Outcomes of mechanically ventilated patients with COVID-19 associated respiratory failure.

PURPOSE: The outcomes of patients requiring invasive mechanical ventilation for COVID-19 remain poorly defined. We sought to determine clinical characteristics and outcomes of patients with COVID-19 managed with invasive mechanical ventilation in an appropriately resourced US health care system.
METHODS: Outcomes of COVID-19 infected patients requiring mechanical ventilation treated within the Inova Health System between March 5, 2020 and April 26, 2020 were evaluated through an electronic medical record review.
RESULTS: 1023 COVID-19 positive patients were admitted to the Inova Health System during the study period. Of these, 164 (16.0%) were managed with invasive mechanical ventilation. All patients were followed to definitive disposition. 70/164 patients (42.7%) had died and 94/164 (57.3%) were still alive. Deceased patients were older (median age of 66 vs. 55, p <0.0001) and had a higher initial d-dimer (2.22 vs. 1.31, p = 0.005) and peak ferritin levels (2998 vs. 2077, p = 0.016) compared to survivors. 84.3% of patients over 70 years old died in the hospital. Conversely, 67.4% of patients age 70 or younger survived to hospital discharge. Younger age, non-Caucasian race and treatment at a tertiary care center were all associated with survivor status.
CONCLUSION: Mortality of patients with COVID-19 requiring invasive mechanical ventilation is high, with particularly daunting mortality seen in patients of advanced age, even in a well-resourced health care system. A substantial proportion of patients requiring invasive mechanical ventilation were not of advanced age, and this group had a reasonable chance for recovery.

Introduction

Since its start in late 2019 in Wuhan, China, the Coronavirus Disease 2019 (COVID-19) has blossomed into a worldwide pandemic, infecting 3 million people and killing over 200,000 [1]. The rapid spread of the disease has been paralleled by an explosion of publications on the topic, with over 7,500 publications produced by a PubMed search of “COVID-19” at the start of May 2020. Despite the intense interest and effort of the medical community to better understand and treat COVID-19, substantial gaps in our knowledge of the disease remain. One particularly important area lacking clarity is the prognosis of COVID-19 patients with acute respiratory failure requiring invasive mechanical ventilation (IMV). Mortality estimates vary substantially, ranging from 16 to 97%, with multiple studies citing mortality in excess of 50% [2-7].

These reports have led to alarming headlines in the lay press, such as “Most COVID Patients Placed On Ventilators Died, New York Study Shows,” the title of an article recently published in U.S. News and World Report [8]. However, there are significant limitations to the available literature. Much of it is derived from centers outside of the United States, where the standard of care and patient populations may differ from those seen in most United States hospitals. In addition, many of these reports come from hospitals that were experiencing a major surge in patient volumes and were forced to use suboptimal equipment and staffing models that varied considerably from typical practice. Using the data from these studies to provide counseling to families and patients with impending respiratory failure may provide an unrealistically grim estimate of the chance of survival, leading some to forego potentially life-saving treatment. We sought to delineate the survival of patients with acute respiratory failure from COVID-19 requiring IMV in a United States hospital system with a high volume of COVID-19 patients, but not surging to a capacity that outstripped the ability to provide critical care in line with the conventional standard of care.

Methods

Data on all COVID-19 positive patients who were placed on IMV for acute respiratory failure within the Inova Health System in Northern Virginia between March 5, 2020 and April 26, 2020 was collected. Outcomes were reassessed on August 19, 2020. The Inova Health System consists of five hospitals including a large tertiary care center and four community hospitals. COVID-19 infection was confirmed by a positive result on polymerase chain reaction testing from either a nasopharyngeal or lower respiratory tract sample. There were no transfers of COVID-19 patients into or out of the Inova Health System during the study period. Transfers within the health system to the tertiary care hospital were analyzed as a single hospitalization attributed to the accepting facility.

All data was collected from the electronic medical record (Epic®). Data collected included patient demographics (race, ethnicity age, gender), comorbidities, adjunctive respiratory treatments [paralysis, prone positioning, inhaled pulmonary vasodilators including inhaled nitric oxide, inhaled epoprostenol, and extracorporeal membrane oxygenation (ECMO)], COVID-19 targeted treatments (clinical trial enrollment, use of toculizumab, hydroxychloroquine, remdesivir, convalescent plasma), secondary infections, and outcomes [extubation, ventilator days, discharge, death, hospital length of stay, development of acute kidney injury, and need for renal replacement therapy (RRT)]. Cause of death was determined by chart review. Immunosuppressed individuals were comprised of solid organ transplant recipients, patients on active chemotherapy, and individuals on chronic immunosuppression for any other indication (at an equivalent of prednisone 20 mg daily or higher). In the event of reintubation, ventilator length of stay was calculated from date of the initial intubation until final extubation. Outcomes were unknown for a subset of patients who remained on ventilator support or hospitalized at the time of data censoring.

The initial and highest values of laboratory data including white blood cell count, ferritin, C-reactive protein (CRP), and D-dimer were also collected. Values listed as greater than or less than the maximal or minimal test value were listed as that cutoff value (e.g. d-dimer > 20 was recorded as 20). Finally, respiratory/ventilator parameters including highest positive end expiratory pressure (PEEP), highest fraction of inspired oxygen (FiO2) required and lowest ratio of pulmonary arterial oxygen tension to FiO2 (P/F ratio) were collected.

The strategy for management of acute respiratory failure was fairly homogenous across the system. Efforts were made to avoid intubation where feasible with the use of reservoir cannulas and high flow nasal cannula (HFNC). Non-invasive ventilation was largely avoided early on due to concerns about aerosolization of the virus, but was increasingly utilized over time. Inhaled nitric oxide was delivered via HFNC in a number of patients. Self-proning was incorporated where appropriate in non-intubated patients. If these measures failed and intubation was required, patients were typically managed initially with moderate PEEP (10–12 cm H20) and a lung protective ventilator strategy targeting tidal volumes of 6 mL/Kg of ideal body weight (IBW) and plateau pressures < 30 cm H20. In patients with compliant lungs, tidal volumes were often liberalized to 7–8 mL/Kg IBW as long as plateau pressure remained < 30 cm H20. Alternatively, some patients were switched to pressure control ventilation. Ultimately the ventilator strategy was left to the discretion of the attending intensivist. Paralysis was frequently utilized in patients with severe ARDS, defined as P/F ratio < 150. Prone positioning was also utilized in patients with a P/F ratio < 150 who required FiO2 of ≥ 60% and PEEP ≥ 10 cm H20. Patients were maintained in the prone position for 16 hours or longer when performed. A conservative fluid strategy was utilized whenever possible, but was not undertaken at the expense of worsening shock. Use of inhaled pulmonary vasodilator therapy was poorly standardized and left to the discretion of the attending intensivist. The choice of sedation and analgesia was also implemented at the discretion of the attending intensivist and was targeted to a Richmond Agitation Sedation Scale (RASS) of 0 to -2 [9]. Patients were considered for venovenous (VV) ECMO if age < 60 years old, on IMV < 10 days, had a P/F ratio < 100 and/or failed lung protective ventilation, despite neuromuscular blockade and prone positioning, or had recalcitrant hypercapnic acidosis affecting perfusion.

Treatments targeting COVID-19 disease were administered at the discretion of the attending physician and were subject to availability. Treatments utilized included hydroxychloroquine, toculizumab, convalescent plasma, remdesivir (either compassionate use or via clinical trial), and sarilumab via clinical trial. Need for and duration of antimicrobial agents was dictated by the attending intensivist, often with input from an Infectious Disease specialist. Use of corticosteroids and anticoagulants was poorly standardized.

This study was approved by the institutional review board (Inova Health System IRB # U20-05-4061). Continuous and categorical variables were presented as the median (IQR) and n (%), respectively, with the exception of length of stay data which was presented as the mean value. The Mann-Whitney U test, Chi-squared, or Fischer’s exact test were used to compare differences between survivors and non-survivors where appropriate. A p value < 0.05 was considered statistically significant. Univariate and multivariate logistic regression analysis of factors potentially associated with mortality were performed. Variables were dropped from the model through use of the likelihood ratio test. All statistical analyses were performed using STATA version 12 (StataCorp LP; College Station, TX, USA).

Results

A total of 1023 COVID-19 positive patients were admitted in our health system during the study period. Exact numbers of patients admitted to ICU beds could not be discerned, as our health system adapted to a contingency status where critically ill patients were managed in both ICU and intermediate care beds. A total of 164 COVID-19 positive patients in our health system required invasive mechanical ventilation during the study period, representing 16.0% of admitted COVID-19 patients. Ninety-four (57.3%) of patients survived to hospital discharge. Table 1 describes the baseline demographics of the IMV patients. The most notable statistically significant demographic difference between the deceased and survivor groups (defined as alive at the time of data censoring) was age, with median ages of 67 vs. 55, respectively. Table 1 provides laboratory and ventilator data on the cohort. The only observed laboratory differences between deceased and survivor groups were a higher initial d-dimer (2.22 vs. 1.31, p = 0.005) and peak ferritin levels (2998 vs. 2077, p = 0.016). No significant difference was found in peak d-dimer, initial ferritin, or initial or peak CRP and WBC. The entire cohort had severe hypoxemic failure with 51.8% having a PaO2/Fio2 ratio < 100 and 86% with a PaO2/FiO2 ratio <200. The deceased cohort had a lower nadir PaO2/FiO2 ratio at 85.5 compared to 115.6 (p = 0.019) for the survivor cohort.

Table 1

Baseline demographics of COVID-19 patients managed with invasive mechanical ventilation stratified by survivor status.

	Total	Non-Survivors	Survivors	p value*
	(N = 164)	(N = 70)	(N = 94)
Male Gender (%)	107 (65%)	43 (40%)	64 (60%)	0.38
Age, median(IQR)	58 (18),	67 (22),	55 (14),	<0.0001
	Range: 16–91	Range: 29–91	Range: 16–84
Race (N = 161)	White 57 (35.4%)	White 36 (51.4%)	White 21(23.1%)	<0.0001
	Black 42 (26.1%)	Black 21 (30.0%)	Black 21 (23.1%)
	Asian 23 (14.3%)	Asian 8 (11.4%)	Asian 15 (16.5%)
	Other 39 (24.2%)	Other 5 (7.1%)	Other 34 (37.4%)
Hispanic Ethnicity	59 (36%)	18 (26%)	41 (43%)	0.018
BMI, median (IQR)	30 (10.4)	30 (11.8)	30 (9.5)	0.8315
	Range: 13–56	Range: 13–56	Range: 18–53
Comorbidities (%)
HTN	85 (52%)	40 (57.1%)	45 (47.9%)	0.24
HLD	47 (28.7%)	21 (30.0%)	26 (27.7%)	0.74
DM	56 (34.1%)	24 (34.3%)	32 (34.0%)	0.97
CAD	11 (6.7%)	7 (10.0%)	4 (4.3%)	0.15
ESRD	5 (3.1%)	4 (5.7%)	1 (1.1%)	0.09
Immunosuppressed	(%)	(%)	(%)
Obesity (BMI ≥ 30)	(%)	(%)	(%)
Morbid Obesity (BMI ≥ 35)	(%)	(%)	(%)
WBC (X 109/L)
Initial	7.44 (3.81)	7.87 (4.73)	7.33 (3.48)	0.49
Peak	16.36 (8.24)	17.9 (12.14)	15.6 (6.77)	0.07
CRP (mg/dL)
Initial	14.1 (13.8)	14 (14.65)	14.1 (14.1)	0.57
Peak	27.3 (14.6)	26.4 (16.7)	28.2 (12.5)	0.28
D-Dimer (ng/mL)
Initial	1.51 (2.35)	2.22 (3.8)	1.31 (1.7)	0.0053
Peak	6.82 (10.58)	7.11 (11.9)	5.96 (10.2)	0.22
Ferritin (ng/mL)
Initial	1106 (1913)	1141 (1920)	1033 (1797)	0.86
Peak	2456 (3904)	2998 (8145)	2077 (3008)	0.0158
Lowest PaO2/FiO2ratio	98.3 (89.1)	85.5 (67.5)	115.6 (84.6)	0.019
≤ 100	n = 85,51.8%	n = 45,64.2%	n = 40,42.6%
101–200	n = 56,34.1%	n = 16,22., %	n = 40,42.6%
201–300	n = 13,7.9%	n = 3, 4.3%	n = 10,10.6%
>300	n = 10,6.1%	n = 6,8.6%	n = 4,4.3%
Maximum PEEP	12 (5)	13 (6)	12 (4)	0.26
	Range: 5–28	Range: 5–24	Range: 5–28

Abbreviations: BMI = Body mass index; IQR = Interquartile Range; HTN = Hypertension; HLD = Hyperlipidemia; DM = Diabetes mellitus; CAD = Coronary Artery Disease; CKD = Chronic kidney Disease WBC = White blood cell count, CRP = C-reactive protein, PEEP = Positive end-expiratory pressure.

* Data shown as median (IQR) or percentage (%).

Comparisons between survivors and non-survivors based on Wilcoxon Rank-sum (Mann-Whitney) test for continuous variables and Chi-square test for categorical variables.

The average time from admission to intubation was 2.5 days (± 3.0 SD) (Range: 0–18 days); however, 43 patients (26%) were intubated on the day of admission. There was no significant difference in the mean time to intubation between the deceased patients and survivors (2.4 vs. 2.7 days, p = 0.54). The mean duration of ventilator support for survivors was 14.6 days (± 12 SD) (Range: 1–59 days). The mean length of stay for patients discharged alive was 24.5 days (± 14.8 SD) (Range: 7–86 days). The ventilator and hospital LOS for deceased patients were 9.3 (± 6.95 SD) and 11 (± 10.04 SD) days respectively. For those who died, the cause of death was hypoxemic respiratory failure in the majority of patients (n = 56, 80%). Other causes of death included shock (n = 9, 12.8%), cerebrovascular accident (n = 2, 2.8%), bowel ischemia (n = 1, 1.4%), subarachnoid hemorrhage (n = 1, 1.4%), and complications of ECMO cannulation (n = 1, 1.4%).

A total of 16 patients in this cohort were treated with both ECMO and IMV, representing 9.7% of the total cohort. 81.25% of the patients on ECMO survived, as compared to 54.7% of those managed with IMV alone. Table 2 summarizes the outcomes for included patients.

Table 2

Patient outcomes.

	Deceased	Survivor*
Total Cohort (n = 164)	70 (42.7%)	94 (57.3%)
IMV only (n = 148)	67 (45.2%)	81(54.7%)
ECMO only (n = 16)	3 (18.8%)	13 (81.3%)

Abbreviations: ECMO = Extracorporeal membrane oxygenation; IMV = Invasive mechanical ventilation.

Table 3 displays the age distribution for deceased patients and survivors. Patients over 70 accounted for over one third of the deaths in the cohort. In fact, over 80% of patients over age 70 died. In the multivariate analysis, the odds ratio of death was 1.07 for age meaning that for every one point increase in age, there was a seven percent increase in the odds of death. No differences were seen with regard to gender, assessed comorbidities, or BMI. White race was found to be associated with deceased status when compared to other races. A substantial portion of the overall cohort reported Hispanic ethnicity (36%); Hispanic ethnicity appeared to be associated with high likelihood of survivor status in comparison to non-Hispanic ethnicity, although that association did not remain after multivariate analysis, likely due to the relationship between race and ethnicity.

Table 3

Age distribution of cohort stratified by survivor status.

Age	Total	Deceased	Survivors*
≤ 40	14 (8.5%)	3 (4.3%)	11 (11.7%)
41–50	31 (18.9%)	7 (10%)	24 (25.5%)
51–60	49 (29.8%)	18 (25.7%)	31 (33.0%)
61–70	38 (23.2%)	15 (21.4%)	23 (24.4%)
> 70	32 (19.5%)	27 (38.6%)	5 (5.3%)

*Survivors were patients alive at the time of the analysis

Fifty-two (55.3%) of the survivors were managed at a tertiary care center, while only 27 (38.6%) deceased patients were cared for at the same tertiary care center. Patients managed at a tertiary care center were statistically more likely to be survivors (p = 0.034). Patients managed at the tertiary care center had access to both extracorporeal membrane oxygenation (ECMO) and clinical trial enrollment. Table 4 summarizes various treatments provided to the cohort of patients. Survivors were more likely to receive tocilizumab or to be on ECMO, although the survival advantage of ECMO use did not persist after adjustment for multiple variables in the logistic regression analysis as it was correlated with access to tertiary care. Table 5 summarizes the findings of the univariate and multivariate analysis. Treatment with tocilizumab was associated with improved survival with an adjusted odd’s ratio of 0.42 (p = 0.45). There was a trend toward increased need for CRRT in deceased patients (35.7% versus 18.1%, p = 0.07). Nearly all patients (n = 152, 92.1%) received antimicrobials for some duration of time. Only 29 patients (17.6%) had confirmed, culture positive secondary infections, 11 (17.4%) in deceased patients and 18 (17.6%) in surviving patients. One patient developed Clostridium difficle infection.

Table 4

Comparison of treatments received by deceased vs. survivor cohorts.

	Total	Deceased	Survivors*	p value
	(N = 164)	(N = 70)	(N = 94)
Care at Tertiary Center	79 (48.2%)	27 (38.6%)	52 (55.3%)	0.034
Antimicrobials	152 (92.7%)	66 (94.3%)	86 (91.5%)	0.50
Hydroxychloroquine	132 (80.5%)	52 (74.3%)	80 (85.1%)	0.08
Toculizumab	49 (29.9%)	13 (18.6%)	36 (38.3%)	0.006
Clinical Trial	15 (9.2%)	4 (5.7%)	11 (11.7%)	0.19
Inhaled Pulmonary Vasodilators	23 (14.0%)	4 (5.7%)	19 (20.2%)	0.008
Paralysis	61 (37.2%)	25 (35.7%)	36 (38.3%)	0.74
Prone Positioning	100 (61.0%)	44 (62.9%)	56 (59.6%)	0.67
CRRT	42 (25.6%)	25 (35.7%)	17 (18.1%)	0.011
ECMO	16 (9.7%)	3 (4.3%)	13 (13.8%)	0.042

Abbreviations: ECMO = Extracorporeal membrane oxygenation; CRRT = Continuous renal replacement therapy

* Comparisons between survivors and non-survivors based on Chi-square tests

Table 5

Odds ratios of death among mechanically ventilated COVID-19 patients.

	Odds Ratio	P value
	(95% Confidence Interval)
Unadjusted: (N = 164)
Age	1.07 (1.04–1.10)	<0.0001
Male Gender	0.75 (0.39–1.43)	0.38
Non-Caucasian Race (N = 161)	0.47 (0.34–0.64)	<0.0001
Hispanic Ethnicity	0.45 (0.23–0.88)	0.0170
Tertiary Care Hospital	0.51 (0.27–0.95)	0.0332
Enrollment in Clinical Trial	0.46 (0.14–1.50)	0.18
ECMO	0.28 (0.08–1.02)	0.03
Tocilizumab	0.37 (0.18–0.76)	0.0055
Hydroxychloroquine	0.51 (0.23–1.10)	0.09
*Adjusted: (N = 161)
Age	1.06 (1.03–1.10)	<0.0001
Non-Caucasian Race	0.54 (0.38–0.77)	0.001
Tertiary Care Hospital	0.44 (0.20–0.94)	0.034
Tocilizumab	0.42 (0.18–0.98)	0.045

Discussion

The primary finding of our analysis is that mortality in COVID-19 patients requiring mechanical ventilation is high, particularly in patients of advanced age. Our study adds to the available literature on outcomes. Our data has the advantage of following all included patients to either death or hospital discharge, whereas most existing studies include patients which are still hospitalized.

Despite the high mortality, the outcomes of mechanically ventilated patients in our health system compare favorably to those reported elsewhere. For instance, in the report by Richardson, et. al. on the Northwell Health System in New York, 1151 patients required IMV. At the time of their report, 24.5% of the patients had died, while only 3.3% were discharged alive, and 72% remained in the hospital [2]. If only those with a confirmed endpoint (death or discharge) from this cohort are analyzed, the reported mortality rate for patients requiring IMV is 88.1% [2]. Data from Wuhan, China reported by Zhou and colleagues found that 31 out of 32 patients (96.8%) treated with IMV died [3]. ICNARC, the Intensive Care National Audit and Research Centre from the United Kingdom, reported that 56.8% of patients treated with “advanced respiratory support”, which can include high flow oxygen, non-invasive ventilation, ECMO or IMV, died in the hospital [7]. Graselli, et. al. reported on 1591 patients hospitalized in the ICU in the Lombardy Region of Italy [5]. They do not specifically report mortality for those managed with IMV, although the majority (72%) required IMV. At the time of data reporting, their ICU mortality rate was 26%, although 58% of the patients were still in the ICU. Finally, a report from Seattle, Washington, USA included data on 24 critically ill COVID-19 patients, 18 of whom required IMV [4]. At the time of data censoring, 50% of the patients died and 5 of 18 (27.7%) were still on mechanical ventilation.

We feel it is particularly reassuring that the death rate in our cohort was not higher, given our system strategy of avoiding intubation unless patients were truly unable to maintain their oxygenation or ventilation despite aggressive management with non-invasive measures. We managed patients with self-proning, inhaled pulmonary vasodilators, and high flow oxygen to avoid intubation whenever possible. Given this approach, the cohort of patients managed with IMV was likely sicker than those reported in some of series and could be expected to have less favorable outcomes. The extreme nature of the severity of illness in our cohort is supported by the median lowest PaO2/FiO2 ratio of 98.3 and the fact that > 85% of the patients had a PaO2/FiO2 ratio < 200.

Patients of advanced age account for the majority of deaths in our cohort. Of 32 patients age 70 of older, 84.3% were deceased at the time the data was censored. However, only 32.6% of patients < 70 years old and 22% of those < age 50 died in the hospital. Indeed, for every increasing year of age in our cohort, there was a 7% increase in the odds of death based on our multivariate logistic regression model. This relationship between advancing age and odds of mortality is consistent with other reports. There are likely multiple reasons for this including more co-morbidities, worse baseline functional status, and variations in the aggressiveness of goals of care. Notably, patients treated with tocilizumab were half as likely to die, although we suspect this result is secondary to inadequate variable control as the recent randomized, double-blinded, placebo controlled COVACTA trial failed to demonstrate a mortality benefit in COVID positive patients treated with tocilizumab [10]. A large number of patients within our cohort were offered adjunct treatments via clinical trials and/or ECMO. While the small number of patients within this cohort does not allow us to make assertions which of these treatment strategies provided the most benefit, the overall survival within our cohort likely supports the notion that availability of advanced interventions at a tertiary care center may play a significant role in improving patient outcomes.

Our study does have some limitations which should be acknowledged.

One limitation of our study is that it may not be generalizable to all health systems. Our health care system has a well-resourced, well-structured and dedicated medical critical care service with high volumes and experience treating ARDS, and hence adept at the application of best practices including proning and lung protective strategies. Additionally, we have a robust and experienced, high volume ECMO program at our tertiary care hospital which bolstered our outcomes. Also, our approach to treatment of hypoxemic respiratory failure prior to intubation may differ from those of other hospitals. Given our aggressive approach to use of noninvasive strategies including self-proning and HFNC, it is possible that our cohort of patients requiring IMV may in fact be sicker than those reported elsewhere. Another issue is that we were unable to provide specific data on use of HFNC or NIV prior to intubation. As mentioned in the methods, our center evolved to a strategy of delayed extubation and aggressive HFNC support relatively early in our surge, although our pre-intubation management strategy was somewhat in flux over time. Finally, since patients frequently had evolving physiology with variable lung compliance and ventilator settings, we are unable to provide specific details on these parameters. However, our ventilator management was in keeping with current best practices and therefore we believe that our survival estimates would be reproducible in similar health care systems.

In conclusion, the need for IMV in COVID-19 is associated with a high mortality in patients with COVID-19. However, successful outcomes are possible, with over 70% of patients younger than 70 still alive at the time of data censoring.

(XLSX)

Click here for additional data file.

18 Aug 2020

PONE-D-20-21719

Outcomes of Mechanically Ventilated Patients with COVID-19 Associated Respiratory Failure

PLOS ONE

Dear Dr. King,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 02 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Andrea Ballotta

Academic Editor

PLOS ONE

Additional Editor Comments:

Thanks for your submission. On the basis of the reviewer's comments i deem the paper not suitable for publication. It needs to be strongly revised. We don't need other partial data. WE need data about medium and long term outcome. The COVID 19 period has also been a sort of nightmare for science. WE must do much better.

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In the ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study.

Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent.

If patients provided informed written consent to have data from their medical records used in research, please include this information.

3. For studies involving humans categorized by race/ethnicity, age, disease/disabilities, religion, sex/gender, sexual orientation, or other socially constructed groupings, authors should:

a) Explicitly describe their methods of categorizing human populations,

b) Define categories in as much detail as the study protocol allows,

c) Justify their choices of definitions and categories,

d) Explain whether (and if so, how) they controlled for confounding variables such as socioeconomic status, nutrition, environmental exposures, or similar factors in their analysis, and

e) Update outmoded terms and potentially stigmatizing labels to more current, acceptable terminology.

Examples: “Caucasian” should be changed to “white” or “of [Western] European descent” (as appropriate).

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear authors,

Thanks for your manuscript. I strongly belive in your message but unfortunately you referred to partial data, particularly about the patient's outcome. My suggestion is to obtain all patient's outcome, complications and ventilation management data and to revised the manuscript results.

Reviewer #2: Thanks for your paper.

To be precise in your analysis you might put the following details in the paper.

I wuold be important to indicate how much time patients has been treated before intubation ( sintoms, NIV and High flow cannula time).

Which was the cutoff for intubation, it was differnt over the time ?

When you indicate the ventilation it has been inappropriate only define the ammount of on ml/kg and plateu.

Is more convenient to indicate at least the driving pressure or better the compliance oder a partition with esophageal baloon.

Which were the indication for ECMo? Murray score Oxigenation index....? And which was the ECMO capability in the number of the admission in ICU .

Finally the big limitation you have also underline is the lack of long term survival which is important to evaluate the treatment.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Mirko Belliato

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

8 Sep 2020

To the Editor,

Thanks to both you and your reviewers for your thoughtful and favorable review of our manuscript. We are pleased to provide a revised version for subsequent review. With this revision we will submit a de-identified version of our dataset as requested. We have attempted to answer the reviewers’ comments below. We look forward to your review.

Respectfully,

Christopher King, MD

Reviewer #1: Dear authors,

Thanks for your manuscript. I strongly believe in your message but unfortunately you referred to partial data, particularly about the patient's outcome. My suggestion is to obtain all patient's outcome, complications and ventilation management data and to revised the manuscript results.

Thank you for your kind words with regards to our manuscript. Initially we hoped to get this manuscript out early to provide information to clinicians with regards to outcomes in intubated COVID-19 patients, but as you point out, a complete data set including all outcomes is more informative. We have now updated all patient outcomes so that all included patients have a clinical outcome of either death or discharge from the hospital. All statistics and tables have been updated in accordance with this change as well. To our knowledge this makes our manuscript the first to report definitive outcomes on an entire cohort of patients with COVID-19 requiring invasive mechanical ventilation.

Reviewer #2: Thanks for your paper.

To be precise in your analysis you might put the following details in the paper.

I wuold be important to indicate how much time patients has been treated before intubation ( sintoms, NIV and High flow cannula time).

We include in the manuscript the time from admission to intubation for all included patients. This is provided in the results section where it states:

“The average time from admission to intubation was 2.5 days (± 3.0 SD) (Range: 0-18 days); however, 43 patients (26%) were intubated on the day of admission. There was no significant difference in the mean time to intubation between the deceased patients and survivors (2.4 vs. 2.7 days, p = 0.54).”

We are unable to provide data on time from symptom onset to intubation. Unfortunately, we are also limited in our ability to report specific data on use of HFNC and NIV. We do mention in the methods section that use of NIV was relatively uncommon. We have added a statement in the limitations section to address this. See the next comment for what was added.

Which was the cutoff for intubation, it was differnt over the time ?

Very early on we had a low threshold for intubation but quickly evolved to a strategy of delayed intubation and reliance on HFNC. The vast majority of patients would have been managed with a greater reliance on HFNC. We added the following to the limitations section “Another issue is that we were unable to provide specific data on use of HFNC or NIV prior to intubation. As mentioned in the methods, our center evolved to a strategy of delayed extubation and aggressive HFNC support relatively early in our surge, although our pre-intubation management strategy was somewhat in flux over time.”

When you indicate the ventilation it has been inappropriate only define the ammount of on ml/kg and plateu.

Is more convenient to indicate at least the driving pressure or better the compliance oder a partition with esophageal baloon.

Patients often had multiple changes in their ventilator settings as their clinical status progressed and had evolving lung compliance over time. Given this we found it difficult to report specifics on ventilator or lung compliance in a manner that we thought would be instructive to providers. We do believe that our ventilator management across the system was consistent with current best practices and would be similar to standard of care ventilator management in similarly structured hospital systems. Our key message in writing the manuscript is to provide information on outcomes of mechanically ventilated patients rather than to attempt to provide guidance on specific ventilator management strategies. We have added a section to the limitations portion of the paper to highlight these issues. It reads “Finally, since patients frequently had evolving physiology with variable lung compliance and ventilator settings, we are unable to provide specific details on these parameters. However, our ventilator management was in keeping with current best practices and therefore we believe that our survival estimates would be reproducible in similar health care systems.” No patients had esophageal balloons placed given concerns for staff exposure to COVID-19.

Which were the indication for ECMo? Murray score Oxigenation index....? And which was the ECMO capability in the number of the admission in ICU .

Our indications for ECMO initiation were in line with those recommended by ELSO. As stated in the text “Patients were considered for venovenous (VV) ECMO if age < 60 years old, on IMV < 10 days, had a P/F ratio < 100 and/or failed lung protective ventilation, despite neuromuscular blockade and prone positioning, or had recalcitrant hypercapnic acidosis affecting perfusion.” Like many other clinical decisions, ECMO was initiated only after a formal consultation by an experience ECMO intensivist and discussion with the cannulating cardiac surgeon. We were not reliant on a single parameter but did require all patients to fail conventional lung protective ventilation, neuromuscular blockade and proning. We expanded our typical ECMO capacity to the point that we were staffing up to 10 ECMO patients simultaneously. This is an increase from our typical maximum capacity of approximately 6 patients.

Finally the big limitation you have also underline is the lack of long term survival which is important to evaluate the treatment.

Thank you for raising this important point. As mentioned above, we have updated outcomes for all patients included in the study. This makes our manuscript unique as it is one of the only studies providing complete outcomes data for critically ill COVID-19 patients.

9 Nov 2020

Outcomes of Mechanically Ventilated Patients with COVID-19 Associated Respiratory Failure

PONE-D-20-21719R1

Dear Dr. King,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Chiara Lazzeri

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: Thanks

Can you please put in the paper ECMO cannulation parameters you have indicated on the answer ?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

12 Nov 2020

PONE-D-20-21719R1

Outcomes of Mechanically Ventilated Patients with COVID-19 Associated Respiratory Failure

Dear Dr. King:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Chiara Lazzeri

Academic Editor

PLOS ONE