ABSTRACT Background In‐hospital use of non‐invasive ventilation improves outcomes in patients with acute respiratory failure caused by acute exacerbation of chronic obstructive pulmonary disease. Prehospital use of non‐invasive ventilation is uncommon due to diagnostic uncertainty and logistical constraints. We hypothesised that prehospital non‐invasive ventilation guided by arterial blood gas analysis would improve early physiological outcomes compared with standard medical treatment alone. Methods This investigator‐initiated, multicentre trial randomised patients with suspected acute respiratory failure due to acute exacerbation of chronic obstructive pulmonary disease with respiratory acidosis defined as pH 6.0 kPa to standard medical treatment alone or standard medical treatment plus non‐invasive ventilation. Patients were attended by prehospital physician‐manned emergency care units and enrolled without prior consent under emergency trial approval. The primary outcome was the change in arterial pH from enrolment to hospital arrival. Results The trial was terminated early due to low recruitment. Out of 111 screened patients, 16 were randomised and 14 were included in the analysis. The median prehospital pH change was 0.09 (interquartile range 0.02–0.12) in the non‐invasive ventilation + standard medical treatment group and 0.02 (0.01–0.03) in the standard medical treatment alone group. Median on‐scene time was 32 min in the non‐invasive ventilation + standard medical treatment group versus 28 min in the standard medical treatment alone group. Recruitment was limited by strict inclusion criteria, procedural complexity and challenges related to the implementation of prehospital arterial blood gas analysis and trial procedures. Missing data further limited the assessment of in‐hospital outcomes. Conclusion The clinical effect of prehospital non‐invasive ventilation could not be determined due to early termination. This trial highlights important challenges relevant to future clinical trials in the prehospital setting, including patient selection, diagnostic requirements and implementation of complex interventions. Trial Registration: NCT06211920
Brendel et al. (Sun,) studied this question.