Abstract
Our hypothesis was that patients managed with noninvasive ventilation (NIV) on the wards could be risk-stratified with initial pulse oximetry/fraction of inspired oxygen (SpO2/FiO2) ratios and tidal volumes (Vte). A prospective study of consecutive patients with acute respiratory failure requiring NIV on the wards was conducted. A multivariate logistic regression model and a negative binomial regression model were used. A total of 403 patients (55.8% women) had a mean age of 65.0 ± 14.9 years with a mean body mass index of 32.1 ± 11.1 kg/m2. The 28-day mortality was 14.1%, and the intubation rate was 16.1%. Pneumonia was associated with the highest 28-day mortality (22.5%) and rate of intubation (36.7%) when compared with chronic obstructive pulmonary disease (4.4% and 7.3%) or congestive heart failure (22.2% and 13.4%). The SpO2/FiO2 groups were <214 (26.6%), 214 -357 (66.0%), and ≥357 (7.4%). Those in the SpO2/FiO2 < 214 group had a higher 28-day mortality rate (odds ratio [OR] = 8.19; 95% confidence interval [CI] 1.02 -65.7), intubation rate (OR = 3.7; 95% CI 1.1 -12.1), intensive care unit admission rate (OR = 2.9; 95% CI 1.2 -7.4), and length of stay (relative risk = 2.0; 95% CI 1.3 -3.0). A Vte/predicted body weight <7.7 mL/kg was associated with increased intubations (OR = 3.1; 95% CI 1.3 -7.4), intensive care unit admissions (OR = 2.5; 95% CI 1.3 -4.6), and 30-day readmissions (OR = 2.9; 95% CI 1.2 -6.8). In conclusion, in patients without acute respiratory distress syndrome who had acute respiratory failure managed with noninvasive ventilation on the wards, severe hypoxemia as assessed by a simple SpO2/FiO2 ≤ 214 was associated with poor outcomes.
Keywords: Intubation, mortality, noninvasive ventilation, SpO2/FiO2 ratio, tidal volume
The use of noninvasive ventilation (NIV) to treat acute respiratory failure is increasing.1–3 NIV use decreases the rate of endotracheal intubation and mortality in patients with chronic obstructive pulmonary disease (COPD) exacerbations4–6 and reduces the rate of endotracheal intubation in patients with cardiogenic pulmonary edema.4,7–9 NIV is commonly used to treat acute respiratory failure on the wards.10,11
The benefits of tidal volume (Vte) reduction may be present in critically ill patients without acute respiratory distress syndrome (ARDS).12,13 However, an optimal expiratory Vte strategy in NIV for acute respiratory failure has not been well established, except perhaps in patients with acute hypoxemic respiratory failure.4 Carteaux et al14 explored the feasibility of a low Vte strategy using NIV in acute hypoxemic respiratory failure and found that higher Vte was associated with failure defined as need of intubation. Frat et al showed that Vte >9 mL/kg predicted body weight (PBW) and severe hypoxemia as indicated by a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2/FiO2) ≤200 mm Hg in patients spontaneously breathing with acute hypoxemic respiratory failure was associated with a higher rate of intubation and mortality.15
More recently, the ratio of oxygen saturation by pulse oximetry/fraction of inspired oxygen (SpO2/FiO2) has been comparable to PaO2/FiO2 values16,17 in evaluating the degree of hypoxemia in patients with acute respiratory failure. The clinical characteristics and outcomes appear similar in patients with ARDS diagnosed with SpO2/FiO2 or PaO2/FiO2 ratios.18 We hypothesized that patients with acute respiratory failure supported with NIV on the wards could be initially risk-stratified using the SpO2/FiO2 ratio and Vte/PBW with lower SpO2/FiO2 ratios.
METHODS
This was a prospective observational study at a 680-bed academic hospital in Central Texas over a year and a half. The study was approved by the institutional review board. Consent was waived due to the observational nature of the study.
Consecutive patients receiving NIV were entered into the study after the primary medical team ordered NIV for any cause of acute respiratory failure. Patients were excluded from the study if they had a do not resuscitate status, were under the age of 18, received NIV after extubation, were on chronic NIV therapy for previously diagnosed sleep apnea or obesity hypoventilation syndrome, used NIV only immediately postoperatively, had a diagnosis of ARDS, were managed with continuous positive airway pressure, or were previously admitted and enrolled in this study (Figure 1).
Figure 1.
Patients screened for study inclusion. CPAP indicates continuous positive airway pressure; DNR/DNI, do not resuscitate/do not intubate; NIV, noninvasive ventilation.
NIV was delivered via Respironics V60, Vision, Focus, or STD 30 using an oral-nasal mask. Clinicians determined bilevel NIV pressure settings based on their assessment of the patient and managed the patients on the general medical floor assisted by respiratory therapists. In our hospital, the respiratory therapy protocol includes titrating FiO2 to a goal SpO2 of 92%. Furthermore, pressure changes were made at the physician’s discretion based on SpO2, blood gases, and clinical response. Escalation of care to the intensive care unit (ICU) was determined by consultation with the ICU team. Patient information collected included demographics, age, gender, height, weight, and calculated body mass index. The patient’s PBW was calculated using the same formula used in the ARDSnet Trial.10 During each 4-hour period, according to our usual practice, the respiratory therapist would document NIV data including the Vte, positive airway pressures, and clinical data, including patients’ SpO2 and FiO2. This information was recorded prospectively. We recorded Vte to reduce the leak error inherent to inspiratory Vte and better represent patients’ actual Vte. The Charlson Comorbidity Index was calculated for each patient based on a validated International Classification of Diseases–9 scoring system.19,20
We used the initial Vte to place patients into quartiles: <7.7 mL/kg PBW, ≥7.7–9.2 mL/kg PBW, ≥9.2–11.0 mL/kg PBW, and ≥11.0 mL/kg PBW. The initial SpO2/FiO2 ratio was used to classify patients into three validated groups derived from PaO2/FiO2 ratios of <200, ≥200–300, and ≥300 that had equivalent SpO2/FiO2 ratios of <214, ≥214<357, and ≥357. All analyses of variables were determined a priori. The primary outcome studied was 28-day mortality. Secondary outcomes were ICU admission, endotracheal intubation, length of stay (LOS), and 30-day readmission rate.
Means and frequencies were calculated for patient characteristics, treatment, and outcomes. Bivariate analyses assessed underlying differences by group status. Chi-square analyses (Fisher’s exact for small expected counts) compared categorical characteristics and outcomes across groups. The nonparametric Kruskal-Wallis test assessed differences for continuous measures between patient groups.
Multivariable logistic regression analyses were used to examine the association between SpO2/FiO2 ratio and Vte/PBW with the outcomes of 28-day mortality, intubation, ICU admission, and 30-day readmission, adjusting for patient age, gender, and comorbidities. A generalized linear model assuming a negative binomial distribution and log link was used to assess the relationship between SpO2/FiO2 ratio and Vte/PBW with hospital LOS, adjusted for the same confounders. In all models, age was divided by 10 to assess a decade effect, and a type I error of α = 0.05 was assumed throughout. All analyses were performed using SAS, Version 9.4 (SAS Institute, Cary, NC).
RESULTS
A total of 1880 consecutive patients who received NIV for any reason were evaluated for entry into the study, and 403 patients were included (Figure 1). We recorded 5,721 4-hour periods, with a median of 8.4 (interquartile range [IQR] 2–11) periods per patient. The mean age of the patients was 65.0 ± 14.9 years, the mean body mass index was 32.1 ± 11.1, and 55.8% were female. The median Vte/PBW was 8.9 (IQR 7.3–11.3) mL/kg, with a median inspiratory positive airway pressure of 12 cm H2O (IQR 10–12) and median expiratory positive airway pressure of 6 cm H2O (IQR 5–6). The 28-day mortality was 14.1%, 16.1% were intubated, 39.2% were admitted to the ICU, the median LOS was 5 days (IQR 3–9), and the 30-day readmission rate was 16.1% (Table 1). The most common cause of death was pneumonia followed by septic shock (Supplementary Table 2).
Table 1.
Characteristics of patients at baseline (N = 403)
| Variable | Resulta |
|---|---|
| Age (years) | 65 (14.9) |
| Male | 178 (44.2%) |
| Female | 225 (55.8%) |
| White | 294 (73%) |
| Black | 61 (15%) |
| Other race | 48 (12%) |
| Body mass index (kg/m2) | 32.1 (11.1) |
| Charlson comorbidity index | 4.3 (2.8) |
| Bilevel NIV | |
| Vte (mL) | 553 (445–683) |
| Vte/PBW (mL/kg) | 8.9 (7.3–11.3) |
| IPAP (cm H2O) | 12.0 (10.0–12.0) |
| EPAP (cm H2O) | 6.0 (5.0–6.0) |
| Vte/PBW quartile groups | |
| <7.7 mL/kg PBW (N = 100) | 6.5 (1.5) |
| ≥7.7–9.2 mL/kg PBW (N = 101) | 8.6 (1.5) |
| ≥9.2–11.0 mL/kg PBW (N = 101) | 9.8 (1.8) |
| ≥11.0 mL/kg PBW (N = 101) | 12.7 (2.9) |
| SpO2/FiO2 groups | |
| <214 (n = 107, 26.6%) | 154.9 (43.3) |
| 214–<357 (n = 266, 66%) | 272.6 (37.0) |
| ≥357 (n = 30, 7.4%) | 438.9 (35.6) |
| Outcomes | |
| 28-day mortalitya | 56 (14.1%) |
| Intubation | 65 (16.1%) |
| ICU admission | 158 (39.2%) |
| Length of stay (days) | 5.0 (3.0–9.0) |
| 30-day readmission | 65 (16.1%) |
Sample size of 398 patients (1% missing).
EPEP indicates expiratory positive airway pressure; FiO2, fraction of inspired oxygen; ICU, intensive care unit; IPAP, inspiratory positive airway pressure; IQR, interquartile range; NIV, noninvasive ventilation; PBW, predicted body weight; SD, standard deviation; SpO2, oxygen saturation by pulse oximetry; Vte, tidal volume. Data presented as N (%), mean (standard deviation), or median (interquartile range).
There was an average of 1.7 respiratory diagnoses per patient. The most common diagnoses were COPD (40.4%), congestive heart failure (CHF) (39.5%), and pneumonia (39.0%), with only a minority having pneumonia as the only diagnosis (n = 49, 12.2%). When comparing those patients who had only one diagnosis, patients with pneumonia had the highest 28-day mortality (22.5%) compared to patients with COPD (4.4%) or CHF (22.2%); in addition, they had the highest intubation rate (36.7%) compared to COPD (7.3%) or CHF (13.4%), the highest ICU admission rate (55.1%) compared to COPD (18.8%) or CHF (40.2%), and the longest LOS (6 days; IQR 4–14) compared to COPD (3 days; IQR 2–6) or CHF (5 days; IQR 3–9) (Table 2, Supplementary Figure 1).
Table 2.
Effect of respiratory diagnosis on outcomes (N = 403)
| Variable |
N (%) or median (quartile 1–3) |
P value | |||
|---|---|---|---|---|---|
| COPDa | CHFa | Pneumoniaa | Otherb | ||
| Exclusive diagnosisc | 69 (17.1%) | 82 (20.3%) | 49 (12.2%) | 72 (18.9%)b | 0.04 |
| Total diagnosesd | 163 (40.4%) | 159 (39.5%) | 157 (39.0%) | 207 (51.4%) | |
| Exclusive outcome | |||||
| 28-day mortality | 3 (4.4%) | 18 (22.2%) | 11 (22.5%) | 7 (9.9%) | 0.004 |
| Intubation | 5 (7.3%) | 11 (13.4%) | 18 (36.7%) | 8 (11.1%) | <0.0001 |
| ICU admission | 13 (18.8%) | 33 (40.2%) | 27 (55.1%) | 23 (31.9%) | 0.0005 |
| LOS median (min-max) | 3 (2–6) | 5 (3–9) | 6 (4–14) | 3 (2–7.5) | 0.0001 |
| 30-day readmission | 11 (15.9%) | 8 (9.8%) | 5 (10.2%) | 12 (16.7%) | 0.49 |
Not COPD, CHF, or pneumonia.
COPD without CHF, pneumonia, or other; CHF without COPD, pneumonia, or other; pneumonia without COPD, CHF, or other; other without COPD, CHF, or pneumonia.
Including COPD, CHF, pneumonia, and other diagnoses.
COPD indicates chronic obstructive pulmonary disease; CHF, congestive heart failure; ICU, intensive care unit; LOS, length of stay.
Irrespective of underlying diagnosis, the distribution of SpO2/FiO2 ratios was <214, 26.6%; ≥214–<357, 66.0%; and ≥357, 7.4%. The 28-day mortality rate was highest in patients with an SpO2/FiO2 ratio <214 (27.4%) when compared with those with 214–<357 (9.9%) or ≥357 (3.3%). An SpO2/FiO2 ratio <214 had the highest rates of ICU admission and endotracheal intubation (51.4% and 29.9%) when compared with those with 214–<357 (35.3% and 10.9%) or ≥357 (30.0% and 13.3%). The median LOS was longest with SpO2/FiO2 ratio <214 (7 days; IQR 4–12) compared to 214–<357 (5 days; IQR 3–8) or ≥357 (4 days; IQR 2–6). The 30-day readmission rate did not vary with the severity of hypoxemia (Table 3). The Bland-Altman plot showed a good level of agreement between the SpO2/FiO2 ratio and PaO2/FiO2 ratio and outcomes in 266 patients with initial arterial blood gases (Supplemental Figure 2).
Table 3.
Effect of SpO2/FiO2 ratio severity group on outcomes (N = 403)
| Variable | SpO2/FiO2 value |
P valuea | ||
|---|---|---|---|---|
| <214 | 214–<357 | ≥357 | ||
| N | 107 (26.6%) | 266 (66%) | 30 (7.4%) | |
| Initial SpO2/FiO2 ratio | 154.9 (43.3) | 272.6 (37.0) | 438.9 (35.6) | |
| Age (years) | 67.2 (13.7) | 65.0 (15.2) | 56.2 (14.1) | 0.002 |
| Women | 54 (50.5%) | 149 (56.0%) | 22 (73.3%) | 0.08 |
| Charlson score | 3.8 (2.5) | 4.5 (2.9) | 4.4 (2.8) | 0.09 |
| Body mass index (kg/m2) | 28.9 (9.4) | 32.9 (11.2) | 35.7 (13.1) | 0.001 |
| 4-hour periods | 5 (2–11) | 5 (2–11) | 5 (2–7) | 0.98 |
| Bilevel Vte/PBW (mL/kg) | 9.1 (7.8–10.4) | 9.2 (7.6–11.3) | 9.4 (8.9–10.9) | 0.30 |
| 28-day mortalityb | 29 (27.4%) | 26 (9.9%) | 1 (3.3%) | <0.0001 |
| Intubation | 32 (29.9%) | 29 (10.9%) | 4 (13.3%) | <0.0001 |
| ICU admission | 55 (51.4%) | 94 (35.3%) | 9 (30.0%) | 0.009 |
| Length of stay | 7 (4–12) | 5 (3–8) | 4 (2–6) | <0.0001 |
| 30-day readmission | 13 (12.2%) | 48 (18.1%) | 4 (13.3%) | 0.34 |
Chi-square analyses for categorical variables and Kruskal-Wallis/ANOVA test for continuous variables, assuming a type I error rate of α = 0.05.
Sample size of N = 398 patients (1% missing): SpO2/FiO2 < 214–106 patients (1% missing), 214 ≤ SpO2/FiO2 < 357–262 patients (1.5% missing), SpO2/FiO2 ≥ 357–30 patients (0% missing).
FiO2 indicates fraction of inspired oxygen; ICU, intensive care unit; PBW, predicted body weight; SpO2, oxygen saturation by pulse oximetry; Vte, tidal volume. Data presented as N (%), mean (standard deviation), or median (interquartile range).
The Vte/PBW were placed in quartiles of <7.7 mL/kg PBW (25%); ≥7.7–9.2 mL/kg PBW (25%); ≥9.2–11.0 mL/kg PBW (25%); and ≥11.0 mL/kg PBW (25%). Patients with Vte <7.7 mL/kg PBW had a higher ICU admission rate (47%), median LOS (6 days; IQR 4–10), and 30-day readmission rate (21%) compared to patients with Vte ≥7.7–9.2 mL/kg PBW (34%, 5 days, 11%); ≥9.2–11.0 mL/kg PBW (49%, 5 days, 24%); and ≥11.0 mL/kg PBW (28%, 4 days, 9%). The 28-day mortality and intubation rate was not significantly different over all quartiles of Vte/PBW.
In terms of the primary outcome, in the multivariable logistic regression models, severe hypoxemia (SpO2/FiO2 ≤214) was associated with increased odds of 28-day mortality (odds ratio [OR] = 8.19; 95% confidence interval [CI] 1.02–65.7). For every decade of increase in age, the odds of 28-day mortality increased (OR = 1.51; 95% CI 1.17–1.93) (Table 4).
Table 4.
Multivariate logistic regression analysis for the primary outcome of 28-day mortality
| Variables | Odds ratio | 95% CI | P value |
|---|---|---|---|
| Age (decade effect) | 1.51 | 1.17–1.93 | 0.001 |
| Female | 1.08 | 0.58–2.01 | 0.81 |
| Charlson Comorbidity Index | 1.09 | 0.99–1.21 | 0.09 |
| SpO2/FiO2 ≤ 214 | 8.19 | 1.02–65.7 | 0.05 |
| SpO2/FiO2 > 214<357 | 2.11 | 0.27–16.9 | 0.48 |
| <7.7 mL/kg PBW | 2.05 | 0.85–4.97 | 0.11 |
| ≥7.7–9.2 mL/kg PBW | 1.45 | 0.57–3.67 | 0.43 |
| ≥9.2–11.0 mL/kg PBW | 0.83 | 0.31–2.27 | 0.72 |
CI indicates confidence interval; FiO2, fraction of inspired oxygen; PBW, predicted body weight; SpO2, oxygen saturation by pulse oximetry.
In terms of secondary outcomes, in the multivariate logistic regression model, severe hypoxemia (OR = 3.7; 95% CI 1.1–12.1) and lower age in decades (OR = 0.7; 95% CI 0.6–0.9) were associated with failure of NIV (intubation). Severe hypoxemia (OR = 2.9; 95% CI 1.2–7.4) and Vte/PBW <7.7 mL/kg and ≥9.2–11.0 mL/kg PBW were associated with ICU admission (OR = 2.5; 95% CI 1.3–4.6 and OR = 2.6; 95% CI 1.4–4.8). Greater comorbidity per Charlson score was associated with a higher 30-day readmission (OR = 1.1; 95% CI 1.0–1.2) (Supplementary Table 1).
DISCUSSION
In our cohort of patients with non-ARDS acute respiratory failure managed with bilevel NIV on the wards, we found that the initial bedside SpO2/FiO2 ratio in patients with hypoxemic respiratory failure was associated with a higher mortality rate, intubation rate, ICU admission rate, and LOS. This is in concordance with the findings of Frat et al in which a PaO2/FiO2 < 200 at 1 h of NIV use in acute hypoxemic respiratory failure predicted intubation (OR 4.26; 95% CI 1.62–11.1, P = 0.003).15 Even as the use of NIV increases on the wards for acute respiratory failure,15 delays in intubation have been associated with higher mortality.21 Obtaining timely arterial blood gases may be challenging on the wards. Thus, a bedside tool to predict poor outcomes may alert clinicians to pursue earlier escalation of care, including intubation.
A second major finding was that patients with pneumonia had higher mortality, higher intubation rates, increased admissions to the ICU, and longer LOS compared with patients with COPD and CHF. In acute hypoxemic respiratory failure, NIV failure leads to a high mortality.22,23 This is of particular concern in patients with ARDS, which is why we excluded such patients from this study.2,4,23–26 Our intubation rate of 16.1% was far lower than the 50% reported in hypoxemic-only respiratory failure14 as a result of our case mix of patients, which included pneumonia (39.0%), COPD (40.4%), and CHF (39.5%). NIV use has proven benefits in patients with COPD and CHF exacerbations.4,5,27
The SpO2/FiO2 ratio has been used in lieu of the PaO2/FiO2 ratio.16–18 Because we did not have patients with an SpO2/FiO2 < 89 or a PaO2/FiO2 < 100, a group with a severe degree of hypoxemia could not be analyzed. We used the initial SpO2/FiO2 ratios, but these outcomes were similar when the initial SpO2/FiO2 ratios were compared to the average SpO2/FiO2 ratios during NIV utilization. Our data limitation didn’t allow us to look at the change in SpO2/FiO2 ratios with the application of NIV and the subsequent impact on outcomes, including the median times to intubation.
We found a low Vte/PBW was not associated with mortality or intubation. However, a Vte/PBW of <7.7 mL/kg PBW was associated with higher odds of ICU admission and an increased risk of prolonged LOS. A “dose escalation” effect with sequentially higher Vte quartiles was not found. This is in contrast to the findings of Frat et al who showed a 50% intubation rate with a Vte/PBW > 9 mL/kg (OR 3.14; 95% CI 1.22–8.06, P = 0.02) and a higher 90-day mortality.15 Notably, their study comprised only patients with acute hypoxemic respiratory failure managed with NIV, unlike our heterogeneous cohort of non-ARDS acute respiratory failure. It is possible that patients with low Vte/PBW may have had decreased lung compliance because of more severe underlying cardiopulmonary disease. However, the degree of hypoxemia did not account for this difference in our analysis. It is also possible that low Vte/PBW may have been a surrogate marker for a rapid shallow breathing pattern not amenable to capture with NIV. Due to the nature of this pragmatic study, we didn’t have all the respiratory rates in these patients.
In our “real-life” study with non-ARDS acute respiratory failure on the wards, we carefully excluded patients who only continued use of home NIV, needed NIV after extubation, had care limitations imposed by advance directives, and received NIV in the immediate postoperative period. Our study is unique in that we suggest using easily available parameters such as the SpO2/FiO2 ratio to risk-stratify those managed with NIV.
Our limitations include lack of severity of illness information such as Simplified Acute Physiology Score (SAPS) or Acute Physiology and Chronic Health Evaluation (APACHE) scores, which are not collected in patients admitted on the general medical floors. An observational study makes the determination of cause and effect difficult. Although we used the SpO2/FiO2 ratio for simplicity, we acknowledge that its accuracy would be lower in quantifying hypoxemia in the presence of high FiO2 compared to the PaO2/FiO2 ratio. Lastly, it is possible that certain indications for NIV or clinician experience with NIV management and patient selection may have introduced bias in our data.
In conclusion, in non-ARDS patients with acute respiratory failure managed on bilevel NIV, severe hypoxemia as assessed by a quick bedside SpO2/FiO2 ≤ 214 was associated with a high 28-day mortality, intubation rate, and ICU admission rate in patients irrespective of etiology. In these patients, a diagnosis of pneumonia was associated with worse outcomes compared with COPD, CHF, or pneumonia with COPD or CHF.
Supplementary Material
References
- 1.Esteban A, Ferguson ND, Meade MO, et al. Evolution of mechanical ventilation in response to clinical research. Am J Respir Crit Care Med. 2008;177(2):170–177. doi: 10.1164/rccm.200706-893OC. [DOI] [PubMed] [Google Scholar]
- 2.Antonelli M, Conti G, Esquinas A, et al. A multiple-center survey on the use in clinical practice of noninvasive ventilation as a first-line intervention for acute respiratory distress syndrome. Crit Care Med. 2007;35(1):18–25. doi: 10.1097/01.CCM.0000251821.44259.F3. [DOI] [PubMed] [Google Scholar]
- 3.Maheshwari V, Paioli D, Rothaar R, Hill NS. Utilization of noninvasive ventilation in acute care hospitals: a regional survey. Chest. 2006;129(5):1226–1233. doi: 10.1378/chest.129.5.1226. [DOI] [PubMed] [Google Scholar]
- 4.Nava S, Hill N. Non-invasive ventilation in acute respiratory failure. Lancet. 2009;374(9685):250–259. doi: 10.1016/S0140-6736(09)60496-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Plant PK, Owen JL, Elliott MW. Early use of non-invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on general respiratory wards: a multicentre randomised controlled trial. Lancet. 2000;355(9219):1931–1935. doi: 10.1016/S0140-6736(00)02323-0. [DOI] [PubMed] [Google Scholar]
- 6.Lightowler JV, Wedzicha JA, Elliott MW, Ram FS. Non-invasive positive pressure ventilation to treat respiratory failure resulting from exacerbations of chronic obstructive pulmonary disease: Cochrane systematic review and meta-analysis. BMJ. 2003;326(7382):185. doi: 10.1136/bmj.326.7382.185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Masip J, Roque M, Sánchez B, Fernández R, Subirana M, Expósito JA. Noninvasive ventilation in acute cardiogenic pulmonary edema: systematic review and meta-analysis. JAMA. 2005;294(24):3124–3130. doi: 10.1001/jama.294.24.3124. [DOI] [PubMed] [Google Scholar]
- 8.Masip J, Betbesé AJ, Páez J, et al. Non-invasive pressure support ventilation versus conventional oxygen therapy in acute cardiogenic pulmonary oedema: a randomised trial. Lancet. 2000;356(9248):2126–2132. doi: 10.1016/S0140-6736(00)03492-9. [DOI] [PubMed] [Google Scholar]
- 9.Gray A, Goodacre S, Newby DE, Masson M, Sampson F, Nicholl J; 3CPO Trialists . Noninvasive ventilation in acute cardiogenic pulmonary edema. N Engl J Med. 2008;359(2):142–151. doi: 10.1056/NEJMoa0707992. [DOI] [PubMed] [Google Scholar]
- 10.Raurell-Torredà M, Argilaga-Molero E, Colomer-Plana M, Ródenas-Fransico A, Ruiz-Garcia MT, Muntaña JU. Optimising non-invasive mechanical ventilation: Which unit should care for these patients? A cohort study. Aust Crit Care. 2017;30(4):225–233. doi: 10.1016/j.aucc.2016.08.005. [DOI] [PubMed] [Google Scholar]
- 11.Ozsancak Ugurlu A, Sidhom SS, Khodabandeh A, et al. Where is noninvasive ventilation actually delivered for acute respiratory failure? Lung. 2015;193(5):779–788. doi: 10.1007/s00408-015-9766-y. [DOI] [PubMed] [Google Scholar]
- 12.Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA. 2012;308(16):1651–1659. doi: 10.1001/jama.2012.13730. [DOI] [PubMed] [Google Scholar]
- 13.Slutsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med. 2013;369(22):2126–2136. doi: 10.1056/NEJMra1208707. [DOI] [PubMed] [Google Scholar]
- 14.Carteaux G, Millán-Guilarte T, De Prost N, et al. Failure of noninvasive ventilation for de novo acute hypoxemic respiratory failure: role of tidal. Crit Care Med. 2016;44(2):282–290. doi: 10.1097/CCM.0000000000001379. [DOI] [PubMed] [Google Scholar]
- 15.Frat JP, Ragot S, Coudroy R, et al. Predictors of intubation in patients with acute hypoxemic respiratory failure treated with a noninvasive oxygenation strategy. Crit Care Med. 2018;46(2):208–215. doi: 10.1097/CCM.0000000000002818. [DOI] [PubMed] [Google Scholar]
- 16.Festic E, Bansal V, Kor DJ, Gajic O; US Critical Illness and Injury Trials Group: Lung Injury Prevention Study Investigators (USCIITG–LIPS). SpO2/FiO2 ratio on hospital admission is an indicator of early acute respiratory distress syndrome development among patients at risk. J Intensive Care Med. 2015;30(4):209–216. doi: 10.1177/0885066613516411. [DOI] [PubMed] [Google Scholar]
- 17.Sanz F, Dean N, Dickerson J, et al. Accuracy of PaO2/FiO2 calculated from SpO2 for severity assessment in ED patients with pneumonia. Respirology. 2015;20(5):813–818. doi: 10.1111/resp.12560. [DOI] [PubMed] [Google Scholar]
- 18.Rice TW, Wheeler AP, Bernard GR, et al; National Institutes of Health, National Heart, Lung, and Blood Institute ARDS Network. Comparison of the SpO2/FiO2 ratio and the PaO2/FiO2 ratio in patients with acute lung injury or ARDS. Chest. 2007; 132(2):410–417. doi: 10.1378/chest.07-0617. [DOI] [PubMed] [Google Scholar]
- 19.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 20.Li B, Evans D, Faris P, Dean S, Quan H. Risk adjustment performance of Charlson and Elixhauser comorbidities in ICD-9 and ICD-10 administrative databases. BMC Health Serv Res. 2008;8(1):12. doi: 10.1186/1472-6963-8-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cabrini L, Esquinas A, Pasin L, et al. An international survey on noninvasive ventilation use for acute respiratory failure in general non-monitored wards. Respir Care. 2015;60(4):586–92. doi: 10.4187/respcare.03593. [DOI] [PubMed] [Google Scholar]
- 22.Bellani G, Laffey JG, Pham T, et al. Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA. 2016;315(8):788–800. doi: 10.1001/jama.2016.0291. [DOI] [PubMed] [Google Scholar]
- 23.Carrillo A, Gonzalez-Diaz G, Ferrer M, et al. Non-invasive ventilation in community-acquired pneumonia and severe acute respiratory failure. Intensive Care Med. 2012;38(3):458–466. doi: 10.1007/s00134-012-2475-6. [DOI] [PubMed] [Google Scholar]
- 24.Thille AW, Contou D, Fragnoli C, et al. Non-invasive ventilation for acute hypoxemic respiratory failure: intubation rate and risk factors. Crit Care. 2013;17(6):R269. doi: 10.1186/cc13103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rana S, Jenad H, Gay PC, Buck CF, Hubmayr RD, Gajic O. Failure of non-invasive ventilation in patients with acute lung injury: observational cohort study. Crit Care. 2006;10(3):R79. doi: 10.1186/cc4923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hess DR. Noninvasive ventilation for acute respiratory failure. Respir Care. 2013;58(6):950–972. doi: 10.4187/respcare.02319. [DOI] [PubMed] [Google Scholar]
- 27.ARDS Definition Task Force , Ranieri VM, Rubenfeld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin definition. JAMA. 2012;307(23):2526–2533. doi: 10.1001/jama.2012.5669. [DOI] [PubMed] [Google Scholar]
- 28.Cabrini L, Landoni G, Oriani A, et al. Noninvasive ventilation and survival in acute care settings: a comprehensive systematic review and metaanalysis of randomized controlled trials. Crit Care Med. 2015;43(4):880–888. doi: 10.1097/CCM.0000000000000819. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.