Abstract
Purpose
We examined whether patients with acute respiratory distress syndrome (ARDS) lacking risk factors are enrolled in therapeutic trials and assessed their clinical characteristics and outcomes.
Methods
We performed a secondary analysis of patient-level data pooled from the ARMA, ALVEOLI, FACTT, ALTA and EDEN ARDSNet randomized controlled trials obtained from the Biologic Specimen and Data Repository Information Coordinating Center of the National Heart, Lung and Blood Institute. We compared baseline characteristics and clinical outcomes (before and after adjustment using Poisson regression model) of ARDS patients with versus without risk factors.
Results
Of 3733 patients with ARDS, 81 (2.2%) did not have an identifiable risk factor. Patients without risk factors were younger, had lower baseline severity of illness, were more likely to have the ARDS resolve rapidly (i.e., within 24 hours) (p<0.001) and they had more ventilator-free days (median 21; p=0.003), more intensive care unit-free days (18; p=0.010), and more non-pulmonary organ failure-free days (24; p<0.001) than comparators (17, 14 and 18, respectively). Differences persisted after adjustment for potential confounders.
Conclusions
Patients with ARDS without identifiable risk factors are enrolled in therapeutic trials and may have better outcomes, including a higher proportion of rapidly resolving ARDS, than those with risk factors.
Keywords: acute lung injury, acute respiratory failure, hypoxemia, intensive care unit, epidemiology
INTRODUCTION
From a historical point of view, identification of a risk factor seems to be an important component of the definition of acute respiratory distress syndrome (ARDS). In their original description of the syndrome in 1967, Ashbaugh et al identified severe trauma, pneumonia and pancreatitis as precipitating factors of ARDS in a series of 12 patients.1 Similarly, in their article on “an expanded definition of adult respiratory distress syndrome” published in 1988, Murray et al proposed that ARDS should be defined by a lung injury score combined with acute onset and a known clinical risk factor.2 Interestingly, the subsequent (released in 1994) American-European Consensus Conference definition of ARDS did not formally require the presence of a risk factor.3 This was considered a limitation,4 which was addressed by the latest Berlin definition of ARDS.5 Indeed, the Berlin definition requires that “for a patient to be defined as having ARDS, the onset must be within one week of a known clinical insult” with the additional clarification that “if none [risk factor is] identified, [there is] need to objectively rule out hydrostatic edema”.5 Accordingly, when dealing with a patient with suspicion of ARDS, clinicians seek risk factors associated with the syndrome.
Yet most observational studies show that there are cases of ARDS when a risk factor cannot be identified.6 In a retrospective observational study by Gibelin et al, 12 (1.8%) of 665 patients with ARDS did not have any identifiable predisposing factors.7 Similarly, in the largest survey on the epidemiology of ARDS (the LUNG SAFE study), a risk factor was not recognized in 8.3% of patients.8 In a recently published ancillary analysis of the LUNG SAFE cohort, de Prost et al studied these patients and concluded that “the outcome of ARDS with no risk factor is comparable to other ARDS”.9 It is not known whether patients with ARDS in the absence of risk factors are enrolled in therapeutic trials and what their characteristics are. By harnessing the high-quality data of the ARDS Network (ARDSNet) randomized controlled trials, which were funded by the National Heart, Lung and Blood Institute (NHLBI), we endeavored to estimate the prevalence, clinical characteristics and outcomes of patients with ARDS with versus (vs.) without identifiable risk factors.
METHODS
Patient Population
The patient population for this analysis consisted of unique patients enrolled in ARDSNet therapeutic clinical trials, namely the ARMA, ALVEOLI, FACTT, ALTA and EDEN trials, which were published between 2000 and 2012.10–14 The compared treatment modalities in these trials were lower vs. higher tidal volume,10 higher vs. lower positive end-expiratory pressure,11 conservative vs. liberal fluid administration,12 aerosolized albuterol vs. placebo,13 and initial trophic vs. full enteral feeding,14 respectively. For this analysis, we excluded individuals from the SAILS trial (rosuvastain vs. placebo) because all patients should have had a particular risk factor, namely sepsis, to be considered eligible for the SAILS trial.15 We also excluded individuals from the LaSRS trial (corticosteroids vs. placebo) because they needed to have late-phase ARDS and therefore were substantially different from participants in other ARDSNet trials.16 Finally, individuals from the OMEGA trial17 were also enrolled in the EDEN trial and therefore were included in our analysis as part of the latter trial.14 To be eligible in the ARDSNet trials10–14 and hence our analysis, subjects had to be endotracheally intubated receiving positive-pressure mechanical ventilation, had to have a partial pressure of arterial oxygen to fraction of inspired oxygen ratio (PaO2:FiO2) of equal to or less than 300, and had to present with bilateral infiltrates on chest radiography consistent with pulmonary edema, without evidence of left atrial hypertension.3 Subjects with underlying illnesses, including chronic heart failure, and those meeting clinical criteria of ARDS for more than 36–48 hours at the time of enrollment were excluded.10–14 The Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC) of NHLBI18 provided us with extensive de-identified patient-level clinical data from ARDSNet trials,10–14 after approving the prospective protocol of our analysis (available as Supplemental Material). Our analysis was approved by the Institutional Review Board at Weill Cornell Medicine (#1706018291), which also gave a waiver of the need for informed consent.
Identification of ARDS patients without risk factor
For each patient included in the ARDSNet trials,10–14 investigators had to specify the risk factor of ARDS by choosing among the following: “trauma”, “sepsis”, “multiple transfusions”, “aspiration”, “pneumonia” or “other”. In detail, ARDSNet investigators had to record a primary (i.e., the most important reason leading to ARDS based on their judgment) and a secondary (i.e., a co-existing but less important) risk factor of ARDS for each patient. For example, for a given patient with ARDS, “trauma” could have been chosen as the primary risk factor and “multiple transfusions” could have been chosen as the secondary risk factor. If “other” was chosen as risk factor of ARDS (either primary or secondary), ARDSNet investigators had to elaborate in a free text field available in each case report form. For patients in whom this option (i.e., “other”) was recorded, we reviewed the text field to adjudicate whether the patient had or did not have an identifiable risk factor. In cases that this text field was not filled by the ARDSNet investigators (which happened 8 times), we considered that this patient did not have an identifiable risk factor. In Supplemental Table 1, we present the exact wording used by ARDSNet investigators to describe primary risk factors other than trauma, sepsis, multiple transfusions, aspiration and pneumonia, and our subsequent adjudication regarding the categorization of patients into the identifiable vs. unidentifiable risk factor groups.
For our main analysis, we considered patients without primary identifiable risk factors. Given that for some patients a secondary but not a primary risk factor had been identified, we carried out a sensitivity analysis, for which we only considered patients with neither primary nor secondary identifiable risk factors.
Two subgroup analyses were also undertaken. One subgroup analysis compared patients with identifiable risk factors indicating direct/pulmonary ARDS (namely, pneumonia or aspiration as primary risk factor) vs. patients with ARDS lacking identifiable risk factors. The other subgroup analysis compared patients with identifiable risk factors indicating indirect/non-pulmonary ARDS (namely, trauma, sepsis or multiple transfusions as primary risk factor) vs. patients with ARDS lacking identifiable risk factors.
Study Outcomes and Statistical analysis
Continuous variables are presented as median (interquartile range, IQR) and differences between them were estimated using non-parametric Mann-Whitney U-test. Categorical variables are presented as count (percentage) and differences between them were estimated using chi-square test or Fisher’s exact test, as appropriate. Analysis of the temporal trend in study prevalence of ARDS without identifiable risk factors was performed via study-level least-squares regression weighting studies by the inverse of the sample size.
All-cause 60-day mortality was the primary outcome of our analysis. Patients discharged from hospital with unassisted breathing before 60 days were considered to be alive at 60 days. Kaplan-Meier analysis was used to estimate time to mortality for ARDS patients with vs. without identifiable risk factors. As a secondary outcome of our analysis, we considered the proportion of patients with vs. without risk factors who had the ARDS resolve rapidly, i.e., they had a PaO2:FiO2 of greater than 300 within 24 hours after enrollment or achieved unassisted breathing within 24 hours after enrollment and remained free from assisted breathing for 48 hours or more.19 Ventilator-free days, intensive care unit (ICU)-free days, and non-pulmonary organ failure-free days also served as secondary outcomes of our analysis. These composite outcomes were calculated by the number of days in the first 28 days that a patient was alive and not on a ventilator, not in the ICU, or free of non-pulmonary organ failure, respectively. To further assess whether outcome differences were due to confounders, a multivariable logistic regression model (for mortality; binary outcome) and a multivariable Poisson regression model (for ventilator-free days, ICU-free days and non-pulmonary organ failure-free days; count outcomes) were used with no imputation for missing data. These models included Acute Physiology and Chronic Health Evaluation (APACHE III) score (a marker of baseline severity of illness), age, and malignancy as covariates. These covariates were chosen a priori in light of recent evidence suggesting that these are the non-modifiable risk factors most strongly associated with worsened outcome from ARDS.20 An additional post-hoc multivariable Poisson regression analysis was carried out, which used sex in addition to APACHE III score, age and malignancy as covariates. All statistical analyses were done with R v3.2.3 (R Core Team, Vienna, Austria). Two-sided p values of less than 0.05 were considered to denote statistical significance.
RESULTS
Prevalence
Of the 3733 unique patients enrolled in the ARDSNet randomized controlled trials,10–14 81 (2.2%) had ARDS without a primary identifiable risk factor. Of these 81 patients, 65 (1.7% of the total population) had neither primary nor secondary identifiable risk factors, while for the remaining 16 patients (0.5% of the total population) a secondary (namely, aspiration, sepsis and trauma in 6, 7 and 3 patients, respectively) but not a primary risk factor had been identified. Enrollment of patients without identifiable risk factors into ARDSNet trials decreased over time, from a prevalence of 3.3% in ARMA10 to 0.9% in the EDEN trial14 (r-squared=0.78, p=0.048).
Clinical Characteristics
In Table 1, baseline characteristics of patients with ARDS are presented. Compared to patients with identifiable risk factors, those without such factors were younger [median 51 (IQR 39–63) vs. 45 (32–59); p=0.006] and had lower severity of illness as assessed by APACHE III score [89 (70–111) vs. 80 (63–108); p=0.04]. Consistently, non-pulmonary organ failures, such as circulatory [2164/3652 (60%) vs. 35/81 (43%); p=0.003] and coagulation (20% vs. 8%; p=0.011), were less common in patients lacking risk factors. In contrast, compared groups did not differ at baseline in terms of PaO2:FiO2 (p=0.30) or severity of ARDS (p=0.73). Also, usage of corticosteroids was not different (p=0.39). The above results persisted in the sensitivity analysis (Supplemental Table 2). With the exception of the usage of corticosteroids, which was more common in patients with identifiable risk factors indicating direct/pulmonary ARDS than in patients without identifiable risk factors (36% vs 22%, p=0.016) (Supplemental Table 3), the above results also persisted in the two subgroup analyses (Supplemental Tables 3 and 4).
Table 1.
Baseline characteristics of patients with acute respiratory distress syndrome with vs. without identifiable risk factors.
| With Risk Factors | Without Risk Factors | p value | |
|---|---|---|---|
| Number of patients | 3652 (98%) | 81 (2%) | |
| Age, years | 51 (39–63) | 45 (32–59) | 0.006 |
| Female sex | 1652 (45%) | 42 (52%) | 0.29 |
| Race | 0.93 | ||
| White | 2684 (74%) | 58 (72%) | |
| Black | 639 (18%) | 15 (19%) | |
| Other | 329 (9%) | 8 (10%) | |
| Body mass index | 27 (23–32) | 27 (23–33) | 0.67 |
| Comorbidity | |||
| Diabetes Mellitus | 709 (20%) | 12 (15%) | 0.40 |
| Malignancy | 165 (5%) | 2 (3%) | 0.58 |
| Cirrhosis | 136 (4%) | 2 (3%) | 0.99 |
| End-stage Renal Disease | 76 (2%) | 3 (4%) | 0.24 |
| Immunosuppression | 388 (11%) | 4 (5%) | 0.14 |
| APACHE III score | 89 (70–111) | 80 (63–108) | 0.04 |
| Non-Pulmonary Organ Failure | |||
| Circulatory | 2164 (60%) | 35 (43%) | 0.003 |
| Coagulation | 703 (20%) | 6 (8%) | 0.011 |
| Hepatic | 569 (17%) | 18 (23%) | 0.20 |
| Renal | 795 (22%) | 15 (19%) | 0.57 |
| PaO2:FiO2* | 124 (86–173) | 138 (96–185) | 0.30 |
| Severity of ARDS** | 0.73 | ||
| Mild | 561 (15%) | 11 (17%) | |
| Moderate | 1805 (49%) | 31 (48%) | |
| Severe | 1286 (35%) | 23 (35%) | |
| Usage of Corticosteroids | 990 (27%) | 18 (22%) | 0.39 |
Abbreviations: vs., versus; APACHE, acute physiology and chronic health evaluation; PaO2:FiO2, partial pressure of arterial oxygen to fraction of inspired oxygen ratio; ARDS, acute respiratory distress syndrome.
Data are presented as n (%) or median (interquartile range). Counts may not add to full group sample size due to missing values.
For 1 (0.03%) patient, data on PaO2:FiO2 at screening were missing and therefore severity of ARDS could not be assessed. This patient had an identifiable risk factor.
Severity of ARDS was categorized based on the Berlin definition.5
Outcomes
All-cause 60-day mortality did not differ significantly between patients with vs. without risk factors [985/3652 (27%) vs. 18/81 (22%); p=0.41] (Table 2). This finding persisted after adjustment for APACHE III score, age, and malignancy (Table 3). Similarly, the Kaplan-Meier analysis did not reveal any differences between the compared groups with regard to the probability of death over time (p=0.2 by log-rank test) (Figure 1). However, there was difference between the compared groups with regard to secondary outcomes (Table 2). Indeed, a greater proportion of patients without risk factors [18/81 (22%); p<0.001] than patients with risk factors [347/3652 (10%)] had the ARDS resolve rapidly. At 24 hours after trial enrollment, median PaO2:FiO2 was 145 (113–200) (p=0.82) in patients without risk factors and 140 (99–188) in those with risk factors. Also, individuals without risk factors had more ventilator-free days [21 (0–26); p=0.003], more ICU-free days [18 (0–24); p=0.010], and more non-pulmonary organ failure-free days [24 (12–27); p<0.001] than comparators [17 (0–24), 14 (0–21) and 18 (2–25), respectively] (Table 2). These differences persisted even after adjustment for APACHE III score, age, and malignancy (Table 3). This was also the case after additional adjustment for sex (Supplemental Table 5). The above results also persisted in the sensitivity analysis in which patients with neither primary nor secondary identifiable risk factors were considered (Supplemental Table 6 for the crude and Supplemental Table 7 for the adjusted association between absence of identifiable risk factors of ARDS and outcomes). This was also the case for the subgroup analysis comparing patients with direct ARDS vs. patients with ARDS without identifiable risk factors (Supplemental Table 8) and for the subgroup analysis comparing patients with indirect ARDS vs. patients with ARDS without identifiable risk factors (Supplemental Table 9).
Table 2.
Outcomes of patients with acute respiratory distress syndrome with vs. without identifiable risk factors.
| Outcome | With Risk Factors (n=3652) | Without Risk Factors (n=81) | p value |
|---|---|---|---|
| 60-day mortality | 985 (27%) | 18 (22%) | 0.41 |
| Rapidly Resolving ARDS | 347 (10%) | 18 (22%) | <0.001 |
| Ventilator-free days | 17 (0–24) | 21 (0–26) | 0.003 |
| ICU-free days | 14 (0–21) | 18 (0–24) | 0.010 |
| Non-pulmonary organ failure-free days | 18 (2–25) | 24 (12–27) | <0.001 |
Abbreviations: vs., versus; ARDS, acute respiratory distress syndrome; ICU, intensive care unit.
Data are presented as n (%) or median (interquartile range).
Patients discharged from hospital with unassisted breathing before 60 days were considered to be alive at 60 days. Rapidly resolving ARDS was defined by a partial pressure of arterial oxygen to fraction of inspired oxygen ratio of greater than 300 within 24 hours following enrollment or by achieving unassisted breathing within 24 hours following enrollment and remaining free from assisted breathing for at least 48 hours. Ventilator-free days, ICU-free days, and non-pulmonary organ failure-free days were calculated by the number of days in the first 28 days that a patient was alive and not on a ventilator, not in the ICU, or free of non-pulmonary organ failure, respectively.
Table 3.
Logistic and Poisson regression analysis for several outcomes of patients with acute respiratory distress syndrome with vs. without identifiable risk factors.
| Crude | Adjusted | ||||
|---|---|---|---|---|---|
| 60-day mortality | Odds ratio | p-value | Odds ratio | p-value | |
| Without risk factor | 0.77 (0.46–1.31) | 0.35 | 1.11 (0.63–1.96) | 0.72 | |
| APACHE III | 1.02 (1.02–1.03) | <0.0001 | |||
| Age | 1.03 (1.02–1.03) | <0.0001 | |||
| Malignancy | 2.62 (1.85–3.71) | <0.0001 | |||
| Ventilator-free days | Relative rate | p-value | Relative rate | p-value | |
| Without risk factor | 1.22 (1.16–1.29) | <0.0001 | 1.10 (1.04–1.16) | 0.0006 | |
| APACHE III | 0.91 (0.90–0.91) | <0.0001 | |||
| Age | 0.94 (0.93–0.95) | <0.0001 | |||
| Malignancy | 0.80 (0.75–0.84) | <0.0001 | |||
| ICU-free days | Relative rate | p-value | Relative rate | p-value | |
| Without risk factor | 1.22 (1.15–1.30) | <0.0001 | 1.11 (1.05–1.18) | 0.0004 | |
| APACHE III | 0.91 (0.90–0.91) | <0.0001 | |||
| Age | 0.95 (0.94–0.96) | <0.0001 | |||
| Malignancy | 0.85 (0.81–0.90) | <0.0001 | |||
| Non-pulmonary organ failure-free days | Relative rate | p-value | Relative rate | p-value | |
| Without risk factor | 1.32 (1.23–1.40) | <0.0001 | 1.20 (1.13–1.27) | <0.0001 | |
| APACHE III | 0.92 (0.92–0.93) | <0.0001 | |||
| Age | 0.96 (0.96–0.97) | <0.0001 | |||
| Malignancy | 0.76 (0.72–0.80) | <0.0001 | |||
Abbreviations: vs., versus; APACHE, acute physiology and chronic health evaluation; ICU, intensive care unit.
Increments of 10 points and 10 years were used for APACHE III and age, respectively.
Patients discharged from hospital with unassisted breathing before 60 days were considered to be alive at 60 days. Ventilator-free days, ICU-free days, and non-pulmonary organ failure-free days were calculated by the number of days in the first 28 days that a patient was alive and not on a ventilator, not in the ICU, or free of non-pulmonary organ failure, respectively.
All multivariable analyses were complete-case. Data on APACHE III, age and malignancy were missing for 81, 0 and 46 patients, respectively.
Figure 1.
Kaplan Meier curves of mortality of patients with acute respiratory distress syndrome with versus without identifiable risk factors (p=0.2 by log-rank test). Patients discharged home were considered alive at 60 days.
DISCUSSION
This secondary analysis of data from 3733 individuals with ARDS enrolled in ARDSNet trials suggests that patients without identifiable risk factors were enrolled in the trials and that their enrollment decreased over time. These patients had distinct baseline characteristics (such as lower severity of illness) and better clinical outcomes other than mortality (namely, proportion of rapidly resolving ARDS, ventilator-free days, ICU-free days, and non-pulmonary organ failure-free days) compared to ARDS patients with risk factors.
We found that a risk factor was generally identified in patients with ARDS enrolled in therapeutic trials. However, patients with ARDS without identifiable risk factors were also enrolled in therapeutic trials but in an ever-decreasing rate. The 2.2% prevalence of ARDS lacking precipitating factors in our secondary analysis of interventional trials10–14 was substantially lower than the 8.3% prevalence reported in the observational LUNG SAFE study.8 This may be anticipated as researchers may be more conservative with ARDS designation when recruiting for interventional trials, especially those involving potentially risky treatments, than for observational studies. On the other hand, the decrease in prevalence of ARDS without identifiable risk factors may be challenging to explain. A clue is that earliest ARDSNet trials were carried out shortly after the release of the American-European Consensus Conference definition3 which had given less importance to the presence of a risk factor compared to the subsequent Berlin definition.5 Thus, one could assume that the recruitment rate of patients without identifiable risk factors into trials merely reflected the evolving appreciation of managing clinicians regarding the importance of the presence of a known risk factor for accurately identifying patients with ARDS.
We found that ARDS patients without identifiable risk factors had different baseline characteristics from those with such factors. This finding corroborates the results of the ancillary study of LUNG SAFE.9 Indeed, both the abovementioned contribution9 and our analysis agree that individuals lacking risk factors had at baseline fewer non-pulmonary organ failures compared to those with risk factors. Also, both studies agree that compared groups did not differ regarding usage of corticosteroids.9 This is interesting because ARDS without predisposing factors presumably includes clinical entities, such as auto-immune or drug-induced disorders, which have been coined as “ARDS imitators or mimics” and may benefit from administration of anti-inflammatory treatment (mainly corticosteroids).21,22
We found that ARDS patients without identifiable risk factors had better outcomes than those with risk factors. This finding seems robust given that it persisted after adjustment for potential confounders, in the sensitivity analysis and in the two subgroup analyses. One might consider it not surprising that patients with ARDS without identifiable risk factors fared better than those with risk factors as the latter would have a contribution of morbidity from their underlying disease state. Also, this finding is in line with the hypothesis that ARDS without identifiable predisposing factors may lack diffuse alveolar damage at histological examination and therefore it may be associated with a benign course.23,24 By performing post-mortem lung histological examination, Thille et al observed that up to 14% of patients meeting clinical criteria of ARDS (including severe ARDS) did not have any pulmonary lesions and the investigators attributed this specific observation to atelectasis.25 Although we could not be sure whether this was also the case for our study population due to unavailability of data from open lung biopsies or necropsy, a conjecture can be made that atelectasis might at least partially explain our finding that 22% of patients without risk factors (including several patients initially having severe ARDS) had the ARDS resolved and they were subsequently extubated within 24 hours. One may therefore infer that, although we tend to associate ARDS in the absence of risk factors with “ARDS mimics” that could benefit from anti-inflammatory treatment,26 it might occasionally reflect atelectasis that can respond rapidly to standard care.27 Practically, when they cannot identify a risk factor in a patient who meets criteria for ARDS, clinicians should keep in mind that this patient might have the ARDS resolve rapidly without the need for aggressive treatment.
Our analysis has limitations. Firstly, although it is based on a large database of 3373 patients with ARDS,10–14 it might still have insufficient power to reveal a statistically significant difference in mortality and this may explain why the 5% absolute decrease in mortality associated with absence of risk factors did not reach statistical significance (Table 2). Secondly, although secondary analyses provide useful insights,28 they are retrospective and vulnerable to residual confounding. Thirdly, we lacked detailed information on whether patients underwent a comprehensive diagnostic work-up, such as performance of bronchoalveolar lavage or immunological tests, according to a systematic protocol to rule out entities mimicking ARDS. We also could not assess whether they underwent measurement of left heart filling pressures to rule out fluid overload. One could therefore be reasonably skeptical whether all patients without risk factors had indeed ARDS and whether they should have been included in the ARDSNet trials. However, given that these trials were meticulously performed, it is anticipated that the ARDSNet investigators rigorously attempted to rule out fluid overload while screening for potential candidates.10–14 Moreover, chronic heart failure was an explicit exclusion criterion of ARDSNet trials10–14 and patients with and without risk factors did not differ in terms of baseline renal comorbidities (Table 1).
Fourthly, one could raise concerns on the validity of the information recorded by ARDSNet investigators regarding risk factors. The observation that for some patients a secondary but not a primary risk factor had been recorded might fuel such concerns about possible data entry mistakes. It is also possible that some patients without identifiable risk factor upon ARDS diagnosis might have a risk factor identified later in their hospitalization based on microbiological or radiological information. Since we did not have access to patient charts to perform a thorough review, we could not rule out the above possibilities. However, these were high-quality randomized controlled trials and it may be reasonably expected that data entry was carefully performed.10–14 Also, it may be reassuring that the baseline characteristics of patients without risk factors in our analysis were similar to those in relevant observational studies9 and that our results persisted in the sensitivity analysis, for which we only considered patients with neither primary nor secondary identifiable risk factors. Finally, one could raise concerns on the accuracy of our adjudication regarding the categorization of patients into the identifiable vs. unidentifiable risk factor groups on the basis that only scarce relevant information was generally given by the ARDSNet investigators. To address this concern and for transparency, we reported the exact wording used by the ARDSNet investigators in Supplemental Table 1. Although we cannot preclude some unintended incidents of misclassification, they seem unlikely to introduce a systematic error undermining the validity of our findings which had strong statistical significance (Tables 2 and 3).
CONCLUSIONS
In conclusion, this analysis suggests that patients with ARDS lacking identifiable risk factors were enrolled in therapeutic trials in a decreasing rate and they had distinct baseline clinical characteristics and better outcomes other than mortality, including a higher proportion of rapidly resolving ARDS, than individuals with risk factors. Clinicians should be aware that some of the patients without risk factors may respond rapidly to standard care.
Supplementary Material
Highlights.
Patients with acute respiratory distress syndrome (ARDS) without identifiable risk factors are enrolled in therapeutic trials and may have better outcomes, including a higher proportion of rapidly resolving ARDS, than those with risk factors.
Clinicians should be aware that some of the patients without risk factors may respond rapidly to standard care.
Acknowledgments
The authors are grateful to the investigators of ARMA, ALVEOLI, FACTT, ALTA and EDEN ARDSNet trials for collecting the data on which this analysis was based and the NHLBI BioLINCC for providing these data. This manuscript was prepared using ARMA, ALVEOLI, FACTT, ALTA, EDEN research materials obtained from the NHLBI BioLINCC and does not necessarily reflect the opinions or views of the ARMA, ALVEOLI, FACTT, ALTA, EDEN, ARDSNet, or the NHLBI.
Funding
This work was supported by Clinical and Translational Science Center at Weill Cornell Medicine grant UL1TR000457-06 (to CO) and National Institutes of Health grants KL2TR000458-10 (to EJS), R01 HL055330 (to AMKC) and P01 HL108801 (to AMKC).
Footnotes
Authors’ Contributions
JSH designed the study, interpreted the data and critically revised the manuscript. EJS designed the study, participated in data cleaning, interpreted the data and critically revised the manuscript. CO contributed to study design, did the data cleaning and analysis, contributed to data interpretation and critically revised the manuscript. AMKC contributed to data interpretation and critically revised the manuscript. IIS conceived the study, designed the study, interpreted the data, critically revised the manuscript and is the guarantor of the study. All authors approved the submission.
Competing interests
The authors declare that they have no competing interests.
Ethics approval and consent to participate
The study was approved by the Institutional Review Board at Weill Cornell Medicine (#1706018291), which gave a waiver of the need for informed consent.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Ashbaugh DG, Bigelow DB, Petty TL, et al. Acute respiratory distress in adults. Lancet. 1967;2:319–323. doi: 10.1016/s0140-6736(67)90168-7. [DOI] [PubMed] [Google Scholar]
- 2.Murray JF, Matthay MA, Luce JM, et al. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis. 1988;138:720–723. doi: 10.1164/ajrccm/138.3.720. [DOI] [PubMed] [Google Scholar]
- 3.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;149:818–824. doi: 10.1164/ajrccm.149.3.7509706. [DOI] [PubMed] [Google Scholar]
- 4.Phua J, Stewart TE, Ferguson ND. Acute respiratory distress syndrome 40 years later: time to revisit its definition. Crit Care Med. 2008;36:2912–2921. doi: 10.1097/CCM.0b013e31817d20bd. [DOI] [PubMed] [Google Scholar]
- 5.Ranieri VM, Rubenfeld GD, Thompson BT, et al. ARDS Definition Task Force: Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307:2526–2533. doi: 10.1001/jama.2012.5669. [DOI] [PubMed] [Google Scholar]
- 6.Pham T, Rubenfeld GD. Fifty Years of Research in ARDS. The Epidemiology of Acute Respiratory Distress Syndrome. A 50th Birthday Review. Am J Respir Crit Care Med. 2017;195:860–870. doi: 10.1164/rccm.201609-1773CP. [DOI] [PubMed] [Google Scholar]
- 7.Gibelin A, Parrot A, Maitre B, et al. Acute respiratory distress syndrome mimickers lacking common risk factors of the Berlin definition. Intensive Care Med. 2016;42:164–172. doi: 10.1007/s00134-015-4064-y. [DOI] [PubMed] [Google Scholar]
- 8.Bellani G, Laffey JG, Pham T, et al. LUNG SAFE Investigators; ESICM Trials Group: Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 2016;315:788–800. doi: 10.1001/jama.2016.0291. [DOI] [PubMed] [Google Scholar]
- 9.de Prost N, Pham T, Carteaux G, et al. LUNG SAFE investigators; ESICM trials group; REVA network: Etiologies, diagnostic work-up and outcomes of acute respiratory distress syndrome with no common risk factor: a prospective multicenter study. Ann Intensive Care. 2017;7:69. doi: 10.1186/s13613-017-0281-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brower RG, Matthay MA, Morris A, et al. 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;342:1301–1308. doi: 10.1056/NEJM200005043421801. [DOI] [PubMed] [Google Scholar]
- 11.Brower RG, Lanken PN, MacIntyre N, et al. National Heart, Lung, and Blood Institute ARDS Clinical Trials Network: Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med. 2004;351:327–336. doi: 10.1056/NEJMoa032193. [DOI] [PubMed] [Google Scholar]
- 12.Wiedemann HP, Wheeler AP, Bernard GR, et al. National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network: Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354:2564–2575. doi: 10.1056/NEJMoa062200. [DOI] [PubMed] [Google Scholar]
- 13.Matthay MA, Brower RG, Carson S, et al. National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network: Randomized, placebo-controlled clinical trial of an aerosolized β2 -agonist for treatment of acute lung injury. Am J Respir Crit Care Med. 2011;184:561–568. doi: 10.1164/rccm.201012-2090OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rice TW, Wheeler AP, Thompson BT, et al. National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network: Initial trophic vs full enteral feeding in patients with acute lung injury: the EDEN randomized trial. JAMA. 2012;307:795–803. doi: 10.1001/jama.2012.137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Truwit JD, Bernard GR, Steingrub J, et al. National Heart, Lung, and Blood Institute ARDS Clinical Trials Network: Rosuvastatin for sepsis-associated acute respiratory distress syndrome. N Engl J Med. 2014;370:2191–2200. doi: 10.1056/NEJMoa1401520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.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;354:1671–1684. doi: 10.1056/NEJMoa051693. [DOI] [PubMed] [Google Scholar]
- 17.Rice TW, Wheeler AP, Thompson BT, et al. NIH NHLBI Acute Respiratory Distress Syndrome Network of Investigators: Enteral omega-3 fatty acid, gamma-linolenic acid, and antioxidant supplementation in acute lung injury. JAMA. 2011;306:1574–1581. doi: 10.1001/jama.2011.1435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Coady SA, Mensah GA, Wagner EL, et al. Use of the National Heart, Lung, and Blood Institute Data Repository. N Engl J Med. 2017;376:1849–1858. doi: 10.1056/NEJMsa1603542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Oromendia C, Siempos II. Reclassification of Acute Respiratory Distress Syndrome: A Secondary Analysis of the ARDS Network Trials. Ann Am Thorac Soc. 2018 May 3; doi: 10.1513/AnnalsATS.201803-192RL. Epub ahead of print. [DOI] [PubMed] [Google Scholar]
- 20.Laffey JG, Bellani G, Pham T, et al. LUNG SAFE Investigators and the ESICM Trials Group: Potentially modifiable factors contributing to outcome from acute respiratory distress syndrome: the LUNG SAFE study. Intensive Care Med. 2016;42:1865–1876. doi: 10.1007/s00134-016-4571-5. [DOI] [PubMed] [Google Scholar]
- 21.Schwarz MI, Albert RK. “Imitators” of the ARDS: implications for diagnosis and treatment. Chest. 2004;125:1530–1535. doi: 10.1378/chest.125.4.1530. [DOI] [PubMed] [Google Scholar]
- 22.Guérin C, Thompson T, Brower R. The ten diseases that look like ARDS. Intensive Care Med. 2015;41:1099–1102. doi: 10.1007/s00134-014-3608-x. [DOI] [PubMed] [Google Scholar]
- 23.Aublanc M, Perinel S, Guérin C. Acute respiratory distress syndrome mimics: the role of lung biopsy. Curr Opin Crit Care. 2017;23:24–29. doi: 10.1097/MCC.0000000000000373. [DOI] [PubMed] [Google Scholar]
- 24.Lorente JA, Cardinal-Fernández P, Muñoz D, et al. Acute respiratory distress syndrome in patients with and without diffuse alveolar damage: an autopsy study. Intensive Care Med. 2015;41:1921–1930. doi: 10.1007/s00134-015-4046-0. [DOI] [PubMed] [Google Scholar]
- 25.Thille AW, Esteban A, Fernández-Segoviano P, et al. Comparison of the Berlin definition for acute respiratory distress syndrome with autopsy. Am J Respir Crit Care Med. 2013;187:761–767. doi: 10.1164/rccm.201211-1981OC. [DOI] [PubMed] [Google Scholar]
- 26.Thompson BT, Chambers RC, Liu KD. Acute respiratory distress syndrome. N Engl J Med. 2017;377:562–572. doi: 10.1056/NEJMra1608077. [DOI] [PubMed] [Google Scholar]
- 27.Siempos II, Berlin DA. Incidence of acute respiratory distress syndrome. JAMA. 2016;316:346. doi: 10.1001/jama.2016.6465. [DOI] [PubMed] [Google Scholar]
- 28.Bauer PR, Gajic O, Nanchal R, et al. ICON Investigators (Supplemental Appendix 1): Association between timing of intubation and outcome in critically ill patients: A secondary analysis of the ICON audit. J Crit Care. 2017;42:1–5. doi: 10.1016/j.jcrc.2017.06.010. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.