The purpose of our trial was to assess the arterial blood gases changes, during upper gastrointestinal endoscopy (UGE), among subjects younger than sixty years without cardiopulmonary disorders or smoking habits, with or without sedation with i.v. midazolam. After informed consent, sixty five patients (29 males and 36 females), whose ages ranged between 16 and 59 years (average 37.5 years) were randomly assigned to two groups (I and II) of 40 and 25 patients, respectively. Group I, underwent UGE without sedation and arterial blood gases from a right radial arterial sample were determined before and during the procedure. Group II underwent UGE with sedation and arterial blood gases from a right radial arterial sample were determined, before sedation, after sedation with i.v. midazolam at a dosage of 0.05 mg/kg, and during the procedure. In group I 55% of patients showed hypoxemia: PO2 decreased 10 mmhg (mean) and oxygen saturation in 1.45%. In group II, 64% of patients showed hypoxemia after sedation, and before UGE, the decrease being even more notorious during the procedure. The PO2 decreased 18.8 mmhg (mean) and oxygen saturation in 3.3% (p 0.001). In both groups no significative changes in PCO2,pH or respiratory rate were observed. However the depth of thorax respiratory expansion was clinically decreased in 70% of patients of group II. We concluded that hypoxaemia occurred during UGE, in previously asymptomatic, middle-age, non smokers patient, without history of cardiac or pulmonary disease. Major changes were observed in subjects sedated with midazolam (p 0.001).