© 2003 American Thoracic Society
Pump and CircumstancesFaculté Xavier Bichat Paris, France Many experimental models have been designed to explore how ventilation could damage lungs and produce permeability-type pulmonary edema. Edema produced by ventilation with a high tidal volume varies in severity depending on the intensity and duration of the challenge and on previous lung state. Alveolar flooding with foam in airways excludes zones from the ventilation, and creates a vicious circle that increases the risk of overinflation of the remaining ventilated areas. This risk exists in patients suffering from the acute respiratory distress syndrome, even if ventilated with low tidal volumes. It is thus important to recruit zones filled with edematous liquid, and a means to achieve this goal is to accelerate alveolar liquid absorption. This may be obtained by activation of molecular mechanisms implicated in vectorial sodium transport by alveolar-bronchiolar cells responsible for liquid absorption, such as the sodium channel or the Na,K-ATPase pump. Unfortunately, overinflation experimentally decreases alveolar liquid resorption by depressing Na,K-ATPase function in the alveolar epithelium (1), for a yet unclear reason. Normal absorption may be restored in rat lungs submitted to overinflation by stimulating sodium transport with ß-adrenergic agonists (2), although these agents have not proven as effective in a similar model (3) or in patients with acute lung injury (4). In this issue of the Journal (pp. 1445–1448), Adir and coworkers (5) explore another possibility, overexpression of Na, K-ATPase ß1 subunit in alveolar epithelial cells, a gene transfer strategy that the same team designed to increase liquid clearance in rats (6). They measured liquid clearance in lungs isolated from rats that were ventilated for 40 minutes with a rather high tidal volume and rate 7 days after infection with the adenovirus (used for gene transfer). They say they increased "lung liquid clearance", but the usual term "alveolar liquid clearance" would have been preferable because they measured the rate at which liquid is removed from airspaces. This difference is not only semantic. It reflects a physiologic reality and the two do not necessarily harmonize. If alveolar clearance is faster than the rate of removing interstitial liquid by lung microcirculation, hydration of the interstitium increases and a back flux may occur that redirects more and more liquid to alveoli over time. Lung clearance rate may thus be much lower than alveolar liquid removal suggests (7). The same is true under mild edematogenous condition when lung interstitium swells because it accumulates both fluid filtrating from capillaries and absorbed from alveolar spaces. Adir and coworkers (5) do not provide undisputable evidence that there was no such back-flux during clearance measurement in their edematous preparation. It would have been preferable to discard this possibility by more complete blockade of active sodium transport (with the use of a sodium channel blocker in addition to ouabain, instead of ouabain alone). Nevertheless, clearance was three times higher in lungs from treated rats than in lungs from control animals ventilated with the same injurious protocol. This clear-cut result was obtained under an aggregate of several helpful conditions. The investigators used an isolated lung preparation in which it is possible to set perfusion so as to reduce filtration pressure. This is not the case during in vivo clearance measurements. In situ, nonperfused lungs can also be used but are likely to underestimate alveolar clearance when resorption rate is high (8), which is the expected effect of Na,K-ATPase ß1 subunit gene transfer. Clearance studies in edematous lungs are problematic because flooding, even when minimal, may cause foam and formation of menisci in airways (9), which preclude homogeneous filling of lungs because of the high surface tension at the trapped gas interface. Uneven filling alters the distribution of exchange surface/volume ratio and affects clearance and permeability measurements. Despite a rather aggressive ventilation protocol, Adir and coworkers (5) apparently produced only mild to moderate edema, with minimal lung damage. The relatively small volume of epithelial lining fluid suggested the absence of alveolar flooding, and the only slightly altered permeability of the alveolar barrier suggested the absence of significant damage. Measurement of epithelial lining fluid volume by dilution of a foreign tracer, however, lacks sensitivity (10), raising some concern about the reported estimate. The extent to which lungs are injured is important because alterations of the alveolar barrier properties are likely to decrease the sodium reflection coefficient and, thus, the efficiency of sodium transport to drive liquid absorption. It would be interesting to evaluate whether this gene treatment would increase clearance in vivo as well and for more severe or different forms of ventilation-induced lung injury. Endothelial defects produced by overinflation may reverse spontaneously (11). Inflammatory phenomena aggravate edema severity when aggressive ventilation is prolonged (12). Reduction of liquid clearance by a similar period of high-volume ventilation reverses a few hours later in vivo (3), but the ability to stimulate clearance by ß-adrenergic agonists is lost. This pattern was not observed in earlier studies on isolated lungs (2). Isolated, liquid-filled lungs are no longer subjected to mechanical stretch or shear stress and are perfused with an artificial medium devoid of inflammatory cells. The course of injury is thus different from that in vivo. Nitric oxide production is implicated in lung endothelial defect due to high-volume ventilation (3, 13). Ultrastructural abnormalities observed after high-volume ventilation (11) are similar to those seen during increased endothelial nitric oxide production (14). Nitric oxide signaling (15) or free radicals (16) may decrease Na,K-ATPase activity. Whether nitric oxide signaling or derived reactive species are implicated in the lower Na,K-ATPase activity that follows high-volume ventilation may be worth investigating. It is astonishing to see the plethora of phenomena that are triggered by simple overdistension of lungs and how sensitive these are to the experimental conditions. FOOTNOTES Conflict of Interest Statement: G.S. has no declared conflict of interest. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
