Preprocedural Prediction of Sedation-Related Hypoxemia in Advanced Gastrointestinal Endoscopy: Development of a Respiratory Risk Score

Aim: Intraprocedural hypoxemia is a common respiratory adverse event during sedated gastrointestinal endoscopy. This study aimed to identify factors associated with hypoxemia and to develop a simple preprocedural risk score using routinely available clinical variables.Material and Methods: This retrospective single-center study included 150 adults who underwent sedated ERCP or EUS between January 2015 and December 2025. Hypoxemia was defined as SpO₂ <90% during the procedure. Clinical, procedural, and sedation-related variables were analyzed. Logistic regression was used to identify factors associated with hypoxemia, and a preprocedural risk score was developed from variables available before sedation.Results: Hypoxemia occurred in 32 patients (21.3%). In multivariable analysis, body mass index, ASA III–IV status, chronic pulmonary disease, lower baseline SpO₂, and procedure duration were associated with hypoxemia. Procedure duration was excluded from the score because it was not preprocedural. The final score included body mass index ≥30 kg/m², ASA III–IV status, chronic pulmonary disease, and baseline SpO₂ ≤95% (range, 0–7 points). Hypoxemia rates increased across low-, intermediate-, and high-risk groups (4.8%, 20.8%, and 51.4%, respectively; p<0.001). The score yielded an AUC of 0.80 (95% CI: 0.72–0.88; p<0.001). At a cut-off of ≥4 points, sensitivity was 56.3%, specificity was 85.6%, positive predictive value was 51.4%, negative predictive value was 87.8%, and accuracy was 79.3%.Conclusions: A simple preprocedural score based on body mass index, ASA physical status, chronic pulmonary disease, and baseline SpO₂ stratified hypoxemia risk in sedated gastroenterological endoscopy. External validation is required before routine clinical use.