0
Original Research: Critical Care |

Detection of Fibroproliferation by Chest High-Resolution CT Scan in Resolving ARDSHigh-Resolution CT Scan Fibroproliferation in ARDS FREE TO VIEW

Ellen L. Burnham, MD; Robert C. Hyzy, MD, FCCP; Robert Paine, III, MD; Aine M. Kelly, MD; Leslie E. Quint, MD; David Lynch, MB; Douglas Curran-Everett, PhD; Marc Moss, MD; Theodore J. Standiford, MD
Author and Funding Information

From the Division of Pulmonary Sciences and Critical Care Medicine (Drs Burnham and Moss), Department of Medicine, University of Colorado School of Medicine, Aurora, CO; the Division of Pulmonary and Critical Care Medicine (Drs Hyzy and Standiford), Department of Medicine, and the Department of Radiology (Drs Kelly and Quint), University of Michigan School of Medicine, Ann Arbor, MI; the Division of Pulmonary and Critical Care Medicine (Dr Paine), Department of Medicine, University of Utah School of Medicine, Salt Lake City, UT; Department of Radiology (Dr Lynch) and the Department of Biostatistics and Bioinformatics (Dr Curran-Everett), National Jewish Health, Denver, CO; and the Department of Biostatistics and Informatics (Dr Curran-Everett), University of Colorado School of Public Health, Aurora, CO.

CORRESPONDENCE TO: Ellen L. Burnham, MD, University of Colorado School of Medicine, Anschutz Medical Campus, 12700 E. 19th Ave, C272, Aurora, CO 80045; e-mail: ellen.burnham@ucdenver.edu


FOR EDITORIAL COMMENT SEE PAGE 1140

Some of this work has been previously reported as an abstract at the American Thoracic Society International Meeting, May 19, 2013, Philadelphia, PA.

FUNDING/SUPPORT: This work was supported by the National Institutes of Health/National Heart, Lung, and Blood Institute [Grant P50 HL074024].

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2014;146(5):1196-1204. doi:10.1378/chest.13-2708
Text Size: A A A
Published online

BACKGROUND:  In ARDS, the extent of fibroproliferative activity on chest high-resolution CT (HRCT) scan has been reported to correlate with poorer short-term outcomes and pulmonary-associated quality of life. However, clinical factors associated with HRCT scan fibroproliferation are incompletely characterized. We questioned if lung compliance assessed at the bedside would be associated with fibroproliferation on HRCT scans obtained during the resolution phase of ARDS.

METHODS:  We used data from a published randomized, controlled clinical trial in ARDS. All patients were cared for using a low tidal volume strategy. Demographic data and ventilator parameters were examined in association with radiologic scores from chest HRCT scans obtained 14 days after diagnosis.

RESULTS:  Data from 82 patients with ARDS were analyzed. Average static respiratory compliance over the first 14 days after diagnosis was inversely associated with chest HRCT scan reticulation (ρ = −0.46); this relationship persisted in multivariable analysis including APACHE (Acute Physiology and Chronic Health Evaluation) II scores, initial Pao2/Fio2, pneumonia diagnosis, and ventilator days. Average static respiratory compliance was also lower among patients with bronchiectasis at day 14 (P = .007). Initial static respiratory compliance obtained within the first day after ARDS diagnosis was correlated inversely with the presence of HRCT scan reticulation (ρ = −0.38) and was lower among patients who demonstrated bronchiectasis on the day 14 HRCT scan (P = .008).

CONCLUSIONS:  In patients with ARDS, diminished lung compliance measured bedside was associated with radiologic fibroproliferation 14 days post diagnosis. Establishing factors that predispose to development of excessive fibroproliferation with subsequent confirmation by chest HRCT scan represents a promising strategy to identify patients with ARDS at risk for poorer clinical outcomes.

Figures in this Article

Although mortality in ARDS has improved with the use of a low tidal volume (Vt) ventilator strategy,1 pulmonary fibroproliferation remains an observable clinical phenomenon in a subset of patients,2 in whom it has been associated with higher mortality and ventilator dependence.25 Moreover, recent data support a negative relationship between fibroproliferation and health-related quality of life in ARDS survivors.6 Nevertheless, factors influencing the development of fibroproliferation are not well understood.

Previous investigations have used chest high-resolution CT (HRCT) scans as a tool in very early ARDS (< 1 week)79 to identify patients with fibroproliferative radiologic abnormalities who are likely to suffer poorer short-term outcomes, such as mortality. However, studies to systematically assess fibroproliferation via HRCT scan during resolving ARDS (> 1 week after diagnosis) have not been performed. Therefore, the specific patient and clinical factors that underlie the development of radiologic fibroproliferation have not been clearly established. It is possible that certain clinical factors associated with fibroproliferation are modifiable and that novel treatment modalities could be developed to limit excessive fibroproliferation. Moreover, using HRCT scanning during the resolution phase of ARDS to identify patients who have excessive fibroproliferation may be a useful strategy to predict who will have an impaired quality of life after hospital discharge.

The mechanical ventilation strategy used to care for patients with ARDS is one clinical factor that may be implicated in the fibroproliferative response. In laboratory investigations focused on characterizing ventilator-associated lung injury, mechanical ventilation has been linked to increased inflammation, altered extracellular matrix composition, enhanced production of profibrotic molecules, and promotion of frank fibrosis within the lung.1012 Analogous observations have been reported in mechanically ventilated patients with ARDS.1315 Although large clinical trials have reassured clinicians that a low Vt strategy1 and judicious use of positive end-expiratory pressure (PEEP)16,17 promote improved short-term outcomes in ARDS, including survival, their effects on early development of fibroproliferation remain unclear. Importantly, recent investigations using chest CT scanning have suggested that lung inhomogeneities typical of ARDS function as “stress raisers” that may locally increase the susceptibility to ventilator-induced lung injury from externally applied pressure or volume,18 highlighting the need to further delineate and characterize the effects of mechanical ventilation on ARDS pathophysiology.

We sought to characterize clinical factors associated with the development of fibroproliferative changes on 14-day chest HRCT scan in a cohort of patients with ARDS enrolled in a multicenter clinical trial.19 We hypothesized that patients with ARDS with the poorest respiratory compliance, who were exposed to an increased duration and intensity of PEEP and plateau pressure (PPlat), would display evidence of increased chest HRCT scan fibroproliferation, as evidenced by increased reticulation and bronchiectasis, 14 days after diagnosis.

Data were derived from a National Institutes of Health-sponsored double-blind randomized controlled trial for acute lung injury of granulocyte macrophage-colony stimulating factor19 that enrolled patients from 2004 to 2009. The study was conducted in accordance with the amended Declaration of Helsinki. Local institutional review boards approved the protocol (institutional review board of the University of Michigan Medical School approval number 2003-0430, Emory University institutional review board approval number 734-2004, Colorado Multiple Institutional Review Board, approval number 06-0573), and written informed consent was obtained from all patients.

Subjects Examined

Individuals meeting criteria for ARDS using the American-European Consensus Conference definition were considered for enrollment in the parent study.20 Patients who had evidence of preexisting lung disease (GOLD [Global Initiative for Chronic Obstructive Lung Disease] stages III or IV COPD, or interstitial lung disease from any cause) were excluded. Subjects received standardized ventilator management according to the ARDS Network protocol, targeting a Vt of 6 mL/kg ideal body weight and PPlat < 30 cm H2O.1 Weaning was similarly based upon protocols developed through the ARDS Network and implemented by respiratory therapists.

For the present study, parent study data were included from patients with chest HRCT scans who were followed clinically for 28 days after diagnosis (or until death or discharge). Clinical data from the parent trial were used in our analyses. Detailed information regarding patients from the parent study who were excluded from the present investigations may be found in e-Table 1. Predicted body weight was calculated using patient height and sex by established methods.1 Ventilator parameters, including Vt (mL), PEEP (cm H2O), and PPlat (cm H2O) were recorded daily in patients on mechanical ventilation, along with the number of mechanically ventilated days. Average Vt, PEEP, and PPlat exposures prior to chest HRCT scan were calculated by summing all available daily values for these parameters while patients were on mechanical ventilation and dividing by the total number of mechanically ventilated days prior to the chest HRCT scan. Static respiratory compliance (Cstat) (mL/cm H2O) was calculated using collected ventilator variables.

HRCT Scans

Chest HRCT scans were performed at 14 (SD, 2) days post ARDS diagnosis. Enrolled patients who were mechanically ventilated at the time of the chest HRCT scan were placed on ≥ 10 cm H2O PEEP during the HRCT scan21; all images were obtained at end-inspiration. Chest HRCT scan sections were obtained with 1.25-mm collimation and reconstructed using a high-spatial-frequency algorithm. All images were viewed at window settings optimized for assessment of lung parenchyma (window width, 1,300 Hounsfield units; window level, −500 Hounsfield units). Completed HRCT scans were evaluated by two chest radiologists using established methods.8,22 Radiologists had previously calibrated their scoring strategy by concomitantly reading chest HRCT scans obtained early in the study (n = 8) to ensure consistent grading of subsequently performed HRCT scans. Five levels of imaging were assessed: aortic arch, 1 cm above the dome of the right hemidiaphragm, and three additional levels equally spaced between these two levels. At each level, lungs were divided into anterior and posterior zones by drawing a horizontal line across the image, creating four quadrants for analysis per level. Radiologic patterns that were quantified in each quadrant included reticulation, ground-glass opacification (GGO), and intense parenchymal opacification (IPO). Reticulation was defined as evidence of innumerable interlacing shadows that were fine, intermediate, or coarse, with associated distortion of the lung architecture, and taken to represent fibrosis. GGO was defined as a hazy increase in lung attenuation with preservation of bronchovascular markings. IPO was characterized by a homogeneous increase in lung attenuation with obscuration of the bronchovascular structures in which an air bronchogram may have been present. The extent of involvement for each pattern was assigned a numerical score, where 0 = no involvement; 1= < 5% involvement/minimal/not normal; 2 = 5% to 25% involvement; 3 = 26% to 49% involvement; 4 = 50% to 75% involvement; and 5 = > 75% involvement. For each radiologic pattern, the average score (range, 0-20) was calculated by summing all quadrant values for each slice, adding these values together, and dividing by the number of slices with complete data. Bronchiectasis was scored as present (1) or absent (0) on each slice, with the total bronchiectasis score calculated by summing values over all slices (range, 0-5). Missing data were scored as 0.6

Statistical Analyses

Analyses with continuous data were performed with Student t test or the Wilcoxon rank sum test. Categorical data were analyzed using Fisher exact test. Correlation coefficients between continuous data were calculated using Spearman correlation coefficient. Linear or logistic regression was used to determine the association between clinical variables, including ventilator parameters, and HRCT scan scores at the 14-day time point. Two-sided P values of < .05 were deemed to be significant. JMP software (version 9) was used in analyses.

The parent study enrolled 132 patients. Eight patients died before the 14-day HRCT scan could be completed. In 42 of the patients, HRCT scan was not performed because of the patient’s being unable to safely tolerate the procedure, the attending physician being unwilling to proceed with study, or investigator withdrawing the patient from the study. Therefore, the cohort included 82 patients who had a chest HRCT scan completed at day 14.

Baseline characteristics for the patients who had chest HRCT scans available are highlighted in Table 1. A similar number of women and men were in the cohort; the majority of patients were white. Survival at 28 days was 96%. On enrollment, the majority of patients had either moderate or severe ARDS by Pao2/Fio2 criteria.23 Using an initial Cstat value of < 40, 68% (36 of 53) of patients were categorized as having severe ARDS, whereas initial corrected minute ventilation ([V. ecorr] > 10 L/min) categorized 64% (47 of 73) as having severe ARDS. Twelve of the initial 82 patients (15%) had a Pao2/Fio2 ratio of ≤ 100 and V. ecorr of ≥ 13 L/min, which would place them in the higher-risk ARDS subgroup.23 At 14 days, in the 42 patients who remained on mechanical ventilation, 5% had a Pao2/Fio2 ratio of ≤ 100, whereas 79% had V. ecorr that would place them in the severe ARDS category.

Table Graphic Jump Location
TABLE 1 ]  Baseline Characteristics of Patients With ARDS

N = 82 with 14-d chest HRCT scans and complete follow-up. Values are means (SD) unless otherwise noted. Earliest values available for patients after initiation of mechanical ventilation presented. APACHE = Acute Physiology and Chronic Health Evaluation; HRCT = high-resolution CT; SOFA = Sequential Organ Failure Assessment.

a 

Minute ventilation calculated correcting for Paco2 using the following equation: [minute ventilation × Paco2]/40. Severe ARDS categorized as minute ventilation > 10 L/min.

b 

Static compliance calculated with plateau, positive end-expiratory pressure, and tidal volume values collected within 24 h of ARDS diagnosis. Severe ARDS categorized as static compliance < 40.

For comparison, characteristics of patients from the parent study who were excluded from these investigations indicated nonsignificant differences in terms of age, sex, and ARDS risk factors; excluded patients were significantly less likely to be white. ARDS severity by either initial Pao2/Fio2 or V. ecorr did not differ; however, a higher percentage of patients with severe ARDS based on Cstat (P = .01) was observed in the excluded group. APACHE (Acute Physiology and Chronic Health Evaluation) II and Sequential Organ Failure Assessment (SOFA) scores indicated a greater severity of illness in excluded patients, with significantly poorer 28-day survival (61%, P < .0001 vs included patients). Please see e-Table 1 for additional details.

In terms of ventilator measurements, the average Vt exposure indexed to predicted body weight (7.2 mL/kg) was increased relative to the goal of 6 mL/kg (Table 2). Both the initial and average Cstat at day 14 would categorize these patients as having severe ARDS. We did not observe significant associations between initial or average Cstat with BMI, ARDS severity, or ARDS risk factor. Comparing physiologic data between included and excluded patients revealed nonsignificant differences in terms of Vt (with or without indexing to predicted body weight) and PEEP values obtained within the first 24 h of ARDS diagnosis. Notably, initial PPlat was higher (P = .04) and initial Cstat was lower (P = .01) in excluded patients.

Table Graphic Jump Location
TABLE 2 ]  Ventilator Characteristics of Enrolled Patients

Values are means (SD). Cstat = static respiratory compliance; n/a = not applicable; PEEP = positive end-expiratory pressure; PPlat = plateau pressure; Vt = tidal volume.

a 

Data from 53 subjects with ARDS available at this time point.

No direct correlations with baseline clinical characteristics including age, sex, race, APACHE II, total SOFA score at week 1, and Pao2/Fio2 ratio on date of enrollment and individual day HRCT scan scores were observed. A diagnosis of pneumonia as a primary or secondary ARDS risk factor (yes or no) correlated with higher 14-day HRCT scan reticulation (5.1 [SD, 3.3] vs 3.3 [SD, 3.2], P = .04) and bronchiectasis (1.3 [SD, 1.8] vs 0.4 [SD, 0.98], P = .01) scores. Reticulation was observed in 87% (71 of 82), whereas bronchiectasis was observed in 34% of all patients (28 of 82) (Table 3). Average reticulation and bronchiectasis scores were most elevated in the basilar regions of the chest (Fig 1).

Table Graphic Jump Location
TABLE 3 ]  Scores for Individual Chest HRCT Scan Patterns Obtained at 14 Days After ARDS Diagnosis

IQR = interquartile range. See Table 1 legend for expansion of other abbreviation.

Figure Jump LinkFigure 1 –  Evidence of fibroproliferative changes on chest HRCT scan were quantitated by radiologists from the level of the aortic arch (slice 1) to 1 cm above the dome of the right hemidiaphragm (slice 5) and at three additional levels spaced equally between these two levels (slices 2-4). On average, fibroproliferative changes were most prominent in slices 3, 4, and 5 although they were observed in all slices examined. A, Bars represent mean scores for reticulation, with 95% CIs. B, Bars represent mean scores for bronchiectasis, with 95% CIs. HRCT = high-resolution CT.Grahic Jump Location

Analyses were performed examining mechanical ventilation characteristics in relationship to 14-day chest HRCT scan individual scores (Table 4). Average daily PPlat exposure was associated with both HRCT scan reticulation (ρ = 0.52) and bronchiectasis (ρ = .42) (Figs 2A2C). Average Cstat was similarly associated with HRCT scan reticulation (ρ = −0.46) but less strongly with bronchiectasis (ρ = −0.31) (Figs 2D2F). Both PPlat exposure and Cstat were also associated with HRCT scan GGO (ρ = 0.48 and ρ = −0.44, respectively). Days on the ventilator correlated only with HRCT scan IPO scores (ρ = 0.55). Average daily PEEP correlated weakly with HRCT scan scores, whereas average daily Vt exposure did not correlate with any radiologic pattern.

Table Graphic Jump Location
TABLE 4 ]  Correlations Between Composite Ventilation Parameters Obtained During First 14 Days After ARDS Diagnosis and Chest HRCT Scan Scores

See Table 1 and 2 legends for expansion of abbreviations.

a 

P ≤ .02.

b 

P ≤ .0007.

Figure Jump LinkFigure 2 –  Ventilator parameters recorded during the first 14 d after ARDS diagnosis were examined in conjunction with HRCT scan radiologic involvement present at 14 (SD, 2) d after ARDS diagnosis. A, B, In this cohort of patients, the average plateau pressure exposure during the first 14 d after diagnosis correlated positively with HRCT scan reticulation scores (A, ρ = 0.52) and bronchiectasis scores (B, ρ = 0.42). C, If patients were stratified according to having no bronchiectasis vs any bronchiectasis, average plateau pressure exposure among the group with any bronchiectasis was significantly higher (P < .0001). D, E, Complementary to this, the average static compliance for patients over the first 14 d after diagnosis calculated using tidal volume, plateau pressure, and positive end-expiratory pressure correlated inversely with HRCT scan reticulation scores (D, ρ = −0.46) and bronchiectasis scores (E, ρ = −0.31). F, The presence of any bronchiectasis on HRCT scan was associated with a lower average static compliance (P = .007). Horizontal bars indicate mean values; diagonal bars are regression lines. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

Secondary analyses were performed examining correlations between highest single recorded ventilator measurement obtained during the first 14 days after ARDS diagnosis and the HRCT scan scores. Similar to what was observed with average exposure values, the single highest PPlat measurement was associated with 14-day reticulation (ρ = 0.33) and bronchiectasis (ρ = 0.33) scores (Fig 3A, Table 5). The highest single recorded PPlat value was also greater among patients with ARDS who exhibited any bronchiectasis (P < .01) (Figure 3B). Finally, we assessed the earliest ventilator measurements obtained after enrollment in association with the 14-day HRCT scan scores. Modest correlations between the earliest PPlat (ρ = 0.32) and Cstat (ρ = −0.38) values and HRCT scan reticulation were observed (Figs 3C, 3E, Table 5). Similarly, the first recorded PPlat value was higher, and first Cstat lower, among patients who exhibited any bronchiectasis on the 14-day HRCT scan (Figs 3D, 3F).

Figure Jump LinkFigure 3 –  The highest single plateau pressure, and the first available recorded plateau pressure or static compliance within 24 h of ARDS diagnosis, were examined in conjunction with HRCT scan reticulation and bronchiectasis present at 14 d (SD, 2) after ARDS diagnosis. A, In this cohort of patients, the highest plateau pressure exposure was associated with 14-d chest HRCT scan reticulation score (ρ = 0.33). B, Among patients who exhibited any bronchiectasis on their 14-d HRCT scan, on average, the highest single plateau pressure value recorded in the days prior to the HRCT scan was higher (P < .01). C, D, Similar relationships between the earliest single plateau pressure exposure obtained within 24 h of ARDS diagnosis and reticulation (C, ρ = 0.32) or bronchiectasis (D, P < .04) were also observed. E, F, Complementary to the earliest plateau pressure/HRCT scan relationships, the first recorded static respiratory compliance was inversely associated with HRCT scan reticulation (E, ρ = −0.38) and was also determined to be lower in patients with evidence of bronchiectasis (F, P = .008). Horizontal bars indicate mean values; diagonal bars are regression lines. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Table Graphic Jump Location
TABLE 5 ]  Correlations Between Single Time Point Mechanical Ventilation Parameters Obtained During the First 14 Days After ARDS Diagnosis and Chest HRCT Scan Scores

See Table 1 and 2 legends for expansion of abbreviations.

a 

Highest values recorded on any single day up until the 14-d chest HRCT scan.

b 

First values recorded within the first 24 h after ARDS diagnosis. Complete data available for 53 subjects at this first time point.

c 

P ≤ .006.

d 

P ≤ .04.

Multiple linear regression analyses were performed to clarify the relationship between mechanical ventilation characteristics and individual chest HRCT scan scores. In the first model, average Cstat prior to chest HRCT scan remained independently associated with reticulation (β = −0.13, P < .0001), accounting for APACHE II scores, number of ventilator days, initial Pao2/Fio2 ratio, and diagnosis of pneumonia (R2 = 0.33 for model). In a separate model containing average PEEP, PPlat, and Vt, along with these same clinical variables, PPlat exposure remained independently associated with reticulation after accounting for other variables in the model (β = 0.095, P < .0001; R2 = 0.36 for model). In multiple logistic regression analysis, average PPlat exposure (β = 0.29, P = .001) and a diagnosis of pneumonia (β = 0.74, P = .04) were associated with the finding of any HRCT scan bronchiectasis, accounting for these same clinical variables in the model (R2 = 0.25 for model).

Finally, given prior correlations between excessive fibroproliferation and poorer ARDS outcomes, we examined HRCT scan reticulation and bronchiectasis scores in conjunction with outcome data from patients with ARDS who remained on the ventilator > 14 days (n = 40), excluding spontaneously breathing patients from these analyses. Fourteen-day chest HRCT scan reticulation scores did not correlate with ventilator-free (ρ = 0.12) or ICU-free (ρ = −0.17) days. Similarly, bronchiectasis scores did not correlate with either ventilator-free (ρ = −0.23) or ICU-free (ρ = −0.27) days.

Our present investigations in a cohort of patients with ARDS who were closely managed on a low Vt ventilator protocol in the context of a clinical trial19 revealed associations between measured PPlat and respiratory compliance with 14-day chest HRCT scan reticulation and bronchiectasis. Associations were consistent whether average PPlat or unique PPlat values were used and in multivariate analyses. Our observations suggest that poor lung compliance, reflected in higher PPlats within days after ARDS diagnosis, corresponds with evidence of enhanced fibroproliferative activity, manifest by increased radiographic reticulation and bronchiectasis on HRCT scan. This work is distinct from previous investigations in which chest CT imaging has been used to assess lung recruitability in the setting of ARDS to adjust the application of PEEP.21,24,25 Our findings suggest the potential of early HRCT scan evaluation in patients with ARDS with the poorest lung compliance to identify individuals who may be at risk for complications related to excessive fibroproliferation.6

Emerging evidence suggests that HRCT scan scores measured very early (within days) after ARDS diagnosis using validated methods7,8 may be helpful in prognosticating short-term outcomes, illustrated in a prospective observational study9 in which evidence for chest HRCT scan fibroproliferation present on the first day of ARDS diagnosis was associated with 60-day mortality. However, our recent investigations suggest that chest HRCT scans obtained at 14 days post ARDS diagnosis identifies a subset of patients who will have persistent radiographic fibroproliferation much later, at 180 days, and also poorer pulmonary-specific quality of life.6 Data from our present work complement these prior observations by demonstrating an association between poor respiratory compliance measured in the first days after ARDS diagnosis with radiographic fibroproliferation at 14 days. These data are clinically meaningful in suggesting that simple bedside ventilator measurements early in ARDS may help to identify and target patients who are most likely to have radiologic fibroproliferation during the resolution of their acute illness; that is, patients with the poorest respiratory compliance. If substantial fibroproliferation is in fact evident on HRCT scan, appropriate follow-up and therapy may be implemented in this selected subset of ARDS survivors to optimize health-related quality of life. In contrast, patients who have relatively normal or rapidly improving respiratory compliance would not be targeted for HRCT imaging, as their likelihood of future pulmonary compromise would likely be low.

Our present work is not without limitations. Although the data support associations between ventilator parameters and radiologic fibroproliferation, this does not indicate causality. The observation that poorer compliance was also associated with HRCT scan GGO and IPO suggests that factors other than fibroproliferation, such as specific ARDS risk factors (eg, pneumonia) may play a role. Moreover, biologic evidence of excessive profibrotic activity would help substantiate a cause-and-effect relationship between the ventilator and fibroproliferation. Patients we included from the parent study were racially homogeneous and had a high survival rate compared with general patients with ARDS. In comparison, patients from the parent study who were excluded from the current investigation had a significantly greater severity of illness, poorer initial lung compliance, and higher mortality. However, the distribution of ARDS severity by initial Pao2/Fio2 values between patients included or excluded from the parent study did not differ and was relatively comparable to other ARDS studies. Moreover, both included and excluded patients were cared for using a stringent ventilator management and weaning strategy as part of a larger randomized controlled trial; this should provide further confidence in the validity of the observations. In summary, the unique characteristics of our study population raise at least two important clinical considerations. First, the fact that the more severely ill patients with ARDS with poorer lung compliance, and those who died before day 14, were excluded from our investigations suggests the possibility that the magnitude of the relationship we report between lung compliance and chest HRCT scan fibroproliferation is an underestimation. Second, the inability to safely perform HRCT imaging in the sickest patients with ARDS perhaps indicates a limitation in the use of this technology for prognostication to a subset of patients who are less ill and who can, therefore, travel safely to obtain the imaging study.

In our cohort of patients with ARDS, diminished respiratory compliance was associated with evidence of increased chest HRCT scan reticulation and bronchiectasis at 14 days after ARDS diagnosis. Specific factors in ventilator management, biologic mediators, and patient characteristics driving fibroproliferation remain to be established. However, HRCT imaging represents a promising tool to identify and quantify fibroproliferative activity in the acute setting and to help predict who may be at risk for poorer outcomes or impaired quality of life.

Author contributions: E. L. B. takes responsibility for the content of the manuscript, including the data and analyses; had full access to all the data in the study; and takes responsibility for the integrity of the data and the accuracy of the data analysis. E. L. B. contributed to the drafting of the submitted article; E. L. B. and M. M. contributed to conception and design; E. L. B., R. C. H., R. P., A. M. K., L. E. Q., M. M., and T. J. S. contributed to acquisition of data; E. L. B., D. L., D. C.-E., and M. M. contributed to data analysis and interpretation; R. C. H., R. P., A. M. K., L. E. Q., D. L., D. C.-E., M. M., T. J. S. contributed to the critical revision of the manuscript for important intellectual content and provided final approval of the version to be published; and E. L. B., R. C. H., R. P., A. M. K., L. E. Q., D. L., D. C.-E., M. M., and T. J. S. agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Lynch’s institution and laboratory receive research support from the National Heart Lung and Blood Institute; Siemens Corporation; Parexel Inc; and Janssen Biotech, Inc (formerly Centocor Biotech, Inc). Dr Lynch is a consultant to Parexel Inc; Boehringer Ingelheim GmbH; Genentech, Inc; Gilead; Veracyte, Inc; and InterMunec. Drs Burnham, Hyzy, Paine, Kelly, Quint, Curran-Everett, Moss, and Standiford have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Role of sponsors: The sponsor provided funding for the parent study but had no role in the development of this manuscript.

Additional information: The e-Table can be found in the Supplemental Materials section of the online article.

APACHE

Acute Physiology and Chronic Health Evaluation

Cstat

static respiratory compliance

GGO

ground-glass opacification

HRCT

high-resolution CT

IPO

intense parenchymal opacification

PEEP

positive end-expiratory pressure

PPlat

plateau pressure

SOFA

Sequential Organ Failure Assessment

V. ecorr

corrected minute ventilation

Vt

tidal volume

The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342(18):1301-1308. [CrossRef] [PubMed]
 
Papazian L, Doddoli C, Chetaille B, et al. A contributive result of open-lung biopsy improves survival in acute respiratory distress syndrome patients. Crit Care Med. 2007;35(3):755-762. [CrossRef] [PubMed]
 
Meduri GU, Headley AS, Golden E, et al. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome: a randomized controlled trial. JAMA. 1998;280(2):159-165. [CrossRef] [PubMed]
 
Steinberg KP, Hudson LD, Goodman RB, et al; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. Efficacy and safety of corticosteroids for persistent acute respiratory distress syndrome. N Engl J Med. 2006;354(16):1671-1684. [CrossRef] [PubMed]
 
Martin C, Papazian L, Payan MJ, Saux P, Gouin F. Pulmonary fibrosis correlates with outcome in adult respiratory distress syndrome. A study in mechanically ventilated patients. Chest. 1995;107(1):196-200. [CrossRef] [PubMed]
 
Burnham EL, Hyzy RC, Paine R III, et al. Chest computed tomography features are associated with poorer quality of life in acute lung injury survivors. Crit Care Med. 2013;;41(2)445-456.
 
Ichikado K, Suga M, Müller NL, et al. Acute interstitial pneumonia: comparison of high-resolution computed tomography findings between survivors and nonsurvivors. Am J Respir Crit Care Med. 2002;165(11):1551-1556. [CrossRef] [PubMed]
 
Ichikado K, Suga M, Muranaka H, et al. Prediction of prognosis for acute respiratory distress syndrome with thin-section CT: validation in 44 cases. Radiology. 2006;238(1):321-329. [CrossRef] [PubMed]
 
Ichikado K, Muranaka H, Gushima Y, et al. Fibroproliferative changes on high-resolution CT in the acute respiratory distress syndrome predict mortality and ventilator dependency: a prospective observational cohort study. BMJ Open. 2012;2(2):e000545. [CrossRef] [PubMed]
 
Cabrera-Benítez NE, Parotto M, Post M, et al. Mechanical stress induces lung fibrosis by epithelial-mesenchymal transition. Crit Care Med. 2012;40(2):510-517. [CrossRef] [PubMed]
 
Caruso P, Meireles SI, Reis LF, Mauad T, Martins MA, Deheinzelin D. Low tidal volume ventilation induces proinflammatory and profibrogenic response in lungs of rats. Intensive Care Med. 2003;29(10):1808-1811. [CrossRef] [PubMed]
 
Hirsch J, Hansen KC, Sapru A, et al. Impact of low and high tidal volumes on the rat alveolar epithelial type II cell proteome. Am J Respir Crit Care Med. 2007;175(10):1006-1013. [CrossRef] [PubMed]
 
Budinger GR, Chandel NS, Donnelly HK, Eisenbart J, Oberoi M, Jain M. Active transforming growth factor-beta1 activates the procollagen I promoter in patients with acute lung injury. Intensive Care Med. 2005;31(1):121-128. [CrossRef] [PubMed]
 
dos Santos CC, Slutsky AS. Protective ventilation of patients with acute respiratory distress syndrome. Crit Care. 2004;8(3):145-147. [CrossRef] [PubMed]
 
Marshall RP, Bellingan G, Webb S, et al. Fibroproliferation occurs early in the acute respiratory distress syndrome and impacts on outcome. Am J Respir Crit Care Med. 2000;162(5):1783-1788. [CrossRef] [PubMed]
 
Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and meta-analysis. JAMA. 2010;303(9):865-873. [CrossRef] [PubMed]
 
Santa Cruz R, Rojas JI, Nervi R, Heredia R, Ciapponi A. High versus low positive end-expiratory pressure (PEEP) levels for mechanically ventilated adult patients with acute lung injury and acute respiratory distress syndrome. Cochrane Database Syst Rev. 2013;6:CD009098. [PubMed]
 
Cressoni M, Cadringher P, Chiurazzi C, et al. Lung inhomogeneity in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med. 2014;189(2):149-158. [PubMed]
 
Paine R III, Standiford TJ, Dechert RE, et al. A randomized trial of recombinant human granulocyte-macrophage colony stimulating factor for patients with acute lung injury. Crit Care Med. 2012;40(1):90-97. [CrossRef] [PubMed]
 
Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med. 1994;149(3 pt 1):818-824. [CrossRef] [PubMed]
 
Crotti S, Mascheroni D, Caironi P, et al. Recruitment and derecruitment during acute respiratory failure: a clinical study. Am J Respir Crit Care Med. 2001;164(1):131-140. [CrossRef] [PubMed]
 
Desai SR, Wells AU, Rubens MB, Evans TW, Hansell DM. Acute respiratory distress syndrome: CT abnormalities at long-term follow-up. Radiology. 1999;210(1):29-35. [CrossRef] [PubMed]
 
Ranieri VM, Rubenfeld GD, Thompson BT, et al; ARDS Definition Task Force. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307(23):2526-2533. [PubMed]
 
Chiumello D, Marino A, Brioni M, et al. Visual anatomical lung CT scan assessment of lung recruitability. Intensive Care Med. 2013;39(1):66-73. [CrossRef] [PubMed]
 
Gattinoni L, Caironi P, Cressoni M, et al. Lung recruitment in patients with the acute respiratory distress syndrome. N Engl J Med. 2006;354(17):1775-1786. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1 –  Evidence of fibroproliferative changes on chest HRCT scan were quantitated by radiologists from the level of the aortic arch (slice 1) to 1 cm above the dome of the right hemidiaphragm (slice 5) and at three additional levels spaced equally between these two levels (slices 2-4). On average, fibroproliferative changes were most prominent in slices 3, 4, and 5 although they were observed in all slices examined. A, Bars represent mean scores for reticulation, with 95% CIs. B, Bars represent mean scores for bronchiectasis, with 95% CIs. HRCT = high-resolution CT.Grahic Jump Location
Figure Jump LinkFigure 2 –  Ventilator parameters recorded during the first 14 d after ARDS diagnosis were examined in conjunction with HRCT scan radiologic involvement present at 14 (SD, 2) d after ARDS diagnosis. A, B, In this cohort of patients, the average plateau pressure exposure during the first 14 d after diagnosis correlated positively with HRCT scan reticulation scores (A, ρ = 0.52) and bronchiectasis scores (B, ρ = 0.42). C, If patients were stratified according to having no bronchiectasis vs any bronchiectasis, average plateau pressure exposure among the group with any bronchiectasis was significantly higher (P < .0001). D, E, Complementary to this, the average static compliance for patients over the first 14 d after diagnosis calculated using tidal volume, plateau pressure, and positive end-expiratory pressure correlated inversely with HRCT scan reticulation scores (D, ρ = −0.46) and bronchiectasis scores (E, ρ = −0.31). F, The presence of any bronchiectasis on HRCT scan was associated with a lower average static compliance (P = .007). Horizontal bars indicate mean values; diagonal bars are regression lines. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location
Figure Jump LinkFigure 3 –  The highest single plateau pressure, and the first available recorded plateau pressure or static compliance within 24 h of ARDS diagnosis, were examined in conjunction with HRCT scan reticulation and bronchiectasis present at 14 d (SD, 2) after ARDS diagnosis. A, In this cohort of patients, the highest plateau pressure exposure was associated with 14-d chest HRCT scan reticulation score (ρ = 0.33). B, Among patients who exhibited any bronchiectasis on their 14-d HRCT scan, on average, the highest single plateau pressure value recorded in the days prior to the HRCT scan was higher (P < .01). C, D, Similar relationships between the earliest single plateau pressure exposure obtained within 24 h of ARDS diagnosis and reticulation (C, ρ = 0.32) or bronchiectasis (D, P < .04) were also observed. E, F, Complementary to the earliest plateau pressure/HRCT scan relationships, the first recorded static respiratory compliance was inversely associated with HRCT scan reticulation (E, ρ = −0.38) and was also determined to be lower in patients with evidence of bronchiectasis (F, P = .008). Horizontal bars indicate mean values; diagonal bars are regression lines. See Figure 1 legend for expansion of abbreviation.Grahic Jump Location

Tables

Table Graphic Jump Location
TABLE 1 ]  Baseline Characteristics of Patients With ARDS

N = 82 with 14-d chest HRCT scans and complete follow-up. Values are means (SD) unless otherwise noted. Earliest values available for patients after initiation of mechanical ventilation presented. APACHE = Acute Physiology and Chronic Health Evaluation; HRCT = high-resolution CT; SOFA = Sequential Organ Failure Assessment.

a 

Minute ventilation calculated correcting for Paco2 using the following equation: [minute ventilation × Paco2]/40. Severe ARDS categorized as minute ventilation > 10 L/min.

b 

Static compliance calculated with plateau, positive end-expiratory pressure, and tidal volume values collected within 24 h of ARDS diagnosis. Severe ARDS categorized as static compliance < 40.

Table Graphic Jump Location
TABLE 2 ]  Ventilator Characteristics of Enrolled Patients

Values are means (SD). Cstat = static respiratory compliance; n/a = not applicable; PEEP = positive end-expiratory pressure; PPlat = plateau pressure; Vt = tidal volume.

a 

Data from 53 subjects with ARDS available at this time point.

Table Graphic Jump Location
TABLE 3 ]  Scores for Individual Chest HRCT Scan Patterns Obtained at 14 Days After ARDS Diagnosis

IQR = interquartile range. See Table 1 legend for expansion of other abbreviation.

Table Graphic Jump Location
TABLE 4 ]  Correlations Between Composite Ventilation Parameters Obtained During First 14 Days After ARDS Diagnosis and Chest HRCT Scan Scores

See Table 1 and 2 legends for expansion of abbreviations.

a 

P ≤ .02.

b 

P ≤ .0007.

Table Graphic Jump Location
TABLE 5 ]  Correlations Between Single Time Point Mechanical Ventilation Parameters Obtained During the First 14 Days After ARDS Diagnosis and Chest HRCT Scan Scores

See Table 1 and 2 legends for expansion of abbreviations.

a 

Highest values recorded on any single day up until the 14-d chest HRCT scan.

b 

First values recorded within the first 24 h after ARDS diagnosis. Complete data available for 53 subjects at this first time point.

c 

P ≤ .006.

d 

P ≤ .04.

References

The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342(18):1301-1308. [CrossRef] [PubMed]
 
Papazian L, Doddoli C, Chetaille B, et al. A contributive result of open-lung biopsy improves survival in acute respiratory distress syndrome patients. Crit Care Med. 2007;35(3):755-762. [CrossRef] [PubMed]
 
Meduri GU, Headley AS, Golden E, et al. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome: a randomized controlled trial. JAMA. 1998;280(2):159-165. [CrossRef] [PubMed]
 
Steinberg KP, Hudson LD, Goodman RB, et al; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. Efficacy and safety of corticosteroids for persistent acute respiratory distress syndrome. N Engl J Med. 2006;354(16):1671-1684. [CrossRef] [PubMed]
 
Martin C, Papazian L, Payan MJ, Saux P, Gouin F. Pulmonary fibrosis correlates with outcome in adult respiratory distress syndrome. A study in mechanically ventilated patients. Chest. 1995;107(1):196-200. [CrossRef] [PubMed]
 
Burnham EL, Hyzy RC, Paine R III, et al. Chest computed tomography features are associated with poorer quality of life in acute lung injury survivors. Crit Care Med. 2013;;41(2)445-456.
 
Ichikado K, Suga M, Müller NL, et al. Acute interstitial pneumonia: comparison of high-resolution computed tomography findings between survivors and nonsurvivors. Am J Respir Crit Care Med. 2002;165(11):1551-1556. [CrossRef] [PubMed]
 
Ichikado K, Suga M, Muranaka H, et al. Prediction of prognosis for acute respiratory distress syndrome with thin-section CT: validation in 44 cases. Radiology. 2006;238(1):321-329. [CrossRef] [PubMed]
 
Ichikado K, Muranaka H, Gushima Y, et al. Fibroproliferative changes on high-resolution CT in the acute respiratory distress syndrome predict mortality and ventilator dependency: a prospective observational cohort study. BMJ Open. 2012;2(2):e000545. [CrossRef] [PubMed]
 
Cabrera-Benítez NE, Parotto M, Post M, et al. Mechanical stress induces lung fibrosis by epithelial-mesenchymal transition. Crit Care Med. 2012;40(2):510-517. [CrossRef] [PubMed]
 
Caruso P, Meireles SI, Reis LF, Mauad T, Martins MA, Deheinzelin D. Low tidal volume ventilation induces proinflammatory and profibrogenic response in lungs of rats. Intensive Care Med. 2003;29(10):1808-1811. [CrossRef] [PubMed]
 
Hirsch J, Hansen KC, Sapru A, et al. Impact of low and high tidal volumes on the rat alveolar epithelial type II cell proteome. Am J Respir Crit Care Med. 2007;175(10):1006-1013. [CrossRef] [PubMed]
 
Budinger GR, Chandel NS, Donnelly HK, Eisenbart J, Oberoi M, Jain M. Active transforming growth factor-beta1 activates the procollagen I promoter in patients with acute lung injury. Intensive Care Med. 2005;31(1):121-128. [CrossRef] [PubMed]
 
dos Santos CC, Slutsky AS. Protective ventilation of patients with acute respiratory distress syndrome. Crit Care. 2004;8(3):145-147. [CrossRef] [PubMed]
 
Marshall RP, Bellingan G, Webb S, et al. Fibroproliferation occurs early in the acute respiratory distress syndrome and impacts on outcome. Am J Respir Crit Care Med. 2000;162(5):1783-1788. [CrossRef] [PubMed]
 
Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and meta-analysis. JAMA. 2010;303(9):865-873. [CrossRef] [PubMed]
 
Santa Cruz R, Rojas JI, Nervi R, Heredia R, Ciapponi A. High versus low positive end-expiratory pressure (PEEP) levels for mechanically ventilated adult patients with acute lung injury and acute respiratory distress syndrome. Cochrane Database Syst Rev. 2013;6:CD009098. [PubMed]
 
Cressoni M, Cadringher P, Chiurazzi C, et al. Lung inhomogeneity in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med. 2014;189(2):149-158. [PubMed]
 
Paine R III, Standiford TJ, Dechert RE, et al. A randomized trial of recombinant human granulocyte-macrophage colony stimulating factor for patients with acute lung injury. Crit Care Med. 2012;40(1):90-97. [CrossRef] [PubMed]
 
Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med. 1994;149(3 pt 1):818-824. [CrossRef] [PubMed]
 
Crotti S, Mascheroni D, Caironi P, et al. Recruitment and derecruitment during acute respiratory failure: a clinical study. Am J Respir Crit Care Med. 2001;164(1):131-140. [CrossRef] [PubMed]
 
Desai SR, Wells AU, Rubens MB, Evans TW, Hansell DM. Acute respiratory distress syndrome: CT abnormalities at long-term follow-up. Radiology. 1999;210(1):29-35. [CrossRef] [PubMed]
 
Ranieri VM, Rubenfeld GD, Thompson BT, et al; ARDS Definition Task Force. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307(23):2526-2533. [PubMed]
 
Chiumello D, Marino A, Brioni M, et al. Visual anatomical lung CT scan assessment of lung recruitability. Intensive Care Med. 2013;39(1):66-73. [CrossRef] [PubMed]
 
Gattinoni L, Caironi P, Cressoni M, et al. Lung recruitment in patients with the acute respiratory distress syndrome. N Engl J Med. 2006;354(17):1775-1786. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Supporting Data

Online Supplement

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543