Article Text

Download PDFPDF

Oxygen therapy in the breathless patient
  1. B D Fox
  1. Dr B D Fox, Pulmonary Institute, Rabin Medical Center (Beilinson Campus), Petach Tiqwa 49100, Israel; benfox{at}

Statistics from

Beasley and colleagues argue persuasively that oxygen delivery to hypoxaemic patients should be optimised to provide adequate oxygen delivery while reducing the adverse effects of hyperoxaemia and preventing delay in identifying a patient with deteriorating gas exchange.1 However, their consistent use of the term high-flow oxygen instead of high-concentration oxygen perpetuates another widespread misunderstanding regarding oxygen therapy—that oxygen flow to the patient has a consistent and predictable relationship to the fractional inspired oxygen concentration (Fio2) delivered to the alveoli. This incorrect assumption threatens to undermine the wisdom and potential benefits of their insightful editorial.

When oxygen is delivered by nasal cannula, Hudson mask or reservoir bag mask, the inspiratory flow generated by the patient will generally exceed the oxygen flow delivered.2 Room air is entrained by the patient and thus the inspired oxygen is diluted. During the expiratory phase, oxygen flow continues and has a variable and unpredictable effect of flushing exhaled gases from the device and filling the upper airways with high concentration oxygen. Therefore, as respiratory rate, inspiratory flow and tidal volume change, so does the Fio2 arriving in the patient’s alveoli. These devices are referred to as “variable performance devices”. Venturi systems blend oxygen and gas at a fixed ratio and the total gas flow delivered to the patient usually exceeds inspiratory flow when Fio2 is <40% (fig 1). These devices, along with gas blenders and mechanical ventilators, are “fixed performance”.

Figure 1 Composition of total gas flows through a typical Venturi valve.

In the case example shown on page 841 of the editorial by Beasley et al,1 the theoretical patient is provided in part (a) with an Fio2 of 0.3 (2–3 litres via nasal cannula) and subsequently (b) with an Fio2 of 0.6 (8–10 litres via a Hudson mask). These numbers are erroneous since both employ a variable performance device.



  • Competing interests: None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Letters
    R Beasley