Authors: Martin Dres, Alexandre Demoule
Categories: Commentaries
Source: Revista Brasileira de Terapia Intensiva
for critically ill patients
Authors: Martin Dres, Alexandre Demoule
The most conventional forms of oxygen delivery rely on facemasks, a nasal cannula or
nasal prongs. However, the use of these methods is limited by certain drawbacks,
including the need for a flow of oxygen higher than 15L/min in case of severe
hypoxemia or the dilution of administered oxygen by entrained room air in cases of
high inspiratory flow. An alternative to conventional oxygen therapy has received
growing heated, humidified, high flow nasal oxygen (HFNO), a technique
that can deliver heated and humidified oxygen, with a controlled fraction of
inspired oxygen (FiO2), at a maximum flow rate of 60L/min via a nasal
cannula. For a decade, the use of HFNO has been considered for patients with
hypoxemic de novo acute respiratory failure (ARF). Recent reports
suggest that HFNO can also be used to secure intubation and to prevent
post-extubation ARF. The purpose of the present review is to provide clinicians with
the most recent information on HFNO and to discuss its benefits and risks in its
most common indications.
In physiological terms, HFNO improves the fraction of inspired oxygen, washes and reduces dead space, generates positive end-expiratory pressure (PEEP) and provides more comfort than cold and dry oxygen.
Patients with ARF have high inspiratory flow rates ranging between 30 and 120L/min.
This flow commonly exceeds the maximum 15L/min flow rate that usual devices deliver.
Subsequently, entrained room air dilutes the oxygen, which in turn decreases the
FiO2. By delivering up to 100% oxygen at a maximum flow rate of
60L/min, HFNO minimizes the entrainment of room air and subsequently increases the
FiO2.
A high rate of airflow delivered directly to the naso-pharynx improves carbon dioxide clearance by flushing the expired carbon dioxide from the upper airway.^(1)^ Subsequently, the dead space attributable to the flush out volume is reduced, thereby improving alveolar ventilation.^(2)^ The reduction of dead space contributes to the observed decrease in both the respiratory rate and the work of breathing.^(3)^
High flow nasal oxygen generates a varying PEEP level.^(4-6)^ In healthy
volunteers treated with HFNO with a closed mouth and a flow rate of 60L/min, the
measured PEEP was as high as 7cmH2O.^(6)^ However, this level of PEEP can decrease easily, as soon as
the mouth is opened. Each 10L/min increase in flow rate increases the mean airway
pressure by 0.69cmH2O when subjects breathe with their mouths closed, and
by 0.35cmH2O when they breathe with their mouths open.^(5)^
Furthermore, HFNO delivers a heated and humidified flow, providing more comfort than dry air.^(7)^ HFNO also increases the water content of mucus, which facilitates secretion removal and avoids desiccation and epithelial injury. Finally, compared to other devices such as non-invasive ventilation (NIV), the tolerance of HFNO may be higher due to its interface, as simple nasal prongs enable patients to speak, eat and drink.
The expected benefits of HFNO depend on its indication and the device to which HFNO is compared (Table 1). To date, the use of HFNO has been suggested in several indications, but few have been rigorously evaluated (Table 2). The two primary indications in terms of level of evidence are (1) to prevent intubation in patients with hypoxemic de novo ARF and (2) to prevent post extubation ARF and subsequent reintubation in the medical ICU or after surgery. Other important indications in which HFNO has been investigated but which lack clear evidence of clinical benefits are preoxygenation before the intubation of severely hypoxemic patients, de novo hypoxemic ARF in immunocompromised hosts and oxygenation to secure flexible bronchoscopy.
lower intubation rate than standard oxygen and non-invasive ventilation in the most severely hypoxemic patients
The largest known study on HFNO in patients with hypoxemic ARF was recently conducted
by Frat et al.^(8)^ This randomized
controlled trial included patients with hypoxemic ARF and a ratio of partial
pressure of arterial oxygen to the fraction of inspired oxygen
(PaO2/FiO2) of 300mmHg or less, mainly due to pneumonia.
Patients were assigned to receive HFNO, standard oxygen therapy delivered through a
facemask, or NIV. The primary outcome was the rate of endotracheal intubation, which
was similar among the three groups (38% in HFNO versus 47% in
conventional oxygen therapy and 50% in NIV). However, in a post hoc analysis
restricted to patients with severe initial hypoxemia defined by a
PaO2/FiO2 ≤ 200mmHg, the intubation rate was
significantly lower in patients who received high flow oxygen than in the other two
groups. Furthermore, mortality, which was a secondary outcome, was significantly
lower in the HFNO group than in the other two groups. Although these results are
promising, they need to be confirmed by further trials before the systematic use of
HFNO in patients with ARF and severe hypoxemia can be recommended.
should be used carefully in high-risk patients
Post extubation is a hazardous period for ICU patients and ARF after planned extubation is associated with remarkably high mortality.^(9)^ In this setting, ARF is related to many factors such as excessive secretions, progressive exhaustion, respiratory muscle weakness, aspiration or fluid overload. A first open label trial comparing HFNO to the use of a Venturi mask for 48 hours postextubation in patients with hypoxemia, but not ARF, led to reports of greater comfort, fewer desaturations, better interface tolerance and a lower reintubation rate with HFNO.^(10)^ In low-risk patients, HFNO applied for 24 hours postextubation compared to a Venturi mask was associated with a significant reduction in the need for reintubation.^(11)^ In high-risk patients, HFNO compared to NIV was not inferior to NIV to prevent reintubation and post extubation ARF.^(12)^
prevent acute respiratory failure in the post-surgical setting
Hypoxemia frequently occurs soon after major surgery,^(13,14)^ and expiratory complications are the second most frequent complications after surgery.^(15)^ In patients undergoing cardiothoracic surgery, NIV has been shown to prevent postoperative ARF. A multicenter randomized controlled trial has compared HFNO to NIV to treat post-extubation ARF or to prevent the occurrence of ARF in patients deemed at-risk.^(16)^ High flow nasal oxygen was not inferior to NIV, with a similar rate of treatment success. ICU mortality was similar in the two groups. A second large randomized controlled trial compared HFNO to standard oxygen in patients undergoing major elective abdominal surgery and deemed at moderate to high risk of postoperative pulmonary complications.^(17)^ The occurrence of postoperative hypoxemia one hour after extubation, of pulmonary complications and the length of hospital stays were similar between the two groups.
The benefit of HFNO in the postoperative setting is not unequivocal. Future studies should aim to identify the subset of patients who are the best candidates to benefit from HFNO.
further investigation
High flow nasal oxygen has been proposed as an alternative to NIV to prevent intubation in hypoxemic de novo ARF in immunocompromised patients. To date, two post hoc analyses of large multicenter randomized controlled trials have been released, with conflicting results.^(18,19)^
High flow nasal oxygen has also been proposed as an alternative to NIV to prevent significant desaturation during endotracheal intubation in patients with severe hypoxemic ARF. Due to population heterogeneity, the results of these studies are conflicting,^(20-23)^ and additional studies targeting more homogeneous populations are needed.
Finally, HFNO may be useful for providing adequate oxygenation and comfort to end-of-life patients with do-not-intubate orders,^(24)^ but further studies are needed.
Clinicians should be aware of the potential harmful effects of HFNO. First, HFNO can mask patient worsening and subsequently delay intubation, which may be harmful. As recently reported by an observational study in de novo ARF patients treated by HFNO, patients intubated after more than 48 hours of treatment had a higher mortality than those intubated within 48 hours.^(25)^ In de novo hypoxemic ARF, HFNO preserves spontaneous breathing, permitting highly negative intrathoracic pressure. Therefore, HFNO can theoretically contribute to lung injury in patients breathing with high drive and large tidal volume.^(26)^ As opposed to NIV, no monitoring of pressures or volume is available for patients breathing with HFNO.
Clinicians willing to use HFNO in ICU patients need to implement the technique cautiously, carefully, and progressively in their unit, as with any new therapy. Identifying the patients who are the most likely to benefit from HFNO is a challenge.
Over the past several years, a growing number of studies have suggested the potential benefit of HFNO in preventing intubation or reintubation in ICU patients who are either admitted for de novo acute respiratory failure or mechanically ventilated for surgery. Although HFNO appears to be a promising therapy in the ICU, additional studies are needed to define more precisely the subgroups of patients who are most likely to benefit from HFNO. Clinicians willing to use HFNO should know that HFNO might have deleterious effects, especially if it is not used adequately. As with any novel therapy, clinicians should learn how to use and implement HFNO progressively and cautiously.