The optimal method of oxygen supplementation during upper gastrointest
inal endoscopy has not been clearly defined. The aim of this study was
to compare oxygen supplementation via nasal prongs with that via a ca
theter passed into the low oropharynx to eliminate the effect of mouth
breathing. Patients were stratified according to the American Society
of Anesthesiologists (ASA) classification of physical status into low
er-risk (ASA 1 and 2) and higher-risk (ASA 3) groups. The lower-risk g
roup received intranasal, intrapharyngeal, or no oxygen supplementatio
n, and higher-risk patients received either intranasal or intrapharyng
eal oxygen. Continuous arterial oxygen saturation (SpO(2)) was recorde
d, using a pulse oximeter, before and during endoscopy. Critical desat
urations (SpO(2) less than or equal to 90%), minimum SpO(2) during end
oscopy, and maximum desaturation from the baseline oxygen on air, were
evaluated. There was no significant difference in the number of patie
nts desaturating, mininum SpO(2), or in the maximum desaturation from
the baseline between the groups receiving intranasal or intrapharyngea
l oxygen supplementation. In lower-risk patients receiving no suppleme
ntary oxygen (n = 27), ten patients (37 %) desaturated, compared,vith
one of 52 patients (2 %) receiving supplementary oxygen (p < 0.001). T
here was also a significant difference between these groups in the min
imum SpO(2) (91 % vs 97 %, p < 0.001) and the maximum desaturation fro
m the baseline (- 5.2 % vs + 0.7 %, p < 0.001) during endoscopy. We co
nclude that the intranasal and intrapharyngeal methods of oxygen suppl
ementation are of similar efficacy, and that supplementary oxygen sign
ificantly decreases the incidence of critical arterial oxygen desatura
tion that occurs even in healthy patients undergoing upper gastrointes
tinal endoscopy.