Background and aim: A number of studies were able to show a reduction of hypoxemia episodes during procedural sedation through the use of capnography (CA). The present study investigates the number of episodes of hypoxemia during percutaneous endoscopic gastrostomy (PEG) placement with propofol sedation comparing standard monitoring (SM) versus SM with additional CA surveillance.
Methods: In this single center randomized controlled trial, 150 patients were prospectively randomized 1:1 in either the SM group or the CA group after stratification for ASA class, PEG method (push or pull method), presence of head and neck tumor, and tracheostomy. CA analysis was performed for all patients but was blinded for the endoscopic team in the SM group.
Results: In the SM group, 57% episodes of hypoxemia (SpO2 < 90% for > 15 s) and 41% episodes of severe hypoxemia (SpO2 < 85% for > 15 s) were observed in comparison with 28% and 20% in the CA group, respectively. Odds ratios for hypoxemia and severe hypoxemia were 0.29 (confidence interval 0.15-0.57; P = 0.0005) and 0.35 (confidence interval 0.17-0.73; P = 0.008) in favor of the CA group. On average, CA was able to detect imminent mild and severe hypoxemia 83 and 99 s before standard monitoring. Standard monitoring represented an independent risk factor for hypoxemia and severe hypoxemia.
Conclusions: Respiratory complications of sedation during PEG placement are frequent events. CA is able to detect imminent hypoxemia at an early time point. This allows an early intervention and consecutively the avoidance of mild and severe hypoxemia. Therefore, CA monitoring can be recommended particularly during PEG insertion procedures.
Keywords: capnography; complication; endoscopy; monitoring; percutaneous endoscopic gastrostomy; sedation.
© 2019 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.