Objective:To investigate the effect and tolerance of non-invasive ventilation (NIV) with helmet in patients with respiratory failure caused by acute exacerbation of chronic obstructive pulmonary disease (AECOPD) and the effect on improving blood gas, alleviating dyspnea and the occurrence of complications.Methods:Patients with AECOPD and respiratory failure admitted to emergency intensive care unit (EICU) and respiratory intensive care unit (RICU) of the First Affiliated Hospital of Zhengzhou University from January 1st, 2018 to May 31st, 2019 were enrolled. After obtaining the informed consent of the patients or their authorized family members, the patients were divided into two groups: the helmet group and the facial mask group by random number table. NIV was carried out by using helmet or facial mask, respectively. During the course of NIV (immediately, 1 hour, 4 hours and at the end of NIV), the tolerance score, blood gas analysis, heart rate (HR), respiratory rate (RR) of patients were monitored, and the incidence of tracheal intubation, in-hospital mortality and complications were observed. Kaplan-Meier survival curve was plotted to analyze the 30-day cumulative survival of the two groups.Results:A total of 82 patients with AECOPD and respiratory failure were included during the study period. After excluding patients with the oxygenation index (PaO 2/FiO 2) > 200 mmHg (1 mmHg = 0.133 kPa), with tracheal intubation or invasive ventilation, suffering from acute myocardial infarction, severe trauma within 2 weeks, excessive secretion, sputum discharge disorder or refusal to participate in the study, 26 patients were finally enrolled in the analysis, randomly assigned to the helmet group and the facial mask group, with 13 patients in each group. The PaO 2/FiO 2 after NIV of patients in both groups was increased significantly as compared with that immediately after NIV, without significant difference between the two groups, but the increase in PaO 2/FiO 2 at the end of NIV compared with immediately after NIV in the helmet group was significantly higher than that in the facial mask group (mmHg: 75.1±73.2 vs. 7.7±86.0, P < 0.05). RR at each time point after NIV in the two groups was lower than that immediately after NIV, especially in the helmet group. There were significant differences between the helmet group and facial mask group at 1 hour, 4 hours, and the end of NIV (times/min: 17.5±4.1 vs. 23.1±6.3 at 1 hour, 16.2±2.5 vs. 20.0±5.5 at 4 hours, 15.5±2.5 vs. 21.2±5.9 at the end of NIV, all P < 0.05). The NIV tolerance score of the helmet group at 4 hours and the end was significantly higher than that of the facial mask group (4 hours: 3.9±0.3 vs. 3.3±0.9, at the end of NIV: 3.8±0.6 vs. 2.9±0.9, both P < 0.05). There was no significant difference in the improvement of pH value, arterial partial pressure of carbon dioxide (PaCO 2), or HR between helmet group and facial mask group. The total number of complications (cases: 3 vs. 8) and the nasal skin lesions (cases: 0 vs. 4) in the helmet group were significantly less than those in the facial mask group (both P < 0.05). Only 2 patients in the helmet group received endotracheal intubation, and 1 of them died; 5 patients in the facial mask group received endotracheal intubation, and 3 of them died; there was no significant difference between the two groups (both P > 0.05). The Kaplan-Meier survival curve analysis showed that the cumulative survival rate of 30 days in the helmet group was lower than that in the facial mask group, but the difference was not statistically significant (Log-Rank test: χ 2 = 1.278, P = 0.258). Conclusion:NIV with helmet has better comfort for patients with AECOPD combined with respiratory failure, and better effect on improving oxygenation and relieving dyspnea, and its effect on carbon dioxide emissions is not inferior to that of traditional mask NIV.