Effect of Awake Prone Positioning on ROX Index in Critically-ill Patients With Respiratory Failure due to COVID-19

Abstract

COVID-19 pneumonia causes acute respiratory distress syndrome (ARDS). Prone positioning (PP) is beneficial to pulmonary physiology and improves oxygenation in patients with ARDS. We aimed to investigate the effect of the PP on oxygenation, respiratory rate (RR) and ROX index in non-intubated patients with COVID-19 associated respiratory failure and to determine whether ROX index predicts intubation. Awake critically-ill patients with confirmed diagnosis of COVID-19 who underwent PP were enrolled in the retrospective, single-center study. Oxygenation parameters were recorded 1 h before PP, during PP and 1 h after return to supine position (after PP). Intubation was defined as the endpoint. Seventy-one patients with a median age of 64 [55-73] years were enrolled in the study. PaO₂/FiO₂ and SpO₂/FiO₂ improved during PP, this improvement did not persist after PP. RR improved during and after PP in both intubated and non-intubated patients (for all P < .001). ROX index improved only in non-intubated patients (P < .001) but not in intubated patients (P = .07). Area under the curve (AUC) of ROX index for intubation before PP, during PP and after PP were 0.74 [0.61-0.88] (P = .002), 0.76 [0.62-0.91] (P = .001), and 0.76 [0.64-0.89] (P = .001), respectively. ROX index >6.83 before PP had a negative predictive value (NPV) of 0.85; ROX index >8.28 during PP had a NPV of 0.88 and ROX index >7.48 after PP had a NPV of 0.85. In logistic regression adjusted for APACHE II score, ROX index ≤6.83 before PP had an odds ratio (OR) 4.47 [1.39-14.38], ROX index ≤8.28 during PP had an OR 7.96 [2.29-27.64] and ROX index ≤7.48 had an OR 3.98 [1.25-12.61] for prediction of intubation. In conclusion, awake PP improves oxygenation and decreases RR. ROX index improved only in non- intubated patients and a higher ROX index predicts lower risk of progressing to mechanical ventilation with intubation.

Keywords

acute respiratory distress syndrome pneumonia acute hypoxemic respiratory failure intensive care severe acute respiratory syndrome corona virus SARS-CoV-2 pandemic oxygen saturation respiratory rate

Introduction

Severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) infection causing coronavirus disease-2019 (COVID-19), dramatically increased the number of hospital admission of patients with acute respiratory failure (ARF) and acute respiratory distress syndrome (ARDS).¹ If these patients need intensive care unit (ICU) admission and require invasive mechanical ventilation (IMV), mortality might hit up to 60%.^2,3

Prone positioning has been shown to improve oxygenation and respiratory mechanics by promoting lung homogeneity and to reduce lung injury and mortality in the intubated patients with moderate to severe ARDS.^4,5 Notably, after the COVID-19 pandemic, interest has been raised for awake PP to reduce the need for IMV.^6-9 The utilization of PP in spontaneously breathing, non-intubated patients treated with oxygen therapy or non-invasive ventilation (NIV) have been reported to improve oxygenation and to decrease respiratory effort.¹⁰ However, predictors for mechanical ventilation with intubation are still scarce in patients who receive PP.^11,12

Respiratory rate-oxygenation (ROX) index, that is, the ratio of oxygen saturation with a pulse oximeter to the fraction of inspired oxygen [SPO₂/FiO₂] divided by the respiratory rate (RR) has been shown to be a predictor for intubation in patients with ARF due to pneumonia treated with high flow nasal oxygen (HFNO).¹³ Few studies during the COVID-19 pandemic have been reported to assess the predictive accuracy of the ROX index for HFNO failure^14-16 or NIV failure.¹⁷ However, the predictive value of ROX index during PP has not been investigated extensively.

This study aimed to investigate the effect of PP on oxygenation, RR and ROX index in non-intubated patients with COVID-19 associated ARF admitted to ICU and to determine predictors for intubation.

Methods

Study Design and Patients

This study is a retrospective, single-center, observational study conducted between March 21^th and October 31^st, 2020 in a tertiary academic hospital ICU. Laboratory-confirmed (all cases in COVID-19 patients had positive results for polymerase chain reaction (PCR) for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)) adult critically-ill awake and spontaneously breathing, non-intubated COVID-19 patients who were admitted to ICU due to ARF and received PP during the ICU stay were included. The study protocol was approved by the University's Ethics Committee (GO: 20-1013, NOV 3 2020).

Variables and Definitions

The data were recorded from electronic and printed medical records.

Patient demographics such as age, sex, comorbidities (hypertension, diabetes mellitus, chronic cardiac and lung diseases, malignancy, chronic renal and liver diseases) were recorded. In addition, acute physiology age and chronic health evaluation II (APACHE II), and sequential organ failure assessment (SOFA) scores during admission, duration of the first PP, use of conventional oxygen treatment, HFNO and NIV during ICU stay, need for endotracheal intubation, ICU length of stay, ICU mortality and 28th-day mortality were also recorded.

The vital signs and physiological parameters (FiO₂, partial pressure of arterial oxygen (PaO₂), SpO₂, FiO₂, PaO₂/FiO₂ ratio, SpO₂/FiO₂ ratio, RR and ROX index) were recorded for three different time points of the first PP because of evaluating the effect of PP on these parameters as 1 h before PP, during PP and 1 h after return to the supine position.

Estimated PaO₂ from SpO₂ was used if an arterial blood gas sample was lacking and estimated FiO₂ from oxygen flow rate was calculated as described elsewhere.¹⁸

Patients who had ≥10% increase in PaO₂/FiO₂ ratio and ≥10% decrease in RR were accepted as responders both before versus during and before versus after. The others were considered as non-responders. Intubation was defined as the end point.

Statistical Analysis

Continuous variables were presented as median [25th-75th percentile] and categorical variables were presented as frequencies and percentages. Mann–Whitney U test was used to compare the continuous variables and x² and Fisher exact test were used to compare categorical variables. Comparisons were made between intubated and non-intubated patients. One-way analysis of variance (ANOVA) for repeated measurement with a post hoc Tukey's correction and Friedman test were used for multiple comparisons were appropriate. Logistic regression analysis was performed in multivariate analysis to determine predictors of intubation and the results were expressed as OR and 95% confidence interval (CI). Receiving operator characteristic (ROC) curves and area under ROC (AUROC) with sensitivities and specificities were performed to measure the effect of RR, ROX and PaO₂/FiO₂ ratio for intubation. Youden index used while finding cut off values. Statistical analyses were performed with SPSS 23.0 (Chicago, IL) and R 3.2. A P-value of less than .05 was considered as statistical significance.

Results

During the study period, 275 patients were diagnosed as COVID-19 associated ARF, and 71 non-intubated patients underwent PP. Of them 19 (26.8%) patients were required intubation during the study period despite PP. Patient characteristics and comparison of patients as non-intubated (patients not required intubation during study period) and required intubation are summarized in Table 1. The median age of the patients was 64 [55-73] years and 67.6% of them were male. The APACHE II and SOFA scores on ICU admission were 14 [11-16] and 3 [3-4], respectively. Patients were admitted to the hospital within 5 [2-7] days and to the ICU within 10 [7-12] days after the onset of symptoms. Estimated PaO₂ from SpO₂ was used at 38 patients before PP, 48 patients during PP and 58 patients after PP. Median PaO₂/FiO₂ ratio on ICU admission was 155 [122-207] mmHg. During the ICU stay, 40 (56.3%) patients were treated with HFNO, 15 of whom during the first PP; 49 (69%) patients were treated with NIV, 5 of whom during the first PP. There were significant changes for PaO₂/FiO₂, SpO₂/FiO₂ and ROX index in all-time points (before, during, after PP), for both non-intubated patients and patients required intubation. The decrease in RR was found to be higher in the non-intubated group. The intubation rate was lower in patients whose RR decreased ≥10% during PP (P = .09). ICU length of stay was 12 [8-19] days and 28^th day mortality was 28.2%. The ICU mortality and 28-day mortality were higher in the patients required intubation group (P < .001).

Table 1.

Characteristics of the Patients and Comparison of Non-Intubated Patients and Patients Required Intubation Despite PP.

	All	Non-intubated n = 52	Required intubation n = 19	P
Age, years	64 [55-73]	63.5 [55-73]	67 [60-74]	.41
Male sex^a	48 (67.6)	34 (70.8)	14 (29.2)	.35
Comorbidities^a
Hypertension	43 (60.6)	32 (61.5)	11 (57.9)	.49
Diabetes mellitus	29 (40.8)	22 (42.3)	7 (36.8)	.44
Chronic cardiac disease	23 (32.4)	16 (30.8)	7 (36.8)	.41
Chronic lung disease	14 (19.7)	8 (15.4)	6 (31.6)	.12
Malignancy	11 (15.5)	7 (13.5)	4 (21.1)	.32
Chronic renal disease	4 (5.6)	3 (5.8)	1 (5.3)	.71
Chronic liver disease	3 (4.2)	2 (3.8)	1 (5.3)	.32
APACHE II score	14 [11-16]	12.5 [10-16]	15 [13-18]	.02
Admission SOFA score	3 [3-4]	3 [3-4]	4 [3-5]	.29
Symptom onset to hospital admission, days	5 [2-7]	5 [2.2-9.2]	4 [2-7]	.37
Symptom onset to ICU admission, days	10 [7-12]	10 [7-12]	9 [5-12]	.60
Admission PaO₂/FiO₂, mmHg	155 [122.0-207.0]	165 [130.5-210.7]	142 [86.0-161.0]	.04
Duration of first PP, hours	3 [2-4]	3 [2-4]	2 [2-5]	.76
ICU admission to PP, days	2.1 [1-3]	1 [1-2.75]	1 [1-3]	.89
PaO₂/FiO₂
Before PP	142 [96-176.5]	150 [97-200]	100 [71-145]	.002
During PP	166 [130-221.5]	200 [146-239]	132 [105-150]	<.001
After PP	145 [94.5-168]	161 [100-186]	100 [77-145]	.003
SaO₂/FiO₂
Before PP	204 [146-251]	218 [150.75-260.25]	130 [111-213]	.002
During PP	215 [152-261]	230 [188.5-262.5]	172 [116-227]	.002
After PP	209 [144.5-250	226 [154.75-254.75]	166 [124-213]	.003
RR
Before PP	28 [25.5-30]	27 [25-30]	29 [26-34]	.13
During PP	22 [20-26.5]	22 [20-26]	26 [22-28]	.03
After PP	26 [22-28]	24 [20-27]	27 [24-30]	.01
ROX index
Before PP	7.5 [5.2-9.2]	7.9 [5.7-9.4]	5.6 [3.5-7.5]	.02
During PP	9.4 [6.9-11.5]	9.7 [7.9-11.8]	6.1 [4.6-8.4]	.02
After PP	8.3 [5.6-9.3	8.4 [6.9-11.6]	5.3 [4.5-8.2]	.04
Oxygen therapy conventional^a	51 (71.8)	40 (76.9)	11 (57.9)	.05
HFNO^a	15 (21.1)	10 (19.2)	5 (26.3)	.01
NIV^a	5 (7.0)	2 (3.8)	3 (15.9)	.92
Responder (after vs before PP)^a
PaO₂/FiO₂	47 (66.2)	35 (37.3)	12 (63.2)	.41
RR	37 (52.1)	25 (48.1)	12 (63.2)	.19
Responder (during vs before PP)^a
PaO₂/FiO₂	21 (29.6)	14 (26.9)	7 (36.8)	.35
RR	23 (32.4)	14 (26.9)	9 (47.4)	.09
ICU length of stay, days	12 [8-19]	11.5 [7-17]	17 [12-22]	.04
28th day mortality^a	16 (22.5)	3 (17.6)	13 (68.4)	<.001
ICU mortality^a	20 (28.2)	4 (20.0)	16 (84.2)	<.001

n (%), others as median [25th-75th percentile].

Abbreviations: APACHE, acute physiology and chronic health evaluation; SOFA, sequential organ failure assessment; ICU, intensive care unit; PaO₂, partial pressure of arterial oxygen; FiO₂, fraction of inspired oxygen; PP, prone positioning; SpO₂, saturation of oxygen with pulse-oximeter; RR, respiratory rate; ROX index, (SpO₂/FiO₂)/RR; HFNO, high flow nasal oxygen; NIV, non-invasive ventilation; Responder, ≥10% increase in PaO₂/FiO₂ ratio and ≥10% decrease in RR.

The impact of PP on PaO₂/FiO₂, SpO₂/FiO₂, RR, and ROX index in non-intubated and required intubation patients is shown in Figure 1. PaO₂/FiO₂ and SpO₂/FiO₂ improved during PP (for required intubation patients P < .001, P = .006, respectively; for non-intubated patients P < .001 for all), however, this improvement did not persist after PP (for patients required intubation P = .87, P = .30, respectively; for non-intubated patients P = .87, P = .23, respectively). RR improved during and after PP in both patients required intubation and non-intubated patients (for all P < .001). Whereas, ROX index improved only in non-intubated patients (P < .001) but not in patients required intubation during study period (P = .07).

In logistic regression models adjusted for APACHE II score, ROX index ≤ 6.83 before PP had an odds ratio (OR) of 4.47 [1.39-14.38], ROX index ≤8.28 during PP had an OR of 7.96 [2.29-27.64] and ROX index ≤7.48 had an OR of 3.98 [1.25-12.61] for prediction of intubation (Table 2).

Table 2.

Logistic Regression Analysis Adjusted by APACHE II for Prediction of Intubation.

	OR	95% CI	P
A
APACHE II	1.11	0.99-1.24	.060
ROX before PP < 6.83	4.47	1.39-14.38	.012
B
APACHE II	1.12	0.99-1.26	.061
ROX during PP < 8.28	7.96	2.29-27.64	.001
C
APACHE II	1.09	0.98-1.22	.101
ROX after PP < 7.48	3.98	1.25-12.61	.019

Models adjusted for APACHE II score, ROX index ≤6.83 before PP (A), ROX index ≤8.28 during PP (B) and ROX index ≤7.48 (C) for prediction of intubation.

ROC curves for ROX index before PP, during PP and after PP as predictors of intubation are seen in Figure 2. Area under the curve (AUC) [95% CI] of ROX index before PP, ROX index during PP and ROX index after PP were 0.74 [0.61-0.88] (P = .002), 0.76 [0.62-0.91] (P = .001), and 0.76 [0.64-0.89] (P = .001), respectively, for intubation. ROX index > 6.83 before PP had a negative predictive value (NPV) of 0.85; ROX index >8.28 during PP had a NPV of 0.88 and ROX index >7.48 after PP had a NPV of 0.85. ROX index cut off value of 8.28 during PP had the highest sensitivity and specificity for prediction of intubation (0.74 and 0.73, respectively).

Discussion

In the current study, PP in awake, spontaneously breathing, non-intubated patients with severe hypoxemic ARF due to COVID-19 was found to be associated with improvement in oxygenation during PP, although the effect in oxygenation was lost after reverting to supine position. RR improved both in non-intubated patients and patients required intubation during and after PP. However, ROX index improved only in non-intubated patients and ROX index was a predictor of intubation with a high NPV. Therefore, a higher ROX index predicts lower risk of progressing to mechanical ventilation with intubation.

A few studies assessed the impact of awake PP in patients with severe hypoxemia and found positive effect on oxygenation and respiratory mechanics before the COVID-19 pandemic.^8-10 In COVID-19 era, several observational and randomized controlled studies have revealed that awake PP avoided intubation especially in patients who received advanced respiratory support such as HFNO, NIV and in the ICU setting.^19,20

However, the improvement in oxygenation was usually temporary, reversing after getting back to supine position in most of the patients.^21,22 Similar to these studies the current study revealed that PP might improve oxygenation during PP but this effect reverted after resupination.

Although the improvement in oxygenation did not continue or even deteriorated, in the current study the reduction in RR persisted after PP. However, this effect was observed both in patients required intubation and non-intubated patients making RR a non-reliable parameter for the prediction of intubation. However, ROX index revealed a sustained improvement only in non-intubated patients during and after PP. Therefore, it could be a better index for predicting intubation compared to RR and oxygenation parameters.

Especially during the first wave of the pandemic, mortality in patients who received IMV was quite high.^1,2 IMV carries relatively higher mortality and complication rates compared to non-invasive respiratory support, however, delaying IMV causes excess mortality as well.^23-25 Therefore, timing is of utmost importance and several studies investigated predictors for intubation in COVID-19 patients with ARF.^14,26,27

Awake PP has been utilized more during the pandemic, however, predictors for intubation in patients who received PP are scarce. Lupieri et al²⁷ investigated in 27 patients who underwent awake PP, whether PaO₂/FiO₂, alveolar–arterial gradient, RR and PaCO₂ could predict the need for intubation and revealed that in non-intubated patients PaO₂/FiO₂ increased and, alveolar–arterial gradient decreased as compared to intubated patients. However, hypoxemia by itself could not be considered an indication of intubation. Downing, et al¹² investigated predictors of intubation in COVID-19 patients undergoing awake proning in the emergency department and found ROX index and PaO₂/FiO₂ measured 24 h after admission to be the variables most likely to predict need for intubation. However, almost 30% of their patients had initial PaO₂/FiO₂ >300 reflecting non-critical illness, although 44% of their patients were intubated. This might not only be due to different patient populations but also due to differences in clinicians’ decisions for intubation at the bedside. Moreover, this study reported ROX index at triage and 24 h later. There was no information in this study about the PaO₂/FiO₂ and ROX index before, during or after PP. So, this study could not give information whether variables during peri-proning period were related to intubation. Nevertheless, similar to this study, ROX index has been shown to be a promising variable for the decision of intubation. Cherian et al¹⁵ revealed that among 59 patients, higher ROX index within 4 days of the institution of awake PP may help identify patients who would not need IMV However, they did not demonstrate any cut-off value for ROX index. In a secondary analysis of a randomized controlled trial comparing awake PP with the control group in patients who received HFNO, baseline low RR, an increase in ROX index after the first PP session, a decrease in lung ultrasound score and >8 h/day PP were associated with survival without intubation.²⁸

ROX index has been shown to be a predictor of HFNO failure in acute hypoxemic respiratory failure.^13,14,29 Although Roca et al¹³ revealed that ROX <3.85 was associated with treatment failure while values >4.88 predicted treatment success in ARF due to pneumonia before the pandemic, in a meta-analysis of Junhai et al²⁸ the cut-off values were reported to change from 2.7 to 9.2. In the study by Downing et al¹² ROX index at triage and 24 h later were 9.9 and 4.6, respectively, in intubated patients, whereas in non-intubated patients the corresponding values were 15.9 and 15.7. Vega et al²⁹ revealed a cut-off value of 5.99 at 12 h predicting intubation in patients receiving HFNO outside the ICU setting. In an another study, ROX index at 24 h with a cut-off point of 5.35 predicts HFNC success in patients with SARS-Cov-2-induced ARF.³⁰ Colaianni-Alfonso et al³¹ revealed a cut-off value of 6.64 having the -highest sensitivity and specificity and AUC values for continuous positive airway pressure (CPAP) treatment failure 24 h after CPAP. Therefore, dynamic (rather than single-point static) evaluation of the ROX index could represent an accurate method for treatment failure,³² and different cut-off values could be used in different settings.

This study has some limitations. First, it is difficult to generalize the results as this study is a retrospective single-center study with limited number of patients and no control group. As we have analyzed the effect of the first PP, the optimum duration and the frequency of PP could not be determined. In addition, timing of recording variables during PP could not be standardized because there was not any protocol for duration and timing of PP. As patients had received different adjunctive respiratory support such as conventional O₂, HFNO, and/or NIV, it is difficult to assess the sole effect of PP. There are many more variables which might be related to intubation which were not considered in this study. Measurements of work of breathing are lacking, however these are difficult to use at the bedside especially in the pandemic. Nevertheless, this study is important as there are few studies investigating predictors for intubation during awake PP in severe COVID-19 patients.

Conclusions

In conclusion, improvement in oxygenation during awake PP is transient, improvement in RR does not discriminate patients who need intubation from those who do not. However, in patients who received awake PP, ROX is a promising index for determining patients with lower risk of progressing to mechanical ventilation with intubation.

Footnotes

Authors’ Contributions

EOE analyzed and interpreted the patient, did analysis of data, wrote the manuscript; BE: collected patient data. BH was a major contributor in writing the manuscript. GG analyzed the data; MY collected data; ITG analyzed the data; NYK collected data; AT was a major contributor in writing the manuscript. All authors read and approved the final manuscript.

Availability of Data and Materials

The datasets generated and/or analyzed during the current study are not publicly available It is not appropriate to share because it is hospital and patient data and important for patient privacy. But are available from the corresponding author on reasonable request.

Consent for Publication

Not applicable.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics Approval

The study protocol was approved by the University's Ethics Committee (GO:20-1013, NOV 3 2020).

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Ebru Ortac Ersoy

Burcin Halacli

Mehmet Yildirim

References

Grasselli

Tonetti

Protti

, et al. Pathophysiology of COVID-19-associated acute respiratory distress syndrome: a multicentre prospective observational study. Lancet Respir Med. 2020;8(12):1201-1208.

Gündoğan

Akbudak

İH

Hancı

, et al. Clinical outcomes and independent risk factors for 90-day mortality in critically ill patients with respiratory failure infected with SARS-CoV-2: a multicenter study in Turkish intensive care units. Balkan Med J. 2021;38(5):296-303.

Auld

Caridi-Scheible

Blum

, et al. ICU And ventilator mortality among critically ill adults with coronavirus disease 2019. Crit Care Med. 2020;48(9):e799-e804.

Rampon

Simpson

Agrawal

. Prone positioning for acute hypoxemic respiratory failure and ARDS. Chest. 2023;163(2):332-340.

Sud

Friedrich

Adhikari

NKJ

, et al. Effect of prone positioning during mechanical ventilation on mortality among patients with acute respiratory distress syndrome: a systematic review and meta-analysis. CMAJ. 2014;186(10):E381-E390.

Bamford

Bentley

Dean

Whitmore

Wilson-Baig

. ICS guidance for prone positioning of the conscious COVID patient 2020; 2020 https://emcrit.org“wp-content“uploads”2020/04.

Scaravilli

Grasselli

Castagna

, et al. Prone positioning improves oxygenation in spontaneously breathing nonintubated patients with hypoxemic acute respiratory failure: a retrospective study. J Crit Care. 2015;30(6):1390-1394.

Tay

CHB

. Awake prone positioning for non-intubated oxygen dependent COVID-19 pneumonia patients. Eur Respir J. 2020;56(1):2001198.

Gürün Kaya

Öz

Erol

Çiftçi

Çiledağ

Kaya

. Prone positioning in non-intubated patients with COVID-19. Tuberk Toraks. 2020;68(3):331-336.

10.

Othman

El-Menshawy

Mohamed

. Effects of awake-prone positioning on oxygenation and physiological outcomes in non-intubated patients with COVID-19: a randomized controlled trial. Nurs Crit Care. 2022:10.1111.

11.

Ibarra-Estrada

Pavlov

, et al. Factors for success of awake prone positioning in patients with COVID-19-induced acute hypoxemic respiratory failure: analysis of a randomized controlled trial. Crit Care. 2022;26(1):84.

12.

Downing

Cardona

Alfalasi

, et al. Predictors of intubation in COVID-19 patients undergoing awake proning in the emergency department. Am J Emerg Med. 2021;49(November):276-286.

13.

Roca

Caralt

Messika

, et al. An index combining respiratory rate and oxygenation to predict outcome of nasal high-flow therapy. Am J Respir Crit Care Med. 2019;199(11):1368-1376.

14.

Mellado-Artigas

Mujica

Ruiz

, et al. Predictors of failure with high-flow nasal oxygen therapy in COVID-19 patients with acute respiratory failure: a multicenter observational study. J Intensive Care. 2021;9(1):23.

15.

Cherian

Roche

, et al. Predictive factors for success of awake proning in hypoxemic respiratory failure secondary to COVID-19: a retrospective cohort study. Respir Med. 2021;181(May):106379.

16.

Zucman

Mullaert

Roux

Roca

Ricard

, Contributors. Prediction of outcome of nasal high flow use during COVID-19-related acute hypoxemic respiratory failure. Intensive Care Med. 2020;46(10):1924-1926.

17.

Innocenti

Lazzari

Paolucci

, et al. Role of prognostic scores in predicting in-hospital mortality and failure of non-invasive ventilation in adults with COVID-19. Intern Emerg Med. 2022;17(8):2367-2377.

18.

https://www.intensive.org“epic2”Documents

19.

Ehrmann

Ibarra-Estrada

, et al. Awake prone positioning for COVID-19 acute hypoxaemic respiratory failure: a randomised, controlled, multinational, open-label meta-trial. Lancet Respir Med. 2021;9(12):1387-1395.

20.

Luo

Pavlov

, et al. Awake prone positioning for non-intubated patients with COVID-19-related acute hypoxaemic respiratory failure: a systematic review and meta-analysis. Lancet Respir Med. 2022;10(6):573-583.

21.

Elharrar

Trigui

Dols

, et al. Use of prone positioning in nonintubated patients with COVID-19 and hypoxemic acute respiratory failure. JAMA. 2020;323(22):2336-2338.

22.

Coppo

Bellani

Winterton

, et al. Feasibility and physiological effects of prone positioning in non-intubated patients with acute respiratory failure due to COVID-19 (PRON-COVID): a prospective cohort study. Lancet Respir Med. 2020;8(8):765-774.

23.

Grieco

Maggiore

Roca

, et al. Non-invasive ventilatory support and high-flow nasal oxygen as first-line treatment of acute hypoxemic respiratory failure and ARDS. Intensive Care Med. 2021;47(8):851-866.

24.

Chiumello

Brioni

. Severe hypoxemia: which strategy to choose. Crit Care. 2016;20(1):132.

25.

Menga

Berardi

Ruggiero

Grieco

Antonelli

. Noninvasive respiratory support for acute respiratory failure due to COVID-19. Curr Opin Crit Care. 2022;28(1):25-50.

26.

Tobin

Jubran

Laghi

. Hypoxaemia does not necessitate tracheal intubation in COVID-19 patients. Comment on Br J Anaesth 2021; 126: 44-7. Br J Anaesth. 2021;126(2):e75-e76.

27.

Lupieri

Boffi

Ltaief

, et al. Response to the first awake prone positioning relates with intubation rate in SARS-CoV-2 patients suffering from acute respiratory failure with moderate to severe hypoxaemia: a retrospective study. Swiss Med Wkly. 2022;152(3536):w30212.

28.

Junhai

Jing

Beibei

. The value of ROX index in predicting the outcome of high flow nasal cannula: a systematic review and meta-analysis. Respir Res. 2022;23(1):33.

29.

Vega

Pisani

Dongilli

Nava

. COVID-19 pneumonia and ROX index: time to set a new threshold for patients admitted outside the ICU. Author’s reply. Pulmonology. 2022;28(4):322-323.

30.

Ferrer

Sancho

Bocigas

, et al. ROX index as predictor of high flow nasal cannula therapy success in acute respiratory failure due to SARS-CoV-2. Respir Med. 2021;189(November-December):106638.

31.

Colaianni-Alfonso

Montiel

Castro-Sayat

Roca

Grieco

. ROX index to predict CPAP outcome in hypoxemic respiratory failure due to COVID-19. Intensive Care Med. 2022;48(12):1818-1819.

32.

Spinelli

Roca

Mauri

. Dynamic assessment of the ROX index during nasal high flow for early identification of non-responders. J Crit Care. 2020;58(August):130-131.