Acute hypoxemic respiratory failure in immunocompromised patients: the Efraim multinational prospective cohort study

Elie Azoulay, Peter Pickkers, Marcio Soares, Anders Perner, Jordi Rello, Philippe R. Bauer, Andry van de Louw, Pleun Hemelaar, Virginie Lemiale, Fabio Silvio Taccone, Ignacio Martin Loeches, Tine Sylvest Meyhoff, Jorge Salluh, Peter Schellongowski, Katerina Rusinova, Nicolas Terzi, Sangeeta Mehta, Massimo Antonelli, Achille Kouatchet, Andreas Barratt-Due, Miia Valkonen, Precious Pearl Landburg, Fabrice Bruneel, Ramin Brandt Bukan, Frédéric Pène, Victoria Metaxa, Anne Sophie Moreau, Virginie Souppart, Gaston Burghi, Christophe Girault, Ulysses V. A. Silva, Luca Montini, François Barbier, Lene B. Nielsen, Benjamin Gaborit, Djamel Mokart, Sylvie Chevret, for the Efraim investigators and the Nine-I study group

In immunocompromised patients with acute hypoxemic respiratory failure (ARF), initial management aims primarily to avoid invasive mechanical ventilation (IMV). To assess the impact of initial management on IMV and mortality rates, we performed a multinational observational prospective cohort study in 16 countries (68 centers). A total of 1611 patients were enrolled (hematological malignancies 51.9%, solid tumors 35.2%, systemic diseases 17.3%, and solid organ transplantation 8.8%). The main ARF etiologies were bacterial (29.5%), viral (15.4%), and fungal infections (14.7%), or undetermined (13.2%). On admission, 915 (56.8%) patients were not intubated. They received standard oxygen (N = 496, 53.9%), high-flow oxygen (HFNC, N = 187, 20.3%), noninvasive ventilation (NIV, N = 153, 17.2%), and NIV + HFNC (N = 79, 8.6%). Factors associated with IMV included age (hazard ratio = 0.92/year, 95% CI 0.86-0.99), day-1 SOFA (1.09/point, 1.06-1.13), day-1 PaO2/FiO2 (1.47, 1.05-2.07), ARF etiology (Pneumocystis jirovecii pneumonia (2.11, 1.42-3.14), invasive pulmonary aspergillosis (1.85, 1.21-2.85), and undetermined cause (1.46, 1.09-1.98). After propensity score matching, HFNC, but not NIV, had an effect on IMV rate (HR = 0.77, 95% CI 0.59-1.00, p = 0.05). ICU, hospital, and day-90 mortality rates were 32.4, 44.1, and 56.4%, respectively. Factors independently associated with hospital mortality included age (odds ratio = 1.18/year, 1.09-1.27), direct admission to the ICU (0.69, 0.54-0.87), day-1 SOFA excluding respiratory score (1.12/point, 1.08-1.16), PaO2/FiO2 < 100 (1.60, 1.03-2.48), and undetermined ARF etiology (1.43, 1.04-1.97). Initial oxygenation strategy did not affect mortality; however, IMV was associated with mortality, the odds ratio depending on IMV conditions: NIV + HFNC failure (2.31, 1.09-4.91), first-line IMV (2.55, 1.94-3.29), NIV failure (3.65, 2.05-6.53), standard oxygen failure (4.16, 2.91-5.93), and HFNC failure (5.54, 3.27-9.38). HFNC has an effect on intubation but not on mortality rates. Failure to identify ARF etiology is associated with higher rates of both intubation and mortality. This suggests that in addition to selecting the appropriate oxygenation device, clinicians should strive to identify the etiology of ARF.