. The effects of the perfusates on pertinent physiological variables have been assessed by a continuous monitoring, the effects on the pulmonary microstructure by qualitative and morphometric analysis of electron micrographs. Important results include the following. 1) Perfusions with isotonic glutaraldehyde at flow rates within the physiological range produce large increases of perfusion pressure and lung weight that reflect intracellular, interstitial, and intra-alveolar edema. 2) No edema occurs if glutaraldehyde is added to isotonic buffer solutions (total osmolarity 510 mosM). 3) Glutaraldehyde as sole perfusate does not fully eliminate the retractive force of lung tissue. Upon release of transpulmonary pressure the lungs retract by an indeterminable amount. 4) Satisfactory results can be obtained by sequential perfusion with osmium tetroxide and uranyl acetate or glutaraldehyde (510 mosM) followed by osmium tetroxide and uranyl acetate. The latter combination yields optimal preparations to study the alveolar and capillary architecture but causes a hyperosmotic volume loss of lung cells (cell shrinkage).

Morphometric evaluation of chorioallantoic oxygen transport in the chick embryo.

A morphologic and morphometric study was conducted to elucidate the mechanism of O2 transport across the inner shell membrane and chorioallantoic blood-gas barrier in chick embryos. The chorioallantois and shell membranes in 16-day-old incubating chicken eggs were fixed in situ, systematically sampled, and prepared for electron microscopic examination. Scanning and transmission electron micrographs are presented to describe the pathway for respiratory gas exchange. Using stereologic methods the dimensions of important gas exchange parameters were measured and from those data the O2 diffusing capacity of the chorioallantois (DCA) was calculated. Our average DCA value of 6.8 microliter O2 . min-1 . Torr-1 is quite similar to previous physiologic estimates. We found that the rate-limiting factor in chorioallantoic O2 uptake is O2-hemoglobin binding in erythrocytes, which is ten times slower than diffusion across the thin (harmonic mean thickness = 0.47 micrometer) blood-gas barrier. Our analysis provides strong support for the chorioallantoic gas exchange model of Piiper et al. (1980), and implies that the inner shell membrane provides a negligible resistance to O2 movement at this developmental age.

Alveolar epithelium permeability to small solutes: developmental changes.

To determine whether alveolar epithelium permeability to small lipid-insoluble solutes changes during development we measured transport across the blood-gas barrier in isolated Ringer-perfused lungs from prenatal, 1-day-old, 4-wk-old, and adult rabbits. Radioactive test molecules, one of which was always sucrose, were dissolved in Ringer solution and instilled into the trachea of degassed lungs. Samples taken from recirculating perfusate were used to calculate permeability-surface area (PS) products. Results were expressed as the ratio (PS)/(PS)sucrose, and as absolute permeability. Lungs from 4-wk-old rabbits were studied most thoroughly; the (PS)/(PS) sucrose ratios obtained are urea 4.0, erythritol 1.3, mannitol 0.98, L-glucose 1.4, and D-glucose 5.6. These and other data imply that the most lipid-insoluble molecules (erythritol, mannitol, L-glucose, and sucrose) are transported by a nonselective bulk process. Urea transport is primarily through lipid membranes; D-glucose seems to involve a special process. Sucrose and L-glucose permeability decreased during development, but their relative permeabilities did not change. Small lipid-insoluble solutes apparently do not cross the alveolar epithelium through small water-filled pores, and their permeability decreases as the animal matures.

Lung tissue volume changes induced by hypertonic NaCl: morphometric evaluation.

Lung tissue was examined to determine how the volumes of alveolar septum components change when NaCl is added to the vascular perfusate, increasing the osmolarity by 70 mosM. Isolated rabbit lungs were perfused with Ringer solution containing dextran, either with or without added NaCl, and fixed by vascular perfusion. Tissue samples from both "control" and "hypertonic" lungs, prepared for electron microscopy, were examined using established morphometric procedures. Volumes of septal cells, interstitial space, capillary lumen, surface-lining layer, and endothelial and epithelial areas were measured, all normalized against the endothelium basement-membrane area. Results showed that hypertonic NaCl caused a reduction in total cell and surface-lining layer volumes but no change in interstitial or capillary lumen volumes. This supports the hypothesis that small molecules have no osmotic effect across the pulmonary capillary endothelium but do cause a fluid flux from cells and across the alveolar epithelium. Areas and volume measurements for different septal cell types suggest a heterogeneous response: epithelial cells showed significant decreases and endothelial cells changed little, if at all.

Effects of glutaraldehyde or osmium tetroxide fixation on the osmotic properties of lung cells.

The osmotic properties of lung cells have been tested before and after perfusion fixation of isolated, perfused lungs with either glutaraldehyde or osmium tetroxide. The testing procedure was to add hypertonic sucrose to the perfusate for several minutes and monitor the lung weight response (an 'osmotic transient'). Each lung was perfused with one or the other fixative solutions for 10 min, then the perfusate was changed back to Ringer-lactate before the post-fixation test was conducted. The results indicate that osmium tetroxide makes the cell membranes as permeable to sucrose as to water, and that sucrose thus causes no osmotic volume change. Glutaraldehyde, on the other hand, apparently preserves the impermeability of the cell membranes to sucrose, but the osmotic volume response is attenuated, indicating that significant changes in the cells have occurred.

Alveolar epithelium transport of albumin and sucrose: concentration difference effect.

Isolated Ringer-perfused rabbit lungs were used to study albumin and sucrose transport across the blood-gas barrier. Lungs were filled to about 20% total lung capacity (TLC) ith Ringer solution containing radioactive albumin and sucrose, and their rate of appearance in the recirculating perfusate was monitored. From this the product permeability x area (PA) was calculated. In the middle of the 180-min experiments, some alveolar fluid was removed. In control experiments, the same fluid was reinstilled, in other experiments new fluid with higher test molecule concentrations was infused. In all experiments the results for both molecules were similar: PA for the second half of the experiment was 80% of that in the first half. The reduction was probably due to a decrease in exchange area. We thus find the albumin and sucrose permeabilities to be proportional to their concentration difference. In addition, the PA for sucrose was roughly four times that of albumin. These results can possibly be explained by a two-compartment model with two parallel pathways across the alveolar epithelium. One pathway would be small pores, whereas the other would involve a bulk flow or pinocytosis process.