OBJECTIVE To determine the effect of sedation using diazepam on hemoglobin oxygen saturation (SpO2) in patients undergoing esophagogastroduodenoscopy (EGD). METHODS 100 consecutive patients scheduled for EGD were randomly allocated to receive 0.03 mL/Kg of either diazepam (5 mg/mL solution) or normal saline intravenously after topical oropharyngeal anesthesia immediately before the procedure. SpO2 was continuously monitored throughout the procedure by an anesthetist who was unaware of the drug received. RESULTS Fall in SpO2 exceeding 4% was noted in 78% of patients in the diazepam group and in 38% of patients in the placebo group (p < 0.001). Fall in SpO2 to suboptimal level (89%) was seen in 20% of patients in the diazepam group and in 10% patients in the placebo group (p < 0.001). The duration of suboptimal SpO2 was similar (means +/- SD being 2.47 +/- 0.10 min in diazepam group and 2.86 +/- 0.32 min in placebo group). CONCLUSIONS Intravenous diazepam administration before EGD produces a significant fall in SpO2 during the procedure, and so should be avoided; continuous monitoring of SpO2 should be done during EGD.