Share this article on:

Safe and Effective Procedural Sedation for Gastrointestinal Endoscopy in Children

van Beek, Elke J.A.H.; Leroy, Piet L.J.M.

Journal of Pediatric Gastroenterology and Nutrition: February 2012 - Volume 54 - Issue 2 - p 171–185
doi: 10.1097/MPG.0b013e31823a2985
Invited Review

Objective: The aim of the study was to assess, by a review of published evidence, the safest and most effective way to provide procedural sedation (PS) in children undergoing gastrointestinal endoscopy (GIE).

Methods: The databases MEDLINE, Cochrane Library, and Embase were used. Search terms “endoscopy, gastrointestinal” or “endoscopy, digestive system” were combined with “sedation,” “conscious sedation,” “moderate sedation,” “deep sedation,” and “hypnotics and sedatives.” The final review was restricted to studies reporting specifically on safety (incidences of adverse events) and/or effectiveness (time characteristics, need for supplemental sedation, need for restraint, procedural success, provider satisfaction, and patient comfort) of PS for GIE in children younger than 18 years.

Results: The search yielded 182 references and the final selection included 11 randomized controlled trials (RCTs) and 15 non-RCTs. Six sedation categories were identified: propofol, opioid/benzodiazepine, premedication, ketamine-, sevoflurane-, and midazolam-based. Only a few RCTs have compared different categories. Opioid/benzodiazepine- and propofol-based PS have a similar safety profile and a low incidence of major adverse events. Propofol-based sedation turned out to be the most effective regimen, with effectiveness comparable to general anesthesia. The addition of midazolam, fentanyl, remifentanil, and/or ketamine to propofol may increase the effectiveness without creating more adverse events. Data on midazolam-, ketamine- and sevoflurane-based sedation were generally too limited to draw conclusions.

Conclusions: Despite a lack of RCTs containing all aspects of effectiveness and safety, the present evidence indicates propofol-based PS to be the best practice for PS in children undergoing GIE. Propofol can be safely administered by specifically trained nonanesthesiologists.

Pediatric Sedation Unit, Maastricht University Medical Centre, Maastricht, The Netherlands.

Address correspondence and reprint requests to Piet L.J.M. Leroy, Department of Pediatrics, Maastricht University Medical Centre, PB 5800, 6202 AZ Maastricht, The Netherlands (e-mail:

Received 29 March, 2011

Accepted 19 September, 2011

Drs van Beek and Leroy participated equally in this study.

The authors report no conflicts of interest.

Gastrointestinal endoscopy (GIE) is a well-established procedure for diagnosis and treatment in pediatric gastroenterology. Despite local anesthesia, reassurance, and distraction techniques, most children are unable to undergo GIE without being physically restrained (1,2). In addition, the recall of an unpleasant GIE generates a more negative attitude toward future endoscopies (3). Therefore, children undergoing GIE will need anesthesia or procedural sedation (PS) to guarantee optimum comfort and cooperation. PS can be defined as the use of sedative, analgesic, or dissociative drugs to relieve anxiety and pain associated with diagnostic and therapeutic procedures, while maintaining spontaneous ventilation (4,5). PS covers a large spectrum of sedatives and involves a wide range of sedation levels (Table 1). Mild sedation, formerly called anxiolysis, is typically the result of 1 standard dose of midazolam or by the breathing of 50% nitrous oxide (6). Moderate sedation, formerly called conscious sedation, is often used incorrectly to describe a state that is probably more like deep sedation; reflex withdrawal to a painful stimulus alone should not be considered as arousable (7). The term deep sedation has been under discussion in some professional groups because it may be indistinguishable from anesthesia. Although this point may be overstated, it has led to the recommendation that the same personnel, equipment, and facilities must manage both deep sedation and anesthesia. Some have proposed other descriptions of deep sedation/anesthesia: the terms light anesthesia or minimal anesthesia may be more appropriate to describe a technique in which the patient seems unconscious, although any appreciable stimulation is likely to rouse them (8,9).



PS provided in a context of inadequate training, competence, or safety precautions is associated with a greater likelihood of potentially fatal complications (4,10). PS by untrained individuals also bares the risk of ineffective sedation, possibly leading to unsuccessful and/or uncomfortable procedures. Pediatric gastroenterologists differ in their opinions on the effectiveness of PS, and many prefer anesthesiologists’ assistance (11). Depending on local anesthesiology resources, traditions, and personal experience, endoscopists have adopted different PS techniques (12,13). This diversity of sedation practices is, at least partially, the consequence of a lack of consensus on the safest and most effective regimen. By systematically reviewing the literature, we aimed to answer the following clinical question: what is the safest and most effective way to provide PS in children undergoing GIE?

Back to Top | Article Outline


The literature was searched in the databases MEDLINE, Cochrane Library, and Embase. The search terms “endoscopy, gastrointestinal” or “endoscopy, digestive system” were combined with the terms “sedation,” “conscious sedation,” “moderate sedation,” “deep sedation,” and “hypnotics and sedatives.” The search was restricted to articles including human subjects ages 0 to 18 years and published between January 1995 and January 2011. Additional records were searched in bibliographies of published studies, review articles, editorials, and guidelines. In the first step, articles addressing a mainly adult population were excluded. Next, both authors reviewed independently the full text of the remaining articles. Authors were not blinded for either the Journal or authors. Exclusion criteria were studies not reporting on the safety or effectiveness of PS, studies focusing on non-GIE procedures, studies on specific endoscopy procedures (eg, endoscopic retrograde cholangiopancreatography) or specific research populations (eg, critically ill children), review articles, editorials, policy statements, guidelines, and case reports. Disagreements were discussed and solved among authors. The final review was restricted to studies that reported specifically on “safety” and/or “effectiveness” of PS for GIE in children. Safety was assessed using incidences of adverse events: hypoxia, hypoventilation or apnea, laryngeal spasm, bradycardia, hypotension, vomiting, complications with fatal outcome or permanent sequelae, and the need for rescue interventions. Effectiveness was assessed using time characteristics (induction, recovery, and total sedation time), the need for supplemental sedation, procedural success, provider satisfaction, and patient comfort (stress or pain scores, recall, parent or patient satisfaction). We defined that an optimally effective PS technique should achieve near 100% predictable procedural success and timing, minimal induction and recovery times, and optimal patient comfort (absence of procedural pain, anxiety, or the need for restraint) (2).

To answer the clinical question, results from randomized controlled trials (RCTs) were used at first. If these sources remained inconclusive, then the results of non-RCTs were used to formulate a deliberate answer. Non-RCTs were classified as B1 for “comparative-prospective,” B2 for “comparative-retrospective,” and C for “noncomparative” studies.

Back to Top | Article Outline


The results of the literature search and selection process are summarized in Figure 1. The final selection included 25 articles, reporting on 26 studies (11 RCTs and 15 non-RCTs). One article included the results from a retrospective comparative study and an RCT (14). RCTs are summarized in an evidence table, listing details on methodology, results, limitations, and conclusions (Table 2). Non-RCT studies are summarized in Table 3. Esophagogastroduodenoscopy (EGD) was the most frequent GIE. Depending on the investigated sedative(s), RCTs can be divided into 3 categories: propofol-based, opioid + benzodiazepine combination, and studies on premedication. Non-RCT literature yielded 3 additional categories, that is, ketamine-, sevoflurane-, and midazolam-based protocols. Results are ordered according to the number of reported procedures. Studies on premedication are reported separately.











Back to Top | Article Outline

Propofol-based PS

Six RCTs (N = 561 procedures) and 4 non-RCTs (N = 3322 procedures) examined the safety and/or effectiveness of propofol-based PS. The majority of published propofol sedations (3420/3883; 88.1%) were performed by nonanesthesiologists, that is, pediatric intensivists (14,15) or specifically trained pediatricians (4,16).

Propofol-based PS is safe. Mild respiratory events occur frequently and major complications may happen rarely, but adverse events do not occur more frequently compared with other sedation regimens. On a total of 3883 reported propofol-based sedations, major respiratory complications like total airway obstruction, deep hypoxia, or apnea occurred 11 times (0.3%). No cases of intubation, resuscitation, permanent sequelae, or death were reported. In the largest study (Larsen et al (15); N = 2332), the incidence was only 0.04%, but the retrospective design may have caused underreporting. In all of the studies, patients routinely received additional oxygen. Overall, the incidence of adverse events is not higher compared with other sedative regimens included in this systematic review. Minor respiratory events, such as temporary desaturation or hypoventilation, occurred more frequently (up to 24% in Kaddu et al's study), particularly in infants younger than 1 year (incidence 35% compared with 12.5% in children older than 10 years; P < 0.02) (15). Three of the 6 children needing assisted ventilation in the study by Barbi et al (4) (N = 811 procedures) were also younger than 1 year. The other 3 and all 14 children with laryngospasm (overall incidence 1.7%) were younger than 4 years. Disma et al (12) reported an incidence of laryngospasm of 3.75% in children sedated with propofol alone. Adding midazolam or fentanyl to propofol resulted in no cases of laryngospasm. Clinically relevant hypotension, hypoperfusion, or bradycardia has not been reported.

Propofol-based PS is greatly effective. None of the retrieved articles reported in itself on all aspects of effectiveness as defined in the methodology section of this systematic review. The reported incidences of incomplete or failed procedure ranged from 0% (14) over 0.05% (15) to 0.4% (4,15). Kaddu et al (17) showed that deep sedation with propofol is an equivalent alternative for general anesthesia (GA). The time to achieve adequate sedation, the mean procedure time, and the recovery time were significantly shorter in children sedated with propofol, compared with the combined sedation with meperidine + midazolam (14). Only 3 studies assessed patient comfort or the need for restraint as outcome parameters for effectiveness. Paspatis et al (18) reported an incidence of extreme discomfort of 14.2% if propofol was used alone. Premedication with oral midazolam improved significantly the level of patient comfort during the procedure (0% extreme discomfort). Khoshoo et al (14) showed that propofol + midazolam was associated with the need for restraint in only 3% of cases, compared with 57% in the meperidine + midazolam group (P < 0.01). Total absence of recall was observed more frequently in the propofol + midazolam group (88% vs 55%; P < 0.05). Elitsur et al (19) reported the absence of pain and recall in 96% of patients. Propofol injection pain was reported to occur in 53% of cases, despite the addition of lidocaine to propofol (4).

The addition of ketamine, midazolam, or fentanyl to propofol-based PS may have beneficial effects on PS quality. Tosun et al (20) compared propofol + ketamine with propofol + fentanyl. Both combinations provided equally effective sedation and similar low incidences of mild respiratory events. The use of ketamine was associated with a better tolerance of endoscope insertion but higher incidences of cough, vomiting, dizziness, and diplopia. A low dose of ketamine before propofol is more effective in reducing propofol infusion pain, compared with a lidocaine–propofol mixture (16). Disma et al (12) showed that the addition of either fentanyl or midazolam caused less adverse events compared with propofol alone. This effect was mainly caused by a lower incidence of coughs and laryngeal spasms. No differences were seen between these 2 additional medication regimes regarding effectiveness or adverse events. Both Disma et al and Elitsur et al (12,19) showed that adding midazolam or fentanyl significantly lowers the necessary dose of propofol.

Back to Top | Article Outline

Opioid + Benzodiazepine-based PS

The effectiveness and/or safety of opioid + benzodiazepine-based sedation were studied in 3 RCTs (N = 94 procedures) and 6 non-RCTs (N = 2326 procedures). The administrating professionals were intensivists (14), gastroenterologists/endoscopists (21–24), or were not reported (13,25,26).

Opioid + benzodiazepine-based sedation is safe. Mild respiratory events occur frequently and major complications may happen rarely. Of a total of 2420 opioid + benzodiazepine-based sedations, major respiratory complications such as apnea, total airway obstruction, or deep desaturation occurred only 5 times (0.2%). In all of the studies, patients received oxygen routinely, except for the studies by Fishbein et al, Chuang et al, and Gilger at al (21,22,26).

Khoshoo et al (14) showed that the use of meperidine + midazolam was associated with a higher need for supplemental oxygen and ventilatory support, compared with the use of propofol + midazolam (17% vs 7% and 7% vs 3%, respectively; P < 0.05 for both differences). Ali et al (25) showed in a small RCT that fentanyl + midazolam and meperidine + midazolam are equally safe. Chuang et al (21) came to the same conclusion in their retrospective comparative study. Fishbein et al (26) compared intravenous (IV) meperidine + midazolam with the combination of intranasal midazolam and IV meperidine. Both regimens were equally safe and none of the patients experienced major adverse events. In their prospective study on IV fentanyl + midazolam, Mamula et al (13) reported 9% respiratory adverse events (8% transient desaturation, 1% prolonged desaturation, 0.2% apnea), 11% mild cardiovascular adverse events (3% hypertension, 8% hypotension), 0.6% rash, 1% agitation, and 5% vomiting. The following rescue interventions were reported: 9.5% extra oxygen therapy, 1.3% tactile stimulation, 0.4% jaw thrust, 0.2% bag and mask ventilation, and 1.6% IV fluid bolus. In a smaller prospective study, Squires et al (24) compared PS using meperidine + midazolam with general inhalational anesthesia. No major adverse events or desaturations were reported. Gilger et al (22) compared retrospectively meperidine + midazolam with meperidine + midazolam + ketamine and midazolam + ketamine. Mild respiratory adverse events, especially desaturations, occurred most frequently in the meperidine + midazolam group (30.7% vs 10.9% [P < 0.001] and 0.6% [P < 0.001], respectively).

The evidence indicates that effectiveness of opioid + benzodiazepine-based PS is suboptimal, especially regarding time characteristics and the need for restraint. None of the retrieved articles reported in itself on all aspects of effectiveness as defined in the methodology section of this systematic review. In their comparative study between propofol + midazolam and meperidine + midazolam, Khoshoo et al (14) reported equally high procedural success rates (±100%); however, the time to achieve adequate sedation, the mean procedure time, and the recovery time were significantly longer if midazolam + meperidine was used. Furthermore, restraint was significantly needed more in the meperidine + midazolam group, suggesting suboptimal effectiveness of this regimen. Fishbein et al (26) reported high incidences of major negative behavior during EGD in children sedated with IV midazolam + meperidine (19/20) or intranasal midazolam + IV meperidine (18/20). The intensity of negative behavior was significantly lower in the group receiving intranasal midazolam premedication. Ali et al (25) could not show any significant difference in effectiveness between fentanyl- and meperidine-based PS, but patient comfort was not an outcome parameter in their study. In the Mamula et al study (13) on fentanyl + midazolam, 11% of the colonoscopies were incomplete because of inadequate sedation. Gremse et al (23) found similar figures in their study on meperidine + midazolam. Gilger et al (22) showed that, compared with ketamine + midazolam, the use of meperidine + midazolam resulted more frequently in inadequate sedation (8.6% vs 3.1%), although the difference was not significant (P = 0.07). Finally, a prospective comparison between meperidine + midazolam and GA showed a lower procedural failure rate (0% vs 4.8%), lower need for restraint (0% vs 13%), and less direct and indirect indications of discomfort in the GA group (24).

Back to Top | Article Outline

Ketamine-based PS

Three non-RCT trials (N = 1056 procedures) have examined the safety and/or effectiveness of ketamine. In 2 studies, gastroenterologists administered ketamine (22,27). Aggarwal et al (28) did not report the administrating professional.

Ketamine-based PS is probably safe, but data are limited. Especially in EGD, ketamine is associated with a high chance of laryngeal spasm. Compared with meperidine + midazolam, Gilger et al (22) reported less desaturation in the ketamine + midazolam group. Aggarwal et al (28) found no particular major adverse events or complications, but it is unclear how vital parameters were assessed. A retrospective study by Green et al (27) reported in more detail on adverse events (Table 2). All of the patients routinely received oxygen therapy. Laryngeal spasm occurred in 9.5% of EGDs. The only significant independent predictor of laryngospasm in a multivariate analysis was age: 13.9% in preschool-age (6 years or older) children and 3.6% in school-age (older than 6 years) children (difference 10.3%, 95% CI 5.5%–14.9%).

Ketamine-based PS is probably effective, but data are limited. Reliable data on patient comfort and the need for restraint are missing. According to Gilger et al (22), inadequate sedation occurs less frequently in ketamine + midazolam sedation compared with meperidine + midazolam. Incidences of failed sedation ranged from 0% to 1.1% (27,29); however, data on patient comfort, the need for restraint, or the ease of the procedure are missing.

Back to Top | Article Outline

Midazolam Alone

One RCT (N = 61 procedures) and 2 non-RCTs (N = 275 procedures) have studied the safety and/or effectiveness of midazolam alone. Either anesthesiologists (30,31) or pediatricians (32) administered midazolam.

Data are too limited to draw conclusions on the safety of midazolam-based PS. Rafeey et al (31) compared oral and IV midazolam. Major adverse events did not occur. The mean oxygen saturation was lower in the IV group. Verhage et al (32) reported an overall incidence of adverse events of 3%. Nature and severity of events were not reported. Flumazenil was needed in 2.7% of cases. In a small comparative study, Lamireau et al (30) compared IV midazolam with halothane anesthesia. Desaturation below 90% occurred significantly more frequently in the midazolam group (50% vs 0%; P < 0.001). Midazolam-based PS is probably ineffective, but data are limited.

Lamireau et al (30) found that incomplete procedures and complete procedures under difficult conditions occurred significantly more frequently in patients sedated with IV midazolam compared with halothane anesthesia (50% vs 0% and 38.9% vs 0%, respectively). Rafeey et al and Verhage et al (31,32) reported procedural success rates of approximately 100%; however, both studies failed to report reliable data on patient comfort, adequacy of sedation, and ease of procedure. The significant increase in heart rate and blood pressure during endoscopy, as registered by Rafeey et al, suggests that midazolam is ineffective in reducing patient discomfort.

Back to Top | Article Outline

Sevoflurane Inhalation for PS

In a retrospective study, Montes and Bohn (33) compared inhaled sevoflurane (N = 67 procedures), administered by an anesthesiologist using an oropharyngeal tube, with 2 sedative regimens administered by pediatric intensivists: propofol alone (N = 114 procedures) and a combination of midazolam and/or fentanyl and/or ketamine and/or propofol (N = 67 procedures).

Sevoflurane-based PS may be safe, but data are limited. No severe adverse events were reported. Respiratory adverse events occurred rarely at similar incidences in the 3 groups. The incidence of hypotension was 0% in the sevoflurane group versus 13.2% and 7.5% in the other groups. None of these hypotensive episodes was considered relevant. Data are too limited to draw conclusions on the effectiveness of sevoflurane-based PS. Compared with the IV regimens, sevoflurane was characterized by a shorter recovery time, earlier discharge, and lower costs. The present study did not report on patient comfort, need for restraint, or overall procedural success; neither did it specify whether and how sevoflurane pollution in the ambient air was avoided (33).

Back to Top | Article Outline


Three RCTs (N = 107 procedures) specifically analyzed the safety and/or effectiveness of midazolam premedication. One RCT (N = 50 procedures) analyzed atropine premedication.

Midazolam premedication is probably safe for children undergoing PS for GIE, but data are limited. None of the studies recorded differences in safety between premedication and non–premedication groups.

Premedication with midazolam has positive effects on patient comfort. There is good evidence that oral midazolam premedication preceding a propofol- or an opioid-based PS improves significantly the ease and comfort of both IV catheter placement and separating the child from the parents. It also increases the level of patient comfort during the procedure (18,34). Following an opioid-based PS, the degree of partial or total amnesia is higher, compared with placebo, if oral midazolam premedication is given (83% vs 55%; P < 0.05) (34). In propofol-based PS, the addition of oral midazolam premedication lowers the propofol dose during the procedure but increases the recovery time (18). Fishbein et al (26) showed that premedication with intranasal midazolam before IV meperidine effectively reduces the intensity of negative behaviors; however, no effect was seen on the total number of negative behaviors. Furthermore, the intranasal administration itself was associated with negative behavior in 50% of cases.

Atropine premedication probably does not increase the safety or effectiveness of PS in children undergoing GIE, but data are limited. Hofley et al (35) compared atropine premedication with placebo in a small group of children sedated with fentanyl + midazolam or meperidine + midazolam. They concluded that atropine did not increase PS safety and provided no appreciable benefits on oral secretions, gastric motility, vomiting, facial flushing, or diphoria.

Back to Top | Article Outline


The available evidence suggests that propofol-based PS is the most effective regimen for PS during GIE in children (Table 4). Propofol guarantees an excellent level of procedural success, optimal timing, and maximal patient comfort, in particular if propofol is preceded by midazolam premedication. The addition to propofol of IV midazolam, fentanyl, or remifentanil may increase the effectiveness without generating more adverse events, provided that sedation is performed by trained PS practitioners under certain quality and safety conditions. Compared with GA, propofol-based PS seems to be an equally effective technique for selected cases of diagnosis and therapy. It must be noted, however, that there is a considerable risk that deep sedation with propofol may result in an actual sedative state close or similar to GA for a limited period of time with an increased risk. In a prospective study of children sedated with propofol by nonanesthesiologists, 90% of children reached a level of “anesthesia” albeit briefly (36). Nevertheless, this systematic review's search category was “sedation” and not “anesthesia,” so that irrespective of the actual sedation level, practitioners believed that they were sedating and not anesthetizing.



Propofol infusion–related pain can be prevented with a low dose of ketamine, which is, in children, a more effective approach compared with adding lidocaine to propofol. Opioid + benzodiazepine-based PS also results in high procedural success rates; however, the longer time needed to achieve sedation, the longer recovery time, and the significantly lower levels of patient comfort are substantial drawbacks compared with propofol-based PS. There is some evidence that the relatively high incidence of incomplete colonoscopies in opioid + benziodiazepine-based PS (up to 10%) is caused by ineffective sedation. We could find no study comparing the effectiveness of propofol and ketamine for GIE in children. The available evidence on ketamine suggests less inadequate sedation, compared with meperidine + midazolam; however, none of the articles reported data on patient comfort, the need for restraint, or the ease of the procedure. It is plausible that IV midazolam alone is not an effective way to provide PS for GIE in children; however, oral midazolam premedication preceding a propofol-based or an opioid-based PS improves significantly the ease and comfort of both IV catheter placement and separation of the child from the parents. It also increases the level of patient comfort during the procedure.

Comparing the safety outcome of propofol-based PS with the more traditional opioid + benzodiazepine-based PS does not reveal striking differences. Mild and transient hypoxia is reported as the most frequent adverse event in both strategies. Major respiratory adverse events, such as deep hypoxia, hypoventilation, apnea, and airway obstruction, seem to occur rarely but at similar rates. As a consequence, both regimens require the same safety precautions. In adult medicine, 3 recent meta-analyses did not show significant differences between propofol-based and traditional PS for hypoxemia and hypotension, except for fewer cardiorespiratory complications with propofol during colonoscopy (37–39). Khoshoo et al published the only available RCT comparing the safety of propofol-based versus opioid + benzodiazepine–based PS in children. In the propofol group, oxygen therapy and ventilatory support were significantly less needed. There is some evidence that laryngospasm occurs in about 9% of children sedated with ketamine for EGD (27), compared with 1.7% to 3.7% if propofol is used alone (4,12) and 0% if propofol is combined with midazolam or fentanyl (12). Therefore, it may be prudent to advise not to use ketamine for upper GIE and to add midazolam or fentanyl to a propofol PS.

The fact that, apart from 1 article (4), none of the studies made use of capnography monitoring may have caused underestimation of the real incidences of airway obstruction and hypoventilation. The relevance of capnography for early detecting imminent respiratory events during PS is well established (40–42). Its application during GIE was reviewed recently (43).

The similar low rate of adverse events in the different sedation regimens is not surprising. It has been shown that PS-related safety is determined by the circumstances, applied safety precautions, and professional skills rather than by specific pharmacological characteristics (10). In most studies included in this systematic review, competent and skilled professionals performed PS in accordance with widely accepted safety guidelines. Adherence to PS safety guidelines reduces the occurrence of PS-related adverse events (36,44,45).

The present review suggests the superior effectiveness of propofol compared with the more traditionally used opioid + benzodiazepine combination. The latter is generally considered safe in a nonanesthesiologist's hands, whereas the administration of the anesthetic propofol is usually restricted for safety reasons to anesthesiologists only. This distinction is remarkable given the results of this systematic review, showing a similar safety profile for both regimens. Furthermore, the vast majority of included propofol sedations were performed by nonanesthesiologists. Because of limited anesthesiology resources, propofol is being administered worldwide by nonanesthesiologists (specially trained nurses or endoscopists) for GIE in selected adult patients (46). An evidence-based guideline on this topic was published recently (47). Excellent evidence exists demonstrating that well-trained nonanesthesiologists may provide propofol sedation safely in children, including for gastrointestinal procedures (48). It has also been shown that within the setting of adequate training and strict safety measures, no differences exist in rates of major complications among different specialists (ie, anesthesiologists and nonanesthesiologists) (49). Appropriate safety precautions, monitoring, and professional skills, rather than professional title, are determinants for the safe and effective use of propofol for PS (50).

The present study has several limitations. At first, few RCTs compared different pharmacological techniques. The fact that different studies used different outcome measures or different definitions for safety and effectiveness is another important limitation. An objective or validated assessment of the more subjective measures such as satisfaction, ease of procedure, or patient comfort is missing in most studies. Consequently, it is presently impossible to summarize the available evidence in meta-analyses or to draw solid conclusions on best practices. Nevertheless, we believe that it is possible to draw prudent conclusions and to formulate practical recommendations. Because most of the studies reviewed included both upper and lower GIE, we choose not to differentiate between them. It is likely that optimal sedation characteristics and issues on safety and effectiveness are not the same for both procedures; however, we believe that the main conclusions would probably remain the same. Psychological distraction techniques may be useful adjuncts or partial alternatives for PS. We deliberately choose not to include these nonpharmacological methods in this systematic review. Finally, in individual patients, the optimal PS strategy may deviate from the conclusions of this systematic review. Although the benzodiazepine + opioid combination is not as good as propofol-based PS, it may be suitable for some children if they can tolerate moderate sedation. Also, unsedated endoscopy in selected greatly motivated children has been reported (51). In this respect, it is remarkable that in the Barbi et al (4) study among children older than 14 years, 26.6% preferred to have no sedation at all during upper GIE.

The limitations of this systematic review clearly identify a gap of scientific knowledge on this topic. Only well-designed, procedure-specific RCTs comparing validated outcome measures on effectiveness and safety between different sedative regimens in large numbers of children may yield a definite answer on our clinical question; however, given the results of this systematic review in general and the findings on patient comfort in particular, one may question whether setting up such trials is still ethically justifiable.

Back to Top | Article Outline


Despite the methodological limitations, the evidence gathered in this systematic review indicates propofol-based PS as the best practice for PS in children undergoing GIE. The addition to propofol of IV midazolam, fentanyl, or remifentanil may increase its safety and effectiveness. Oral midazolam premedication leads to a lower mean dose of propofol, less painful and easier IV access, easier separation from the parents, and greater patient comfort during endoscopy. A low dose of ketamine is effective in reducing propofol infusion pain. Propofol is likely to lead to an unconscious state in many children. Although this may not be unsafe, it does mean that it is not strictly sedation. Practitioners should be trained to manage this depth of sedation (ie, anesthesia) and parents and children warned that this may happen. Conversely, children who expect to be oblivious to the GIE during sedation with propofol should be told that they may be sedated (not anesthetized) and that they may recall some details of the procedure. Propofol can be safely administered to children by specifically trained nonanesthesiologists who provide PS in adherence to established safety guidelines. In the absence of these professionals and settings, it is advisable to refer children who need to undergo a GIE to an anesthesiologist for propofol-based PS or anesthesia.

Back to Top | Article Outline


1. Koh JL, Black DD, Leatherman IK, et al. Experience with an anesthesiologist interventional model for endoscopy in a pediatric hospital. J Pediatr Gastroenterol Nutr 2001; 33:314–318.
2. Leroy PL, Schipper DM, Knape HJ. Professional skills and competence for safe and effective procedural sedation in children: recommendations based on a systematic review of the literature. Int J Pediatr 2010;934298 [e-pub June 28].
3. Lewis Claar R, Walker LS, Barnard JA. Children's knowledge, anticipatory anxiety, procedural distress, and recall of esophagogastroduodenoscopy. J Pediatr Gastroenterol Nutr 2002; 34:68–72.
4. Barbi E, Petaros P, Badina L, et al. Deep sedation with propofol for upper gastrointestinal endoscopy in children, administered by specially trained pediatricians: a prospective case series with emphasis on side effects. Endoscopy 2006; 38:368–375.
5. Krauss B, Green SM. Procedural sedation and analgesia in children. Lancet 2006; 367:766–780.
6. Babl FE, Oakley E, Seaman C, et al. High-concentration nitrous oxide for procedural sedation in children: adverse events and depth of sedation. Pediatrics 2008; 121:e528–e532.
7. Cote CJ. Round and round we go: sedation—what is it, who does it, and have we made things safer for children? Paediatr Anaesth 2008; 18:3–8.
8. Sury MR, Smith JH. Deep sedation and minimal anesthesia. Paediatr Anaesth 2008; 18:18–24.
9. Vangerven M, Van Hemelrijck J, Wouters P, et al. Light anaesthesia with propofol for paediatric MRI. Anaesthesia 1992; 47:706–707.
10. Cote CJ, Notterman DA, Karl HW, et al. Adverse sedation events in pediatrics: a critical incident analysis of contributing factors. Pediatrics 2000; 105:805–814.
11. Lightdale JR, Mahoney LB, Schwarz SM, et al. Methods of sedation in pediatric endoscopy: a survey of NASPGHAN members. J Pediatr Gastroenterol Nutr 2007; 45:500–502.
12. Disma N, Astuto M, Rizzo G, et al. Propofol sedation with fentanyl or midazolam during oesophagogastroduodenoscopy in children. Eur J Anaesthesiol 2005; 22:848–852.
13. Mamula P, Markowitz JE, Neiswender K, et al. Safety of intravenous midazolam and fentanyl for pediatric GI endoscopy: prospective study of 1578 endoscopies. Gastrointest Endosc 2007; 65:203–210.
14. Khoshoo V, Thoppil D, Landry L, et al. Propofol versus midazolam plus meperidine for sedation during ambulatory esophagogastroduodenoscopy. J Pediatr Gastroenterol Nutr 2003; 37:146–149.
15. Larsen R, Galloway D, Wadera S, et al. Safety of propofol sedation for pediatric outpatient procedures. Clin Pediatr (Phila) 2009; 48:819–823.
16. Barbi E, Marchetti F, Gerarduzzi T, et al. Pretreatment with intravenous ketamine reduces propofol injection pain. Paediatr Anaesth 2003; 13:764–768.
17. Kaddu R, Bhattacharya D, Metriyakool K, et al. Propofol compared with general anesthesia for pediatric GI endoscopy: is propofol better? Gastrointest Endosc 2002; 55:27–32.
18. Paspatis GA, Charoniti I, Manolaraki M, et al. Synergistic sedation with oral midazolam as a premedication and intravenous propofol versus intravenous propofol alone in upper gastrointestinal endoscopies in children: a prospective, randomized study. J Pediatr Gastroenterol Nutr 2006; 43:195–199.
19. Elitsur Y, Blankenship P, Lawrence Z. Propofol sedation for endoscopic procedures in children. Endoscopy 2000; 32:788–791.
20. Tosun Z, Aksu R, Guler G, et al. Propofol-ketamine vs propofol-fentanyl for sedation during pediatric upper gastrointestinal endoscopy. Paediatr Anaesth 2007; 17:983–988.
21. Chuang E, Wenner WJ Jr, Piccoli DA, et al. Intravenous sedation in pediatric upper gastrointestinal endoscopy. Gastrointest Endosc 1995; 42:156–160.
22. Gilger MA, Spearman RS, Dietrich CL, et al. Safety and effectiveness of ketamine as a sedative agent for pediatric GI endoscopy. Gastrointest Endosc 2004; 59:659–663.
23. Gremse DA, Kumar S, Sacks AI. Conscious sedation with high-dose midazolam for pediatric gastrointestinal endoscopy. South Med J 1997; 90:821–825.
24. 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.
25. Ali S, Davidson DL, Gremse DA. Comparison of fentanyl versus meperidine for analgesia in pediatric gastrointestinal endoscopy. Dig Dis Sci 2004; 49:888–891.
26. Fishbein M, Lugo RA, Woodland J, et al. Evaluation of intranasal midazolam in children undergoing esophagogastroduodenoscopy. J Pediatr Gastroenterol Nutr 1997; 25:261–266.
27. Green SM, Klooster M, Harris T, et al. Ketamine sedation for pediatric gastroenterology procedures. J Pediatr Gastroenterol Nutr 2001; 32:26–33.
28. Aggarwal A, Ganguly S, Anand VK, et al. Efficacy and safety of intravenous ketamine for sedation and analgesia during pediatric endoscopic procedures. Indian Pediatr 1998; 35:1211–1214.
29. Agrawal D, Manzi SF, Gupta R, et al. Preprocedural fasting state and adverse events in children undergoing procedural sedation and analgesia in a pediatric emergency department. Ann Emerg Med 2003; 42:636–646.
30. 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.
31. Rafeey M, Ghojazadeh M, Feizo Allah Zadeh H, et al. Use of oral midazolam in pediatric upper gastrointestinal endoscopy. Pediatr Int 2010;52:191–5.
32. Verhage J, Mulder CJ, Willekens FL. Intravenous midazolam sedation in pediatric diagnostic upper digestive endoscopy. A prospective study in a general hospital. Rom J Gastroenterol 2003; 12:273–276.
33. Montes RG, Bohn RA. Deep sedation with inhaled sevoflurane for pediatric outpatient gastrointestinal endoscopy. J Pediatr Gastroenterol Nutr 2000; 31:41–46.
34. Liacouras CA, Mascarenhas M, Poon C, et al. Placebo-controlled trial assessing the use of oral midazolam as a premedication to conscious sedation for pediatric endoscopy. Gastrointest Endosc 1998; 47:455–460.
35. Hofley MA, Hofley PM, Keon TP, et al. A placebo-controlled trial using intravenous atropine as an adjunct to conscious sedation in pediatric esophagogastroduodenoscopy. Gastrointest Endosc 1995; 42:457–460.
36. Barbi E, Gerarduzzi T, Marchetti F, et al. Deep sedation with propofol by nonanesthesiologists: a prospective pediatric experience. Arch Pediatr Adolesc Med 2003; 157:1097–1103.
37. McQuaid KR, Laine L. A systematic review and meta-analysis of randomized, controlled trials of moderate sedation for routine endoscopic procedures. Gastrointest Endosc 2008; 67:910–923.
38. Qadeer MA, Vargo JJ, Khandwala F, et al. Propofol versus traditional sedative agents for gastrointestinal endoscopy: a meta-analysis. Clin Gastroenterol Hepatol 2005; 3:1049–1056.
39. Singh H, Poluha W, Cheung M, et al. Propofol for sedation during colonoscopy. Cochrane Database Syst Rev. 2008;4:CD006268.
40. Cote CJ, Wilson S. Guidelines for monitoring and management of pediatric patients during and after sedation for diagnostic and therapeutic procedures: an update. Pediatrics 2006; 118:2587–2602.
41. Krauss B, Hess DR. Capnography for procedural sedation and analgesia in the emergency department. Ann Emerg Med 2007; 50:172–181.
42. Yarchi D, Cohen A, Umansky T, et al. Assessment of end-tidal carbon dioxide during pediatric and adult sedation for endoscopic procedures. Gastrointest Endosc 2009; 69:877–882.
43. Gerstenberger PD. Capnography and patient safety for endoscopy. Clin Gastroenterol Hepatol 2010;8:423–5.
44. Hoffman GM, Nowakowski R, Troshynski TJ, et al. Risk reduction in pediatric procedural sedation by application of an American Academy of Pediatrics/American Society of Anesthesiologists process model. Pediatrics 2002; 109:236–243.
45. Pitetti R, Davis PJ, Redlinger R, et al. Effect on hospital-wide sedation practices after implementation of the 2001 JCAHO procedural sedation and analgesia guidelines. Arch Pediatr Adolesc Med 2006; 160:211–216.
46. Rex DK, Deenadayalu VP, Eid E, et al. Endoscopist-directed administration of propofol: a worldwide safety experience. Gastroenterology 2009; 137:1229–1237.
47. Dumonceau JM, Riphaus A, Aparicio JR, et al. European Society of Gastrointestinal Endoscopy, European Society of Gastroenterology and Endoscopy Nurses and Associates, and the European Society of Anaesthesiology Guideline: non-anesthesiologist administration of propofol for GI endoscopy. Endoscopy 2010;42:960–74.
48. Cravero JP, Beach ML, Blike GT, et al. The incidence and nature of adverse events during pediatric sedation/anesthesia with propofol for procedures outside the operating room: a report from the Pediatric Sedation Research Consortium. Anesth Analg 2009; 108:795–804.
49. Couloures KG, Beach M, Cravero JP, et al. Impact of provider specialty on pediatric procedural sedation complication rates. Pediatrics 2011; 127:e1154–e1160.
50. Green SM, Krauss B. Barriers to propofol use in emergency medicine. Ann Emerg Med 2008; 52:392–398.
51. Bishop PR, Nowicki MJ, May WL, et al. Unsedated upper endoscopy in children. Gastrointest Endosc 2002; 55:624–630.
52. Abu-Shahwan I, Mack D. Propofol and remifentanil for deep sedation in children undergoing gastrointestinal endoscopy. Paediatr Anaesth 2007; 17:460–463.

children; effectiveness; gastrointestinal endoscopy; procedural sedation; safety

Copyright 2012 by ESPGHAN and NASPGHAN