To the Editor:

In recent article of Kang et al. [1] comparing deep sedation versus conscious sedation for gastroscopy in obese patients, patients’ positions during gastroscopy were not provided. Available literature indicates that positions can significantly change severity of airway obstruction and affect the occurrence of hypoxemia during gastrointestinal endoscopy with sedation [2]. As various positions are often needed due to the preference of endoscopists and the convenience of procedure, we are concerned that significant imbalance in patients’ positions between groups might have biased the incidence of oxygen desaturation and hypoxemia. Furthermore, oxygen flows of 2–4 L/min were supplied via a nasal cannula during gastroscopy. As oxygen flow is an important determinant for efficacy of supplemental oxygen via nasal cannula, design of inconsistent oxygen flows may have confounded the interpretation of their results. Most important, mask-assisted ventilation was used for management of hypoxemia. However, it was unclear what type of face mask was used and whether assisted ventilation needed to interrupt endoscopic procedure. If so, it was difficult to understand why endoscopists’ satisfaction scores were similar between groups. In fact, interruption of endoscopy by airway management is a main reason for dissatisfaction of endoscopists. In addition, dosage of sulfentanil in their table 3 should be μg/kg, rather than mg/kg.