• Anahita Rouzé, Ignacio Martin-Loeches, Pedro Povoa, Demosthenes Makris, Antonio Artigas, Mathilde Bouchereau, Fabien Lambiotte, Matthieu Metzelard, Pierre Cuchet, Claire Boulle Geronimi, Marie Labruyere, Fabienne Tamion, Martine Nyunga, Charles-Edouard Luyt, Julien Labreuche, Olivier Pouly, Justine Bardin, Anastasia Saade, Pierre Asfar, Jean-Luc Baudel, Alexandra Beurton, Denis Garot, Iliana Ioannidou, Louis Kreitmann, Jean-François Llitjos, Eleni Magira, Bruno Mégarbane, David Meguerditchian, Edgar Moglia, Armand Mekontso-Dessap, Jean Reignier, Matthieu Turpin, Alexandre Pierre, Gaetan Plantefeve, Christophe Vinsonneau, Pierre-Edouard Floch, Nicolas Weiss, Adrian Ceccato, Antoni Torres, Alain Duhamel, Saad Nseir, and coVAPid study Group.
• CHU de Lille, Médecine Intensive-Réanimation, 59000, Lille, France.
• Intensive Care Med. 2021 Feb 1; 47 (2): 188-198.

PurposeAlthough patients with SARS-CoV-2 infection have several risk factors for ventilator-associated lower respiratory tract infections (VA-LRTI), the reported incidence of hospital-acquired infections is low. We aimed to determine the relationship between SARS-CoV-2 pneumonia, as compared to influenza pneumonia or no viral infection, and the incidence of VA-LRTI.MethodsMulticenter retrospective European cohort performed in 36 ICUs. All adult patients receiving invasive mechanical ventilation > 48 h were eligible if they had: SARS-CoV-2 pneumonia, influenza pneumonia, or no viral infection at ICU admission. VA-LRTI, including ventilator-associated tracheobronchitis (VAT) and ventilator-associated pneumonia (VAP), were diagnosed using clinical, radiological and quantitative microbiological criteria. All VA-LRTI were prospectively identified, and chest-X rays were analyzed by at least two physicians. Cumulative incidence of first episodes of VA-LRTI was estimated using the Kalbfleisch and Prentice method, and compared using Fine-and Gray models.Results1576 patients were included (568 in SARS-CoV-2, 482 in influenza, and 526 in no viral infection groups). VA-LRTI incidence was significantly higher in SARS-CoV-2 patients (287, 50.5%), as compared to influenza patients (146, 30.3%, adjusted sub hazard ratio (sHR) 1.60 (95% confidence interval (CI) 1.26 to 2.04)) or patients with no viral infection (133, 25.3%, adjusted sHR 1.7 (95% CI 1.2 to 2.39)). Gram-negative bacilli were responsible for a large proportion (82% to 89.7%) of VA-LRTI, mainly Pseudomonas aeruginosa, Enterobacter spp., and Klebsiella spp.ConclusionsThe incidence of VA-LRTI is significantly higher in patients with SARS-CoV-2 infection, as compared to patients with influenza pneumonia, or no viral infection after statistical adjustment, but residual confounding may still play a role in the effect estimates.

23 keywords...

hide…