ANNUAL REVIEWS
Annu. Rev. Med. 1990.41:7-/3 Copyright © 1990 by Annual Reviews Inc. All rights reserved
Further
Quick links to online content
THE PATHOPHYSIOLOGY OF
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
PLEURAL EFFUSIONS Steven A. Sahn, M.D. Department of Medicine, Division of Pulmonary and Critical Care Medicine, Medical University of South Carolina, Charleston, South Carolina 29425 KEY
WORDS:
parietal pleura, visceral pleura, stomata, lymphatics.
ABSTRACT
Two features of human parietal pleura explain its role in the formation and removal of pleural liquid and protein in the normal state: the proximity of the microvessels to the pleural surface and the presence of stomata situated between mesothelial cells. For pleural fluid to accumulate in disease, there must be increased production from increased hydrostatic pressure, decreased oncotic or pleural pressure, increased microvascular permeability, or peritoneal-pleural movement. The rate of formation must overwhelm lymphatic clearance, which may be decreased by hydrostatic forces or blocked by malignant infiltration. INTRODUCTION The pleural space is real, approximately 10 to 20 !lm wide, and encompasses the area between the mesothelium of the parietal and visceral pleurae (1 ). There are important distinguishing features between the two pleural surfaces. Only the parietal pleura has stomata, 2-12-!lm openings situated between mesothelial cells (2, 3). Stomata are the usual exit points for pleural liquid, protein, and cells that are removed from the pleural space (3, 4). These stomata communicate directly with lymphatic lacunae, the roofs of which contain bundles of collagen. The lymphatic lacunae con verge into collecting lymphatics that empty into lymphatic channels that course along the ribs and drain into the mediastinal lymph nodes. The 7 0066-4219/90/0401-0007$02.00
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
8
SAHN
microvessels of the parietal pleura are closer to the pleural surface (approximately 10-12 ,urn) than are the microvessels of the visceral pleura, which are 18 to 56 ,urn from the pleural space (5); this difference is directly related to pleural liquid formation in normal man. The parietal pleura tends to be uniform in thickness between species, in contrast to the visceral pleura, the thickness of which is extremely variable. Humans have intermediate visceral pleural thickness; sheep and cattle have thick visceral pleurae, while the dog, cat, and rabbit have thin visceral pleurae. Visceral pleural thickness is related to its inherent blood supply. Species with thick and intermediate visceral pleurae are predominantly supplied by the bronchial circulation, while species with thin visceral pleurae are supplied by the pulmonary circulation (6, 7). Thus, in humans the entire visceral pleura is supplied by the bronchial circulation (5, 7-9). The terminal arteries and arterioles end in a capillary network much larger than those of alveolar capillaries (10). The venous return from the subvisceral pleural capillaries drains largely into the pulmonary veins (9, 11). The human parietal pleura is supplied by branches of the arteries that flow to the adjacent chest wall; the venous system of the parietal pleura drains into the bronchial veins (12). PHYSIOLOGY OF PLEURAL FLUID FORMATION AND REMOVAL IN NORMAL MAN Influx and exit of liquid and protein of the pleural space are balanced so that a constant volume and protein concentration exist in the normal state. With changes in microvascular hydrostatic or oncotic pressure, pleural pressure, and microvascular permeability, or with impaired lymphatic drainage, this delicate balance is disturbed: fluid accumulates and protein concentration changes. Several misconceptions have led to erroneous con clusions concerning normal pleural fluid formation and removal. It was thought that the arterial blood supply to the visceral pleura in humans was from the pulmonary circulation, that the absorption of protein-free liquid occurred across the visceral pleura, and that there was a high rate of pleural liquid and protein turnover (5, II). Rather, pleural fluid should be considered as interstitial liquid of the parietal pleura (5). Since the parietal pleura is supplied by the systemic circulation and the pressure of the pleural space is subatmospheric and probably less than the pressure in the interstitium of the subpleural space, there is a pressure gradient from the pleural interstitium to the pleural space (13). Extrapolating from the peritoneum, it appears that the microvascular endothelium is the only important barrier to solute and water exchange in the pleural space as well. The contribution to pleural liquid and protein formation from the
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
PLEURAL EFFUSIONS
9
visceral pleura in normal man is probably minimal because the distance between the microvessels and the mesothelium is relatively large and bccause of thc lower filtration pressure in the visceral pleural micro circulation as bronchial venules empty into the pulmonary veins with their lower pressure. Thus, even though liquid moves into the visceral pleural interstitium, most of the interstitial liquid probably is reabsorbed in the normal state before it can move a great distance along the pressure gradient. Investigation in several species demonstrates that the volume of liquid in the normal pleura is small and approximates 0.1 to 0.2 mljkg body weight in both pleural spaces (5,14-16). Protein concentration of normal pleural fluid is consistent with the information known about interstitial liquid. The initial filtrate from the systemic circulation has a protein con centration of approximately 0.3 to 0.4 gjdl (17). However, the majority of the filtered liquid probably is absorbed rapidly into the lower pressure venules, which thus concentrates the interstitial fluid protein and produces the normal pleural fluid protein of 1.0 to 1.5 gjd!' Studies of equilibration of radiolabeled albumin from the intravascular to the pleural space in awake sheep provide evidence for a low filtration rate. An entry rate of 0.01 mlj(kg x hr) or 7 mljday in a 30-kg sheep has been noted (18). This is in marked contrast to previous work, which found entry rates of the magnitude of 0.47 mlj(kg x hr) or approximately 225 mljday in a 20-kg dog (16). The high entry rate documented in prior experiments probably can be explained by pleural catheter inflammation (19). Pleural liquid and protein have been shown to leave the pleural space of sheep at approximately 0.28 mlj(kg x hr) following creation of a sterile hydrothorax at a protein concentration of I gjdl (20). This same study also demonstrated the capacity to increase pleural lymphatic flow at least 20-fold over baseline. These data show the reserve of pleural lymphatics to be large, and for a pleural effusion to form, fluid formation rate must be very high or lymphatic flow impaired. Once excess fluid formation ceases, normal lymphatics probably can remove close to 500 ml of transu dative fluid per day in a 70-kg man. Thus, it appears that pleural liquid is formed and removed slowly and has a lower protein concentration than both lung and peripheral lymph. The parietal pleura appears to be the "business end" of liquid and protein exchange in the pleural space. A low protein filtrate from the systemic vessels enters the parietal pleural interstitial space, is concentrated, and subsequently leaks through the mesothelium. Both pleural liquid and protein exit by the parietal pleural stomata. Since lymphatic drainage of the pleural space has a large reserve, accumulation of an abnormal amount of pleural fluid represents markedly increased fluid formation, decreased
10
SAHN
clearance, or blockage of the clearance pathways, or a combination of the three mechanisms. MECHANISMS OF PLEURAL FLUID
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
ACCUMULATION IN DISEASE An increased rate of formation of pleural fluid can occur with increase in microvascular hydrostatic pressure, decrease in microvascular oncotic pressure, decrease in pressure in the pleural space, and increase in micro vascular permeability; movement of fiuid from the peritoneal to the pleural space is also a cause of pleural fluid accumulation (21). A decreased clearance rate can be found with systemic venous hypertension (22), and blockage of clearance pathways has been documented with malignant involvement of the pleural lymphatics (23). An acute experiment in dogs indicated that systemic venous pressure is the most important determinant of hydrostatic pleural fluid formation and volume (24). However, recent clinical data suggest that increased pulmonary venous pressure is the most important factor in the develop ment of pleural effusions in congestive heart failure (25, 26). It is probable that the fluid leaks into the interstitium of the lung and moves across the visceral. mesothelium along an interstitial-pleural pressure gradient. Preliminary observations from our laboratory have noted that the greatest incidence of pleural effusions is seen in patients with congestive heart failure and both chronic systemic and pulmonary venous hypertension. A decrease in oncotic pressure in the microvascular circulation will increase the tendency to form pleural interstitial liquid. This will result in the increased entry of low protein fluid into the pleural space from the parietal pleural surface. The clinical observation that this mechanism is an unusual cause of a large pleural effusion probably can be explained by the functional reserve of the parietal pleural lymphatic system. However, if the visceral pleura also is involved in pleural leak or if lymphatic drainage is impaired, larger volumes of fluid can accumulate. When there is a decrease in pressure in the pleural space as in atelectasis, the pressure gradient from the pleural interstitium to the pleural space increases and favors the formation of pleural liquid. Furthermore, the separation of the lung from the chest wall could decrease pleural space fluid movement during respiration and inhibit optimal parietal pleural lymphatic drainage. Clinically, moderate to large pleural effusions occur only with complete lung collapse; with minor atelectasis, once a given volume of fluid has formed, pleural pressure returns toward normal and the increased rate of pleural liquid formation ceases. Increased permeability of the microvascular circulation is the mech-
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
PLEURAL EFFUSIONS
11
anism of pleural fluid formation that occurs with inflammation of both pulmonary and pleural microvessels regardless of etiology. Liquid and protein leak across the lung and pleural microvessels at an increased rate. The pleural entry rate could overwhelm the pleural reabsorptive capacity; however, decreased clearance may also be implicated when the parietal pleural stomata are occluded by fibrin, debris, and mesothelial swelling secondary to the inflammatory process. It has been observed retrospectively that pleural fluid occurred in 36% of patients with ARDS compared to 40% with hydrostatic pulmonary edema (27). In a prospective study of eight patients with ARDS, all had pleural fluid documented by ultrasound and two autopsied patients had exudative effusions (28). Experimental studies also have documented pleural fluid associated with ARDS and support its origin from lung edema (29, 30). These effusions usually are small and may be unrecognized clinically as the patient has diffuse alveolar infiltrates and chest radiographs are obtained with the patient in the supine position, which negates the sign
of costophrenic angle blunting. The mechanism of pleural fluid formation in ARDS is movement of extravascular lung water along the interstitial pleural pressure gradient. The use of positive pressure ventilation in the treatment of ARDS would tend to inhibit pleural fluid accumulation by decreasing the interstitial-pleural gradient. A blockage at any juncture in the lymphatic drainage system from the stomata to the mediastinal lymph nodes, whether due to tumor, fibrosis, or lymphatic abnormalities, will result in pleural fluid accumulation. Fluid formation may occur slowly as in patients with yellow nail syndrome, who have a congenital lymphatic abnormality, or it may occur relatively rapidly in the patient with malignant involvement ofiymphatics, who, in addition, has pleural seeding with tumor and increased fluid formation from increased microvascular permeability (21). Large effusions from malig nancy will result only with significant involvement of the lymphatic system. Any fluid, whcther high or low in protein concentration, can movc from the peritoneal to pleural cavity through either diaphragmatic defects or diaphragmatic lymphatics. The diaphragmatic defects usually are small, being < 1 cm in diameter (31). Fluid moves from the peritoneal to pleural space because of the pressure gradient across the diaphragm. When an acute, massive pleural effusion develops with ascites, the portal of entry into the pleural space is usually a diaphragmatic defect resulting in rapid movement of fluid into the pleural space that exceeds the exit rate of the pleural lymphatics. It would appear that for a massive pleural effusion to evolve, a "ball-valve" mechanism must be operative for fluid accumulation would cease when pleural and peritoneal pressures equalize.
12
SAHN
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
Literature Cited 1. Agostoni, E., D'Angelo, E. 1969. Thick ness and pressure of the pleural liquid at various heights and with various hydrothoraces. Respir. Physiol. 6: 33042 2. Leak. L. Y., Rahil. L. 1978. Permeability of the diaphragmatic mesothelium: the ultrastructural basis for "stomata." Am. J. Anat. 151: 557-94 3. Wang, N. S. 1975. The preformed stomas connecting the pleural cavity and the lymphatics in the parietal pleura. Am. Rev. Respir. Dis. 111: 12-20 4. Courtice, F. C., Simmonds, W. 1. 19 54. Physiological significance of lymph drainage of the serous cavities and lungs. Physiol. Rev. 34: 419-48 5. Staub, N. C., Weiner-Kronish, 1. P., Albertine, K. H. 1985. Transport through the pleura. Physiology of nor mal liquid and solute exchange in the pleural space. In The Pleura in Health and Disease, ed. 1. Chretein, J. Bignon, A. Hirsch, pp. 169-93. New York: Mar cel Dekker 6. McLaughlin, R. F., Tyler, W. S.,
Canada, R. O. 1961. A study
of
the
subgross pulmonary anatomy in various animals. Am. J. Anat. 108: 149-65 7. McLaughlin, R. F. 1983. Bronchial artery distribution in various animals and humans. Am. Rev. Respir. Dis. 128: S57-S58 8. McLaughlin, R. F., Tyler, W. S., Canada, R. O. 1966. Subgross pul monary anatomy of the rabbit, rat, and guinea pig with additional notes on the human lung. Am. Rev. Respir. Dis. 94: 380-87 9. Naigaishi, C. 1972. Functional Anatomy and Histology of the Lung, pp. 79-179. Baltimore: Univ. Park 10. Krahl,Y.E.1964.Anatomy ofthe mam malian lung. In Handbook of Physiology and Respiration, ed. W. O. Fenn, H. Rahn, 1(3): 2\3-84. Washington, DC: Am. Physio!. Soc. 11. Albertine, K. H., Weiner-Kronish, J. P., Roos, P. J., Staub, N. C. 1982. Structure, blood supply and lymphatic vessels of the sheep's visceral pleura. Am. J. Anat. 165: 227-94 12. Testut, L. 1930. Traite d'Anatomie Humaine tom. III, p. 1003. Paris: Doin. 8th cd. 13. Bhattacharya, J., Gropper, M. A., Staub, N. C. 1984. Interstitial pleural pressure gradient measured by micro puncture in excised dog lung. J. Appl. Physiol. 56: 271-77 14. Miserocchi, G., Agostoni, E. 1971. Con-
15.
16.
17.
tents of the pleural space. J. Appl. Physiol. 30: 200-13 Sahn, S. A., Willcox, M. L., Good, J. T. Jr., Potts, D. E., Filley, G. F. 1979. Characteristics of normal rabbit pleural fluid: physiologic and biochemical impli cations. Lung 156: 63-69 Stewart, P. B., Burgen, A. S. Y. 1958. The turnover of fluid in the dog's pleural cavity. J. Lab. Clin. Invest. 52: 21230 Landis, E. M., Pappenheimer, J. R. 1963. Exchange of substances through the capillary walls. In Handbook of Physiology, Sect. 2, 2: 961 1034. Wash ington, DC: Am. Physio!. Soc. Weiner-Kronish, J. P., Albertine, K. H., Licko, Y., Staub, N. C. 1984. Protein egress and entry rates in pleural fluid and plasma in sheep. J. Appl. Physiol. 56: 459-63 Weiner-Kronish, J. P.,Albertine, K. H., Roos, P. J., Staub, N. C. 1982. Pleural fluid dynamics in sheep are altered by a pleural catheter. Fed. Proc. 41: 1127 Broaddus, V. c., Wiener-Kronish, J. P., -
18.
19.
20.
Berthiaume, Y., Staub, N. C. 1988.
Removal of pleural liquid and protein by lymphatics in a wake sheep. J. App/. Physiol. 64: 384-90 21. Sahn, S. A. 1988. The state of the art. Thc pleura. Am. Rev. Respir. Div. 138: 184-234 22. Broaddus, V. c., Araya, M., Staub, N. C. 1989. Evidence of pleural lymphatic absorption ceases with elevation of sys temic venous pressure in volume-loaded rabbits. Am. Rev. Respir. Dis. 139: A413 23. Meyer, P. C. 1966. Metastatic carcinoma of the pleura. Thorax 21: 437-43 24. Mellins, R. B., Levine , O. R., Fishman, A. P. 1970. Effect of systemic and pul monary venous hypertension on pleural and pericardial fluid accumulation. J. Appl. Physiol. 29: 564-69 25. Wiener-Kronish, J. P., Matthay, M. A., Callen, P. W., Filly, R. A., Gamsu, G., et al. 1985. Relationship of pleural effusions to pulmonary hemodynamics in patients with congestive heart failure. Am. Rev. Respir. Dis. 132: 1253-56 26. Weiner-Kronish. J. P.. Goldstein, R., Matthay, M. A., Biondi, J. W., Broaddus, V. c., et al. 1987. Lack of association of pleural effusion with chronic pulmonary arterial and right atrial hypertension. Chesl 92: 967-70 27. Aberle, D. R., Wiener-Kronish, J. P., Webb, W. R., Matthay, M. A. 1987. Diagnosis of hydrostatic versus in creased permeability pulmonary edema
PLEURAL EFFUSIONS
on the basis of chest radiographic criteria in critically ill patients. Am. Rev. Respir. Dis. 135: A215 28. Wiener-Kronish, 1. P., Goldstein, R., Matthay, M. A. 1988. Pleural effusions are frequently associated with the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 137: 227 29. Wiener-Kronish, 1. P., Broaddus, V. C., Albertine, M. A., Gropper, M. A., Mat thay, M. A., et al. 1988. Relationship of
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
pleural
effusions to increased perme-
13
ability pulmonary edema in anesthetized sheep. 1. CUn. Invest. 82: 1422-29
30. Miller, K. S., Harley, R. A., Sahn, S. A. 1989. Pleural effusions associated with ethchlorvynol lung injury result from
visceral pleural leak. Am. Rev . Respir. Dis. 140: 764-70 31. Lieberman, F. L., Hidemura, R., Peters, R. L., Reynolds, T. B. 1966. Patho genesis and treatment of hydrothorax complicating cirrhosis with ascites. Ann. Intern. Med. 64: 341-51
Annu. Rev. Med. 1990.41:7-/3 Copyright © 1990 by Annual Reviews Inc. All rights reserved
Further
Quick links to online content
THE PATHOPHYSIOLOGY OF
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
PLEURAL EFFUSIONS Steven A. Sahn, M.D. Department of Medicine, Division of Pulmonary and Critical Care Medicine, Medical University of South Carolina, Charleston, South Carolina 29425 KEY
WORDS:
parietal pleura, visceral pleura, stomata, lymphatics.
ABSTRACT
Two features of human parietal pleura explain its role in the formation and removal of pleural liquid and protein in the normal state: the proximity of the microvessels to the pleural surface and the presence of stomata situated between mesothelial cells. For pleural fluid to accumulate in disease, there must be increased production from increased hydrostatic pressure, decreased oncotic or pleural pressure, increased microvascular permeability, or peritoneal-pleural movement. The rate of formation must overwhelm lymphatic clearance, which may be decreased by hydrostatic forces or blocked by malignant infiltration. INTRODUCTION The pleural space is real, approximately 10 to 20 !lm wide, and encompasses the area between the mesothelium of the parietal and visceral pleurae (1 ). There are important distinguishing features between the two pleural surfaces. Only the parietal pleura has stomata, 2-12-!lm openings situated between mesothelial cells (2, 3). Stomata are the usual exit points for pleural liquid, protein, and cells that are removed from the pleural space (3, 4). These stomata communicate directly with lymphatic lacunae, the roofs of which contain bundles of collagen. The lymphatic lacunae con verge into collecting lymphatics that empty into lymphatic channels that course along the ribs and drain into the mediastinal lymph nodes. The 7 0066-4219/90/0401-0007$02.00
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
8
SAHN
microvessels of the parietal pleura are closer to the pleural surface (approximately 10-12 ,urn) than are the microvessels of the visceral pleura, which are 18 to 56 ,urn from the pleural space (5); this difference is directly related to pleural liquid formation in normal man. The parietal pleura tends to be uniform in thickness between species, in contrast to the visceral pleura, the thickness of which is extremely variable. Humans have intermediate visceral pleural thickness; sheep and cattle have thick visceral pleurae, while the dog, cat, and rabbit have thin visceral pleurae. Visceral pleural thickness is related to its inherent blood supply. Species with thick and intermediate visceral pleurae are predominantly supplied by the bronchial circulation, while species with thin visceral pleurae are supplied by the pulmonary circulation (6, 7). Thus, in humans the entire visceral pleura is supplied by the bronchial circulation (5, 7-9). The terminal arteries and arterioles end in a capillary network much larger than those of alveolar capillaries (10). The venous return from the subvisceral pleural capillaries drains largely into the pulmonary veins (9, 11). The human parietal pleura is supplied by branches of the arteries that flow to the adjacent chest wall; the venous system of the parietal pleura drains into the bronchial veins (12). PHYSIOLOGY OF PLEURAL FLUID FORMATION AND REMOVAL IN NORMAL MAN Influx and exit of liquid and protein of the pleural space are balanced so that a constant volume and protein concentration exist in the normal state. With changes in microvascular hydrostatic or oncotic pressure, pleural pressure, and microvascular permeability, or with impaired lymphatic drainage, this delicate balance is disturbed: fluid accumulates and protein concentration changes. Several misconceptions have led to erroneous con clusions concerning normal pleural fluid formation and removal. It was thought that the arterial blood supply to the visceral pleura in humans was from the pulmonary circulation, that the absorption of protein-free liquid occurred across the visceral pleura, and that there was a high rate of pleural liquid and protein turnover (5, II). Rather, pleural fluid should be considered as interstitial liquid of the parietal pleura (5). Since the parietal pleura is supplied by the systemic circulation and the pressure of the pleural space is subatmospheric and probably less than the pressure in the interstitium of the subpleural space, there is a pressure gradient from the pleural interstitium to the pleural space (13). Extrapolating from the peritoneum, it appears that the microvascular endothelium is the only important barrier to solute and water exchange in the pleural space as well. The contribution to pleural liquid and protein formation from the
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
PLEURAL EFFUSIONS
9
visceral pleura in normal man is probably minimal because the distance between the microvessels and the mesothelium is relatively large and bccause of thc lower filtration pressure in the visceral pleural micro circulation as bronchial venules empty into the pulmonary veins with their lower pressure. Thus, even though liquid moves into the visceral pleural interstitium, most of the interstitial liquid probably is reabsorbed in the normal state before it can move a great distance along the pressure gradient. Investigation in several species demonstrates that the volume of liquid in the normal pleura is small and approximates 0.1 to 0.2 mljkg body weight in both pleural spaces (5,14-16). Protein concentration of normal pleural fluid is consistent with the information known about interstitial liquid. The initial filtrate from the systemic circulation has a protein con centration of approximately 0.3 to 0.4 gjdl (17). However, the majority of the filtered liquid probably is absorbed rapidly into the lower pressure venules, which thus concentrates the interstitial fluid protein and produces the normal pleural fluid protein of 1.0 to 1.5 gjd!' Studies of equilibration of radiolabeled albumin from the intravascular to the pleural space in awake sheep provide evidence for a low filtration rate. An entry rate of 0.01 mlj(kg x hr) or 7 mljday in a 30-kg sheep has been noted (18). This is in marked contrast to previous work, which found entry rates of the magnitude of 0.47 mlj(kg x hr) or approximately 225 mljday in a 20-kg dog (16). The high entry rate documented in prior experiments probably can be explained by pleural catheter inflammation (19). Pleural liquid and protein have been shown to leave the pleural space of sheep at approximately 0.28 mlj(kg x hr) following creation of a sterile hydrothorax at a protein concentration of I gjdl (20). This same study also demonstrated the capacity to increase pleural lymphatic flow at least 20-fold over baseline. These data show the reserve of pleural lymphatics to be large, and for a pleural effusion to form, fluid formation rate must be very high or lymphatic flow impaired. Once excess fluid formation ceases, normal lymphatics probably can remove close to 500 ml of transu dative fluid per day in a 70-kg man. Thus, it appears that pleural liquid is formed and removed slowly and has a lower protein concentration than both lung and peripheral lymph. The parietal pleura appears to be the "business end" of liquid and protein exchange in the pleural space. A low protein filtrate from the systemic vessels enters the parietal pleural interstitial space, is concentrated, and subsequently leaks through the mesothelium. Both pleural liquid and protein exit by the parietal pleural stomata. Since lymphatic drainage of the pleural space has a large reserve, accumulation of an abnormal amount of pleural fluid represents markedly increased fluid formation, decreased
10
SAHN
clearance, or blockage of the clearance pathways, or a combination of the three mechanisms. MECHANISMS OF PLEURAL FLUID
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
ACCUMULATION IN DISEASE An increased rate of formation of pleural fluid can occur with increase in microvascular hydrostatic pressure, decrease in microvascular oncotic pressure, decrease in pressure in the pleural space, and increase in micro vascular permeability; movement of fiuid from the peritoneal to the pleural space is also a cause of pleural fluid accumulation (21). A decreased clearance rate can be found with systemic venous hypertension (22), and blockage of clearance pathways has been documented with malignant involvement of the pleural lymphatics (23). An acute experiment in dogs indicated that systemic venous pressure is the most important determinant of hydrostatic pleural fluid formation and volume (24). However, recent clinical data suggest that increased pulmonary venous pressure is the most important factor in the develop ment of pleural effusions in congestive heart failure (25, 26). It is probable that the fluid leaks into the interstitium of the lung and moves across the visceral. mesothelium along an interstitial-pleural pressure gradient. Preliminary observations from our laboratory have noted that the greatest incidence of pleural effusions is seen in patients with congestive heart failure and both chronic systemic and pulmonary venous hypertension. A decrease in oncotic pressure in the microvascular circulation will increase the tendency to form pleural interstitial liquid. This will result in the increased entry of low protein fluid into the pleural space from the parietal pleural surface. The clinical observation that this mechanism is an unusual cause of a large pleural effusion probably can be explained by the functional reserve of the parietal pleural lymphatic system. However, if the visceral pleura also is involved in pleural leak or if lymphatic drainage is impaired, larger volumes of fluid can accumulate. When there is a decrease in pressure in the pleural space as in atelectasis, the pressure gradient from the pleural interstitium to the pleural space increases and favors the formation of pleural liquid. Furthermore, the separation of the lung from the chest wall could decrease pleural space fluid movement during respiration and inhibit optimal parietal pleural lymphatic drainage. Clinically, moderate to large pleural effusions occur only with complete lung collapse; with minor atelectasis, once a given volume of fluid has formed, pleural pressure returns toward normal and the increased rate of pleural liquid formation ceases. Increased permeability of the microvascular circulation is the mech-
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
PLEURAL EFFUSIONS
11
anism of pleural fluid formation that occurs with inflammation of both pulmonary and pleural microvessels regardless of etiology. Liquid and protein leak across the lung and pleural microvessels at an increased rate. The pleural entry rate could overwhelm the pleural reabsorptive capacity; however, decreased clearance may also be implicated when the parietal pleural stomata are occluded by fibrin, debris, and mesothelial swelling secondary to the inflammatory process. It has been observed retrospectively that pleural fluid occurred in 36% of patients with ARDS compared to 40% with hydrostatic pulmonary edema (27). In a prospective study of eight patients with ARDS, all had pleural fluid documented by ultrasound and two autopsied patients had exudative effusions (28). Experimental studies also have documented pleural fluid associated with ARDS and support its origin from lung edema (29, 30). These effusions usually are small and may be unrecognized clinically as the patient has diffuse alveolar infiltrates and chest radiographs are obtained with the patient in the supine position, which negates the sign
of costophrenic angle blunting. The mechanism of pleural fluid formation in ARDS is movement of extravascular lung water along the interstitial pleural pressure gradient. The use of positive pressure ventilation in the treatment of ARDS would tend to inhibit pleural fluid accumulation by decreasing the interstitial-pleural gradient. A blockage at any juncture in the lymphatic drainage system from the stomata to the mediastinal lymph nodes, whether due to tumor, fibrosis, or lymphatic abnormalities, will result in pleural fluid accumulation. Fluid formation may occur slowly as in patients with yellow nail syndrome, who have a congenital lymphatic abnormality, or it may occur relatively rapidly in the patient with malignant involvement ofiymphatics, who, in addition, has pleural seeding with tumor and increased fluid formation from increased microvascular permeability (21). Large effusions from malig nancy will result only with significant involvement of the lymphatic system. Any fluid, whcther high or low in protein concentration, can movc from the peritoneal to pleural cavity through either diaphragmatic defects or diaphragmatic lymphatics. The diaphragmatic defects usually are small, being < 1 cm in diameter (31). Fluid moves from the peritoneal to pleural space because of the pressure gradient across the diaphragm. When an acute, massive pleural effusion develops with ascites, the portal of entry into the pleural space is usually a diaphragmatic defect resulting in rapid movement of fluid into the pleural space that exceeds the exit rate of the pleural lymphatics. It would appear that for a massive pleural effusion to evolve, a "ball-valve" mechanism must be operative for fluid accumulation would cease when pleural and peritoneal pressures equalize.
12
SAHN
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
Literature Cited 1. Agostoni, E., D'Angelo, E. 1969. Thick ness and pressure of the pleural liquid at various heights and with various hydrothoraces. Respir. Physiol. 6: 33042 2. Leak. L. Y., Rahil. L. 1978. Permeability of the diaphragmatic mesothelium: the ultrastructural basis for "stomata." Am. J. Anat. 151: 557-94 3. Wang, N. S. 1975. The preformed stomas connecting the pleural cavity and the lymphatics in the parietal pleura. Am. Rev. Respir. Dis. 111: 12-20 4. Courtice, F. C., Simmonds, W. 1. 19 54. Physiological significance of lymph drainage of the serous cavities and lungs. Physiol. Rev. 34: 419-48 5. Staub, N. C., Weiner-Kronish, 1. P., Albertine, K. H. 1985. Transport through the pleura. Physiology of nor mal liquid and solute exchange in the pleural space. In The Pleura in Health and Disease, ed. 1. Chretein, J. Bignon, A. Hirsch, pp. 169-93. New York: Mar cel Dekker 6. McLaughlin, R. F., Tyler, W. S.,
Canada, R. O. 1961. A study
of
the
subgross pulmonary anatomy in various animals. Am. J. Anat. 108: 149-65 7. McLaughlin, R. F. 1983. Bronchial artery distribution in various animals and humans. Am. Rev. Respir. Dis. 128: S57-S58 8. McLaughlin, R. F., Tyler, W. S., Canada, R. O. 1966. Subgross pul monary anatomy of the rabbit, rat, and guinea pig with additional notes on the human lung. Am. Rev. Respir. Dis. 94: 380-87 9. Naigaishi, C. 1972. Functional Anatomy and Histology of the Lung, pp. 79-179. Baltimore: Univ. Park 10. Krahl,Y.E.1964.Anatomy ofthe mam malian lung. In Handbook of Physiology and Respiration, ed. W. O. Fenn, H. Rahn, 1(3): 2\3-84. Washington, DC: Am. Physio!. Soc. 11. Albertine, K. H., Weiner-Kronish, J. P., Roos, P. J., Staub, N. C. 1982. Structure, blood supply and lymphatic vessels of the sheep's visceral pleura. Am. J. Anat. 165: 227-94 12. Testut, L. 1930. Traite d'Anatomie Humaine tom. III, p. 1003. Paris: Doin. 8th cd. 13. Bhattacharya, J., Gropper, M. A., Staub, N. C. 1984. Interstitial pleural pressure gradient measured by micro puncture in excised dog lung. J. Appl. Physiol. 56: 271-77 14. Miserocchi, G., Agostoni, E. 1971. Con-
15.
16.
17.
tents of the pleural space. J. Appl. Physiol. 30: 200-13 Sahn, S. A., Willcox, M. L., Good, J. T. Jr., Potts, D. E., Filley, G. F. 1979. Characteristics of normal rabbit pleural fluid: physiologic and biochemical impli cations. Lung 156: 63-69 Stewart, P. B., Burgen, A. S. Y. 1958. The turnover of fluid in the dog's pleural cavity. J. Lab. Clin. Invest. 52: 21230 Landis, E. M., Pappenheimer, J. R. 1963. Exchange of substances through the capillary walls. In Handbook of Physiology, Sect. 2, 2: 961 1034. Wash ington, DC: Am. Physio!. Soc. Weiner-Kronish, J. P., Albertine, K. H., Licko, Y., Staub, N. C. 1984. Protein egress and entry rates in pleural fluid and plasma in sheep. J. Appl. Physiol. 56: 459-63 Weiner-Kronish, J. P.,Albertine, K. H., Roos, P. J., Staub, N. C. 1982. Pleural fluid dynamics in sheep are altered by a pleural catheter. Fed. Proc. 41: 1127 Broaddus, V. c., Wiener-Kronish, J. P., -
18.
19.
20.
Berthiaume, Y., Staub, N. C. 1988.
Removal of pleural liquid and protein by lymphatics in a wake sheep. J. App/. Physiol. 64: 384-90 21. Sahn, S. A. 1988. The state of the art. Thc pleura. Am. Rev. Respir. Div. 138: 184-234 22. Broaddus, V. c., Araya, M., Staub, N. C. 1989. Evidence of pleural lymphatic absorption ceases with elevation of sys temic venous pressure in volume-loaded rabbits. Am. Rev. Respir. Dis. 139: A413 23. Meyer, P. C. 1966. Metastatic carcinoma of the pleura. Thorax 21: 437-43 24. Mellins, R. B., Levine , O. R., Fishman, A. P. 1970. Effect of systemic and pul monary venous hypertension on pleural and pericardial fluid accumulation. J. Appl. Physiol. 29: 564-69 25. Wiener-Kronish, J. P., Matthay, M. A., Callen, P. W., Filly, R. A., Gamsu, G., et al. 1985. Relationship of pleural effusions to pulmonary hemodynamics in patients with congestive heart failure. Am. Rev. Respir. Dis. 132: 1253-56 26. Weiner-Kronish. J. P.. Goldstein, R., Matthay, M. A., Biondi, J. W., Broaddus, V. c., et al. 1987. Lack of association of pleural effusion with chronic pulmonary arterial and right atrial hypertension. Chesl 92: 967-70 27. Aberle, D. R., Wiener-Kronish, J. P., Webb, W. R., Matthay, M. A. 1987. Diagnosis of hydrostatic versus in creased permeability pulmonary edema
PLEURAL EFFUSIONS
on the basis of chest radiographic criteria in critically ill patients. Am. Rev. Respir. Dis. 135: A215 28. Wiener-Kronish, 1. P., Goldstein, R., Matthay, M. A. 1988. Pleural effusions are frequently associated with the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 137: 227 29. Wiener-Kronish, 1. P., Broaddus, V. C., Albertine, M. A., Gropper, M. A., Mat thay, M. A., et al. 1988. Relationship of
Annu. Rev. Med. 1990.41:7-13. Downloaded from www.annualreviews.org Access provided by University of Laval on 01/22/15. For personal use only.
pleural
effusions to increased perme-
13
ability pulmonary edema in anesthetized sheep. 1. CUn. Invest. 82: 1422-29
30. Miller, K. S., Harley, R. A., Sahn, S. A. 1989. Pleural effusions associated with ethchlorvynol lung injury result from
visceral pleural leak. Am. Rev . Respir. Dis. 140: 764-70 31. Lieberman, F. L., Hidemura, R., Peters, R. L., Reynolds, T. B. 1966. Patho genesis and treatment of hydrothorax complicating cirrhosis with ascites. Ann. Intern. Med. 64: 341-51
