Deep Sedation For Upper Gastrointestinal Endoscopy in Children

To the Editor: In their review, Tolia et al. (J Pediatr Gastroenterol Nutr 2000;30:477–85) raise the following important question: Should pediatric gastroenterologists administer the sedation? We think that, in view of adverse events described in the literature, deep sedation should be performed by an anesthesiologist. In the present report, we recommend this guideline on the basis of our 18-month experience with the practice of deep sedation procedures (May 1997 through November 1999). Anesthesia was administered by a pediatric anesthetist in a gastroenterology room adjacent to a recovery room. All patients had been fasting overnight before anesthesia and received local application of an anesthetic cream before the placement of an intravenous catheter. Monitoring included continuous electrocardiography, measurement of oxygen saturation by pulse oximetry (Spo2), noninvasive blood pressure determinations at 3-minute intervals, and respiratory rate. All patients received 0.06 mg/kg intravenous midazolam with 3 μg/kg intravenous alfentanil followed 3 minutes later by the same dose of alfentanil. Then a bolus of propofol (1 mg/kg), was infused with additional 0.5-mg/kg injections to allow the procedure to be performed. Pharyngeal local anesthesia was not administered. Infants were placed in the left lateral position. Most children systematically received nasal oxygen supplementation (3 L/min). Otherwise, nasal oxygen was administered if Spo2 declined below 90%. We recorded clinical data and side effects throughout the procedure. At the end, the operators reported a satisfaction score on a scale of four items (from very satisfactory to another procedure required). Patients were monitored for 1 hour in the recovery room. One hundred ninety-seven consecutive children with a median age of 8 years and a mean weight of 27.6 ± 17.2 kg, underwent 189 diagnostic procedures and 8 percutaneous gastrostomies. One hundred seventy-two patients were of American Society of Anesthesiologists (ASA) physical status grade I or II, and 25 were of ASA grade III or IV. The mean duration of the procedure was 11.9 ± 8 minutes, and the mean dose of propofol was 3 ± 1.7 mg/kg. One hundred seventy-two patients systematically received oxygen supplementation, and hypoxemia developed in one child (Spo2 <90%). Twenty-five patients had no oxygen supplementation at the beginning of the procedure, and hypoxemia developed in 17. None of the patients experienced hemodynamic instability. No aspiration, laryngospasm, vomiting, or dysrhythmia was observed. No intubation was required. Side effects were inadvertent movement (37 patients), cough (15 patients), and rash (3 patients). All procedures were completed. Operator and anesthesiologist were satisfied and very satisfied in 169 cases. Alfentanil and propofol are both rapid-and short-acting drugs. The interaction of midazolam-alfentanil-propofol produces a profound hypnotic synergism without affecting the recovery phase when used for a very brief procedure (1). The mean dose of propofol was low compared with the short duration of the procedure, which may be explained by the combination of the three drugs used. The low incidence of hypoxemia in our patients is not surprising. Most of them received nasal oxygen. Hypoxemia appeared essentially in patients without oxygen supplementation. Low respiratory depression may also result from the low dose of opioid used. Hypoxemia during endoscopy is also known to occur in unsedated patients (2). For this reason, oxygen supplementation is recommended during this procedure. The state and risks of deep sedation may be indistinguishable from those of general anesthesia (3). However, no intubation, mechanical ventilation, or muscle relaxant were required with deep sedation. We agree that control of the patient's airway would be best achieved under general anesthesia. In our experience, we reported neither aspiration nor need for intubation. The presence of an anesthesiologist should allow earlier detection of compromised airways and management of this situation before airway obstruction and major hypoxemia occur. The discomfort and the unpredictable risk that conscious sedation might turn to deep sedation or general anesthesia without the presence of an anesthesiologist prompts us to prefer deep sedation. Deep sedation for upper gastrointestinal endoscopy seems a safe and effective procedure when performed by anesthesiologists. The main benefit in using the three drugs in combination is that it probably increases comfort without major side effects such as respiratory depression or hemodynamic instability. However, the superiority of this procedure over general anesthesia requires further large-scale controlled randomized studies.