Dexmedetomidine for sedation during noninvasive ventilation in COVID-19 patients from the Helmet-COVID randomized clinical trial post-hoc analysis.

Dexmedetomidine may be used in patients managed by noninvasive ventilation (NIV), but its associations with intubation, mortality and health-related quality of life (HRQoL) are unclear. Thus, we aimed to assess the use of dexmedetomidine in patients with COVID-19-induced acute hypoxemic respiratory failure requiring NIV and its associations with the 28-day intubation rate, 28- and 180-day mortality rate, and HRQoL. This post-hoc study of the Helmet-COVID trial analyzed data from NIV-managed patients who did and did not receive dexmedetomidine and assessed the association of dexmedetomidine with clinical outcomes. We evaluated HRQoL on day 180 using EuroQoL-5D-5L index values and the EQ visual analog scale (EQ-VAS). Non-survivors were assigned a value of 0 on the EQ-5D-5L and EQ-VAS. We used a generalized linear mixed model adjusted for the enrollment center and other baseline covariables. Of 290 patients managed with NIV, 107 (36.8%) were treated with dexmedetomidine. The median highest daily dose was 0.6 mcg/kg/h (interquartile range [IQR] 0.4, 0.9) for a median of 2.5 days (IQR: 1, 5). Baseline demographics and severity of respiratory failure, including the PaO2:FiO2 and degree of hypercapnia were largely comparable between both groups. During the ICU stay, dexmedetomidine-treated patients were more likely to have received helmet NIV (70 of 107, 65.4%), awake proning (45 of 107, 42.1%), vasopressors and renal replacement therapy (57.9 and 17.8%, respectively). Dexmedetomidine-treated patients had higher ICU and hospital mortality, higher intubation rate, and fewer ICU-free days compared to those not receiving dexmedetomidine; differences that could not be fully explained by confounding alone. There were no significant differences in HRQoL outcomes at 180 days. Median EQ-5D-5L index values and EQ-VAS median scores showed no significant differences between the dexmedetomidine and non-dexmedetomidine groups. Among patients with COVID-19-induced acute hypoxemic respiratory failure managed with NIV, dexmedetomidine use was common. Dexmedetomidine-treated patients had a different disease trajectory than non-dexmedetomidine-treated patients. The worse outcomes observed with dexmedetomidine were not entirely explained by differences in baseline characteristics, suggesting effects of time-varying confounders. Randomized controlled trials are required to assess effects of dexmedetomidine on patient-centered outcomes.Trial registration: NCT04477668; registered on July 20, 2020.