To evaluate the feasibility and effect of invasive mechanical ventilation followed by high flow nasal cannula oxygen therapy (HFNC) in AECOPD patients.

AECOPD patients who underwent invasive mechanical ventilation were enrolled in the study. When pulmonary infection was significantly controlled (the period of pulmonary infection control was called PIC window), the patients were randomized into the invasive mechanical ventilation followed by non-invasive mechanical ventilation therapy group (NIV group) and the invasive mechanical ventilation followed by HFNC therapy group (HFNC group). The collected data included: general information, APACHEⅡ score, SOFA score, CPIS score, the levels of inflammatory markers (T, WBC, PCT, hs-CRP) in PIC window, respiratory (RR, PaO₂, PaCO₂, P/F, breathing comfort score) and circulatory indicators(MAP, HR, Lac) at the start of PIC window and two hours after extubation, the incidence of re-intubation rate in 48 hours, the incidence of facial stress injury, and hospital mortality.

(1) 73 patients were enrolled in the study, 38 patients in the NIV group and 35 patients in the HFNC group. General conditions, the scores of APACHEⅡ, SOFA, and CPIS were similar between the two groups. (2) There was no significant difference between the two groups regarding T, WBC, PCT, hs-CRP, MAP, HR, Lac, RR, PaO₂, PaCO₂, P/F, breathing comfort score, R, Cstat, and PEEPi-stat at the beginning of PIC window. (3) Two hours after extubation the HR, RR and PaCO₂ in the HFNC group were significantly lower than those of the NIV group [(85.42±11.80) times/min vs (99.38±11.01) times/min, t=3.717, P=0.001; (21.26±5.23) times/min vs (26.88±9.26) times/min, t=2.254, P=0.033; (48.14±5.51) mmHg vs (51.48±4.32) mmHg, t=2.057, P=0.047]. The comfort score after sequential treatment for two hours in the HFNC group was (3.92±0.79), which was significantly higher than that in the NIV group (1.83±0.57) (t=-7.358, P=0.000). Moreover, the incidence of facial stress injury in the HFNC group was considerably lower than that in the NIV group (0 vs 21.05%, χ²=8.275, P=0.004). There was no significantly difference between the two groups regarding the incidence of reintubation in 48 hours, the length of ICU stay and the hospital mortality (P>0.05).

The therapeutic effect of invasive mechanical ventilation followed by HFNC are comparable with that of the invasive mechanical ventilation followed by non-invasive mechanical ventilation regarding oxygenation level, the incidence of reintubation in 48 hours, the length of ICU stay, and hospital mortality. Noticeably, the patients using HFNC has significantly lower level of PaCO₂, incidence of facial stress injury and respiratory comfort score than those using invasive mechanical ventilation following non-invasive mechanical ventilation.