Propofol sedation/anesthesia for children undergoing diagnostic and therapeutic procedures outside the operating room has been an area of significant controversy over the last 10 yrs. Professional organizations have proposed conflicting guidelines on the appropriateness or inappropriateness of the use of propofol in this clinical setting. Many published review articles and editorials have described the practice of propofol sedation/anesthesia by various pediatric subspecialists. Reports of propofol sedation/anesthesia for procedures in children (by all types of sedation/anesthesia providers) are common in the pediatric literature. In particular, emergency medicine specialists and pediatric intensivists have embraced propofol sedation/anesthesia because of its effectiveness and the rapidity of onset and emergence.1–3
Of all patients receiving deep sedation/anesthesia for diagnostic and therapeutic procedures, the pediatric population is the subgroup at the highest risk level and with the lowest tolerance for error.4,5 In addition, deep sedation/anesthesia is required more often in children to achieve acceptable conditions during a procedure. (We cannot simply tell a 2 year old to hold still for an hour-long magnetic resonance imaging [MRI] scan.) The terminology used concerning propofol sedation/anesthesia is particularly problematic, since many providers refer to this practice as “sedation” or “procedural sedation and analgesia” when often the patients might easily meet the definition of what anesthesiologists and The Joint Commission would term “anesthesia.”6 This confusion is understandable, since children can easily slip from one level to a deeper level, and one would have to constantly be stimulating children to test their responsiveness to truly define their state.7
Published reports of sedation/anesthesia practice with propofol have been limited to relatively small studies (hundreds of patients) that have involved a single institution’s experience with the drug in a selected group of patients. These studies are under-powered to report infrequent adverse events, although almost all conclude that their practice is indeed safe. To allow a more comprehensive approach to the study of the safety and reliability of pediatric anesthesia, the Pediatric Sedation Research Consortium (PSRC) was created as a collaborative group of institutions dedicated to improving sedation/anesthesia practice through sharing of prospective observational outcome data regarding sedation/anesthesia encounters. Research support from the National Patient Safety Foundation has funded the design, development and management of the database. To shed some light on the nature of propofol anesthesia, we report the experience of sedation/anesthesia involving a large number of patients (cared for in a variety of settings) to describe the nature and frequency of adverse events when using propofol for procedural sedation/anesthesia outside the operating room.
The data collection methodology used by the PSRC has been described in a report on the first 30,000 sedation/anesthesia administrations (all drugs).8 Thirty-seven locations, including large children’s hospitals, children’s hospitals within hospitals and general/community hospitals, were self-selected for involvement in the PSRC data sharing group. There was no specific selection criterion for participation in the consortium; however, any interested institutions were required to identify a primary investigator and agree to a standardized methodology for data collection and quality oversight from sedation/anesthesia sites at their location. All institutions had to agree to perform periodic audits of records to assure data integrity. Thus, the PSRC consists of anesthesiologists, pediatric medical subspecialists, emergency physicians, pediatric intensivists, nurses, physician assistants and health care research personnel who seek to continuously improve the quality, safety, effectiveness and cost of pediatric sedation/anesthesia practice. The group maintains a prospective registry of patients receiving sedation/anesthesia at various locations within participating institutions. The IRBs (or equivalents) of all participating centers approved this study. A list of participating locations (along with numbers of records submitted) is presented in Appendix.
The consortium first met as a study group in May 2003 in Chicago, IL, where the group decided on a mission statement and priorities for data collection. Decisions were based on guidelines from the American Academy of Pediatrics,9 American Society of Anesthesiologists (ASA),10 and American College of Emergency Physicians11 regarding sedation/anesthesia of pediatric patients, a review of the literature and the consensus of the consortium members. A web-based data collection tool was developed to collect the desired data. The general categories of data collected on each patient are outlined in Table 1. For a more detailed description of the logic and questions used in this data tool, please see “Web Tool Content” on the consortium website at www.pediatricsedationrc.org. Subsequent to this meeting, a “User Manual” for the web tool was written to delineate terms and data required for each field (www.pediatricsedationrc.org).
The web-based tool consisted of 25 primary “screens.” The data collection system was designed to ask one question per screen, dynamically generating an interface for each subsequent question based on the responses from the previous question. The development of standard answer sets allowed clear coding and interpretation of responses, as well as rapid movement through the survey. The system included computer code designed to validate data at the time of data entry (preventing logical errors) and branching logic which ensured that only relevant questions are asked, thus minimizing the total number of questions asked in each survey.
The study data are collected using the Sybase database management system, with interfaces dynamically generated using Java and html coding. The study website and data entry portal were secured using Secure Socket Layer, and each study participant is authenticated through the website and authorized to access only the parts of the website relevant to his or her institution. All data collected in this study met Health Insurance Portability and Accountability Act requirements for de-identification. No patient-identifiable data were transmitted during this study.
After 3 mo (5000 records) of pilot testing, the content of the data collection tool was refined at the next meeting of all consortium members in May 2004. The study was a prospective observational evaluation of sedation/anesthesia practice. No alteration of sedation/anesthesia practice was made at any participating institution for the purposes of this study.
The data tool included three separate screens or question sets that related to complications during anesthesia. One screen collected data on “complications during the procedure.” The options available in this category are defined in Table 2. As with all screens in the tool, questions included logic that prompted further questions to clearly define the nature of the “complication” selected. For example, if “desaturation” was selected, further screens would be activated to define the level and duration of desaturation involved in a given incident. The selections in this category are listed in Table 3.
In all of the categories of data collection, participants were allowed to write in any additional complications that occurred during the course of the procedure that are not accounted for by our standard list. All of these free text entries were then evaluated individually. In cases where it was clear that the written complication fit into an existing category, the appropriate selection box was clicked. In cases where the written complication was distinct from any standard choice on the collection tool, the choice was recorded and grouped with similar complications that were submitted for other patients.
All participating institutions (and primary investigators) were blinded to the data submitted from any individual institution other than their own. Study authors were also blinded to referring institution. All primary investigators were charged with selecting locations within their institutions where collection of sedation/anesthesia data was feasible and assuring nonselective data submission through regular inventories of their data submissions. This project does not intend to collect all sedation/anesthesia interactions from each institution, but rather those from locations where data collection can be assured. All primary investigators were required to perform data audits on 10 charts every 6 mo and report accuracy of the data transmitted. In addition, these investigators were required to review total counts of propofol sedation/anesthesia performed in their institution (independently recorded) versus that of the number of records submitted to the PSRC. Any discrepancies in numbers provided versus propofol sedation/anesthesia performed at the institution required a complete review of the data-gathering methodology at the institution.
The data for this study were collected by the PSRC between July 2004 and September 2007. Our primary outcome analysis was the rate of adverse events and unplanned airway interventions during sedation/anesthesia activity using propofol.
During the study period, of the 88,672 records approved for evaluation, 445 records were eliminated from consideration because of a lapse in IRB approval at one institution and inadequate data audits at three institutions. Of the eligible records, 49,836 were submitted using propofol as the sole or primary sedative for the procedure. The data presented in Table 4 represent a cumulative total of the descriptive information and complication data selected for the population studied. The data were calculated by summing the total number of selection boxes clicked in answer to each question.
The age, sex distribution, and ASA physical status score demographics of the study population are also presented in Table 4. Because nil per os (NPO) status is thought to bear on the likelihood of aspiration or pulmonary complications, we present the data on NPO status for the study population in Table 5. Our data tool did not distinguish clear fluids from other fluids. In addition, we were not able to distinguish between “fatty” solids (hamburger) and a non-fatty meal, such as toast.
The distribution of providers responsible for oversight of the sedation/anesthesia (including pediatric subspecialty training) is presented in Table 6. It should be noted that some of these providers would have been delivering the medication and monitoring the patient (alone), while others would have simply been responsible for the conduct of the anesthesia, and may not have been supervising other providers. Since the physician providing oversight of sedation/anesthesia is ultimately responsible for the sedation/anesthesia, we chose to focus on this portion of the demographic data for this report. The data reflect a preponderance of anesthesiologists, emergency physicians and intensivists involved in our study cohort. All of the institutions involved in this study have pediatric sedation services, meaning they have a specific service that treats children and focuses their professional practice in this area. Although we do not have a comparison group for these providers, we believe this is an important characteristic of the provider systems and may have as much or more impact on their performance than the exact training of each individual sedation provider.
The distribution of procedure categories performed is outlined in Table 7. In lieu of listing the hundreds of procedures performed, we have listed categories. Although we do not specifically present painful versus non-painful procedures, one can reasonably assume that the fracture reductions were painful, and that 98% of the radiological procedures consisted of MRI and computed tomography scans that are not painful.
The “Primary Diagnosis” groups (the diagnosis for which the test/procedure was being performed) for the patients in the database are listed in Table 8. The distribution of drugs used in addition to propofol is presented in Table 9. Because this dataset was so large, we have chosen to present the frequency and type of additional medications used for the patients who received propofol, not the exact dosing or intervals for dosing of these additional medications.
To offer a clearer idea of the rate of adverse events that would be of most interest to sedation providers, we subdivided the complication data into general categories: all inclusive versus pulmonary complications (apnea, aspiration, desaturation, emergency anesthesia consult, unexpected bag-mask ventilation and unexpected intubation). A summary of the total number of adverse events in general categories is included in Table 10.
Specific adverse events reported to the database are listed in Table 11. Airway interventions were coded in our database and were clearly differentiated between planned and unplanned (Table 12). For instance, if the provider intended to deliver a dose of propofol that would be expected to require airway management (such as a jaw lift), this was not collected as a complication. Only airway interventions that would not have been expected (according to the sedation/anesthesia provider) to be part of the normal management of a patient receiving sedation/anesthesia (with the dose and timing used) were recorded in the adverse event category. Several categories of complications were not initially included in the data collection tool. These included “allergic reaction,” “coughing that interfered with the anesthesia” and “IV related problems.” After reviewing the first 10,000 records, these complications were submitted in free text format more often than one per thousand propofol sedation/anesthesia and were therefore thought to warrant their own categories on the data collection tool.
We evaluated the complications reported based on a number of factors. Table 13 lists the reports of all possible complications versus provider type, ASA status, age, NPO status, and the addition of opiates to propofol sedation/anesthesia. Table 14 considers these same factors versus reported pulmonary complications.
We collected data on whether the procedure for which the sedation/anesthesia being delivered was completed. Furthermore, we wanted to know if the sedation/anesthesia was not completed due to problems with the sedation/anesthesia (itself) or due to technical issues not related to the procedural anesthesia (equipment breakdown etc.). These data are presented in Table 15.
The consortium recorded two cases where cardiopulmonary resuscitation (CPR) was required. The first was a 9-year-old male who was to undergo a bronchoscopy for chronic cough under propofol sedation/anesthesia in a pediatric intensive care unit. The child had a history of tracheo-esophageal fistula that had been repaired at the time of birth. The description of the case states that the patient experienced laryngospasm and profound hypoxia leading to bradycardia requiring CPR, and that he rapidly responded to an epinephrine bolus. Two hours after the episode, he was back to his baseline state. He was discharged home after overnight observation. The second case involved an athletic appearing, previously healthy, 16-year-old male who had a history of gastrointestinal bleeding and who was undergoing a colonoscopy. This patient received 195 mg of propofol over 13 min, at which point he became apneic and his heart rate and arterial blood pressure decreased acutely, resulting in asystole for approximately 30 s. He responded to CPR in addition to epinephrine and atropine. He was awake and speaking clearly 30 min after the event. Subsequent cardiology evaluation revealed no underlying pathology, but he was thought to be intravascularly depleted at the time of anesthesia. He was admitted for overnight observation and discharged the following day.
Four aspiration episodes were also reported. In each case, emesis was observed during the procedural sedation/anesthesia and a marked change in respiratory status were noted in the immediate peri-procedural timeframe, including new cough (primarily) and new persistent oxygen requirement. All patients had significant oxygen desaturation which resolved with positive pressure ventilation and high percentage oxygen therapy. All recovered without significant sequellae and were discharged either on the day of the procedure or the following day. Data on radiographs and other diagnostic interventions were not available. Summary data from these cases are presented in Table 16.
The administration of propofol to children undergoing sedation/anesthesia for diagnostic and therapeutic procedures remains controversial. For example, the ASA recommends that only professionals trained in the delivery of general anesthesia should deliver deep sedation/anesthesia.12 Given the potency of propofol and the nature of pediatric patients, essentially all children administered propofol would clearly be categorized as being deeply sedated or anesthetized, as documented by Reeves et al.13 Despite the ASA guideline, propofol sedation/anesthesia is delivered to children for procedures in emergency departments, intensive care units, and sedation/anesthesia units all over the United States (and around the world) by pediatric generalists and subspecialists every day. Furthermore, other professional organizations have written guidelines and recommendations for propofol use by their constituents.14 Unfortunately, all of the positions taken by professional societies are based on collective opinion and analysis of relatively small, observational, single-center studies, because there simply have not been any large multicenter reports on the safety of propofol anesthesia or on the complications that may occur during propofol anesthesia. In this study, we present the largest experience with propofol sedation/anesthesia for children outside of the operating room that has been published to date. Our results are not presented primarily as a comparison between various providers or care settings, but rather as a general overview of the risks involved in this practice, as experienced in our study consortium.
This epidemiological investigation was conceived to fill the void of direct data relating to the incidence and nature of adverse events in pediatric propofol sedation/anesthesia outside the operating room. Voluntary enrollment of 37 PSRC locations (two from one institution) has allowed approximately 50,000 sedation/anesthesia encounters using propofol to be captured in a database. These data clearly show that serious adverse events are quite rare in the practice of pediatric sedation/anesthesia for procedures within our consortium; no deaths occurred and two cardiac arrests were reported (both responded almost immediately to treatment and suffered no long-term injury). The observed (low) incidence of mortality is not unexpected and is consistent with the low incidence of mortality currently associated with the provision of general anesthesia.15 Four aspiration episodes were reported, yielding a rate consistent with previously reported incidences for propofol sedation/anesthesia and sedation/anesthesia practice. However, more minor, but potentially serious, adverse events clearly are not as rare. Approximately 1 in 65 propofol sedation/anesthesias was associated with stridor, laryngospasm, airway obstruction, wheezing or central apnea, any of which could progress to poor outcomes if not managed well. Indeed, 1 in 70 propofol sedation/anesthesia administrations required airway and ventilation interventions, including oral/nasal airway placement, to bag-mask ventilation, to emergency tracheal intubation. The ability of the sedation/anesthesia systems involved in this study to deliver these types of interventions was critical in preventing more serious adverse events.
A careful analysis of our results does not simply reassure providers that propofol sedation/anesthesia of children is safe, but it helps define the competencies required to deliver this care. Specifically, the incidence of apnea and airway obstruction found in this study adds weight to the argument that credentialed providers of deep sedation/anesthesia must also demonstrate proficiency in airway obstruction and respiratory depression management, or have immediate and completely reliable access to such assistance. We believe our findings call for some form of training and testing that is more realistic than that provided by a Pediatric Advanced Life Support class that is given, and tested, on mannequins. Furthermore, we would argue that our data strongly indicate that a monitor of ventilation (such as end-tidal CO2) should be required when providing sedation/anesthesia with propofol, given the significant likelihood of airway obstruction or central apnea. Similarly, the fact that airway secretions appeared as a management issue in many of our patients indicates the absolute need to have suction equipment present for all cases of deep sedation/anesthesia in children.
As part of this study we evaluated data on events we categorized as adverse events with respect to several factors. This type of analysis is quite difficult, since it requires that multiple factors be controlled when considering any individual variable. For instance, when considering the incidence of adverse events for any provider category, one must consider the nature of the patients and procedures that affect the outcomes for each provider. With this in mind, the data reflect the fact that when all possible adverse events are considered, anesthesiologists report fewer issues than other providers. However, when pulmonary complications are considered, no difference in frequency is observed. As expected, higher ASA status, very young patients and the use of adjunctive opiates are all factors associated with a higher rate of general adverse events and pulmonary complications. Intake of solids <8 h before the sedation/anesthesia is associated with higher risk for complications, whereas liquid intake <2 h before the sedation/anesthesia does not appear to be similarly problematic. All of these data must be considered in view of the fact that the overall rate of what we would consider very serious adverse events was extremely low. In addition we must recognize that events considered complications or adverse events are always a matter of some subjective opinion and may or may not represent significant risk to patients.
The two cardiac arrest events reported in this paper are actually the cause of some concern and further reinforce the concept that the data presented here do not simply endorse the care provided. The first involved a patient who had a history of complex tracheal surgery and was undergoing bronchoscopy with sedation/anesthesia provided in an intensive care environment. Although it is impossible to know if the course of this case would have been different if managed by anesthesiologists who may administer propofol in such situations. Careful consideration must be given to complex patients undergoing invasive procedures where deep sedation/anesthesia is required. These patients may well be best served by anesthesiology-trained professionals. The second case was notable for an, otherwise well, adolescent receiving propofol after a gastrointestinal bleed. Although the details of the initial (bleed) resuscitation are not available to us, we would suggest that this case calls into question the knowledge of the sedation providers involved vis a vis the impact of propofol on the cardiovascular stability of a patient who might have been under-resuscitated. We would once again suggest that emergency sedation on potentially unstable patients may require a level of insight and preparation that comes with specific anesthesiology-based training.
The data on our cases of aspiration are particularly interesting. Although we lack definitive radiographic proof that each of these cases represented an aspiration injury, the documentation of emesis combined with new respiratory symptoms is highly suggestive of this event. Notably, none of these episodes led to persistent symptoms or subsequent clinical problems. NPO status was also of interest for the aspiration cases, all of which were at least 2 h NPO for liquids and least 6 h NPO for solids. Although these data do not discount the importance of NPO status, the fact that these cases occurred at all raises the possibility that other factors (such as coexisting pathology) may highly influence the risk of aspiration in the setting of pediatric anesthesia. We also note that most of the cases reported to our database had appropriate fasting intervals and there is no way to predict from our data what the incidence of aspiration might have been with/without fasting.
Prolonged anesthesia/recovery (meaning recovery time at least three times the average for the anesthesia for any reason) and “failed anesthesia” rates (procedures that could not be completed because of problems with the sedation/anesthesia or the procedure) were also delineated by these data, occurring 30 and 81 times, respectively, per 10,000 propofol sedation/anesthesia encounters. The fact that <1% of propofol sedation/anesthesia produced conditions that would not allow the procedure to be completed is an astonishing record of effectiveness. However, we must recognize that the cohort of participants collecting data in this consortium represents specialized pediatric sedation/anesthesia care. The previously reported incidence of failed sedations using a variety of anesthetic drugs ranges from 0.2% to 15%.16,17 These studies did not have dedicated sedation/anesthesia programs and consisted of retrospective chart reviews by procedure type, such as MRI.
We have outlined the limitations of our database in a previous publication. We recognize that the institutions involved in this study selected themselves for inclusion and are voluntarily reporting their outcomes. Thus, it is very likely that we are looking at a highly motivated and organized set of sedation/anesthesia systems that would outperform other less controlled systems and that, in fact, may represent best practice. This bias is similar to that present in almost all studies of sedation/anesthesia practice from single centers that have chosen to study their own outcomes using propofol. Indeed, the observed rates of complications and unplanned treatments are consistent with practice patterns of a highly competent cohort and they may not be applicable to all clinical settings in which sedation/anesthesia care is provided. We must also recognize that it is always difficult to assure that all adverse outcomes which occurred during the study period were reported, especially when dealing with a study group as large and geographically varied as this one. We have tried to address this issue with blinded data submission (no disincentive to report based on embarrassment), along with regular data audits by each institutional principal investigator. Once again, we would suggest that lack of reporting of adverse sedation/anesthesia outcomes is a problem that could occur in any study looking at this practice. We believe that the fact that these data were being used for institutional quality assurance purposes along with the audit safeguards makes selective data reporting unlikely.
Delineating adverse events surrounding airway management proved particularly difficult and could be interpreted as a weakness in our study. Does one count every single airway intervention during a deep anesthesia administration or simply those that were considered unanticipated, and therefore much more of a safety hazard? We chose to separate planned (jaw thrust required after a bolus of propofol given for a bone marrow biopsy) from unplanned (bag-mask ventilation in the middle of a MRI scan) airway interventions to avoid counting events as adverse when they were expected and well managed. We fully understand that there is some subjectivity built into this type of reporting and that our event reporting reflects this bias.
This study is prospective and observational in nature. As there was no control group, we did not attempt to draw direct cause and effect from these data as they pertain to adverse outcomes. Rather, we concentrated on reporting the complications that rise to the level of concern, with the caveat that this information is a marked improvement over previous investigations that were limited in patient numbers, geography and scope. We recognize that the methodologies used to perform sedation for the multitude of procedures captured in this database are extremely varied. They range from moderate sedation to deep sedation to general anesthesia. We wanted to consider all of these levels of care, since it is impossible to completely differentiate these levels and (as a whole) they represent the range of care that is provided using propofol to accomplish procedural sedation/anesthesia care. We also recognize that a percentage of these cases involved other sedatives or analgesics added to propofol primarily as a preanesthesia sedative, with propofol as the primary drug. Since this is the nature of the pediatric sedation/anesthesia practice using propofol, we feel it is necessary to include all of these methods in our analysis to obtain an accurate picture of the aggregate risk to children in this practice. The data from this study will be used to focus research efforts to improve the safety and reliability of pediatric propofol anesthesia. The potential to link markers of risk and adverse outcomes represents a significant research potential.
Data collection in the PSRC is continuing, and future analysis will focus on evaluating the association of adverse outcomes with various provider types, monitoring standards, and medications used. Additional efforts will be aimed at evaluating effectiveness and efficiency of various sedation/anesthesia systems with the possibility of defining and promoting the system characteristics that lead to the best outcomes.
In summary, we conclude that, in the hospital setting of those institutions participating in the PSRC, the reported incidence of serious adverse events in pediatric propofol sedation/anesthesia is low. However, the reported incidence of events that have potential to harm and that require timely rescue interventions is significant, occurring once per 89 propofol administrations. The reported incidence of these adverse events will direct targeted research, and support the continued efforts of those who seek to encourage improved safety and reliability in the provision of pediatric sedation/anesthesia.
1. Sacchetti A, Stander E, Ferguson N, Maniar G, Valko P. Pediatric Procedural Sedation in the Community Emergency Department: results from the ProSCED registry. Pediatr Emerg Care 2007;23:218–22
2.Pershad J, Wan J, Anghelescu DL. Comparison of propofol with pentobarbital/midazolam/fentanyl sedation for magnetic resonance imaging of the brain in children. Pediatrics 2007;120:e629–36
3.Wheeler DS, Vaux KK, Ponaman ML, Poss BW. The safe and effective use of propofol sedation in children undergoing diagnostic and therapeutic procedures: experience in a pediatric ICU and a review of the literature. Pediatr Emerg Care 2003;19:385–92
4.Cote CJ, Notterman DA, Karl HW, Weinberg JA, McCloskey C. Adverse sedation events in pediatrics: a critical incident analysis of contributing factors. Pediatrics 2000;105:805–14
5.Cote CJ, Karl HW, Notterman DA, Weinberg JA, McCloskey C. Adverse sedation events in pediatrics: analysis of medications used for sedation. Pediatrics 2000;106:633–44
6.Vardi A, Salem Y, Padeh S, Paret G, Barzilay Z. Is propofol safe for procedural sedation in children? A prospective evaluation of propofol versus ketamine in pediatric critical care. Crit Care Med 2002;30:1231–6
7.Joint Commission on Accreditation of Healthcare Organizations. Sedation and anesthesia care standards. Oakbrook Terrace, IL: Joint Commission on Accreditation of Healthcare Organizations, 2003
8.Cravero JP, Blike GT, Beach M, Gallagher SM, Hertzog JH, Havidich JE, Gelman B; Pediatric Sedation Research Consortium. The incidence and nature of adverse events during pediatric sedation/anesthesia for procedures outside the operating room—report from the Pediatric Sedation Research Consortium. Pediatrics 2006;118:1087–96
9.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;118:2587–602
10.American Society of Anesthesiologists Task Force on Sedation and Analgesia by Non-Anesthesiologists. Practice guidelines for sedation and analgesia by non-anesthesiologists. Anesthesiology 2002;96:1004–17
11.American College of Emergency Physicians. Clinical policy for procedural sedation and analgesia in the emergency department. Ann Emerg Med 1998;31:663–77
12.Statement on granting privileges to non-anesthesiologist practitioners for personally administering deep sedation or supervising deep sedation by individuals who are not anesthesia professionals, 2006. Available at: http://www.asahq.org/publicationsAndServices/standards/39.pdf
13.Reeves ST, Havidich JE, Tobin DP. Conscious sedation of children with propofol is anything but conscious. Pediatrics 2004;114:e74–6
14.Miner JR, Burton JH. Clinical practice advisory: emergency department procedural sedation with propofol. Ann Emerg Med 2007;50:182–7
15.VanderWalt J. Searching for the Holy Grail: measuring risk in paediatric anaesthesia. Paediatr Anaesth 2001;11:637–41
16.Malviya S. Sedation and general anaesthesia in children undergoing MRI and CT: adverse events and outcomes. Br J Anaesth 2000;84:743–8
17.Greenberg SB, Faerber EN, Aspinall CL, Adams RC. High dose chloral hydrate sedation for children undergoing MR imaging: safety and efficacy in relation to age. AJR Am J Roentgenol 1993;161:639–41