Journal of Applied Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 87: 1301-1312, 1999;
8750-7587/99 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Verghese, G. M.
Right arrow Articles by Matthay, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Verghese, G. M.
Right arrow Articles by Matthay, M. A.
Vol. 87, Issue 4, 1301-1312, October 1999

INVITED REVIEW
Alveolar epithelial fluid transport and the resolution of clinically severe hydrostatic pulmonary edema

G. M. Verghese, L. B. Ware, B. A. Matthay, and M. A. Matthay

Departments of Medicine and Anesthesia and the Cardiovascular Research Institute, University of California, San Francisco, California 94143-0130


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

To characterize the rate and regulation of alveolar fluid clearance in the uninjured human lung, pulmonary edema fluid and plasma were sampled within the first 4 h after tracheal intubation in 65 mechanically ventilated patients with severe hydrostatic pulmonary edema. Alveolar fluid clearance was calculated from the change in pulmonary edema fluid protein concentration over time. Overall, 75% of patients had intact alveolar fluid clearance (>= 3%/h). Maximal alveolar fluid clearance (>= 14%/h) was present in 38% of patients, with a mean rate of 25 ± 12%/h. Hemodynamic factors (including pulmonary arterial wedge pressure and left ventricular ejection fraction) and plasma epinephrine levels did not correlate with impaired or intact alveolar fluid clearance. Impaired alveolar fluid clearance was associated with a lower arterial pH and a higher Simplified Acute Physiology Score II. These factors may be markers of systemic hypoperfusion, which has been reported to impair alveolar fluid clearance by oxidant-mediated mechanisms. Finally, intact alveolar fluid clearance was associated with a greater improvement in oxygenation at 24 h along with a trend toward shorter duration of mechanical ventilation and an 18% lower hospital mortality. In summary, alveolar fluid clearance in humans may be rapid in the absence of alveolar epithelial injury. Catecholamine-independent factors are important in the regulation of alveolar fluid clearance in patients with severe hydrostatic pulmonary edema.

beta -agonist; congestive heart failure; alveolar fluid clearance; mechanical ventilation; left atrial hypertension


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

SEVERE HYDROSTATIC pulmonary edema is an important cause of acute respiratory failure that occurs in a variety of clinical settings, including chronic congestive heart failure, acute myocardial infarction, and intravascular volume overload. Resolution of pulmonary edema depends on the clearance of fluid from the alveolar space, a process that requires an intact, functional alveolar epithelium (27, 39). The primary driving force for alveolar fluid clearance is the active transport of sodium from the alveolar space to the interstitium by alveolar epithelial type II cells (4, 17, 26, 36). As a result, water is passively transported through specialized water channels, the aquaporins, which are probably located primarily on alveolar epithelial type I cells (15, 38). The rate of alveolar fluid clearance has been measured experimentally in several species (4-6, 25, 30, 56). However, in humans, the rate of alveolar fluid clearance has only been measured in the ex vivo human lung (54). The maximal capacity of the uninjured human lung to transport alveolar edema fluid is unknown. Furthermore, extrapolation from ex vivo preparations may underestimate the in vivo alveolar fluid clearance capacity by as much as 50% (4, 30, 53), thus emphasizing the need for in vivo measurements of alveolar fluid clearance.

The factors that regulate alveolar fluid clearance also have been studied in experimental models. Alveolar fluid clearance can be increased by a number of mechanisms, including catecholamine-dependent and -independent pathways (36, 38). Exogenous administration of beta 2-agonists increases alveolar fluid clearance in several species (5, 6, 13, 25, 30, 35, 59), including the ex vivo human lung (53). Endogenous catecholamines also increase alveolar fluid clearance in several models, including acute left atrial hypertension in sheep (9), neurogenic pulmonary edema in dogs (32), and short-term shock in rats (48, 49). The catecholamine-independent mechanisms that upregulate alveolar fluid clearance include growth factors (8, 23, 58, 62), cytokines [tumor necrosis factor-alpha (TNF-alpha )] (50), dopamine (3), alveolar epithelial type II cell hyperplasia (22, 62), and glucocorticoids (21). However, the relative contribution of these catecholamine-dependent and -independent pathways to regulation of alveolar fluid clearance in humans with hydrostatic pulmonary edema is unknown. Furthermore, although there has been speculation regarding the clinical utility of accelerating alveolar fluid clearance with exogenous beta -agonists (6, 38, 53, 54, 59), no one has studied the in vivo effects of beta -adrenergic stimulation on alveolar fluid clearance in humans.

To gain insight into the resolution phase of clinical hydrostatic pulmonary edema, we studied 65 mechanically ventilated patients with severe hydrostatic pulmonary edema. The first objective was to determine the rate of alveolar fluid clearance in patients with severe hydrostatic pulmonary edema by serial sampling of undiluted pulmonary edema fluid. The second objective was to test the hypothesis that catecholamine-dependent mechanisms can upregulate alveolar fluid clearance in humans. The third objective was to study other catecholamine-independent factors that could modulate net alveolar fluid clearance, including medications such as glucocorticoids, hemodynamics, mechanical ventilation, and the underlying clinical cause of hydrostatic pulmonary edema. The final objective was to determine whether the presence of intact alveolar fluid clearance is associated with better clinical outcomes in critically ill patients with hydrostatic pulmonary edema.


    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Patient selection. Patients with acute hydrostatic pulmonary edema were identified prospectively from patients admitted to the intensive care units of Moffitt-Long Hospital, University of California, San Francisco, and San Francisco General Hospital between 1985 and 1998. Inclusion criteria included acute respiratory failure requiring mechanical ventilation and aspirable pulmonary edema fluid from at least two time points within the first 4 h after endotracheal intubation. The definition of acute hydrostatic pulmonary edema was based on standard clinical criteria (1) and the presence of a transudative pulmonary edema fluid with an initial ratio of edema fluid to plasma protein of <0.65 (18, 39). The clinical diagnosis of hydrostatic pulmonary edema was confirmed by detailed review of patient records and chest radiographs by two of the authors (G. M. Verghese and L. B. Ware). Specific criteria used to define hydrostatic pulmonary edema included central venous pressure (CVP) >= 14 mmHg, pulmonary arterial wedge pressure (PAWP) >= 18 mmHg, or a cardiac ejection fraction <= 45% by echocardiogram, radionuclide, or contrast ventriculography, and/or positive physical findings, including a third heart sound and jugular venous distension. Patients with a possible cause for acute lung injury were excluded, including those with sepsis, aspiration of gastric contents, or pneumonia. This study was approved by the Committee for Human Research at the University of California, San Francisco.

Sampling of pulmonary edema fluid. Pulmonary edema samples were collected by trained respiratory therapists or physicians under the supervision of the authors as previously described (39); this method has been well validated experimentally compared with alveolar fluid obtained by micropuncture (5, 6). Briefly, a soft 14-Fr-gauge suction catheter was advanced into a wedged position in a distal bronchus via the endotracheal tube. Pulmonary edema fluid was collected in a suction trap by gentle suction. Simultaneous plasma samples were obtained by venipuncture or aspiration from an indwelling venous catheter. Pulmonary edema fluid was then centrifuged at 14,000 g for 20 min, and plasma samples were centrifuged at 3,000 g for 10 min. The supernatant was aspirated and stored at -70°C. In some cases, plasma and pulmonary edema fluid were stored at 4°C before centrifugation, usually for <4 h but occasionally for up to 24 h.

Measurement of protein concentration in edema fluid and plasma. The total protein concentration was measured in edema fluid and plasma specimens in duplicate by the biuret method as previously described (39). If edema fluid volume was insufficient for measurement by the biuret method (<1% of samples), the total protein concentration was measured by refractometry.

Calculation of rate of alveolar fluid clearance. The rate of alveolar fluid clearance was calculated as the percentage of alveolar fluid volume reabsorbed per hour, as described and validated in prior clinical and experimental studies (6, 30, 38, 39). Briefly, on the basis of the observation that removal of protein from the alveoli is very slow relative to the rate of removal of liquid, the percentage of alveolar edema fluid volume reabsorbed may be estimated by the following equation
Percent alveolar fluid clearance 
= 100 × [1 − (initial edema protein/final edema protein)]

Categories of alveolar fluid clearance. Extrapolation from studies in the ex vivo human lung was used to define three categories of alveolar fluid clearance: impaired, submaximal, and maximal. Previous studies suggested that alveolar fluid clearance rates measured in in vivo lungs are approximately twice the ex vivo rates of clearance in the same species (24, 30, 53). Because maximal clearance in the terbutaline-stimulated ex vivo human lung was 7%/h (54), maximal alveolar fluid clearance was defined as greater than two times that rate, or >= 14%/h. Submaximal alveolar fluid clearance was defined as greater than the basal rate (>= 3%/h) in the ex vivo preparation (54) to two times the stimulated rate (<14%/h), and impaired clearance as less than the basal rate (<3%/h). For most analyses, the submaximal and maximal groups were combined into one group, labeled as intact clearance (>= 3%/h).

Epinephrine assays. Plasma epinephrine concentrations were measured in stored plasma samples by standard high-performance liquid chromatography techniques as previously described (49). Several studies have reported that catecholamine levels in plasma or urine remain stable during prolonged storage at -20°C (42, 47, 63). The stability of catecholamines during storage at 4°C was confirmed by measuring epinephrine levels in blood samples from normal volunteers after the addition of epinephrine to a concentration of 20,000 pg/ml. Declines in epinephrine concentrations during storage at 4°C, compared with specimens that were centrifuged, frozen, and stored immediately, were only 1.9 ± 1.7% in plasma stored for 24 h at 4°C and 8.4 ± 3.4% in whole blood stored at 4°C for 24 h.

Measurements of epinephrine were made coincidently with pulmonary edema fluid sampling. Plasma epinephrine was measured in samples collected from 10 intervals in each of 3 categories of alveolar fluid clearance rate: maximal (>= 14%/h), submaximal (>= 3%/h, <14%/h), and impaired (<3%/h). Reported epinephrine concentrations reflect the average of the plasma epinephrine levels at the beginning and end of each sampling period. In a few patients, levels of epinephrine were measured in pulmonary edema fluid by using the same methods.

Clinical data. Clinical data were obtained from the medical record by using a standardized data-collection form. Demographic data included age, gender, race, and smoking history. Physiological variables recorded included hemodynamic parameters (PAWP, CVP, systolic, diastolic, and mean arterial blood pressures, heart rate, cardiac output, systemic vascular resistance, and left ventricular ejection fraction); respiratory and ventilatory parameters [arterial pH, PO2, PCO2, alveolar-arterial PO2 difference, arterial PO2-to-inspiratory O2 fraction ratio, tidal volume, minute ventilation, positive end-expiratory pressure (PEEP), peak inspiratory pressure, and static compliance]; and multiorgan system function (measurements of renal, hepatic, central nervous system, gastrointestinal, and hematologic function, net fluid balance, and weight). Left ventricular ejection fraction was classified as normal or mildly decreased (>45%), moderately decreased (35-45%), or severely decreased (<35%) on the basis of results of echocardiography, or radionuclide or contrast ventriculography. In addition, medications administered during the period of edema fluid collection were recorded, including vasoactive agents, inhaled beta -agonists, beta -antagonists, digoxin, diuretics, systemic glucocorticoids, antiarrhythmic agents, and narcotics.

The Simplified Acute Physiology Score II (SAPS II) and the Lung Injury Score (LIS) were calculated as before (10, 61), according to the published algorithms (33, 43), during the 24-h period beginning at the time of initial edema sampling. The LIS is a four-point score based on oxygenation, level of PEEP, static thoracic compliance, and chest radiographic findings (43). In the 10 patients in whom chest radiographs were not available, the chest radiographic score was omitted from the calculation of the LIS; changes in the chest radiographic score contribute very little to changes in the LIS over time (11). The primary etiology of pulmonary edema was assessed on the basis of a detailed review of the medical chart and was classified as acute exacerbation of chronic congestive heart failure, acute myocardial infarction, or volume overload. Outcome variables included days of intensive care, death before hospital discharge, and days of unassisted ventilation during a 28-day period, as previously described (1).

Data analysis. Data were managed by using Microsoft Excel 5.0 (Microsoft, Redmond, WA). Statistical analysis was done by using SPSS 6.1.1 for Macintosh (SPSS, Chicago, IL). Continuous variables were compared by using Student's t-test or analysis of variance, and multiple comparisons were analyzed by the Student-Newman-Keuls test. Categorical variables were compared by chi 2 analysis. Nonparametric data were analyzed by using the Mann-Whitney U-test or the Kruskal-Wallis test with Dunn's test for multiple comparisons. Statistical significance was defined as P <=  0.05. All data are reported as means ± SD unless otherwise noted.


    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Patient characteristics. Sixty-five mechanically ventilated patients with severe hydrostatic pulmonary edema were studied in the first 4 h after endotracheal intubation and mechanical ventilation. A summary of demographic and clinical data is shown in Table 1. The majority of patients had pulmonary edema because of exacerbation of chronic congestive heart failure, acute myocardial infarction, and volume overload or diastolic dysfunction. The mean initial edema fluid-to-plasma protein ratio was 0.47 ± 0.10, consistent with pulmonary edema of hydrostatic origin. Notably, the overall severity of illness was quite high, as evidenced by the high SAPS II scores and the overall hospital mortality rate of 31%. The average LIS of 2.9 ± 0.6 is comparable to scores reported in patients with severe acute lung injury (10, 61).

                              
View this table:
[in this window]
[in a new window]
  		Table 1.   Clinical characteristics of 65 ventilated patients with severe hydrostatic pulmonary edema

Alveolar fluid clearance. All 65 patients had at least 2 pulmonary edema fluid samples obtained during the first 4 h after intubation and mechanical ventilation. Alveolar fluid clearance was measured from the change in protein concentration between the first and last samples taken within these first 4 h. The range of observed rates of alveolar fluid clearance was from a minimum of 0%/h (no net alveolar fluid clearance) to a maximum of 61%/h. Overall, mean alveolar fluid clearance in the first 4 h was 13%/h. To analyze the data for clinical variables that might be associated with the rate of alveolar fluid clearance, patients were classified into two categories: intact alveolar fluid clearance (>= 3%/h) and impaired alveolar fluid clearance (<3%/h). The threshold of 3%/h was based on the basal rate of clearance measured in the ex vivo human lung (54). A total of 75% of patients had intact alveolar fluid clearance, and 25% had impaired alveolar fluid clearance (Fig. 1). For some analyses, the group with intact clearance was further broken down into a group with submaximal clearance (>= 3, <14%/h) and maximal clearance (>= 14%/h), as described in METHODS. In the maximal group, 38% of the patients (n = 25) had a mean rate of alveolar fluid clearance of 25 ± 12%/h (Fig. 2). Overall, 37% of the patients (n = 24) fell into the submaximal group, with a mean alveolar fluid clearance of 8 ± 3%/h.


View larger version (11K):
[in this window]
[in a new window]
  		Fig. 1.   Alveolar fluid clearance is intact in majority (75%) of patients with severe hydrostatic pulmonary edema. Alveolar fluid clearance was measured by serial protein concentrations in pulmonary edema fluid from 65 mechanically ventilated patients with hydrostatic pulmonary edema. Intact clearance was defined as >= 3%/h and impaired clearance as <3%/h. Each  represents rate of alveolar fluid clearance in single patient. n, No. of subjects.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 2.   Rates of alveolar fluid clearance in 65 patients with severe hydrostatic pulmonary edema. Groups were defined as follows: impaired <3%/h, submaximal >= 3 to <14%/h, maximal >= 14%/h. Solid bars show percentage of 65 patients in each group. n, No. of subjects.

To determine whether a rise in plasma protein could account for any of the observed alveolar fluid clearance, plasma protein concentrations were compared at the beginning and end of each time interval over which alveolar fluid clearance was measured. In the patients with intact alveolar fluid clearance, mean plasma protein was 5.6 ± 1.7 g/dl at the time of the initial edema fluid sample and was 5.7 ± 1.5 g/dl at the time of the final edema fluid sample, a nonsignificant difference. Similarly, in the patients with impaired alveolar fluid clearance, mean plasma protein was 6.3 ± 1.7 g/dl at the time of the initial edema fluid sample and 5.8 ± 1.4 g/dl at the time of the final edema fluid sample, also a nonsignificant difference. Thus changes in the transvascular osmotic pressure did not appear to influence the rate of alveolar fluid clearance in these patients.

Relationship between baseline clinical characteristics and alveolar fluid clearance. A univariate analysis was done to evaluate the association of selected clinical and demographic parameters with the presence or absence of alveolar fluid clearance (Table 2). Demographic factors such as age, gender, and smoking status were not different between the impaired and intact alveolar fluid clearance groups. There was a trend toward an association of race (Caucasian) with intact alveolar fluid clearance (P = 0.11). LIS, the initial edema fluid-to-plasma protein ratio, and the maximum temperature during the first 24 h were not associated with intact or impaired clearance. Interestingly, there was a trend toward higher clearance rates in patients with chronic congestive heart failure. As shown in Table 2, 35% of patients with intact alveolar fluid clearance had chronic congestive heart failure compared with only 13% of patients with impaired clearance, although the difference did not reach statistical significance because of the relatively small number of patients in the congestive heart failure group (n = 19). Expressed another way, 89% of chronic congestive heart failure patients (17 of 19) had intact clearance vs. only 69% of patients with hydrostatic edema from other causes (31 of 45) (P = 0.12).

                              
View this table:
[in this window]
[in a new window]
  		Table 2.   Comparison of baseline clinical characteristics in patients with intact alveolar fluid clearance vs. those with impaired alveolar fluid clearance

Plasma epinephrine levels. Because experimental studies have demonstrated that endogenous catecholamines can upregulate alveolar fluid clearance, plasma epinephrine levels were measured in a subset of 30 patients. Plasma epinephrine concentrations were measured in stored plasma samples paired with initial and final edema fluid specimens from 10 patients randomly selected from each category of alveolar fluid clearance (maximal, submaximal, and impaired). Average interval epinephrine concentrations in the plasma ranged from 99 to 19,170 pg/ml (10-12 to 10-7 M). Reported maximum plasma epinephrine concentrations in unstressed, supine, normotensive control human subjects are ~100-150 pg/ml (51). There were no significant differences in the plasma epinephrine levels in relationship to the rates of alveolar fluid clearance (Fig. 3). Similarly, there was no threshold level of plasma epinephrine that differentiated patients with impaired alveolar fluid clearance rates from those with maximal or submaximal rates of fluid clearance. When the administration of exogenous catecholamines (dobutamine or inhaled beta 2-agonists) was considered along with endogenous epinephrine levels, there was still no correlation with alveolar fluid clearance (data not shown). Furthermore, 2 of 10 patients with alveolar clearance rates >14%/h had plasma epinephrine levels within the normal range, and the patient with the highest measured plasma epinephrine level had no detectable alveolar fluid clearance (Fig. 3).


View larger version (15K):
[in this window]
[in a new window]
  		Fig. 3.   Plasma epinephrine levels are not associated with rate of alveolar fluid clearance in patients with hydrostatic pulmonary edema. Box plot summary of mean plasma epinephrine level in patients with impaired, submaximal, or maximal alveolar fluid clearance. There were 10 patients tested in each of the 3 groups. Each box encompasses 50% of data; horizontal line, median; error bars encompass 80% of data. x, Outliers; , mean. P = not significant between groups.

Pharmacological agents. The effect of vasoactive medications on alveolar fluid clearance, including those causing beta -adrenergic stimulation or blockade and inhaled beta -agonists, was evaluated. Medications with potential effects on a variety of ion channels or transporters, such as digoxin, diuretics, cardiac antiarrhythmic agents, and exogenous glucocorticoids, were also studied. The results are summarized in Tables 3 and 4.

                              
View this table:
[in this window]
[in a new window]
  		Table 3.   Effect of aerosolized beta -agonist on resolution of pulmonary edema


                              
View this table:
[in this window]
[in a new window]
  		Table 4.   Percentage of patients receiving selected pharmacological agents in 2 groups: intact vs. impaired alveolar fluid clearance

Although nearly twice as many patients in the group with intact alveolar fluid clearance received an inhaled beta -agonist compared with the group with impaired clearance, this difference did not reach statistical significance. A power calculation revealed that this study had a power of only 0.10 to detect the observed beta -agonist effect because of the small number of patients (n = 13) that received inhaled beta -agonists (Table 3). However, the overall positive predictive value for alveolar fluid clearance associated with inhaled beta -agonists was 85% (11/13). There was no association between intravenous dobutamine, which also has beta -agonist activity, and alveolar fluid clearance. Although some experimental data suggest that both dopamine or glucocorticoids can augment alveolar fluid clearance (3, 21), there was also no significant association between treatment with dopamine and the presence of intact alveolar fluid clearance in this study (Table 4).

Twenty-three percent of the patients in the group with intact alveolar fluid clearance received digoxin, an inhibitor of Na+-K+-ATPase, whereas no patient in the group with impaired clearance received this medication (P = 0.05). Because of the association noted above between chronic congestive heart failure and intact alveolar fluid clearance, the patients receiving digoxin were categorized by cause of hydrostatic pulmonary edema. Of the 11 patients who received digoxin, 9 had chronic congestive heart failure, suggesting that the association with intact alveolar fluid clearance may have been related more to the underlying cause of pulmonary edema than to the pharmacological effect of digoxin.

Cardiac and hemodynamic indexes. Several indexes of cardiac function and hemodynamic measurements were tested for an association with the presence or absence of intact alveolar fluid clearance. A total of 75% (49 of 65) of the patients had objective measurement of cardiac function by echocardiography or radionuclide or contrast ventriculography within 48 h of pulmonary edema fluid sampling. Of these patients, 35% had normal or mildly decreased left ventricular function, 30% had moderately severe left ventricular dysfunction, and 35% had severely decreased left ventricular function. There was no significant difference in the distribution of ejection fractions between the group with intact alveolar fluid clearance and the group with no detectable alveolar fluid clearance (Table 5).

                              
View this table:
[in this window]
[in a new window]
  		Table 5.   Selected cardiac and hemodynamic parameters in patients with intact vs. impaired alveolar fluid clearance

CVP was measured in a total of 27 patients, whereas 11 patients had measurement of PAWP. As shown in Table 5, there was no association between a decline in CVP or PAWP and the presence or absence of alveolar fluid clearance. In fact, in the group with impaired alveolar fluid clearance, all of the patients had a fall in PAWP in both the first 4 h and the first 24 h, with an average decline of 8 ± 7 mmHg in the first 4 h and 10 ± 5 mmHg in the first 24 h. Furthermore, a moderately elevated PAWP did not inhibit alveolar fluid clearance. A total of 9 of 11 patients had an initial PAWP >18 mmHg; of these 9 patients, 7 (78%) had intact alveolar fluid clearance with a mean clearance rate of 8 ± 7%/h.

Effect of mechanical ventilation on alveolar fluid clearance. To determine whether the mode of mechanical ventilation was associated with rate of alveolar fluid clearance, the average PEEP and tidal volume per kilogram during the 4-h sampling period were compared between the group of patients with intact alveolar fluid clearance and those with no detectable alveolar fluid clearance. As shown in Fig. 4, there was no difference in level of PEEP or tidal volume between the two groups. Similarly, there was no difference between the two groups in the use of volume-controlled vs. pressure-controlled ventilation (data not shown).


View larger version (13K):
[in this window]
[in a new window]
  		Fig. 4.   Neither level of positive end-expiratory pressure (A) nor tidal volume/kg (B) was different between patients with intact alveolar fluid clearance and patients with impaired alveolar fluid clearance. Data are means ± SD. P = not significant between intact and impaired clearance groups.

SAPS II and acidosis. Interestingly, mean arterial pH at the time of edema fluid sampling was 7.21 ± 0.2 in the group with impaired clearance compared with 7.31 ± 0.1 in the group with intact alveolar fluid clearance (P = 0.02). In addition, SAPS II values were significantly higher in the patients with impaired clearance (50 ± 18) compared to those with intact clearance (40 ± 16, P = 0.04). Because both low arterial pH and a high SAPS II may be markers of poor systemic perfusion, the lowest systolic blood pressure in the first 24 h was also compared with alveolar fluid clearance. In the group with impaired alveolar fluid clearance, the lowest systolic blood pressure was 106 ± 50 mmHg, compared with 98 ± 39 mmHg in the group with intact clearance (P = 0.54).

Mortality, duration of mechanical ventilation, and improvement in oxygenation. Relevant clinical outcomes are summarized in Table 6. The group of patients with intact alveolar fluid clearance had an in-hospital death rate of 26% compared with 44% in the group with impaired clearance, a difference of 18%. Because this finding did not reach statistical significance (P = 0.20), a power calculation was done; this study had a power of only 0.25 to detect the observed 18% reduction in mortality. Interestingly, the median duration of unassisted ventilation in the intact clearance group was almost three times longer than in the group of patients with impaired alveolar fluid clearance, with a median of 23 days vs. 8 days in the impaired clearance group (P = 0.10). In addition, greater improvements in oxygenation at both 4 and 24 h were observed in the group with intact alveolar fluid clearance. The improvement in alveolar-arterial oxygen difference at 24 h achieved statistical significance (P = 0.03).

                              
View this table:
[in this window]
[in a new window]
  		Table 6.   Comparison of outcomes in patients with intact vs. impaired alveolar fluid clearance


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

This study investigated alveolar fluid clearance in 65 mechanically ventilated patients with severe hydrostatic pulmonary edema. The primary objectives were to characterize human alveolar fluid clearance in the absence of injury to the alveolar epithelium and to investigate the role for catecholamine-dependent and -independent mechanisms in regulating alveolar fluid clearance in patients with hydrostatic pulmonary edema. The major findings can be summarized as follows. First, alveolar fluid clearance was intact in 75% of patients (Fig. 1). Maximal clearance (>= 14%/h) occurred in 38% of the patients (Fig. 2), indicating that, in the absence of alveolar epithelial injury, alveolar fluid clearance in humans can be very rapid. Second, elevated levels of endogenous catecholamines (epinephrine) were not associated with intact alveolar fluid clearance (Fig. 3), although administration of inhaled beta -agonist had an 85% positive predictive value for the presence of intact alveolar fluid clearance. Third, patients with chronic congestive heart failure tended to have faster rates of alveolar fluid clearance than those with pulmonary edema from other causes, indicating that chronic left atrial hypertension may upregulate alveolar fluid clearance (Table 2). Fourth, the degree of cardiac dysfunction was not clearly related to the capacity of the alveolar epithelium to remove alveolar edema fluid (Table 5). Fifth, a low arterial pH was associated with impaired alveolar fluid clearance, suggesting that systemic hypoperfusion may contribute to downregulation of alveolar fluid clearance in some patients with severe hydrostatic pulmonary edema. Finally, intact alveolar fluid clearance was associated with more rapid improvement in arterial oxygenation, with trends toward improved mortality and shorter duration of mechanical ventilation (Table 6).

Rate of alveolar fluid clearance in hydrostatic pulmonary edema. The first objective of the study was to characterize alveolar fluid clearance in patients with alveolar edema without overt evidence of injury to the alveolar epithelium. Patients with a hydrostatic cause for pulmonary edema were selected on the basis of strict clinical criteria; those with any clinical evidence of a cause for acute lung injury such as sepsis, aspiration, or pneumonia were excluded. Prior morphometric studies in acute experimental hydrostatic edema have shown no clear evidence of alveolar epithelial injury (in the absence of severe hydrostatic stress), with normal cell morphology, intact intercellular junctions, and intact alveolar epithelial and endothelial basement membranes (12, 14). In the present study, the mean initial edema fluid-to-plasma protein ratio, a measure of permeability of the alveolar capillary barrier (27, 39), was 0.47 ± 0.10, confirming that pulmonary edema occurred in this patient population as a result of hydrostatic forces rather than increased permeability. Remarkably, 75% of the patients had intact alveolar fluid clearance despite the presence of severe pulmonary edema. The group of patients with maximal alveolar fluid clearance comprised over one-third (38%) of the population and had a mean alveolar fluid clearance rate of 25 ± 12%/h. Furthermore, there was no evidence that a change in plasma protein contributed to the clearance of fluid from the alveolar space. Thus the maximal capacity of the uninjured human lung to actively transport alveolar edema fluid is much higher than previously estimated from ex vivo studies, confirming that ex vivo studies can significantly underestimate in vivo alveolar fluid clearance rates.

The method used to measure alveolar fluid clearance in the present study has been validated in a prior clinical study (39). However, because the measurement of alveolar fluid clearance is made in the presence of preexisting pulmonary edema, rather than by instilling a protein-containing solution into the lung as is done in animal studies, there are some limitations to this approach. The most obvious limitation is that the total volume of pulmonary edema is not known. Thus, although the clearance is expressed as the percentage of total edema fluid cleared per hour, the absolute volume cleared cannot be calculated and could differ significantly among patients. However, there are several reasons to believe that this is not a major factor in interpreting our results. First, all patients had severe pulmonary edema requiring mechanical ventilation, yet not so severe as to preclude a minimal level of life-sustaining oxygenation. This limits to some degree the possible range for the volume of pulmonary edema. Second, indexes of lung edema, such as initial oxygenation or the initial LIS, did not differentiate between patients with intact or impaired alveolar fluid clearance (Table 2). Third, the rate of clearance, expressed as the percentage of total edema cleared per hour, correlated well with other physiological indexes of the resolution of pulmonary edema, such as improvement in oxygenation (Table 6). For these reasons, we believe that the method used to calculated the rate of alveolar fluid clearance has value, particularly when employed in the context of clinically relevant physiological indexes.

A second possible limitation of the method used to measure alveolar fluid clearance is that sampling of pulmonary edema fluid is done in a blind fashion, and serial samples could originate from different parts of the lung. Although this is a theoretical concern, radiographic studies have shown that the sampling catheter goes to the right lower lobe 90% of the time. Furthermore, when serial edema fluid samples were taken at intervals of only 1 or 2 min in a small subset of patients, similar protein concentrations were measured.

Comparison of human alveolar fluid clearance with that in other species. The present study allows a preliminary comparison of maximal rates of alveolar fluid clearance measured in humans with hydrostatic pulmonary edema with maximal rates measured in other species. However, it should be noted, as discussed above, that the method used to calculate clearance in human patients differs slightly from that used in animal studies. Maximal alveolar fluid clearance in humans, as approximated in this study, was 25 ± 12%/h in the 38% of patients with maximal clearance. This rapid clearance is in a range similar to the published maximally stimulated rates for rats (35 ± 8%/h) and mice (48 ± 5%/h) but much higher than those reported in larger animals such as dogs (6 ± 2%/h), sheep (13 ± 4%/h), and rabbits (13 ± 4%/h) (5, 6, 30, 44, 56). This finding lends validity to the use of rats and mice for investigations of alveolar fluid clearance. The possibility of expanded use of mice to study alveolar epithelial fluid transport is particularly important given the availability of murine transgenic models (25).

In addition to providing the first characterization of alveolar fluid clearance in a large patient population with intact alveolar epithelium, this study also provides a well-characterized control population for future studies of alveolar fluid clearance in human acute lung injury. For obvious reasons, pulmonary edema fluid cannot be collected from normal control patients; this large group of hydrostatic pulmonary edema patients provides an ideal control group, with an LIS comparable to those in patients with acute lung injury (61) and a relatively high SAPS II indicative of a high severity of illness.

Contribution of endogenous catecholamines. The second objective of this study was to investigate the role of catecholamines in upregulating alveolar fluid clearance in humans. Endogenous epinephrine has been shown to upregulate alveolar fluid clearance under several experimental conditions, including hypovolemic and septic shock in rats (41, 49), neurogenic pulmonary edema in dogs (32), fetal lung fluid clearance in newborn guinea pigs (20), and moderate left atrial hypertension in sheep (9). Furthermore, increased levels of circulating catecholamines have been widely reported in both acute myocardial infarction and congestive heart failure (46, 52). We therefore hypothesized that endogenous epinephrine might play an important role in upregulating alveolar fluid clearance in humans with hydrostatic pulmonary edema. However, although plasma epinephrine concentrations were >300 pg/ml in 73% of patients (twice the level of normal control subjects), there was no correlation between plasma epinephrine levels and the rate of alveolar fluid clearance (Fig. 3). In fact, the highest plasma epinephrine concentration (19,170 pg/ml) was measured in a patient with no net alveolar fluid clearance, and two of the patients with maximal alveolar fluid clearance had normal plasma epinephrine levels. Why did endogenous epinephrine levels fail to correlate with alveolar fluid clearance? A recent experimental study in rats with sustained hypovolemic shock reported that production of oxidants prevented the normal upregulation of alveolar fluid clearance by beta -agonists (40). Oxidant production may also be an important modulator in humans with hydrostatic pulmonary edema, particularly those with shock, as discussed later. Thus, although these results do not definitively exclude a contribution of endogenous catecholamines to regulation of human in vivo alveolar fluid clearance, other mechanisms must also play an important role in both the upregulation and downregulation of alveolar fluid clearance in the clinical setting.

Exogenous catecholamines. The potential for exogenous catecholamines to stimulate alveolar fluid clearance in patients with pulmonary edema has particular clinical relevance because both intravenous and inhaled beta -agonists are already in wide clinical use for other purposes. A number of experimental studies have shown that exogenous beta 2-agonists, including terbutaline, salmeterol, epinephrine, and dobutamine, can stimulate alveolar fluid clearance (5, 6, 13, 35, 53, 59). In the present study, there was no correlation between intravenously administered dobutamine and alveolar fluid clearance. An inhaled beta -agonist (albuterol) was administered to only 20% of the patients (n = 13). Interestingly, twice as many patients in the group with intact alveolar fluid clearance received the beta -agonist as in the group with no alveolar fluid clearance, and the administration of aerosolized beta -agonist therapy had an 85% positive predictive value for the preservation of alveolar fluid clearance. However, given the small number of patients who received albuterol, these differences did not reach statistical significance. A power calculation showed that the study had a power of only 0.10 to detect the difference observed. When exogenous beta -agonists were considered in conjunction with endogenous levels of epinephrine, there was still no correlation with alveolar fluid clearance. A randomized clinical trial of inhaled beta -agonists will be required to definitively test the hypothesis that inhaled beta -agonists may stimulate alveolar fluid clearance in the resolution phase of hydrostatic pulmonary edema. Furthermore, on the basis of experimental data in the sheep and the ex vivo human lung (9, 53), a lipophilic beta -agonist such as salmeterol might have more benefit than the nonlipophilic beta -agonist albuterol, which the patients in this study received.

Role of catecholamine-independent mechanisms in alveolar fluid clearance. The lack of correlation between alveolar fluid clearance and either endogenous or exogenous catecholamines suggests that other catecholamine-independent mechanisms may be important in both the stimulation and inhibition of alveolar fluid clearance. A number of other pharmacological agents in addition to beta -agonists have been found to enhance alveolar fluid clearance, including glucocorticoids (21) and dopamine (3). Both of these agents act by non-beta -agonist mechanisms. However, neither was associated with a significant increase in alveolar fluid clearance in the study population. Similarly, neither the administration of furosemide, a blocker of NaCl cotransport, nor a variety of antiarrhythmics affected alveolar fluid clearance. However, digoxin, a ouabain derivative, was significantly associated with intact alveolar fluid clearance. This finding was unexpected because studies of both acute and chronic administration of digoxin in patients with chronic heart failure have reported that digoxin downregulates circulating catecholamines. A drop in catecholamine levels might be expected to diminish alveolar fluid clearance (19, 31). Furthermore, ouabain is known to inhibit Na+-K+-ATPase activity, the major mechanism for sodium transport across the basolateral membrane of alveolar epithelial type II cells (49). However, it seems likely that the levels of digoxin administered clinically are insufficient to block the alveolar epithelial Na+-K+-ATPase. In prior sheep studies levels of ouabain that could inhibit alveolar fluid clearance uniformly caused fatal arrhythmias (55, 57). In fact, the finding that digoxin administration was associated with intact rather than impaired alveolar fluid clearance provides reassurance that clinical administration of digoxin does not inhibit alveolar fluid clearance. Further analysis revealed that 9 of the 11 patients who received digoxin had chronic congestive heart failure; as discussed below, this appears to be the most likely explanation for the observed association of digoxin with intact alveolar fluid clearance, rather than a direct pharmacological effect.

Overall, the subgroup of patients with an acute exacerbation of chronic congestive heart failure tended to have higher rates of alveolar fluid clearance than those with a more acute cause of hydrostatic pulmonary edema, such as acute myocardial infarction or acute volume overload (Table 2). This finding was not related to left ventricular ejection fraction or levels of central venous or left atrial pressure. There are several mechanisms by which chronically elevated left ventricular pressure could upregulate alveolar fluid clearance. First, patients with chronic congestive heart failure have been shown to have high levels of circulating catecholamines (46). However, endogenous catecholamines were not independently associated with alveolar fluid clearance in the present study. Second, levels of circulating TNF-alpha , a cytokine that has been shown to upregulate alveolar fluid clearance experimentally (50), are elevated in patients with chronic congestive heart failure (16, 34). Although TNF-alpha was not measured in this study, a previous investigation has detected TNF-alpha in the pulmonary edema fluid of patients with hydrostatic pulmonary edema (37), providing evidence that the proinflammatory cytokine TNF-alpha may be present in the alveolar compartment even in the setting of pure hydrostatic pulmonary edema. A third mechanism is alveolar epithelial type II cell proliferation, which has been shown to upregulate alveolar fluid clearance in several experimental models (22, 62). Interestingly, alveolar epithelial type II cell hyperplasia has been reported in an ultrastructural study of lung tissue from patients with chronic congestive heart failure (2). Furthermore, a recent study found detectable levels of potent alveolar type II cell mitogens, hepatocyte and keratinocyte growth factors, in the pulmonary edema fluid of patients with hydrostatic pulmonary edema (61). Finally, histological studies of lung biopsy tissue from patients with chronically elevated left atrial pressure have shown dilated lymphatics (28). Enhanced fluid clearance by the lymphatics from the interstitium could both inhibit the formation of alveolar edema and promote alveolar fluid clearance by removing transported edema fluid from the lung interstitium more rapidly.

An additional catecholamine-independent mechanism that could downregulate alveolar fluid clearance is the production of atrial natriuretic factor. Atrial natriuretic factor decreased alveolar epithelial sodium transport in vitro (9) and in an isolated perfused rat lung model (45). We did not measure atrial natriuretic factor in these studies. However, the levels of atrial natriuretic factor measured in a recent study of acute moderate left atrial hypertension in sheep were at least one order of magnitude below the levels needed for an in vitro effect in cultured alveolar type II cells (9).

Effect of hemodynamics and mechanical ventilation on alveolar fluid clearance. Although the number of patients in the study with invasive hemodynamic monitoring was relatively small, several conclusions can be drawn. An important question, previously addressed only in one animal study, is whether the presence of moderate left atrial hypertension would reduce the rate of alveolar fluid clearance. One might postulate that left atrial hypertension, by increasing net transvascular fluid flux, would lead to a decrease in net alveolar fluid clearance. However, alveolar fluid clearance in sheep was unchanged in the setting of moderate left atrial hypertension (9). The present study extends this finding to humans. Of the patients with PAWP >18 mmHg during the time of pulmonary edema fluid sampling, 7 of 9, or 78%, had intact alveolar fluid clearance. Thus, in humans, as was reported in sheep, moderately elevated left atrial pressure was not found to decrease alveolar fluid clearance. In addition, a decline in PAWP or CVP was not predictive of the presence or absence of alveolar fluid clearance (Table 5). Finally, left ventricular ejection fraction, a measure of left ventricular function, was not associated with impaired or intact alveolar fluid clearance (Table 5).

In patients with hydrostatic pulmonary edema, mechanical ventilation could have either beneficial or detrimental effects on alveolar fluid clearance. Beneficial effects of positive pressure ventilation that might indirectly increase alveolar fluid clearance include decreased preload and afterload and decreased myocardial oxygen consumption due to decreased work of breathing (7). These factors could reduce the formation of pulmonary edema and thus promote net alveolar fluid clearance. There is no experimental evidence that positive pressure ventilation itself hastens alveolar fluid clearance beyond the indirect physiological changes described above. Experimentally, mechanical ventilation at high tidal volumes can promote lung injury (60); injury to the alveolar epithelium might impair alveolar fluid clearance (39). In addition, high levels of PEEP may raise CVP, inhibiting lung lymphatic drainage from the interstitium and thereby limiting alveolar fluid clearance. In this study, no difference in levels of PEEP or tidal volume per kilogram was found between patients with intact and impaired alveolar fluid clearance (Fig. 4). Thus there is no evidence to suggest that mechanical ventilation had a major impact on the rate of alveolar fluid clearance in the absence of lung injury.

Effect of pH and SAPS II on alveolar fluid clearance. Although alveolar fluid clearance was intact in the majority of patients in this study, 25% of the patients had no net alveolar fluid clearance in the first 4 h after endotracheal intubation. This impairment of alveolar fluid clearance was not explained by hemodynamic factors or cardiovascular function. However, two factors were significantly associated with impaired alveolar fluid clearance: arterial pH and SAPS II. Arterial pH at the time of initial edema fluid sampling was significantly lower (7.21 ± 0.2) in the patients with no alveolar fluid clearance compared with those with intact clearance (7.31 ± 0.1). It seems unlikely that low pH, in and of itself, directly inhibited alveolar epithelial fluid transport. In an in situ, nonperfused sheep model, alveolar fluid clearance was rapid despite a pH of <7.0 (55). A more plausible explanation is that low arterial pH was a marker of shock. Interestingly, in a rat model of hypovolemic shock, neither endogenous nor exogenous beta -agonists produced the expected increase in alveolar fluid clearance (40). However, pretreatment with antioxidants restored the normal response to beta -agonist stimulation. Furthermore, in vitro, the production of oxidants decreased sodium uptake by alveolar epithelial type II cells and decreased the function of the apical epithelial sodium channel (29, 36). Thus the lower arterial pH in the patients with impaired alveolar fluid clearance may be a marker of oxidant-mediated decrease in alveolar fluid clearance. Interestingly, nitrate and nitrite, markers of oxidant production, have recently been reported in the pulmonary edema fluid of patients with hydrostatic pulmonary edema, suggesting that oxidant production may occur even in the absence of overt lung injury (64). The association of a higher SAPS II value with impaired alveolar fluid clearance may also be a marker of more severe shock. Several components of the SAPS II relate directly to systemic perfusion, including heart rate, blood pressure, urine output, blood urea nitrogen, and serum bicarbonate.

Mortality, duration of mechanical ventilation, and oxygenation. Intact alveolar fluid clearance was associated with significantly improved oxygenation at 24 h compared with the patients with impaired clearance. There was also a strong trend toward shorter duration of mechanical ventilation in the group with intact clearance, with a median number of days of unassisted ventilation of 23 vs. 8 days in the patients with impaired alveolar fluid clearance. These findings suggest that intact alveolar fluid clearance is critical to recovery from hydrostatic pulmonary edema. Interestingly, there was even a trend toward a decrease in hospital mortality in the patients with intact alveolar fluid clearance, with 18% lower mortality in the group with intact alveolar fluid clearance. However, it seems unlikely that alveolar fluid clearance was the primary determinant of mortality, particularly because the severity of illness, as measured by SAPS II, was significantly higher in the group of patients with impaired alveolar fluid clearance. Nevertheless, the observed association of both improved pulmonary outcomes and the trend toward decreased mortality with intact alveolar fluid clearance emphasizes the critical role the alveolar epithelium plays in the resolution of pulmonary edema.

Conclusions. In conclusion, there were several novel findings in this study of alveolar fluid clearance in severe hydrostatic pulmonary edema. First, this study demonstrates that techniques for measuring alveolar fluid clearance can be applied to a large number of critically ill patients. Despite severe alveolar flooding in these patients, alveolar fluid clearance was intact in the majority (75%) of patients. Several factors were identified that may regulate clinical alveolar fluid clearance. Although endogenous epinephrine levels were not associated with the rate of alveolar fluid clearance, administration of an inhaled beta -agonist had an 85% positive predictive value for intact alveolar fluid clearance. Chronic congestive heart failure also was associated with more rapid alveolar fluid clearance. Impaired alveolar fluid clearance was associated with a lower arterial pH and a higher severity of illness as measured by SAPS II. Finally, intact alveolar fluid clearance was associated with improved physiological and clinical outcomes. The critical importance of alveolar fluid clearance to the resolution of hydrostatic pulmonary edema emphasizes the need for further studies in humans to better define the mechanisms that modulate alveolar fluid clearance in critically ill patients.


    ACKNOWLEDGEMENTS

G. M. Verghese and L. B. Ware made equal contributions to this study.


    FOOTNOTES

The authors thank Gunnard Modin for assistance with the statistical analysis, Rich Kallet and Brian Daniel for help in collecting pulmonary edema fluid, and Dr. Reid Thompson for helpful review of the manuscript.

This work was supported by National Heart, Lung, and Blood Institute Grant HL-51856.

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Address for reprint requests and other correspondence: L. B. Ware, Cardiovascular Research Institute, Box 0130, 505 Parnassus Ave., Univ. of California, San Francisco, San Francisco, CA 94143-0130 (E-mail: lware{at}itsa.ucsf.edu).

Received 5 February 1999; accepted in final form 9 June 1999.


    REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1.   Artigas, A., G. Bernard, J. Carlet, D. Dreyfuss, L. Gattinoni, L. Hudson, M. Lamy, J. Marini, M. Matthay, and M. Pinsky. The American-European consensus conference on ARDS, part 2. Ventilatory, pharmacologic, supportive therapy, study design strategies and issues related to recovery and remodeling. Am. J. Respir. Crit. Care Med. 157: 1332-1347, 1998[Abstract/Free Full Text].

2.   Bachofen, H., M. Bachofen, and E. R. Weibel. Ultrastructural aspects of pulmonary edema. J. Thorac. Imaging 3: 1-7, 1988[Medline].

3.   Barnard, M., W. Olivera, D. Rutschman, A. Bertorello, A. Katz, and J. Sznajder. Dopamine stimulates sodium transport and liquid clearance in rat lung epithelium. Am. J. Respir. Crit. Care Med. 156: 709-714, 1997[Abstract/Free Full Text].

4.   Basset, G., C. Crone, and G. Saumon. Significance of active ion transport in transalveolar water absorption: a study on isolated rat lung. J. Physiol. (Lond.) 384: 311-324, 1987[Abstract/Free Full Text].

5.   Berthiaume, Y., V. C. Broaddus, M. A. Gropper, T. Tanita, and M. A. Matthay. Alveolar liquid and protein clearance from normal dog lungs. J. Appl. Physiol. 65: 585-593, 1988[Abstract/Free Full Text].

6.   Berthiaume, Y., N. C. Staub, and M. A. Matthay. Beta-adrenergic agonists increase lung liquid clearance in anesthetized sheep. J. Clin. Invest. 79: 335-343, 1987.

7.   Berthiaume, Y., L. B. Ware, and M. A. Matthay. Treatment of acute pulmonary edema and the acute respiratory distress syndrome. In: Pulmonary Edema, edited by M. A. Matthay, and D. H. Ingbar. New York: Dekker, 1998, p. 575-631.

8.   Borok, Z., A. Hami, S. Danto, R. Lubman, K. Kim, and E. Crandall. Effects of EGF on alveolar epithelial junctional permeability and sodium transport. Am. J. Physiol. 270 (Lung Cell. Mol. Physiol. 14): L559-L565, 1996[Abstract/Free Full Text].

9.   Campbell, A. R., H. G. Folkesson, Y. Berthiaume, J. Gutkowska, S. Suzuki, and M. A. Matthay. Alveolar epithelial fluid clearance persists in the presence of moderate left atrial hypertension in sheep. J. Appl. Physiol. 86: 139-151, 1999[Abstract/Free Full Text].

10.   Chesnutt, A., M. Matthay, F. Tibayan, and J. Clark. Early detection of type III procollagen peptide in acute lung injury. Pathogenetic and prognostic significance. Am. J. Respir. Crit. Care Med. 156: 840-845, 1997[Abstract/Free Full Text].

11.   Clark, J. G., J. A. Milberg, K. P. Steinberg, and L. D. Hudson. Type III procollagen peptide in the adult respiratory distress syndrome. Association of increased peptide levels in bronchoalveolar lavage fluid with increased risk of death. Ann. Intern. Med. 122: 17-23, 1995[Abstract/Free Full Text].

12.   Cotrell, T. S., R. Levine, R. Senior, J. Wiener, D. Spiro, and A. P. Fishman. Electron microscopic alterations at the alveolar level in pulmonary edema. Circ. Res. 21: 783-797, 1967[Abstract].

13.   Crandall, E. D., T. A. Heming, R. L. Palombo, and B. E. Goodman. Effects of terbutaline on sodium transport in isolated perfused rat lung. J. Appl. Physiol. 60: 289-294, 1986[Abstract/Free Full Text].

14.   DeFouw, D. O., and P. B. Berendsen. Morphological changes in isolated perfused dog lungs after acute hydrostatic edema. Circ. Res. 43: 72-82, 1978[Abstract].

15.   Dobbs, L. G., R. Gonzalez, M. A. Matthay, E. P. Carter, L. Allen, and A. S. Verkman. Highly water-permeable type I alveolar epithelial cells confer high water permeability between the airspace and vasculature in rat lung. Proc. Natl. Acad. Sci. USA 95: 2991-2996, 1998[Abstract/Free Full Text].

16.   Dutka, D. P., J. S. Elborn, F. Delamere, D. J. Shale, and G. K. Morris. Tumor necrosis factor alpha in severe congestive cardiac failure. Br. Heart J. 70: 141-143, 1993[Abstract/Free Full Text].

17.   Effros, R. M., A. Hacker, P. Silverman, and J. Hukkanen. Protein concentrations have little effect on reabsorption of fluid from isolated rat lungs. J. Appl. Physiol. 70: 416-422, 1991[Abstract/Free Full Text].

18.   Fein, A., R. F. Grossman, J. G. Jones, E. Overland, L. Pitts, J. F. Murrary, and N. C. Staub. The value of edema protein measurements in patients with pulmonary edema. Am. J. Med. 67: 32-39, 1979[Medline].

19.   Ferguson, D. W. Digitalis and neurohumoral abnormalities in heart failure and implications for therapy. Am. J. Cardiol. 69: 24G-32G, 1992[Medline].

20.   Finley, N., A. Norlin, D. Baines, and H. G. Folkesson. Alveolar epithelial fluid clearance is mediated by endogenous catecholamines at birth in guinea pigs. J. Clin. Invest. 101: 972-981, 1998[Medline].

21.   Folkesson, H., and M. Matthay. Dexamethasone upregulates epithelial liquid clearance in anesthetized ventilated rats (Abstract). FASEB J. 11: A561, 1997.

22.   Folkesson, H. G., G. Nitenberg, B. L. Oliver, C. Jayr, K. H. Albertine, and M. A. Matthay. Upregulation of alveolar epithelial fluid transport after subacute lung injury in rats from bleomycin. Am. J. Physiol. 275 (Lung Cell. Mol. Physiol. 19): L478-L490, 1998[Abstract/Free Full Text].

23.   Folkesson, H. G., J. F. Pittet, G. Nitenberg, and M. A. Matthay. Transforming growth factor-alpha increases alveolar liquid clearance in anesthetized, ventilated rats. Am. J. Physiol. 271 (Lung Cell. Mol. Physiol. 15): L236-L244, 1996[Abstract/Free Full Text].

24.   Fukuda, N., H. G. Folkesson, Y. Wang, and M. A. Matthay. Basal alveolar epithelial fluid clearance in ventilated mice is very fast: results of a novel in vivo murine model (Abstract). Am. J. Respir. Crit. Care Med. 159: A292, 1999.

25.   Garat, C., E. Carter, and M. Matthay. New in situ mouse model to quantify alveolar epithelial fluid clearance. J. Appl. Physiol. 84: 1763-1767, 1998[Abstract/Free Full Text].

26.   Goodman, B. E., R. S. Fleischer, and E. D. Crandall. Evidence for active Na+ transport by cultured monolayers of pulmonary alveolar epithelial cells. Am. J. Physiol. 245 (Cell Physiol. 14): C78-C83, 1983[Abstract/Free Full Text].

27.   Hastings, R. H., M. Grady, T. Sakuma, and M. A. Matthay. Clearance of different-sized proteins from the alveolar space in humans and rabbits. J. Appl. Physiol. 73: 1310-1316, 1992[Abstract/Free Full Text].

28.   Heath, D., and P. Hicken. The relation between left atrial hypertension and lymphatic distension in lung biopsies. Thorax 15: 54-58, 1960.

29.   Hu, P., H. Ischiropoulos, J. S. Beckman, and S. Matalon. Peroxynitrite inhibition of oxygen consumption and sodium transport in alveolar type II cells. Am. J. Physiol. 266 (Lung Cell. Mol. Physiol. 10): L628-L634, 1994[Abstract/Free Full Text].

30.   Jayr, C., C. Garat, M. Meignan, J. F. Pittet, M. Zelter, and M. A. Matthay. Alveolar liquid and protein clearance in anesthetized, ventilated rats. J. Appl. Physiol. 76: 2626-2642, 1994.

31.   Krum, H., J. T. J. Bigger, R. L. Goldsmith, and M. Packer. Effect of long-term digoxin therapy on autonomic function in patients with chronic heart failure. J. Am. Coll. Cardiol. 25: 289-294, 1995[Abstract].

32.   Lane, S. M., K. C. Maender, N. E. Awender, and M. B. Maron. Adrenal epinephrine increased alveolar liquid clearance in neurogenic pulmonary edema. Am. J. Respir. Crit. Care Med. 158: 760-768, 1998[Abstract/Free Full Text].

33.   LeGall, J., S. Lemeshow, and F. Saulnier. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA 270: 2957-2963, 1993[Abstract].

34.   Levine, B., J. Kalman, L. Mayer, H. Fillit, and M. Packer. Elevated levels of circulating tumor necrosis factor in severe chronic heart failure. N. Engl. J. Med. 323: 236-241, 1990[Abstract].

35.   Maron, M. B. Dose-response relationship between plasma epinephrine concentration and alveolar liquid clearance in dogs. J. Appl. Physiol. 85: 1702-1707, 1998[Abstract/Free Full Text].

36.   Matalon, S., D. J. Benos, and R. M. Jackson. Biophysical and molecular properties of amiloride-inhibitable sodium channels in alveolar epithelial cells. Am. J. Physiol. 271 (Lung Cell. Mol. Physiol. 15): L1-L22, 1996[Abstract/Free Full Text].

37.   Matthay, M. A., H. G. Folkesson, A. Campagna, and F. Kheradmand. Alveolar epithelial barrier and acute lung injury. New Horizons 1: 613-622, 1993[Medline].

38.   Matthay, M. A., H. G. Folkesson, and A. S. Verkman. Salt and water transport across alveolar and distal airway epithelia in the adult lung. Am. J. Physiol. 270 (Lung Cell. Mol. Physiol. 14): L487-L503, 1996[Abstract/Free Full Text].

39.   Matthay, M. A., and J. P. Wiener-Kronish. Intact epithelial barrier function is critical for the resolution of alveolar edema in humans. Am. Rev. Respir. Dis. 142: 1250-1257, 1990[Medline].

40.   Modelska, K., M. A. Matthay, L. A. Brown, L. Deutsch, L. Lu, and J. F. Pittet. Inhibition of beta -adrenergic-dependent alveolar epithelial clearance by oxidant mechanisms after hemorrhagic shock. Am. J. Physiol. 276 ( Lung Cell. Mol. Physiol. 20): L844-857, 1999[Abstract/Free Full Text].

41.   Modelska, K., M. A. Matthay, M. C. McElroy, and J. F. Pittet. Upregulation of alveolar liquid clearance after fluid resuscitation for hemorrhagic shock in rats. Am. J. Physiol. 273 (Lung Cell. Mol. Physiol. 17): L305-L314, 1997[Abstract/Free Full Text].

42.   Moyer, T., N. Jiang, G. Tyce, and S. Sheps. Analysis of urinary catecholamines by liquid chromatography with amperometric detection: methodology and clinical interpretation of results. Clin. Chem. 25: 256-263, 1979[Abstract/Free Full Text].

43.   Murray, J., M. Matthay, J. Luce, and M. Flick. An expanded definition of the adult respiratory distress syndrom