Humidified High-Flow Nasal Cannula Oxygen—More Than Just Supplemental Oxygen*

    loading  Checking for direct PDF access through Ovid

Excerpt

Oxygen supplementation through nasal cannulae is a common therapy for hypoxemic patients. Conventional cannulae can provide up to 8 L/min of oxygen, and wider bore (so called “high flow”) cannulae can deliver up to 15 L/min. A bubble humidifier is often added with higher flows. More recently, cannulae and flow meters capable of up to 60 L/min of oxygen flow have been introduced. Flows at this rate always require heat and humidity and are thought to provide physiologic benefits beyond simple increases in oxygen content (see below). Despite these differences, the term “high flow” oxygen therapy is also used to describe these newer systems leading to confusion in differentiating them from the simple 8–15 L/min systems described above. Because of this, some have recommended the term “humidified high-flow nasal cannula (HHFNC)” oxygen therapy to describe devices providing heated and humidified oxygen at 20–60 L/min, a term I will employ here.
HHFNC oxygen therapy approximates normal inspiratory flow and is quite comfortable. Because of this flow, anatomic dead spaces in the naso and oropharynx are flushed with fresh gas reducing functional dead space. Turbulent (not laminar) gas flow also develops in the posterior pharynx which may also facilitate gas mixing. Together these effects may reduce ventilation demands. Depending upon oronasal anatomy and mouth opening, HHFNC oxygen can also generate a certain level of CPAP in the posterior pharynx. This has been estimated to be as high as 5–8 cm H2O under certain circumstances. This CPAP effect may prevent alveolar derecruitment, improve respiratory system mechanics, reduce the breath initiation muscle loads associated with intrinsic positive end-expiratory pressure (PEEP), and perhaps help unload ventilatory muscles during inspiration (1–3).
In this issue of Critical Care Medicine, Delorme et al (4) use esophageal pressure measurements to demonstrate in 12 patients with both hypercapneic and hypoxemic respiratory failure that HHFNC oxygen compared with standard low-flow nasal cannula oxygen therapy significantly reduces ventilatory muscle loading. Despite these interesting results, it is difficult with the data reported to sort out the mechanism(s) of this reduced effort. Indeed, with multiple potential mechanisms in play and only 12 patients with a wide range of respiratory mechanics studied, it is not surprising that it would difficult to identify mechanism(s) that may be important in one type of patient but which may be irrelevant in another.
In these patients, overall minute ventilation demand did not appear to change. However, there was a trend toward lower PCO2 levels suggesting that a reduced dead space may have played a role in some patients. Further assessments of functional dead space and gas mixing were not performed, and thus the role of these effects in individual patients cannot be determined.
Evaluating the CPAP effect in these patients is also difficult because airway opening pressure was not measured. Accurate measurements of transpulmonary pressure at end inspiration and end expiration thus cannot be made, and the calculations of compliance in this study (4) can be called into question. End-expiratory lung volumes were also not assessed. Further sorting out the effects of CPAP/PEEP on mechanics and possible intrinsic PEEP muscle loading is therefore made even more difficult. Nevertheless, the authors argue that a PEEP effect was unlikely as the overall end-expiratory esophageal pressure did not change. However, the authors go on to also argue that an increase in end-expiratory lung volume from HHFNC oxygen might have improved compliance and reduced mechanical loading.
At the end of the day, it is likely that many of the proposed mechanisms associated with HHFNC oxygen therapy were operational to varying degrees in reducing muscle loading in these patients. However, we must await future studies to better characterize these phenomenon.
    loading  Loading Related Articles