Objective       To study the best oxygen inhalation model during painless gastroscopy.      Methods       Five hundred and sixty-six patients were randomly divided into non pre-oxygen inhalation group (oxygen was supplied at the time of anesthesia)

2-3 L/min intra-operative oxygen flow group (group Ⅰ)

3-4 L/min intra-operative oxygen flow group (group Ⅲ)

4-5 L/min intra-operative oxygen flow group (group Ⅴ). Respiratory rate and SpO

2

of the patients were recorded before operation

after re-oxygenation

2 min after anesthesia

intra- and post- operation

respectively and pre-oxygen inhalation group (

oxygen-supply begins at 2-3 L/min two minutes before the anethesia): 2-3 L/min intra-operative oxygen flow group (group Ⅱ)

3-4 L/min intra-operative oxygen flow group (group Ⅳ)

4-5 L/min intra-operative oxygen flow group (group Ⅵ).      Results       The SpO

2

of patients was differently decreased after anesthesia and reached its lowest value 2 min after anesthesia. The SpO

2

was higher in pre-oxygen inhalation group than in non pre-oxygenation group (P0.05). The incidence of hypoxemia was significantly lower in pre-oxygenation group than in non pre-oxygen inhalation group (11/93

8/95 and 10/91 vs 28/98

25/94 and 27/95

P0.05 or P0.01）. However，no significant difference was observed in SpO

2

and incidence of hypoxemia in pre-oxygen inhalation group and non pre-oxygenation group receiving oxygen flow at different speeds.      Conclusion       Oxygen inhalation through a nasal tube at a speed of 2-3 L/min during painless gastroscopy is a safe

effective

and simple procedure for oxygen supply.