Secondary Logo

Journal Logo

Paediatric anaesthesia

Incidence and risk factors for adverse events during monitored anaesthesia care for gastrointestinal endoscopy in children

A prospective observational study

Najafi, Nadia; Veyckemans, Francis; Vanhonacker, Domien; Legrand, Catherine; Van de Velde, Anne; Vandenplas, Yvan; Poelaert, Jan

Author Information
European Journal of Anaesthesiology: June 2019 - Volume 36 - Issue 6 - p 390-399
doi: 10.1097/EJA.0000000000000995



Gastrointestinal endoscopy is a commonly performed procedure in children. It comprised 4.8% of all cases referred to the recent Anaesthesia PRactice in Children Observational Trial survey.1 Safe and complete performance of endoscopy procedure as well as poor expected level of cooperation necessitate its conduct under general anaesthesia or deep sedation in children.2,3 Despite an abundance of data, lack of uniformity in general anaesthesia or sedation techniques and reporting of sedation-related adverse events impedes solid comparison of outcomes in children. Most studies have regarded adverse respiratory events (AREs) as the primary cause of adverse events and identified risk factors such as age, obesity, respiratory comorbidity and upper gastrointestinal endoscopy in sedation practices.2,4–10 Whether anaesthetic drugs also contribute to the occurrence of adverse events and the magnitude of their contribution is uncertain.

The American Society of Anesthesiologists (ASA) has defined Monitored Anaesthesia Care (MAC) and declared its position to allow the safe administration of a maximal depth of sedation necessary to achieve ideal procedural conditions by qualified personnel.11 Although general anaesthesia is preferred in some circumstances, MAC may be applicable to most cases of paediatric gastrointestinal endoscopy procedures. Better understanding of its risk factors for adverse events may therefore improve outcome in children.

At our university teaching hospital, therapeutic and urgent gastrointestinal endoscopy procedures are performed in the operating room under general anaesthesia whereas diagnostic elective gastrointestinal endoscopy procedures are generally performed outside the operating room under MAC by using natural airways with the child in lateral position. Therefore, the main aims of this cohort study were to determine the prevalence of adverse events during MAC and to identify their predicting factors, including different anaesthetic/sedative drugs.


This observational cohort was approved by Institutional Review Board (O.G. 016, B.U.N.143201730771-2016/418) of the Universitair Ziekenhuis Brussel, Brussels, Belgium on 8 February 2017 and was performed in accordance with the Helsinki Declaration on Patient Safety in Anaesthesiology12 and Strengthening the Reporting of Observational Studies in Epidemiology statement.13 Parental consent to review data was waived but informed consent was obtained for sedation purposes in line with the Joint Commission on International Accreditation.

The prospectively collected electronic data of all children requiring MAC provided by anaesthesiologists outside the operating room between 1 January 2010 and 31 August 2016 were analysed retrospectively.

Children aged up to 16 years who underwent elective or diagnostic gastrointestinal endoscopy were reviewed for eligibility. Children presenting with ASA physical status at least 4, those requiring mechanical ventilation, and therapeutic or urgent gastrointestinal endoscopy were not included as they were managed in the operating room. Endoscopy was postponed in children presenting with an acute respiratory illness which was defined as the presence of fever, purulent nasal congestion or wheeze on auscultation.

In our institution, diagnostic gastrointestinal endoscopies are carried out in a central endoscopy suite, outside the operating rooms, but equipped with modern anaesthetic facilities and a separate 4-bed postanaesthesia care unit (PACU). Gastrointestinal endoscopy is performed or directly supervised by experienced paediatric gastroenterologists whereas MAC is provided or directly supervised by experienced (paediatric) anaesthesiologists. Presedation evaluation, fasting recommendations, periprocedural monitoring and assessment of sedation level based on the Ramsay Sedation Scale14 are described in a previous study.15 Intravenous access was established by qualified paediatric nurses following application of local anaesthetics and prior to the sedation session. If a child had known difficult intravenous access or when it proved difficult after evaluation by nurses or on specific request of child or his parents, this was obtained after inhalational induction using sevoflurane. The choice of anaesthetic drugs was left to the discretion of the anaesthesia provider with no restrictions and their induction dosages were titrated to the child's response. Following induction of anaesthesia and according to departmental policy, oxygen was given to all patients via nasal prongs equipped with a sample port for end-tidal carbon dioxide analysis (Intellivue MP30, Philips, Böblingen, Germany). The child was then placed in the left lateral position. Once a minimum Ramsay Sedation Scale of 5 had been achieved and following confirmation of respiratory and haemodynamic stability, a bite block was inserted and the endoscopic procedure was allowed to proceed. Minimal movement of hands at the initiation of endoscopy was tolerated provided that no restraint was needed. The technical conditions of the procedure were assessed by the endoscopist as satisfactory, difficult or impossible and documented in the electronic medical records. According to institutional agreement, children were maintained in the lateral position for upper gastrointestinal endoscopy, but during lower gastrointestinal endoscopy, the child's position could be changed from lateral to supine and vice versa if needed. Deep sedation was maintained with repeated boluses of propofol (Propolipid, Fresenius Kabi, Schelle, Belgium) for upper gastrointestinal endoscopy or a continuous infusion for lower gastrointestinal endoscopy. Propofol dosages (10 to 15 mg kg−1 h−1) varied depending on the degree of stimulation and the child's needs. After the procedure, the patients were transferred to the PACU where their vital parameters were monitored and recorded at 15-min intervals. Patients were discharged from PACU and then to home when they maintained a modified Aldrete Score of at least 9 and more than 12, respectively.16

Full details of patients, procedures and anaesthesia-related characteristics were collected by using structured data collection forms. Children's characteristics and comorbidities were predefined. Anthropometric data (weight, height, BMI) were translated to the child's size for example underweight, normal weight, overweight or (morbid) obesity based on international definitions and national growth chart (see Table, Supplemental Digital Content 1,,17,18 There is a large referral to our university hospital to investigate gastro-oesophageal reflux disease (GORD) and allergy. Additional investigations (e.g. placement of an impedance pH-study probe) were performed after endoscopy with children still asleep and blood was sampled for tests at the time of intravenous line insertion or during endoscopy. Results were retrieved following longitudinal follow-up.

The complete details of type, time of occurrence, interventions applied and outcome of adverse events were documented. According to the recent publication of the International Committee for the Advancement of Procedural Sedation, adverse events were divided into minor, intermediate and sentinels based upon their associated interventions and patient-centred outcomes.19 AREs were predefined in line with existing literature: apnoea as cessation of respiratory effort and the absence of a capnogram; bronchospasm as increased respiratory effort and wheeze on auscultation; laryngospasm as complete airway obstruction; and upper airway obstruction as a combination of airway obstruction with snoring.1,20 Any unplanned escalation of care due to adverse events was also detailed and follow-up was continued up to 4 months, if necessary.

Statistical methods

Primary outcome was the overall occurrence of adverse events and their predicting factors. Secondary outcome was to determine the impact of different anaesthetic/sedative agents. Child-related, procedure-related and anaesthesia-related characteristics were considered as independent variables. Data are expressed as mean ± SD or as number (%) as appropriate. For univariate analysis, the Student t test, χ2 test or Fisher exact test were used as appropriate. Risk ratio and odds ratio (OR) were obtained from a univariate logistic regression. A multivariate logistic regression model was built considering all clinically meaningful variables significant in the univariate analysis (unless specified otherwise) and applying a backward selection procedure. Model fit and stability were assessed by a likelihood ratio test by using the Hosmer and Lemeshow Goodness of Fit test. A two-tailed P value less than 0.05 was considered significant. Statistics were computed using SPSS version 25 (IBM, Chicago, Illinois, USA) and R software (R Development Core Team 2013, Vienna, Austria).


The cohort study included 4710 procedures performed in 3435 children (Fig. 1). Endoscopy was postponed in 52 children who presented with an acute respiratory illness, as defined earlier. The data from these children are not included in the final analysis as they were not anaesthetised. Table 1 summarises demographics and child-related, procedure-related and anaesthesia-related characteristics studied. Sexes were equally distributed. Mean age was 8.5 ± 4.4 years (range 3 months to 16 years). Although 2718 children (79.1%) were aged at least 4 years, 36 (1.0%) and 309 (9.0%) were less than 1 year and less than 2 years, respectively. Children's size, as defined above, deviated from normal in 1071 children (31.0%): 14.8% were underweight, 10.1% overweight and 6.1% (morbidly) obese. These data were missing in 20 patients, none of whom experienced an adverse event. Overall, 64.0% of patients were ASA physical status 1 or 2 and 36.0% ASA 3. The following comorbidities were documented: respiratory comorbidities in 1299 children (37.8%); recent respiratory infection in 799 (23.3%); both respiratory comorbidity and infection in 736 (21.4%); psychological disorders in 605 (17.6%); neurological comorbidities in 505 (14.7%); history of prematurity in 219 (6.4%) and congenital cardiac abnormalities in 106 (3.1%). Primary or secondary GORD was diagnosed in 1385 children (41.1%) and allergy in 1551 (45.4%).

Fig. 1:
Study flow diagram.
Table 1:
Demographics and child, procedure and anaesthesia-related characteristics of patients included in study

Upper gastrointestinal endoscopy, with or without lower gastrointestinal endoscopy, was performed in 87.1% of children. The reason for the procedure was gastrointestinal-related problems in 87.8% of children; the remaining 12.2% complained mainly of recurrent respiratory problems with or without gastrointestinal problems. Only a few children (<0.5%) were sedated by an anaesthesiologist trainee under direct supervision. Propofol was administered in 3428 patients (99.8%), in co-administration with ketamine in 525 (15.3%) and after sevoflurane induction in 77 (2.2%). A combination of sevoflurane, propofol and ketamine was used in 14 cases and another 11 children received sufentanil, etomidate or midazolam alone or in combination. The mean duration of the procedure, and times to discharge from PACU and to home were 18 ± 13, 36 ± 11 and 108 ± 16 min, respectively.

Table 2 outlines the type and the timing of occurrence of adverse events. Overall, 116 (3.4%) adverse events and 113 AREs (3.3%) were reported, of which 55 (1.6%) were minor, 58 (1.7%) intermediate and three (0.09%) sentinel events. Minor AREs were easily managed by supplemental oxygen, airway repositioning or suctioning. One child required additional sedatives in PACU for myoclonus. Another child experienced local redness at the site of injection. No intervention was performed and extensive allergic testing did not reveal any allergy.

Table 2:
Type and timing of the occurrence of adverse events and interventions performed19

Table 3 details the data of 11 children (0.3%) in whom MAC failed. Unplanned tracheal intubation was needed for three sentinel events, two because of apnoea and one because of severe laryngospasm. The tracheas of the first two children were successfully extubated at the end of the procedure. The third patient was admitted to the paediatric intensive care unit and was the only child in whom medical care was escalated. In eight out of 11 children, the procedure was interrupted because the ARE recurred. No child experienced pulmonary aspiration or cardiovascular events. No other sedation-related adverse events were identified on the paediatric ward and no procedure-related adverse events were reported. Endoscopists scored the ease of endoscopy performance as satisfactory in 98.5% of all children.

Table 3:
Details of children who had failed monitored anaesthesia care for gastrointestinal endoscopy procedures

After univariate analysis, 19 potential predictors were found to be significantly associated with the occurrence of adverse events (Table 4). They were: ASA physical status; age; weight; height; children's size; the presence of respiratory comorbidities, or recent respiratory infection, or both; neurological comorbidities; history of prematurity; psychological disorders; GORD; allergy to food; upper versus lower gastrointestinal endoscopy; respiratory reason for endoscopy with or without gastrointestinal problems; induction dose of propofol; induction dose of ketamine; co-administration of ketamine and propofol and propofol administration after sevoflurane induction.

Table 4:
Univariate analysis: risk factors associated with the occurrence of adverse events

Significant factors were entered in the multivariate analysis with the exception of weight and height (which are proxy for age and children's size), psychological disorders and reason for endoscopy, as these were highly associated with neurological and respiratory comorbidities respectively (P < 0.0001), and also induction dose of ketamine as this was only administered in 525 children. Applying a backward selection procedure revealed 12 independent predictors of adverse events (Table 5): age (OR 0.92, 95% CI 0.86 to 0.97, P = 0.002); underweight (OR 1.78, CI 1.01 to 3.06, P = 0.039); overweight (OR 2.20, CI 0.99 to 4.51, P = 0.039); (morbid) obesity (OR 4.25, CI 1.43 to 11.42, P = 0.006); presence of respiratory comorbidities only (OR 8.18, CI 4.20 to 15.95, P < 0.001); recent respiratory infection only (OR 23.55, CI 7.21 to 68.07, P < 0.001); both respiratory comorbidities and recent respiratory infection (OR 17.46, CI 9.33 to 33.55, P < 0.001); neurological comorbidities (OR 2.18, CI 1.22 to 3.80, P = 0.007); upper gastrointestinal endoscopy (OR 5.66, CI 2.38 to 16.80, P < 0.001); propofol in co-administration with ketamine (OR 10.34, CI 6.37 to 16.88, P < 0. 001); propofol administration after sevoflurane induction (OR 44.95, CI 15.20 to 131.12, P < 0.001); and propofol induction dose (OR 18.97, CI 11.58 to 32.09, P < 0.001).

Table 5:
Multivariate logistic regression analysis to predict the occurrence of adverse events

Analysing age as a continuous variable demonstrated that for each increasing year of age, the odds of developing adverse events decreased by 0.9 (CI 0.87 to 0.97). However, this alteration assumes a linear relationship. To explore further the potential effect of young age on the occurrence of adverse events, the population was arbitrarily divided into two groups: less than 2 years and at least 2 years of age. Posthoc secondary univariate analyses revealed a significantly higher risk of adverse events (risk ratio 3.7, CI 2.4 to 5.5; OR 3.9, CI 2.6 to 6.1; P < 0.0001) in children below 2 years of age, and we also observed significantly more respiratory comorbidities in association with recent respiratory infections (P < 0.0001) in this age group compared with children at least 2 years of age. Given the small number of children less than 1 year of age, separate analysis was not possible for that age group.


In this large database, we determined the prevalence and predictors of adverse events associated with MAC for paediatric gastrointestinal endoscopy. Young age, underweight, overweight, (morbid) obesity, the presence of respiratory comorbidities with or without recent respiratory infection, neurological comorbidities, upper gastrointestinal endoscopy, co-administration of propofol with ketamine or after sevoflurane induction, and induction dose of propofol were identified as independent risk factors.

In line with previous research, our data emphasised that AREs are predominant in paediatric sedation practices4–6,10,21,22 and identified young age, obesity, respiratory comorbidities and upper gastrointestinal endoscopy as independent risk factors.4,7,8 However, our findings provide additional and new information on MAC safety issues. In clinical practice, a clear-cut distinction between a recent respiratory infection and a chronic respiratory illness may be challenging in small children and studies assessing the effect of both comorbidities on the rate of adverse events in sedation practices are limited.4,5,23 In our cohort, 799 children (23.3%) presented with a recent respiratory infection, of whom 736 (21.4%) also had respiratory comorbidities. However, the procedures could not be postponed for at least 2 weeks, as recommended,20 because the gastrointestinal endoscopy was required in 420 children (12.2%) specifically because of recurrent respiratory problems, and in the others either a diagnosis needed to be confirmed or the benefits for the individual child overweighed those of cancellation. Both conditions of respiratory comorbidities and recent respiratory infection regardless of combination or not may be associated with airway hypersensitivity or impaired respiratory function and increased significantly the adverse events rate in our population. Our findings therefore emphasise that, in addition to respiratory comorbidities, a recent respiratory infection (within 2 weeks before the procedure) should also be considered when evaluating the individual risk of MAC.

The contribution of neurological pathologies as a risk factor for deep sedation is poorly explored. They represent a wide spectrum of significant comorbidities, reduced physiological reserves and polypharmacy, all of which may complicate sedation management. Unlike previous studies,24,25 the presence of neurological comorbidities was identified as an independent predictor in our cohort and increased the odds of adverse events by 2.2.

Anthropometric parameters are generally used to describe patient population. When they reflected child's size, they also proved to be useful as predictors of adverse events in our series. In addition to childhood (morbid) obesity, we also demonstrated, respectively, 2.2 and 1.8 times higher odds of adverse events in children who were either overweight or underweight when compared children of normal weight. This reflects the early impact of obesity on the respiratory system in overweight children. Being underweight is mostly the result of already established significant comorbidities and is characterised by muscular weakness. When anaesthetic drugs are administered, the restricted respiratory physiology may inappropriately compensate for suppressed respiratory drive, predisposing these children to adverse events. Although our findings merit further research, they highlight the need for awareness and vigilant monitoring when sedating this population.

ASA physical status traditionally helps the anaesthesiologist to estimate the severity of accompanying diseases with relevance to both anaesthesia and institutional resources. Although ASA physical status 3 was represented in more than a third of our population, we failed to demonstrate any predictive role for adverse events when adjusting for other factors. This finding is in line with some other reports.20,26 In our series, a child-specific risk profile based on age, size, respiratory and neurological diseases helped more reliably to predict the occurrence of adverse events. Therefore, we suggest that both ASA physical status and these risk factors should be used as complementary to each other when evaluating the risk of MAC.

The majority of adverse events occurred at induction or initiation of the procedure and their incidence varied significantly with the combined use of ketamine or sevoflurane with propofol, all of which are potent anaesthetic agents. Propofol and sevoflurane have been shown to cause dynamic airways collapse and to suppress respiratory drive whereas ketamine increases the risk of adverse events in a dose-dependent fashion.22,27–29 The combination of these drugs has probably amplified these effects and contributed to the increased rate of adverse events in children in our cohort who received a combination of drugs. Insufficient information exists to guide dosing regimens for propofol in paediatric MAC. In our study, we also sought to investigate the relationship between propofol induction dose and adverse events during paediatric endoscopy. We demonstrated that the odds of adverse events increased with each increasing unit dose of propofol (1 mg kg−1). However, this finding needs to be explored further in prospective dose–response studies and should be interpreted cautiously as the propofol dosages both at induction and maintenance were not predefined in this observational cohort. Moreover, propofol dose calculation normalised for total body weight is considered to be different in diverse paediatric age and weight groups.

Both allergy and GORD were extensively investigated and probably overreported in our population by comparison with the normal population because of a positive bias of recruitment in an academic paediatric gastroenterology department. Nevertheless, neither allergy nor GORD could be identified as an independent risk predictor when adjusting for other factors. This underlines the fact that identification of children at risk should rely on clinical, rather than laboratory or endoscopy assessment.

Biber et al.4 reported an overall adverse event rate of 4.8% during procedural sedation for gastrointestinal endoscopy in a paediatric population that consisted of 8.1% of ASA physical status 3 children. Despite a high proportion of ASA 3 patients in the present report (36%), the overall adverse event rate is comparatively low (3.4%). All adverse events also had an uneventful course for several reasons, among which were the experience of the team and high case volume, which probably allowed the early recognition and appropriate treatment of adverse events before they became severe.

Patino et al.7 demonstrated a significantly higher incidence of adverse events in children undergoing upper gastrointestinal endoscopy in the supine position using the native airway (45.9%). In our cohort, whether the child's position independently affected the outcome remains undetermined. Upper gastrointestinal endoscopies were exclusively performed in the lateral position. Lower gastrointestinal endoscopies were initiated in the lateral position but the child's position could change at a later stage, if necessary. Lateral positioning theoretically decreases the risk of airway obstruction compared with supine positioning. In addition, previous reports suggested that the more the airway stimulation, the higher the risk of developing adverse events, especially in small children with airway hypersensitivity.20 Based on their experience, the authors believe that MAC using propofol alone adequately blunted airway reflexes while using natural airways in the lateral position and contributed to the low adverse event rate in our series. Moreover, according to our endoscopists, our technique gives easy access to introduce the endoscope into the oropharynx and to enter the oesophagus.

The present cohort has limitations due to its observational, retrospective and single-centre nature, which may reduce the generalisability of our data. We also acknowledge that many minor interventions performed by anaesthesiologists, such as airway repositioning, were perhaps not always documented. This might have led to underreporting of minor adverse events. Except for AREs, very few sedation-related adverse events were documented in our report. This reflects either underreporting or self-limiting character of those adverse events but deserves more attention in reporting in future. Moreover, we cannot exclude that some episodes of oxygen desaturation, which were reported as such, were actually a consequence of another concomitant adverse event.

Our data also have strengths. Our findings demonstrated that MAC can provide optimal procedural conditions with a high safety profile and may be used as an alternative for the majority of paediatric gastrointestinal endoscopy procedures provided that effective presedation screening and expertise are in place. Young age proved to be an important risk factor not only because of the overall recognised vulnerability of this population but also as a result of significant comorbidities these children carry for which they require endoscopy procedures. Moreover, our cohort highlights the impact of new patient's comorbidities, for example underweight and overweight status, respiratory infections and neurological comorbidities. Propofol induction dose as well as its use in combination with ketamine and after sevoflurane induction appeared to be important and these need to be explored in prospective studies.

In conclusion, while adverse event occurred rarely in our unit, predictive factors were identified. These risk factors could be included in a risk assessment sheet based on clinical information and they should allow modification of the anaesthetic management to reduce the risk of adverse events.

Acknowledgements relating to this article

Assistance with the study: the authors would like to express their gratitude to all nurse practitioners from the Department of Anaesthesiology and Perioperative Medicine for their professional assistance during the sedation procedures, to Mrs Nicole Stijlemans for her outstanding support and assistance in collecting data and to Mrs Veerle Van Mossevelde for her proficient assistance in the completion of the administrative tasks.

Financial support and sponsorship: the study was supported by departmental funds.

Conflicts of interest: none.

Presentation: none.


1. Habre W, Disma N, Virag K, et al. APRICOT Group of the European Society of Anaesthesiology Clinical Trial Network. Incidence of severe critical events in paediatric anaesthesia (APRICOT): a prospective multicentre observational study in 261 hospitals in Europe. Lancet Respir Med 2017; 5:412–425.
2. Chung HK, Lightdale JR. Sedation and monitoring in the pediatric patient during gastrointestinal endoscopy. Gastrointest Endosc Clin N Am 2016; 26:507–525.
3. Lightdale JR, Acosta R, Shergill AK, et al. American Society for Gastrointestinal Endoscopy Standards of Practice Committee. Modifications in endoscopic practice for pediatric patients. Gastrointest Endosc 2014; 79:699–710.
4. Biber JL, Allareddy V, Allareddy V, et al. Prevalence and predictors of adverse events during procedural sedation anesthesia-outside the operating room for esophagogastro-duodenoscopy and colonoscopy in children: age is an independent predictor of outcomes. Pediatr Crit Care Med 2015; 16:e251–e259.
5. Barry N, Miller M, Ryshen K, et al. Etiology of postanesthetic and postsedation events on the inpatient ward: data from a rapid response team at a tertiary care children's hospital. Paediatr Anaesth 2016; 26:504–511.
6. Krauss B, Green SM. Procedural sedation and analgesia in children. Lancet 2006; 367:766–780.
7. Patino M, Glynn S, Soberano M, et al. Comparison of different anesthesia techniques during esophagogastroduedenoscopy in children: a randomised trial. Paediatr Anaesth 2015; 25:1013–1019.
8. Potié A, Prégardien C, Pirotte T, et al. Evaluation of the Explorer Endoscopy Mask® for esogastroduodenoscopy in children: a retrospective study of 173 cases. Paediatr Anaesth 2016; 26:649–654.
9. van Beek EJ, Leroy PL. Safe and effective procedural sedation for gastrointestinal endoscopy in children. J Pediatr Gastroenterol Nutr 2012; 54:171–185.
10. Rajasekaran S, Hackbarth RM, Davis AT, et al. The safety of propofol sedation for elective nonintubated esophagogastroduodenoscopy in pediatric patients. Pediatr Crit Care Med 2014; 15:e261–e269.
11. American Society of Anesthesiologists. Position on monitored anesthesia care. Available at: Approved by the House of Delegates on October 25, 2005, and last amended on October 17, 2018.
12. Mellin-Olsen J, Staender S, Whitaker DK, et al. The Helsinki declaration on patient safety in anaesthesiology. Eur J Anaesthesiol 2010; 27:592–597.
13. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med 2007; 4:e296.
14. Ramsay MA, Savege TM, Simpson BR, et al. Controlled sedation with alphaxalone-alphadolone. Br Med J 1974; 2:656–659.
15. Najafi N, Veyckemans F, Van de Velde A, et al. Usability of dexmedetomidine for deep sedation in infants and small children with respiratory morbidities. Acta Anaesthesiol Scand 2016; 60:865–873.
16. Aldrete JA. The postanesthesia recovery score revisited. J Clin Anesth 1995; 7:89–91.
17. Cole TJ, Flegal KM, Nicholls D, et al. Body mass index cut offs to define thinness in children and adolescents: international survey. BMJ 2007; 335:194–201.
18. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320:1240–1243.
19. Roback MG, Green SM, Andolfatto G, et al. Tracking and Reporting Outcomes Of Procedural Sedation (TROOPS): standardized quality improvement and research tools from the International Committee for the Advancement of Procedural Sedation. Br J Anaesth 2018; 120:164–172.
20. von Ungern-Sternberg BS, Boda K, Chambers NA, et al. Risk assessment for respiratory complications in paediatric anaesthesia: a prospective cohort study. Lancet 2010; 376:773–783.
21. Bhatt M, Johnson DW, Chan J, et al. Risk factors for adverse events in emergency department procedural sedation for children. JAMA Pediatr 2017; 171:957–964.
22. Green SM, Roback MG, Krauss B, et al. Predictors of airway and respiratory adverse events with ketamine sedation in the emergency department: an individual-patient data meta-analysis of 8,282 children. Ann Emerg Med 2009; 54:158–168.
23. Mallory MD, Travers C, McCracken CE, et al. Upper respiratory infections and airway adverse events in pediatric procedural sedation. Pediatrics 2017; 140:e20170009.
24. Grunwell JR, McCracken C, Fortenberry J, et al. Risk factors leading to failed procedural sedation in children outside the operating room. Pediatr Emerg Care 2014; 30:381–387.
25. Kiringoda R, Thurm AE, Hirschtritt ME, et al. Risks of propofol sedation/anesthesia for imaging studies in pediatric research: eight years of experience in a clinical research center. Arch Pediatr Adolesc Med 2010; 164:554–560.
26. Malviya S, Voepel-Lewis T, Chiravuri SD, et al. Does an objective system-based approach improve assessment of perioperative risk in children? A preliminary evaluation of the ‘NARCO’. Br J Anaesth 2011; 106:352–358.
27. Ehsan Z, Mahmoud M, Shott SR, et al. The effects of anesthesia and opioids on the upper airway: a systematic review. Laryngoscope 2016; 126:270–284.
28. Simons JC, Pierce E, Diaz-Gil D, et al. Effects of depth of propofol and sevoflurane anesthesia on upper airway collapsibility, respiratory genioglossus activation, and breathing in healthy volunteers. Anesthesiology 2016; 125:525–534.
29. Grunwell JR, Travers C, Stormorken AG, et al. Pediatric procedural sedation using the combination of ketamine and propofol outside of the emergency department: a report from the Pediatric Sedation Research Consortium. Pediatr Crit Care Med 2017; 18:e356–e363.

Supplemental Digital Content

© 2019 European Society of Anaesthesiology