INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mead and colleagues (1) postulated that airspaces inflate nonuniformly as a result of interdependence of atelectatic regions surrounded by fully expanded areas. The concept of pulmonary inhomogeneity, a phenomenon expected because of instability and collapse of airspaces and dynamic nonuniformity of ventilation, was then introduced.

The presence of discrete mechanical unevenness was clearly demonstrated in normal lungs (2), an event that is dramatically increased in the acute respiratory distress syndrome (ARDS) (3). ARDS lungs can be theoretically divided into three compartments: a normal, an intermediary, and a diseased zone, the last two potentially recruitable by increasing pressures. All these zones respond to positive transpulmonary pressure in different ways. Normal units expand uniformly and overinflate at high-pressure levels. Intermediary components undergo an increase in pressure without an important change in volume until a critical point is reached where the alveoli suddenly pop open. Finally, diseased regions also show the latter behavior but higher pressures are required to open the distal airways (4, 5).

The aim of the present study is to evaluate histoarchitecture of distal lung parenchyma at different positive transpulmonary pressures, both in inflation and deflation of excised normal and damaged rat lungs. The basic approach is a morphological and morphometric analysis of the unevenness of alveolar geometry, mainly at low lung volumes, in an experimental model of acute lung injury with paraquat. Paraquat is a herbicide that induces alveolar epithelial damage due to its accumulation in pneumocytes type II, leading to surfactant dysfunction (6). It is an experimental model of diffuse alveolar damage of low cost, rapid effect, and simplicity of administration that has been widely used to study acute lung injury (7, 8).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Groups

Fifty non-specific-pathogen free Wistar rats weighing 190 to 330 g were used. These rats were randomly allocated to experimental and control groups with 25 animals each. Both groups were further divided in subgroups of five rats which were characterized as follows:

Control, pressure of 5 cm H₂O, inspiratory phase (Con P5ins)

Control, pressure of 15 cm H₂O, inspiratory phase (Con P15ins)

Control, pressure of 25 cm H₂O (Con P25)

Control, pressure of 15 cm H₂O, expiratory phase (Con P15ex)

Control, pressure of 5 cm H₂O, expiratory phase (Con P5ex)

Paraquat, pressure of 5 cm H₂O, inspiratory phase (Pq P5ins)

Paraquat, pressure of 15 cm H₂O, inspiratory phase (Pq P15ins)

Paraquat, pressure of 25 cm H₂O (Pq P25)

Paraquat, pressure of 15 cm H₂O, expiratory phase (Pq P15ex)

Paraquat, pressure of 5 cm H₂O, expiratory phase (Pq P5ex)

Lung Preparation and Apparatus

Paraquat was injected intraperitoneally (25 mg/kg of body weight) 24 h prior to the measurements. The animals were anesthetized with pentobarbital sodium (30 mg/kg body weight, intraperitoneally) and killed by exsanguination via the abdominal aorta. The lungs were surgically exposed, allowed to collapse freely, and removed en bloc. The heart was dissected away. In order to measure pressure-volume relationships, the lungs were placed in a 280-ml plethysmograph connected to a graded pipette. The tracheal cannula was connected to a 20-ml syringe and to a small vertical column partially filled with water (Figure 1). When lungs were inflated, volume was measured by the displacement of a detergent meniscus in the pipette. Initially, lungs were subjected to three inflations up to 30 cm H₂O, and then allowed to return to their resting volume. This initial larger inflation is usually done in experiments with excised lungs in order to get rid of atelectatic areas due to thoracotomy (9, 10). Lungs were filled with air in small increments of volume until the desired pressure level was achieved. For measurements during deflation, lungs were inflated until 25 cm H₂O and then deflated to the desired point. Thirty seconds were allowed at each step for equilibration before tracheal pressure was recorded.

View larger version (47K): [in this window] [in a new window]   Figure 1.   Schematic representation of the system used to obtain the pressure-volume curve in the excised rat lungs.

Morphological and Morphometric Evaluation

Immediately after the achievement of the desired pressure level, the tracheal cannula was occluded and lungs were quick-frozen by immersion in liquid nitrogen. Fixation was made with Carnoy's solution (ethanol:chloroform:acetic acid 60:30:10 by volume) at 70° C; after 24 h lungs were kept at 20° C and the concentration of ethanol was progressively increased to reach absolute ethanol. Sections representing the central and peripheral areas of the lungs were sampled and processed for paraffin embedding. Histological sections of 5 µm were obtained and stained with hematoxylin-eosin.

Morphometric analysis was made with an integrating eyepiece with a coherent system made of a 100-point grid consisting of 50 lines of known length, coupled to a conventional light microscope. The volume fraction of collapsed and normal pulmonary areas, as well as the fraction of the lung occupied by large-volume gas-exchanging airspaces (LVGEASstructures with a morphology distinct from that of alveoli and wider than 120 µm), were determined by point-counting procedures, made at a magnification of ×40 across 10 microscopic noncoincident fields (11, 12).

After this procedure, 10 fields per animal were focused at ×400 and the number of intercepts of the lines with alveolar epithelium and points falling on air or tissue were counted.

Lung surface density can be estimated by intersection counting, so that the probability of a test line of given length to intercept the alveolar septum is directly proportional to its surface density in the lung parenchyma. The surface-to-volume ratio was, therefore, calculated by

where l is the number of intersections of the alveolar septum with a test line and L is the cumulative length of the test line obtained by

where P is the number of points counted over the airspaces and d is the length of the test line in µm.

The surface-to-volume (S/V) relation is considered in this work as a shape estimator of gas-exchanging parenchyma. It describes if a given increase in volume (V) occurs with changes in surface (S) due to alveolar recruitment, or without changes in surface, without opening of distal units. Its usefulness has been proved in several morphometric studies (13).

Statistical Analysis

The significance of the results obtained was assessed by means of analysis of variance (ANOVA) for independent samples, taking into account two factors: pressure and treatment (paraquat or control), using SPSS version 6.1 statistical software (SPSS Inc., Chicago, IL) (16). The level of significance was set at 5%.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

LVGEAS

Figure 2 depicts the results (means and standard errors) for the proportion of volume occupied by LVGEAS for control and paraquat groups in the five levels of pressure considered. There was a significant effect of pressure (p < 0.001) and treatment (p < 0.001) and a nonsignificant degree of interaction between pressure and treatment (p = 0.07). For all pressure levels studied, animals treated with paraquat presented a significantly larger proportion of their pulmonary gas volume contained within LVGEAS in comparison to controls. The relative proportion of LVGEAS exhibited an increase (more pronounced in paraquat-treated animals) when transpulmonary pressure changed from 5 to 15 cm H₂O during inflation. When the maximal inflation pressure was achieved (25 cm H₂O), the volume fraction of LVGEAS decreased in both groups, suggesting that a larger proportion of small ventilating units (possibly alveoli) was recruited. During deflation, the LVGEAS fraction remained virtually unchanged in control animals. In the paraquat group there was a sharp increase of the fraction of LVGEAS at low expiratory pressure, suggesting that some degree of alveolar instability might have occurred, with gas transference from small units to larger ones.

View larger version (13K): [in this window] [in a new window]   Figure 2.   Graphic representation of the areal fraction of LVGEAS as a function of transpulmonary pressure (P5, 15, and 25 cm H2O during inspiration [-ins] and expiration [-ex]), for control and paraquat-treated animals. Each point represents the mean of five animals; bars represent standard error of the mean. A significant effect of treatment and pressure (p < 0.001) was determined, whereas treatment  pressure interaction disclosed a marginal level of significance (p = 0.07), ANOVA.

Surface-to-volume Ratio

Figure 3 presents the means and standard errors of the S/V ratio. Paraquat-treated animals presented significantly smaller S/V ratios (p < 0.001) for all points of pressure studied, indicating that the amount of gas-exchanging surface available for a given lung volume is reduced. This finding is coherent with the presence of larger areas of uneven distribution of air (collapse and overinflation) within lung parenchyma in paraquat-treated animals. The S/V configuration was significantly (p = 0.002) affected by changes in both transpulmonary pressure and pulmonary volume history. When transpulmonary pressure changed from 5 to 15 cm H₂O during inflation, S/V ratio decreased in both groups, suggesting that volume changes probably induced some degree of parenchyma distention. After overcoming a critical pressure (between 15 and 25 cm H₂O), recruitment of more compartmentalized units (with a larger S/V ratio) probably occurred. During deflation, S/V ratio did not appreciably change in both groups. Interestingly enough, a high degree of alveolar recruitment led the S/V ratio of paraquat-treated animals to approach that of the control group during lung emptying. This finding suggests that a full inflation, followed by alveolar pressurization, may improve parenchyma geometry (in terms of S/V ratio) in this model of alveolar instability.

View larger version (14K): [in this window] [in a new window]   Figure 3.   Graphic representation of the fraction of S/V as a function of transpulmonary pressure (P5, 15, and 25 cm H2O during inspiration [-ins] and expiration [-ex]), for control and paraquat-treated animals. Each point represents the mean of five animals; bars represent standard error of the mean. A significant effect pressure (p < 0.001) and treatment (p = 0.002) was determined, ANOVA.

Descriptive Histopathology

A better description of the conformational changes of distal pulmonary parenchyma may be provided by the set of histopathological fields depicted in Figure 4. There is some degree of uneven gas distribution during midinflation, especially in paraquat-treated animals (Figure 4C). A full inflation makes paraquat-treated animals exhibit a more uniform gas distribution (Figure 4E). Alveolar collapse was easily identified at end-deflation in paraquat-treated animals (Figure 4I), but not in control animals (Figure 4J).

View larger version (88K): [in this window] [in a new window]   Figure 4.   Representative panel illustrating the histopathological patterns of distal pulmonary parenchyma studied at different levels of transpulmonary pressure during inflation and deflation. In comparison with control (right column), paraquat-treated lung (left column) presented evidence of dilation of large-volume gas-exchange airspaces and alveolar collapse at low pressures (A through D and I through J ). A full inflation makes paraquat-treated animals exhibit a more uniform gas distribution (E ). Photographs were taken at an original magnification of ×100 from slides stained by hematoxylin-eosin. A = paraquat lung at 5 cm H2O, B = control lung at 5 cm H2O; inflation. C = paraquat lung at 15 cm H2O, D = control lung at 15 cm H2O; inflation. E = paraquat lung at 25 cm H2O, F = control lung at 25 cm H2O. G = paraquat lung at 15 cm H2O, H = control lung at 15 cm H2O; deflation. I = paraquat lung at 5 cm H2O, J = control lung at 5 cm H2O; deflation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mead suggested thirty years ago that the static hysteresis of the pressure-volume curve in both living and excised lungs could be due to recruitment of units during inflation (17). This concept indicates that critical opening pressures do exist along lung parenchyma, which are mostly regulated by surface tension. Given the wide variation and dynamism of ARDS mechanics, many investigators have used the static pressure-volume curve of the respiratory system to determine the tidal pressure range that avoids the low compliance regions associated with collapse and overdistension. An inflection zone, the lower one of the pressure-volume curve (Pflex), is accepted as the "ideal positive end-expiratory pressure" ("Ideal PEEP") (18, 19). Some investigators, however, have suggested that the Pflex may overestimate the least PEEP level actually required to maintain alveolar stability (20, 21).

The management of a patient with respiratory failure involves two conflicting goals: to provide ventilatory support to maintain adequate gas exchange and to avoid pressure and/or volume trauma promoted by inadequate volume distribution during positive pressure ventilation. The perfect matching of these two aims is often difficult to achieve in clinical practice and, in this scenario, it is quite possible that inadequate volume distribution may play a significant additional role in the pathogenesis of pulmonary deterioration in patients under mechanical ventilation (22). The use of the "ideal PEEP" combined with low tidal volumes has been claimed to influence the prognosis of ARDS patients, simply by avoiding mechanical stress at the dependent pulmonary zones (23, 24). There is a growing body of experimental evidence suggesting that high transpulmonary pressures may produce both epithelial and endothelial damage (25). In addition, it is plausible that the "shearing stress" caused by the amplitude of ventilation with persistent collapse and/or tidal opening and reclosure of injured tissue may cause lesion too (28). Thus, the rationale of keeping a constant pressurization of the lungs at end-expiration is to reduce the shearing stress applied by positive pressure ventilation to the pulmonary compartments with a smaller time constant.

The present study was designed to assess the changes of pulmonary histoarchitecture at different pressure points during inspiration and expiration. The idea was to determine the volume history of the gas-exchanging territory of the lung: LVGEAS. The term LVGEAS had to be created because of the difficulty to adequately classify structures larger than alveoli at lower lung volumes, that can be alveolar ducts, alveolar sacs, or distended alveoli, especially in paraquat-treated animals. As may be observed in Figure 4, it is quite difficult to distinguish respiratory bronchioli from alveolar ducts (or even overinflated alveoli) in the presence of uneven volume distribution.

We observed in paraquat-treated animals evidence of alveolar instability close to the resting volume, a situation that is minimized near total lung capacity (Figures 2-4). Our results indicate that LVGEAS are under stress during inflation from the resting volume (from 5 to 15 cm H₂O), a situation that is partially abolished at the same pressure level during deflation. Thus, our data provide structural plausibility for the notion that a significant degree of recruitment followed by alveolar pressurization may lessen parenchyma distortion caused by mechanical inhomogeneity.

In conclusion, this work demonstrates that morphological evidence of uneven distribution of inspired air may be partially reversed by applying larger alveolar pressures, that should be maintained at the end-expiratory volume to avoid alveolar collapse. Ventilation above this pressure level, where inhomogeneity is corrected, is able to maintain recruitment of gas-exchange units and theoretically may improve oxygenation.