• Andrea Cortegiani, Federico Longhini, Annalisa Carlucci, Raffaele Scala, Paolo Groff, Andrea Bruni, Eugenio Garofalo, Maria Rita Taliani, Uberto Maccari, Luigi Vetrugno, Enrico Lupia, Giovanni Misseri, Vittoria Comellini, Antonino Giarratano, Stefano Nava, Paolo Navalesi, and Cesare Gregoretti.
• Department of Surgical, Oncological and Oral Science (Di.Chir.On.S.), Section of Anesthesia, Analgesia, Intensive Care and Emergency, Policlinico Paolo Giaccone, University of Palermo, Via del vespro 129, 90127, Palermo, Italy. andrea.cortegiani@unipa.it.
• Trials. 2019 Jul 22; 20 (1): 450.

BackgroundNoninvasive ventilation (NIV) is indicated to treat respiratory acidosis due to exacerbation of chronic obstructive pulmonary disease (COPD). Recent nonrandomized studies also demonstrated some physiological effects of high-flow nasal therapy (HFNT) in COPD patients. We designed a prospective, unblinded, multicenter, randomized controlled trial to assess the noninferiority of HFNT compared to NIV with respect to the reduction of arterial partial pressure of carbon dioxide (PaCO2) in patients with hypercapnic acute respiratory failure with mild-to-moderate respiratory acidosis.MethodsWe will enroll adult patients with acute hypercapnic respiratory failure, as defined by arterial pH between 7.25 and 7.35 and PaCO2 ≥ 55 mmHg. Patients will be randomly assigned 1:1 to receive NIV or HFNT. NIV will be applied through a mask with a dedicated ventilator in pressure support mode. Positive end-expiratory pressure will be set at 3-5 cmH2O with inspiratory support to obtain a tidal volume between 6 and 8 ml/kg of ideal body weight. HFNT will be initially set at a temperature of 37 °C and a flow of 60 L/min. At 2 and 6 h we will assess arterial blood gases, vital parameters, respiratory rate, treatment intolerance and failure, need for endotracheal intubation, time spent under mechanical ventilation (both invasive and NIV), intensive care unit and hospital length of stay, and hospital mortality. Based on an α error of 5% and a β error of 80%, with a standard deviation for PaCO2 equal to 15 mmHg and a noninferiority limit of 10 mmHg, we computed a sample size of 56 patients. Considering potential drop-outs and nonparametric analysis, the final computed sample size was 80 patients (40 per group).DiscussionHFNT is more comfortable than NIV in COPD patients recovering from an episode of exacerbation. If HFNT would not be inferior to NIV, HFNT could be considered as an alternative to NIV to treat COPD patients with mild-to-moderate respiratory acidosis.Trial RegistrationClinicalTrials.gov, NCT03370666 . Registered on December 12, 2017.

32 keywords...

hide…