Abstract
BACKGROUND: Arterial hypoxemia occurs in about 2.5-69% of cases during fiberoptic bronchoscopy and may necessitate administration of supplemental oxygen. Whether routine supplementary administration is indicated for all patients is a debated issue. In this prospective randomized study, we assessed the incidence of systemic desaturation (SpO(2) <90% or a >4% decrease lasting for more than 60 s) and wanted to find out whether cerebral desaturation occurs in parallel with systemic changes. PATIENTS AND METHODS: 92 consecutive patients scheduled for diagnostic bronchoscopy were randomly assigned to the no oxygen (O(2)- group), 2 l/min supplemental O(2), or 4 l/min supplemental O(2) groups. Primary end points were systemic and cerebral desaturation rate during the procedure. Secondary end points were to delineate the main risk factors of systemic and cerebral desaturation. RESULTS: In the entire cohort, systemic desaturation occurred in 18.5% of patients (n = 17), corresponding to 5 patients (16%) in the O(2) (-)group, 6 patients (19%) in the 2 l/min group, and 6 patients (20%) in 4 l/min group, respectively. In the O(2) (-) group, the probability of desaturation was 41.7 times higher than that in the 2 l/min group (p=0.014 s), while there was no difference in the probabilities of desaturation between the 2 l/min and 4 l/min groups (p=0.22). Cerebral desaturation (more than 20% rSO(2) decrease compared to baseline) did not occur in any patients in the three groups. Systemic desaturation developed earlier, and recovery after desaturation was longer in the O(2) (-) group. Male gender, smoking, and systemic oxygen saturation at baseline and FEV1% were the most significant factors contributing to systemic desaturation during bronchoscopy. CONCLUSIONS: Administration of supplemental oxygen does not prevent systemic desaturation during flexible bronchoscopy, but may contribute to the shortening of desaturation episodes and faster normalization of oxygen saturation. According to our results, 2 l/min supplemental oxygen should routinely be administered to patients throughout the procedure. This trial is registered with NCT04002609.