Guest Skeptic: Dr. Justin Morgenstern. Justin is an emergency physician and the director of simulation education at Markham Stouffville Hospital in Ontario. He loves skepticism and medical education, especially when it is free and open access. He is the author of the #FOAMed blog First10EM.com and is an associate editor of Emergency Medicine Cases.
Case: A 60-year-old male with no major past medical history presents to your community emergency department with a three-day history of cough, fever and increasing shortness of breath. He is tachypneic with a respiratory rate of 28 and an oxygen saturation of 89% despite facemask oxygen, but he is mentating normally. Your Respiratory Therapist asks you whether you want her to try the new high flow nasal oxygen machine we have?
Background: High flow nasal oxygen is a novel device that actively humidifies and heats air to make flows of up to 60 liters a minutes tolerable. These incredibly high flows are important, because in order to provide 100% fiO2 to patients in respiratory distress, we must be able to match their minute ventilation.
High flow nasal oxygen also offers the theoretical benefits of low levels of PEEP and allows for much easier communication with our patients.
It has gained popularity in pediatrics for the treatment of severe bronchiolitis and it is seeing wider adoption among adult patients.
Clinical Question: In adult ICU patients with hypoxic respiratory distress not due to chronic obstructive pulmonary disease, asthma, or congestive heart failure, who do not require immediate intubation, does high-flow oxygen therapy or noninvasive ventilation therapy compared to standard therapy alone, reduce the rate of endotracheal intubation and improve outcomes?
Reference: Frat JP et al. High-Flow Oxygen through Nasal Cannula in Acute Hypoxemic Respiratory Failure. NEJM 2015.
Population: Adult ICU patients with acute hypoxic respiratory failure.
Inclusion criteria: Must have all four: Respiratory rate > 25/min, PO2/FiO2 ratio ≤ 300, a PCO2 of ≤ 45mmHg and no clinical underlying chronic respiratory failure.
Exclusion criteria: PaCO2 >45mmHG, exacerbation asthma or COPD, cardiogenic pulmonary edema, hemodynamic instability, use of vasopressors, GCS of 12 or less, contraindications to non-invasive ventilation, urgent need to intubate, DNR order.
High flow nasal oxygen (HFNO),
Non-invasive positive pressure ventilation (NIV)
Comparison: Standard face mask oxygen
Primary: Rate of intubation at 28 days
Secondary: Mortality in the ICU, mortality at 90 days, duration of ICU stay and number of ventilator free days. This is different than the two secondary outcomes listed in ClinicalTrails.gov that included ventilator free days at 28 days and ICU morbidity at 28 days.
Author’s Conclusions: In patients with nonhypercapnic acute hypoxemic respiratory failure, treatment with high-flow oxygen, standard oxygen, or noninvasive ventilation did not result in significantly different intubation rates. There was a significant difference in favor of high-flow oxygen in 90-day mortality.
Quality Checklist for Randomized Clinical Trials:
The study population included or focused on those in the ED. NO. These were adult patients from 23 ICUs in France and Belgium)
The patients were adequately randomized. YES. Permuted blocks of six in 1:1:1 ratio.
The randomization process was concealed. YES
The patients were analyzed in the groups to which they were randomized. YES. They performed an intention to treat analysis.
The study patients were recruited consecutively (i.e. no selection bias). YES
The patients in both groups were similar with respect to prognostic factors. YES
All participants (patients, clinicians, outcome assessors) were unaware of group allocation. NO. Clinicians and patients were aware of their group assignment. We are not sure if the assessors were blinded.
All groups were treated equally except for the intervention. NO. For the standard and nasal oxygen group, a trial of non-invasive ventilation was allowed before intubation – but the number is not presented. Also, this is an open label study, so other unreported treatment differences may have occurred.
Follow-up was complete (i.e. at least 80% for both groups). YES. 100% 90 days follow up.
All patient-important outcomes were considered. YES
The treatment effect was large enough and precise enough to be clinically significant. NO
Key Results: A total of 310 patients were included in the analysis, 106 in the high flow oxygen group, 94 in the standard oxygen group, and 110 in the noninvasive ventilation group.
Primary Outcome: Rate of intubation at 28 days – No statistical difference.
Standard O2: 47% (95% CI 37-57%)
High Flow Nasal O2: 38% (95% CI 29-47%)
Non-Invasive Ventilation: 50% (95% CI 41-59%)
Not statistically significant (p=0.18)
Secondary Outcome – ICU Mortality at 90d: Standard 23% vs. High flow 12% vs. NIV 28% (statistically different)
Standard O2: 23% (95% CI 16-33%)
High Flow Nasal O2: 12% (95% CI 7-20%)
Non-Invasive Ventilation: 28% (95% CI 21-37%)
HR 2.01 Standard vs. HFNO (95% CI 1.01-3.99 p=0.046)
HR 2.50 NIV vs. HFNO (95% CI 1.31-4.78 p=0.006)
Talk Nerdy: This was a multicenter trial attempting to ask an important question in critical care medicine. However there were a number of concerns/limitations in this study that threaten the validity of the conclusions.
Exclusion/Inclusion – We see a lot of patients that require respiratory support. They had close to 5,000 patients with acute respiratory failure. Over 2,000 patients were excluded with almost 1,400 for chronic lung disease and close to 700 because of cardiogenic pulmonary edema. We know from many studies that patients with COPD and cardiogenic pulmonary edema have a mortality benefit from NIV (Ram et al and Vital et al ).Of the 525 eligible for inclusion, 160 had “logistic” reasons. It is not clear what that means but it could have introduced some selection bias and negated the recruitment of consecutive patients.The vast majority of these patients (>80%) had a diagnosis of pneumonia. This makes it difficult to extrapolate to all patients with acute hypoxic respiratory distress.
Not Blinded – This could have introduced some bias into the results by influencing the treating staff not to intubate the intervention group compared to the control. However, they did have pre-specified criteria for endotracheal intubation, some of which was objective while other criteria was subjective (copious tracheal secretions).
Contamination– There was a fair amount of contamination between the groups. This was because it was left up to the treating physicians to try NIV for the standard group or HFNO group if the patient was looking worse. For example, 50% of the non-invasive group was on positive pressure for 8 hours or less each day, and the remainder of the time they were wearing a standard facemask. Similarly, the high flow oxygen group may have been put on non-invasive ventilation before intubation, although they don’t tell us how many actually were. This amount of cross-over/contamination between groups could have impacted the ITT analysis.
Power – This was a negative trail. They powered the study to detect a 20% difference in intubation rate and failed to achieve this result. That does not mean there is no difference or that the treatments are equivalent. This was a superiority trial and we can only conclude that the treatments were not show to have 20% superiority to the control.
Secondary Outcome – They highlight in their conclusion the significant difference in mortality at 90 days. The study was not powered for mortality but rather to demonstrate an absolute difference of 20% in intubation rates assuming an intubation rate of 60%. In addition, if you search ClinicalTrial.gov for their protocol it states the secondary outcome was ICU morbidity at 28 days, not mortality and mechanical ventilation-free to day 28. This is in contrast to the listed secondary outcomes of mortality in the ICU, mortality at 90 days, the number of ventilatory-free days between day 1 and day 28, and the duration of ICU stay. The lack of power and differences between published paper and original protocol should make us more skeptical of the secondary results.
Comment on author’s conclusion compared to SGEM Conclusion: We agree that the three different oxygenation strategies did not result in significantly different intubation rates.
We are less enthusiastic about the secondary endpoint that showed decreased mortality with high flow nasal oxygen, but think this is an interesting finding that should generate a hypothesis for future research.
SGEM Bottom Line: High flow nasal oxygen can be another tool in the toolbox to address patients in acute respiratory distress, but does not appear to be superior to standard care or NIV in this study.
Case Resolution: You try the high flow nasal oxygen but the patient continues to get clinically worse and you switch over to NIV.
Clinically Application: I am comfortable using NIV and will probably continue doing this given the good evidence to support this practice. However, for those patients not able to tolerate NIV, high flow nasal oxygen would be a reasonable option to try.
What do I tell my patient? Your current oxygen level is low and I think you require supplemental oxygen. We have three different types of oxygen mask: the usual one you are wearing, a tight fitting high-pressure mask, and a mask that only goes in the nose. The current evidence indicates they will probably all work for you, so you should tell us which seems the most comfortable.
Keener Kontest: Last week’s winner was Greg Rodgers for his second win in a row. Greg was able to name FIVE risk factors associated with spontaneous SAH included. Greg will be receiving a double strength cool skeptical prize.
Listen to the podcast for this week’s question. If you know the answer, then be the first person to email TheSGEM@gmail.com with “keener” in the subject line to win a cool skeptical prize.
Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.