help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by UMBERTO MEDURI, G.
Right arrow Articles by POSTLETHWAITE, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by UMBERTO MEDURI, G.
Right arrow Articles by POSTLETHWAITE, A.
Am. J. Respir. Crit. Care Med., Volume 158, Number 5, November 1998, 1432-1441

Procollagen Types I and III Aminoterminal Propeptide Levels during Acute Respiratory Distress Syndrome and in Response to Methylprednisolone Treatment

G. UMBERTO MEDURI, ELIZABETH A. TOLLEY, ALBERT CHINN, FRANKIE STENTZ, and ARNOLD POSTLETHWAITE

Departments of Medicine and Preventive Medicine, and Memphis Lung Research Program, University of Tennessee, Baptist Memorial Hospitals, and Veterans Affairs Medical Center, Memphis, Tennessee

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Ineffective lung repair in patients with unresolving acute respiratory distress syndrome (ARDS) is accompanied by progressive fibroproliferation, inability to improve lung injury score (LIS), progressive multiple organ dysfunction syndrome (MODS), and an unfavorable outcome. Our aim was to investigate the relationship between fibrogenesis, pulmonary and extrapulmonary organ dysfunction, and outcome during the natural course of ARDS and in response to prolonged methylprednisolone treatment. We investigated 29 patients with ARDS. We obtained serial measurements of plasma and BAL procollagen aminoterminal propeptide type I (PINP) and type III (PIIINP) levels and components of the lung injury score (LIS) and MODS score. A reduction in LIS greater than one point from day 1 to day 7 of ARDS divided patients in improvers (group 1, n = 7) and nonimprovers (n = 22). Nonimprovers included those who were recruited (day 9 ± 3 of ARDS) into a prospective, randomized, double-blind, placebo-controlled trial investigating prolonged methylprednisolone therapy in unresolving ARDS (group 2, n = 17), and those who died (all by day 10 of ARDS) prior to meeting eligibility criteria for the randomized trial (group 3, n = 5). On day 1 of ARDS, plasma PINP or PIIINP levels were elevated in all patients. By day 7 of ARDS, mean plasma PINP or PIIINP levels were unchanged in group 1 but increased significantly in group 2 (p = 0.0002) and group 3 (p = 0.03). On day 7, patients with plasma PINP levels less than 100 ng/ml were 2.5 times more likely to survive (95% CI: 0.855-7.314), and patients with plasma PIIINP levels greater than 25 ng/ml were nine times more likely to die (95% CI: 1.418-55.556). In group 2, patients taking placebo (n = 6) had no change in plasma PINP or PIIINP levels over time, while patients treated with methylprednisolone (n = 11) had a rapid and sustained reduction in mean plasma and bronchoalveolar lavage (BAL) PINP and PIIINP levels. By day 3 of treatment, mean plasma PINP and PIIINP levels (ng/ml) decreased from 100 ± 9 to 45 ± 8 (p = 0.0001) and 31 ± 3 to 12 ± 3 (p = 0.0008), respectively. After 8 to 15 d of methylprednisolone, mean BAL PINP and PIIINP levels (ng/ml) decreased from 63 ± 25 to 6 ± 23 (p = 0.002) and 42 ± 5 to 10 ± 3 (p = 0.003), respectively. Estimated partial correlation coefficients indicated that as plasma PINP and PIIINP levels decreased over the first 7 d of methylprednisolone treatment, positive end-expiratory pressure, creatinine, bilirubin, and temperature also decreased, while PaO2:FIO2 increased. In early ARDS, plasma PINP and PIIINP levels are elevated and continue to increase over time in those not improving. Among nonimprovers, those randomized to prolonged methylprednisolone treatment had a rapid and significant reduction in plasma and BAL aminoterminal propeptide levels and similar changes in lung injury and MODS scores. These findings provide additional evidence of an association between biological efficacy and physiologic response during prolonged methylprednisolone treatment of unresolving ARDS.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

The host defense response to an insult consists of an interactive network of simultaneously activated pathways---inflammation, coagulation, and tissue repair---that work in synergy to increase the host's chance of survival (1). Tissue repair consists of angiogenesis, epithelial growth, fibroblast migration and proliferation (fibroproliferation), and deposition of extracellular matrix (fibrogenesis). Cytokines of the interleukin-1 (IL-1) and tumor necrosis factor (TNF) families are proximal mediators of all key aspects of the host defense response, including inflammation (2), coagulation (3), and tissue repair (4).

A uniform connective tissue repair response has been demonstrated to occur in various organs following different types of injury (7). Pulmonary fibroproliferation in acute respiratory distress syndrome (ARDS) may share a common pathogenetic mechanism with other fibroproliferative diseases where, in the absence of inhibitory signals, the continued production of inflammatory mediators sustains connective tissue accumulation, which results in permanent alteration in tissue structure and function (maladaptive response) (8). Within this theoretical paradigm, defined by Bitterman and Henke (9) as the "linear response" to injury, fibrosis ensues when the inflammatory response is overexuberant and prolonged. In support of this hypothesis, we previously reported that: (1) on day 1 of ARDS and over time, nonsurvivors had significantly higher plasma and bronchoalveolar lavage (BAL) inflammatory cytokine (TNF-alpha , IL-beta , IL-6, and IL-8) levels than survivors (10, 11); and (2) lung histology in patients with unresolving ARDS (open lung biopsies obtained 8 to 22 d into ARDS) showed new injury to previously spared endothelial and epithelial surfaces occurring concurrently with an amplified reparative (coagulation and fibroproliferation with deposition of collagen) process over previously injured areas (airspaces, interstitium, respiratory bronchioles, and walls of the intra-acinar microvessels) (12).

IL-1 and TNF-alpha are directly fibrogenic (i.e., lead to increased deposition of new connective tissue matrix) by upregulating fibroblast functions (fibroblast growth, collagen synthesis, glycosaminoglycam synthesis, fibronectin synthesis, and production of collagenase and tissue inhibitor of metalloproteinases [TIMP]) (13). During tissue injury and repair, intact and fragmented matrix components are liberated into the extracellular fluid and circulation (7). Types I and III collagens are synthesized as procollagens, which have propeptides at both ends of their constituent alpha  chains (14). Newly synthesized procollagen is secreted by fibroblasts and other mesenchymal cells, after which specific endopeptidases cleave the aminoterminal and carboxyterminal propeptides from the remainder of the alpha  chains forming collagen molecules, and giving rise to N-terminal and C-terminal propeptides (15). Several studies have substantiated that elevated plasma procollagen aminoterminal propeptide levels reflect collagen synthesis at the site of disease, and may be used as a marker of the reparative process independent of etiology (7, 16). Prior ARDS studies have found that nonsurvivors have persistent elevation over time of circulating and BAL type III procollagen aminoterminal propeptide (PIIINP) levels (17).

Glucocorticoids have a broad range of inhibitory effects on the host defense response, including inhibition of cytokine transcription and cellular action, growth factor production by a variety of different cells, and inhibition of fibroblast proliferation and deposition of collagen (1, 21, 22). Recently we completed a prospective, randomized, double-blind, placebo-controlled trial and found that prolonged methylprednisolone therapy in patients with unresolving ARDS was associated with a rapid improvement in lung injury (23) and multiple organ dysfunction syndrome (MODS) scores (24) and a significant reduction in mortality (25).

The purpose of this study was to quantify the relationship among plasma and BAL type I procollagen aminoterminal propeptide (PINP) and PIIINP levels (markers of new collagen synthesis), pulmonary and extrapulmonary organ dysfunction, and outcome during the natural course of ARDS and in response to prolonged methylprednisolone treatment. The secondary purpose was to evaluate the relationship between PINP and PIIINP levels during progression of ARDS, and the clinical utility of propeptide levels for monitoring disease activity and response to treatment.

    METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Study Design

This study was conducted between October 1994 and November 1996 in the intensive care units of Baptist Memorial Medical Center and East hospitals, the Regional Medical Center, and the University of Tennessee Bowld Medical Center, all in Memphis, Tennessee. The study protocol was approved by each institutional review board, and informed consent was obtained prior to enrollment. An active effort was made to identify and recruit eligible patients with ARDS. Patients older than 18 yr were eligible if they met previously described ARDS consensus criteria (26). The ventilator management followed recently developed guidelines aimed at limiting plateau pressure to less than 35 cm H2O (27). Positive end-expiratory pressure (PEEP) was adjusted up to 20 cm H2O to maintain FIO2 less than or equal to 0.6.

The precipitating cause of ARDS was classified as either direct or indirect lung injury (10). The presence or absence of improvement in lung function (as defined by the lung injury score) by day 7 of ARDS was used to divide the patient sample into improvers and nonimprovers. On day 7 of mechanical ventilation, improvers (group 1) had a reduction in lung injury score (LIS) greater than one point from day 1 of ARDS (23). Nonimprovers included those who were recruited (day 9 ± 3 of ARDS) into the prospective, randomized, double-blind, placebo-controlled trial (group 2) (25), and those who died (all by day 10 of ARDS) prior to meeting eligibility criteria for the randomized trial (group 3). In patients with a new documented serious infection, appropriate antibiotic therapy for three or more days was required prior to entry in the randomized trial. Randomization was done on a 2:1 basis; 16 patients received methylprednisolone and eight received placebo. Data from 17 of the 24 randomized patients are reported in the present article. Methylprednisolone or placebo was given daily as intravenous push every 6 h (one-fourth of the daily dose) and changed to a single oral dose when oral intake was restored. A loading dose of 2 mg/kg was followed by 2 mg/kg/d from day 1 to 14, 1 mg/kg/d from day 15 to 21, 0.5 mg/kg/d from day 22 to 28, 0.25 mg/kg/d on days 29 and 30, and 0.125 mg/kg/d on days 31 and 32. According to the protocol, randomized patients failing to improve LIS by at least one point after 10 d of treatment were blindly crossed over to the alternative treatment arm.

During the course of ARDS, the following data were obtained on days 1, 3, 5, and 7 of ARDS, and on days 0 (day of randomization and prior to initiation of study drug), +3, +5, +7, +10, +14, and +21 of methylprednisolone therapy: components of the LIS while intubated (23), MODS score (24), systemic inflammatory response syndrome (SIRS) score (28), infections defined by strict criteria (29), ventilatory requirements, and hemodynamic variables in patients with pulmonary artery catheter in place. Resolution of individual organ dysfunction followed consensus guidelines (24). SIRS, sepsis, and shock were defined following consensus guidelines (28). Patients were classified as survivors if they were discharged alive from the ICU and did not require mechanical ventilation. Death was defined as associated with ARDS if it occurred while the patient was still intubated.

Plasma and Bronchoalveolar Lavage Collection

Blood samples were obtained on days 1, 3, 5, 7, and 10 of ARDS and, in patients recruited in the randomized study, on the day of randomization (day 0) and on days 3 (+3), 5 (+5), 7 (+7), and 10 (+10) of treatment. All blood samples were obtained from a central venous line or an antecubital venipuncture, placed in a vacutainer tube containing edetic acid, and transported to the laboratory for immediate processing. Blood samples were centrifuged at 1,500 g for 10 min, and plasma was aspirated and stored at -70° C. Bilateral BAL specimens were collected in patients without contraindication to bronchoscopy, following a previously reported methodology (11), on day 1 of ARDS and every 7 d or when the patient developed clinical manifestations suggesting ventilator-associated pneumonia (VAP), defined as the development of fever, purulent tracheal secretions, and new or progressive densities visible on chest radiograph. Right and left BAL samples were kept separate and processed immediately for quantitative bacterial culture and cytology following a previously described method (30). The remainder of each BAL specimen was centrifuged at 3,500 rpm for 10 min in a Beckman TJ6 centrifuge, and the supernatants were collected and stored at -70° C.

Procollagen Aminoterminal Propeptide Level Measurements

Plasma and BAL fluid specimens were analyzed for PIIINP and PINP content. The plasma or BAL which had been stored at -70° C was thawed immediately prior to performing the standards. PIIINP was assayed by double antibody radioimmunoassay using standards, PIIINP-specific antibodies, and other reagents supplied and utilized according to the methodology specifications of the manufacturer (Orion Diagnostica, Espoo, Finland, purchased through INCSTAR Corporation, Stillwater, MN). For PIIINP, 200 µl of specimen were used for the assay. Specimen values that read above the highest standard on the standard curve were diluted with zero standard, whereas those specimens with levels below the lowest detectable dose of 0.13 ng/L were concentrated 10-fold on a Savant Speed Vac concentrator and reassayed. The intra-assay and interassay variability was equal to or less than 4% and 5%, respectively. The normal range of serum levels of nonsmoking individuals was determined to be 1.6-4.0 ng/L. PINP was also assayed by a double antibody radioimmunoassay using standards, PINP specific antibodies, and other reagents supplied by and according to specifications of the same manufacturer (Orion). Specimen volume used for this assay was 50 µl. Specimens requiring dilution or concentration (lowest detectable dose was 1.5 ng/L) were diluted with the zero standard for the PINP assay or concentrated as for the PIIINP assay. The intra-assay variability and interassay variability were 4.5% and 5.5%, respectively. The normal range determined for serum levels of PINP in nonsmokers was 20-42 ng/L. Prior investigators have reported the results in units per ml (U/ml), and 1 U/ml corresponds to about 15 ng/ ml quantified from the original version of the assay (26).

Statistical Analysis

At the onset of ARDS (day 1), clinical variables were used to dichotomize patients; mean PINP and PIIINP values for patients in these categories were compared with Student's or Behrens-Fisher t tests. For each of the groups, changes in mean plasma PINP and PIIINP values on days 3, 5, and 7 were detected by using preplanned contrasts with the respective mean value at the onset of ARDS. For these contrasts, least-square means were estimated from repeated measures (split-plot) analysis of variance (ANOVA) (31), in which the effects of group and patient within group were designated as main plot effects and the effect of day and the interaction effect of day with group were the subplot effects. Contrasts were made with the residual mean square used as the denominator. In order to determine whether the profiles for mean plasma PINP and PIIINP values of these groups were the same as those for clinical variables serially measured during the first 7 d, similar analyses with a different estimate of the residual mean square were conducted. Finally, contrasts of mean values of survivors (excluding the patients who were randomized to the methylprednisolone treatment arm) and nonsurvivors with their respective baseline values were completed using similar analyses. For ANOVA, values for PINP and PIIINP were transformed using natural logarithms, least-squares means, and standard errors; actual values are reported in table and figures.

Data for PINP and PIIINP in BAL fluid were sparse. Values from BAL fluid obtained during the first 72 h were considered baseline values and are referred to as values for day 1 of ARDS. Values from BAL fluid obtained during days 5 through 10 are considered follow-up values and are referred to as values for day 5. Almost all patients had only one of these values; for patients with both values, only the first was considered in the statistical analyses. For each of the groups, changes in mean BAL PINP and PIIINP values over the first 10 d of ARDS were detected by using preplanned contrasts. For these contrasts least-square means were estimated from two-way ANOVA, in which group and day were main effects and the interaction effect of group with day was the effect of primary interest.

Estimated Pearson product moment correlation coefficients were used to quantify the relationships among PINP and PIIINP in plasma and BAL fluid on days 1, 3, 5, and 7 of ARDS. For these analyses, the actual day on which the BAL fluid was obtained was used. Similar analyses were conducted to quantify the relationships among PINP and PIIINP in plasma and BAL fluid and clinical variables during the first week of ARDS. In addition, after adjusting for between-patient variability, partial correlation coefficients were estimated to quantify the relationships among plasma PINP and PIIINP and clinical variables.

Finally, to evaluate the effects of treatment with methylprednisolone on plasma PINP and PIIINP values, data from the patients randomized to the treatment and placebo arms were analyzed. There were no protocol violations; all data were analyzed according to the randomization scheme. Least-square means were estimated from repeated measures (split-plot) ANOVA. For each arm mean values on days 3, 5, 7, and 10 of treatment were contrasted with baseline values as described above. In addition, for patients randomized to the treatment arm, Pearson product moment and partial correlation coefficients were estimated as previously described.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

Patient Sample and Clinical Variables

Over the study period 72 patients with ARDS were prospectively evaluated for this study (Figure 1). Twenty-four patients improved LIS by day 7 (group 1) and three died (one prior to day 10). Of the remaining 48 patients, 33 (group 2) were evaluated for a prospective, randomized, double-blind, placebo-controlled trial investigating prolonged methylprednisolone therapy in unresolving ARDS (25). Nine of these 33 patients were excluded for the following reasons: five refusals to participate, one life expectancy less than 3 mo, one pregnancy, one disease process requiring methylprednisolone, and one participation in another investigational study. Fifteen nonimprovers died prior to meeting eligibility for the randomized study, and all by day 10 of ARDS (group 3). Forty-three of the 72 evaluated patients could not be included in this report for the following reasons: 23 (seven in group 2) because blood and BAL specimens were either not obtained or were obtained at insufficiently frequent intervals, eight were on glucocorticoids for reasons other than ARDS, nine had exclusion criteria for the randomized study, and three for other reasons. Thus, this study reports data from only 29 patients, seven in group 1, 17 in group 2, and five in group 3. 


View larger version (10K):
[in this window]
[in a new window]
  		Figure 1.   Patients evaluated and recruitment.

All seven patients in group 1 survived, and the mean (± SD) duration of mechanical ventilation was 8.5 ± 2 days. Of the 17 patients in group 2, 11 were randomized to methylprednisolone and six to placebo. All patients randomized to methylprednisolone improved and survived. In the placebo group, two patients improved and survived, three failed to improve and died (two died 7 d after randomization), and one improved after crossover from placebo to methylprednisolone and survived. Two of the five patients in group 3 died on day 3 of ARDS.

Table 1 shows the clinical and physiologic characteristics of the three groups at the onset of ARDS. The changes in mean LIS over time in the three groups are shown in Figure 2. Although the three groups had similar mean LIS on day 1 of ARDS, improvers (group 1), in contrast to nonimprovers (groups 2 and 3), had a significant reduction in mean LIS by days 5 (p = 0.05) and 7 (p = 0.0001). On ARDS day 1, patients in group 3 had a higher mean MODS score (3 ± 0.3) than patients in groups 1 (2 ± 0.3, p = 0.03) and 2 (2 ± 0.2, p = 0.02). By day 7 of ARDS, a greater than one-point decline in MODS was seen in four patients (57%) in group 1 and in six patients (38%) in group 2. 

                              
View this table:
[in this window]
[in a new window]
 

TABLE 1

CLINICAL AND PHYSIOLOGICAL CHARACTERISTICS AT THE ONSET OF ARDS


View larger version (36K):
[in this window]
[in a new window]
  		Figure 2.   Changes in LIS during the first 7 d of ARDS in patients of groups 1, 2, and 3. p Values refer to changes from day 1 of ARDS: *p = 0.05; dagger p = 0.0001. The three groups are similar on day 1 of ARDS. Group 1 (improvers) differed from group 2 on days 5 (p = 0.008) and 7 (p = 0.0001), and from group 3 on days 3 (p = 0.001), 5 (p = 0.002), and 7 (p = 0.008). Group 2 differed from group 3 on days 1 (p = 0.02) and 3 (p = 0.005).

Plasma Procollagen Aminoterminal Propeptide Levels

From the 29 patients, 147 blood specimens were available for analysis of procollagen aminoterminal peptide levels. All plasma samples had elevated PINP and PIIINP levels when compared with the values obtained in normal control subjects (PIIINP, 1.6-4.0 ng/L; PINP, 20-42 ng/L). Table 2 shows the relationship among clinical variables and plasma PINP and PIIINP levels, as well as the ratio of PINP to PIIINP at the onset of ARDS. Patients with MODS scores greater than 3 or those with liver failure or renal failure had higher mean values for plasma PINP and PIIINP. However, a significant elevation was detected only for plasma PIIINP levels (ng/ml) in patients with liver failure (30 ± 10 versus 13 ± 2; p = 0.04).

                              
View this table:
[in this window]
[in a new window]
 

TABLE 2

RELATIONSHIP BETWEEN CLINICAL VARIABLES AND PLASMA TYPE I, TYPE III, AND RATIO OF TYPE I TO TYPE III PROCOLLAGEN AMINOTERMINAL PROPEPTIDE LEVELS ON DAY 1 OF ARDS

Figure 3 shows the changes in mean plasma PINP and PIIINP levels during the first 7 d of ARDS in patients of groups 1, 2, and 3. The estimated Pearson correlation coefficients between plasma PINP and PIIINP levels on day 1 to 7 of ARDS ranged from 0.66 to 0.85 (all p < 0.002), thereby indicating that during early ARDS both propeptides are markers of the same fibrogenerative process regardless of outcome. On day 1 of ARDS, the three groups had similar mean plasma PINP and PIIINP levels. Within each group, the profiles of plasma PINP and PIIINP over time reflected similar changes; thus, the mean ratio of PINP to PIIINP (not shown) remained relatively constant (within one point in each group). Group 1 (improvers) showed no change in mean plasma PINP or PIIINP levels over time. For group 2, mean plasma PINP and PIIINP levels increased over time and were significantly higher by day 5 (p = 0.004). For group 3, mean plasma PINP and PIIINP levels increased over time and for PIIINP reached significance by day 7 (p = 0.03). By day 7, group 1 had significantly lower mean PINP and PIIINP levels compared with group 2 (p = 0.0001) and group 3 (p = 0.02).


View larger version (43K):
[in this window]
[in a new window]
  		Figure 3.   The top panel shows plasma type I procollagen aminoterminal propeptide levels during the first 7 d of ARDS in patients of groups 1, 2, and 3. The three groups are similar on day 1 of ARDS. Group 1 (improvers) had no change in plasma type I procollagen aminoterminal propeptide levels over time. Group 2 mean plasma type I procollagen aminoterminal propeptide levels increased over time and were significantly higher by day 5. p Values refer to changes from day 1 of ARDS: *p = 0.002; dagger p = 0.0002. By day 7, group 1 differed significantly from group 2 (p = 0.0001) and group 3 (p = 0.02). The n value in groups 1 and 2 is smaller on day 1 because a few patients had a delayed recruitment, and in group 3 is smaller on days 5 and 7 because of early deaths. The bottom panel shows plasma type III procollagen aminoterminal propeptide levels during the first 7 d of ARDS in patients of groups 1, 2, and 3. The three groups are similar on day 1 of ARDS. Group 1 (improvers) had no change in mean plasma type III procollagen aminoterminal propeptide levels over time. Groups 2 and 3 mean plasma type III procollagen aminoterminal propeptide levels increased over time and were significantly higher by day 7. p Values refer to changes from day 1 of ARDS: Dagger p = 0.0004; §p = 0.0001. By day 7, group 1 differed significantly from group 2 (p = 0.0001) and group 3 (p = 0.002). The n value in groups 1 and 2 is smaller on day 1 because a few patients had a delayed recruitment, and in group 3 is smaller on days 5 and 7 because of early deaths.

Survivors included all group 1 subjects and survivors who were randomized to the placebo arm; nonsurvivors included all group 3 subjects and those randomized to the placebo arm who did not survive; subjects randomized to methylprednisolone were excluded from this analysis. Figure 4 shows the changes in mean plasma PINP and PIIINP levels during the first 7 d of ARDS in survivors and nonsurvivors. Although at the onset of ARDS mean PINP and PIIINP levels were similar in both nonsurvivors and survivors, procollagen aminoterminal propeptide levels increased over time in nonsurvivors and were significantly higher than those in survivors by day 7 (p = 0.0001). In both survivors and nonsurvivors, the mean ratio of PINP to PIIINP remained relatively constant over time (range, 3.4 to 4.4). On day 7, patients with plasma PINP levels less than 100 ng/ml (8 of 10 survived) were 2.5 times more likely to survive (95% CI: 0.855-7.314) compared with those with higher value (3 of 3 died), and patients with plasma PIIINP levels greater than 25 ng/ml (5 of 6 died) were nine times more likely to die (95% CI: 1.418-55.556) compared with those with lower value (8 of 8 survived). Among the 14 patients with plasma obtained immediately prior to randomization, plasma PIIINP levels greater than 25 ng/ml were observed in five of nine patients randomized to methylprednisolone, and in five of six (three nonsurvivors and two of the three survivors) patients randomized to placebo.


View larger version (24K):
[in this window]
[in a new window]
  		Figure 4.   The top panel shows plasma type I procollagen aminoterminal propeptide levels during the first 7 d of ARDS in survivors (n = 9) and nonsurvivors (n = 7). Patients randomized to prolonged methylprednisolone treatment were excluded. Survivors had no change in plasma type I procollagen aminoterminal propeptide levels over time. For nonsurvivors mean plasma type I procollagen aminoterminal propeptide levels increased over time and were significantly higher by day 5. p Values in the figure refer to changes from day 1 of ARDS. *p = 0.05; dagger p = 0.002. By day 7, survivors differed significantly from nonsurvivors (p = 0.0001). The bottom panel shows plasma type III procollagen aminoterminal propeptide levels during the first 7 d of ARDS in survivors (n = 9) and nonsurvivors (n = 7). Patients randomized to prolonged methylprednisolone treatment were excluded. Survivors had higher plasma type I procollagen aminoterminal propeptide levels on day 5. For nonsurvivors mean plasma type III procollagen aminoterminal propeptide levels increased over time and were significantly higher by day 5. p Values refer to changes from day 1 of ARDS: Dagger p = 0.0008; §p = 0.02; p = 0.0004. By day 7, survivors differed significantly from nonsurvivors (p = 0.0001).

BAL Procollagen Aminoterminal Propeptide Levels

The following trends are reported because these were the same as those reported for plasma, and BAL procollagen aminoterminal peptides levels are a more direct measure of fibroproliferative changes in the lung. Unfortunately, for these comparisons power was lacking due to insufficient numbers of samples and greater variability compared with plasma levels. In 29 patients, 55 BAL specimens were available. On days 1-3 of ARDS, no significant differences were found in mean BAL procollagen aminoterminal propeptide levels (ng/ml) between improvers and nonimprovers (PINP: 46 ± 40 versus 64 ± 38; p = 0.8, and PIIINP: 19 ± 16 versus 23 ± 13; p = 0.8) or between patients with direct and indirect injury (PINP: 41 ± 31 versus 73 ± 45; p = 0.6, and PIIINP: 16 ± 13 versus 27 ± 15; p = 0.6). Over the first 10 d of ARDS, mean BAL PINP and PIIINP levels decreased in group 1 and increased in both group 2 and group 3, without reaching statistical significance. Mean BAL PINP levels (ng/ml) on days 1-3 compared with those on days 5-10 were higher for group 1 (46 ± 44 versus 20 ± 50; p = 0.8), and lower for group 2 (73 ± 40 versus 90 ± 33; p = 0.6). Mean BAL PIIINP levels (ng/ml) on days 1-3 compared with those on days 5-10 were lower for group 1 (19 ± 18 versus 36 ± 21; p = 0.7), and group 2 (29 ± 18 versus 50 ± 13; p = 0.6). No serial BALs were available for group 3. The estimated Pearson correlation coefficients between BAL PINP and PIIINP on days 1-3 and days 5-10 were 0.97 and 0.83, respectively (both p < 0.002), indicating that both propeptides are markers of the same fibrogenerative process regardless of outcome. When patients randomized to methylprednisolone were excluded, a significant estimated partial correlation was found between plasma and BAL procollagen aminoterminal propeptide levels on days 1-3 of ARDS (PINP: r = 0.58; p = 0.03, and PIIINP: r = 0.56; p = 0.04) but not days 5-10 (after adjusting for time), thereby indicating that for these patients plasma propeptides are probable markers of fibrogenesis occurring not only in the lung but also in extrapulmonary organs. On days 1-3, mean BAL PINP and PIIINP levels (ng/ml) were numerically higher in nonsurvivors than survivors (PINP: 163 ± 862 versus 20 ± 8; p = 0.1, and PIIINP: 93 ± 9 versus 37 ± 18; p = 0.4).

Relationship between Procollagen Aminoterminal Propeptide Levels and Clinical Variables on Days 1 to 7 of ARDS

Correlation analysis for plasma PINP and PIIINP on days 1 to 7 of ARDS with LIS (Figure 2) and with its individual components showed that estimated Pearson coefficients were significant on day 3 for Cst (PIIINP: r = -0.61; p = 0.002), on day 5 for PaO2:FIO2 (PIIINP: r = -0.53; p = 0.02) and LIS (PIIINP: r = 0.46; p = 0.04), and on day 7 for PEEP (PINP: r = 0.65; p = 0.004; and PIIINP: r = 0.68; p = 0.002), and PaO2:FIO2 (PIIINP: r = -0.57; p = 0.02), and LIS (PINP: r = 0.58; p = 0.02; PIIINP: r = 0.66; p = 0.003). After adjusting for individual patient variability, significant estimated partial correlations were found between plasma procollagen aminoterminal propeptide levels and PEEP (PINP: r = 0.32; p = 0.045, and PIIINP: r = 0.4; p = 0.02), and after adjusting for time, between BAL PIIINP and Cst (r = -0.75; p = 0.03) and PaO2:FIO2 (r = 0.69; p = 0.05).Thus, plasma propeptides provided additional potentially important information about ongoing dysregulated inflammation in the lung and elsewhere beyond that provided by the usual physiologic measurements of lung function. In contrast, these physiologic measurements and BAL propeptides appeared to be markers of the same process in the lung; the amount of shared information appeared sufficiently large to justify using LIS scores and the individual components, instead of BAL propeptides, for clinical decision making.

Correlation analysis for plasma PINP and PIIINP on days 1 to 7 of ARDS with MODS scores and with its individual components showed the estimated Pearson correlation coefficients were significant for bilirubin on both day 1 (PINP: r = 0.5; p = 0.07, and PIIINP: r = 0.75; p = 0.001) and day 3 (PINP: r = 0.67; p = 0.003, and PIIINP: r = 0.54; p = 0.03), and MODS score on both day 1 (PIIINP: r = 0.42; p = 0.06) and day 3 (PIIINP: r = 0.42; p = 0.04). After adjusting for individual patient variability, significant partial correlations were found between plasma procollagen aminoterminal propeptide levels and serum creatinine (PINP: r = 0.42; p = 0.009, and PIIINP: r = 0.47; p = 0.003). Therefore, plasma, but not BAL, propeptides and MODS scores and specific individual components provided additional potentially important information about ongoing extrapulmonary dysregulated inflammation.

Changes in Procollagen Aminoterminal Propeptide Levels and Clinical Variables during Methylprednisolone Treatment

Of the 17 nonimprovers, 11 were randomized to the methylprednisolone treatment arm. Figure 5 shows changes in mean plasma PINP and PIIINP levels, LIS, and MODS scores before and during prolonged methylprednisolone treatment. Mean plasma PINP and PIIINP levels prior to initiation of treatment were higher but not significantly different from day 1 of ARDS. Methylprednisolone treatment was associated with a rapid and sustained reduction in mean plasma PINP and PIIINP levels. Mean plasma PINP levels (ng/ml) were significantly lower on days +3 compared with day 0 (45 ± 8 versus 101 ± 9; p = 0.0001), day +7 compared with day +3 (33 ± 9 versus 45 ± 8; p = 0.02), and day +10 compared with day +5 (26 ± 10 versus 37 ± 9; p = 0.04). Mean plasma PIIINP levels (ng/ml) were significantly lower on day +3 compared with day 0 (12 ± 3 versus 31 ± 3; p = 0.0007), and day +5 compared with day +3 (9 ± 3 versus 12 ± 3; p = 0.01). Generally, there were no differences in the ratio of PINP to PIIINP over time.


View larger version (34K):
[in this window]
[in a new window]
  		Figure 5.   Plasma type I and type III procollagen aminoterminal propeptide levels, MODS score, and PaO2:FIO2 before and during prolonged methylprednisolone treatment. *p = 0.005; dagger p < 0.0001; Dagger p = 0.002; p = 0.02; §p = 0.0007.

Estimated partial correlation coefficients (adjusting for between patient variability) among plasma PINP and PIIINP levels and variables of the lung injury and MODS scores during the first 7 d of methylprednisolone treatment were similar to those estimated for the first 7 d of ARDS. Thus, during methylprednisolone treatment plasma propeptides provided additional potentially important information about resolving dysregulated inflammation in the lung and elsewhere, beyond that provided by LIS and MODS scores and various individual components. For patients randomized to methylprednisolone, those with liver failure had higher mean values for bilirubin and plasma PINP on days 1 and 3 of treatment, but not PIIINP, compared with those without liver failure.

Figure 6 shows changes in mean BAL PINP and PIIINP levels before and during prolonged methylprednisolone treatment. Mean BAL PINP and PIIINP levels (ng/ml) prior to initiation of treatment (day 0) were similar to the one obtained on day 1 of ARDS (PINP: 63 ± 25 versus 54 ± 44; p = 0.7, and PIIINP: 42 ± 11 versus 36 ± 27; p = 0.8). Methylprednisolone treatment was associated with a reduction in BAL procollagen aminoterminal propeptide levels significant by day 8-15 (p = 0.003).


View larger version (33K):
[in this window]
[in a new window]
  		Figure 6.   Bronchoalveolar lavage type I and type III procollagen aminoterminal propeptide levels, before and during prolonged prednisolone treatment. *p Values refer to changes from the day of randomization; dagger p = 0.003. Values obtained within 72 h prior to randomization.

Figure 7 shows plasma PINP and PIIINP levels in relation to pulmonary and extrapulmonary organ function in one patient randomized to placebo (ARDS day 7) and later blindly crossed over (ARDS day 17) to methylprednisolone treatment because of failure to improve. No reduction was observed in LIS and procollagen aminoterminal propeptide levels during the 10 d of placebo administration. After crossover, a rapid and significant reduction was observed in PINP and PIIINP levels, with a similar change in LIS and MODS score, and the patient survived.


View larger version (28K):
[in this window]
[in a new window]
  		Figure 7.   Plasma type I and type III procollagen aminoterminal propeptide levels in relation to changes in pulmonary and extrapulmonary function in one patient randomized to placebo and later blindly switched to methylprednisolone treatment because he failed to improve. Randomized = day of randomization to placebo; crossed over = day the patient therapy was blindly changed from placebo to methylprednisolone.

To evaluate the relative response to methylprednisolone treatment, mean plasma PINP and PIIINP levels of the randomized patients were compared with those of the control patients. At the time of randomization, the six control patients had, in comparison to patients randomized to methylprednisolone, similar mean plasma PINP (142 ± 17 versus 101 ± 9; p = 0.16) and higher mean PIIINP (48 ± 6 versus 31 ± 3; p = 0.009) levels. No reduction was observed in mean plasma PINP and PIIINP levels within 10 d of randomization to placebo. However, in a subgroup analysis, control survivors, in contrast to nonsurvivors, had a slow reduction in plasma procollagen aminoterminal propeptide levels over time that reached significance for PIIINP on day +7 (p = 0.04).

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

To our knowledge this is the first report evaluating the relationship between fibrogenesis, organ dysfunction, and outcome during the natural evolution of ARDS and in response to prolonged methylprednisolone treatment. Fibrogenesis was monitored by obtaining serial measurements of PINP and PIIINP levels in both circulating plasma and BAL. Pulmonary and extrapulmonary organ dysfunctions were assessed following recently developed scoring systems (23, 24). We have found that fibrogenesis occurs early in the course of ARDS and that, in the first week of respiratory failure, patients who did not have improved lung function had significant elevations in mean plasma PINP and PIIINP levels. The positive linear relationship between increasing levels of plasma PINP and PIIINP and disease progression (e.g., MODS and LIS scores) supports a "linear" concept of tissue response to injury in ARDS (9). Among nonimprovers, those randomized to methylprednisolone treatment, in contrast to control patients, had a rapid and sustained reduction in plasma and BAL PINP and PIIINP levels. This antifibrotic action of methylprednisolone was associated with resolution of pulmonary and extrapulmonary organ dysfunction and provides additional evidence of the association between biologic efficacy and physiologic response during glucocorticoid treatment of unresolving ARDS.

Relationship between Procollagen Peptide Levels and Clinical Variables at the Onset of ARDS

Elevated plasma and BAL procollagen PINP and PIIINP levels on day 1 of ARDS indicate that collagen synthesis is an early event in the host defense response to an insult. This finding is in agreement with prior clinical reports on circulating (19) and BAL PIIINP levels (32) in patients with ARDS and with a recent investigation comparing lung tissue obtained before and after coronary artery surgery (33). In the latter study, patients undergoing cardiopulmonary bypass were found to have a 91 ± 68% increase in procollagen type I messenger RNA expression in conjunction with interstitial polymorphonuclear cell infiltration and edema in lung biopsies obtained within 2 h of lung injury (33).

Relationship between Procollagen Peptide Levels, Improvement in Lung Function, and Outcome

Fibrogenesis is an integral part of the host defense response to an insult and involves a standard sequence of events that is similar regardless of the tissue involved or the type of injury (7). In the present study over the first 7 d of ARDS, increased fibrogenesis as measured by higher plasma procollagen peptide levels was associated with failure to improve lung function and increased risk of mortality. Although mean plasma PINP and PIIINP levels were similar at the onset of ARDS, during the first week of observation they continued to increase significantly in nonimprovers and in nonsurvivors, while a declining trend was observed in improvers and survivors. These findings are in agreement with three prior reports on serial measurements of circulating (17, 19) and BAL PIIINP levels (20) in patients with ARDS, and with postmortem morphometric studies (34).

In agreement with prior reports (17, 19, 20), we have found moderate-to-strong associations between plasma and BAL procollagen aminoterminal propeptide levels and some components of the lung injury score over time. Histologic findings and pathophysiologic correlates in patients with unresolving ARDS submitted to open lung biopsy were previously described (12). Dysregulated fibroproliferation and collagen deposition transforms the initially fibrinous intra-alveolar exudate into a loosely formed myxoid connective tissue matrix and eventually into dense acellular fibrosis with lung remodeling (12). In agreement with others (35, 36), we previously reported that failure to improve gas exchange by day 7 of ARDS is associated with a mortality in excess of 80% (37). On day 7 of ARDS, patients with plasma PIIINP levels greater than 25 ng/ml were nine times more likely to die (95% CI: 1.418- 55.556), and compared with survivors, nonsurvivors had a trend toward higher BAL PINP (164 ± 122 versus 20 ± 14; p = 0.1) and PIIINP (92 ± 11 versus 36 ± 32; p = 0.4) levels. Clark and collaborators (20) previously reported that a BAL PIIINP level greater than 1.75 U/ml (equivalent to 26 ng/ml) (19) on day 7 of ARDS was strongly associated with death even after multivariate analysis controlled for other risk factors (20).

The findings of this study are similar to those we previously reported for plasma and BAL inflammatory cytokines over time (10, 11) and provide additional support for the theoretical paradigm of "linear response" to injury in ARDS, where duration and magnitude of the host response dictate the ultimate reparative outcome (9). If the injury is minor, the host response is curtailed, appropriate connective tissue regeneration occurs, and normal structure and function are restored (adaptive response). If the injury is severe and the host response is prolonged, then tissue reconstruction often includes excess proliferation of mesenchymal cells and deposition of collagen, in tandem with persistent inflammation and coagulation (maladaptive response) (8, 12). Prior evidence substantiating that an exaggerated and protracted host response is a critical determinant of patient outcome in ARDS was provided by several ARDS studies that have investigated clinical, laboratory, and physiologic expressions of inflammation, coagulation, and fibroproliferation over time (38), and by a recent report on infections and the systemic inflammatory response in ARDS (37).

Relationship between Type I and III Collagen during Evolution of Fibrosis

To our knowledge this is the first study reporting plasma and BAL PINP and PIIINP levels over time in ARDS. A prior study obtained a single measurement of multiple collagen markers in early ARDS and found a significant positive linear correlation between serum and BAL levels for both type I and type III collagen (39). We have found moderate-to-strong correlations over time between plasma PINP and PIIINP levels, and strong correlations between BAL PINP and PIIINP levels. Immunohistochemical studies of collagen types in fibrotic lung have shown that there is an increased proportion of type III collagen in early active fibrosis, while late fibrosis is characterized by increased deposition of the more rigid type I collagen (40). The reason(s) for the change from type III to type I collagen synthesis and deposition is (are) not known, but it has been suggested that higher fibroblast density earlier in the injured tissue may be part of the explanation (41). It is known that in vitro both lung and dermal fibroblasts synthesize predominantly type III collagen when plated at high density compared with when plated at low density (42). Another explanation for increased type III collagen production after ARDS injury may be the influence of alveolar macrophages. In vitro co-culture of alveolar macrophages with lung fibroblasts results predominantly in synthesis of type III collagen (43). We did not observe a temporal variation in the ratio of PINP to PIIINP. The mean plasma PINP to PIIINP ratio was consistently preserved at 3.4 to 4.4 over time within each group, and even during methylprednisolone treatment. The reason for persistence of the PINP/PIIINP ratio may be that the length of the period of sampling (0-10 d) was insufficient to show a greater predominance of PINP. However, several factors determine the levels of PINP and PIIINP in plasma or BAL, including: (1) rate of synthesis of procollagen alpha I (I) and procollagen alpha II (III); (2) cleavage rates by types I and III N-terminal proteases; and (3) clearance rates from plasma and body fluids. PIIINP is cleaved at a slower rate than PINP and would tend to make plasma PIIINP levels underestimate type III procollagen synthesis (16).

Procollagen Peptides as a Reflection of Whole Body Repair

We have observed a significant correlation between plasma PINP and PIIINP levels, bilirubin, creatinine, and MODS score, a finding in agreement with a prior report in trauma patients (19). Although it is yet unclear if increased propeptide levels in patients with elevated bilirubin or creatinine reflect increased organ production or impaired excretion, recent evidence suggests that in patients with MODS, circulating procollagen aminoterminal propeptide levels reflect the "whole body" collagen formation, including lungs, liver, kidneys, and wounds (19).

Changes during Methylprednisolone Treatment

Glucocorticoids inhibit both fibroblast proliferation and deposition of collagen (21). In response to prolonged methylprednisolone treatment, we have observed a rapid and significant reduction in plasma and BAL aminoterminal propeptide levels and similar changes (although delayed) in lung injury and MODS scores. The significant correlation detected during therapy among these variables indicates a likely casual link between reduction in fibrogenesis and resolution of pulmonary and extrapulmonary organ dysfunction. Two prior clinical reports have described a reduction in circulatory PIIINP levels during prolonged glucocorticoid treatment of patients with usual interstitial pneumonitis (16), and pneumocystis pneumonia (44). Both studies found that a reduction in circulatory peptide levels correlated with improvement in lung physiology (16, 44).

We previously reported that patients with unresolving ARDS and histologic evidence of dense acellular fibrosis at open lung biopsy do not improve with methylprednisolone rescue treatment (45). The striking physiologic and biologic response to methylprednisolone treatment reported in this article indicates that end-stage fibrosis is less likely to develop before days 7-10 of ARDS (timing of randomization). The treatment-induced decline in procollagen aminoterminal propeptide levels is similar to the response previously observed for clinical and laboratory parameters of local and systemic inflammation (46). In that study, methylprednisolone treatment was associated with a significant and parallel reduction in plasma and BAL inflammatory cytokine levels, LIS, and BAL indices of pulmonary vascular permeability (BAL albumin and percentage of polymorphonuclear cells) (46). The significant correlation between changes in PINP and PIIINP levels and variables of pulmonary and extrapulmonary organ dysfunction before and during prolonged methylprednisolone treatment further validates the interdependence between host response, organ function, and outcome. By monitoring plasma procollagen aminoterminal propeptide levels over time, we have observed that, in the absence of treatment, patients with unresolving ARDS had either a sustained collagen deposition and died, or a slow and delayed resolution and survived. Although unproven, it is reasonable to speculate that methylprednisolone treatment may accelerate resolution of fibroproliferation in the latter group (survivors in the placebo group) and thus improve long-term pulmonary function. Complications of prolonged methylprednisolone treatment were previously reported (25). Our findings indicate that plasma PINP and PIIINP levels could be a useful marker to monitor treatment response in ARDS and could be included in studies evaluating dose responses, timing, and duration of methylprednisolone administration. On the other hand, our work indicates that monitoring simple physiologic parameters is very helpful, since we found moderate-to-strong associations between PINP and PIIINP levels and components of LIS over time, and a clear demarcation in outcome among improvers and nonimprovers by day 7 of ARDS.

Terminating the randomized sequential trial very quickly produced two major consequences, relevant to this report. The first was unavoidable because stopping any trial early biases the estimate of the treatment effect. In this trial estimates of the large effects of methylprednisolone on the propeptides (PINP and PIIINP) and improvement in LIS and MODS scores were probably greater than those estimates that would have been if the trial had not been terminated. The second consequence involved potential lack of comparability between the two groups of randomized patients to the extent that the treatment effect might merely have reflected some confounding variable, such as severity of illness. When the total sample size is relatively small, the process of randomization may not prevent some important variable from being confounded with both treatment group and outcome. Because group 2 included a subset of the patients who were randomized, the potential for lack of comparability between the two subgroups is greater. Based on APACHE III score at randomization, the most severely ill patient was randomized to the placebo group and is included in this report. Although estimates of the positive effects of methylprednisolone may have been biased, it is extremely unlikely that any confounding variable could have explained the reductions in propeptides following methylprednisolone administration or the differences in responses between the two groups after randomization.

In conclusion, we have found that plasma PINP and PIIINP levels are elevated in early ARDS. Over the first 7 d of ARDS, increased fibrogenesis, as measured by higher plasma procollagen peptide levels, was associated with failure to improve lung function and increased risk of mortality. Furthermore, we observed significant correlations between plasma PINP and PIIINP levels, bilirubin, creatinine, and MODS score. The strong associations between increasing plasma PINP and PIIINP levels and disease progression, in addition to our prior report on pro-inflammatory cytokine levels over time (10, 11), support the theoretical paradigm of a "linear response" to injury, where the fibrotic process is driven by an overexuberant and prolonged inflammatory response (9). Among nonimprovers, those randomized to prolonged methylprednisolone treatment had a rapid and significant reduction in plasma and BAL aminoterminal propeptide levels and similar changes in lung injury and MODS scores. These findings provide additional evidence of an association between biological efficacy and physiologic response during prolonged methylprednisolone treatment of unresolving ARDS. Plasma PINP or PIIINP levels could be a useful marker to monitor disease progression and response to treatment in ARDS.

    Footnotes

Correspondence and requests for reprints should be addressed to Dr. G. Umberto Meduri, University of Tennessee, Memphis, Division of Pulmonary and Critical Care Medicine, 956 Court Avenue, Room H316, Memphis, TN 38163. E-mail: umeduri{at}utmem1.utmem.edu

(Received in original form January 27, 1998 and in revised form May 28, 1998).

Acknowledgments: The writers wish to acknowledge the invaluable assistance received from David Armbruster, Ph.D., Johnny Belenchia, M.D., Stephanie Carsons, R.N., Emmel Golden, M.D., Kenneth V. Leeper, M.D., Reba Umberger, R.N., and Muhammad Zaman, M.D. The APACHE Medical System Inc. graciously assisted in the calculation of the Acute Physiologic and Chronic Health Evaluation (APACHE) III score on ICU admission (47).

Supported by the Baptist Memorial Health Care Foundation and by the University of Tennessee Clinical Research Center.

    References
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES

1. Meduri, G. U.. 1996. The role of the host defense response in the progression and outcome of ARDS: pathophysiological correlations and response to glucocorticoid treatment. Eur. Respir. J. 9: 2650-2670 [Abstract].

2. Strieter, R. M., N. W. Lukacs, T. J. Standiford, and S. L. Kunkel. 1993. Cytokines and lung inflammation: mechanisms of neutrophil recruitment to the lung. Thorax 48: 765-769 [Free Full Text].

3. Van Der Poll, T., H. R. Buller, H. T. Cate, C. H. Wortel, K. A. Bauer, S. J. K. VanDeventer, C. E. Hack, H. P. Sauerwein, R. D. Rosenberg, and J. W. Cate. 1990. Activation of coagulation after administration of tumor necrosis factor to normal subjects. N. Engl. J. Med. 322: 1622-1627 [Abstract].

4. Elias, J. A., B. Freundlich, J. A. Kern, and J. Rosenbloom. 1990. Cytokine networks in the regulation of inflammation and fibrosis in the lung. Chest 97: 1439-1445 [Abstract/Free Full Text].

5. King, R. J., M. B. Jones, and P. Minoo. 1989. Regulation of lung cell proliferation by polypeptide growth factors. Am. J. Physiol. 257: L23-L38 [Abstract/Free Full Text].

6. Postlethwaite, A. E., and J. M. Seyer. 1990. Stimulation of fibroblast chemotaxis by human recombinant tumor necrosis factor alpha  (TNF-alpha ) and a synthetic TNF-alpha 31-68 peptide. J. Exp. Med. 172: 1749-1756 [Abstract/Free Full Text].

7. Horslev-Petersen, K.. 1990. Circulating extracellular matrix components as markers for connective tissue response to inflammation. Danish Medical Bulletin 37: 308-329 [Medline].

8. Kovacs, E. J., and L. A. Dipietro. 1994. Fibrogenic cytokines and connective tissue production. FASEB J. 8: 854-861 [Abstract].

9. Bitterman, P. B., and C. A. Henke. 1991. Fibroproliferative disorders. Chest 99: 81S-84S .

10. Meduri, G. U., S. Headley, G. Kohler, F. Stentz, E. Tolley, R. Umberger, and K. Leeper. 1995. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS: plasma IL-beta and IL-6 are consistent and efficient predictors of outcome over time. Chest 107: 1062-1073 [Abstract/Free Full Text].

11. Meduri, G. U., G. Kohler, S. Headley, E. Tolley, F. Stentz, and A. Postlethwaite. 1995. Inflammatory cytokines in the BAL of patients with ARDS: persistent elevation over time predicts poor outcome. Chest 108: 1303-1314 [Abstract/Free Full Text].

12. Meduri, G. U., M. ElTorky, and H. T. Winer-Muram. 1995. The fibroproliferative phase of late adult respiratory distress syndrome. Semin. Respir. Infect. 10: 154-175 [Medline].

13. Postlethwaite, A. E.. 1995. Role of T cells and cytokines in effecting fibrosis. Intern. Rev. Immunol. 12: 247-258 [Medline].

14. Fessler, J. H., and L. I. Fessler. 1978. Biosynthesis of procollagen. Ann. Rev. Biochem. 47: 129-162 [Medline].

15. Kessler, E., and B. Goldberg. 1978. A method for assaying the activity of the endopeptidases which excises the nonhelical carboxyterminal extensions from type I procollagen. Anal. Biochem. 86: 463-469 [Medline].

16. Kirk, J. E., E. D. Bateman, P. Haslam, G. J. Laurent, and M. Turner-Warwick. 1984. Serum type III procollagen peptide concentration in cryptogenic fibrosing alveolitis and its clinical relevance. Thorax 39: 726-732 [Abstract/Free Full Text].

17. Entzian, P., A. Huckstadt, H. Kreipe, and J. Barth. 1990. Determination of serum concentrations of type III procollagen peptide in mechanically ventilated patients: pronounced augmented concentrations in the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 142: 1079-1082 [Medline].

18. Kropf, J., E. Grobe, M. Knoch, M. Lammers, A. M. Gressner, and H. Lennartz. 1991. The prognostic value of extracellular matrix component concentrations in serum during treatment of adult respiratory distress syndrome with extracorporeal CO2 removal. Eur. J. Clin. Chem. Clin. Biochem. 29: 805-812 [Medline].

19. Waydhas, C., D. Nast-Kolb, A. Trupka, S. Lenk, K. H. Duswald, L. Schweiberer, and M. Jochum. 1993. Increased serum concentrations of procollagen type III peptide in severely injured patients: an indicator of fibrosing activity? Crit. Care Med. 21: 240-247 [Medline].

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

21. Durant, S., D. Duval, and F. Homo-Delarche. 1986. Factors involved in the control of fibroblast proliferation by glucocorticoids: a review. Endocr. Rev. 7: 254-268 [Medline].

22. Meduri, G. U., J. M. Belenchia, R. J. Estes, R. G. Wunderink, M. ElTorky, and K. V. Leeper Jr.. 1991. Fibroproliferative phase of ARDS: clinical findings and effects of corticosteroids. Chest 100: 943-952 [Abstract/Free Full Text].

23. Murray, J. F., M. A. Matthay, J. M. Luce, and M. R. Flick. 1988. An expanded definition of the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 138: 720-723 [Medline].

24. Bernard, G. R., G. Doig, L. D. Hudson, S. Lemeshow, J. C. Marshall, J. Russell, W. Sibbald, C. L. Sprung, J. L. Vincent, A. P. Wheeler, and Ad Hoc Committees for Organ Failure. 1995. Quantification of organ failure for clinical trials (abstract). Am. J. Respir. Crit. Care Med. 151: A323 .

25. Meduri, G. U., A. S. Headley, E. Golden, S. J. Carson, R. A. Umberger, T. Kelso, and E. A. Tolley. 1998. Prolonged methylprednisolone treatment improves lung function and outcome in patients with unresolving acute respiratory distress syndrome: results of a randomized, double-blind, placebo-controlled trial. J.A.M.A. 280: 159-165 [Abstract/Free Full Text].

26. Bernard, G. R., A. Artigas, K. L. Brigham, J. Carlet, K. Falke, L. Hudson, M. Lamy, J. R. Legall, A. Morris, R. Spragg, and the Consensus Committee. 1994. The American-European Consensus Conference on ARDS: definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am. J. Respir. Crit. Care Med. 149: 818-824 [Abstract].

27. Slutsky, A. S.. 1993. Mechanical ventilation. Chest 104: 1833-1059 [Free Full Text].

28. Bone, R. C., R. A. Balk, F. B. Cerra, R. P. Dellinger, A. M. Fein, W. A. Knaus, R. M. H. Schein, and W. J. Sibbald. 1992. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 101: 1644-1655 [Abstract/Free Full Text].

29. Meduri, G. U., G. L. Mauldin, R. G. Wunderink, K. V. Leeper, C. B. Jones, E. Tolley, and G. Mayhall. 1994. Causes of fever and pulmonary densities in patients with clinical manifestations of ventilator- associated pneumonia. Chest 106: 221-235 [Abstract/Free Full Text].

30. Meduri, G. U., and J. Chastre. 1992. The standardization of bronchoscopic techniques for ventilator-associated pneumonia. Chest 102: 557S-564S .

31. Sokal, R. R., and F. J. Rohlf. 1981. Biometry: The Principles and Practice of Statistics in Biological Research, 2nd ed. W. H. Freeman and Co., New York. 215-216.

32. Chesnut, A. N., M. A. Matthay, and J. G. Clark. 1996. Type III procollagen peptide is present in pulmonary edema fluid on day one in patients with acute respiratory distress syndrome (ARDS) (abstract). Am. J. Respir. Crit. Care Med. 153: A11 .

33. Deheinzelin, D., F. B. Jatene, P. H. N. Saldiva, and R. R. Brentani. 1997. Upregulation of collagen messenger RNA expression occurs immediately after lung damage. Chest 112: 1184-1188 [Abstract/Free Full Text].

34. Fukuda, Y., M. Ishizaki, Y. Masuda, G. Kimura, O. Kawanami, and Y. Masugi. 1987. The role of intraalveolar fibrosis in the process of pulmonary structural remodeling in patients with diffuse alveolar damage. Am. J. Pathol. 126: 171-182 [Abstract].

35. Bernard, G. R., J. M. Luce, C. L. Sprung, J. E. Rinaldo, R. M. Tate, W. J. Sibbald, K. Kariman, S. Higgins, R. Bradley, C. A. Metz, T. R. Harris, and K. L. Brigham. 1987. High-dose corticosteroids in patients with the adult respiratory distress syndrome. N. Engl. J. Med. 317: 1565-1570 [Abstract].

36. Bone, R. C., R. Maunder, G. Slotman, H. Silverman, T. M. Hyers, M. D. Kerstein, J. J. Urpsprung, and the Prostaglandin E1 Study Group. 1989. An early test of survival in patients with the adult respiratory distress syndrome: the PaO2/FIO2 ratio and its response to conventional therapy. Chest 96: 849-851 [Abstract/Free Full Text].

37. Headley, A. S., E. Tolley, and G. U. Meduri. 1997. Infections and the inflammatory response in acute respiratory distress syndrome. Chest 111: 1306-1321 [Abstract/Free Full Text].

38. Meduri, G. U.. 1997. Host defense response and outcome in ARDS. Chest 112: 1154-1158 [Free Full Text].

39. Farjanel, J., D. J. Hartmann, B. Guidet, L. Luquel, and G. Offenstadt. 1993. Four markers of collagen metabolism as possible indicators of disease in the adult respiratory distress syndrome. Am. Rev. Respir. Dis. 147: 1091-1099 [Medline].

40. Low, R. B., M. S. Giancola, T. E. King, J. Chapitis, P. Vacek, and G. S. Davis. 1992. Serum and bronchoalveolar lavage of N-terminal type III procollagen peptides in idiopathic pulmonary fibrosis. Am. Rev. Respir. Dis. 146: 701-706 [Medline].

41. Krane, S. M. 1984. Genetic and acquired disorders of collagen deposition. In K. A. Piez and A. H. Reddi, editors. Extracellular Matrix Biochemistry. Elsevier, New York. 413-463.

42. Abe, S., B. U. Steinmann, L. M. Wahl, and G. R. Martin. 1979. High cell density alters the ratio of type III to type I collagen synthesis by fibroblasts. Nature 279: 442-444 .

43. Kelley, J., L. Trombley, E. J. Kovacs, G. S. Davis, and M. Absher. 1981. Pulmonary macrophages alter the collagen phenotype of lung fibroblasts. J. Cell. Physiol. 109: 353-361 [Medline].

44. Bentsen, K. D., T. L. Nielsen, J. K. M. Eaftinck, Schattenkerk, B. N. Jensen, J. D. Lundgren, and the Copenhagen-Amsterdam PCP-prednisolone Study Group. 1993. Serum type III procollagen peptide in patients with Pneumocystis carinii infection. Am. Rev. Respir. Dis. 148: 1558-1562 [Medline].

45. Meduri, G. U., A. J. Chinn, K. V. Leeper, R. G. Wunderink, E. Tolley, H. T. Winer-Muram, V. Khare, and M. ElTorky. 1994. Corticosteroid rescue treatment of progressive fibroproliferation in late ARDS: patterns of response and predictors of outcome. Chest 105: 1516-1527 [Abstract/Free Full Text].

46. Meduri, G. U., S. Headley, E. Tolley, M. Shelby, F. Stentz, and A. Postlethwaite. 1995. Plasma and BAL cytokine response to corticosteroid rescue treatment in late ARDS. Chest 108: 1315-1325 [Abstract/Free Full Text].

47. Knaus, W. A., D. P. Wagner, E. A. Draper, J. E. Zimmerman, M. Bergner, P. G. Bastos, C. A. Sirio, D. J. Murphy, T. Lotring, A. Damiano, and F. E. Harrell. 1991. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 100: 1619-1636 [Abstract/Free Full Text].




This article has been cited by other articles:


Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
D. E. McClintock, B. Starcher, M. D. Eisner, B. T. Thompson, D. L. Hayden, G. D. Church, M. A. Matthay, and the National Heart, Lung, Blood Institute Acute Re
Higher urine desmosine levels are associated with mortality in patients with acute lung injury
Am J Physiol Lung Cell Mol Physiol, October 1, 2006; 291(4): L566 - L571.
[Abstract] [Full Text] [PDF]

Home page
 J. Clin. Endocrinol. Metab.Home page
A. Dovio, L. Perazzolo, G. Osella, M. Ventura, A. Termine, E. Milano, A. Bertolotto, and A. Angeli
Immediate Fall of Bone Formation and Transient Increase of Bone Resorption in the Course of High-Dose, Short-Term Glucocorticoid Therapy in Young Patients with Multiple Sclerosis
J. Clin. Endocrinol. Metab., October 1, 2004; 89(10): 4923 - 4928.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
P. R. M. Rocco, A. B. Souza, D. S. Faffe, C. P. Passaro, F. B. Santos, E. M. Negri, J. G. M. Lima, R. S. Contador, V. L. Capelozzi, and W. A. Zin
Effect of Corticosteroid on Lung Parenchyma Remodeling at an Early Phase of Acute Lung Injury
Am. J. Respir. Crit. Care Med., September 15, 2003; 168(6): 677 - 684.
[Abstract] [Full Text] [PDF]