RCT: High-Flow Nasal Oxygen Noninferior to Noninvasive Ventilation for Most Acute Respiratory Failure Causes

Background: Acute respiratory failure (ARF) arises from diverse etiologies and can manifest as hypoxemic or hypercapnic events. High-flow nasal oxygen (HFNO) and noninvasive ventilation (NIV) are common noninvasive respiratory support modalities, but robust comparative data in various ARF subgroups have been limited. Prior research suggests NIV may benefit chronic obstructive pulmonary disease (COPD) exacerbations and acute cardiogenic pulmonary edema (ACPE), yet for hypoxemic failure (including COVID-19 and immunocompromised populations), HFNO is often favored for its comfort and physiological advantages. The RENOVATE trial was designed to assess whether HFNO is noninferior to NIV for preventing intubation or death among five distinct groups of patients with ARF.

Objective: To determine if HFNO is noninferior to NIV in terms of the composite outcome of endotracheal intubation or death within seven days in patients with ARF, categorized into five subgroups: (1) nonimmunocompromised with hypoxemic ARF, (2) immunocompromised with hypoxemic ARF, (3) COPD exacerbation with respiratory acidosis, (4) ACPE, and (5) hypoxemic COVID-19.

Methods: This multicenter, adaptive, noninferiority randomized clinical trial enrolled 1800 hospitalized adults across 33 Brazilian centers. Patients were stratified by ARF etiology and randomized 1:1 to receive either HFNO or NIV. Treatment protocols allowed HFNO escalation to NIV (particularly for COPD or ACPE) if needed. The primary outcome was defined using a Bayesian hierarchical model with dynamic borrowing across subgroups; noninferiority was met if the posterior probability for an odds ratio (OR) below 1.55 reached ≥0.992. Predefined futility and superiority thresholds guided interim analyses, with a maximum sample size of 2000.

Results: Of 1800 randomized patients, 1766 completed the study (mean age 64 years; 40% women). The primary outcome (intubation or death by day 7) occurred in 39.0% (HFNO) vs 38.1% (NIV). HFNO was noninferior in four subgroups:

• Nonimmunocompromised with hypoxemia: 32.5% vs 33.1% (OR 1.02; posterior probability of noninferiority 0.999).
• COPD exacerbation with respiratory acidosis: 28.6% vs 26.2% (OR 1.05; probability 0.992).
• ACPE: 10.3% vs 21.3% (OR 0.97; probability 0.997).
• Hypoxemic COVID-19: 51.3% vs 47.0% (OR 1.13; probability 0.997).

The immunocompromised subgroup stopped enrollment early for futility; final results there did not confirm noninferiority (57.1% vs 36.4%; OR 1.07; probability 0.989). No significant differences in 28- or 90-day mortality emerged, although mortality rates were generally higher than in some previous trials. Comfort scores favored HFNO, and rates of serious adverse events were similar between groups.

Conclusions: In four of five ARF subgroups, HFNO met predefined noninferiority criteria compared with NIV regarding endotracheal intubation or death at seven days. However, immunocompromised patients with hypoxemic ARF remain an area of uncertainty, as do smaller subgroups (e.g., COPD) under non-borrowing analyses. Clinicians may consider HFNO as an alternative initial approach, recognizing that rescue NIV may still be necessary, particularly in COPD exacerbations.

Implications for Practice: These findings support using HFNO for a broad range of ARF etiologies as a first-line therapy. HFNO’s ease of use, patient comfort, and comparable safety profile may make it especially appealing. Nevertheless, clinicians should remain vigilant in immunocompromised patients and in COPD exacerbations when hypercapnia is pronounced. Potential cost variations between HFNO and NIV may influence real-world adoption, and local resources, staff expertise, and patient tolerance should guide final decisions.

Study Strengths and Limitations: Strengths include a large, diverse sample and a robust Bayesian adaptive design that allowed dynamic borrowing across subgroups. This approach increased precision but also introduced heterogeneity concerns. Some patient groups (particularly immunocompromised and COPD) were relatively small, limiting definitive conclusions in those strata. Additionally, early stopping for futility in immunocompromised patients curtailed full enrollment, and the trial compared HFNO only with face-mask NIV (rather than alternatives such as helmet CPAP).

Future Research: Further large-scale studies should refine whether HFNO can supplant NIV in COPD exacerbations and immunocompromised populations. Investigations on cost-effectiveness, patient-centered outcomes (comfort, quality of life), and comparative models (e.g., helmet NIV) are also warranted.

Reference:
• RENOVATE Investigators and the BRICNet Authors. High-Flow Nasal Oxygen vs Noninvasive Ventilation in Patients With Acute Respiratory Failure: The RENOVATE Randomized Clinical Trial. JAMA. Published online December 10, 2024. DOI: http://doi.org/10.1001/jama.2024.26244
• Frat JP, Le Pape S, Thille AW. Editorial: Is High-Flow Oxygen the Standard for All Patients With Acute Respiratory Failure? JAMA. Published online December 10, 2024. DOI: http://doi.org/10.1001/jama.2024.25906
• Freund Y, Vromant A. Editorial: Reevaluating Respiratory Support in Acute Respiratory Failure—Insights From the RENOVATE Trial and Implications for Practice. JAMA. Published online December 10, 2024. DOI: http://doi.org/10.1001/jama.2024.25869