Permeability of parietal pleura to liquid and proteins

Pleural function and lymphatics

The pleural space plays an important role in respiratory function as the negative intrapleural pressure regimen ensures lung expansion and in the mean time maintains the tight mechanical coupling between the lung and the chest wall. The efficiency of the lung-chest wall coupling depends upon pleural liquid volume, which in turn reflects the balance between the filtration of fluid into and its egress out of the cavity. While filtration occurs through a single mechanism passively driving fluid from the interstitium of the parietal pleura into the cavity, several mechanisms may co-operate to remove pleural fluid. Among these, the pleural lymphatic system emerges as the most important one in quantitative terms and the only one able to cope with variable pleural fluid volume and drainage requirements. In this review, we present a detailed account of the actual knowledge on: (a) the complex morphology of the pleural lymphatic system, (b) the mechanism supporting pleural lymph formation and propulsion, (c) the dependence of pleural lymphatic function upon local tissue mechanics and (d) the effect of lymphatic inefficiency in the development of clinically severe pleural and, more in general, respiratory pathologies.

Isolation of interstitial fluid and demonstration of local proinflammatory cytokine production and increased absorptive gradient in chronic peritoneal dialysis

In peritoneal dialysis (PD) patients, the frequent exposure to "unphysiological" dialysis fluids elicits a chronic state of a low-grade peritoneal inflammation leading to interstitial matrix remodeling and angiogenesis. Proinflammatory cytokines are important regulators involved in this inflammatory process that ultimately leads to dysfunction of the peritoneum as a dialysis membrane. We aimed to measure the local concentrations of proinflammatory cytokines in the peritoneal interstitial fluid (IF). Furthermore, we wanted to assess how the driving forces for fluid and solute exchanges are affected in a remodeled interstitial matrix and thus measured the colloid osmotic pressure (COP) gradient in rats that were exposed to chronic PD. After 8 wk of peritoneal dialysis, IF from peritoneum was isolated using a centrifugation method, and was analyzed for cytokine content and COP along with plasma. For several of the proinflammatory cytokines there were gradients from IF to plasma, showing local production. For some cytokines, the concentration in IF was increased severalfold, whereas IL-18 was increased systemically due to PD. Furthermore, the presence of the catheter per se seemed to increase cytokine levels. COP in IF was significantly decreased in the PD group, while collagen and hyaluronan content was increased. Collectively, our data suggest that the increased levels of proinflammatory cytokines after PD may be an integral component of the development of fibrosis and angiogenesis commonly seen in PD patients, and the decreased COP in IF after chronic PD may shift the Starling equilibrium across peritoneal capillaries to an absorptive state.

Pleural liquid and its exchanges

After an account on morphological features of visceral and parietal pleura, mechanical coupling between lung and chest wall is outlined. Volume of pleural liquid is considered along with its thickness in various regions, and its composition. Pleural liquid pressure (P(liq)) and pressure exerted by lung recoil in various species and postures are then compared, and the vertical gradient of P(liq) considered. Implications of lower P(liq) in the lung zone than in the costo-phrenic sinus at iso-height are pointed out. Mesothelial permeability to H(2)O, Cl(-), Na(+), mannitol, sucrose, inulin, albumin, and various size dextrans is provided, along with paracellular "pore" radius of mesothelium. Pleural liquid is produced by filtration from parietal pleura capillaries according to Starling forces. It is removed by absorption in visceral pleura capillaries according to Starling forces (at least in some species), lymphatic drainage through stomata of parietal mesothelium (essential to remove cells, particles, and large macromolecules), solute-coupled liquid absorption, and transcytosis through mesothelium.

Transport properties of the mesothelium and interstitium measured in rabbit pericardium

The contribution of the pleural mesothelium to pleural liquid and protein transport is still vigorously debated. Recent in vitro studies of stripped pleural membrane and free-standing pericardium have demonstrated active ion solute coupled transport of liquid and transcytosis of protein. However, the relative contribution of the passive transport properties of the pleural mesothelium compared to the pleural interstitium has not been extensively studied. In in vitro studies, we measured the albumin diffusion coefficient, reflection coefficient, hydraulic conductivity and electrical resistance of rabbit pericardium. We used two techniques, treatment with 40 muM nocodazole and a 1-min hypotonic cell lysis with distilled water, to eliminate the effect of the two mesothelial layers on diffusional and hydraulic resistances. Each technique increased the albumin diffusion coefficient and hydraulic conductivity 3- to 4-fold. In hydraulic conductivity experiments using tracer 125I-albumin, nocodazole reduced the reflection coefficient to zero, rendering the pericardium completely permeable to albumin. We applied the cell-lysis technique to the pleural and pericardial mesothelium in sequence to evaluate the separate contribution of each mesothelium. Both diffusional and hydraulic resistances, but not electrical resistance, of the mesothelium were overestimated by the cell-lysis technique. The pleural mesothelium contributed at most 30% of diffusional resistance, 10% of hydraulic resistance and 14% of electrical resistance of the total pericardial resistances. We conclude that the pleural mesothelium is not the primary barrier to protein diffusion or bulk flow of liquid from the pericardial microcirculation to the pleural liquid.

Contribution of lymphatic drainage through stomata to albumin removal from pleural space

The contribution of lymphatic drainage through the stomata of parietal mesothelium to the overall removal of labeled albumin from the pleural space was found 89% in sheep with very large hydrothoraces (10 ml/kg), a condition involving a approximately 20 times increase in lymphatic drainage [Broaddus et al., J. Appl. Physiol. 64 (1988) 384]. We determined this contribution in anesthetized rabbits with small (0.12 ml/kg) and large (2.4 ml/kg) hydrothoraces of Ringer-albumin with labeled albumin and labeled dextran-2000 kDa. This dextran was used as marker of liquid removal through the stomata because it should essentially leave the pleural space through the stomata only, owing to its size. The removal of labeled albumin by lymphatic drainage through the stomata was 39% of the overall removal in the small hydrothoraces, and 64% in the large ones. Hence, lymphatic drainage through the stomata does not contribute most of protein and liquid removal from the pleural space under physiological conditions, as it has been maintained. It markedly increases with the increase in pleural liquid volume.

Pleural mechanics and fluid exchange

The pleural space separating the lung and chest wall of mammals contains a small amount of liquid that lubricates the pleural surfaces during breathing. Recent studies have pointed to a conceptual understanding of the pleural space that is different from the one advocated some 30 years ago in this journal. The fundamental concept is that pleural surface pressure, the result of the opposing recoils of the lung and chest wall, is the major determinant of the pressure in the pleural liquid. Pleural liquid is not in hydrostatic equilibrium because the vertical gradient in pleural liquid pressure, determined by the vertical gradient in pleural surface pressure, does not equal the hydrostatic gradient. As a result, a viscous flow of pleural liquid occurs in the pleural space. Ventilatory and cardiogenic motions serve to redistribute pleural liquid and minimize contact between the pleural surfaces. Pleural liquid is a microvascular filtrate from parietal pleural capillaries in the chest wall. Homeostasis in pleural liquid volume is achieved by an adjustment of the pleural liquid thickness to the filtration rate that is matched by an outflow via lymphatic stomata.

Effect of concentration on restriction and diffusion of albumin in the excised rat diaphragm

In tissue samples of rat diaphragm mounted between two chambers, we measured the flow of albumin solution (0-5 g/dl) containing radioactive tracer (125)I-albumin in response to a driving pressure of 20 cmH(2)O. The ratio of the albumin concentration of the output solution to that of the input (sieving ratio, C(out)/C(in)) was measured from solution radioactivity. C(out)/C(in) increased monotonically from 0.5 with the flow of approximately 0 g/dl albumin solution (tracer) to 0.9 with the flow of 5 g/dl albumin solution. We modeled the tissue as a membrane subjected to flows of high Peclet No. with a reflection coefficient sigma = 1 - C(out)/C(in). Values of sigma decreased from 0.5 with Ringer solution to 0.1 with 5 g/dl albumin solution. Hydraulic conductivity measured with the flow of Ringer solution increased with the flow of 5 g/dl albumin solution. Wet-to-dry weight ratio and radioactivity of tissue samples immersed in 0.01-5 g/dl albumin solutions indicated a 40% increase in tissue water, associated with an albumin volume fraction of 0.3 measured at 0.5-2 h. The slower rate of albumin uptake occurring up to 20-30 h indicated intracellular diffusion that was equal with 1 and 5 g/dl albumin solution but reduced with a 0.01 g/dl albumin solution. The results suggest that interstitial pores increase in size in response to an increase in albumin concentration. We postulate a two-pore model made of intracellular pores that coalesce into a set of larger pores by osmotic flow.

Albumin transcytosis from the pleural space

Occurrence of transcytosis in pleural mesothelium was verified by measuring removal of labeled macromolecules from pleural liquid in experiments without and with nocodazole. To this end, we injected 0.3 ml of Ringer-albumin with 750 microg of albumin-Texas red or with 600 microg of dextran 70-Texas red in the right pleural space of anesthetized rabbits, and after 3 h we measured pleural liquid volume, labeled macromolecule concentration, and, hence, labeled macromolecule quantity in the liquid of this space. Labeled albumin left was 318 +/- 28 microg in control and 419 +/- 17 microg in nocodazole experiments (means +/- SE); hence, whereas ventilation was similar its removal was greater (P < 0.01) in control experiments. Labeled dextran left was 283 +/- 10 microg in control and 381 +/- 21 microg in nocodazole experiments; hence, whereas ventilation was similar its removal was greater (P < 0.01) in control experiments. These findings indicate occurrence of transcytosis from the pleural space. Liquid removed by transcytosis was 0.05 ml/h. This amount times unlabeled albumin concentration under physiological conditions (10 mg/ml) times lumen-vesicle partition coefficient for albumin (0.78) provides fluid-phase albumin transcytosis: approximately 203 microg. h(-1) kg(-2/3). Transcytosis might contribute a relevant part of protein and liquid removal from the pleural space.

Effect of concentration and hydration on restriction of albumin by lung interstitium

In previous studies, the flow of albumin solution through hydrated lung interstitial segments was higher than a prior flow of Ringer solution (A. Tajaddinni et al., 1994, J. Appl. Physiol. 76, 578-583). We wondered whether this effect was caused by an increased pore size. We measured the flow of albumin solutions through interstitial segments subjected to a driving pressure of 5 cm H(2)O and various mean interstitial pressures (P(if)). The ratio of albumin concentration (C(alb)) of the output solution to that of the input solution (C(out)/C(in), sieving ratio) was measured using tracer (125)I-albumin. At normal hydration (0 cm H(2)O P(if)), C(out)/C(in) was minimal (0.6) with the flow of Ringer solution, increased to 0.8 with the flow of 5 g/dl albumin solution, and increased to 1 with increased hydration at 15 cm H(2)O P(if). We modeled the interstitium as a membrane subjected to flows of high Peclet numbers. Accordingly, the albumin reflection coefficient [sigma = 1 - (C(out)/C(in))] at 0 cm H(2)O P(if) was 0.4 with the flow of Ringer solution and decreased to 0 at 5 g/dl C(alb) and 15 cm H(2)O P(if). This behavior suggests that the flow of albumin occurred through interstitial pores that increased in size as either C(alb) or hydration increased. We conceive of an interstitium that consists of pores with permeable moveable walls across which osmotic interaction occurs between the pore liquid and the surrounding tissue.

Steady-state pleural fluid flow and pressure and the effects of lung buoyancy

Both theoretical and experimental studies of pleural fluid dynamics and lung buoyancy during steady-state, apneic conditions are presented. The theory shows that steady-state, top-to-bottom pleural-liquid flow creates a pressure distribution that opposes lung buoyancy. These two forces may balance, permitting dynamic lung floating, but when they do not, pleural-pleural contact is required. The animal experiments examine pleural-liquid pressure distributions in response to simulated reduced gravity, achieved by lung inflation with perfluorocarbon liquid as compared to air. The resulting decrease in lung buoyancy modifies the force balance in the pleural fluid, which is reflected in its vertical pressure gradient. The data and model show that the decrease in buoyancy with perfluorocarbon inflation causes the vertical pressure gradient to approach hydrostatic. In the microgravity analogue, the pleural pressures would be toward a more uniform distribution, consistent with ventilation studies during space flight. The pleural liquid turnover predicted by the model is computed and found to be comparable to experimental values from the literature. The model provides the flow field, which can be used to develop a full transport theory for molecular and cellular constituents that are found in pleural fluid.

Macromolecule transfer through mesothelium and connective tissue

Diffusional permeability (P) to inulin (P(in)), albumin (P(alb)), and dextrans [70 (P(dx 70)), 150 (P(dx 150)), 550 (P(dx 550)), and 2, 000 (P(dx 2,000))] was determined in specimens of parietal pericardium of rabbits, which may be obtained with less damage than pleura. P(in), P(alb), P(dx 70), P(dx 150), P(dx 550), and P(dx 2, 000) were 0.51 +/- 0.06 (SE), 0.18 +/- 0.03, 0.097 +/- 0.021, 0. 047 +/- 0.011, 0.025 +/- 0.004, and 0.021 +/- 0.005 x 10(-5) cm/s, respectively. P(in), P(alb), and P(dx 70) of connective tissue, obtained after removal of mesothelium from specimens, were 10.3 +/- 1.42, 2.97 +/- 0.38, and 2.31 +/- 0.16 x 10(-5) cm/s, respectively. Hence, P(in), P(alb), and P(dx 70) of mesothelium were 0.54, 0.20, and 0.10 x 10(-5) cm/s, respectively. Inulin (like small solutes) fitted the relationship P-solute radius for restricted diffusion with a 6-nm "pore" radius, whereas macromolecules were much above it. Hence, macromolecule transfer mainly occurs through "large pores" and/or transcytosis. In line with this, the addition of phospholipids on the luminal side (which decreases pore radius to approximately 1.5 nm) halved P(in) but did not change P(alb) and P(dx 70). P(in) is roughly similar in mesothelium and capillary endothelium, whereas P to macromolecules is greater in mesothelium. The albumin diffusion coefficient through connective tissue was 17% of that in water. Mesothelium provides 92% of resistance to albumin diffusion through the pericardium.

Hydraulic conductivity, albumin reflection and diffusion coefficients of pig mediastinal pleura

Hydraulic conductivity (L), albumin reflection coefficient (sigma), and albumin diffusion coefficient (D) were measured across pig mediastinal pleura. The tissue (7 mm diameter) was bonded between two chambers. Flow (Q) of lactated Ringer solution between the chambers was measured in turn at driving pressures (DeltaP) of 2, 4, and 6 cm H(2)O. Value of L was proportional to the slope of the Q-DeltaP curve. Then Q was measured in turn at three albumin osmotic pressure differences (Deltapi equivalent to -1, -2, and -3 g/dl albumin concentration difference, DeltaC) with DeltaP constant at either 2, 3, 4, or 6 cm H(2)O. From Starling's equation, magnitude of sigma was the slope of the Q-Deltapi curve divided by the slope of the Q-DeltaP curve. We measured the diffusion of 0, 2, 5, and 10 g/dl albumin with tracer (125)I-albumin. Tracer mass (M) that diffused across the pleura was measured for 10 h using a well-type NaI(T1) detector. D was calculated from the slope of the M-time curve. Values of L averaged 2.0 x 10(-8) cm(3). s(-1). dyne(-1) (n = 23). Values of sigma were small (0.02-0.05) and sigma increased as flow increased 20-fold. D (n = 24) increased 3-fold from 2.7 x 10(-8) cm(2)/s as DeltaC increased from 0 to 10 g/dl. The small values of sigma indicated that mediastinal pleura provided little restriction to the passage of protein.

Mechanical coupling and liquid exchanges in the pleural space

The pleural space provides the mechanical coupling between lung and chest wall: two views about this coupling are reported and discussed. Information on volume, composition, thickness, and pressure of the pleural liquid under physiologic conditions in a few species is provided. The Starling pressures of the parietal pleura filtering liquid into pleural space, and those of the visceral pleura absorbing liquid from the space are considered along with the permeability of the mesothelium. Information on the lymphatic drainage through the parietal pleura and on the solute-coupled liquid absorption from the pleural space under physiologic conditions and with various kinds of hydrothorax are provided.

Clara cell protein (CC16) in pleural fluids: a marker of leakage through the visceral pleura

Pleural fluid (PF) proteins either derive from serum by diffusion or are locally secreted within the pleural space. Another hypothetical origin is a leakage of lung secretory proteins across the visceral pleura. To test this hypothesis, we investigated the occurrence, sources, and determinants in PF of CC16, a small-size and readily diffusible protein of 16 kDa secreted by bronchiolar Clara cells. CC16 concentration was determined by a sensitive latex immunoassay in serum and PF of 117 subjects (86 exudates and 31 transudates) and, for purpose of comparison, in ascites samples from another group of 38 subjects (7 exudates and 31 transudates). CC16 was also studied in serum and PF of normal rats and in rats with pleural exudate induced by alpha-naphthyl-thiourea (ANTU). The levels of CC16 in PF and ascites were highly correlated with that in serum, suggesting a diffusional exchange across the pleural/blood and peritoneal/blood barriers. Whereas CC16 occurs at similar levels in ascites and serum, the protein was found to be more concentrated in PF than in serum in both humans (geometric mean in microg/L, 26.2 versus 14.6, p < 0.0001) and rats (213 versus 16.2, p < 0.001). A local synthesis of CC16 appeared unlikely in view of the lack of CC16-immunostaining in pleura of both species. The only plausible explanation for these findings is that CC16 in PF originates from two sources: diffusion from plasma and a leakage from the lung into the pleural space across the semipermeable visceral pleura. This interpretation is supported by a markedly increased leakage of CC16 in experimental exudates induced by ANTU and the finding of high CC16 concentrations in human transudates associated with congestive heart failure, two conditions wherein PF has been shown to arise from the interstitial spaces of the lung.

Cytokines and soluble cytokine receptors in pleural effusions from septic and nonseptic patients

The balance between proinflammatory cytokines and their inhibitors has rarely been investigated in pleural effusions of nonmalignant or noninfectious origin. To evaluate the impact of a lung and/or intrathoracic infection in such a circumstance, we compared the levels of proinflammatory cytokines (interleukin-8 [IL-8]); tumor necrosis factor-alpha (TNF-alpha); the cytokine antagonists and inhibitors (IL-1 receptor antagonist [IL-1ra]) and soluble TNF receptors Types I and II (sTNFRI, sTNFRII); and antiinflammatory cytokines (transforming growth factor-beta [TGF-beta]) in pleural effusion and plasma from septic (n = 15) and nonseptic (n = 9) patients. In addition, we analyzed the levels of IL-6 and its soluble receptor (sIL-6R). Bronchoalveolar lavage fluids (BALFs) were also studied in a few septic patients. High and nonsignificantly different levels of cytokines and inhibitors were detected in both groups of patients. The levels of IL-6 and sTNFRI and sTNFRII in pleural effusion were higher than in plasma, whereas the levels of IL-1ra and sIL-6R were higher in plasma. The levels of sIL-6R influenced the bioactivity of IL-6. There was no correlation between the levels of cytokines in plasma and in pleural effusion. In contrast, a significant correlation was observed for the soluble receptors sIL-6R (r = 0.67, p < 0.001), sTNFRI (r = 0.76, p < 0.001) and sTNFRII (r = 0.66, p = 0.001). Furthermore, a high correlation was found between the levels of both forms of sTNFRs in plasma (r = 0.95, p < 0.001) and in pleural effusion (r = 0.79, p < 0.001). In addition, a correlation was observed between the levels of TGF-beta in pleural effusion and in BALF. The highest levels of some markers in plasma and of others in pleura argue in favor of both a systemic and a compartmentalized response, independently of the presence of infection. Because cytokines can be trapped by the surrounding cells in their environment, measurable levels of cytokines in biologic fluids represent the "tip of the iceberg," which is not the case for soluble receptors. The correlations of these latter markers between plasma and pleura strongly suggest that exchanges between both compartments can occur in both directions.