Flexible bronchoscopy in the intensive care unit: the FLEXICARE survey

Flexible bronchoscopy (FB) is a key diagnostic and therapeutic procedure in the intensive care unit (ICU). However, significant heterogeneity exists regarding indications, peri-procedural management, training and safety practices across ICUs. Data on real-world practices remain limited.

An international, online cross-sectional survey was conducted by the European Respiratory Society's Acute Critical Care Group between May and December 2023. The questionnaire, designed according to Checklist for Reporting Of Survey Studies (CROSS) and Academy of Critical Care: Development, Evaluation and Methodology (ACCADEMY) standards, targeted healthcare professionals performing FB in ICUs. It addressed procedural protocols, sedation, ventilation strategies, complications, microbiologic sampling, tracheostomy practices and training. Only fully completed responses were included in the analysis.

Data from 266 respondents across 73 countries were analysed. Training in FB varied widely: 41% reported no formal training, and only 61% had access to a 24/7 bronchoscopy service. Sedation practices were heterogeneous. Ventilator settings were adjusted differently based on ICU certification, with ICU-certified physicians more likely to reduce positive end-expiratory pressure during FB (55% versus 42%; p=0.031). Bronchoalveolar lavage was the most common sampling method, with a median saline volume of 60-120 mL and a recovery rate of 48±18%. Microbiological findings led to therapeutic changes in 48±21% of cases. FB guidance during percutaneous tracheostomy was applied in only 56% of cases.

Substantial variability exists in FB practices across ICUs, particularly in training, sedation, ventilation adjustments and microbiologic sampling. Despite frequent therapeutic consequences, standardised approaches are lacking. These findings underscore the need for international consensus guidelines and structured training programmes to harmonise FB practices in critical care settings.