Noninvasive Ventilatory Support As First-Line Treatment for COVID-19 Critically Ill Patients with Acute Hypoxemic Respiratory Failure

Background: Although widely applied, noninvasive ventilatory support (NIVS) efficacy in COVID-19 disease is unknown. Early identification of treatment failure is warranted to avoid delays in endotracheal intubation and protective ventilation. We conducted a study to determine the rate of and factors associated to NIVS failure in critically ill patients with COVID-19 disease, and to compare NIVS failure rate in COVID-19 patients to that of a matched cohort with hypoxemic respiratory failure of other origins. Methods: All consecutive patients receiving first-line treatment NIVS for hypoxemic respiratory failure due to COVID-19 in the ICU of a University Hospital in Italy up to April 20th, 2020, were studied: laboratory data were collected on arrival, 28-day outcome was recorded. After one-to-one propensity score matching based on simplified acute physiology (SAPS) II score, age, PaO2/FiO2 and PaCO2 at arrival, NIVS failure rate in COVID-19 patients was compared to a previously published cohort who received NIVS during hypoxemic respiratory failure from other causes. Results: Eighty-five patients received first-line treatment with NIVS, mainly with helmet noninvasive ventilation and high-flow nasal cannula. Fifty-two patients (61%) needed endotracheal intubation. Independent predictors of NIVS failure were SAPSII score (adjusted hazard ratio 1.039 [1.018-1.061], <0.001), and serum lactate de-hydrogenase at enrolment (adjusted hazard ratio: 1.002 [1.000-1.004], p=0.013): 88% of patients with SAPSII≥33 and serum lactate de-hydrogenase≥405 Units/Liter needed endotracheal intubation vs. 32% of patients with SAPS<33 and lactate de-hydrogenase<405 Units/Liter. In the propensity-matched cohorts (108 patients), COVID-19 patients showed higher risk of NIVS failure than other causes of hypoxemic respiratory failure (59% vs. 35%, p=0.02), with an adjusted hazard ratio of 2 (CI95% [1.1-3.6], p=0.01). Conclusions: COVID-19 patients receiving first-line NIVS are burdened by high risk of needing endotracheal intubation: this appears greater than that of patients affected by hypoxemic respiratory failure of other origins. In order to not delay endotracheal intubation, if logistically sustainable, NIVS use in the intensive care unit should be avoided in severe patients (SAPSII score≥33) with serum lactate de-hydrogenase≥405 Units/Liter.