Arterial oxygen desaturation during endoscopic ultrasonography combined with gastroscopy: A safety evaluation in out-patients

Background and Study Aims: Data concerning potential clinically relevant ar terial oxygen desaturation (SaO(2) <90.0%) during endoscopic ultrasonograph y (EUS) combined with a prior conventional gastroscopy are not available, b ut are nevertheless important in view of the increasing numbers of examinat ions and relevant safety standards. We therefore continuously measured SaO( 2) before and during the periods of gastroscopy and subsequent EUS in order to answer the question of whether EUS has additional effects on arterial o xygen desaturation, A secondary aim of the study was to define patients wit h particular risks for decreased SaO(2). Patients and Methods: SaO(2) was continuously monitored in 96 outpatients b efore and during the endoscopic procedures. After sedation with midazolam, an exploratory gastroscopy was performed, followed by endoscopic ultrasonog raphy, Results: In all patients undergoing gastroscopy followed by EUS (n = 96), S aO(2) values significantly decreased from 96.5 +/- 2.6% to 92.4 +/- 4.7% (P < 0.0001). Analysis of the time-related SaO(2) courses showed that the min imal SaO(2) was reached after 7.5 minutes. The SaO(2) curves in patients wi th desaturation (n = 29) were similar to those in patients who had no desat uration, except for a significantly lower nadir (88.6 +/- 5.3 %, P < 0.0001 ). The SaO(2) time courses in all patients showed no further decrease or a second minimum during the EUS period, as might have been expected with the insertion of a second instrument. Baseline (pre-procedure) and minimal SaO( 2) were significantly correlated (T-s: 0.604, P < 0.038). Quantitative anal ysis yielded the lowest significant percentage of desaturation, with baseli ne SaO(2) equal to or above 97.5 % (6.3 %, P < 0.038). Conclusions: EUS following an exploratory gastroscopy under mild sedation c auses no additional reduction in SaO(2). Patients with an initial arterial oxygen saturation equal to or above 97.5 % may have a low probability of de saturation during the combined procedure after sedation with one single mid azolam dose, and are therefore unlikely to need further monitoring after an initial assessment. Patients below this limit, however, should be continuo usly monitored, due to the increasing probability of desaturation below thi s limit.