Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Predictors of Intubation in Patients With Acute Hypoxemic Respiratory Failure Treated With a Noninvasive Oxygenation Strategy*

Frat, Jean-Pierre MD1,2,3; Ragot, Stéphanie PhD4,5,6; Coudroy, Rémi MD1,2,3; Constantin, Jean-Michel PhD7,8; Girault, Christophe MD9; Prat, Gwénael MD10; Boulain, Thierry MD11; Demoule, Alexandre PhD12,13; Ricard, Jean-Damien PhD14,15,16; Razazi, Keyvan MD17; Lascarrou, Jean-Baptiste MD18; Devaquet, Jérôme MD19; Mira, Jean-Paul PhD20; Argaud, Laurent PhD21; Chakarian, Jean-Charles MD22; Fartoukh, Muriel PhD23; Nseir, Saad PhD24; Mercat, Alain PhD25; Brochard, Laurent MD26,27; Robert, René PhD1,2,3; Thille, Arnaud W. PhD1,2,3; for the REVA network

doi: 10.1097/CCM.0000000000002818
Clinical Investigations

Objectives: In patients with acute hypoxemic respiratory failure, noninvasive ventilation and high-flow nasal cannula oxygen are alternative strategies to conventional oxygen therapy. Endotracheal intubation is frequently needed in these patients with a risk of delay, and early predictors of failure may help clinicians to decide early. We aimed to identify factors associated with intubation in patients with acute hypoxemic respiratory failure treated with different noninvasive oxygenation techniques.

Design: Post hoc analysis of a randomized clinical trial.

Setting: Twenty-three ICUs.

Patients: Patients with a respiratory rate greater than 25 breaths/min and a Pao2/Fio2 ratio less than or equal to 300 mm Hg.

Intervention: Patients were treated with standard oxygen, high-flow nasal cannula oxygen, or noninvasive ventilation.

Measurement and Main Results: Respiratory variables one hour after treatment initiation. Under standard oxygen, patients with a respiratory rate greater than or equal to 30 breaths/min were more likely to need intubation (odds ratio, 2.76; 95% CI, 1.13–6.75; p = 0.03). One hour after high-flow nasal cannula oxygen initiation, increased heart rate was the only factor associated with intubation. One hour after noninvasive ventilation initiation, a Pao2/Fio2 ratio less than or equal to 200 mm Hg and a tidal volume greater than 9 mL/kg of predicted body weight were independent predictors of intubation (adjusted odds ratio, 4.26; 95% CI, 1.62–11.16; p = 0.003 and adjusted odds ratio, 3.14; 95% CI, 1.22–8.06; p = 0.02, respectively). A tidal volume above 9 mL/kg during noninvasive ventilation remained independently associated with 90-day mortality.

Conclusions: In patients with acute hypoxemic respiratory failure breathing spontaneously, the respiratory rate was a predictor of intubation under standard oxygen, but not under high-flow nasal cannula oxygen or noninvasive ventilation. A Pao2/Fio2 below 200 mm Hg and a high tidal volume greater than 9 mL/kg were the two strong predictors of intubation under noninvasive ventilation.

1CHU de Poitiers, Réanimation Médicale, Poitiers, France.

2INSERM, CIC-1402, équipe 5 ALIVE, Poitiers, France.

3Université de Poitiers, Faculté de Médecine et de Pharmacie de Poitiers, Poitiers, France.

4INSERM, CIC-1402, Biostatistics, Poitiers, France.

5CIC-1402, Poitiers, France.

6Université de Poitiers, Faculté de Médecine et de Pharmacie de Poitiers, Poitiers, France.

7CHU Clermont-Ferrand, Pôle de Médecine Périopératoire, Clermont-Ferrand, France.

8R2D2, EA-7281, Auvergne University, Clermont-Ferrand, France.

9Department of Medical Intensive Care, Normandie Univ, UNIROUEN, EA3830-GRHV, Rouen University Hospital, Rouen, France.

10CHU de la Cavale Blanche, Service de Réanimation Médicale, Brest, France.

11Centre Hospitalier Régional d’Orléans, Réanimation médico-chirurgicale, Orléans, France.

12Groupe Hospitalier Universitaire Pitié Salpêtrière, Service de Pneumologie et Réanimation Médicale, Paris, France.

13Université Pierre et Marie Curie - Paris 6, Paris, France.

14Assistance Publique des Hôpitaux de Paris, Hôpital Louis Mourier, Service de Réanimation Médico-Chirurgicale, F-92700, Colombes, France.

15Université Paris Diderot, UMR IAME 1137, Sorbonne Paris Cité, F-75018, Paris, France.

16INSERM, IAME 1137, F-75018, Paris, France.

17AP-HP, Hôpitaux universitaires Henri Mondor, DHU A-TVB, Service de Réanimation Médicale, Créteil, 94010 France, GRC CARMAS, Créteil, France.

18Centre Hospitalier Départemental de La Roche sur Yon, Service de Réanimation Polyvalente, La Roche sur Yon, France.

19Hôpital Foch, Réanimation Polyvalente, Suresnes, France.

20Assistance Publique - Hôpitaux de Paris, Groupe Hospitalier Universitaire de Paris Centre, Hôpital Cochin, Réanimation Médicale, and Université Paris Descartes, Paris, France.

21Hospices Civils de Lyon, Groupement Hospitalier Universitaire Edouard Herriot, Service de Réanimation Médicale, Lyon, France.

22Centre Hospitalier de Roanne, Réanimation Polyvalente, Roanne, France.

23Assistance Publique–Hôpitaux de Paris, Hôpital Tenon, Service de Réanimation, Paris, France.

24CHU de Lille, Centre de Réanimation, Université de Lille, Faculté de Médecine, Lille, France.

25CHU Angers, Département de Réanimation Médicale et Médecine Hyperbare, Angers, France.

26Keenan Research Centre and Critical Care Department, St Michael’s Hospital, Toronto, ON, Canada.

27Interdepartmental Division of Critical Care Medicine, University of Toronto, Toronto, ON, Canada.

*See also p. 330.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (

This study was coordinated at Réanimation Médicale, CHU de Poitiers, Poitiers, France and performed in 23 centers in France and Belgium.

Dr. Frat reports grants, personal fees, and nonfinancial support from the “Fisher & Paykel HealthCare” firm, during the conduct of the study; personal fees and nonfinancial support from SOS oxygène, outside the submitted work. Dr. Coudroy reports nonfinancial support from MSD, outside the submitted work. Dr. Girault reports nonfinancial support and other from Fisher & Paykel Healthcare, during the conduct of the study; personal fees from Fisher & Paykel Healthcare, outside the submitted work. Dr. Demoule reports personal fees from Covidien, grants and personal fees from Maquet, grants from Philips, personal fees from MSD, nonfinancial support from Dräger, outside the submitted work. Dr. Ricard reports coverage by Fisher & Paykel Healthcare of expenses to attend scientific meetings. Dr. Mercat reports personal fees from Faron Pharmaceuticals, personal fees from Air Liquide Medical Systems, grants and personal fees from Fisher-Paykel, grants and personal fees from Covidien, outside the submitted work. In addition, Dr. Mercat has a patent General Electric licensed. Dr. Brochard reports grants and nonfinancial support from Fisher & Paykel, grants and nonfinancial support from Covidien, nonfinancial support from Philips, nonfinancial support from Maquet, nonfinancial support from General Electric, grants from Air Liquide, outside the submitted work. Dr. Robert reports nonfinancial support from Fresenius Medical Care, from Baxter Gambro, outside the submitted work. The remaining authors have disclosed that they do not have any potential conflicts of interest.

For information regarding this article, E-mail:

Copyright © 2017 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.