Esophagogastroduodenoscopy (EGD) is considered an essential diagnostic and therapeutic procedure in the pediatric population. Although generally safe, EGD has the potential for complications. The most frequently reported adverse effect is hypoxia associated with the administration of sedative medications (1,2). Between 0.03% and 1.8% of children undergoing EGD with conscious sedation have minor complications, some of which may prevent completion of the examination and hypoxia requiring intervention (3); major complications are less common. Transient oxygen desaturation (<90%) occurs in up to 68% of patients undergoing EGD (4).
In our department, on the basis of our experience and previous reports (3–5), we routinely perform EGD in patients younger than 10 years and are under general anesthesia. In the past, oxygen was administered through a nasal cannula or the patient was intubated. However, a nasal cannula impedes intervention by the anesthetist in case of severe desaturation. In this situation, the procedure is terminated and endotracheal intubation is usually performed. We developed a simple method of introducing an endoscope through an endoscopic mask. To our knowledge, there are no reports concerning the use of this mask for pediatric EGD. Here we report our experience using the endoscopic mask as a conduit for flexible EGD.
We developed a simple method of introducing an endoscope through an endoscopic mask. This face mask, which was first described by Mallios (6) in 1980 and later modified by Patil et al (7), was primarily developed for flexible fiberoptic bronchoscopy (8). It has a membrane with an opening valve through which a video endoscope is introduced, thereby facilitating adequate oxygenation and controlling ventilation when necessary as well as supplemental oxygen and inhalational anesthesia throughout the procedure. Currently, we use a modified endoscopy mask developed by Erb et al (8) that is manufactured by VBM Medizintechnik GmbH (Sulz/Neckar, Germany).
Children between 6 months and 10 years of age undergoing diagnostic EGD, with American Society of Anesthesiologists (ASA) physical status I and II, constituted the study population. The following exclusion criteria were applied: therapeutic EGD, hemoglobin concentration below 10 g/dL, active asthmatic patients, allergy to anesthetic agents, and neck or maxillofacial congenital malformations.
All children fasted for at least 6 hours for solids and 2 hours for clear fluids. No premedication was used. Children were anesthetized in the presence of their parents in a special induction room. Anesthesia was induced by a senior anesthesiologist and a resident.
Heart rate, heart auscultation, noninvasive systolic and diastolic blood pressure, SaO2, end-tidal CO2 (ETCO2), and temperature were monitored in all patients with an AS/3 patient monitor (Datex Ohmeda, General Electric, Madison, WI, USA). These parameters were recorded every 1 min during the procedure by the resident anesthesiologist. All children were hydrated with lactated Ringer solution at a rate of 4 to 6 mL/kg/hour. Anesthesia was induced with N2O/O2 (50%/50%) and sevofluran up to 6% using a Bain circuit through a regular face mask. After the eyelash reflex disappeared, the regular face mask was replaced by the endoscopic mask (VBM Medizintechnik GmbH). This mask comes in 3 sizes for the pediatric population: size 0 with a silicone membrane measuring 2.0 mm in diameter designed for neonates, size 2 with a 2-mm diameter port for infants, and size 3 with a 3-mm diameter port for children (Fig. 1). Only sizes 2 and 3 were used in our study. The masks are completely transparent and latex free (Fig. 1).
The flexible endoscope is passed through the valve of the mask into the oropharynx, although the patient is breathing spontaneously or with assisted manual ventilation when anesthetized. The silicone diaphragm covering the valve ensures that air leakage is minimal, and mask ventilation can be continued safely throughout the procedure.
End-tidal isoflurane was maintained between 1 and 1.5 minimal alveolar anesthetic concentration. Isoflurane concentrations were increased whenever heart rate and blood pressure increased by more than 20%, that is, higher than the baseline values. Manual-assisted ventilation was performed to keep ETCO2 between 30 and 48 mmHg. An intravenous catheter was inserted after the patient was asleep. After induction of anesthesia, the patient was placed in the left lateral position for the procedure. All procedures were performed by the same pediatric gastroenterologist (R.S.) using the Olympus GIF 160 pediatric endoscope or the GIFQ 145 endoscope (Olympus, Hamburg, Germany) for older patients, inserted through the silicone membrane of the endoscopy mask and through an appropriate pediatric bite block (Fig. 2). The patient's demographic data were collected by the resident, and hemodynamic and respiratory data were recorded throughout the procedure.
Complications or events such as desaturation (O2 saturation less than 90% for more than 1 min), upper airway obstruction episodes (detected by chest wall abnormalities and capnography in addition to chest auscultation), and any other respiratory event that caused an immediate change of the ventilation technique or airway device were recorded.
Ventilation was rated “satisfactory” when spontaneous ventilation was uneventful (symmetrical and normal chest movement, regular ETCO2 waveform). Ventilation was rated “difficult” when signs of airway obstruction were noted and resolved by external airway maneuvers (EAMs), namely, extension of the neck, chin lift, jaw thrust, or by changing the size of the face mask. Ventilation was rated “impossible” when EAM did not result in resolution of the problem and it became necessary to insert an endotracheal tube.
Analysis of variance (ANOVA) with repeated measures was used to assess differences between the hemodynamic parameters over time. Confidence intervals were calculated to assess the relative risk of complications.
A total of 262 children underwent upper endoscopy during the 2 years of the study. Of these, 22 patients were excluded: 3 had a high ASA score, 2 had anatomic malformations in the upper airway, 8 had undergone therapeutic procedures, 5 were younger than 6 months, and 4 underwent both upper endoscopy and colonoscopy and, therefore, were not eligible for the study. Consequently, 240 children (122 boys and 118 girls) participated in the study. Age range was 7 to 135 months (mean 60.7 months ± 34.4).
All patients maintained a stable hemodynamic status throughout the procedure (Table 1). Although respiratory rate and systolic and diastolic blood pressures changed significantly over time, the differences were not clinically relevant. Ventilation was regarded as satisfactory in 230 patients. Only 10 patients required active manipulation by the anesthesiologist (4%, 95% confidence interval [CI] 1.6%–6.7%). Nine patients had difficult ventilation, which was resolved by applying external airway maneuvers. Ventilation was impossible in only 1 patient (10 months old) and endotracheal intubation was required. There were no procedure-related complications.
Several factors may contribute to hypoxia in pediatric patients undergoing upper endoscopy. Because the endoscope is being inserted, it may obstruct the pharynx or compress the trachea, thereby resulting in desaturation during this stage of the procedure (9). This is observed more frequently by a large-diameter endoscope (10,11) and in small children (10). In small infants and children, gastric distension due to air insufflation may also lead to severe hypoxia (12). Medications used for sedation are potent central nervous system depressants and can lead to hypoventilation, particularly when several drugs are combined (13). Conscious sedation may, therefore, be problematic, especially in infants and young children. Oxygen desaturation is also frequent in adults undergoing EGD under conscious sedation (14–16) and is often accompanied by cardiac dysrhythmia or myocardial ischemia (16,17). In children, oxygen desaturation has been reported in 37% to 68% of patients during EGD performed with conscious sedation (10,18,19), and cardiac dysrhythmia occurred in half of them (18). These complications have been observed less frequently in recent years (3). Lightdale et al (20) conducted a randomized controlled study of 163 children undergoing 174 elective gastrointestinal procedures with moderate sedation in a pediatric endoscopy unit. Their aim was to determine whether intervention based on capnography indications of alveolar hypoventilation reduced the incidence of arterial oxygen desaturation during procedural sedation in children. Capnography indicated alveolar hypoventilation during 56% of the procedures, and apnea occurred during 24% of them. Lamireau et al (4) compared the occurrence of oxygen desaturation during intravenous sedation to that during general anesthesia and found significantly lower baseline oxygen saturation in the sedation group compared with the general anesthesia group (89 ± 5 vs 97 ± 1; P < 0.001). In addition, oxygen desaturation (<95%) events were more frequent in the sedation group than in the general anesthesia group (89% vs 5.5%). Profound desaturation (<90%) occurred in 9 patients of the sedation group and in none of the patients in the general anesthesia group (50% vs 0%). In the general anesthesia group, heart rate and mean arterial pressure remained stable throughout the procedure, whereas in the sedation group, heart rate and mean arterial pressure increased significantly during the procedure. Squires et al (5) reported that children ages 3 to 9 years were more difficult to sedate, and that children ages 6 to 9 years required the highest doses of midazolam and meperidine per kilogram body weight and had higher anxiety levels that resulted in higher pulse rates.
When EGD is performed in patients under general anesthesia, it is well tolerated and does not cause retching, pain, or discomfort. This is confirmed by the absence of changes in heart rate and mean arterial pressure during the procedure (4). The anesthetic risk is considered low (21), which is in agreement with our experience.
There is a consensus that tracheal intubation under general anesthesia results in better control of ventilation and oxygen delivery than mask ventilation. Endotracheal intubation is considered the gold standard for protecting the airway from aspiration and ensuring adequate ventilation (22). Endotracheal intubation may be beneficial for infants because they are more likely to develop upper airway obstruction and gastric distention with longer periods of mask ventilation than older children. However, laryngospasm or bronchospasm, airway edema, and postoperative tracheal stenosis have been associated with endotracheal intubation, especially in young children (22).
In our study we used a modified endoscopic mask. The size of endoscopic mask should ensure a perfect seal around the mouth and nose, and the mask should be individualized for each child. Long-term face mask ventilation carries a significant risk for gastric insufflation, especially in infants (22). However, this is of less consequence during endoscopic procedures. Dead space is greater with the face mask than with the endotracheal tube. The smaller the patient, the greater is this concern. However, this was not a problem in our study. The endoscopic masks are made of lightweight and transparent material so that it is possible to monitor the color of the perioral skin, to evaluate the humidity of the expired gases, and to recognize oral or gastric secretions (22) immediately.
A modified laryngeal mask has been used for upper endoscopic procedures. Lopez-Gil et al (23) compared the use of the ProSeal laryngeal mask airway (LMA) with the drain tube as a conduit to the stomach with the nasal cannula with conventional oral access to the stomach in 60 children. They used the Olympus LF-DP fiberscope, which has an external diameter of 3.1 mm. This endoscope, which is designed for bronchoscopy, is not suitable for upper endoscopy and cannot replace the routine gastroscopes. No differences in the ease of performing the procedure were found. However, hypoxia was more frequent in the nasal cannula group (20% vs 0%). Recovery scores were similar.
Fuentes-Garcia et al (24) compared the safety and efficacy of LMA with endotracheal intubation in children undergoing elective diagnostic upper gastrointestinal endoscopies. Sixty ASA I–III patients were randomly allocated to endotracheal intubation or LMA. The cardiovascular and respiratory parameters evaluated in the study varied during the procedure. However, they remained within physiological ranges and did not differ between groups. The median recovery time was 4 min and the time-to-discharge was 58 min in the endotracheal intubation group versus 3 min and 50 min, respectively, in the LMA group (P > 0.10). The authors concluded that LMA was as effective and safe as endotracheal intubation for securing the airways of children undergoing diagnostic upper endoscopy. They had a 3% failure rate (1 patient) with LMA. The use of LMA for EGD has some limitations, that is, it makes it difficult to inspect the upper part of the esophagus and pharynx, and regular endoscopes, including pediatric endoscopes, cannot be inserted in this device. In addition, LMA may affect the correct positioning of the laryngeal mask itself.
The face mask we used was originally developed for flexible bronchoscopy to guarantee the supply of oxygen and anesthesia throughout the procedure. It has also been used to facilitate fiberoptic intubation in children (25,26). There are several versions of this mask (8,27) and we used the one that was modified according to the dimensions given by Erb et al (8). Most of our patients remained stable throughout the procedure, as can be seen in Table 1. Although there were statistically significant hemodynamic changes (blood pressure and respiratory rate) over time, they were not clinically relevant. In addition, although not measured, our impression is that the use of the mask did not delay the start of the procedure or the recovery time.
On the basis of our results, it appears reasonable to suggest that an endoscopic mask be used as a first choice for EGD under general anesthesia in children. Endotracheal intubation should be reserved for cases in which airway problems occur during the procedure or in small neonates and infants. The mask ensures a safe procedure without the need of endotracheal intubation in most cases. We, therefore, recommend its routine use in upper endoscopic procedures for healthy children who do not have a significant anesthesia risk (ASA physical status of I and II) and older than 6 months.
We thank Dr Ana Vayntrub for help in the arrangement of data. We thank Dr Ada Tamir for statistics consultation, and Prof Michael Jaffe for useful remarks. We are grateful to Jean Ann Gilder text editing.
1. Thakkar K, El-Serag HB, Mattek N, et al
. Complications of pediatric EGD: a 4-year experience in PEDS-CORI. Gastrointest Endosc 2007; 65:213–221.
2. Ament ME. Complications of pediatric EGD: more questions than answers. Gastrointest Endosc 2007; 65:222–223.
3. Rothbaum RJ. Complications of pediatric endoscopy. Gastrointest Endosc Clin N Am 1996; 6:445–459.
4. Lamireau T, Dubreuil M, Daconceicao M. Oxygen saturation during esophagogastroduodenoscopy in children: general anesthesia versus intravenous sedation. J Pediatr Gastroenterol Nutr 1998; 27:172–175.
5. Squires RH Jr, Morriss F, Schluterman S, et al
. Efficacy, safety, and cost of intravenous sedation versus general anesthesia in children undergoing endoscopic procedures. Gastrointest Endosc 1995; 41:99–104.
6. Mallios C. A modification of the Laerdal anaesthesia mask for nasotracheal intubation with the fiberoptic laryngoscope. Anaesthesia 1980; 35:599–600.
7. Patil V, Stehling LC, Zauder HL, et al
. Mechanical aids for fiberoptic endoscopy. Anesthesiology 1982; 57:69–70.
8. Erb T, Hammer J, Rutishauser M, et al
. Fibreoptic bronchoscopy in sedated infants facilitated by an airway endoscopy mask. Paediatr Anaesth 1999; 9:47–52.
9. Rimmer KP, Graham K, Whitelaw WA, et al
. Mechanisms of hypoxemia during panendoscopy. J Clin Gastroenterol 1989; 11:17–22.
10. Casteel HB, Fiedorek SC, Kiel EA. Arterial blood oxygen desaturation in infants and children during upper gastrointestinal endoscopy. Gastrointest Endosc 1990; 36:489–493.
11. Bendig DW. Pulse oximetry and upper intestinal endoscopy in infants and children. J Pediatr Gastroenterol Nutr 1991; 12:39–43.
12. Brock-Utne JG, Moynihan RJ. Patient draping contributing to a near disaster (desaturation during endoscopy in a 2-year-old). Paediatr Anaesth 1992; 2:333–334.
13. Arrowsmith JB, Gerstman BB, Fleischer DE, et al
. Results from the American Society for Gastrointestinal Endoscopy/U.S. Food and Drug Administration collaborative study on complication rates and drug use during gastrointestinal endoscopy. Gastrointest Endosc 1991; 37:421–427.
14. O'Connor KW, Jones S. Oxygen desaturation is common and clinically underappreciated during elective endoscopic procedures. Gastrointest Endosc 1990; 36(3 Suppl):S2–S4.
15. Bell GD, Reeve PA, Moshiri M, et al
. Intravenous midazolam: a study of the degree of oxygen desaturation occurring during upper gastrointestinal endoscopy. Br J Clin Pharmacol 1987; 23:703–708.
16. Lieberman DA, Wuerker CK, Katon RM. Cardiopulmonary risk of esophagogastroduodenoscopy. Role of endoscope diameter and systemic sedation. Gastroenterology 1985; 88:468–472.
17. Murray AW, Morran CG, Kenny GN, et al
. Examination of cardiorespiratory changes during upper gastrointestinal endoscopy. Comparison of monitoring of arterial oxygen saturation, arterial pressure and the electrocardiogram. Anaesthesia 1991; 46:181–184.
18. Gilger MA, Jeiven SD, Barrish JO, et al
. Oxygen desaturation and cardiac arrhythmias in children during esophagogastroduodenoscopy using conscious sedation. Gastrointest Endosc 1993; 39:392–395.
19. Chuah SY, Crowson CP, Dronfield MW. Topical anaesthesia in upper gastrointestinal endoscopy. BMJ 1991; 303:695.
20. Lightdale JR, Goldmann DA, Feldman HA, et al
. Microstream capnography improves patient monitoring during moderate sedation: a randomized, controlled trial. Pediatrics 2006; 117:e1170–e1178.
21. Cohen MM, Cameron CB, Duncan PG. Pediatric anesthesia morbidity and mortality in the perioperative period. Anesth Analg 1990; 70:160–167.
22. Brambrink AM, Braun U. Airway management in infants and children. Best Pract Res Clin Anaesthesiol 2005; 19:675–697.
23. Lopez-Gil M, Brimacombe J, Az-Reganon G. Anesthesia for pediatric gastroscopy: a study comparing the ProSeal laryngeal mask airway with nasal cannulae. Paediatr Anaesth 2006; 16:1032–1035.
24. Fuentes-Garcia VE, Morales-Perez E, Ramirez-Mora JC, et al
. A randomized trial comparing laryngeal mask airway to endotracheal tube in children undergoing upper gastrointestinal endoscopy. Acta Biomed 2006; 77:90–94.
25. Frei FJ, Ummenhofer W. A special mask for teaching fiber-optic intubation in pediatric patients. Anesth Analg 1993; 76:458.
26. Wilton NC. Aids for fiberoptically guided intubation in children. Anesthesiology 1991; 75:549–550.
27. Kitamura S, Fukumitsu K, Kinouchi K, et al
. A new modification of anaesthesia mask for fibreoptic intubation in children. Paediatr Anaesth 1999; 9:119–122.