RESUMEN
BACKGROUND: Bronchoscopy is widely used and regarded as standard of care in most intensive care units (ICUs). Data concerning recommendations for on-call bronchoscopy are lacking. OBJECTIVES: Evaluation of recommendations, complications, and outcome of on-call bronchoscopies. METHOD: A retrospective single-centre analysis was conducted in a large university hospital. All on-call bronchoscopies performed outside normal working hours in the year before (period 1) and after (period 2) establishing a catalogue of recommendations for indications of on-call bronchoscopy on November 1, 2016, were included. RESULTS: Overall, 924 bronchoscopies in 538 patients were analysed. A relative reduction of 83.6% from 794 bronchoscopies in 432 patients (1.84 per patient) during period 1 to 130 in 107 patients (1.21 per patient) during period 2 was observed. Most bronchoscopies (812/924, 87.9%) were performed in ICUs, and 416 patients (77.3%) were intubated. Bronchoscopies for excessive secretions decreased significantly during period 2. Fifty-three of 130 bronchoscopies (40.8%) fulfilled the specified recommendations during period 2, in comparison with 16.8% in period 1 (p < 0.001). Complications were recorded in 58 of 924 procedures (6.3%) and were more frequent in period 2, especially moderate bleeding. In-hospital mortality of patients undergoing on-call bronchoscopy did not differ between periods and was 28.7 and 30.2% in periods 1 and 2, respectively. CONCLUSION: The introduction of recommendations for on-call bronchoscopy led to a significant decline of on-call bronchoscopies without negatively affecting outcome. More evidence is needed in on-call bronchoscopy, especially for ICU patients with intrinsic higher complication rates.