High-flow oxygen therapy performs as well as non-invasive ventilation

“The Renovate trial addresses a debate that has been going on for several years on the choice of high-flow oxygen therapy or non-invasive ventilation for patients in respiratory distress,” reports Professor Marc Leone, from the North Hospital of Assistance publique-Hôpitaux de Marseille (AP-HM), vice-president of the French Society of Anesthesia and Resuscitation (Sfar). It appears from this Brazilian study that high flow is equivalent to non-invasive ventilation and is suitable for all patients, with the exception of those hypoxic and immunocompromised for whom NIV will be more beneficial. “The results indicate that there is no additional risk, apart from for the immunocompromised, in putting patients on high flow,” analyzes Professor Leone, who did not participate in Renovate.

The study sought to evaluate whether high-flow oxygen therapy (HTO) was non-inferior to non-invasive ventilation (NIV) on the rate of endotracheal intubation or death at 7 days in patients with respiratory distress. Five groups were concerned: non-immune deficient hypoxemic patients; immunodeficient hypoxemics; with exacerbation of chronic obstructive pulmonary disease (COPD) with respiratory acidosis; with cardiogenic pulmonary edema (CPO); and hypoxemic with Covid-19 (these patients were added after the start of inclusion in November 2019).

A more comfortable option

Appearing more than ten years ago, OHD particularly developed during the Covid-19 period and is commonly used as a supportive therapy in cases of respiratory distress. “ High-flow oxygen therapy, compared to conventional oxygen therapy, provides warmed and humidified oxygen. It has the advantage of recruiting alveoli in the lung”adds the anesthetist-resuscitator. OHD has other physiological advantages, according to the Renovate authors, being more comfortable for the patient than NIV. They also mention the possibility of “eat, drink and speak more easily than with NIV”. These results are published in the Jama.

This randomized non-inferiority trial included 1,766 adult patients classified into five respiratory distress groups in 33 Brazilian hospitals. Patients were placed either on OHD (n = 883) or on NIV (n = 883), and the non-inferiority margin was set between 0.992 and 1.55. Overall, considering all patients, 39% of the OHD group were intubated or died within 7 days compared to 38.1% of the NIV group. The incidence of severe adverse events was similar in the OHD and NIV groups. The secondary endpoint was the 28-day mortality rate and the authors found no difference between OHD and NIV. Finally, the tertiary outcome was patient comfort, which was higher in the OHD group compared to the NIV group.

Confirmation required for certain subgroups

OHD was non-inferior to NIV for non-immune deficient hypoxemic patients (32.5% intubation or death within 7 days versus 33.1%, non-inferiority: 0.999), with COPD exacerbation (28, 6 versus 26.2%, non-inferiority: 0.992), with cardiogenic pulmonary edema (10.3 versus 21.3%, non-inferiority: 0.997) and hypoxemic with Covid-19 (51.3 versus 47%, non-inferiority: 0.997). On the other hand, OHD was lower than NIV in immunocompromised hypoxemic patients: the authors found 57.1% intubation or death within 7 days compared to 36.4% in the NIV group (non-inferiority: 0.989) .

However, the authors note that the small number of patients in some groups calls for future studies in COPD (n = 77), immunocompromised (n = 50), and OPC (n = 272) patients. Moreover, “quite surprisingly, subgroup comparisons even show that high flow is slightly better than NIV in hypercapnic patients, that is to say those with COPD or OPC, in whom NIV is usually favorite »comments Professor Leone. For the Brazilian team, “if the recommendations today indicate the use of NIV for respiratory distress caused by exacerbations of COPD and cardiogenic pulmonary edema, they were in fact based on a comparison of NIV with conventional oxygen therapy” . Moreover, “ OHD is preferred over low flow to treat patients with an acute hypoxemic episode, also including the immunocompromised and Covid-19”.

Valuable information for practice

“This is not the first study to present such results, continues Professor Leone. Thus, for teams mastering both techniques, the choice can be made based on patient comfort and tolerance. For others, there is no danger in putting a patient in respiratory distress on high-flow oxygen therapy. » NIV is in fact a more technical procedure which requires expertise as well as medical and paramedical time. “In addition, for the patient, NIV can be more uncomfortable and impose variations in respiratory pressure”he adds.

Two editorials published in the Jamasigned by the French, advocate caution. In the first, Professor Yonathan Freund and Dr Amélie Vromant (AP-HP) highlight the small sample size for patients with COPD and those with heart failure and raise “the uncertain role of OHD in hypercapnic patients (such as COPD) due to the not fully known physiological effects on the elevation of PaCO levels₂ », despite the four studies cited in the article. The two Parisian specialists regret that Renovate was not more patient-centered in understanding the comfort criterion of these two techniques.

For their part, the authors of the other editorial (3), Professors Jean-Pierre Frat and Arnaud Thille, and Dr Sylvain Le Pape, all three from Poitiers University Hospital, tempered: “Rather than suggesting that OHD can universally replace NIV for all etiologies of respiratory distress,” these results indicate “that initiating treatment with OHD is generally not harmful.” According to them, for clinical practice, “this interpretation provides information giving clinicians time to make a diagnosis and find the cause of respiratory distress before choosing the most appropriate option”; they thus qualify OHD as “safe transition therapy”.

(1) Renovate Investigators, Jama, December 2024.
DOI: 10.1001/jama.2024.26244
(2) Y. Freund et al., Jama, December 2024.
DOI: 10.1001/jama.2024.25869
(3) J.-P. Frat et al, Jama, December 2024.
DOI: 10.1001/jama.2024.25906