Skip to main content
PMC home page
American Journal of Respiratory and Critical Care Medicine logoLink to American Journal of Respiratory and Critical Care Medicine
. 2018 Jun 15;197(12):1575–1585. doi: 10.1164/rccm.201712-2530OC

Early Intravascular Events Are Associated with Development of Acute Respiratory Distress Syndrome. A Substudy of the LIPS-A Clinical Trial

Raja-Elie E Abdulnour 1,*, Tina Gunderson 2,*, Ioanna Barkas 1, Jack Y Timmons 1, Cindy Barnig 1,3, Michelle Gong 4,5, Daryl J Kor 6, Ognjen Gajic 7, Daniel Talmor 8, Rickey E Carter 2, Bruce D Levy 1,✉
PMCID: PMC6006404  PMID: 29782179

Abstract

Rationale: Acute respiratory distress syndrome (ARDS) is a devastating illness with limited therapeutic options. A better understanding of early biochemical and immunological events in ARDS could inform the development of new preventive and treatment strategies.

Objectives: To determine select peripheral blood lipid mediator and leukocyte responses in patients at risk for ARDS.

Methods: Patients at risk for ARDS were randomized as part of a multicenter, double-blind clinical trial of aspirin versus placebo (the LIPS-A [Lung Injury Prevention Study with Aspirin] trial; NCT01504867). Plasma thromboxane B2 (TXB2), aspirin-triggered lipoxin A4 (15-epi-LXA4, ATL), and peripheral blood leukocyte number and activation were determined on enrollment and after treatment with either aspirin or placebo.

Measurements and Main Results: Thirty-three of 367 subjects (9.0%) developed ARDS after randomization. Baseline ATL levels, total monocyte counts, intermediate monocyte counts, and monocyte–platelet aggregates were associated with the development of ARDS. Peripheral blood neutrophil count and monocyte–platelet aggregates significantly decreased over time. Of note, nine subjects developed ARDS after randomization yet before study drug initiation, including seven subjects assigned to aspirin treatment. Subjects without ARDS at the time of first dose demonstrated a lower incidence of ARDS with aspirin treatment. Compared with placebo, aspirin significantly decreased TXB2 and increased the ATL/TXB2 ratio.

Conclusions: Biomarkers of intravascular monocyte activation in at-risk patients were associated with development of ARDS. The potential clinical benefit of early aspirin for prevention of ARDS remains uncertain. Together, results of the biochemical and immunological analyses provide a window into the early pathogenesis of human ARDS and represent potential vascular biomarkers of ARDS risk.

Clinical trial registered with www.clinicaltrials.gov (NCT01504867).

Keywords: ARDS; monocyte–platelet aggregates; aspirin; lipoxin; thromboxane

At a Glance Commentary

Scientific Knowledge on the Subject

Leukocyte and platelet activation are associated with acute respiratory distress syndrome (ARDS) pathogenesis, and aspirin exposure may decrease risk for ARDS. The intention-to-treat analysis in the LIPS-A (Lung Injury Prevention Study with Aspirin) prevention trial of aspirin to decrease ARDS incidence did not show benefit, but the strength of this finding was limited by low event rates.

What This Study Adds to the Field

In the LIPS-A trial, biochemical and immunological markers of vascular monocyte activation were associated with the development of ARDS, including increased levels of 15-epi-LXA4 (aspirin-triggered lipoxin A4), total monocyte counts, intermediate monocyte counts, and monocyte–platelet aggregates. Subjects at risk for ARDS demonstrated aspirin treatment–dependent changes in the relative plasma levels of 15-epi-LXA4 and thromboxane B2, and a trend for decreased ARDS incidence.

Acute respiratory distress syndrome (ARDS) remains a life-threatening illness characterized by dysregulated inflammation and alveolar-capillary barrier disruption (1). Given the limited number of effective treatments for established ARDS, identification of early molecular and cellular events associated with development of ARDS can complement clinical and immunological markers of established ARDS (2, 3) and may lead to therapeutic strategies aimed at ARDS prevention (4).

Preclinical investigation has identified pathogenic roles for vascular inflammatory responses in the development of acute lung injury. Intravascular neutrophil and platelet activation result in heterotypic neutrophil–platelet aggregate (Ne-PA) formation, with subsequent secondary capture of neutrophils, endothelial cell activation, and barrier dysfunction (5). These events are regulated in part by lipid mediators; thromboxane A2 (TXA2) released by platelets promotes neutrophil adhesion to endothelial cells (5, 6), whereas the specialized proresolving mediator lipoxin A4 (LXA4) and its epimer 15-epi-LXA4 (aspirin-triggered LXA4 [ATL]), produced during neutrophil–platelet interactions by transcellular biosynthesis (7, 8), restrain neutrophil activation (5). Platelet activation and aggregates have also been identified in other forms of lung disease and with hypoxemia (9, 10). Therefore, these intravascular events may serve as biomarkers of early vascular inflammation in ARDS.

Of interest, blocking Ne-PA formation mitigates experimental ARDS (5, 6, 11). Aspirin increases plasma levels of ATL relative to thromboxane B2 (TXB2, stable metabolite of TXA2) in humans (12) and is organ protective in select murine models of ARDS (5). Thus, harnessing these biochemical mechanisms could prevent ARDS.

To test the hypothesis that circulating bioactive lipid mediators and intravascular leukocyte and platelet activation are associated with the development of ARDS, plasma levels of TXB2 and ATL and flow cytometry parameters of circulating leukocyte and platelet activation were determined in patients at risk for ARDS who were enrolled in the LIPS-A (Lung Injury Prevention Study with Aspirin) trial (NCT01504867 [13]). Some of the results of these studies have been previously reported in the form of an abstract (14).

Methods

Study Design and Patient Selection

The biospecimens included in this study were obtained through a preplanned ancillary study to the LIPS-A clinical trial, a multicenter, double-blind, placebo-controlled, randomized clinical trial of aspirin or placebo in patients at risk for ARDS. The full study design, study procedures, and results of the primary study are published elsewhere (13, 15).

Sample Collection and Analysis

Peripheral blood was collected at baseline (after randomization, before administration of study drug), on study Day 1 (within 24 hours after the first dose), and on study Day 3.

Statistical Analysis

Descriptive data are summarized as mean (confidence interval [CI]), median (interquartile range [IQR]), or n (%). To estimate the discriminatory ability of the baseline variables for the development of ARDS, standard receiver operating characteristic (ROC) analyses were conducted in subjects who: 1) had baseline data present; 2) did not develop ARDS between randomization and 12 hours after baseline draw; 3) did not receive their first dose of medication before their baseline draw; and 4) had a Lung Injury Prediction Score (LIPS) greater than or equal to 4. To compare the longitudinal profiles of the biochemical markers over time, a linear mixed model was used. Post hoc contrasts were constructed from the model to compare differences between groups at each time point. For the overall test for differences, the treatment by time interaction was used. Additional details on study design, sample preparation and measurements, and statistical analyses are provided in the online supplement.

Results

Timeline for ARDS Development

To determine the time of ARDS diagnosis relative to the time of randomization and drug administration, a timeline was established. ARDS developed in 33 of 367 (9.0%) subjects enrolled in this substudy. Of these subjects, nine developed ARDS after randomization and before the intervention, including seven patients in the aspirin group (Figure 1A). Of the 24 subjects who developed ARDS after the first treatment dose, 10 patients developed ARDS after initiation of aspirin treatment and 14 patients after placebo (Figure 1B). Despite the low number of events, there was a trend for decreased ARDS after aspirin treatment (Figure 1B). Demographics were similar between aspirin- and placebo-exposed groups (Table 1), including median LIPS (6 [IQR, 5–7.5] vs. 5.5 [IQR, 4.5–7]).

Figure 1.

Figure 1.

Open in a new tab

Timeline of cohort and acute respiratory distress syndrome (ARDS) incidence. (A) ARDS development compared with time of randomization and first dose of study drug (placebo or loading dose of aspirin) for all subjects with flow cytometry or lipid mediator results. ARDS was adjudicated using modified Berlin criteria (see Methods). The latest of the qualifying criteria defined the time of ARDS development. (B) ARDS incidence after study drug loading dose.

Table 1.

Demographics of As-Treated Subjects with Usable Flow Cytometry or Lipid Samples by Treatment Arm Assignment

  Aspirin (n = 184) Placebo (n = 183) Total (n = 367) P Value
Age, median (Q1–Q3) 57 (44–67) 57 (46.5–68) 57 (45–68) 0.459
         
Sex, n (%)       0.567
 Female 85 (46.2) 91 (49.7) 176 (48)  
 Male 99 (53.8) 92 (50.3) 191 (52)  
         
Race, n (%)       0.641
 White 134 (72.8) 128 (69.9) 262 (71.4)  
 Black or African American 31 (16.8) 35 (19.1) 66 (18)  
 American Indian/Alaska 0 (0) 2 (1.09) 2 (0.545)  
 Native Asian 4 (2.17) 3 (1.64) 7 (1.91)  
 Native Hawaiian or other Pacific islander 0 (0) 0 (0) 0 (0)  
 Unknown or unreported ethnicity 15 (8.15) 15 (8.2) 30 (8.17)  
         
Ethnicity, n (%)       0.842
 Not Hispanic or Latino 151 (82.1) 148 (80.9) 299 (81.5)  
 Hispanic or Latino 19 (10.3) 18 (9.84) 37 (10.1)  
 Unknown 14 (7.61) 17 (9.29) 31 (8.45)  
         
LIPS, median (Q1–Q3) 6 (5–7.5) 5.5 (4.5–7) 6 (5–7) 0.304
Open in a new tab

Definition of abbreviations: LIPS = Lung Injury Prediction Score; Q = quartile.

Plasma Mediator Levels and Development of ARDS

To determine if ATL, TXB2, or the ATL/TXB2 ratio were associated with the development of ARDS, baseline plasma levels were compared between subjects who developed ARDS after the first blood draw and subjects who did not (No-ARDS). Of the 367 subjects with analyzable samples, 345 subjects were included in this analysis (see online supplement for inclusion criteria). ARDS occurred in 22 patients, whereas 323 patients did not develop ARDS, with similar demographics between the two groups (Table 2). Plasma ATL at baseline draw was median 5.29 ng/ml (IQR, 4.2–7.11 ng/ml) in ARDS (Figure 2A), which was significantly higher than in No-ARDS (4.03 ng/ml; IQR, 2.17–6.62 ng/ml; P = 0.04). The area under the ROC curve was 0.66 (CI, 0.57–0.75). Baseline plasma TXB2 levels (Figure 2B) and the ATL/TXB2 ratio (Figure 2C) were not significantly different with ARDS. The area under the ROC curve was 0.50 (CI, 0.37–0.63) for TXB2, and 0.52 (CI, 0.38–0.66) for the ATL/TXB2 ratio (Figures 2B and 2C). Together, these findings indicate that elevated baseline plasma ATL levels were associated with the development of ARDS.

Table 2.

Demographics by Acute Respiratory Distress Syndrome Status

  ARDS (n = 22) No ARDS (n = 323) Total (n = 345) P Value
Age, median (Q1–Q3) 52 (43.2–59.5) 57 (45–68) 56 (45–67) 0.262
         
Sex, n (%)       0.666
 Female 12 (54.5) 153 (47.4) 165 (47.8)  
 Male 10 (45.5) 170 (52.6) 180 (52.2)  
         
Race, n (%)       0.496
 White 13 (59.1) 238 (73.7) 251 (72.8)  
 Black or African American 6 (27.3) 53 (16.4) 59 (17.1)  
 American Indian/Alaska 0 (0) 2 (0.619) 2 (0.58)  
 Native Asian 0 (0) 5 (1.55) 5 (1.45)  
 Native Hawaiian or other Pacific islander 0 (0) 0 (0) 0 (0)  
 Unknown or unreported ethnicity 3 (13.6) 25 (7.74) 28 (8.12)  
         
Ethnicity, n (%)       0.338
 Not Hispanic or Latino 17 (77.3) 264 (81.7) 281 (81.4)  
 Hispanic or Latino 4 (18.2) 30 (9.29) 34 (9.86)  
 Unknown 1 (4.55) 29 (8.98) 30 (8.7)  
         
LIPS, median (Q1–Q3) 6 (5–6.88) 5.5 (5–7) 6 (5–7) 0.773
Open in a new tab

Definition of abbreviations: ARDS = acute respiratory distress syndrome; LIPS = Lung Injury Prediction Score; Q = quartile.

Figure 2.

Figure 2.

Open in a new tab

Baseline plasma aspirin-triggered lipoxin A4 (ATL) and thromboxane B2 (TXB2) levels before the development of acute respiratory distress syndrome (ARDS). The distribution of plasma (A) ATL, (B) TXB2, and (C) ATL/TXB2 at baseline in subjects who developed ARDS more than 12 hours after baseline draw versus those who never developed ARDS is shown. Insets: receiver operating characteristic curves for prediction of ARDS. Area under the curve and 95% confidence interval calculated using the Delong method and the Mann-Whitney test for differences between ARDS and no-ARDS group at baseline are provided. AUC = area under the curve; No-ARDS = did not develop ARDS.

Peripheral Blood Leukocyte Responses and Development of ARDS

Next, the relationship between peripheral blood leukocyte responses and the development of ARDS was determined. Baseline neutrophil counts (Figure 3A) and Ne-PA (Figure 3B) were not significantly different in the ARDS cohort. The area under the ROC curve was 0.66 (CI, 0.48–0.84) for neutrophil counts and 0.56 (CI, 0.34–0.77) for Ne-PA. Neutrophil CD11b surface expression measured as CD11b mean fluorescence intensity (MFI; Figure 3C) was also not associated with ARDS development, with an area under the ROC curve of 0.71 (CI, 0.59–0.84; P = 0.06). In contrast, baseline peripheral blood total monocyte counts were significantly increased in subjects who developed ARDS (median [IQR], 7.3 × 108 [4.4–10.8 × 108] vs. 3.7 × 108 [1.8–7.2 × 108]; P = 0.05), with an area under the ROC curve of 0.71 (CI, 0.53–0.90; Figure 3D). In addition, monocyte–platelet aggregates (Mo-PA) were significantly higher in ARDS (median [IQR], 72.4% [30.7–75%] vs. 16.8% [8.81–36.2%]; P = 0.02), and the area under the ROC curve was 0.74 (CI, 0.48–0.99; Figure 3E). Monocyte CD11b surface expression was not significantly different in the ARDS cohort, with an area under the ROC curve of 0.57 (CI, 0.36–0.78; Figure 3F). In addition to total monocytes, subjects who developed ARDS also had significantly higher intermediate monocyte (IntMo) counts (median [IQR], 2.2 × 108 [0.9–7.2 × 108] vs. 0.4 × 108 [0.2–1.3 × 108]; P = 0.02), with an area under the ROC curve of 0.74 (CI, 0.53–0.95) (Figure 3G). Increased IntMo-PA were observed in patients who developed ARDS (median [IQR], 69.8% [34–79%] vs. 31% [14.5–54.6%]; P = 0.07), with an area under the ROC curve of 0.70 (0.46–0.93; Figure 3H). Together, these findings indicate that the baseline peripheral blood total monocyte counts, IntMo counts, and circulating Mo-PA were associated with the development of ARDS.

Figure 3.

Figure 3.

Open in a new tab

Baseline whole-blood leukocyte counts and activation before the development of acute respiratory distress syndrome (ARDS). The distribution of (A) neutrophil counts, (B) neutrophil–platelet aggregates, (C) neutrophil CD11b surface expression, (D) monocyte counts, (E) monocyte–platelet aggregates, (F) monocyte CD11b surface expression, (G) intermediate monocyte counts, and (H) intermediate monocyte–platelet aggregates in whole blood of subjects who developed ARDS more than 12 hours after baseline draw versus those who never developed ARDS is shown. Insets: Receiver operating characteristic curve for prediction of ARDS. Area under the curve and 95% confidence interval calculated using the Delong method and the Mann-Whitney test for differences between ARDS and no-ARDS group at baseline are provided. Percent (%) parent in B, E, and H refers to total neutrophils, total monocytes, and intermediate monocytes, respectively. AUC = area under the curve; IntMo = intermediate monocyte; IntMo-PA = intermediate monocyte–platelet aggregates; MFI = mean fluorescence intensity; Mo = monocyte; Mo-PA = monocyte–platelet aggregates; Ne = neutrophil; Ne-PA = neutrophil–platelet aggregates; No-ARDS = did not develop ARDS.

Plasma Lipid Mediator Levels and Impact of Aspirin in Patients at Risk for ARDS

To determine time- and treatment-related changes in lipid mediator levels from patients at risk for ARDS, a linear mixed model analysis on as-treated subjects was performed. Baseline mean levels of TXB2 and ATL and the ATL/TXB2 ratio were similar between the aspirin and placebo groups (Figure 4). In the placebo cohort, plasma TXB2 was 0.66 ng/ml (CI, 0.55–0.8 ng/ml) at baseline, 0.63 ng/ml (CI, 0.51–0.77 ng/ml) at Day 1, and 0.88 ng/ml (CI, 0.70–0.11 ng/ml) at Day 3, which was significantly increased compared with baseline (Figure 4A). Baseline ATL was 4.01 ng/ml (CI, 3.52–4.57 ng/ml) and was not significantly different on Day 1 and Day 3 (Figure 4B). The placebo arm’s baseline plasma ATL/TXB2 ratio was 6.58 (CI, 5.42–8.00), then decreased to 5.6 (CI, 4.57–6.85) on Day 1 and 4.71 (CI, 3.73–5.95) on Day 3 (Figure 4C), which was significantly decreased from baseline. These findings indicate a time-dependent association of TXB2 plasma levels and the ATL/TXB2 ratio. Aspirin significantly decreased TXB2 relative to placebo, from 0.59 ng/ml (CI, 0.49–0.72 ng/ml) before treatment to 0.24 ng/ml (CI, 0.20–0.30 ng/ml) on Day 1 and 0.25 ng/ml (CI, 0.20–0.31 ng/ml) on Day 3 (Figure 4A). Plasma ATL levels did not significantly change with treatment (Figure 4B). As such, the ATL/TXB2 ratio increased after aspirin treatment; the ratio increased from pretreatment level of 5.61 (CI, 4.63–6.80) to 13.1 (CI, 10.7–16.0) on Day 1 and 13.8 (CI, 11.0–17.2) on Day 3 (Figure 4C). These findings indicate treatment-dependent changes in plasma TXB2 with aspirin that influenced arachidonic acid metabolism to increase the ATL/TXB2 ratio.

Figure 4.

Figure 4.

Open in a new tab

Aspirin and draw-time effect on plasma aspirin-triggered lipoxin A4 (ATL) and thromboxane B2 (TXB2). Plasma (A) TXB2, (B) ATL, and (C) ATL/TXB2 levels at baseline and after (Day 1 and Day 3) aspirin or placebo are shown. The inset shows linear mixed-effects model results for effect of treatment arm, time, and treatment × time interactions. P values listed below each draw time are for the between-group comparison at that draw based on the model estimates. Error bars for 90% Wald confidence intervals are shown. Units for ATL and TXB2 are ng/ml.

Peripheral Blood Neutrophil Responses in Patients at Risk for ARDS and Impact of Aspirin

Next, time- and treatment-related changes in leukocyte responses from patients at risk for ARDS were determined. Peripheral blood mean neutrophil counts, Ne-PA, and neutrophil activation (i.e., CD11b surface expression) were quantified. In the placebo cohort, initial neutrophil counts were 9.92 × 103 cells/μl (CI, 8.77–11.20 × 103 cells/μl) and significantly decreased over time to 8.3 × 103 cells/μl (CI, 7.33–9.32 × 103 cells/μl) on Day 1 and 6.3 × 103 cells/μl (CI, 5.58–7.22 × 103 cells/μl) on Day 3 (Figure 5A). The percent of total neutrophils that were Ne-PA was 4.53% (CI, 3.79–5.27%) at baseline, 4.35% (CI, 3.63–5.06%) on Day 1, and 4.31% (CI, 3.52–5.1%) on Day 3 (Figure 5B). Mean neutrophil CD11b surface expression did not significantly change from baseline to Days 1 and 3 (Figure 5C). Aspirin treatment did not significantly impact neutrophil counts, Ne-PA, and neutrophil CD11b expression (Figures 5A–5C); however, CD11b surface expression was significantly lower with aspirin at all time points, including at baseline before aspirin exposure (Figure 5C). These findings indicate a time-dependent effect on peripheral blood neutrophil counts that is likely independent of aspirin administration.

Figure 5.

Figure 5.

Open in a new tab

Whole-blood leukocyte counts and activation. Whole-blood (A) neutrophil count, (B) neutrophil–platelet aggregates, (C) neutrophil CD11b surface expression, (D) monocyte count, (E) monocyte–platelet aggregates, and (F) monocyte CD11b surface expression at baseline and after (Day 1 and Day 3) aspirin or placebo are shown. The inset shows linear mixed-effects model results for effect of treatment arm, time, and treatment × time interactions. P values listed below each draw time are for the between-group comparison at that draw based on the model estimates. Error bars for 90% Wald confidence intervals are shown. Percent (%) parent in B and E refers to total neutrophils and monocytes, respectively. MFI = mean fluorescence intensity; Mo = monocyte; Mo-PA = monocyte–platelet aggregates; Ne = neutrophil; Ne-PA = neutrophil–platelet aggregates.

Peripheral Blood Monocyte Responses in Patients at Risk for ARDS and Impact of Aspirin

Next, peripheral blood total monocyte counts, Mo-PA, and mean CD11b surface expression were determined. In the placebo cohort, peripheral blood monocyte counts were not significantly different during the course of the study (Figure 5D). The mean percent of total monocytes that were Mo-PA was 26.1% at baseline (CI, 22.1–30.1%), 24.5% on Day 1 (CI, 20.6–28.4%), and 19.3% on Day 3 (CI, 15.1–23.6%), which was significantly decreased compared with baseline (Figure 5E). Monocyte CD11b surface expression did not significantly change from baseline to Days 1 and 3 (Figure 5F). Aspirin treatment did not significantly change total monocyte counts, Mo-PA, or monocyte CD11b expression (Figures 5D–5F); however, like neutrophil CD11b MFI, monocyte CD11b MFI was different between treatment groups at all three time points, including at baseline before aspirin exposure (Figure 5F). Together, these results indicate time-dependent changes in peripheral blood Mo-PA that are likely independent of aspirin administration.

Plasma Mediator and Cellular Correlation Analysis

To uncover potential relationships between the measures of biochemical mediators and peripheral blood cells analyzed here, determinations from all available samples were used to calculate correlations (see Methods; Figure 6). Plasma TXB2 and ATL/TXB2 were inversely related across samples (r = −0.68); however, there were no significant correlations between the lipid mediators and measures of cell counts or activation. Monocyte counts correlated positively with Mo-PA (r = 0.74), and CD11b surface expression on neutrophils correlated positively with CD11b surface expression on monocytes (r = 0.80). Of interest, Ne-PA and Mo-PA were inversely related (r = −0.60). Together, these correlations suggest that these lipid mediator levels and leukocyte numbers and activation were under distinct regulation.

Figure 6.

Figure 6.

Open in a new tab

Plasma lipid mediators and peripheral blood leukocyte responses correlogram. A Spearman correlation matrix between plasma lipids aspirin-triggered lipoxin A4 (ATL) and thromboxane B2 (TXB2) and whole-blood leukocyte markers in all samples is shown. Under the diagonal list of parameters, Spearman correlation values are shown for each pair of intersecting parameters. Above the diagonal, the correlations are represented as pie charts, with a full pie representing a correlation of +1.0 or −1.0. Color scale (right bar) shows Spearman correlation values >0 as blue and <0 as red. Percent (%) parent refers to total neutrophils (Ne) and monocytes (Mo). MFI = median fluorescence intensity; Mo-PA = monocyte–platelet aggregates; Ne-PA = neutrophil–platelet aggregates.

Discussion

Preclinical research has increased our understanding of intravascular events underlying ARDS pathogenesis (1). Patients are diagnosed with ARDS at the height of their illness, rendering investigation of disease pathogenesis in humans challenging. By its design, the LIPS-A trial identified patients at risk for development of ARDS, enabling investigation of early molecular and cellular pathogenesis. Results presented here demonstrate that peripheral blood total monocyte counts, Mo-PA, and plasma ATL at baseline were associated with ARDS development. The IntMo subset of total monocytes was also increased in patients who developed ARDS, and IntMo-PA were increased but did not reach statistical significance. There were time-dependent and treatment (aspirin)-dependent changes in plasma levels of TXB2 and the ratio of ATL/TXB2. Time-dependent, but not treatment-dependent, changes were also identified in peripheral blood neutrophil counts and Mo-PA. Collectively, these findings indicate an association between select intravascular responses and ARDS development and demonstrate aspirin-dependent regulation of intravascular arachidonic acid metabolism.

Timeline analysis of disease onset relative to randomization and treatment administration in the LIPS-A trial revealed that several patients developed ARDS after randomization but before study drug administration. Among these subjects, disproportionately more were assigned to receive aspirin. Exclusion of patients with established disease revealed a decreased incidence of ARDS with aspirin that nearly reached statistical significance, despite the overall low number of patients developing the disease. In view of the findings presented here, including the treatment effects on the biochemical markers, a beneficial effect for aspirin in ARDS prevention remains possible.

TXA2 generation by platelets plays an important role in inflammatory responses (16). Activated platelets can also release arachidonic acid for conversion to LXA4 (7), which, in contrast to TXA2, promotes the resolution of lung inflammation and injury in experimental models (17). Aspirin acetylation of cyclooxygenase (COX)-1 irreversibly inhibits platelet TXA2 production, whereas aspirin-acetylated COX-2 participates in production of 15-epi-LXA4 (i.e., ATL) (18). Preclinical studies of lung injury indicate a pathogenic role for TXA2 (19) and a protective role for ATL (20, 21), suggesting that the impact of aspirin on the relative levels of these lipid mediators (i.e., ATL/TXB2 ratio) could have accounted for differences in ARDS development with aspirin. As seen herein, TXB2 plasma levels were decreased by aspirin. In contrast, plasma ATL levels here were already higher than ATL plasma levels reported in healthy subjects (12) and not further increased with aspirin, indicating that aspirin acetylation of COX-2 alone was not sufficient to drive further increases in plasma ATL production. Although originally coined ATL, 15-epi-LXA4 is also produced in the absence of aspirin via arachidonic acid metabolism by cytochrome P450 enzymes and 5-lipoxygenase (22), so ATL detection in the placebo group likely resulted from activation of this biosynthetic pathway. In healthy subjects, aspirin treatment increases ATL levels (12), suggesting that increased oxidative stress in patients at risk for ARDS may have been a significant factor in limiting further increases in ATL production, similar to that seen in severe asthma (10). It is notable that increased plasma ATL levels at baseline were associated with development of ARDS, consistent with engagement of counterregulatory pathways that were ultimately insufficient to prevent the development of ARDS in some patients. As such, patients at risk for ARDS demonstrate specific plasma lipid mediator responses relative to healthy subjects that can be targeted for prevention or ARDS.

The role of neutrophil activation and recruitment to the lung in development of ARDS is well established (1). Secondary capture of neutrophils contributes in preclinical models to leukocyte recruitment in response to acute lung injury (6, 11, 21, 23) and represents a potential therapeutic target. Total neutrophil counts decreased over the course of the study, similar to time-dependent decrements in tumor necrosis factor-α, IL-1β, and IL-6 noted in LIPS-A (13). In contrast, Ne-PA did not demonstrate time-dependent changes in patients at risk for ARDS, and values remained within the expected range for healthy donors (24, 25), which may explain why aspirin did not impact Ne-PA in these at-risk patients, in distinction to findings from experimental lung injury in animal models (21). These results suggest a role for peripheral neutrophils in ARDS development, but the role of Ne-PA in the initiation of human ARDS requires additional investigation.

Human monocytes can be classified into “classical” CD14HiCD16Lo, “intermediate” CD14HiCD16Hi, and “non-classical” CD14LoCD16Hi subsets (26). Acute inflammation can rapidly decrease circulating monocyte counts, with subsequent mobilization of classical monocytes from the bone marrow that mature in the circulation to intermediate monocytes and then nonclassical monocytes (27, 28). These monocyte subsets differ in their inflammatory properties (29, 30). Pathogenic roles for monocytes in ARDS development are apparent in experimental models of acute lung injury. Peripheral blood monocytes infiltrate inflamed lungs in a CD11/CD18-dependent manner (31), and monocyte depletion before injury reduces measures of lung injury (32). Conversely, monocytes recruited to injured lung can promote resolution of inflammation and a return to homeostasis (33, 34). Mo-PA increase in patients with acute myocardial infarction and stroke, correlating with disease severity (25, 35). Platelet–monocyte interactions, including aggregate formation, can promote intercellular signaling with inflammatory mediator generation and alteration of the inflammatory milieu (16, 36), potentially driving the development of ARDS. Mo-PA are increased in patients diagnosed with ARDS and H1N1 (37) but had not yet been implicated in early in ARDS pathogenesis. Here, peripheral blood Mo-PA demonstrated time-dependent changes in patients at risk for ARDS and were higher than in healthy control subjects examined in Reference 25. Of note, Mo-PA were specifically associated with ARDS development. In addition, expansion of IntMo in patients who develop ARDS implies activation of unique signaling pathways related to monocyte responses. Together, these results suggest that peripheral blood total monocytes and Mo-PA could play previously unappreciated and pivotal roles early in development of ARDS and can serve as potential biomarkers for ARDS development. Future studies may enable more detailed functional analyses of Mo-PA and monocyte subsets.

Inhibition of TXA2 without significant increase in ATL was not sufficient to regulate peripheral blood leukocyte responses in early ARDS, emphasizing that antiinflammation does not equal proresolution (38). In view of the inability of aspirin to substantially increase ATL plasma levels, it is likely that direct administration of lipoxins or other specialized proresolving mediators would be needed to counter the strong prophlogistic factors in early ARDS and those at risk. Exogenous ATL can decrease platelet-activating factor–stimulated Ne-PA in whole blood (10), suggesting that pharmacological increases in plasma ATL levels could provide the counterregulation for pathogenic leukocyte–platelet aggregates.

Effective medical therapies for established ARDS are lacking, emphasizing the need for early identification of at-risk patients to facilitate the development of preventive strategies (4). The LIPS identifies patients at risk for developing ARDS (39), and here, a LIPS of 4 or higher successfully identified at-risk patients, albeit with a lower than expected ARDS incidence rate (39). The addition of functional biomarkers, such as Mo-PA, to the clinical parameters may increase both the LIPS sensitivity and specificity.

The biochemical circuits regulating TXA2 and ATL generation play critical roles in the pathogenesis and resolution of acute inflammatory responses in model systems (6, 7, 11, 38). Here, correlation analyses of the intravascular biochemical and immunological biomarkers demonstrated engagement of these pathways. Plasma ATL/TXB2 ratio showed a strong negative correlation with plasma TXB2 levels that was principally driven by the changes in TXB2, emphasizing that distinct biosynthetic pathways may be selectively regulated in platelets from subjects at risk for ARDS. In addition, neutrophil activation correlated with monocyte activation, indicating concurrent roles for both cell types on patient presentation for host responses. Ne-PA and Mo-PA were negatively correlated, consistent with differences in the regulation of these cell–cell interactions during acute lung inflammatory responses. Distinct from lipoxins’ antineutrophil actions, these mediators can promote monocyte and macrophage functional responses (40), so it was interesting to not find a significant correlation between monocyte responses and lipid mediators, suggesting an uncoupling of the immune regulation by lipoxins in the setting of early ARDS.

Here, we have uncovered intravascular responses that may contribute to early ARDS pathogenesis. A limitation of this work is that no external cohort was available to validate our findings. In addition, our results report unadjusted P values, given the exploratory nature of the experiment. This combination increases our risk for false discoveries; however, correction for multiple comparisons in exploratory studies where biomarkers may not be independent of each other can be overly conservative. We have attempted to mitigate this limitation by reporting effect sizes and confidence intervals for all findings, showing as much of the data graphically as possible, and by reporting quantitative P values throughout. This latter approach allows application of post hoc corrections, such as the Bonferroni correction, while interpreting the results. In addition, there are potential limitations to biochemical and immunological phenotyping methods used here that were associated with the multicenter design of this trial; blood samples were handled by multiple operators in multiple sites before fixation, leading to blood clot formation and exclusion of samples. Furthermore, some cell subsets were not present in sufficient numbers to allow measurement of specific analyses; monocyte subsets were not numerous enough to allow CD11b surface expression determinations and linear mixed model analyses.

In summary, our findings have demonstrated that specific plasma lipid mediators and peripheral blood leukocyte responses are present in patients at risk for ARDS and indicative of clinical outcomes and aspirin’s biochemical actions. These results have uncovered an association between peripheral blood monocytes and monocyte–platelet interactions and the development of ARDS, supporting a pivotal role for intravascular events in early ARDS pathogenesis.

Acknowledgments

Acknowledgment

The authors thank Sook Hwa Tan for technical assistance.

Footnotes

Supported by NIH grants U01 HL108712 (B.D.L.) and K08 HL130540 (R.E.A.).

Author Contributions: R.E.A. designed and performed experiments, analyzed and interpreted data, and wrote the manuscript. T.G. and R.E.C. analyzed and interpreted data and wrote the manuscript. I.B., J.Y.T., C.B., M.G., D.J.K., O.G., and D.T. collected specimens and contributed to manuscript preparation. B.D.L. conceived the study, designed experiments, analyzed and interpreted data, and wrote the manuscript.

This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.

Originally Published in Press as DOI: 10.1164/rccm.201712-2530OC on May 21, 2018

Author disclosures are available with the text of this article at www.atsjournals.org.

References

  • 1.Matthay MA, Ware LB, Zimmerman GA. The acute respiratory distress syndrome. J Clin Invest. 2012;122:2731–2740. doi: 10.1172/JCI60331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Famous KR, Delucchi K, Ware LB, Kangelaris KN, Liu KD, Thompson BT, et al. ARDS Network. Acute respiratory distress syndrome subphenotypes respond differently to randomized fluid management strategy. Am J Respir Crit Care Med. 2017;195:331–338. doi: 10.1164/rccm.201603-0645OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Zhao Z, Wickersham N, Kangelaris KN, May AK, Bernard GR, Matthay MA, et al. External validation of a biomarker and clinical prediction model for hospital mortality in acute respiratory distress syndrome. Intensive Care Med. 2017;43:1123–1131. doi: 10.1007/s00134-017-4854-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Yadav H, Thompson BT, Gajic O. Fifty years of research in ARDS: is acute respiratory distress syndrome a preventable disease? Am J Respir Crit Care Med. 2017;195:725–736. doi: 10.1164/rccm.201609-1767CI. [DOI] [PubMed] [Google Scholar]
  • 5.Looney MR, Nguyen JX, Hu Y, Van Ziffle JA, Lowell CA, Matthay MA. Platelet depletion and aspirin treatment protect mice in a two-event model of transfusion-related acute lung injury. J Clin Invest. 2009;119:3450–3461. doi: 10.1172/JCI38432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Zarbock A, Singbartl K, Ley K. Complete reversal of acid-induced acute lung injury by blocking of platelet-neutrophil aggregation. J Clin Invest. 2006;116:3211–3219. doi: 10.1172/JCI29499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Serhan CN, Sheppard KA. Lipoxin formation during human neutrophil-platelet interactions. Evidence for the transformation of leukotriene A4 by platelet 12-lipoxygenase in vitro. J Clin Invest. 1990;85:772–780. doi: 10.1172/JCI114503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Fiore S, Serhan CN. Formation of lipoxins and leukotrienes during receptor-mediated interactions of human platelets and recombinant human granulocyte/macrophage colony-stimulating factor-primed neutrophils. J Exp Med. 1990;172:1451–1457. doi: 10.1084/jem.172.5.1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Wedzicha JA, Syndercombe-Court D, Tan KC. Increased platelet aggregate formation in patients with chronic airflow obstruction and hypoxaemia. Thorax. 1991;46:504–507. doi: 10.1136/thx.46.7.504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ono E, Dutile S, Kazani S, Wechsler ME, Yang J, Hammock BD, et al. National Heart, Lung, and Blood Institute’s Asthma Clinical Research Network. Lipoxin generation is related to soluble epoxide hydrolase activity in severe asthma. Am J Respir Crit Care Med. 2014;190:886–897. doi: 10.1164/rccm.201403-0544OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Abdulnour RE, Dalli J, Colby JK, Krishnamoorthy N, Timmons JY, Tan SH, et al. Maresin 1 biosynthesis during platelet-neutrophil interactions is organ-protective. Proc Natl Acad Sci USA. 2014;111:16526–16531. doi: 10.1073/pnas.1407123111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Chiang N, Bermudez EA, Ridker PM, Hurwitz S, Serhan CN. Aspirin triggers antiinflammatory 15-epi-lipoxin A4 and inhibits thromboxane in a randomized human trial. Proc Natl Acad Sci USA. 2004;101:15178–15183. doi: 10.1073/pnas.0405445101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kor DJ, Carter RE, Park PK, Festic E, Banner-Goodspeed VM, Hinds R, et al. US Critical Illness and Injury Trials Group: Lung Injury Prevention with Aspirin Study Group (USCIITG: LIPS-A) Effect of aspirin on development of ARDS in at-risk patients presenting to the emergency department: the LIPS-A randomized clinical trial. JAMA. 2016;315:2406–2414. doi: 10.1001/jama.2016.6330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Abdulnour RE, Gunderson T, Barkas I, Timmons JY, Barnig C, Gong MN, et al. Early vascular events are associated with development of ARDS [abstract] Am J Respir Crit Care Med. 2017;195:A7659. doi: 10.1164/rccm.201712-2530OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Kor DJ, Talmor DS, Banner-Goodspeed VM, Carter RE, Hinds R, Park PK, et al. US Critical Illness and Injury Trials Group Lung Injury Prevention with Aspirin Study Group. Lung Injury Prevention with Aspirin (LIPS-A): a protocol for a multicentre randomised clinical trial in medical patients at high risk of acute lung injury BMJ Open 20122e001606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Rondina MT, Weyrich AS, Zimmerman GA. Platelets as cellular effectors of inflammation in vascular diseases. Circ Res. 2013;112:1506–1519. doi: 10.1161/CIRCRESAHA.113.300512. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Planagumà A, Pfeffer MA, Rubin G, Croze R, Uddin M, Serhan CN, et al. Lovastatin decreases acute mucosal inflammation via 15-epi-lipoxin A4. Mucosal Immunol. 2010;3:270–279. doi: 10.1038/mi.2009.141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Clària J, Serhan CN. Aspirin triggers previously undescribed bioactive eicosanoids by human endothelial cell-leukocyte interactions. Proc Natl Acad Sci USA. 1995;92:9475–9479. doi: 10.1073/pnas.92.21.9475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kuebler WM. Selectins revisited: the emerging role of platelets in inflammatory lung disease. J Clin Invest. 2006;116:3106–3108. doi: 10.1172/JCI30664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Fukunaga K, Kohli P, Bonnans C, Fredenburgh LE, Levy BD. Cyclooxygenase 2 plays a pivotal role in the resolution of acute lung injury. J Immunol. 2005;174:5033–5039. doi: 10.4049/jimmunol.174.8.5033. [DOI] [PubMed] [Google Scholar]
  • 21.Ortiz-Muñoz G, Mallavia B, Bins A, Headley M, Krummel MF, Looney MR. Aspirin-triggered 15-epi-lipoxin A4 regulates neutrophil-platelet aggregation and attenuates acute lung injury in mice. Blood. 2014;124:2625–2634. doi: 10.1182/blood-2014-03-562876. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Clària J, Lee MH, Serhan CN. Aspirin-triggered lipoxins (15-epi-LX) are generated by the human lung adenocarcinoma cell line (A549)-neutrophil interactions and are potent inhibitors of cell proliferation. Mol Med. 1996;2:583–596. [PMC free article] [PubMed] [Google Scholar]
  • 23.Eickmeier O, Seki H, Haworth O, Hilberath JN, Gao F, Uddin M, et al. Aspirin-triggered resolvin D1 reduces mucosal inflammation and promotes resolution in a murine model of acute lung injury. Mucosal Immunol. 2013;6:256–266. doi: 10.1038/mi.2012.66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Gerrits AJ, Frelinger AL, 3rd, Michelson AD. Whole blood analysis of leukocyte-platelet aggregates. Curr Protoc Cytom. 2016;78:6.15.1–6.15.10. doi: 10.1002/cpcy.8. [DOI] [PubMed] [Google Scholar]
  • 25.Michelson AD, Barnard MR, Krueger LA, Valeri CR, Furman MI. Circulating monocyte-platelet aggregates are a more sensitive marker of in vivo platelet activation than platelet surface P-selectin: studies in baboons, human coronary intervention, and human acute myocardial infarction. Circulation. 2001;104:1533–1537. doi: 10.1161/hc3801.095588. [DOI] [PubMed] [Google Scholar]
  • 26.Ziegler-Heitbrock L. Monocyte subsets in man and other species. Cell Immunol. 2014;289:135–139. doi: 10.1016/j.cellimm.2014.03.019. [DOI] [PubMed] [Google Scholar]
  • 27.Patel AA, Zhang Y, Fullerton JN, Boelen L, Rongvaux A, Maini AA, et al. The fate and lifespan of human monocyte subsets in steady state and systemic inflammation. J Exp Med. 2017;214:1913–1923. doi: 10.1084/jem.20170355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Tak T, van Groenendael R, Pickkers P, Koenderman L. Monocyte subsets are differentially lost from the circulation during acute inflammation induced by human experimental endotoxemia. J Innate Immun. 2017;9:464–474. doi: 10.1159/000475665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Tapp LD, Shantsila E, Wrigley BJ, Pamukcu B, Lip GY. The CD14++CD16+ monocyte subset and monocyte-platelet interactions in patients with ST-elevation myocardial infarction. J Thromb Haemost. 2012;10:1231–1241. doi: 10.1111/j.1538-7836.2011.04603.x. [DOI] [PubMed] [Google Scholar]
  • 30.Zawada AM, Rogacev KS, Rotter B, Winter P, Marell RR, Fliser D, et al. SuperSAGE evidence for CD14++CD16+ monocytes as a third monocyte subset. Blood. 2011;118:e50–e61. doi: 10.1182/blood-2011-01-326827. [DOI] [PubMed] [Google Scholar]
  • 31.Li XC, Miyasaka M, Issekutz TB. Blood monocyte migration to acute lung inflammation involves both CD11/CD18 and very late activation antigen-4-dependent and independent pathways. J Immunol. 1998;161:6258–6264. [PubMed] [Google Scholar]
  • 32.Dhaliwal K, Scholefield E, Ferenbach D, Gibbons M, Duffin R, Dorward DA, et al. Monocytes control second-phase neutrophil emigration in established lipopolysaccharide-induced murine lung injury. Am J Respir Crit Care Med. 2012;186:514–524. doi: 10.1164/rccm.201112-2132OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Herold S, Tabar TS, Janssen H, Hoegner K, Cabanski M, Lewe-Schlosser P, et al. Exudate macrophages attenuate lung injury by the release of IL-1 receptor antagonist in gram-negative pneumonia. Am J Respir Crit Care Med. 2011;183:1380–1390. doi: 10.1164/rccm.201009-1431OC. [DOI] [PubMed] [Google Scholar]
  • 34.Abdulnour RE, Sham HP, Douda DN, Colas RA, Dalli J, Bai Y, et al. Aspirin-triggered resolvin D1 is produced during self-resolving gram-negative bacterial pneumonia and regulates host immune responses for the resolution of lung inflammation. Mucosal Immunol. 2016;9:1278–1287. doi: 10.1038/mi.2015.129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.McCabe DJ, Harrison P, Mackie IJ, Sidhu PS, Purdy G, Lawrie AS, et al. Platelet degranulation and monocyte-platelet complex formation are increased in the acute and convalescent phases after ischaemic stroke or transient ischaemic attack. Br J Haematol. 2004;125:777–787. doi: 10.1111/j.1365-2141.2004.04983.x. [DOI] [PubMed] [Google Scholar]
  • 36.Vieira-de-Abreu A, Campbell RA, Weyrich AS, Zimmerman GA. Platelets: versatile effector cells in hemostasis, inflammation, and the immune continuum. Semin Immunopathol. 2012;34:5–30. doi: 10.1007/s00281-011-0286-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rondina MT, Brewster B, Grissom CK, Zimmerman GA, Kastendieck DH, Harris ES, et al. In vivo platelet activation in critically ill patients with primary 2009 influenza A(H1N1) Chest. 2012;141:1490–1495. doi: 10.1378/chest.11-2860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Levy BD, Serhan CN. Resolution of acute inflammation in the lung. Annu Rev Physiol. 2014;76:467–492. doi: 10.1146/annurev-physiol-021113-170408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Gajic O, Dabbagh O, Park PK, Adesanya A, Chang SY, Hou P, et al. U.S. Critical Illness and Injury Trials Group: Lung Injury Prevention Study Investigators (USCIITG-LIPS) Early identification of patients at risk of acute lung injury: evaluation of lung injury prediction score in a multicenter cohort study. Am J Respir Crit Care Med. 2011;183:462–470. doi: 10.1164/rccm.201004-0549OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Dalli J, Serhan CN. Specific lipid mediator signatures of human phagocytes: microparticles stimulate macrophage efferocytosis and pro-resolving mediators. Blood. 2012;120:e60–e72. doi: 10.1182/blood-2012-04-423525. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from American Journal of Respiratory and Critical Care Medicine are provided here courtesy of American Thoracic Society

RESOURCES