Deep sedation controlled by the staff of gastrointestinal endoscopy units is currently controversial. In the last few years, numerous studies have provided data supporting the safety of propofol use in these techniques. We present a large series of patients who underwent gastroscopy or colonoscopy under endoscopist-controlled deep sedation. A total of 875 procedures (297 gastroscopies and 578 colonoscopies) were included. In all procedures intravenous propofol with or without intravenous midazolam was administered. In gastroscopies, complications attributable to the sedation were found in only 6.7% of the patients, mostly due to desaturation, which was resolved without the need for intubation. In colonoscopies, complications were found in 11.2%, the most frequent being bradycardia and desaturation, none of which were serious. No association was found between the presence of complications and the propofol dose administered. In the group of patients undergoing colonoscopy, simultaneous midazolam administration allowed reduction of the propofol dose required to achieve deep sedation. In conclusion, propofol shows a good safety profile and excellent tolerance in patients undergoing gastroscopy and colonoscopy and can be administrated by the endoscopy team. At least in the case of colonoscopy, the associated use of midazolam allows the propofol dose to be decreased, thus, theoretically, reducing the drug's adverse effects.