0
Background: Post-extubation failure and noninvasive ventilation (NIV)
0
Post-extubation respiratory failure requiring re-intubation is a major setback for any critically ill patient. Current evidence mainly involves the pre-emptive use of NIV to reduce post-extubation failure, particularly in hypercapnic COPD patients. However, in practice this is difficult to implement widely. Except for the sickest COPD patients, it’s often difficult to persuade the patient to continue NIV long enough to benefit. Many patients with respiratory failure aren’t ill enough to justify NIV, yet remain at risk for extubation failure. Some patients have contraindications to NIV.
0
Physiology of post-extubation respiratory failure and role of pre-emptive ventilatory support
0
The most common cause of post-extubation respiratory failure is inability of the respiratory muscles to sustain the work of breathing, leading to progressive muscle fatigue (figure below). The first sign of decompensation is usually a gradually rising respiratory rate (a rapid-shallow breathing pattern is the body’s compensatory response to an excessive work of breathing). Based on this model, when evaluating an intervention to avoid reintubation the key variables pay attention to are respiratory rate and carbon dioxide tension (not oxygenation).
0
Post-extubation success often depends on the balanceof respiratory demands with muscle capability (MacIntyre 2004).
0
It’s important to differentiate between pre-emptive respiratory support (initiated at time of extubation) vs. salvage support (initiated later on, when the patient develops respiratory failure). NIV is only shown to be effective for pre-emptive support. This makes sense. Post-extubation failure is usually due to excessive work of breathing, with accumulating respiratory muscle fatigue. By the time the patient becomes significantly dyspneic, the respiratory muscles may already be fatigued and it may be too late to intervene.
0
Theory & previous data regarding HFNC after extubation
0
In addition to supporting oxygenation, HFNC may improve ventilation by reducing the anatomic dead space. The anatomic dead space is the volume of air in each breath which never reaches the alveoli and doesn’t participate in gas exchange. Anatomically this is equal to the volume between the alveoli and the source of fresh gas (normally the mouth/nose). HFNC pushes fresh gas into the airway, thereby decreasing the anatomic dead space (which now shrinks to the volume between the alveoli and, perhaps, the tracheal level). Decreasing the anatomic dead space improves the efficiency of ventilation, reducing the work of breathing (see image above).
0
In theory, HFNC may be a good option for a wide range of patients following extubation to reduce the risk of re-intubation. HFNC is noninvasive and very well tolerated. Practically speaking, a patient placed on HFNC will typically remain on HFNC overnight whereas its harder to achieve compliance with NIV. Unlike NIV, HFNC allows patients to eat and expectorate secretions.
0
There is currently little evidence regarding pre-emptive use of HFNC after extubation. Parke 2013 randomized 340 cardiac surgery patients to usual care vs. HFNC (at 45 liters/minute) from extubation through two days postoperatively. HFNC caused a small yet statistically significant reduction in arterial carbon dioxide tension, but no difference in respiratory rate. Patients receiving HFNC were less likely to require escalation to NIV (28% vs. 45%, p = 0.001). Only two patients required intubation. Overall the patients in this study may not have been ill enough to detect a meaningful clinical benefit from HFNC (patients who weren’t ready for extubation by 10:00 AM on the first postoperative day were excluded).
0
Rittayami 2014 performed a randomized crossover study of 17 respiratory ICU patients following extubation. The etiology of respiratory failure was mainly COPD or pneumonia. Following extubation patients were randomized to receive either non-rebreather facemask or HFNC at 35 liters/minute for 30 minutes, followed by the alternative therapy for 30 minutes. HFNC caused immediate reduction in respiratory rate, tachycardia, and subjective dyspnea (see below).
0
0
New study in the American Journal of Respiratory and Critical Care Medicine
0
Maggiore 2014 is a prospective randomized trial of 105 patients in a with PaO2/FiO2<300 before extubation. Patients had been intubated mainly due to pneumonia, multiple trauma, or atelectasis. Patients were randomized to receive oxygen via venturi mask or HFNC at 50 liters/minute flow following extubation. The primary endpoint was the relationship of arterial PaO2 to the set inspired oxygen fraction (FiO2). Secondary endpoints included patient discomfort, adverse events, and reintubation.
0
From 24-48 hours after extubation, the ratio of PaO2/FiO2 was higher among patients receiving HFNC (below figure, top left). Patients randomized to HFNC had better tolerance of the interface, fewer episodes of desaturation, less interface displacement, and a substantially reduced reintubation rate (4% vs 21%, p = 0.01). There was an immediate reduction in respiratory rate among the HFNC group as well as a trend toward reduced PaCO2, suggesting that the HFNC improved ventilation efficiency.
0
0
Weaknesses This is an industry-funded study, and their primary outcome (PaO2/set FiO2) seems a bit unusual. PaO2/FiO2 is a standard measure of the oxygenation function of the lung. However, for this measurement to have meaning, the FiO2 must be the effective FiO2 of inhaled gas in the trachea (not the set FiO2 of gas administered to the patient). Patients using low-flow oxygen delivery devices (including venturi masks) may have an effective FiO2 which is below the set FiO2 due to entrainment of room air (as discussed in the paper’s introduction). In contrast, HFNC at 50 liters/minute flow will achieve an effective FiO2 very close the set FiO2. Therefore, by selecting PaO2/(set FiO2) as the primary outcome, the design may give an unfair advantage to the HFNC group. Additionally, this is also not a patient-centered outcome.
0
Another weakness of the study was that it did not include any patients with obstructive lung disease. This may be because any patient with a combination of pre-extubation tachypnea and hypercapnia was excluded from the study and provided with post-extubation NIV. This impairs generalizability of results to patients with COPD or asthma.
0
Strengths: Despite choice of primary endpoint, there were substantial improvements in many secondary outcomes. Reintubation rate is the most important patient-centered “hard” clinical endpoint, and this was substantially reduced in the HFNC group. Based on their data, the number needed to treat (NNT) is six (for every six patients treated with NFNC, one reintubation would be avoided). By demonstrating improvement in carbon dioxide tension as well as respiratory rate, this study supports the concept that HFNC improves ventilation, reduces work of breathing, and prevents reintubation.
0
Conclusions
0
One of the primary goals of critical care is to safely and efficiently liberate patients from mechanical ventilation. Post-extubation respiratory failure requiring reintubation is a huge setback, and any way to avoid this is very welcome. HFNC is emerging as a safe and well-tolerated approach to support a patient’s oxygenation and ventilation. Maruzio et al. suggests that initiation of HFNC immediately following extubation reduces the risk of reintubation. Currently, post-extubation NIV is supported by a much greater body of evidence, and remains the first-line approach when indicated. For patients who aren’t candidates for post-extubation NIV or who are unable to tolerate NIV, HFNC may be an attractive option. For maximal efficacy, the flow rate should probably be kept high (i.e., 40-50 liters/minute) to reduce the work of breathing.
Image credits: First image from Ward JJ, Respiratory Care 2013 (nice review of HFNC).
Latest posts by Josh Farkas (see all)
- PulmCrit Blogitorial – Use of ECGs for management of (sub)massive PE - March 24, 2024
- PulmCrit Wee: Propofol induced eyelid opening apraxia – the struggle is real - March 20, 2024
- PulmCrit wee: Why I like central lines for GI bleed resuscitation - March 13, 2024
Hi, thanks for your posts, i think you’re doing a great job! I’m confused about what you say about Parke 2013 : “Patients receiving HFNC were less likely to require escalation to NIV (28% vs. 45%, p = 0.001)”. These figures seem huge to me and I can’t find them in the paper from Parke et al. Moreover, one could consider HFNC with 45L/min is already a kind of NIV considering a PEEP maybe about 4-5 mmHg. I’m working in anesthesia field in a french hospital, dealing frequently with major abdominal surgery and using HFNC quiet frequently. Thanks again for… Read more »
The numbers are from the final sentence of the results section, which reads: “Escalation in respiratory support at any time in the study occurred in 47 patients (27.8%) allocated to NHF compared with 77 (45%) standard care (OR 0.47, 95% CI 0.29–0.7, P=0.001).” Table 3 in the study presents qualitatively similar data although the numbers don’t match up exactly. Your point is very well taken: given that HFNC already represents a higher level of support than standard therapy, this is an unfair comparison. Hard to argue this. Honestly, as far as escalation in support goes, the only clinically meaningful outcome… Read more »