Comments
Abstract
Rationale: Patients with severe coronavirus disease (COVID-19) meet clinical criteria for the acute respiratory distress syndrome (ARDS), yet early reports suggested they differ physiologically and clinically from patients with non–COVID-19 ARDS, prompting treatment recommendations that deviate from standard evidence-based practices for ARDS.
Objectives: To compare respiratory physiology, clinical outcomes, and extrapulmonary clinical features of severe COVID-19 with non–COVID-19 ARDS.
Methods: We performed a retrospective cohort study, comparing 130 consecutive mechanically ventilated patients with severe COVID-19 with 382 consecutive mechanically ventilated patients with non–COVID-19 ARDS. Initial respiratory physiology and 28-day outcomes were compared. Extrapulmonary manifestations (inflammation, extrapulmonary organ injury, and coagulation) were compared in an exploratory analysis.
Results: Comparison of patients with COVID-19 and non–COVID-19 ARDS suggested small differences in respiratory compliance, ventilatory efficiency, and oxygenation. The 28-day mortality was 30% in patients with COVID-19 and 38% in patients with non–COVID-19 ARDS. In adjusted analysis, point estimates of differences in time to breathing unassisted at 28 days (adjusted subdistributional hazards ratio, 0.98 [95% confidence interval (CI), 0.77–1.26]) and 28-day mortality (risk ratio, 1.01 [95% CI, 0.72–1.42]) were small for COVID-19 versus non–COVID-19 ARDS, although the confidence intervals for these estimates include moderate differences. Patients with COVID-19 had lower neutrophil counts but did not differ in lymphocyte count or other measures of systemic inflammation.
Conclusions: In this single-center cohort, we found no evidence for large differences between COVID-19 and non–COVID-19 ARDS. Many key clinical features of severe COVID-19 were similar to those of non–COVID-19 ARDS, including respiratory physiology and clinical outcomes, although our sample size precludes definitive conclusions. Further studies are needed to define COVID-19–specific pathophysiology before a deviation from evidence-based treatment practices can be recommended.
Coronavirus disease (COVID-19) causes hypoxic respiratory failure via severe lung injury. Most patients with severe COVID-19 meet clinical criteria for acute respiratory distress syndrome (ARDS), defined by hypoxemia and bilateral opacities on chest radiographs without evidence of left-sided cardiac dysfunction (1). Yet early reports suggested that the respiratory physiology among patients with severe COVID-19 might differ from that of non–COVID-19 ARDS, reflecting distinct pathophysiology, and thus should be managed differently (2, 3). In addition, early reports suggested greater mortality among mechanically ventilated patients with COVID-19 than historic ARDS cohorts (4–6). The question of whether severe COVID-19 represents “atypical ARDS” or even non-ARDS pathophysiology has been a major source of controversy (7–9).
In addition, numerous extrapulmonary clinical features have been identified as abnormal in COVID-19, which prompted widespread variation in routine intensive care unit (ICU) management. Patients with COVID-19 have been described as hyperinflammatory, exhibiting a “cytokine storm,” prompting empiric and unproven use of potent, targeted immune modulators (e.g., interleukin-6 inhibition via monoclonal antibodies) (10, 11). Patients with COVID-19 have been described as hypercoagulable, with disordered indices of coagulation, prompting widespread escalation of routine ICU anticoagulation practices, even including empiric administration of thrombolytic therapy (12, 13). Finally, mechanically ventilated patients with COVID-19 develop high rates of extrapulmonary organ injury, prompting suggestions that conventional ARDS ventilator strategies may compromise patient hemodynamics and perpetuate multiorgan failure (3, 9, 14, 15). Some authors have even recommended deviating from standard ARDS evidence-based management strategies in severe COVID-19 (3, 9, 16), although others have strongly cautioned against such strategies (17, 18).
Yet these clinical features—variable respiratory mechanics, hyperinflammation, hypercoagulability, and extrapulmonary organ injury—are common among critically ill patients without COVID-19, including patients with ARDS. No study to date has systematically compared all of these clinical features of COVID-19 with a cohort of patients with non–COVID-19 ARDS; the disease-specific pathophysiologic features of severe COVID-19 have yet to be elucidated. We performed a retrospective cohort study comparing patients with COVID-19 who received invasive mechanical ventilation with historical control subjects with ARDS from the same tertiary care center, hypothesizing that there would be no large differences between groups in these key clinical features.
We performed a single-center retrospective cohort study. We studied consecutive adult patients (age ⩾18 yr) with COVID-19 requiring invasive mechanical ventilation who were admitted between March 1 and June 31, 2020. All patients were observed 28 days after the onset of invasive mechanical ventilation. A CONSORT diagram is provided in Figure E1 in the online supplement. COVID-19 was diagnosed using real-time reverse transcriptase–polymerase chain reaction for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). We excluded patients who were transferred for extracorporeal membrane oxygenation or initiated on invasive ventilation before the date of transfer. We excluded patients who lacked documentation in the electronic medical record of 1) height and weight, 2) arterial oxygen pressure (PaO2)/fraction of inspired oxygen (FiO2) ratio (or arterial oxygen saturation [SaO2]/FiO2 ratio [19] if no arterial blood gas was obtained), or 3) static lung compliance measurement (all within the first 24 h of invasive mechanical ventilation).
We compared patients with COVID-19 with consecutive adult patients with ARDS not caused by COVID-19 admitted between January 1, 2016, and December 31, 2017, at the same center. Portions of this cohort have been described in previous studies (20, 21). ARDS was determined using formal post hoc adjudication by critical care–trained physicians using the Berlin Criteria as previously described (20). The additional inclusion and exclusion criteria were the same as the COVID-19 cohort.
We collected baseline demographic data, including age, sex, self-reported race, height, and weight for all patients. Median values of PaO2/FiO2 ratio (or SaO2/FiO2 ratio), oxygenation index (or oxygen saturation index if no arterial blood gas was obtained [22]), and ventilatory ratio (23) were calculated from the electronic health record during the first 24 hours of mechanical ventilation. All values of static lung compliance measured at the same positive end-expiratory pressure (PEEP) were also collected during the first 24 hours and the median value was determined. The highest sequential organ failure assessment (SOFA) score over the first 24 hours of mechanical ventilation was determined. We also extracted laboratory values from the electronic health record closest to the initiation of invasive mechanical ventilation, all within 48 hours of intubation. Analyses were not adjusted for multiplicity and thus should be considered exploratory.
We compared lung compliance, ventilatory ratio, and oxygenation index before and after matching patients based on age, sex, body mass index (BMI), and the initial PEEP. Patients were matched on these characteristics because they may have an effect on respiratory mechanics independent of ARDS or COVID-19. We performed coarsened exact matching to compare these physiologic parameters in a 1:1 matched subgroup. Coarsened exact matching is a commonly used matching approach in which exact matching is performed after first “coarsening” the data (24, 25). For example, BMI is first coarsened into meaningful subgroups based on user-defined cut points or a default binning method, and then exact matches are made within subgroups. Additional matching details are described in the online supplement and matching results are shown in Table E1. We compared median physiologic variables in the unmatched and matched cohorts and performed quantile regression to estimate 95% confidence intervals (CIs) of differences in medians.
We compared time to breathing unassisted from mechanical ventilation before 28 days, which is measured from the first to the last time the patient received invasive mechanical ventilation during the hospitalization. We plotted a time-to-event curve for the time to breathing unassisted from mechanical ventilation before 28 days. To account for the competing risk of death, patients who died before Day 28 were assigned more than 28 days of invasive mechanical ventilation. We also determined patient status at 28 days, which included death, still hospitalized receiving invasive mechanical ventilation, hospitalized but breathing unassisted, and discharged. Because our institution collects death data after discharge for some patients, if there was a record that the patient died within 28 days of invasive mechanical ventilation, but after discharge, they were included as a death.
To compare time to breathing unassisted in an adjusted analysis, we fit a competing risk survival model with death as competing risk and reported the subdistributional hazards ratio (SHR). We adjusted for age, sex, self-reported race, BMI, and maximum SOFA score during the first 24 hours of mechanical ventilation. We did not adjust for measures of severity of gas exchange (e.g., oxygenation index) as these may be in the causal pathway between COVID-19 or ARDS and death. To compare 28-day mortality, we fit a log binomial model adjusting for the same covariates. To estimate adjusted mortality rates, we calculated the average marginal effect in each group after fitting the model.
Certain laboratory tests were routinely obtained in all patients with COVID-19 but not among patients with non–COVID-19 ARDS, for whom specific laboratory testing was performed based on provider discretion. This resulted in differences in missingness between groups. Median differences in laboratory results were compared in patients who had the test performed. In an additional analysis described in the online supplement, laboratory results were compared after multiple imputations (26). The missingness rates and variables used in multiple imputations are described in Table E5.
As an additional sensitivity analysis, physiologic characteristics and outcomes among patients with COVID-19 were also compared with patients with non–COVID-19 ARDS and pneumonia (as adjudicated by reviewing physicians). All statistical analysis was performed in Stata version 16.1 (STATA). The institutional review board of the University of Michigan Healthcare System (HUM00104714) approved the human study protocol and deemed that informed consent was not required.
We identified 130 consecutive patients with COVID-19 who received invasive mechanical ventilation and 382 historical patients with non–COVID-19 ARDS who met all study inclusion criteria (Figure E1). The baseline characteristics of both groups are shown in Table 1. Of the mechanically ventilated patients with COVID-19, 123/130 (95%) met Berlin Criteria for ARDS as adjudicated post hoc. There were differences in patient age, sex, self-reported race, and BMI between groups. The median delivered tidal volume was 5.9 ml/kg predicted body weight (interquartile range [IQR], 5.2–6.9) in COVID-19 and 6.0 ml/kg predicted body weight (IQR, 5.4–6.8) in non–COVID-19 ARDS, whereas median delivered PEEP was 12 (IQR, 8,14) and 8 (IQR, 5,12) for COVID-19 and non–COVID-19 ARDS, respectively. There were 182/382 (47.6%) patients with non-COVID ARDS and 66/130 (50.7%) patients with COVID-19 who had a PaO2/FiO2 ratio < 100 at any point during invasive mechanical ventilation. There were 33/130 (25%) patients with COVID-19 and 9/382 (2%) patients with non–COVID-19 ARDS who received prone positioning during the hospitalization.
| Characteristics | COVID-19 (n = 130) | Non–COVID-19 ARDS (n = 382) |
|---|---|---|
| Age, median (IQR), yr | 64 (56–71) | 59 (45–68) |
| Male, n (%) | 85 (65) | 231 (60) |
| Race, n (%) | ||
| White | 49 (38) | 315 (82) |
| Black | 55 (42) | 38 (10) |
| Other/unknown | 26 (20) | 29 (8) |
| BMI, median (IQR), kg/m2 | 31 (27–38) | 28 (24–34) |
| ARDS risk factor, n (%) | ||
| Pneumonia | — | 229 (60) |
| Aspiration | — | 82 (21) |
| Sepsis | — | 122 (32) |
| Trauma | — | 25 (7) |
| PaO2/FiO2 ratio, n (%) | ||
| <100 | 7 (5) | 25 (7) |
| 100–200 | 78 (60) | 220 (58) |
| >200 | 45 (35) | 137 (36) |
| SOFA score, median (IQR) | 11 (10–13) | 12 (10–15) |
| Vt, median (IQR) | 5.9 (5.2–6.9) | 6.0 (5.4–6.8) |
| PEEP, first 24 h, median (IQR) | 12 (8–14) | 8 (5–12) |
| Plateau pressure, first 24 h, median (IQR) | 25 (21–28) | 23 (20–27) |
| Driving pressure, first 24 h, median (IQR) | 11 (10–13) | 13 (11–17) |
| Received prone positioning, n (%) | 33 (25) | 9 (2) |
| Outcomes at 28 d after mechanical ventilation | ||
| Ventilator-free days, median (IQR) | 9 (0–19) | 9 (0–23) |
| Discharged alive, n (%) | 57 (44) | 176 (46) |
| Breathing unassisted, still hospitalized, n (%) | 20 (15) | 36 (9) |
| Invasive ventilation, n (%) | 14 (11) | 25 (7) |
| Died, n (%) | 39 (30) | 145 (38) |
First, we compared respiratory physiology at the onset of invasive mechanical ventilation in patients with COVID-19 and non–COVID-19 ARDS (see Figure 1A and Table E2). Respiratory compliance was low in both cohorts, with only 13.1% of patients with COVID-19 and 10.5% of patients with non–COVID-19 ARDS having respiratory compliance greater than 50 ml/cm H2O, and only 3.1% of patients with COVID-19 and 3.4% of patients with non–COVID-19 ARDS having respiratory compliance equal to or greater than 70 ml/cm H2O (normal respiratory compliance in mechanically ventilated patients without a respiratory disease [27, 28]). The median respiratory compliance was 34.6 ml/H2O in patients with COVID-19 and 30.0 ml/H2O in patients with non–COVID-19 ARDS (difference: 4.6 ml/H2O; 95% CI, 1.7 to 7.6). The median ventilatory ratio (an index of ventilatory efficiency, which correlates with dead space [29]) was 1.35 in COVID-19 and 1.41 in non–COVID-19 ARDS (difference: −0.05; 95% CI, −0.16 to 0.05). The median oxygenation index was 10.2 in COVID-19 and 8.1 in non–COVID-19 ARDS (difference: 2.1; 95% CI, 0.9 to 3.3). After excluding patients who received prone positioning during this comparison period, results were not different (Table E2).
Figure 1. Respiratory physiology of mechanically ventilated patients with coronavirus disease (COVID-19) and non–COVID-19 acute respiratory distress syndrome (ARDS). Measurements were collected during the first 24 hours of invasive mechanical ventilation. The middle bar represents the median value with outer bars representing the interquartile range. (A) All subjects with COVID-19 and non–COVID-19 ARDS. (B) Matched subgroup based on age, sex, body mass index, and administered positive end-expiratory pressure.
[More] [Minimize]As patients with COVID-19 and patients with non–COVID-19 ARDS differed in age, sex, BMI, and initial delivered PEEP, all of which impact respiratory mechanics, we successfully matched a 1:1 subgroup of 82 patients with COVID-19 with 82 patients with non-COVID ARDS based on age, sex, BMI, and PEEP (Table E1). In this matched subgroup (Figure 1B), median compliance was 32.1 in COVID-19 and 30.8 in non–COVID-19 ARDS (difference: 1.1; 95% CI, −3.6 to 5.8), median ventilatory ratio was 1.33 in COVID-19 and 1.54 in non–COVID-19 ARDS (difference: −0.21; 95% CI, −0.35 to −0.07), median oxygenation index was 9.8 in COVID-19 and 9.7 in non–COVID-19 ARDS (difference: 0.2; 95% CI, −1.9 to 2.2).
We next compared 28-day clinical outcomes among patients with COVID-19 and non–COVID-19 ARDS (Figure 2). We did not detect a difference in time to breathing unassisted in patients with COVID-19 and ARDS when accounting for death as a competing risk, in both an unadjusted analysis (SHR, 0.98 [95% CI, 0.77–1.26]) and analysis adjusting for age, sex, race, BMI, and SOFA score (SHR, 0.97 [95% CI, 0.73–1.29]). Unadjusted mortality at 28-days was 30% in COVID-19 compared with 38% in non-COVID ARDS (risk ratio, 0.79 [95% CI, 0.59–1.06]). Adjusted 28-day mortality 36.1% in patients with COVID-19 and was 35.7% in non–COVID-19 ARDS (adjusted risk ratio, 1.01 [95% CI, 0.72–1.42]). When the analysis was limited to patients with non–COVID-19 ARDS and pneumonia as an identified risk factor, initial pulmonary physiology was similar although ventilator-free days were higher in pneumonia causing ARDS compared with COVID-19 (Table E3).
Figure 2. Comparison of time to unassisted breathing in mechanically ventilated patients with coronavirus disease (COVID-19) and non–COVID-19 ARDS. When plotting the time-to-event curve, outcomes were censored at 28 days and patients who died before 28 days were assigned 28 days of mechanical ventilation to account for death as a competing risk. The unadjusted subdistributional HR for time to breathing unassisted was calculated using a competing risk survival model, which also accounts for death as a competing risk. ARDS = acute respiratory distress syndrome; CI = confidence interval; HR = hazard ratio.
[More] [Minimize]We performed an exploratory analysis to identify potential differences in systemic inflammation, extrapulmonary organ injury, and disordered coagulation among patients with COVID-19 and patients with non–COVID-19 ARDS with laboratory results performed (Figure 3 and Table E4). Indices of systemic inflammation and extrapulmonary organ injury were largely similar in patients with COVID-19 and patients with non–COVID-19 ARDS at the initiation of mechanical ventilation. Total white blood cell count was lower in patients with COVID-19 (8.9 K/μl [IQR, 6.7–11.3] than in patients with non–COVID-19 ARDS (12.2 K/μl [IQR, 7.1–18]). This difference was reflected in measured neutrophil counts, which were 7.5 K/μl [IQR, 5.3–9.6] among patients with COVID-19 versus 9.6 K/μl [IQR, 5.2–14.7] among patients with non–COVID-19 ARDS. COVID-19 and non–COVID-19 ARDS did not have detectable differences in lymphocyte count or nonspecific indices of systematic inflammation (erythrocyte sedimentation rate and C-reactive protein), although there was considerable missing data among patients with non–COVID-19 ARDS for these indices.
Figure 3. Extrapulmonary clinical features in mechanically ventilated patients with coronavirus disease (COVID-19) and non–COVID-19 acute respiratory distress syndrome (ARDS). (A–C) Comparison of indices of inflammation (A), extrapulmonary organ injury (B), and coagulation (C) were performed using measurements performed within 48 hours of invasive mechanical ventilation initiation. The middle bar represents the median value with outer bars representing the interquartile range. ALT = alanine transaminase; AST = aspartate transaminase.
[More] [Minimize]Procalcitonin was frequently elevated in patients with COVID-19 (0.42 ng/ml [IQR: 0.22, 1.19] but higher in non-COVID ARDS (1.33 ng/ml [IQR, 0.27–9.61]). At the time of invasive mechanical ventilation, 46% of patients with COVID-19 with a procalcitonin concentration measured (55/119) had values above the normal reference value for “high probability for bacterial infection” (⩾0.50 ng/ml), as compared with 64% (145/228) of patients with non–COVID-19 ARDS. In contrast, patients with COVID-19 and patients with non–COVID-19 ARDS did not differ in their indices of renal dysfunction (blood urea nitrogen and serum creatinine) or liver injury (aspartate transaminase and alanine transaminase) at the initiation of mechanical ventilation (Table E4).
The disordered coagulation of patients with COVID-19 may have been distinct from that of patients with non–COVID-19 ARDS. Platelet count was higher in patients with COVID-19 (206 K/μl [IQR, 175–272] versus 177 K/μl [IQR, 89–262]; difference: 28 [95% CI, 3–53]), as was fibrinogen (572 mg/dl [IQR, 474–650] versus 339 mg/dl [IQR, 227–468]). Elevated D-dimer concentrations were common in both cohorts when measured, and higher among patients with non–COVID-19 ARDS (4.9 mg/L [IQR, 2–12.5] versus 2.0 mg/L [IQR, 1.1–6.9]), although there were many missing D-dimer measurements among patients with ARDS (79%, 81/382). The analysis comparing laboratory results with missing data imputed using multiple imputations was largely similar (Table E5).
We compared a moderate-sized series of patients with COVID-19 who received invasive mechanical ventilation with patients with non–COVID-19 ARDS who also received invasive mechanical ventilation at the same center. We did not identify major differences between the two groups in their respiratory physiology or 28-day clinical outcomes.
Early anecdotal reports suggested that most patients with severe COVID-19 exhibit “atypical ARDS” with “near-normal respiratory system compliance,” reflecting unique, non-ARDS pathophysiology (2, 3, 9, 16). Yet our findings that patients with COVID-19 have profoundly impaired respiratory mechanics, comparable to those of patients with non–COVID-19 ARDS, are aligned with numerous cohort studies now representing over 1,800 mechanically ventilated patients with COVID-19 across medical centers and continents (8, 30–40). Our results also validate several recent studies that directly compared respiratory mechanics in cohorts of patients with COVID-19 and non–COVID-19 ARDS and found no meaningful difference in respiratory system compliance or other indices of respiratory mechanics (33, 38, 39). A recently published series of 32 patients with COVID-19, which reported preserved respiratory compliance in COVID-19 compared with non–COVID-19 ARDS, was an outlier among this group (41). The findings of our single-center study should be interpreted in the context of this now considerable literature (e.g., in a systematic meta-analysis).
Although both of our cohorts (COVID-19 and non–COVID-19 ARDS) had considerable physiologic heterogeneity, very few patients in either cohort had “near-normal” respiratory compliance (27, 28). In fact, the respiratory compliance of our COVID-19 cohort (34.6 ml/H2O) was nearly identical to that of the 4,188-patient cohort used to derive and validate the Berlin Definition of ARDS (34.0 ml/H2O) (1) or that of recent clinical trials of patients with severe ARDS (e.g., the PROSEVA trial, 35 ml/H2O [42]). Although authors have conjectured that impaired respiratory compliance is a temporally late feature of COVID-19 lung injury, with patients initially presenting with preserved compliance (3, 9, 16), four studies comprising over 600 mechanically ventilated patients with COVID-19 have reported serial measurements of respiratory compliance (8, 34–36), none finding any temporal trend toward decreased compliance after initiation of mechanical ventilation. A recent meta-analysis of three studies also found no correlation between duration of symptoms and respiratory compliance in patients with COVID-19 (33, 43). Although our confidence intervals exclude large differences in physiologic measurements across COVID-19 and non–COVID-19 ARDS cohorts, the range of possible differences supported by our compliance data could correspond to a 2–4% difference in predicted mortality based on recent work in ARDS cohorts (44). Taken together, our study and other large series do not support the claim that patients with COVID-19 exhibit unique or atypical respiratory mechanics compared with other causes of ARDS.
The 28-day mortality rate of 30% among our COVID-19 cohort contrasts starkly with initial reports describing very high mortality among mechanically ventilated patients with COVID-19 (88–97%) (4, 45). This discrepancy is at least partly explained by the administrative censoring used in early studies, in which only patients with an outcome of death or discharge were analyzed at the time of study reporting, excluding many other patients still hospitalized. In contrast, we restricted our analysis to patients with COVID-19 who were observed at least 28 days after the onset of invasive mechanical ventilation. Both our clinical outcomes and our ventilator management (assessed in terms of lung-protective ventilation and PEEP strategies) were comparable to those of patients with non–COVID-19 ARDS, similar to reports at other centers. These data do not support the widespread speculation that conventional, evidence-based ventilator strategies for ARDS may cause harm in COVID-19 due to the purportedly unique pathophysiology of the disease (3, 14, 15).
We observed considerable heterogeneity among patients with COVID-19, both in their respiratory physiology and in their extrapulmonary disease manifestations. The heterogeneity of respiratory physiology (in terms of both severity of lung injury and respiratory mechanics) was comparable to that of patients with non–COVID-19 ARDS, a surprising finding given that whereas COVID-19 is a pathogen-mediated disease, ARDS is a syndrome arising from highly divergent underlying etiologies. Given the well-described heterogeneity of ARDS, there have been multiple recent efforts to identify more uniform ARDS subgroups, such as by etiology (46), chest imaging findings (47), and blood biomarkers reflecting hyperinflammatory or hypoinflammatory subphenotypes (48–50). Two recent reports have confirmed the presence of subphenotypes among critically ill patients with COVID-19 that parallel those observed in non–COVID-19 ARDS (51) and sepsis (52). The observation that patients with severe COVID-19 still exhibit considerable heterogeneity suggests that clustering patients with ARDS by etiology (e.g., COVID-19) may not, by itself, be a sufficient strategy for identifying homogenous subgroups. Other underlying sources of biologic variation (e.g., genetic (53, 54), metabolomic (55, 56), and microbiologic (57–59) variation across patients) should be explored to identify clinically, prognostically, and therapeutically distinct subgroups.
Our study found that many extrapulmonary clinical features of severe COVID-19 are common among patients with non–COVID-19 causes of ARDS. Although early reports identified features purportedly specific to SARS-CoV-2 pathogenesis, these studies only compared patients with COVID-19 to reference ranges derived from healthy subjects. As an example, early reports noted immune dysfunction in severe COVID-19, manifesting with lymphopenia (60) and reflecting a supposed “cytokine storm.” (10) Yet our study found similar frequency and severity of lymphopenia and nonspecific indices of systemic inflammation in patients with COVID-19 and non–COVID-19 ARDS. Systemic concentrations of inflammatory cytokines are actually lower in COVID-19 than in other causes of ARDS (11, 61), and the clinical efficacy of cytokine inhibition in COVID-19 patients has been conflicting (62–67). Taken together, these data underscore the critical importance of rigorously testing early pathophysiologic hypotheses before modifying clinical practice.
Similarly, early speculation blamed inappropriate mechanical ventilation, rather than SARS-CoV-2 infection, for the high rates of extrapulmonary organ injury observed in patients with severe COVID-19 (3, 14, 15, 68). Yet our data suggest that extrapulmonary organ injury both is common among patients with COVID-19 at the time of intubation and may be comparable in frequency to that of non–COVID-19 ARDS. Although our data suggest that the pattern of discorded coagulation may be distinct among patients with COVID-19 (with higher platelet and fibrinogen values than non–COVID-19 ARDS patients), elevated D-dimer concentrations (which have been used in patients with COVID-19 to justify an escalation of anticoagulation regimens and even empiric thrombolysis [12, 13]) are extremely common when tested in patients with non–COVID-19 ARDS. Although we are confident that COVID-19–specific features of pathogenesis will be identified, our study underscores the importance of basing inferences on adequately powered cohorts with appropriate control arms, as opposed to underpowered, uncontrolled, and anecdotal experience (7). We emphasize that our single-center study should be interpreted in the context of the now considerable published literature of mechanically ventilated patients with COVID-19 (e.g., via meta-analysis or systematic review). Specifically, the high percentage of missing data in our comparison of coagulation parameters is an important source of bias, and these comparisons are no more than exploratory.
The current study has several limitations. Because patients were analyzed at a single center, they were all treated under largely standardized institutional practices, which may strengthen the between-group comparisons. However, the non–COVID-19 ARDS cohort was managed during a different time period than the COVID-19 cohort; thus, some important variation in clinical practice was observed (e.g., use of prone positioning). In addition, the racial demographics of our COVID-19 and non–COVID-19 ARDS cohorts differed, attributable to a high number of emergency department–to–emergency department transfers from a nearby urban center during the COVID-19 surge. Although the study’s single-center nature may also limit generalizability, both the respiratory physiology and clinical outcomes of our COVID-19 cohort were similar to those of other recent well-powered COVID-19 cohorts (8, 30–40), suggesting it is likely representative of patients with severe COVID-19. We analyzed data recorded in the electronic medical record, which introduces variation in availability and reliability of recorded measurements and precludes analysis of respiratory parameters not routinely recorded. Other practice changes during the COVID-19 surge at our institution also prevented reliable comparison of venous thromboembolic events between groups. Although our patients without COVID-19 were studied over a 2-year period, our patients with COVID-19 were studied over a single 4-month period, introducing a potential influence of seasonality. Comparison of some extrapulmonary features of disease was limited by missing laboratory values for many patients without COVID-19. Although these comparisons were robust to multiple imputations, we emphasize that this analysis is exploratory and may be confounded by other factors in clinician ordering practices.
Within the limits of our sample size, we identified only small differences in respiratory physiology between patients with COVID-19 and non–COVID-19 ARDS. Clinical outcomes were also similar between groups, although the smaller sample size of this analysis precludes more definitive conclusions. Many key clinical features of COVID-19 (hyperinflammation, hypercoagulability, and extrapulmonary organ injury) are also common in non–COVID-19 ARDS. Although disease-specific features of SARS-CoV-2 pathogenesis will surely be elicited, we found no physiologic evidence that lung injury from COVID-19 falls outside of the heterogenous spectrum of ARDS arising from other etiologies. Further studies are needed to define COVID-19–specific clinical features before deviation from evidence-based treatment practices can be recommended.
The authors thank Ann Wolski, Eric Wilson, and Ryan Townshend for their assistance with data extraction from the electronic medical record. They also thank Chris Gillies, Sardar Ansari, Jonathan Motyka, and the Michigan Center for Integrative Research in Critical Care for data infrastructure and bioinformatic resources.
| 1. | Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, Fan E, et al.; ARDS Definition Task Force. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012;307:2526–2533. |
| 2. | Gattinoni L, Coppola S, Cressoni M, Busana M, Rossi S, Chiumello D. COVID-19 does not lead to a “typical” acute respiratory distress syndrome. Am J Respir Crit Care Med 2020;201:1299–1300. |
| 3. | Marini JJ, Gattinoni L. Management of COVID-19 respiratory distress. JAMA 2020;323:2329–2330. |
| 4. | Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn T, Davidson KW, et al.; the Northwell COVID-19 Research Consortium. Presenting characteristics, comorbidities, and outcomes among 5,700 patients hospitalized with COVID-19 in the New York city area. JAMA 2020;323:2052–2059. |
| 5. | Arentz M, Yim E, Klaff L, Lokhandwala S, Riedo FX, Chong M, et al. Characteristics and outcomes of 21 critically ill patients with COVID-19 in Washington state. JAMA 2020;323:1612–1614. |
| 6. | Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 2020;395:497–506. |
| 7. | Bos LDJ, Sinha P, Dickson RP. The perils of premature phenotyping in COVID-19: a call for caution. Eur Respir J 2020;56:2001768. |
| 8. | Ziehr DR, Alladina J, Petri CR, Maley JH, Moskowitz A, Medoff BD, et al. Respiratory pathophysiology of mechanically ventilated patients with COVID-19: a cohort study. Am J Respir Crit Care Med 2020;201:1560–1564. |
| 9. | Gattinoni L, Chiumello D, Rossi S. COVID-19 pneumonia: ARDS or not? Crit Care 2020;24:154. |
| 10. | Mehta P, McAuley DF, Brown M, Sanchez E, Tattersall RS, Manson JJ; HLH Across Speciality Collaboration, UK. COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet 2020;395:1033–1034. |
| 11. | Sinha P, Matthay MA, Calfee CS. Is a “cytokine storm” relevant to COVID-19? JAMA Intern Med 2020;180:1152–1154. |
| 12. | Poor HD, Ventetuolo CE, Tolbert T, Chun G, Serrao G, Zeidman A, et al. COVID-19 critical illness pathophysiology driven by diffuse pulmonary thrombi and pulmonary endothelial dysfunction responsive to thrombolysis. Clin Transl Med 2020;10:e44. |
| 13. | McGonagle D, O’Donnell JS, Sharif K, Emery P, Bridgewood C. Immune mechanisms of pulmonary intravascular coagulopathy in COVID-19 pneumonia. Lancet Rheumatol 2020;2:e437–e445. |
| 14. | Ducharme J. Why ventilators may not be working as well for COVID-19 patients as doctors hoped. New York, NY: Time Magazine; 2020. [accessed 2020 May 1]. Available from: https://time.com/5820556/ventilators-covid-19/. |
| 15. | Stobbe M. Some doctors moving away from ventilators for virus patients. New York, NY:Associated Press; 2020 [accessed 2020 May 1]. Available from: https://apnews.com/article/8ccd325c2be9bf454c2128dcb7bd616d. |
| 16. | Gattinoni L, Chiumello D, Caironi P, Busana M, Romitti F, Brazzi L, et al. COVID-19 pneumonia: different respiratory treatments for different phenotypes? Intensive Care Med 2020;46:1099–1102. |
| 17. | Fan E, Beitler JR, Brochard L, Calfee CS, Ferguson ND, Slutsky AS, et al. COVID-19-associated acute respiratory distress syndrome: is a different approach to management warranted? Lancet Respir Med 2020;8:816–821. |
| 18. | Rice TW, Janz DR. In defense of evidence-based medicine for the treatment of COVID-19 acute respiratory distress syndrome. Ann Am Thorac Soc 2020;17:787–789. |
| 19. | Brown SM, Grissom CK, Moss M, Rice TW, Schoenfeld D, Hou PC, et al.; NIH/NHLBI PETAL Network Collaborators. Nonlinear imputation of Pao2/Fio2 from Spo2/Fio2 among patients with acute respiratory distress syndrome. Chest 2016;150:307–313. |
| 20. | Sjoding MW, Hofer TP, Co I, Courey A, Cooke CR, Iwashyna TJ. Interobserver reliability of the Berlin ARDS definition and strategies to improve the reliability of ARDS Diagnosis. Chest 2018;153:361–367. |
| 21. | Sjoding MW, Hofer TP, Co I, McSparron JI, Iwashyna TJ. Differences between patients in whom physicians agree and disagree about the diagnosis of acute respiratory distress syndrome. Ann Am Thorac Soc 2019;16:258–264. |
| 22. | DesPrez K, McNeil JB, Wang C, Bastarache JA, Shaver CM, Ware LB. Oxygenation saturation index predicts clinical outcomes in ARDS. Chest 2017;152:1151–1158. |
| 23. | Sinha P, Fauvel NJ, Singh S, Soni N. Ventilatory ratio: a simple bedside measure of ventilation. Br J Anaesth 2009;102:692–697. |
| 24. | Blackwell M, Iacus S, King G, Porro G. cem: coarsened exact matching in Stata. Stata J 2009;9:524–546. |
| 25. | Iacus SM, King G, Porro G. Causal inference without balance checking: coarsened exact matching. Polit Anal 2012;20:1–24. |
| 26. | Royston P. Multiple imputation of missing values. Stata J 2004;4:227–241. |
| 27. | Dohi S, Gold MI. Pulmonary mechanics during general anaesthesia. The influence of mechanical irritation on the airway. Br J Anaesth 1979;51:205–214. |
| 28. | Bromage PR. Total respiratory compliance in anaesthetized subjects and modifications produced by noxious stimuli. Clin Sci 1958;17: 217–236. |
| 29. | Sinha P, Calfee CS, Beitler JR, Soni N, Ho K, Matthay MA, et al. Physiologic analysis and clinical performance of the ventilatory ratio in acute respiratory distress syndrome. Am J Respir Crit Care Med 2019;199:333–341. |
| 30. | Auld SC, Caridi-Scheible M, Blum JM, Robichaux C, Kraft C, Jacob JT, et al.; Emory COVID-19 Quality and Clinical Research Collaborative. ICU and ventilator mortality among critically ill adults with coronavirus disease 2019. Crit Care Med 2020;48:e799–e804. |
| 31. | Bos LDJ, Paulus F, Vlaar APJ, Beenen LFM, Schultz MJ. Subphenotyping acute respiratory distress syndrome in patients with COVID-19: consequences for ventilator management. Ann Am Thorac Soc 2020;17:1161–1163. |
| 32. | Cummings MJ, Baldwin MR, Abrams D, Jacobson SD, Meyer BJ, Balough EM, et al. Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study. Lancet 2020;395:1763–1770. |
| 33. | Haudebourg AF, Perier F, Tuffet S, de Prost N, Razazi K, Mekontso Dessap A, et al. Respiratory mechanics of COVID-19- versus non-COVID-19-associated acute respiratory distress syndrome. Am J Respir Crit Care Med 2020;202:287–290. |
| 34. | Schenck EJ, Hoffman K, Goyal P, Choi J, Torres L, Rajwani K, et al. Respiratory mechanics and gas exchange in COVID-19-associated respiratory failure. Ann Am Thorac Soc 2020;17:1158–1161. |
| 35. | Bhatraju PK, Ghassemieh BJ, Nichols M, Kim R, Jerome KR, Nalla AK, et al. Covid-19 in critically ill patients in the Seattle region - case series. N Engl J Med 2020;382:2012–2022. |
| 36. | Ferrando C, Suarez-Sipmann F, Mellado-Artigas R, Hernández M, Gea A, Arruti E, et al.; COVID-19 Spanish ICU Network. Clinical features, ventilatory management, and outcome of ARDS caused by COVID-19 are similar to other causes of ARDS. Intensive Care Med 2020;46:2200–2211. [Published erratum appears in Intensive Care Med 47:144–146.] |
| 37. | Mitra AR, Fergusson NA, Lloyd-Smith E, Wormsbecker A, Foster D, Karpov A, et al. Baseline characteristics and outcomes of patients with COVID-19 admitted to intensive care units in Vancouver, Canada: a case series. CMAJ 2020;192:E694–E701. |
| 38. | Grasselli G, Tonetti T, Protti A, Langer T, Girardis M, Bellani G, et al.; collaborators. Pathophysiology of COVID-19-associated acute respiratory distress syndrome: a multicentre prospective observational study. Lancet Respir Med 2020;8:1201–1208. |
| 39. | Brault C, Zerbib Y, Kontar L, Fouquet U, Carpentier M, Metzelard M, et al. COVID-19– versus non–COVID-19–related acute respiratory distress syndrome: differences and similarities. Am J Respir Crit Care Med 2020;202:1301–1304. |
| 40. | Barbeta E, Motos A, Torres A, Ceccato A, Ferrer M, Cilloniz C, et al.; Covid Clinic Critical Care Group. SARS-CoV-2–induced acute respiratory distress syndrome: pulmonary mechanics and gas-exchange abnormalities. Ann Am Thorac Soc 2020;17:1164–1168. |
| 41. | Chiumello D, Busana M, Coppola S, Romitti F, Formenti P, Bonifazi M, et al. Physiological and quantitative CT-scan characterization of COVID-19 and typical ARDS: a matched cohort study. Intensive Care Med 2020;46:2187–2196. |
| 42. | Guérin C, Reignier J, Richard JC, Beuret P, Gacouin A, Boulain T, et al.; PROSEVA Study Group. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med 2013;368:2159–2168. |
| 43. | Bos LDJ, Sinha P, Dickson RP. Response to “COVID-19 conundrum: clinical phenotyping based on pathophysiology as a promising approach to guide therapy in a novel illness” and “Strengthening the foundation of the house of CARDS by phenotyping on the fly” and “COVID-19 phenotypes: leading or misleading?”. Eur Respir J 2020;56:2002756. |
| 44. | Sedhai YR, Yuan M, Ketcham SW, Co I, Claar DD, McSparron JI, et al. Validating measures of disease severity in acute respiratory distress syndrome. Ann Am Thorac Soc [online ahead of print] 21 Dec 2020; DOI: 10.1513/AnnalsATS.202007-772OC. |
| 45. | Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet 2020;395:1054–1062. |
| 46. | Calfee CS, Janz DR, Bernard GR, May AK, Kangelaris KN, Matthay MA, et al. Distinct molecular phenotypes of direct vs indirect ARDS in single-center and multicenter studies. Chest 2015;147:1539–1548. |
| 47. | Constantin JM, Jabaudon M, Lefrant JY, Jaber S, Quenot JP, Langeron O, et al.; AZUREA Network. Personalised mechanical ventilation tailored to lung morphology versus low positive end-expiratory pressure for patients with acute respiratory distress syndrome in France (the LIVE study): a multicentre, single-blind, randomised controlled trial. Lancet Respir Med 2019;7:870–880. |
| 48. | Calfee CS, Delucchi K, Parsons PE, Thompson BT, Ware LB, Matthay MA; NHLBI ARDS Network. Subphenotypes in acute respiratory distress syndrome: latent class analysis of data from two randomised controlled trials. Lancet Respir Med 2014;2:611–620. |
| 49. | Delucchi K, Famous KR, Ware LB, Parsons PE, Thompson BT, Calfee CS; ARDS Network. Stability of ARDS subphenotypes over time in two randomised controlled trials. Thorax 2018;73:439–445. |
| 50. | Bos LDJ, Scicluna BP, Ong DSY, Cremer O, van der Poll T, Schultz MJ. Understanding heterogeneity in biologic phenotypes of acute respiratory distress syndrome by leukocyte expression profiles. Am J Respir Crit Care Med 2019;200:42–50. |
| 51. | Sinha P, Calfee CS, Cherian S, Brealey D, Cutler S, King C, et al. Prevalence of phenotypes of acute respiratory distress syndrome in critically ill patients with COVID-19: a prospective observational study. Lancet Respir Med 2020;8:1209–1218. |
| 52. | Bhavani SV, Huang ES, Verhoef PA, Churpek MM. Novel temperature trajectory subphenotypes in COVID-19. Chest 2020;158:2436–2439. |
| 53. | Bime C, Pouladi N, Sammani S, Batai K, Casanova N, Zhou T, et al. Genome-wide association study in African Americans with acute respiratory distress syndrome identifies the selectin P ligand gene as a risk factor. Am J Respir Crit Care Med 2018;197:1421–1432. |
| 54. | Morrell ED, O’Mahony DS, Glavan BJ, Harju-Baker S, Nguyen C, Gunderson S, et al. Genetic variation in MAP3K1 associates with ventilator-free days in acute respiratory distress syndrome. Am J Respir Cell Mol Biol 2018;58:117–125. |
| 55. | Evans CR, Karnovsky A, Kovach MA, Standiford TJ, Burant CF, Stringer KA. Untargeted LC-MS metabolomics of bronchoalveolar lavage fluid differentiates acute respiratory distress syndrome from health. J Proteome Res 2014;13:640–649. |
| 56. | Rogers AJ, Contrepois K, Wu M, Zheng M, Peltz G, Ware LB, et al. Profiling of ARDS pulmonary edema fluid identifies a metabolically distinct subset. Am J Physiol Lung Cell Mol Physiol 2017;312:L703–L709. |
| 57. | Dickson RP, Schultz MJ, van der Poll T, Schouten LR, Falkowski NR, Luth JE, et al.; Biomarker Analysis in Septic ICU Patients (BASIC) Consortium. Lung microbiota predict clinical outcomes in critically ill patients. Am J Respir Crit Care Med 2020;201:555–563. |
| 58. | Ashley S, Sjoding M, Popova A, Cui T, Hoostal M, Schmidt T, et al. Lung and gut microbiota are altered by hyperoxia and contribute to oxygen-induced lung injury. Sci Transl Med 2020;12:eaau9959. |
| 59. | Dickson RP, Singer BH, Newstead MW, Falkowski NR, Erb-Downward JR, Standiford TJ, et al. Enrichment of the lung microbiome with gut bacteria in sepsis and the acute respiratory distress syndrome. Nat Microbiol 2016;1:16113. |
| 60. | Yang W, Cao Q, Qin L, Wang X, Cheng Z, Pan A, et al. Clinical characteristics and imaging manifestations of the 2019 novel coronavirus disease (COVID-19):a multi-center study in Wenzhou city, Zhejiang, China. J Infect 2020;80:388–393. |
| 61. | Leisman DE, Ronner L, Pinotti R, Taylor MD, Sinha P, Calfee CS, et al. Cytokine elevation in severe and critical COVID-19: a rapid systematic review, meta-analysis, and comparison with other inflammatory syndromes. Lancet Respir Med 2020;8:1233–1244. |
| 62. | Salama C, Han J, Yau L, Reiss WG, Kramer B, Neidhart JD, et al. Tocilizumab in patients hospitalized with covid-19 pneumonia. N Engl J Med 2021;384:20–30. |
| 63. | Gordon AC, Mouncey PR, Al-Beidh F, Rowan KM, Nichol AD, Arabi YM, et al. Interleukin-6 receptor antagonists in critically ill patients with Covid-19 – preliminary report [preprint]. medRxiv; 2021 [accessed 2021 Jan 15]. Available from: https://www.medrxiv.org/content/10.1101/2021.01.07.21249390v2. |
| 64. | Salvarani C, Dolci G, Massari M, Merlo DF, Cavuto S, Savoldi L, et al.; RCT-TCZ-COVID-19 Study Group. Effect of tocilizumab vs standard care on clinical worsening in patients hospitalized with COVID-19 pneumonia: a randomized clinical trial. JAMA Intern Med 2021;181:24–31. |
| 65. | Stone JH, Frigault MJ, Serling-Boyd NJ, Fernandes AD, Harvey L, Foulkes AS, et al.; BACC Bay Tocilizumab Trial Investigators. Efficacy of tocilizumab in patients hospitalized with Covid-19. N Engl J Med 2020;383:2333–2344. |
| 66. | Rosas I, Bräu N, Waters M, Go RC, Hunter BD, Bhagani S, et al. Tocilizumab in hospitalized patients with COVID-19 pneumonia [preprint]. medRxiv; 2020 [accessed 2021 Jan 15]. Available from: https://www.medrxiv.org/content/10.1101/2020.08.27.20183442v2. |
| 67. | Veiga VC, Prats JAGG, Farias DLC, Rosa RG, Dourado LK, Zampieri FG, et al.; Coalition covid-19 Brazil VI Investigators. Effect of tocilizumab on clinical outcomes at 15 days in patients with severe or critical coronavirus disease 2019: randomised controlled trial. BMJ 2021; 372:n84. |
| 68. | Savel RH, Shiloh AL, Saunders PC, Kupfer Y. Mechanical ventilation during the coronavirus disease 2019 pandemic: combating the tsunami of misinformation from mainstream and social media. Crit Care Med 2020;48:1398–1400. |
Supported by National Institutes for Health grants R01HL144599, R01LM013325, R21AI137669, K23HL13064, K01HL136687, and F32HL149337.
Author Contributions: Conception and design: M.W.S. and R.P.D. Acquisition of data: M.W.S., A.K.S., S.G.K., and C.A.B. Analysis and interpretation of data: M.W.S., A.J.A., A.K.S., S.G.K., C.A.B., I.C., D.C., J.I.M., and R.P.D. Drafting or revising of the manuscript: M.W.S., A.J.A., A.K.S., S.G.K., C.A.B., I.C., D.C., J.I.M., and R.P.D. Final approval of manuscript: M.W.S., A.J.A., A.K.S., S.G.K., C.A.B., I.C., D.C., J.I.M., and R.P.D.
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.
Author disclosures are available with the text of this article at www.atsjournals.org.
