clinical policy_ procedural sedation and analgesia in the emergency department[1]

PAIN MANAGEMENT/CLINICAL POLICY

Clinical Policy: Procedural Sedation and Analgesia in theEmergency Department

From the American College of Emergency Physicians Clinical Policies Subcommittee (Writing Committee) on

Procedural Sedation and Analgesia:

Steven A. Godwin, MD, Chair

David A. Caro, MD

Stephen J. Wolf, MD

Andy S. Jagoda, MD

Ronald Charles, MD

Benjamin E. Marett, RN, MSN, CEN, CNA, COHN-S (ENA

Representative 2002-2003)

Jessie Moore, RN, MSN, CEN (ENA Representative 2001-

2002)

Members of the American College of Emergency Physicians Clinical Policies Committee (Oversight Committee) included:

William C. Dalsey, MD (Chair 2000-2002, Co-Chair 2002-

2003)

Andy S. Jagoda, MD (Co-Chair 2002-2003, Chair 2003-

2005)

Wyatt W. Decker, MD

Jonathan A. Edlow, MD

Francis M. Fesmire, MD

Steven A. Godwin, MD

Sigrid A. Hahn, MD (EMRA Representative 2003-2004)

John M. Howell, MD

Shkelzen Hoxhaj, MD (EMRA Representative 2002-2003)

J. Stephen Huff, MD

Edwin K. Kuffner, MD

JoAnn Lazarus, RN, MSN, CEN (ENA Representative 2003-

2004)

Thomas W. Lukens, MD, PhD

Benjamin E. Marett, RN, MSN, CEN, CNA, COHN-S (ENA

Representative 2002-2003)

Michael Moon, RN, CNS, MSN, CEN (ENA Representative

2003-2004)

Jessie Moore, RN, MSN, CEN (ENA Representative 2001-

2002)

Devorah Nazarian, MD

Scott M. Silvers, MD

Edward P. Sloan, MD, MPH

Robert L. Wears, MD, MS, Methodologist

Stephen J. Wolf, MD

Susan M. Nedza, MD, MBA (Board Liaison 2001-2003)

John Skiendzielewski, MD (Board Liaison 2003-2004)

Rhonda Whitson, RHIA, Staff Liaison, Clinical Policies

Committee and Subcommittees

Approved by the ACEP Board of Directors October 20,

2004.

Policy statements and clinical policies are the official policies of the American College of EmergencyPhysicians and, as such, are not subject to the same peer review process as articles appearing in theprint journal. Policy statements and clinical policies of ACEP do not necessarily reflect the policiesand beliefs of Annals of Emergency Medicine and its editors.

0196-0644/$-see front matterCopyright ª 2005 by the American College of Emergency Physicians.doi:10.1016/j.annemergmed.2004.11.002

[Ann Emerg Med. 2005;45:177-196.]

PREFACEEmergency medicine training provides physicians with

expertise in critical care, airway and pain management.

Accordingly, expertise in procedural sedation and analgesia isa core competency in emergency medicine practice.1,2 Emer-gency physicians routinely provide sedation, analgesia, re-spiratory, and hemodynamic management of critically illpatients.

Volume 45, no. 2 : February 2005 Annals of Emergency Medicine 177

REV 5.0 DTD � YMEM1440_proof � 7 January 2005 � 6:06 pm

Procedural sedation has received a great amount of attentionin recent years. Several groups have produced documentscovering its use, including the Joint Commission on Accred-itation of Healthcare Organizations (JCAHO), which has madeit an area of intense review. Unfortunately, most of thesepromulgated advisory materials are not truly evidence-based.

The following clinical policy, developed by the ClinicalPolicies Committee of the American College of EmergencyPhysicians (ACEP), updates the original clinical policyapproved on January 13, 1998.3 This clinical policy attempts toremove the bias from recommendations for procedural sedationby creating a document that is, to the degree possible, evidence-based. There remains a relative lack of high-quality data in someareas of procedural sedation. It must be carefully noted,however, that despite the statements made in this policy,individual institutions will still be accredited on the basis of thecriteria of the respective accrediting organization, such as theJCAHO.

A MEDLINE search of original research on sedation andanalgesia from the past decade revealed that emergencyphysicians are doing the majority of research in this area. Asubstantial body of evidence supports the routine safe use ofprocedural sedation and analgesia by emergency physicians.4-21

INTRODUCTIONThe emergency department (ED) is a unique environment

where a variety of patients with emergent and urgent conditionsare managed. Many of these conditions result in significant painand are associated with varying degrees of anxiety. Accordingly,the management of sedation and analgesia is an importantcomponent of comprehensive emergency medical care forpatients of all ages and, therefore, a primary concern for theemergency physician. Pain control often is not adequatelyprovided for a variety of reasons, including fear of oversedation,concern of altering physical findings, or underestimation ofpatient needs.22,23 However, proactively addressing pain andanxiety may improve quality of care and patient satisfaction byfacilitating interventional procedures and minimizing patientsuffering. Many of the drugs used for sedation and analgesia havethe potential to cause central nervous system, respiratory, orcardiac depression. According to the JCAHO 2004Comprehensive Accreditation Manual for Hospitals, the standardsfor sedation and anesthesia care apply when patients in anysetting receive, for any purpose and by any route, moderate ordeep sedation, as well as general, spinal, or other major regionalsedation and anesthesia.24 To minimize complications, theappropriate drug(s) and dosages must be chosen, monitored, andadministered in the proper setting, and a patient evaluationshould be performed before, during, and after their use. Manydrug and dosing regimens used in procedural sedation andanalgesia have been discussed in-depth in other publications.25-28

It is good medical practice to discuss with patients allmedications and interventions that will be provided. Thediscussion should include the risks, benefits, potential sideeffects, and alternatives. There is no literature to support that

the use of an informed consent form separate from the generalinformed consent obtained at registration in the ED has aneffect on patient satisfaction or on clinical outcome. In somecases, procedural sedation and analgesia is provided in situationswhen the patient is in severe pain or extremely anxious becauseof the circumstances surrounding the ED visit; such situationslimit the patient’s ability to comprehend issues presented in theinformed consent process. In other circumstances, the informedconsent process is limited by an altered mental status, whichaffects the patient’s capacity to understand risk and benefit.Procedural sedation and analgesia under implied consent maybe appropriate in these circumstances.

Recommendations offered in this policy are not intended torepresent the only diagnostic and management options that theemergency physician should consider. ACEP clearly recognizesthe importance of the individual physician’s judgment. Rather,this guideline defines for the physician those strategies for whichmedical literature exists to provide support for answers to thecritical questions addressed in this policy.

DEFINITIONSProcedural sedation refers to a technique of administering

sedatives or dissociative agents with or without analgesics toinduce a state that allows the patient to tolerate unpleasantprocedures while maintaining cardiorespiratory function.3

Procedural sedation and analgesia is intended to result ina depressed level of consciousness that allows the patient tomaintain oxygenation and airway control independently.Moderate sedation, previously referred to as ‘‘conscioussedation,’’ is defined as a drug-induced depression of con-sciousness during which patients respond purposefully.24

Specifically, the drugs, doses, and techniques used are not likelyto produce a loss of protective airway reflexes. As part of thecontinuum of sedation, deep sedation is defined as a drug-induced depression of consciousness during which patientscannot be easily aroused but respond purposefully after repeatedor painful stimulation. These patients may require assistance inmaintaining airway patency and ventilatory effort.24 Generalanesthesia is defined as a drug-induced loss of consciousnessduring which patients are not arousable and may have animpaired cardiorespiratory function requiring varying degrees ofsupport.24 The patient under general anesthesia is profoundlycompromised and does not exhibit movement or autonomicnervous system responses to a standard surgical stimulus. Onthe basis of recognition that proper administration of sedativemedications is a continuum and it is often difficult to predicthow an individual will respond to a specific sedative agent,practitioners should possess the skills required to rescuea patient 1 level greater than the intended level of sedation.Therefore, should deep sedation be required to performa procedure, the practitioner is expected to be competent inskills involving cardiovascular support and airway managementas in general anesthesia.26 Due to emphasis in the emergencymedicine training curriculum, these qualities are now consid-ered core skills for all board-certified emergency physicians.1,2

Clinical Policy

178 Annals of Emergency Medicine Volume 45, no. 2 : February 2005


A separate category of sedation has been suggested in aneffort to better classify and describe the dissociative sedativeagents. Dissociative sedation is described as a ‘‘trancelikecataleptic state characterized by profound analgesia andamnesia, with retention of protective airway reflexes, sponta-neous respirations, and cardiopulmonary stability.’’ The termsmoderate, deep, and general anesthesia sedation refer to formsof sedation that do not apply to dissociative sedation.29

METHODOLOGYThis clinical policy was created after careful review and

critical analysis of the peer-reviewed literature. A MEDLINEsearch of English-language articles published between January1992 and January 2004 was performed using combinations ofthe key words ‘‘conscious sedation,’’ ‘‘moderate sedation,’’ ‘‘deepsedation,’’ ‘‘analgesia,’’ ‘‘sedation,’’ ‘‘standards,’’ ‘‘guidelines,’’‘‘complications,’’ and ‘‘emergency department.’’ Terms werethen exploded as appropriate. Abstracts and articles werereviewed by subcommittee members, and pertinent articles wereselected. These articles were evaluated, and those addressing thequestions considered in this document were chosen for grading.Subcommittee members also supplied references from bibliog-raphies of initially selected articles or from their own files.

The reasons for developing clinical policies in emergencymedicine and the approaches used in their development havebeen enumerated.30 This policy is a product of the ACEPclinical policy development process and is based on the existingliterature; where literature was not available, consensus ofemergency physicians was used. Expert review comments werereceived from emergency physicians, individual members ofACEP’s Pediatric Emergency Medicine Committee and Sec-tion, and individual members of the American Society ofAnesthesiologists. Their responses were used to further refineand enhance this policy. Clinical policies are scheduled forrevision every 3 years; however, interim reviews are conductedwhen technology or the practice environment changes signif-icantly.

All publications were graded by at least 2 of the sub-committee members into 1 of 3 categories of strength ofevidence. Some articles were downgraded on the basis ofa standardized formula that considers the size of studypopulation, methodology, validity of conclusions, and potentialsources of bias (Appendix A).

During the review process, all articles were given a baseline‘‘strength of evidence’’ by the subcommittee members accordingto the following criteria:

Strength of evidence Class IdInterventional studiesincluding clinical trials, observational studies includingprospective cohort studies, aggregate studies includingmeta-analyses of randomized clinical trials only.

Strength of evidence Class IIdObservational studiesincluding retrospective cohort studies, case-controlled studies,aggregate studies including other meta-analyses.

Strength of evidence Class IIIdDescriptive cross-sectionalstudies, observational reports including case series and case

reports, consensus studies including published panel consensusby acknowledged groups of experts.

Strength of evidence Class I and II articles were then rated onelements subcommittee members believed were most importantin creating a quality work. Class I and II articles with significantflaws or design bias were downgraded on the basis of a setformula (Appendix B). Strength of evidence Class III articleswere downgraded if they demonstrated significant flaws or bias.Articles downgraded below strength of evidence Class III weregiven an ‘‘X’’ rating and were not used in formulatingrecommendations in this policy. An Evidentiary Table wasconstructed and is included in this policy.

Recommendations regarding patient management were thenmade according to the following criteria:

Level A recommendations. Generally accepted principles forpatient management that reflect a high degree of clinicalcertainty (ie, based on strength of evidence Class I oroverwhelming evidence from strength of evidence Class IIstudies that directly address all the issues).

Level B recommendations. Recommendations for patientmanagement that may identify a particular strategy or range ofmanagement strategies that reflect moderate clinical certainty (ie,based on strength of evidence Class II studies that directly addressthe issue, decision analysis that directly addresses the issue, orstrong consensus of strength of evidence Class III studies).

Level C recommendations. Other strategies for patientmanagement based on preliminary, inconclusive, or conflictingevidence, or in the absence of any published literature, based onpanel consensus.

There are certain circumstances in which the recommenda-tions stemming from a body of evidence should not be rated ashighly as the individual studies on which they are based. Factorssuch as heterogeneity of results, uncertainty about effectmagnitude and consequences, strength of prior beliefs, andpublication bias, among others, might lead to such a down-grading of recommendations.

Scope of Application. This guideline is intended forphysicians working in hospital-based EDs.

Inclusion Criteria. This clinical policy is intended for: (1)ED patients of all ages who have emergent or urgent conditionsthat require pain and/or anxiety management to successfullyaccomplish an interventional or diagnostic procedure. Aseparate multidisciplinary clinical policy has been developedhighlighting medications and particular considerations com-monly encountered in the practice of sedation and analgesia inchildren.31

(2) High-risk patients (eg, those with underlying cardiopul-monary disorders, multiple trauma, head trauma, or who haveingested a central nervous system depressant such as alcohol) areincluded with the understanding that these patients are atincreased risk of complications from procedural sedation andanalgesia.4,5,10,26

Exclusion Criteria. Excluded from this guideline are: (1)patients receiving inhalational anesthetics, (2) patients whoreceive analgesia for pain control without sedatives, (3) patients

Clinical Policy



who receive sedation solely for the purpose of managingbehavioral emergencies, and (4) patients who are intubated.

CRITICAL QUESTIONSI. What are the personnel requirements needed to provideprocedural sedation and analgesia in the ED?

Personnel providing procedural sedation and analgesia musthave an understanding of the drugs administered, the ability tomonitor the patient’s response to the medications given, and theskills necessary to intervene in managing all potentialcomplications.

JCAHO anesthesia standards reinforce that sedation-to-anesthesia is a continuum, and it is not always possible topredict how individual patients receiving medications willrespond. It is therefore important for individual institutions toensure that all individuals performing moderate or deepsedation are trained to: (1) administer pharmacologic agents topredictably achieve desired levels of sedation, (2) monitorpatients and maintain them at desired levels of sedation, and (3)manage complications observed during procedural sedation andanalgesia.24 Procedural sedation and analgesia at both moderateand deep levels has been demonstrated to be both safe andeffective when properly administered by experienced emergencyphysicians.4-6,8-10,12,15,18,19

The literature does not provide clear evidence on the numberof personnel necessary to safely provide procedural sedation andanalgesia. The presence of a support person assumes increasedimportance when the physician is involved in a procedure thatprecludes the ability to continually assess the patient’s clinicalstatus. However, there are situations where exceptions arepermissible, such as when low doses of pharmacologic agents areused and the physician is able to maintain visual or verbalcommunication with the patient. During procedural sedationand analgesia, there must be an individual available who iscapable of recognizing and managing respiratory and hemody-namic emergencies. As a result, all physicians who are workingor consulting in the ED should coordinate all proceduresrequiring procedural sedation and analgesia with the ED staff.

Personnel Recommendations: What are the personnelrequirements needed to provide procedural sedation andanalgesia in the ED?

Level A recommendations. None specified.Level B recommendations. None specified.Level C recommendations. During moderate and deep

sedation, a qualified support person should be present forcontinuous monitoring of the patient.

Procedural sedation and analgesia in the ED must besupervised by an emergency physician or other appropriatelytrained and credentialed specialist.

II. What are the key components of the patient assessmentbefore initiating procedural sedation?

There is a lack of outcome-based studies to mandate moreextensive evaluation beyond vital signs, mental status, and

airway and cardiopulmonary assessment. An increased risk withprocedural sedation and analgesia may exist in select subsets ofpatients such as those with difficult facial or neck anatomy,patients who are very old, or patients with underlyingcardiopulmonary disease.

There is no literature to support the need for routinediagnostic testing before procedural sedation and analgesia;diagnostic testing is driven by the patient’s status.

Patient Assessment Recommendations: What are the keycomponents of the patient assessment before initiatingprocedural sedation?

Level A recommendations. None specified.Level B recommendations. None specified.Level C recommendations. Obtain a history and perform

a physical examination to identify medical illnesses, medica-tions, allergies, and anatomic features that may affect proceduralsedation and analgesia and airway management.

No routine diagnostic testing is required before proceduralsedation.

III. Is preprocedural fasting necessary before initiatingprocedural sedation?

The combination of vomiting and loss of airway protectivereflexes is an extremely rare occurrence with procedural sedationand analgesia, making aspiration an unlikely event.6,12,32-35

Recent studies and consensus statements have recommendedvarying fasting periods based on the specific substance ingested,but fasting patterns remain variable with no uniformly acceptedpractice standards.26,32,35-41 Much of the aspiration data havebeen extrapolated from the general anesthesia literature wherethe potential of aspiration is increased with manipulation of theairway during intubation and extubation. Thorough reviews ofthis topic demonstrate a lack of evidence that gastric emptyinghas any impact on the incidence of complications or onoutcome in procedural sedation and analgesia. A prospectiveobservational study of 1,014 children identified no differencewith airway complications, emesis, or other adverse eventsbetween patients classified by their preprocedural fasting status.Of the 509 (56%) patients who did not meet preproceduralfasting guidelines for elective procedures as suggested by theAmerican Academy of Pediatrics and the American Society ofAnesthesiologists, no episodes of aspiration were documented.The authors correctly acknowledge that the study is under-powered to detect significant differences in the rate of emesiswith and without aspiration due to the extremely rare incidenceof these combined events.35 Despite the paucity of data and inrecognition of the potential risk for aspiration, a number ofpublications encourage careful consideration of timing anddepth of procedural sedation and analgesia in the absence of anadequate fasting period.32,41,42 In addition, pharmacologicagents including antacids have not been shown to improveoutcomes and are no longer recommended as standardpractice.32,41,43 No study has determined a necessary fastingperiod before initiation of procedural sedation and analgesia.

Clinical Policy



There is insufficient evidence to determine absolute recommen-dations. Although recent food intake is not a contraindicationfor administering procedural sedation and analgesia, the emer-gency physician must weigh the risk of pulmonary aspirationand the benefits of providing procedural sedation and analgesiain accordance with the needs of each individual patient.

Preprocedure Fasting Recommendations: Is preproceduralfasting necessary before initiating procedural sedation?

Level A recommendations. None specified.Level B recommendations. None specified.Level C recommendations. Recent food intake is not

a contraindication for administering procedural sedation andanalgesia, but should be considered in choosing the timing andtarget level of sedation.

IV. What equipment and supplies are required to provideprocedural sedation and analgesia?

Although rare, procedural sedation and analgesia may result inan allergic reaction, respiratory arrest, or cardiopulmonary arrest.The incidence of complications is dependent on the drugs used,rate and dose of administration, and patient sensitivities.Consequently, the appropriate equipment to monitor thepatient’s condition, to manage airways, allergic reactions, anddrug overdoses, and to treat respiratory and cardiopulmonaryarrest should be readily available; use of specific monitoringequipment is discussed in the subsequent sections.

Supportive equipment for procedural sedation and analgesiaincludes oxygen, suction, medications, and advanced lifesupport equipment (eg, a bag-mask ventilation device, andintubation equipment). The need for intravenous access isdependent on the medications, the dose, and the routeused.4,12,15,44,45 Ketamine, which has been shown to exhibita wide safety margin with preservation of protective airwayreflexes and no cardiovascular suppression, is often administeredintramuscularly.45 This is especially beneficial in the pediatricpopulation where additional needle sticks may not be desirable.Procedural sedation and analgesia with opioids or benzodiaze-pines may result in respiratory suppression or transienthypotension despite route of administration; therefore, physi-cians may consider intravenous access necessary in selectpatients.4-6 When opioids or benzodiazepines are used, theopioid antagonist naloxone and the benzodiazepine antagonistflumazenil should be available.6,46,47

Equipment and Supplies Recommendations: What equip-ment and supplies are required to provide proceduralsedation and analgesia?

Level A recommendations. None specified.Level B recommendations. None specified.Level C recommendations. Oxygen, suction, reversal agents,

and advanced life support medications and equipment shouldbe available when procedural sedation and analgesia is used.

Intravenous access should be maintained when intravenousprocedural sedation and analgesia is provided. Intravenous

access may not be necessary when procedural sedation andanalgesia is provided by other routes.

V. What assessment and monitoring are required to provideprocedural sedation in the ED?

Monitoring the patient’s condition involves visual observa-tion and assessment of the level of consciousness andphysiologic changes. The monitoring process should bedocumented. The components of monitoring may include butare not limited to the level of consciousness, respiratory rate,oxygen saturation, exhaled carbon dioxide, heart rate, bloodpressure, and cardiac rhythm monitoring. The patient’s abilityto follow commands is a method of monitoring level ofconsciousness. There are times when patients receive proceduralsedation and analgesia and then must be transported outside theED. In such cases, every attempt must be made to provide thesame level of monitoring during the transport that was usedwithin the ED.

In general, documentation of the patient’s preprocedure statusand clinical status during and after the procedure is recom-mended. The available literature provides little guidance as to theminimum frequency at which vital signs should be recorded.

Vital sign monitoring includes assessment of bloodpressure, pulse rate, respiratory rate, and pulse oximetry. Aprospectively collected database of 1,367 pediatric patientsfound the highest risk of serious adverse events occurredwithin 25 minutes of receiving the last dose of intravenousmedications.48 Adverse events occurred in 184 (13%) of thepatients. Of those, 169 (92%) occurred during the procedureand 14 (8%) occurred in patients after the procedure. Themedian time for serious adverse events occurred approximately2 minutes after administration of the final medication.48 In 1study using midazolam and fentanyl, all cases of apneaoccurred within 5 minutes of drug administration.49 Inanother study, using diazepam and fentanyl, all episodes ofdesaturation occurred within 20 minutes.50 Despite a recom-mendation in a consensus statement,26 there is no evidencethat cardiac monitoring during procedural sedation andanalgesia is of benefit, especially if the patient has nounderlying cardiopulmonary disease.

General Recommendations: What assessment and monitor-ing are required to provide procedural sedation in the ED?

Level A recommendations. None specified.Level B recommendations. None specified.Level C recommendations. Obtain and document vital signs

before, during, and after procedural sedation and analgesia.Monitor the patient’s appearance and ability to respond toverbal stimuli during and after procedural sedation andanalgesia.

VI. How should respiratory status be assessed?Pulse Oximetry. Pulse oximetry provides continuous non-

invasive estimates of arterial oxygen saturation and is a reliabletool in detecting early decreases in oxygen saturation and

Clinical Policy



changes in the patient’s heart rate.51 Under most circumstances,there is excellent correlation between the pulse oximetersaturation, measured by spectrophotometry, and arterialhemoglobin oxygen saturation measured by oximetry. However,when the hemoglobin saturation decreases below 80%, accuracymay be affected.51,52 The limitations of oximetry include itsinability to detect early decreases in the adequacy of ventilationand, thus, the detection of the onset of hypercarbia that mayoccur before the development of apnea. The administration ofoxygen during procedural sedation and analgesia may delay theonset of hypoxemia and, thus, the detection of hypoventilation.

Many studies have demonstrated the utility of pulse oximetryin detecting decreases in oxygen saturation; most of these studiesdefine hypoxemia as a saturation of less than 90%.4,13,45,49

It has been clearly demonstrated that many of the drugs usedin procedural sedation and analgesia predispose the patient tothe development of hypoxemia, and that drug combinations,especially benzodiazepines and opioids, have a potentiatingeffect in suppressing respirations.49,53 Despite the evidence thatdesaturation may occur during procedural sedation andanalgesia, there is little information regarding the clinicalsignificance of transient desaturation. Studies have demon-strated that decreases in oxygen saturation occur without clinicalconsequence, and in fact, transient decreases in oxygensaturation have been reported during sleep in healthyvolunteers. In 1 series, 43% of asymptomatic men haddesaturation to below 90% during sleep, with 13% below75%.54 The results from at least 2 studies suggested that theonly consistent predictor of desaturation during proceduralsedation and analgesia was age greater than 55 years.10,55

It has been hypothesized that failure to properly monitoroxygenation has led to drug-related deaths in proceduralsedation and analgesia.49 There are no studies showing thatdetection of a decrease in oxygen saturation alone duringprocedural sedation and analgesia has an impact on patientoutcome. When the patient’s level of consciousness orrespiratory efforts can be adequately assessed, proceduralsedation and analgesia has an extremely low risk of morbidityand mortality.6,9,13,18,19,45,48,50,56,57

However, the rarity of adverse events results in low statisticalpower for detecting decreases in the adverse event rate. Withoutdevaluing its utility as a monitoring adjunct, pulse oximetryshould not substitute for clinical assessment during proceduralsedation and analgesia, but rather be utilized as a reliableadjunct.

Capnometry. Procedural sedation and analgesia may causehypoventilation49,58 and an increase in end-tidal carbon dioxide(ETCO2).

59 Capnometry is a technique used to monitor ETCO2

and, therefore, may detect early cases of inadequate ventilationbefore oxygen desaturation takes place.60,61 An increase inETCO2 might be the only clue to hypoventilation and potentialrespiratory compromise.62 There is an excellent correlationbetween PACO2 and ETCO2 even when the ETCO2 is measuredthrough a nasal cannula while the patient is receivingoxygen.61

Theoretically, early detection of hypoventilation withcapnometry may be beneficial. However, there is no evidencethat this benefit has an impact on patient outcome when used inprocedural sedation and analgesia. In 1 study of 27 ED patientsreceiving procedural sedation and analgesia with a benzodiaze-pine and/or opioids, the average ETCO2 increased from 36 to 42mm Hg while the oxygen saturation decreased from 98% to94%, but this study had low statistical power to detectdifferences in outcome. These changes were without clinicalconsequence.61 One patient in this study had a desaturationlevel of 83%, but responded to verbal and tactile stimuli. Amore recent prospective study evaluated whether ETCO2 couldbe used to detect respiratory depression and measure depth ofsedation.60 No correlation was found between the ETCO2 andthe level of sedation measured by the Observer’s Assessment ofAlertness and Sedation Scale. A post-hoc analysis of the 74patients revealed that all patients with respiratory depressiondemonstrated an ETCO2 greater than 50 mm Hg, an absentwaveform, or an absolute change from baseline in ETCO2 greaterthan 10 mm Hg. The authors conclude that in the presence ofETCO2 monitoring, these identifiers may allow more rapididentification of hypoventilation than pulse oximetry alone. Inthe study, pulse oximetry would have identified only 11 of the33 patients meeting the predetermined definitions for re-spiratory depression of an oxygen saturation less than 90%,ETCO2 of greater than 50 mm Hg, or an absent waveform. Thisstudy does not address the question as to whether earlierrecognition of hypoventilation has any significant impact onoutcomes. Hypoventilation and subsequent transient hyper-capnea without hypoxemia during procedural sedation andanalgesia has no identified impact on outcome.

Bispectral Index. The Bispectral Index has been validated asan objective measure of sedation depth in the operating room,which has led some researchers to suggest its utility inprocedural sedation. The Bispectral Index reflects the varyinglevels of a patient’s sedation as expressed by mathematicalevaluations involving relevant, descriptive electroencephalo-graphic measures from the frontal cortex.63-65 Although earlyevidence is supportive,21,66 there is insufficient evidence toadvocate its routine use in procedural sedation andanalgesia.

Respiratory Status Recommendations: How shouldrespiratory status be assessed?

Level A recommendations. None specified.Level B recommendations. Pulse oximetry should be used in

patients at increased risk of developing hypoxemia, such aswhen high doses of drugs or multiple drugs are used, or whentreating patients with significant comorbidity.

Level C recommendations. When the patient’s level ofconsciousness is minimally depressed and verbal communica-tion can be continually monitored, pulse oximetry may not benecessary.

Consider capnometry to provide additional informationregarding early identification of hypoventilation.

Clinical Policy



VII. Can ketamine, midazolam, fentanyl, propofol, andetomidate be safely administered for procedural sedationand analgesia in the ED?

Agents such as ketamine result in a dissociative state in whicha patient may not speak or respond purposefully to verbalcommands. Use of ketamine in the doses recommended forprocedural sedation and analgesia does not result in a loss ofprotective reflexes. The medical literature documents the safetyof its use for procedural sedation and analgesia in pediatricpopulations.44,45,67 In a well-designed randomized controlledtrial in 260 children aged 5 to 15 years, Kennedy et al67 foundthat a ketamine and midazolam combination was safer andmore efficacious than a fentanyl and midazolam combinationfor sedation in orthopedic procedures. Hypoxia occurred in 6%of patients receiving ketamine and midazolam versus 20% ofpatients in the fentanyl and midazolam group. Efficacy asdetermined by objective measures of physician and parentalsatisfaction were thought to be superior in the ketamine andmidazolam study arm.67 In a consecutive case series of 1,022children, Green et al45 report that ketamine at doses of 4 to 5mg/kg intramuscularly produced adequate sedation in 98% ofchildren. They reported airway complications in 1.4% ofpatients that included laryngospasm, apnea, and respiratorydepression, all of which were quickly identified and treatedwithout intubation or sequelae. Emesis occurred in 6.7%without evidence of aspiration.

In regard to nondissociative sedation agents such asmidazolam and fentanyl, a key to minimizing complications inprocedural sedation and analgesia is the titration of drugs to thedesired effect. Rapid administration of drugs may be associatedwith hypotension or respiratory depression. In addition, thecombination of drugs may accentuate the potential side effectsassociated with each drug individually. In 1 study, use ofbenzodiazepines alone resulted in no significant respiratorydepression, whereas use of an opioid alone caused hypoxemia in50% of volunteers and caused a decrease in ventilatory responseto carbon dioxide, but did not cause apnea. When thebenzodiazepine and opioid were used together, hypoxemiaoccurred in 92% of subjects, and apnea occurred in 50%.49

Although there was no clinical correlation of these findings topatient outcome, this study does suggest that the combined useof benzodiazepines and opioids increases the risk of respiratorycompromise. Pohlgeers et al50 also reported no association ofbenzodiazepine dose with desaturation. High doses of opioidswithout careful and slow titration also increased the risk ofrespiratory compromise. In a case report, Yaster et al47

presented a child who had a respiratory arrest after 10 mg/kg offentanyl was given over approximately 4 minutes.

It has been suggested that when both a benzodiazepine andan opioid are used, the opioid, which poses the greater risk ofrespiratory depression, should be given first and the benzodi-azepine dose titrated.68

There is a growing body of evidence supporting the safe use ofpropofol for procedural sedation by emergency physicians. Ina prospective observational study performed in the ED, propofol-

induced procedural sedation was reported to have the lowest rateof respiratory depression when compared with methohexital,fentanyl/midazolam, and etomidate.21 There were no significantcomplications. A prospective randomized trial with 103 patientsreceiving propofol or methohexital within the ED compared thedepth of sedation with Bispectral Index scores and rates ofrespiratory depression assessed by ETCO2. Of note, approximatelyone half of the patients in both groups met predefined criteria forrespiratory depression. The study did not detect a difference ineither the level of sedation by Bispectral Index or the level ofsubclinical respiratory depression between the 2 agents. Theauthors found no significant adverse events with either sedative.20

In another prospective study of 43 children comparingmidazolam and propofol, the authors reported successfulsedation but also significant hypoxemia and oversedation;however, no significant complications were reported.15 Inanother pediatric study involving 40 patients, Skokan et al16

reported a significant incidence of oxygen desaturation; however,again, there were no clinical sequelae.

The literature regarding safety and efficacy of etomidate inthe ED is also mounting. A prospective, double-blinded,randomized trial of 46 adult patients undergoing anteriorshoulder dislocation reduction in the ED compared midazolamwith etomidate.17 Burton et al17 found approximately a 90%procedural success rate for both groups. Patients experienced noepisodes of hypotension or arrhythmia. Adverse respiratoryevents in the 19 etomidate patients included 4 episodes ofdesaturation with one 75-year-old patient requiring bag-maskventilation for a brief period of apnea. There was no significantdifference between the groups with episodes of hypoxia. Patientand physician visual analog pain assessment scores were similarbetween groups. Dickinson et al,11 in a retrospective review of 53children, with the majority of these children older than 10 years,described the drug’s use in the reduction of pediatric orthopedicprocedures. The authors found an 83% success rate after the firstattempt. Documented side effects included 1 patient with nauseaand 1 patient requiring a fluid bolus for transient hypotension.There were no reported incidences of patients who requiredassisted ventilation of any kind.11 Ruth et al12 performeda prospective study on 51 consecutive patients. Ninety-eightpercent of the patients achieved adequate sedation: 5 patientsdesaturated below 90% and 1 patient vomited; however, nopatient required assistance with ventilation or had a clinicallysignificant complication. Vinson and Bradbury10 retrospectivelyreviewed 134 patients treated with etomidate for proceduralsedation. Ninety-five percent of the patients were ‘‘extremely’’satisfied with their care. They reported a 5% rate of adverseevents including 5 cases of desaturation below 94%, all of whomrecovered uneventfully. Another smaller retrospective chartreview of 46 adults and 2 children identified 4/48 (8%) patientswith procedural failure.13 These authors found 10/48 (21%)patients with adverse reactions. Respiratory complicationsincluded 1 patient who developed transient apnea requiring bag-mask ventilation, while another patient required a nonrebreatherfor a desaturation below 90%. Other commonly reported side

Clinical Policy



effects of etomidate include pain with injection and myoclo-nus.12,17,69-71 Myoclonus reports ranging from 0% to 21% ofpatients receiving etomidate usually last less than 1 minute, butcan be dramatic and may resemble seizure activity.10,12,17,70

These tremors are generally benign and not epileptigenicactivity, but they can result in brief desaturation.69,72

In conclusion, there is a growing body of literaturesupporting the safe use of a large variety of agents for proceduralsedation and analgesia in the ED. Ketamine, midazolam,fentanyl, propofol, and etomidate are just a few of these agentsin common usage. Respiratory depression is the mostconcerning of side effects related to the use of these agents;however, careful preparation and administration have beenshown to prevent harmful sequelae.

Drug Administration Recommendations: Can ketamine,midazolam, fentanyl, propofol, and etomidate be safelyadministered for procedural sedation and analgesia in theED?

Level A recommendations. Ketamine can be safely adminis-tered to children for procedural sedation and analgesia in the ED.

Level B recommendations. Propofol can be safely admin-istered for procedural sedation and analgesia in the ED.

Nondissociative sedation agents should be titrated to clinicaleffect to maximize safety during procedural sedation in the ED.

The combination of fentanyl and midazolam is effective forprocedural sedation and analgesia in the ED.

Level C recommendations. Etomidate can be safelyadministered for procedural sedation and analgesia in the ED.

REFERENCES1. Core Content Task Force. The model of the clinical practice of

emergency medicine. Ann Emerg Med. 2001;37:745-770.2. Task Force on the Core Content for Emergency Medicine

Revision. Core content for emergency medicine. Ann Emerg Med.1997;29:792-811.

3. American College of Emergency Physicians. Clinical policy forprocedural sedation and analgesia in the emergency department.Ann Emerg Med. 1998;31:663-677.

4. Terndrup TE, Cantor RM, Madden CM. Intramuscular meperidine,promethazine, and chlorpromazine: analysis of use and compli-cations in 487 pediatric emergency department patients. AnnEmerg Med. 1989;18:528-533.

5. Chudnofsky CR, Wright SW, Dronen SC, et al. The safety offentanyl use in the emergency department. Ann Emerg Med.1989;18:635-639.

6. Chudnofsky CR, Weber JE, Stoyanoff PJ, et al. A combination ofmidazolam and ketamine for procedural sedation and analgesia inadult emergency department patients. Acad Emerg Med. 2000;7:228-235.

7. Wathen JE, Roback MG, Mackenzie T, et al. Does midazolam alterthe clinical effects of intravenous ketamine sedation in children?A double-blind, randomized, controlled, emergency departmenttrial. Ann Emerg Med. 2000;36:579-588.

8. Dachs RJ, Innes GM. Intravenous ketamine sedation of pediatricpatients in the emergency department. Ann Emerg Med. 1997;29:146-150.

9. Barsan WG, Tomassoni AJ, Seger D, et al. Safety assessment ofhigh-dose narcotic analgesia for emergency department proce-dures. Ann Emerg Med. 1993;22:1444-1449.

10. Vinson DR, Bradbury DR. Etomidate for procedural sedation inemergency medicine. Ann Emerg Med. 2002;39:592-598.

11. Dickinson R, Singer AJ, Carrion W. Etomidate for pediatric sedationprior to fracture reduction. Acad Emerg Med. 2001;8:74-77.

12. Ruth WJ, Burton JH, Bock AJ. Intravenous etomidate for pro-cedural sedation in emergency department patients. Acad EmergMed. 2001;8:13-18.

13. Keim SM, Erstad BL, Sakles JC, et al. Etomidate for proceduralsedation in the emergency department. Pharmacotherapy. 2002;22:586-592.

14. Swanson ER, Seaberg DC, Mathias S. The use of propofol forsedation in the emergency department. Acad Emerg Med. 1996;3:234-238.

15. Havel CJ Jr, Strait RT, Hennes H. A clinical trial of propofol vs.midazolam for procedural sedation in a pediatric emergencydepartment. Acad Emerg Med. 1999;6:989-997.

16. Skokan EG, Pribble C, Bassett KE, et al. Use of propofol sedationin a pediatric emergency department: a prospective study. ClinPediatr (Phila). 2001;40:663-671.

17. Burton JH, Bock AJ, Strout TD, et al. Etomidate and midazolam forreduction of anterior shoulder dislocation: a randomized, con-trolled trial. Ann Emerg Med. 2002;40:496-504.

18. Guenther E, Pribble CG, Junkins EP Jr, et al. Propofol sedation byemergency physicians for elective pediatric outpatient proce-dures. Ann Emerg Med. 2003;42:783-791.

19. Bassett KE, Anderson JL, Pribble CG, et al. Propofol for proceduralsedation in children in the emergency department. Ann EmergMed. 2003;42:773-782.

20. Miner JR, Biros M, Krieg S, et al. Randomized clinical trial ofpropofol versus methohexital for procedural sedation duringfracture and dislocation reduction in the emergency department.Acad Emerg Med. 2003;10:931-937.

21. Miner JR, Biros MH, Heegaard W, et al. Bispectral electroenceph-alographic analysis of patients undergoing procedural sedation inthe emergency department. Acad Emerg Med. 2003;10:638-643.

22. Wilson JE, Pendleton JM. Oligoanalgesia in the emergencydepartment. Am J Emerg Med. 1989;7:620-623.

23. Brown JC, Klein EJ, Lewis CW, et al. Emergency departmentanalgesia for fracture pain. Ann Emerg Med. 2003;42:197-205.

24. Joint Commission on Accreditation of Healthcare Organizations.Comprehensive Accreditation Manual for Hospitals, The OfficialHandbook. Chicago, IL: JCAHO Publication; 2004.

25. Krauss B, Green SM. Sedation and analgesia for procedures inchildren. N Engl J Med. 2000;342:938-945.

26. American Society of Anesthesiologists Task Force on Sedationand Analgesia by Non-Anesthesiologists. Practice guidelines forsedation and analgesia by non-anesthesiologists. Anesthesiology.2002;96:1004-1017.

27. Blackburn P, Vissers R. Pharmacology of emergency departmentpain management and conscious sedation. Emerg Med Clin NorthAm. 2000;18:803-827.

28. Glauser J, Cullison B. Procedural sedation: general principles andspecific pharmacotherapeutic strategies-part 1. EmergencyMedicine Reports. 2002;23:247-260.

29. Green SM, Krauss B. The semantics of ketamine. Ann Emerg Med.2000;36:480-482.

30. Schriger DL, Cantrill SV, Greene CS. The origins, benefits, harms,and implication of emergency medicine clinical policies. AnnEmerg Med. 1993;22:597-602.

31. Mace SE, Barata IA, Cravero JP, et al. EMSC Grant PanelWriting Committee on Pharmacologic Agents Used in PediatricSedation and Analgesia in the Emergency Department. Clinicalpolicy: evidence-based approach to pharmacologic agents usedin pediatric sedation and analgesia in the emergency de-partment. Ann Emerg Med. 2004;44:342-377.

Clinical Policy



32. Green SM, Krauss B. Pulmonary aspiration risk during emergencydepartment procedural sedation: an examination of the role offasting and sedation depth. Acad Emerg Med. 2002;9:35-42.

33. Pacifico A, Cedillo-Salazar FR, Nasir N Jr, et al. Conscioussedation with combined hypnotic agents for implantation ofcardioverter-defibrillators. J Am Coll Cardiol. 1997;30:769-773.

34. Pena BM, Krauss B. Adverse events of procedural sedation andanalgesia in a pediatric emergency department. Ann Emerg Med.1999;34:483-491.

35. Agrawal D, Manzi SF, Gupta R, et al. Preprocedural fasting stateand adverse events in children undergoing procedural sedationand analgesia in a pediatric emergency department. Ann EmergMed. 2003;42:636-646.

36. Ferrari LR, Rooney FM, Rockoff MA. Preoperative fastingpractices in pediatrics. Anesthesiology. 1999;90:978-980.

37. Maekawa N, Mikawa K, Yaku H, et al. Effects of 2-, 4- and 12-hourfasting intervals on preoperative gastric fluid pH and volume, andplasma glucose and lipid homeostasis in children. Acta Anaes-thesiol Scand. 1993;37:783-787.

38. Splinter WM, Schaefer JD, Zunder IH. Clear fluids three hoursbefore surgery do not affect the gastric fluid contents of children.Can J Anaesth. 1990;37:498-501.

39. Soreide E, Hausken T, Soreide JA, et al. Gastric emptying ofa light hospital breakfast. A study using real time ultrasonogra-phy. Acta Anaesthesiol Scand. 1996;40:549-553.

40. Ingebo KR, Rayhorn NJ, Hecht RM, et al. Sedation in children:adequacy of two-hour fasting. J Pediatr. 1997;131:155-158.

41. American Society of Anesthesiologists Task Force on Preopera-tive Fasting. Practice guidelines for preoperative fasting and theuse of pharmacologic agents to reduce the risk of pulmonaryaspiration: application to healthy patients undergoing electiveprocedures. Anesthesiology. 1999;90:896-905.

42. Green SM. Fasting is a considerationdnot a necessitydforemergency department procedural sedation and analgesia. AnnEmerg Med. 2003;42:647-650.

43. Engelhardt T, Webster NR. Pulmonary aspiration of gastriccontents in anesthesia. Br J Anaesth. 1999;83:453-460.

44. Green SM, Nakamura R, Johnson NE. Ketamine sedation forpediatric procedures: part 1, a prospective series. Ann EmergMed. 1990;19:1024-1032.

45. Green SM, Rothrock SG, Lynch EL, et al. Intramuscular ketaminefor pediatric sedation in the emergency department: safety profilein 1,022 cases. Ann Emerg Med. 1998;31:688-697.

46. Chudnofsky CR. Safety and efficacy of flumazenil in reversingconscious sedation in the emergency department. EmergencyMedicine Conscious Sedation Study Group. Acad Emerg Med.1997;4:944-950.

47. Yaster M, Nichols DG, Deshpande JK, et al. Midazolam-fentanylintravenous sedation in children: case report of respiratory arrest.Pediatrics. 1990;86:463-467.

48. Newman DH, Azer MM, Pitetti RD, et al. When is a patient safe fordischarge after procedural sedation? The timing of adverse effectevents in 1,367 pediatric procedural sedations. Ann Emerg Med.2003;42:627-635.

49. Bailey PL, Pace NL, Ashburn MA, et al. Frequent hypoxemia andapnea after sedation with midazolam and fentanyl. Anesthesiol-ogy. 1990;73:826-830.

50. Pohlgeers AP, Friedland LR, Keegan-Jones L. Combinationfentanyl and diazepam for pediatric conscious sedation. AcadEmerg Med. 1995;2:879-883.

51. American Medical Association, Council on Scientific Affairs. Theuse of pulse oximetry during conscious sedation. JAMA. 1993;270:1463-1468.

52. Aughey K, Hess D, Eitel D, et al. An evaluation of pulse oximetry inprehospital care. Ann Emerg Med. 1991;20:887-891.

53. Rubin DM, Eisig S, Freeman K, et al. Effect of supplemental gases onend-tidal CO2 and oxygen saturation in patients undergoing fentanyland midazolam outpatient sedation. Anesth Prog. 1997;44:1-4.

54. Block AJ, Boysen PG, Wynne JW, et al. Sleep apnea hypopnea andoxygen desaturation in normal subjects: a strong male pre-dominance. N Engl J Med. 1979;300:513-517.

55. Bilotta JJ, Floyd JL, Waye JD. Arterial oxygen desaturation duringambulatory colonoscopy: predictability, incidence, and clinicalinsignificance. Gastrointest Endosc. 1990;36:S5-S8.

56. Arrowsmith JB, Gerstman BB, Fleischer DE, et al. Results fromthe American Society for Gastrointestinal Endoscopy/US Foodand Drug Administration collaborative study on complicationrates and drug use during gastrointestinal endoscopy. Gastro-intest Endosc. 1991;37:421-427.

57. Ceravolo FJ, Meyers HE, Michael JJ, et al. Full dentitionperiodontal surgery using intravenous conscious-sedation. Areport of 10,000 cases. J Periodontal. 1986;57:383-384.

58. Tobias JD. End-tidal carbon dioxide monitoring during sedationwith a combination of midazolam and ketamine for childrenundergoing painful, invasive procedures. Pediatr Emerg Care.1999;15:173-175.

59. McQuillen KK, Steele DW. Capnometry during sedation/analgesiain the pediatric emergency department. Pediatr Emerg Care.2000;16:401-404.

60. Miner JR, Heegaard W, Plummer D. End-tidal carbon dioxidemonitoring during procedural sedation. Acad Emerg Med. 2002;9:275-280.

61. Wright SW. Conscious sedation in the emergency department:the value of capnography and pulse oximetry. Ann Emerg Med.1992;21:551-555.

62. Hart LS, Berns SD, Houck CS, et al. The value of end-tidal CO2

monitoring when comparing three methods of conscious sedationfor children undergoing painful procedures in the emergencydepartment. Pediatr Emerg Care. 1997;13:189-193.

63. Rosow C, Manberg PJ. Bispectral index monitoring. AnesthesiolCl in North America: Annual Anesth Pharm. 1998;2:89-107.

64. Bower AL, Ripepi A, Dilger J, et al. Bispectral index monitoring ofsedation during endoscopy. Gastrointest Endosc. 2000;52:192-196.

65. Gill M, Green SM, Krauss B. A study of the Bispectral IndexMonitor during procedural sedation and analgesia in theemergency department. Ann Emerg Med. 2003;41:234-241.

66. Agrawal D, Feldman HA, Krauss B, et al. Bispectral indexmonitoring quantifies depth of sedation during emergency de-partment procedural sedation and analgesia in children. AnnEmerg Med. 2004;43:247-255.

67. Kennedy RM, Porter FL, Miller JP, et al. Comparison of fentanyl/midazolam with ketamine/midazolam for pediatric orthopedicemergencies. Pediatrics. 1998;102:956-963.

68. Bell GD, McCloy RF, Charlton JE, et al. Recommendations forstandards of sedation and patient monitoring during gastrointes-tinal endoscopy. Gut. 1991;32:823-827.

69. Helmers JH, Adam AA, Giezen J. Pain and myoclonus duringinduction with etomidate. A double-blind, controlled evaluation ofthe influence of droperidol and fentanyl. Acta Anaesthesiol Belg.1981;32:141-147.

70. Van Keulen SG, Burton JH. Myoclonus associated with etomidatefor ED procedural sedation and analgesia. Am J Emerg Med.2003;21:556-558.

71. McDowall RH, Scher CS, Barst SM. Total intravenous anesthesiafor children undergoing brief diagnostic or therapeutic proce-dures. J Clin Anesth. 1995;7:273-280.

72. Modica PA, Tempelhoff R. Intracranial pressure during inductionof anaesthesia and tracheal intubation with etomidate-inducedEEG burst suppression. Can J Anaesth. 1992;39:236-241.

Clinical Policy



Evidentiary Table.

Study Design Findings Limitations Conclusions Grade

Terndrup et al4 Descriptive retrospectivechart review; 487pediatric patientsidentified who receiveddemerol, phenergan,and thorazine for pro-cedural sedation andanalgesia in the ED

3 (0.6%) patients found to have sig-nificant respiratory depression re-quiring naloxone

Design Demerol, phenergan, andthorazine combinationappears to be safe inpediatric ED patients

III

Chudnofskyet al5

Retrospective chartreview

841 patients receiving fentanyl; aver-age dose 180 mg (range 25–1,400mg); 6 (0.7%) respiratory compro-mise; 3 (0.4%) hypotension; nosequelae associated with comorbidfactors (ie, alcohol, head trauma)

Design Fentanyl is safe when usedin the ED; the safetywindow can be maxi-mized by careful dosingand titration of fentanylduring administration;the authors suggest thatnaloxone be availableduring procedural seda-tion and analgesia

III

Chudnofskyet al6

Prospective, observa-tional trial

ED study of 70 patients aged >18 y allof whom received 0.07 mg/kg of IVmidazolam followed by 2 mg/kg ofIV ketamine; indications for proce-dural sedation and analgesia includ-ed incision and drainage (26%),fracture/joint reduction (26%), andother (8%); there were no episodesof hallucinations, delirium, or otheremergency reactions; 18 (25%)patients recalled dreaming duringsedation, 2 (3%) were unpleasant,and 3 (4%) were both pleasant andunpleasant

No control groupand lack of inter-rater reliability inrecognition ofemergence reac-tions

Midazolam and ketaminein combination provideeffective and safe pro-cedural analgesia inadult ED patients

II

Wathen et al7 Double-blind, random-ized, controlled study

ED study of 266 patients aged 4.5 moto 16 y to evaluate frequency andseverity of adverse effects in patientreceiving ketamine with or withoutmidazolam for sedation; 129patients received ketamine and 137received ketamine and midazolam;the incidence of emergence reactionwas not affected by the addition ofmidazolam

No majorlimitations

Ketamine and combinedketamine and midazolamprovide equally effectivesedation

I

Dachs andInnes8

Observational study 30 children aged 18 mo to 8 y wereenrolled; a bolus of 1.5 mg/kgproduced adequate sedation in17/18 (94%) patients

Design Ketamine is an effectivemethod of producingsedation and analgesiain children

III

Barsan et al9 Prospective, multicenterclinical trial

72 patients receiving 1.5–3 mg/kgof IV meperidine; complications in-cluded 3 patients with nausea, 6with vomiting, and 1 with hypoten-sion; no reversal with naloxone wasneeded

Limited number ofpatients to dem-onstrate compli-cations thatoccur with lowfrequency

High-dose narcotic analge-sia is safe when used inselected patients beforepainful procedures

II

Clinical Policy



Evidentiary Table (continued).


Vinson andBradbury10

Retrospective observa-tional study in 3 affili-ated suburban EDs;134 patients enrolled,aged 6–93 y; chart re-view performed withprospective question-naire sent to patientswith 120 (90%) com-pleted

Moderate sedation achieved in 48(32%) patients and deep sedationinduced in 102 (68%) patients asmeasured by the Aldrete Postanes-thetic Recovery Score; 5 (4.7%) of 7adverse reactions were the result ofO2 desaturation requiring face maskO2; 4 of 5 also received bag-assistedventilation; these patients were >55y and had received relatively higherdoses of etomidate (0.23 mg/kg vs0.19 mg/kg in the nonrespiratorycompromised patients); no intuba-tions; 2 patients experienced eme-sis; 114 (94%) responders statedthey would be ‘‘extremely’’ willing tohave this medication again

Design; no stan-dardized dosing;adjunctive medi-cations given in23% of proce-dures; onlypatients meetingASA status I or IIwere candidatesfor procedural se-dation; somepatients were en-tered into thestudy more thanonce

The authors concludedthat etomidate appearsto be a brief, safe, andeffective drug for emer-gency procedural seda-tion; however, as theauthors point out, thepower of the study is notenough to evaluate in-cidence of all complica-tions

III

Dickinsonet al11

Retrospective descrip-tive study; 53 charts ofchildren aged <18 ywho received IV etomi-date were studied

There were no major adverse effects; 1patient reported nausea and anotherwas given a fluid bolus for transienthypotension; no patient requiredventilatory assistance

Retrospective de-sign; selectionbias; documenta-tion concerns asdemonstrated byno reported casesof myoclonus

Results suggest thatetomidate may be usedsafely in children but theauthors warn that fur-ther larger prospectivestudies are indicated

III

Ruth et al12 2-phase feasibility study;retrospective pilotfollowed by a prospec-tive descriptive study

Procedural success was achieved in56/60 (93%) patients, with ade-quate sedation as documented bythe physician in 59/60 (98%)patients; 12 complications werereported, including O2 desaturation<90% (5), myoclonus (4), vomiting(1), pain with injection (1), anda ‘‘brief’’ bradycardic episode; noneof the patients required ventilatoryassistance

13 patients hadmissing nursingrecords and an-other 4 patientslacked depth ofsedation informa-tion

Etomidate administeredIV for procedural seda-tion in the ED was botheffective and safe in thisgroup of patients

III

Keim et al13 Retrospective chart re-view; the charts of 46adults and 2 childrenwere reviewed

10/48 (21%) patients had adversereactions; 1 (2%) patient had tran-sient apnea requiring bag-mask ven-tilation (this patient had multipledoses of analgesics as well); 1 (2%)patient required a nonrebreather fora desaturation <90%; emesis in 2(4%) patients; anxiety in 2 (4%)patients; and 4 (8%) patients hadfailed procedures

Design; not anetomidate-onlystudy so someadverse sideeffects may berelated to othersedatives andanalgesics; docu-mentationconcerns as dem-onstrated by noreported cases ofmyoclonus

The authors concludethat although furtherstudy is indicated,etomidate holds promiseas a procedural agent inthe ED; the authors fur-ther stress the need foradequate monitoring inthe ED when usingagents that may inducedeep sedation

III

Swansonet al14

Convenience sample; 20patients received 2mg/kg of fentanyl andpropofol (0.21 mg/kg/min)

3 patients had pain on injection; 1patient experienced hypotension <1min and 1 patient had apnea <30 s,responded to stimulation; 1 patienthad apnea with desaturation to 86%requiring ‘‘brief’’ bag-mask ventila-tion

Design, limitedstudy power

Further study required III

Clinical Policy





Havel et al15 Prospective, randomized,blinded, comparisontrial; 89 patients, aged2–18 y received mor-phine for pain, andthen midazolam orpropofol for sedation inthe ED

Sedation scores and rates of O2

desaturation similar betweengroups; recovery time for the propo-fol group was 14.9G11.1 min,compared with 76.4G47.5 min forthe midazolam group (P<.001)

Sample size smallfor comparison ofcomplicationrates

Propofol induced sedationas effectively as mida-zolam but with a shorterrecovery time; compli-cation rates for propofoland midazolam werecomparable

I

Skokan et al16 Prospective, observa-tional study of pediat-ric patients receivingpropofol for proceduralsedation and analgesiain the ED

Mean dose 3.3 mg/kg; 30% experi-enced O2 desaturation, but only 1required assistance with ventilation;mean recovery time 18 min

Not controlled andnot randomized

Propofol provides safe andeffective sedation forprocedural sedation andanalgesia

III

Burton et al17 Prospective, double-blinded, randomizedtrial of etomidate andmidazolam for proce-dural sedation in ante-rior shoulderdislocation reductions;opiates for analgesiagiven at physician dis-cretion; all patientsaged >18 y

19/41 patients included for analysis inthe etomidate group; the mediantime for procedural sedation was 10min for etomidate and 23 min formidazolam (P<.001); no significantdifference for the medial lowest PARscore was observed betweenagents; 1 patient in each grouprequired brief bag-mask ventilation;myoclonus was noted in the etomi-date group in 4/19 (21%) patients;successful reduction occurred in allbut 2 patients in each group

No predefined tar-get depth of se-dation; limitedpower to identifypotential compli-cations

Etomidate provides effec-tive sedation for perfor-mance of shoulderdislocation reduction;duration of proceduralsedation with etomidateis shorter than sedationperformed with midazo-lam

II

Guentheret al18

Prospective, observa-tional trial evaluatingpropofol for proceduralsedation in 87 patientswith 291 separatesedations; all seda-tions performed in EDsedation unit; propofoldosing was an initialdose of 1 mg/kgfollowed by 0.5 mg/kgsupplements as need-ed; fentanyl used asa single bolus dose foranalgesia in 97% ofcases; median age 6 y

100% success rate for all procedures;success defined as completion ofprocedure with no adverse events;93% of children maintained satura-tions >90%, 4% developed partialairway obstruction corrected withjaw thrust, 3/291 (1%) experiencedapnea mandating brief bag-maskventilation; transient decrease insystolic blood pressure noted in allbut 4 patients; no evidence of poorprofusion noted; 1 episode of eme-sis; no report of aspiration

Patients werepreselected andscheduled forelective proce-dures; no objec-tive monitoringfor depth of seda-tion; no ETCO2

monitoring

Propofol is safe and effec-tive when administeredby pediatric emergencyphysicians in ED-associ-ated sedation units

II

Bassett et al19 Prospective observation-al study of 392 con-secutive patients aged1–18 y (median age 8y) with 393 proce-dures; all participantsreceived propofol witha narcotic for analge-sia; initial dosing ofpropofol of 1 mg/kgwith subsequent dos-ing of 0.5 mg/kg atphysician discretion

O2 saturation was maintained >90% in95% of patients; median duration ofhypoxia ranged from 1–3 min with nopatients requiring intubation; partialairway obstruction was noted in 11/393 (3%) patients; an additional 3/393 (0.08%) patients required bag-mask ventilation for 5 s to 1 min;23/393 (6%) patients experiencedclinically insignificant transient bra-dycardia; 331/393 (84%) patientsexperienced clinically insignificantdecreases in systolic blood pressurewith no episodes of poor profusiondocumented; all procedures werecompleted successfully

Design; depth of se-dation was notobjectivelyscored; no stan-dardization fortiming of repeatdosing of seda-tive agent; noETCO2 monitoringfor detection ofhypoventilation

Propofol is safe and effec-tive when administeredin the ED

II

Clinical Policy





Miner et al20 Prospective, randomizedcomparison of metho-hexital and propofol forprocedural sedationand analgesia in theED; morphine used onall patients for paincontrol

Respiratory depression rate of 48% formethohexital and 49% for propofol(P=.88); BIS 66.2 for methohexitaland 66 for propofol (P=.50); 11/34patients receiving single dose ofmedication experienced respiratorydepression; 39/69 patients receiv-ing multiple doses experienced re-spiratory depression (P=.021)

Not blinded Both agents equally effi-cacious for proceduralsedation and analgesiafor fracture reduction inthe ED

II

Miner et al21 Prospective, observa-tional study usinga convenience samplethat included patientsreceiving propofol,methohexital, etomi-date, and fentanyl/midazolam to deter-mine if a correlationexisted between theBIS and the rate ofrespiratory depression,patients’ pain percep-tion, recall of the pro-cedure, andsatisfaction

108 patients were divided into 4groups on the basis of the lowestBIS recorded during the procedure;no serious adverse effects werenoted; the rate of respiratory de-pression in the 2 groups with scores>70 was significantly lower than the2 groups with scores%69 (P=.019);the group with the lowest BIS (>85)had significantly higher rates of pain(P=.003) and recall (P=.001) com-pared with the combined group ofpatients (<85)

Design; patients’state of healthbefore study wasnot described

The optimally sedatedpatients were repre-sented by those with thelowest BIS between 70and 85 as illustrated bythe same VAS outcomeas more deeply sedatedpatients and the samerate of respiratory de-pression as patients whowere less sedated

III

Wilson andPendleton22

Chart review 198 charts reviewed of patients ad-mitted through the ED with a widerange of acutely painful medical andsurgical conditions

Design Both emergency physi-cians and consultantswere providing a signifi-cant degree of oligo-analgesia

III

Brown et al23 Retrospective cohort Analysis of the ED component of theNCHS NHAMCS; total of 2,828isolated closed extremity and clavi-cle fractures; 73% of patients aged0–R70 y received analgesia, and54% received narcotic analgesiaamong patients with documentedmoderate and severe pain

Design Patients with documentedfractures frequently donot receive analgesia;compared with adults,pediatric patients areless likely to receiveanalgesics includingnarcotics

III

JCAHO24 Standards document Design III

Krauss andGreen25

Review Design III

American Socie-ty of Anes-thesiologistsTask Force onSedation andAnalgesia byNon-Anes-thesiolo-gists26

Clinical guideline Design ASA clinical guideline III

EMSC GrantPanel on Phar-macologicAgents Usedin PediatricSedation andAnalgesia inthe ED31

Evidence-based review ofthe safety and efficacyof etomidate,fentanyl/midazolam,ketamine, methohexi-tal, pentobarbital, andpropofol in children

Design The policy provides an ev-idence-based approachto drugs used in pediat-ric sedation and analge-sia in the ED

III

Clinical Policy





Green andKrauss32

Review Review of pulmonary aspiration riskduring ED procedural sedation andanalgesia with extensive list ofreferences

Design The risk of aspiration dur-ing ED procedural seda-tion and analgesiaappears to be extremelylow; also there is nocompelling evidencethat mandates specifictime frames for preseda-tion fasting for eithersolids or liquids

III

Pacifico et al33 Case series of 162patients undergoingICD implantation underprocedural sedationwith combination ofmidazolam, morphine,promethazine, andetomidate

No episodes of apnea, hypoxia, orhypotension

Not randomized,not blinded;small samplesize

ICD implantation is safeunder moderate anddeep sedation protocolsusing combinations ofintravenous anestheticagents including etomi-date, midazolam, mor-phine, and promethazine

III

Pena andKrauss34

Prospective observation-al study; 1,180patients <21 y under-going procedural seda-tion in ED for painfulprocedures and diag-nostic imaging

2.3% experienced adverse events in-cluding hypoxia, paradoxical reac-tions, laryngospasm, bradycardia,stridor, or emesis

Not randomized,not blinded

2.3% adverse reaction ratewith no serious compli-cations

III

Agrawal et al35 Prospective, observa-tional study

1,014 patients studied; inadequatefasting documentation in 11%; nosignificant difference in adverseoutcomes between fasting andnonfasting patients

Inadequate powerto determinedifferences in as-piration afteremesis;nonblinded seda-tion recorders(ED nurses caringfor patients); nofollow-up ofpatients after EDdischarge

56% of patients not fasted;no difference in adverseevents between fastedand unfasted groups

III

Ferrari et al36 Survey of preoperativefasting guidelines atpediatric anesthesiafellowship programs

44/55 institutions responded; no clearconsensus on guidelines; in 50% ofinstitutions: clear liquids wereallowed up to 2 h before anesthesiaand solids were restricted aftermidnight in all children >3 y

Design; evaluatespractices in prep-aration for gener-al anesthesia andnot proceduralsedation

There is a large degree ofvariation in preoperativefasting practices forchildren undergoingelective surgery

III

Maekawaet al37

Prospective randomizedtrial of 105 pediatricpatients aged 1–14 y;the study evaluatesthe effects of preoper-ative fasting at 2-h, 4-h, and 12-h intervals

Gastric aspirates taken from patientsimmediately after intubationrevealed no significant differencebetween groups with gastricvolumes or gastric fluid pH

Insufficient gastricaspiratesvolumesprohibited pHevaluation in 28samples; howev-er, there was nodifference be-tween groups

There was no significantdifference in the gastricpH and gastric volumesof patients who werefasted for 2 h, 4 h, and12 h preoperatively

I

Clinical Policy





Splinter et al38 Prospective randomizedsingle-blinded trial of121 patients aged 2–12 y; the study evalu-ates the effects of un-limited clear fluids upto 3 h before generalanesthesia inductioncompared withprolonged fast

Prolonged fast with mean duration of14 h demonstrated no significantdifference in gastric volumes andgastric pH compared with 3-h fast

No major limitations Drinking clear fluids up to3 h before surgery hasno significant effect ongastric volume or gastricpH

I

Soreide et al39 Prospective observation-al study of healthy fe-male volunteers inwhich ultrasono-graphy was used todetermine the amountof food particlesremaining after a lightmeal

Subjects were found to have solid foodparticles after 2 h, and all volunteersdemonstrated empty stomachs after4 h

Unable to evaluatepH changes

In healthy subjects, com-plete stomach emptyingafter a light meal maynot occur until 240 minafter ingestion

II

Ingebo et al40 Prospective observation-al study of 285 pediat-ric patients aged 1–18.6 y; gastriccontents were collect-ed immediately after IVsedation via endo-scopic suction; dura-tion of fasting afterclear fluid ingestionranged from 0.5–24 h(mean 6.7G5.3 h)

The results demonstrated no signifi-cant difference between gastricvolumes and gastric pH when com-paring groups with fasting times>30 min

Compared data tohistorical stand-ards; does notspecifically lookat time from in-gestion of milkand solids beforesedation

Fasting >2 h after inges-tion of clear liquids doesnot significantly changegastric pH or gastricvolumes during proce-dures

II

American Socie-ty of Anes-thesiologistsTask Force onPreoperativeFasting41

Clinical guideline Design ASA guideline for preoper-ative fasting

III

Engelhardt andWebster43

Review Review of risk factors of gastricaspiration during general anesthesia

Design The incidence of pulmo-nary aspiration in gener-al surgical patients isonly slightly greater inthe pediatric and ob-stetric patients

III

Green et al44 Prospective uncontrolledtrial; 108 childrenaged 14 mo to 13 y,mean age 54 mo,using IM ketamine 4mg/kg

6 emesis during recovery; 1 case ofemesis and laryngospasm withtransient cyanosis with no adversesequelae

Design; no validatedoutcome measure

Ketamine can be used ef-fectively in the ED, butproviders should haveexpertise in airway man-agement

II

Green et al45 Retrospective observa-tional study; an IMketamine protocol wasfollowed in a collectionof 1,022 pediatriccases aged <15 y; 431prospectively collect-ed data sheets werethen reviewed forcomplications, adjunc-tive medications, etc

Transient airway complications oc-curred in 1.4% of patients, withlaryngospasm occurring in 4/431patients and apnea in 2/431patients

Design; only 431 ofthe treatedpatients had datasheets completed

Ketamine has a wide safe-ty margin and may beadministered safely inthe ED with appropriatemonitoring and definedprotocols; IV accessmay not be required withketamine sedation

II

Clinical Policy





Chudnofsky46 Multicenter, randomized,double-blind controlledtrial; patients received2 mg/kg fentanylfollowed by midazolamtitrated to sedation;patients then receivedplacebo or flumazenil

Flumazenil dosing did not decreasetime to discharge

No major limitations Flumazenil is safe and ef-fective in reversingmidazolam-induced se-dation in ED patients

I

Yaster et al47 Case report Large incremental doses of both mida-zolam and fentanyl over a 4-minperiod; no supplemental O2 avail-able; respiratory arrest, ‘‘instanta-neous resolution’’ with naloxone

Design Reversal agents should beavailable during proce-dural sedation and anal-gesia

III

Newman et al48 Prospective cohort inwhich a database re-view was performedevaluating timing of184 adverse eventsdocumented in 1,341sedation events ina pediatric ED; medianage with adverseevents was 64.4 mo

160/184 records for timing of eventswere complete; 92% of adverseevents occurred during the proce-dure with 8% occurring after theprocedure; serious adverse eventsoccurred a median of 2 min after thefinal medication dosing; no primaryserious adverse effects occurred>25 min after final medication

Incomplete data-base; no stan-dardized dis-charge criteria

In children without seriousadverse effects duringprocedural sedation,discharge from the EDmay be safe approxi-mately 30 min after finalmedication administra-tion

II

Bailey et al49 Randomized, double-blind crossover study

The authors investigated the respira-tory effects of midazolam (0.05 mg/kg) and fentanyl (2 mg/kg) in 12volunteers; apnea was defined as norespirations for 15 s and was ob-served in 50% of volunteers receiv-ing both drugs; no complicationsnoted; midazolam alone caused nohypoxemia; fentanyl alone causedhypoxemia in 50% without clinicalcorrelate; midazolam plus fentanylcaused hypoxemia in 92% withoutclinical correlate and apnea in 50%

Small sample size;selected volun-teers

Patients receiving bothmidazolam and an opiateare at risk for hypoxiaand apnea; adequateprecautions includingmonitoring of patientoxygenation with pulseoximetry, the adminis-tration of supplementalO2, and the availabilityof persons skilled in air-way management arerecommended

II

Pohlgeerset al50

Retrospective chart re-view of a standardizedprotocol in 133 con-secutive patients

11% desaturation to <90%, 0.7%vomited, 0.7% pruritus

Design The combination of diaze-pam and fentanyl formoderate sedation issafe in the pediatric pa-tient requiring emergentorthopedic procedures

III

American Medi-cal Associa-tion Councilon ScientificAffairs51

Literature review Design III

Aughey et al52 Prospective, cross-sec-tional study

Paired measurements of pulse oxime-try vs hemoglobin saturation

Design Strong correlation wasfound between studygroups

II

Rubin et al53 Randomized controlledtrial; 46 ASA I and IIstudy subjects re-ceived 0.7 mg/kg offentanyl and titratedmidazolam

1 group received nitrous oxide and theother group received 100% O2 withETCO2 and O2 saturation measured

Low power to showdifference

No significant difference inETCO2 and O2 betweenstudy groups

III

Block et al54 Prospective observation-al trial

Desaturation in 43% of asymptomaticmales documented during sleep

No major limitations Significant desaturationoccurs during sleep innormal study patients

I

Clinical Policy





Bilotta et al55 103 consecutivepatients aged 22–96 yundergoing colono-scopy; diazepam plusmeperidine

41% desaturation rate; age correlatedto desaturation (P<.05), but pulseoximetry and other historical meas-ures did not

No clinical correla-tion; not blinded

No adverse outcomes;oximetry monitoring maynot be clinically useful

III

Arrowsmithet al56

Aggregate study usingcomputer-based datacollection system:19,363 proceduresusing midazolam, diaz-epam with or withoutnarcotics

0.3 deaths/1,000 procedures; 5.4serious cardiopulmonary events/1,000 procedures (not defined)

Study limited byretrospective de-sign and lack ofinformation onindications forprocedures andon patient clinicalstatus; 17 sites,1 site reported54% of the mida-zolam-associatedcardiorespiratoryevents

Serious cardiorespiratoryevents and death areuncommon during pro-cedural sedation; con-comitant use ofnarcotics and urgentand emergent proce-dures, however, did in-crease the risk ofserious cardiorespirato-ry events

III

Ceravolo et al57 10,000 periodontalcases over a period of11 y using IV meperi-dine, diazepam, andmethohexital in 7,443patients; diazepamplus methohexital in2,557 patients (ages9–78 y)

4.1% phlebitis; 0.37% with nausea;0 cases of vomiting

Retrospective; lackof generalizability

No major complicationsrecorded in 10,000cases of procedural se-dation and analgesia

III

Tobias58 Prospective case cohort;50 children undergoingprocedural sedationand analgesia withmidazolam andketamine had side-stream ETCO2 measured

3 episodes O2 desaturation; 2episodes of hypercapnia (ETCO2 >50mm Hg); 1 episode apnea detectedonly by ETCO2 monitor

Not randomized, notblinded, not con-trolled

The addition of capnome-try provides an addition-al monitor to detectrespiratory depression

III

McQuillen andSteele59

Prospective, observation-al series of 106 chil-dren aged 1–16 yundergoing proceduralsedation and analge-sia; main outcomemeasure was change inETCO2

ETCO2 increased a mean of 6.7 mm Hg(range 0.16–22.3; P<.00001)

Not randomized ETCO2 is useful in assessingventilatory effort duringprocedural sedation andanalgesia

II

Miner et al60 Prospective observation-al study; 74 adultstreated with variousagents, monitored withcapnometry, oximetry,OAA/S

Respiratory depression in 45% ofpatients; oximetry detected onethird of the patients with respiratorydepression; no correlation betweenOAA/S and capnometry; capnome-try changes (ETCO2 >50 mm Hg,absent ETCO2 waveform, or change inETCO2 >10 mm Hg) able to detect allclinical cases of respiratory depres-sion

Post-hoc analysis ofETCO2 perfor-mance

Capnometry may add tosafety of procedural se-dation and analgesia byearly detection of hypo-ventilation

II

Wright61 Prospective, nonblinded,nonrandomized, non-controlled observation-al trial; 27 treated withbenzodiazepine and/ornarcotic

Average ETCO2 increased from 36 to 42mm Hg; O2 saturation decreasedfrom 98% to 94%; 1 episode ofapnea lasting 30 s responded toverbal stimuli

No clinical correla-tion

Pulse oximetry recommen-ded for detection of un-recognized hypoxemiaduring conscious seda-tion

II

Clinical Policy





Hart et al62 42 patients randomlyassigned to fentanyl,fentanyl-midazolam, orMPC; ETCO2 prospec-tively measured duringprocedural sedationand analgesia

Respiratory depression occurred in20% of fentanyl, 23% of fentanyl-midazolam, and 11% of MPCpatients, respectively (P=NS);capnometry detected subclinical re-spiratory depression better thanoximetry

Small number ofstudy subjects

Capnometry provides earli-er detection of subclini-cal respiratorydepression than doespulse oximetry or respi-ratory rate alone

II

Rosow andManberg63

Summary article Design III

Bower et al64 Prospective, observa-tional study of 50adults undergoing pro-cedural sedation andanalgesia for endosco-py; BIS analysis com-pared to the OAA/S

Temporal correlations between BISanalysis and OAA/S (r=0.59;P<.0001)

Not blinded BIS monitoring temporallycorrelates with theOAA/S scale, providingan objective measure ofsedation

II

Gill et al65 Prospective observation-al study measuring BISand the modifiedRamsay SedationScale score in EDpatients undergoingprocedural sedationand analgesia; conve-nience sample of 37patients; 270 pairedreadings obtained

Statistically significant (P<.0005) butclinically moderate correlation(Spearman r=0.690) that showed‘‘wide variability’’; ROC curves dem-onstrate moderate discriminatorypower at all Ramsay Sedation Scalescore thresholds

Convenience sam-ple design; smallsample size;Ramsay SedationScale scoreassigned by mul-tiple investiga-tors

BIS monitoring reliablypredicts patients under-going procedural seda-tion and analgesia whoare sedated to the levelof general anesthesia,but didn’t discriminatebetween mild-moderatesedation or moderate-deep sedation

III

Agrawal et al66 Prospective observation-al trial; evaluated abil-ity of BIS to monitordepth of sedation innondissociative proce-dural sedation; conve-nience sample of 20children undergoingprocedural sedation inthe ED; 217 pairedreadings obtained

Significant correlation between BISand Ramsay Sedation Scale scores(P<.001); BIS showed increasedvariability at moderate-to-deeplevels of sedation as measured bythe Ramsay Sedation Scale score;ROC curves demonstrate moderate-to-high discriminatory power atmoderate Ramsay Sedation Scalescore thresholds

Convenience sam-ple design; smallsample size

BIS monitoring correlatedwith traditional clinicalsedation scores; BISvalues between 60 and90 predicted with mod-erate accuracy andreliability clinical levelsof sedation

II

Kennedy et al67 Prospective, single-blinded, randomizedcontrolled trial ofketamine and midazo-lam vs fentanyl andmidazolam; 260patients aged 5–15 yundergoing orthopedicprocedures; allpatients were ASA I-II;videotaped blindedassessments of ade-quacy of sedation,documented efficacyassessments, and ob-jective monitoringwere used

Ketamine and midazolam resulted inhypoxic events in 6% of patientscompared with 25% in the fentanyland midazolam group (P=.001);there was documented greater phy-sician satisfaction with ketamineand midazolam measured by VAS8.71G2.21 vs 9.61G0.78 (P=.001)with fentanyl and midazolam anda lower OSBD-R scored by blindedobservers and blinded treatingphysicians 1.08G1.12 vs2.70G2.16 (P<.0001)

Only evaluated chil-dren aged 5–15 y;unclear if theseresults apply out-side of this agerange especiallyin the very youngand elderly

Both drug regimens aresafe and effective in thestudy groups; however,ketamine and midazolamappears more effica-cious and demonstrateda higher safety profilewith fewer episodes ofdocumented hypoxiaand/or airway interven-tions

I

Bell et al68 British Society of Gastro-enterology, endoscopyguidelines

Design Policy guideline III

Clinical Policy





Helmers et al69 Prospective double-blindcontrolled trial todetermine if IV injec-tion of droperidol orfentanyl before etomi-date could attenuateside effects of pain andmyoclonus; 83patients aged 14–78 y;a severity score wasassigned to pain andinvoluntary move-ments

The incidence of pain on delivery inpatients was: etomidate+normalsaline 17% (5/29), etomidate+droperidol 16.7% (4/24), etomi-date+fentanyl 18.5% (5/27);incidence of myoclonus was:etomidate+normal saline 37.9%(11/29), etomidate+droperidol12.5% (3/24), etomidate+fentanyl11% (3/27); P<.5 for the groupsusing etomidate+another drug vsetomidate alone

Not a sedationstudy; sideeffects measuredafter initial induc-tion only; poten-tial bias inscoring of painand myoclonus

Both fentanyl and droperi-dol showed a significantdifference in attenuatingmyoclonus after etomi-date administration

II

Van Keulen andBurton70

Case report of myoclonusduring proceduralsedation after etomi-date administration

3 cases of myoclonus described rang-ing from brief, mild tremors to severemyoclonic tightening with at least 1patient experiencing transient desa-turation to 85%

Design Physicians must be pre-pared to provide briefperiods of supportincluding respiratoryassistance when etomi-date-associated myoclo-nus is encountered

III

McDowallet al71

Retrospective review of971 pediatric oncologypatients at a universityhospital; 101 receivedIV etomidate (0.3 mg/kg) combined with ei-ther fentanyl or alfen-tanil for briefdiagnostic or thera-peutic procedures

Etomidate was associated with moreepisodes of vomiting (9.9%) andagitation (4%) compared withpropofol (0.5% and 1.2%, respec-tively); no procedural failuresreported with etomidate

Retrospectivedesign; efficacywas not objec-tively measured

Etomidate is an effectiveagent but had morevomiting and agitationcompared with propofol;however, propofol hada greater incidence ofhypoxia (15.7%) thanetomidate (2%)

III

Modica andTempelhoff 72

Prospective observation-al study; included 8patients with space-occupying lesions andundergoing surgery;aged 18–71 y; anes-thesia was inducedwith 0.2-mg/kg ofetomidate followed bya 20-mg/min infusion

Patients were monitored during induc-tion with IV etomidate for EEG burstsuppression and changes in intra-cranial pressure; the etomidatebolus required to reach burst sup-pression was 1.28G0.11 mg/kg

Small sample size Etomidate resulted insignificant reduction inintracranial pressureduring intubation

II

ASA, American Society of Anesthesiologists; BIS, Bispectral Index Score; EEG, electroencephalogram; ICD, implantable cardioverter defibrillator; IM, intra-

muscular; IV, intravenous; MPC, meperidine-promethazine-chlorpromazine; NCHS, National Center for Health Statistics; NHAMCS, National Hospital

Ambulatory Medical Care Survey; NS, not significant; OAA/S, Observer’s Assessment of Alertness and Sedation; PAR, postanesthetic recovery; ROC, receiver

operating characteristic; VAS, visual analog scale.

Clinical Policy



Appendix A. Literature classification schema.*

Design/

Class Therapyy Diagnosisz Prognosisx

1 Randomized,controlledtrial ormeta-analysesof randomizedtrials

Prospectivecohortusing acriterionstandard

Populationprospectivecohort

2 Nonrandomizedtrial

Retrospectiveobservational

Retrospectivecohort

Case control

3 Case series Case series Case seriesCase report Case report Case reportOther (eg,

consensus,review)

Other (eg,consensus,review)

Other (eg,consensus,review)

*Some designs (eg, surveys) will not fit this schema and should be

assessed individually.yObjective is to measure therapeutic efficacy comparing R2 interventions.zObjective is to determine the sensitivity and specificity of diagnostic tests.xObjective is to predict outcome including mortality and morbidity.

Appendix B. Approach to downgrading strength of evidence.

Design/Class

Downgrading 1 2 3

None I II III1 level II III X2 levels III X XFatally flawed X X X

Clinical Policy



clinical policy_ procedural sedation and analgesia in the emergency department[1]

Documents