Previous studies suggest that prone positioning (PP) can increase PaO2/FiO2 and reduce mortality in moderate to severe acute respiratory distress syndrome (ARDS). The aim of our study was to determine whether the early use of PP combined with non-invasive ventilation (NIV) or high-flow nasal cannula (HFNC) can avoid the need for intubation in moderate to severe ARDS patients.
Methods
This prospective observational cohort study was performed in two teaching hospitals. Non-intubated moderate to severe ARDS patients were included and were placed in PP with NIV or with HFNC. The efficacy in improving oxygenation with four support methods—HFNC, HFNC+PP, NIV, NIV+PP—were evaluated by blood gas analysis. The primary outcome was the rate of intubation.
Results
Between January 2018 and April 2019, 20 ARDS patients were enrolled. The main causes of ARDS were pneumonia due to influenza (9 cases, 45%) and other viruses (2 cases, 10%). Ten cases were moderate ARDS and 10 cases were severe. Eleven patients avoided intubation (success group), and 9 patients were intubated (failure group). All 7 patients with a PaO2/FiO2 < 100 mmHg on NIV required intubation. PaO2/FiO2 in HFNC+PP were significantly higher in the success group than in the failure group (125 ± 41 mmHg vs 119 ± 19 mmHg, P = 0.043). PaO2/FiO2 demonstrated an upward trend in patients with all four support strategies: HFNC < HFNC+PP ≤ NIV < NIV+PP. The average duration for PP was 2 h twice daily.
Conclusions
Early application of PP with HFNC, especially in patients with moderate ARDS and baseline SpO2 > 95%, may help avoid intubation. The PP was well tolerated, and the efficacy on PaO2/FiO2 of the four support strategies was HFNC < HFNC+PP ≤ NIV < NIV+PP. Severe ARDS patients were not appropriate candidates for HFNC/NIV+PP.
Trial registration
ChiCTR, ChiCTR1900023564. Registered 1 June 2019 (retrospectively registered)
Hinweise
Lin Ding, Li Wang and Wanhong Ma contributed equally to this work.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Abkürzungen
ARDS
Acute respiratory distress syndrome
ECMO
Extracorporeal membrane oxygenation
HFNC
High-flow nasal cannula
ICU
Intensive care unit
NIV
Non-invasive ventilation
NMB
Neuromuscularblocker
PP
Prone positioning
RCT
Randomized controlled trial
TPP
Trans-pulmonary pressure
VILI
Ventilator-induced lung injury
Introduction
Acute respiratory distress syndrome (ARDS) has a high mortality of 25~40%, even with improvement in supportive therapies. Previous studies suggest that prone positioning (PP) can increase the average ratio of arterial oxygen tension to the fraction of inspired oxygen (PaO2/FiO2) by + 35 mmHg and reduce mortality in moderate to severe ARDS, especially when combined with neuromuscular blocker (NMB) and low tidal volume ventilation, which decrease the risk of ventilator-induced lung injury (VILI) [1‐5]. However, PP was only recommended in severe ARDS with PaO2/FiO2 < 100 mmHg, and the actual use of PP is less than 33% of severe ARDS patients [6].
Early use of non-invasive ventilation (NIV) can reduce the need for intubation of mild ARDS patients [7‐12]. In a few observational studies [13‐16], lower rates of intubation were seen among hypoxemic patients receiving high-flow nasal cannula (HFNC) than among those receiving NIV or standard oxygen therapy. HFNC provides a lower trans-pulmonary pressure (TPP) potentially resulting in less VILI than NIV.
Anzeige
From a theoretical and physiological point of view, NIV and HFNC may both be beneficial in patients with ARDS. However, these two techniques work via different mechanisms. NIV applies end-expiratory positive airway pressure (PEEP) and pressure support (PS). Optimally, PEEP increases functional residual capacity and opens collapsed alveoli, thereby improving ventilation–perfusion matching and reducing intrapulmonary shunt, as well as improving lung compliance, thus reducing respiratory load. The latter assists respiratory muscles during inspiration, reducing work of breathing and dyspnea. In contrast, HFNC only generates a small positive pressure spike at end-expiration that depends on the nasal airflow and the extent of mouth opening. HFNC appears to improve oxygenation primarily by flushing the nasal airspaces, reducing anatomical dead space. In addition, by delivering warm, well-humidified gas through the nostrils and avoiding the discomfort generated by the pressure that NIV masks exert on the facial skin, HFNC is tolerated much better than NIV and can be applied continuously for long periods of time [17]. The major goal of NIV and HFNC in treating ARDS is to achieve a sufficient oxygenation to avoid endotracheal intubation. However, NIV and HFNC are only “partially support” therapies. They do not address the underlying pathology of ARDS sufficiently, such as the ventilation–perfusion matching caused by atelectasis or consolidation in the dependent areas when supported with HFNC [18]. High-level PS in combination with deep inspiratory efforts could generate high tidal volumes and excessive trans-pulmonary pressures, increasing lung stress and contributing to VILI, maybe a risk for NIV failure [19]. In this regard, combining PP with these non-invasive respiratory supports in ARDS may result in better physiological effects on ventilation–perfusion mismatch, a better drainage for purulent lung infection-induced ARDS, and greater homogeneity in ARDS mechanics while receiving positive pressure support.
Based on these potential beneficial mechanisms, we performed a prospective observational study to test the hypothesis that early use of PP combined with either NIV or HFNC can avoid the need for intubation in moderate to severe ARDS patients. We also evaluated the effects of PP combined with HFNC or NIV to improve PaO2/FiO2 compared with HFNC or NIV support only, and the safety of the PP therapy in awake, non-intubated ARDS patients.
Methods
Study design
This prospective observational cohort study was performed in a respiratory intensive care unit (ICU) of two university teaching hospitals. This study was registered online (ChiCTR1900023564). The study was approved by the ethics committees of both participating institutions. All participating subjects or their next of kin provided written informed consent.
Patient selection
The diagnosis of moderate to severe ARDS was made when a patient met the Berlin definition criteria [20]. ARDS patients admitted to the respiratory ICU were evaluated with arterial blood gas analysis after a PEEP of 5 cmH2O supported by NIV (CPAP/BiPAP mode) with FiO2 0.5 for at least 30 min. ARDS patients were included if their PaO2/FiO2 was less than 200 mmHg on this level of support.
Anzeige
Exclusion criteria were (1) signs of respiratory fatigue (RR > 40/min, PaCO2 > 50 mmHg/pH < 7.30, and obvious accessory respiratory muscle use), (2) immediate need for intubation (PaO2/FiO2 < 50 mmHg, unable to protect airway or change of mental status), (3) unstable hemodynamic status, and (4) inability to collaborate with PP with agitation or refusal.
Interventions
Indications for PP: Included patients were all switched to HFNC after initial evaluation by NIV. The target SpO2 was > 90% with a FiO2 equal to or lower than 0.6. Patients were placed in PP in two conditions with HFNC: (1) if the patients had a stable SpO2, they were repositioned to PP 1 h after the HFNC was initiated; (2) if patient’s SpO2 was consistently < 90% on HFNC for more than 10 min, they were put in the prone position with the same setting of HFNC. If the patients had a consistent SpO2 < 90% when on NIV evaluation with a FiO2 of 0.6, they were put in a prone position with the same setting of NIV (as shown in Fig. 1). Abbreviations for four interventions applied were (1) HFNC, high-flow nasal cannula support alone; (2) HFNC+PP, high-flow nasal cannula therapy combined with prone positioning; (3) NIV, non-invasive ventilation support alone; and (4) NIV+PP, non-invasive ventilation combined with prone positioning.
×
Duration and frequency of PP: patients remained in PP with HFNC or NIV for at least 30 min; if patients tolerated PP well, PP would last until the patients felt too tired to maintain that position. The PP was performed at least two times a day for the first 3 days after the patient inclusion. No sedation was used during the PP. The patients were monitored by bedside respiratory therapists and nurses for their comfort and tolerance for the PP every 15 min.
Patients who received NIV were ventilated using the CPAP or bilevel positive airways pressure S/T mode (BiPAP Vision or V60; Respironics Inc., Murrysville, PA) via an oral-nasal face mask. HFNC was delivered with a max flow rate of 60 L/min and a maximum FiO2 of 0.9 (AIRVO2; Fisher & Paykel Healthcare Limited, Auckland, New Zealand) via a specialized nasal cannula (F&P Optiflow™).
Outcomes
The primary outcome for the efficacy of PP combined with HFNC/NIV was the rate of avoidance for intubation. Patients who avoided intubation were classified as success while those needing intubation were classified as failure.
Secondary outcome for efficacy of PP combined with HFNC/NIV was the increase in PaO2/FiO2 from HFNC alone to HFNC+PP, to NIV alone, and to NIV+PP. The threshold of PaO2/FiO2 for successful PP cases was also assessed. Safety outcomes for PP combinations were the time duration (tolerance) for each PP therapy session.
Data collection
The following information of all patients was collected in a data file: patients’ characteristics, including age, sex, medical history, reasons for ARDS, the laboratory and microbiology findings, and outcome and cause of death, complications including barotrauma, the different respiratory support methods used, and the time of PP. The arterial blood gases were obtained after patients received one mode of support for more than 30 min.
Statistical analysis
Based on the 61 to 77% intubation rate for ARDS patients reported in previous studies [21, 22], we estimated a total of at least 18 subjects with an expected intubation rate of 75% in the moderate to severe ARDS patients with NIV or HFNC support would have sufficient power to detect a 40% reduction in intubation (to 45% = 75% × [1–0.4]) in the PP patients in our cases, with a confidence level [1 − α] = 95% and power level [1 − β] = 80%.
Anzeige
Quantitative continuous variables were given as either means (±standard deviations [SD]) or medians (with inter-quartile ranges [IQR]) were compared using the unpaired Student’s t test or the Mann-Whitney test. Qualitative or categorical variables were compared with the chi-square test or Fisher’s exact test. P values < 0.05 were considered significant. All analyses were performed with SPSS software for Windows (release 11.5).
Results
Patient characteristics
Between January 2018 and April 2019, a total of 20 patients were included from the two centers. The average age was 50 ± 10 years old, and 65% (13/20) were male. The main cause of ARDS was viral pneumonia, due to influenza (9 cases, 45%) and other viruses (2 cases, 10%). Ten cases met the criteria for moderate ARDS and ten cases were severe.
Twelve patients were attempted on HFNC+PP, of whom seven required escalation to NIV, with two of those patients receiving NIV+PP for further support. HFNC could not be attempted on seven patients who subsequently received NIV+PP. One patient required NIV+PP without attempting PP on HFNC when HFNC alone failed.
Primary outcome
Eleven patients (11/20, 55%) avoided intubation, and nine patients were intubated. Three of the nine (33%) intubated patients needed extracorporeal membrane oxygenation (ECMO) support. Only one patient died in the entire group.
Anzeige
Secondary outcomes
As shown in Fig. 2, the PaO2/FiO2 showed a trend of increase in transitions from HFNC to HFNC+PP, to NIV, and to NIV+PP. Only two cases had a lower PaO2/FiO2 when PP was added to NIV (cases 9 and 15), and in two cases, HFNC+PP had a higher PaO2/FiO2 than NIV (cases 12 and 17).
For the 11 patients in the success group, as shown in Table 1, eight patients received HFNC+PP, of whom six patients required change to NIV, with one patient requiring the addition of PP to NIV for further support. Three patients received NIV+PP after evaluation, and one patient received HFNC+PP and NIV+PP. PaO2/FiO2 was significantly higher in HFNC+PP than in HFNC (130 ± 35 mmHg vs 95 ± 22 mmHg, P = 0.016). PaO2/FiO2 had an upward trend when PP was added to NIV (166 ± 12 mmHg vs 140 ± 30 mmHg, P = 0.133, panel a of Fig. 2).
Table 1
Clinical characteristics and outcomes of patients in the success group
For the nine patients in the failure group (Table 2), four patients received HFNC+PP, of whom two required NIV, and five patients received NIV+PP after initial evaluation. Excluding the one patient who was unable to cooperate with PP on NIV, PaO2/FiO2 were significantly higher in NIV+PP compared to NIV (111 ± 20 mmHg vs 77 ± 14 mmHg, P = 0.011, panel b of Fig. 2).
Table 2
Clinical characteristics and outcomes of patients in the failure group
Regarding safety, two patients could not tolerate PP. As shown in panel b of Fig. 2, all seven patients with a PaO2/FiO2 < 100 mmHg when initially evaluated on NIV only were ultimately intubated.
Comparison of success and failure patients
Characteristics of success and failure patients are shown in Table 4. SpO2 was significantly higher before PP in the success group than in the failure group (95% ± 1% vs 93% ± 3%, P = 0.006). PaO2/FiO2 in those evaluated on HFNC+PP was significantly higher in the success group than in the failure group (125 ± 41 mmHg vs 119 ± 19 mmHg, P = 0.043). No significant difference in total days, frequency, and duration of PP between the successful and the failure groups was demonstrated (Table 5).
Table 4
Comparisons between patients avoid intubation and received invasive ventilation
Comparisons of PP duration between success group and failure group
Success, n = 11
Failure, n = 9
P value
Total days
4.3 ± 3.7
1.9 ± 1.2
0.065
Frequency during a day
2.4 ± 1.5
1.6 ± 0.5
0.130
Duration of one course of PP (h)
1.8 ± 0.7
1.9 ± 1.4
0.826
PP prone position
Discussion
The main strength of our study is that, for the first time, we evaluated the early use of PP combined with HFNC/NIV in non-intubated awake patients with moderate to severe ARDS. Our study revealed that early PP combined with HFNC/NIV may avoid the need for intubation in up to half of the patients with moderate to severe ARDS; when PP was added, PaO2/FiO2 increased by 25 to 35 mmHg compared with the prior HFNC or NIV; and PP was safely performed and well tolerated by the moderate ARDS patients.
ARDS was initially reported with mortality as high as 90% [23]. Although outcome has improved in recent decades, mortality of ARDS currently approximates 32% and 45% for patients with PaO2/FiO2 of 100–200 mmHg and < 100 mmHg, respectively [24]. High mortality warrants ongoing research efforts for alternative therapeutic strategies. PP during invasive mechanical ventilation for ARDS has been extensively studied and demonstrates decreased mortality and improved oxygenation and lung recruitment. As PP was recommended for patients with a threshold PaO2/FiO2 < 150 mmHg based on previous studies, we chose to test the effects of PP in non-intubated ARDS patients with a PaO2/FiO2 < 200 mmHg, which includes moderate and severe ARDS patients.
In our study, there were 11 patients avoided intubation, including 8 (73%) patients of moderate ARDS and 3 (27%) patients of severe ARDS. Nine patients were intubated, including 2 (22%) patients of moderate ARDS and 7 (78%) patients of severe ARDS. Although no differences were statistically significant due to the small sample size of our study, moderate ARDS had a higher proportion in the successful group than in the failure group (73% vs 27%, P = 0.174). Moreover, SpO2 was significantly higher at inclusion point in the success group than in the failure group. Furthermore, although HFNC+PP or NIV+PP can improve the oxygenation compared to the HFNC or NIV single support in most severe ARDS patients, the intubation rate was very high and three patients needed ECMO for further support after intubation. Therefore, our results suggest that patients who meet the criteria for severe ARDS patients are not appropriate candidates for PP therapy. The late addition of PP to NIV in severe ARDS results in high risk for delayed intubation and treatment failure. Conversely, early application of PP in patients with moderate ARDS and an initial SpO2 > 95% on NIV may benefit and avoid the need for invasive ventilation.
Prone positioning may decrease the need for intubation and even mortality by improving the oxygenation of ARDS [1, 25‐28]. In our study, the addition of PP actually showed an improvement for oxygenation in most cases. In the success group, when compared with HFNC or NIV support alone, PaO2/FiO2 increased significantly after PP was added. In the failure group, PaO2/FiO2 increased significantly in NIV+PP than NIV support alone. PaO2/FiO2 was also higher in HFNC+PP than HFNC. As a result, the efficacy strength for PaO2/FiO2 improvement of the four support strategies was HFNC < HFNC+PP ≤ NIV < NIV+PP.
One interesting point is that in our study, the effects of HFNC+PP were compared with those of NIV for the first time, and HFNC+PP showed similar or less effects in oxygenation improvement than NIV. When a high level of positive pressure was applied by NIV in ARDS patients with spontaneous breathing, high driving pressure plus the high level of airway positive pressure delivered by NIV may lead to a greater possibility of VILI and poor outcome [13, 19]. HFNC provides a lower level of positive pressure than NIV and thus may be better in avoiding VILI in non-intubated ARDS patients with spontaneous breathing and a high TPP. Conversely, in our study, PaO2/FiO2 was higher with NIV than HFNC+PP in the success group. Our results suggest that for the patient with good cooperation with NIV, NIV results in a higher PaO2/FiO2 than HFNC+PP and may be the first choice for improving oxygenation. In patients who cannot tolerate NIV, HFNC+PP may be an effective alternative.
When PP was performed in intubated ARDS patients, high dose of sedation or even neuromuscular blockers may be needed [4, 6]. However, in the present study, our patients generally showed good tolerance for PP while awake and receiving HFNC or NIV. The average duration of one course of PP was 2 h. In our study, ARDS was mainly caused by infectious pneumonia, and airway drainage of secretions may be important for treatment. Prone positioning in awake patients promotes better drainage of the airway and, especially when combined with HFNC, may be one reason for successful avoidance of intubation in our study.
Several limitations of our study exist. First, the small sample size was prone to bias, yielding spurious findings on statistical analysis. Increasing the sample size and collecting more cases in further study may avoid this limitation. Second, not all patients were managed with all four support strategies, possibly causing misclassification bias. Finally, this two-center observational, non-randomized, uncontrolled trial cannot prove the beneficial effects of adding PP to HFNC/NIV in early ARDS. However, our study informs the design of a future multi-center prospective randomized controlled trial (RCT) of PP in non-intubated moderate ARDS patients to answer this question.
Conclusion
Early application of PP with HFNC or NIV, especially in patients with non-intubated moderate ARDS and with SpO2 > 95%, may avoid intubation. PP was generally well tolerated, and the efficacy on PaO2/FiO2 of the four support strategy was HFNC < HFNC+PP ≤ NIV < NIV+PP. Severe ARDS patients were not appropriate candidates for NFNC/NIV+PP, and the risk of complications with delayed intubation may be increased. A multi-center prospective RCT is warranted in the future in non-intubated moderate ARDS patients to get a settled conclusion.
Acknowledgements
We thanked professor Richard G Wunderink (from Pulmonary and Critical Care Division of Northwestern Memorial Hospital, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA) for his advices and help for the manuscript revision.
Ethics approval and consent to participate
This study is approved by the ethics committee of Beijing Chao-Yang Hospital.
Consent for publication
Each patient gave written informed consent for their data to be used for research and publication.
Competing interests
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.