INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HYDROSTATIC PULMONARY EDEMA is associated with an increase in the microvascular pressure gradient in the pulmonary circulation in the presence of elevated left atrial pressures (LAPs) (13, 24). Hydrostatic edema is the result of an imbalance in the Starling forces across the pulmonary capillary endothelial barrier (26), which results in increased flux of fluid from the pulmonary capillaries into the interstitial space, as seen in patients with congestive heart failure (14, 24). A pressure gradient exists between the alveolar and extra-alveolar compartments of the interstitial space (5); therefore, transuded fluid first moves to the extra-alveolar interstitium. A significant portion of edema fluid is then directly absorbed into the circulation or is returned to the circulation via the lymphatic system. Thus the lung interstitium and lymphatic drainage constitute the first line of defense against edema formation (24, 27). However, if the increase in microvascular pressures persists, these defense mechanisms become overwhelmed, and edema appears in the interstitium and the alveolus.

Pulmonary edema resolution in mammalian lungs is effected by active Na⁺ transport across the alveolar epithelium (23). Alveolar fluid reabsorption is regulated in epithelial cells by the rate of Na⁺ entry via the apical Na⁺ channels coupled to the rate of Na⁺ extrusion via the basolateral Na⁺-K⁺-ATPase (6, 16-18, 25). Water moves out of the alveolar space, following the osmotic gradients generated by active Na⁺ transport through water channels located in the alveolar epithelium (10) and other pathways. The rate of lung edema clearance has been measured in several species under normal and pathological conditions (9, 15, 19-22). However, active Na⁺ transport and the rate of fluid reabsorption in lungs exposed to elevated LAP have only been partially studied (2, 4, 7, 12)

This study was designed to examine the effects of increasing LAP on alveolar epithelial fluid reabsorption in an isolated, perfused rat lung model. The data show that increasing LAP results in increased solute permeability and decreased active Na⁺ transport and fluid reabsorption. Restoration of pulmonary circulation pressures to control values restores the lung permeability to solutes and edema clearance.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Pathogen-free male Sprague-Dawley rats weighing 280-320 g were purchased from Harlan Sprague Dawley (Indianapolis, IN). In all, 78 rats were studied. All animals were provided food and water ad libitum and were maintained on a 12:12-h light-dark cycle. All animals were cared for in accordance with NIH guidelines. Benzamil, amiloride, and ouabain were purchased from Sigma (St. Louis, MO).

Specific protocols. Alveolar epithelial Na⁺ transport and lung permeability to small and large solutes were examined in rat lungs instilled with 5 ml of buffered salt albumin solution (BSA) into the airspaces and exposed to increased LAPs (5, 10, 15, and 20 cmH₂O) over 1 h (n = 6 lungs/group) and were compared with those in control rat lungs exposed to a LAP of 0 cmH₂O (n = 10 lungs).

4

4

4

Isolated lungs. The isolated, perfused lung preparation was performed as previously described (15, 19, 21, 22). Briefly, rats were anesthetized with 50 mg/kg body wt of pentobarbital sodium; tracheotomized; and mechanically ventilated with a tidal volume of 2.5 ml, a peak airway pressure of 8-10 cmH₂O, and 100% oxygen for 5 min. The chest was opened via a median sternotomy after which 400 U of heparin sodium were injected into the right ventricle. After exsanguination, the heart and lungs were removed en bloc. The pulmonary artery and left atrium were catheterized, and the pulmonary circulation was flushed of remaining blood by perfusion with BSA containing 135.5 mM Na⁺, 119.1 mM Cl, 25 mM HCO, 4.1 mM K⁺, 2.8 mM Mg²⁺, 2.5 mM Ca²⁺, 0.8 mM SO, 8.3 mM glucose, and 3% bovine serum albumin at an osmolality of 300 mosmol/kgH₂O. The BSA was maintained at pH 7.4 by bubbling with a mixture of 5% CO₂ and 95% O₂ as needed. Two sequential bronchoalveolar lavages were performed with 3 ml of BSA containing 0.1 mg/ml of Evans blue dye (EBD; Sigma), 0.02 µCi/ml of ²²Na⁺ (Dupont NEN, Boston, MA), and 0.12 µCi/ml of [³H]mannitol (Dupont NEN). The volume of the epithelial lining fluid (ELF) was estimated by the dilution of EBD in the first bronchoalveolar lavage. The lungs were then instilled with the volume necessary to leave 5 ml in the alveolar space. Finally, the lungs were immersed in a "pleural bath" reservoir containing 100 ml of BSA maintained at 37°C. This allowed us to follow markers that had moved across the pleural membrane or had been drained by the lung lymphatics.

Calculations. The alveolar lining fluid volume (V_ELF) was calculated by instilling 3 ml of fluid [initial volume (V₀)] containing a known initial concentration of albumin tagged with EBD ([EBD]₀) into the airspace. After brief mixing, a sample was removed, and the EBD concentration at time t ([EBD]_t) was estimated. The amount of EBD was the same in the instillate (V₀[EBD]₀) and in the lung after initial mixing {(V₀ + V_ELF) × [EBD]_t}. Equating the two yields

(1)

or

(2)

Similarly, the alveolar fluid volume at time t (V_t) was estimated by

(3)

The movement of Na⁺ from the alveolar space during a defined period of time was assumed to be accompanied by isotonic water flux and is given as J_Na,net = J_Na,out  J_Na,in, where J_Na,net is the net or active Na⁺ transport, J_Na,out is the total or unidirectional Na⁺ outflux from the rat airspaces, J_Na,in is the passive bidirectional flux of Na⁺ between the airspace and the other compartments, and [Na⁺] is the constant Na⁺ concentration in the BSA. The volume flux, J = J_Na,net/[Na⁺], is the average rate of change in the volume and is given as

(4)

As described by Rutschman et al. (19), the passive movement of ²²Na⁺, J_Na,in, is given as

(5)

where C₀ is the initial [²²Na⁺] and C_t is the [²²Na⁺] at time t.

(6)

Perfusion fixation. Before fixation, the lungs were perfused with BSA with LAPs of 0, 5, 10, 15, or 20 cmH₂O over 1 h. The lungs were then perfused for 15 min with a phosphate-buffered 1.5% glutaraldehyde solution (pH 7.4) while left LAPs were maintained. The lungs were then cut into small pieces, fixed in 2.5% glutaraldehyde in 0.1 M PBS (pH 7.4) at 4°C overnight, rinsed for 45 min in PBS, postfixed in 1% osmium tetroxide in PBS for 1-2 h at room temperature, dehydrated in a graded series of ethanols, and embedded in Epon. Semithin sections were cut on a microtome, stained with uranyl acetate and lead citrate, and examined with a transmission electron microscope.

Data analysis. Data are means ± SE; n is the number of animals in each experimental group. When comparisons were made between two experimental groups, an unpaired Student's t-test was used. When more than one comparison was made, analysis of variance was used with Tukey's procedure to determine where the differences were. Results were considered significant when P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Alveolar epithelial permeability. Alveolar epithelial permeability to small solutes (²²Na⁺ and [³H]mannitol) did not change in rats exposed to mild left atrial hypertension over 1 h (LAP 5 and 10 cmH₂O) compared with that in control rats (LAP 0 cmH₂O). However, lung permeability to small solutes increased progressively in rat lungs exposed to high pulmonary circulation pressures (LAP 20 cmH₂O; Fig. 1A).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   A: passive 22Na+ and [3H]mannitol movement increased in rats exposed to elevated left atrial pressure (LAP). Values are means ± SE. Significantly different from control group (LAP 0 cmH2O): *P < 0.05; **P < 0.01. B: movement of albumin from the pulmonary circulation into the airspace progressively increased in rats exposed to left atrial hypertension. Values are means ± SE. ***P < 0.001 compared with control group (LAP 0 cmH2O).

Alveolar fluid clearance. Vectorial Na⁺ transport and alveolar fluid reabsorption did not change in rats exposed to mild left atrial hypertension (LAP 5 and 10 cmH₂O) compared with that in control rat lungs (LAP 0 cmH₂O; 0.50 ± 0.02 ml/h; Fig. 2). Active Na⁺ transport and alveolar fluid clearance decreased by ~50% in rats exposed to a LAP of 15 cmH₂O compared with levels in control rats (P < 0.01). However, a LAP of 20 cmH₂O caused edema, precluding the ability to assess alveolar fluid reabsorption.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Alveolar fluid clearance decreased in rats exposed to moderate LAP (15 cmH2O). Values are means ± SE. There was a complete disruption of alveolocapillary barrier in rats exposed to high LAP (20 cmH2O). **Significantly different from control group (LAP 0 cmH2O): P < 0.01; ***P < 0.001.

4

5

View larger version (9K): [in this window] [in a new window]   Fig. 3.   The Na+ channel blocker benzamil inhibited ~80% of alveolar fluid clearance in rats exposed to normal and moderately elevated LAP (0 and 15 cmH2O, respectively). Values are means ± SE. ***P < 0.001 compared with control rats (LAP 0 cmH2O). &P < 0.001 compared with moderate left atrial hypertension (LAP 15 cmH2O).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   A: alveolar epithelial Na+ transport and alveolar fluid clearance were completely inhibited by exposing rat lungs to hypothermia (4°C) in presence of normal and moderately elevated LAP (0 and 15 cmH2O, respectively). Values are means ± SE. ***P < 0.001 compared with control group (LAP 0 cmH2O). B: amiloride instilled into airspaces and ouabain perfused through the pulmonary circulation inhibited alveolar fluid clearance in rats exposed to normal and mild to moderate elevated LAP (0 and 15 cmH2O, respectively). Values are means ± SE. ***P < 0.001 compared with control group (LAP 0 cmH2O).

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Alveolar fluid clearance inhibited by mild to moderate left atrial hypertension rapidly recovers after normalization of LAP (0 cmH2O). Values are means ± SE. ***P < 0.001 compared with moderately elevated LAP (15 cmH2O).

Changes in alveolar morphology. In rats exposed to mild left atrial hypertension over 1 h (LAP 0, 5, and 10 cmH₂O), no alveolocapillary disruptions were observed with histological analysis. However, in rat lungs exposed to moderate or high pulmonary circulation pressures (LAP 15 and 20 cmH₂O), lesions in the alveolocapillary barrier were observed in the form of alveolocapillary blebs found ultrastructrually (Fig. 6).

View larger version (83K): [in this window] [in a new window]   Fig. 6.   In rats exposed to mild left atrial hypertension (LAP 0, 5, and 10 cmH2O for 1 h; A-C, respectively), epithelial disruptions were not observed by histological analysis with electron microscopy. D and E: in rat lungs exposed to increasing pulmonary circulation pressures (LAP 15 and 20 cmH2O), lesions in the alveolocapillary barrier were observed in the form of alveolocapillary blebs (*). F and G: enlarged images of alveolocapillary blebs depicted in D and E, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Pulmonary edema formation depends on changes in hydrostatic or colloid-osmotic pressure gradients in the pulmonary circulation or increased alveolocapillary barrier permeability (24). In patients with acute or chronic heart failure, hydrostatic pulmonary edema results from increased microvascular pressure, which results from high LAP (27). Interstitial edema forms when an imbalance exists between the rate of fluid filtration into the pulmonary interstitium and the alveolus and the rate of alveolar fluid reabsorption (13, 24, 27). It is well known that when LAP increases, edema increases, and alveolar flooding must be resolved if a patient with pulmonary edema is to recover (8). However, the mechanisms of fluid resolution associated with increasing hydrostatic pulmonary circulation pressures have not been fully elucidated. Because alveolar fluid reabsorption is effected by active Na⁺ transport across the alveolar epithelium (10, 16, 19, 21, 23), we studied alveolar fluid clearance while increasing the hydrostatic pulmonary circulation pressures.

As shown in Fig. 2, alveolar fluid reabsorption was normal in animals exposed to low to mild left atrial hypertension (LAP 5 and 10 cmH₂O) for 1 h. However, active Na⁺ transport was inhibited by higher hydrostatic pulmonary circulation pressures (LAP 15 cmH₂O). The lung permeability to small and large solutes increased progressively in rats exposed to high LAP (Fig. 1), which is concordant with a previous report in which sheep ventilated with high LAP (24 cmH₂O) had a 30% reduction in alveolar fluid clearance (7).

Hydrostatic pulmonary edema (LAP 15 cmH₂O) in rats decreased active Na⁺ transport and alveolar fluid clearance, with concomitant changes in lung permeability to small (Na⁺ and mannitol) and large (albumin) solutes. Rat lungs exposed to either hypothermia (4°C) or both amiloride and ouabain had no significant movement of fluid from the pulmonary circulation into the alveolar space at a LAP of 15 cmH₂O compared with control rat lungs (LAP 0 cmH₂O; Fig. 4). These data suggest that at 15 cmH₂O, there is no significant edema formation regardless of the fact that the ability of the lung to clear edema is impaired. These data demonstrate further that the isolated perfused rat lung model can be used to accurately assess alveolar fluid clearance, as previously reported in normal lungs and in models of mild to moderate lung injury such as hyperoxia, mechanical ventilation, and increased LAP (15, 19, 21, 22). However, increasing LAP to 20 cmH₂O resulted in significant edema formation, which precluded us from assessing alveolar fluid reabsorption rates in this model.

The mechanisms responsible for the decrease in active Na⁺ transport in this model have not been elucidated. A recent report by Fukuda et al. (12) suggests that slower rates of alveolar fluid reabsorption may be related to the accumulation of interstitial fluid in the lung, which would represent a physical barrier to fluid reabsorption. Another possibility is that high pressures by mechanotransduction could disrupt mechanisms that regulate active Na⁺ transport, such as apical Na⁺ channels and/or Na⁺-K⁺-ATPase. Two recent reports (1, 11) suggest that catecholamines may accelerate the rate of alveolar fluid reabsorption in rats and sheep, respectively. In the report by Frank et al. (11), rats that were administered salmeterol had a 62% reduction in excess lung water; however, there was no difference observed in excess lung water at 4 h compared with that in control rats. Additionally, the rate of fluid clearance after the induction of left atrial hypertension was similar to control rates. The report by Azzam et al. (1) demonstrated that dopamine and isoproterenol increased alveolar fluid reasborption and active Na⁺ transport, possibly due to the recruitment of Na/K pumps from intracellular pools to the plasma membrane of the alveolar epithelium. The understanding of these mechanisms will be important in the regulation of alveolar fluid reabsorption in patients with hydrostatic pulmonary edema (28).

To examine the mechanisms contributing to active Na⁺ transport in the rat alveolar epithelium, we studied the effects of the Na⁺ channel blocker benzamil in rats exposed to elevated LAPs. Benzamil significantly inhibited the basal alveolar fluid reabsorption in the lungs of rats exposed to a moderate LAP (15 cmH₂O) and in control rat lungs (LAP 0 cmH₂O). To examine whether the alveolar fluid reabsorption could be restored after a transient increase in LAP, we studied a group of rats with a moderate LAP (15 cmH₂O) in the first hour and allowed the lungs to recover during the second and third hours (LAP 0 cmH₂O). As shown in Fig. 5, alveolar fluid clearance was restored to basal levels after the normalization of LAP, suggesting that alveolar fluid clearance inhibition was transient, reversible, and directly related to high pulmonary circulation pressures. These data suggest that transport proteins such as Na⁺-K⁺-ATPase that are responsible for active Na⁺ transport and alveolar fluid reabsorption are not irreversibly damaged or removed from the cells. These data, along with other reports (1, 22), provide a basis for use of agents such as catecholamines to recruit more Na/K pump molecules to the plasma membrane of the alveolar epithelium during periods of increased LAP. A possible explanation for this phenomenon is the reversible and transient stretching of pores across the alveolocapillary barrier and/or transient blebbing of the alveolocapillary barrier (3, 4). This could account for the reversible nature of this mild injury. In contrast, there have been reports (3, 4) suggesting that high microvascular pressures are associated with endothelial and epithelial breaks. This type of alveolocapillary damage is inconsistent with the recovery data in our physiological model and morphological tissue analysis and is possibly related to the time and magnitude of increased pressures across the pulmonary circulation.

With electron microscopy analysis, we observed alveolocapillary blebbing in lungs exposed to increased LAP that were similar to those observed by Bachofen et al. (3, 4) (Fig. 6). This was associated with decreased active Na⁺ transport and alveolar edema formation in lungs exposed to high LAP. In the endothelial cell layer, the blebbing probably occurred by the opening of the relatively weak intercellular junctions. On pressure release, endothelial cells exhibited a high repair capacity (3, 4), accounting for the repair process and providing the explanation of the recovery from increased LAP on the alveolocapillary barrier. In view of the decreased active Na⁺ transport and alveolar edema formation in the lungs exposed to a high LAP (20 cmH₂O), we had expected to observe more changes in the alveolocapillary barrier. However, it is also possible that our fixation methods were not adequate for the visualization of the epithelial lesions. Finally, a limitation to the present experimental design was that the size of the peribronchial cuff was not examined to provide an assessment as to the degree of interstitial edema.

In summary, we report here that alveolar fluid reabsorption decreases in the presence of mild to moderate left atrial hypertension in association with morphological changes across the alveolocapillary barrier. However, these effects are reversible, and the rate of alveolar fluid reabsorption is rapidly restored to basal levels after normalization of LAP.