Ketamine-Propofol Versus Propofol Alone for Procedural Sedation in the Emergency Department: A Systematic Review and Meta-analysis

PROGRESSIVE CLINICAL PRACTICE Ketamine-Propofol Versus Propofol Alone for Procedural Sedation in the Emergency Department: A Systematic Review and Meta-analysis Justin W. Yan, MD, MSc, FRCPC, Shelley L. McLeod, MSc, and Alla Iansavitchene, MLIS Abstract Objectives: Propofol is an agent commonly used for procedural sedation and analgesia (PSA) in the emergency department (ED), but it can cause respiratory depression and hypotension. The combination of ketamine-propofol (K-P) is an alternative that theoretically provides a reduction in adverse events compared to propofol. The primary objective of this review was to determine if K-P has a lower frequency of adverse respiratory events in patients undergoing PSA in the ED than propofol alone. Secondary objectives were to compare the proportion of overall adverse events, sedation time, procedure time, and recovery time between K-P and propofol. Methods: Electronic searches of Medline, EMBASE, Cochrane Central Register of Controlled Trials, and CINAHL were conducted and reference lists were hand-searched. Randomized controlled trials (RCTs) published in English comparing the use of K-P to propofol alone for PSA in the ED were included. Results: Six RCTs were included with a combined total of 932 patients (K-P = 520, propofol = 412). Five RCTs reported the proportion of adverse respiratory events; the pooled estimate revealed fewer adverse respiratory events with K-P compared to propofol (29.0% vs. 35.4%; risk ratio [RR] = 0.82; 95% confidence interval [CI] = 0.68 to 0.99). There was no significant difference with respect to the proportion of overall adverse events (38.8% vs. 42.5%; RR = 0.88; 95% CI = 0.75 to 1.04). Procedure time was similar when the groups were compared. Conclusions: The premise of combining ketamine with propofol is based on the many synergies that theoretically exist between these two agents. In this study, K-P had a lower frequency of adverse respiratory events in patients undergoing PSA in the ED compared to propofol alone. ACADEMIC EMERGENCY MEDICINE 2015;22:1003 1013 2015 by the Society for Academic Emergency Medicine INTRODUCTION Patients often require procedural sedation and analgesia (PSA) while undergoing painful procedures in the emergency department (ED), such as fracture reduction, abscess drainage, and electrical cardioversion. Propofol is a sedative/hypnotic that is frequently used for procedural sedation and offers several advantages compared to other pharmacologic agents, including ease of titration, rapid onset of action, and brief duration of effect. However, propofol may cause dose-dependent adverse events such as respiratory depression and hemodynamic compromise. 1 Although propofol has amnestic and antiemetic properties, it is not an analgesic, so opioids and other agents are often used to provide pain control. 2 It has been suggested that the combination of ketamine and propofol (K-P), or ketofol if the two agents From the Division of Emergency Medicine, Department of Medicine, Schulich School of Medicine and Dentistry, The University of Western Ontario (JWY, SLM), London, Ontario; and the London Health Sciences Centre (JWY, SML, AI), London, Ontario, Canada. Received February 2, 2015; revisions received March 11 and April 21, 2015; accepted April 22, 2015. Presented at the Society for Academic Emergency Medicine Annual Meeting, Dallas, TX, May 2014; and the Canadian Association of Emergency Physicians Annual Conference, Ottawa, Ontario, June 2014. The authors have no relevant financial information or potential conflicts to disclose. Supervising Editor: Alan E. Jones, MD. Address for correspondence and reprints: Justin W. Yan, MD; e-mail: jyan2009@meds.uwo.ca. 2015 by the Society for Academic Emergency Medicine ISSN 1069-6563 1003 doi: 10.1111/acem.12737 PII ISSN 1069-6563583 1003

1004 Yan et al. KETAMINE-PROPOFOL VS. PROPOFOL FOR PROCEDURAL SEDATION IN THE ED are mixed in a single syringe, for procedural sedation allows for a reduction in the required dose of each individual agent, potentially mitigating the risk of adverse respiratory events compared to propofol alone or in combination with opioids. 3 Furthermore, adding ketamine to propofol may reduce the potential for hemodynamic instability while providing deep sedation and analgesia to patients. 4,5 However, ketamine may precipitate nausea or vomiting, while propofol has intrinsic antiemetic properties. 6 Postprocedural agitation is a common side effect of using ketamine, while propofol is known to be an anxiolytic. 7 11 Previous systematic reviews have attempted to summarize the clinical efficacy and frequency of adverse events when K-P is used for procedural sedation compared to propofol. In a systematic review by Loh and Dalen, 12 there was insufficient evidence to recommend the routine use of combination therapy with K-P for procedural sedation. The authors suggested that further trials of the safety and efficacy of K-P for sedation in ED procedures are needed. Similarly, Slavik and Zed 13 concluded in their systematic review that combination therapy did not demonstrate superior clinical efficacy compared with propofol alone for PSA. Conflicting data exist regarding the frequency of hemodynamic and respiratory complications in patients receiving K-P compared with propofol monotherapy. The authors suggested that at higher doses, the addition of ketamine to propofol might actually increase adverse events. However, these two systematic reviews included very few studies that took place specifically within the ED setting, where procedural sedation commonly occurs. The primary objective of this systematic review was to determine if K-P has a lower frequency of adverse respiratory events in patients undergoing procedural sedation in the ED compared to propofol alone. Secondary objectives were to compare the proportion of overall adverse events, sedation time, procedure time, and recovery time between K-P and propofol. METHODS Literature Search Strategy The systematic literature searches were conducted in MEDLINE (1946 to September 2014), Embase Classic plus Embase (1947 to Week 36, 2014), CINAHL (1982 to September 2014), and the Cochrane Central Register of Controlled Trials (September 2014) by a research librarian with formal training in electronic literature searching, in consultation with the review authors. A sensitive search strategy (Data Supplement S1, available as supporting information in the online version of this paper) included a combination of subject headings and free-text words using various spelling and endings, such as, but not limited to, the following terms: hypnotics and sedatives, ketamine, esketamin, ketanest, ketalar, ketaset, propofol, diprivan, ketamine-propofol, ketofol, emergency service, hospital, emergency medical services, and emergency. Study Setting and Population Studies involving ED patients undergoing procedural sedation for any nonelective painful procedures were eligible for inclusion. Painful procedures included orthopedic manipulation (e.g., reduction of a fracture or dislocation), electrical cardioversion, abscess drainage, burn-dressing changes, wound debridement, suturing of lacerations, or foreign body removal. Trials that involved procedures in the operating room or clinic setting and those evaluating procedural sedation performed by nonemergency physicians were excluded. There was no age restriction. Studies were considered eligible if the intervention in the trial was procedural sedation with intravenous (IV) administration of K-P or ketofol (if given in a single syringe), and the comparator was IV administration of propofol alone. Studies that used different sedative or analgesic agents (e.g., benzodiazepines, opioids) alone or in combination with propofol were excluded from this review. The searches were restricted to studies published in the English language. An optimized hedges filter and keywords were used to refine search results to randomized controlled trials (RCTs) and systematic reviews published on the topic. The search strategies were modified for each particular database using prespecified terms, search filters, and fields. Reference lists of retrieved studies were hand-searched for relevant citations, and the regulatory website ClinicalTrials.gov was searched to identify any unpublished trials. The authors independently screened the search output to identify potentially eligible trials, the full texts of which were retrieved and assessed for inclusion (Figure 1). Outcome Measures The primary outcome was the occurrence of an adverse respiratory event as defined by each trial. The various definitions of adverse respiratory event that were used in each individual study are presented in Table 1. Secondary outcomes included the occurrence of any adverse event (e.g., nausea or vomiting, rash, agitation or emergence phenomena, and hypotension), sedation time, procedure time, and recovery time. Studies that did not report any of these outcomes were excluded from the analysis. Risk of Bias Assessment Using a standardized data collection form, two reviewers independently extracted data on patient demographics, sample size, type of painful procedure, dose and type of pharmacologic agent used, and all outcome data. Risk of the individual trials was independently assessed using the Cochrane Collaboration s tool for assessing risk as described in section 8.5 of the Cochrane Handbook for Systematic Reviews of Interventions, 14 and discrepancies in quality assessment scores were resolved by discussion. The following domains were assessed as having either low, unclear (uncertain), or high risk : random sequence generation, allocation concealment, blinding of participants/ personnel, blinding of outcomes assessment, incomplete outcome data, and selective outcome reporting. Data Analysis The extent of agreement between reviewers during final study selection was estimated using Cohen s kappa statistic and percent agreement. Individual study results

ACADEMIC EMERGENCY MEDICINE September 2015, Vol. 22, No. 9 www.aemj.org 1005 1,688 citations identified from electronic search (no filters) 437 citations identified from filtered electronic search 1,251 citations excluded after English, human, and RCT filters applied 156 duplicate citations excluded 281 titles, keywords, and abstracts screened 260 citations did not meet eligibility criteria (e. g. non-ed setting, different pharmacologic comparators) 21 potentially relevant studies retrieved in full text for further scrutiny 6 studies included in review 15 studies excluded: - further duplicates (3) - did not meet inclusion criteria due to non-ed setting (11) or assessing outcome of pain on injection only (1) Figure 1. Flow diagram of included studies. RCT = randomized controlled trial. were combined using Review Manager 5.2.4. For the outcomes of adverse respiratory events and overall adverse events, data were pooled using random-effects models to account for both within-study and betweenstudy heterogeneity and reported as risk ratios (RR) with 95% confidence intervals (CIs). RRs were computed such that a value < 1 indicated that K-P was better than treatment with propofol alone for procedural sedation. Statistical significance was defined as p < 0.05 or 95% CI of the RR that excluded unity. Data on sedation, procedure, and recovery times could not be pooled due to the reporting of medians; these outcomes have been reported descriptively. Statistical heterogeneity was assessed using the I 2 statistic. The I 2 describes the percentage of variability in the effect estimates that is due to underlying differences between the studies rather than chance. I 2 values of 50% indicate substantial heterogeneity. To explain possible heterogeneity, a priori subgroup analyses were planned to investigate pediatric versus adult patients, American Society of Anesthesiologists physical classification, 15 type of procedure, and study drug dosage. The Grading of Recommendations, Assessment, Development and Evaluation (GRADE) criteria were used to evaluate the quality of evidence and strength of recommendation by each outcome and presented using the GRADEpro Framework (version 2015). 16,17 RESULTS The search strategy yielded 1,688 potentially relevant citations. After eliminating duplicate citations and studies that did not meet eligibility criteria, 21 full-text articles were retrieved for complete review (Figure 1). Fifteen studies were subsequently excluded, leaving six RCTs included in the review with a combined total of 932 patients: 520 in the K-P group and 412 in the propofol group. 1,18 22 Percent agreement for final selection of included trials was 95.2%, with very good inter-rater agreement (j = 0.88; 95% CI = 0.68 to 1.0). A summary of the characteristics of the included trials can be viewed in Table 2. Two of the studies included only adult patients, 1,18 and two included both adult and pediatric patients (one excluded age < 14 years and one included age 2 years). 19,20 The trial by Del Pizzo et al. 21 included pediatric patients aged 3 to 18 years, and the patient population was unclear in one trial. 22 The majority of procedures performed were orthopedic in nature (>65%), followed by abscess drainage (>20%). Four of the studies were published journal articles, 1,18 20 and two were peer-reviewed abstracts selected for presentation at national and international conferences. 21,22 Two of the included studies were randomized, blinded, three-arm clinical trials of patients undergoing procedural sedation in the ED. 18,21 In the trial by Miner et al., 18 adult patients were randomized to receive 1 mg/kg propofol, 0.5 mg/kg ketamine combined 0.5 mg/kg with propofol (1:1), or 0.8 mg/kg propofol combined with 0.2 mg/kg ketamine (4:1) in an IV bolus, followed by half of the initial dose every 3 minutes as needed for sedation. In the study by David and Shipp, 20 fentanyl at a dose of 0.5 to 1.0 lg/kg was administered 5 minutes prior to sedation in all patients, regardless of whether they received K-P or propofol. In the single-blinded RCT by Del Pizzo et al., 21 children aged 3 to 18 years were randomized into one of three sedation groups: 1) propofol alone (initial dose of 1 mg/

1006 Yan et al. KETAMINE-PROPOFOL VS. PROPOFOL FOR PROCEDURAL SEDATION IN THE ED Table 1 Definitions of Adverse Respiratory Events From the Included Trials Trial Outcome Components of Adverse Respiratory Event Outcome Airway Interventions Andolfatto (2012) 19 Number and proportion of patients experiencing a respiratory adverse event as defined by the Quebec Criteria David (2011) 20 Respiratory depression Miner (2015) 18 Relative number and proportion of subjects experiencing airway or respiratory adverse events leading to an intervention O2 desaturation decrease in O2 saturation resulting in intervention Central apnea cessation of ventilatory effort* resulting in intervention Partial or complete upper airway obstruction stridor or snoring or ventilatory effort with no air exchange, with response to intervention Laryngospasm partial or complete upper airway obstruction, with O 2 desaturation caused by involuntary and sustained closure of the vocal cord not responding to intervention Clinically apparent pulmonary aspiration presence before the end of the recovery phase of new physical symptom, requirement for supplemental O2 to maintain baseline oxygenation, chest radiograph findings of focal infiltrate, consolidation, or atelectasis Requirement for airway intervention ETCO2 increase of greater than 5 mm Hg of 10 or more seconds in duration Respiratory rate less than 8 breaths/min of 10 or more seconds in duration SaO 2 less than 90% of 10 or more seconds in duration Apnea greater than 15 seconds (defined as no visible respiratory effort) Hypoxia O2 saturation of < 92% at any time during the procedure Central apnea absent ETCO2 waveform > 6 seconds during the procedure Subclinical respiratory depression decrease in ETCO2 > 10 mm Hg recorded during the procedure or by physician report of partial upper airway obstruction Complete airway obstruction absent ETCO 2 waveform with physician report of ventilatory effort Laryngospasm partial or complete airway obstruction caused by involuntary closure of the vocal cords not relieved by intervention, by physician report Vigorous tactile stimulation, airway repositioning, suctioning, supplemental O2 delivery, any airway placement, BVM ventilation Need for jaw repositioning or use of BVM Increased supplemental O 2, airway adjunct use (oral airway or nasal trumpet), repositioning, verbal or physical stimulus to induce ventilation, BVM ventilation Aspiration clinically apparent aspiration after vomiting, by physician report Phillips (2010) 1 Presence of respiratory depression O2 saturation < 90% or requirement for airway intervention Sawas (2013) 22 Clinical or subclinical respiratory depression Clinical respiratory depression: requirement for airway intervention Subclinical respiratory depression: the occurrence of any of the following during the sedation: an increase or decrease in ETCO 2 by 10 points from baseline, an ETCO2 greater than 50 mm Hg or a loss of ETCO2 for greater than 15 seconds (apnea) Airway intervention not defined Placement of an airway, airway maneuvers, BVM ventilation, or intubation *No observable respiratory effort and loss of capnographic waveform. Cough, crackles/rales, decreased breath sounds, tachypnea, wheeze, rhonchi, respiratory distress. BVM = bag-valve-mask; ETCO2 = end tidal carbon dioxide; SaO2 = oxygen saturation.

ACADEMIC EMERGENCY MEDICINE September 2015, Vol. 22, No. 9 www.aemj.org 1007 Table 2 Characteristics of Included Trials Trial Procedure Type Initial Dose of K-P Initial Dose of Propofol Outcomes K-P Propofol Andolfatto (2012), 19 Canada David (2011), 20 United States Del Pizzo (2011), 21 United States Miner (2015), 18 United States Phillips (2010), 1 United States Sawas (2013), 22 United States 171 orthopedic, 51 abscess drainage, 38 cardioversion, 24 other 169 orthopedic, 14 suturing, 10 other 57 orthopedic 0.3 mg/kg K, 1 mg/kg P, 0.3 mg/kg K and P combined 117 orthopedic, 146 abscess drainage, 6 chest tube placement; 2 cardioversion 0.375 mg/kg K and P combined 0.75 mg/kg Adverse respiratory events;* sedation consistency, efficacy, and time; induction time; other adverse events 0.5 mg/kg K, 1 mg/kg P 1 mg/kg Adverse respiratory events;* satisfaction; sedation quality; total propofol administered 1:1 0.5 mg/kg K and P combined; 4:1 0.8 mg/kg K, 0.2 mg/kg P 1 mg/kg Sedation time; procedure time; recovery time; ED length of stay 1 mg/kg Adverse airway and respiratory events;* sedation depth; efficacy; procedure and recovery time; patient satisfaction, pain and procedural recall 28 orthopedic 0.75 mg/kg K and P combined 0.5 1.5 mg/kg Adverse respiratory events;* procedural success; BIS score; adverse events; recovery time; vital signs 82 orthopedic, 4 abscess drainage, 7 cardioversion, 5 other, 1 unknown 142 142 97 96 38 19 1:1 = 85 4:1 = 96 90 14 14 Unclear (1:1 ratio) Unclear Adverse respiratory events* 51 48 BIS = bispectral index score; ETCO2 = end-tidal carbon dioxide; K = ketamine; K-P = ketamine-propofol; P = propofol. *Adverse respiratory events defined by Table 1. Abstracts presented at national and international emergency medicine conferences

1008 Yan et al. KETAMINE-PROPOFOL VS. PROPOFOL FOR PROCEDURAL SEDATION IN THE ED Table 3 Risk of Bias Summary for Included Trials Trial Random Sequence Generation Allocation Concealment Blinding of Patients/ Personnel Blinding of Outcome Assessment Lost to Follow-up (%) Free of Selective Outcome Reporting Other Bias Andolfatto (2012) 19 Low Low Low High 0 Low Low David (2011) 20 Low Low Low High 0 Low Low Del Pizzo (2011) 21 Low Unclear Unclear* High 0 High Low Miner (2015) 18 Low Low Low High 0 Low Low Phillips (2010) 1 Low Unclear Unclear* High 0 Low Low Sawas (2013) 22 Low Unclear Unclear* High 0 Unclear Unclear Summary score Low risk Unclear risk Unclear risk High risk Low risk Unclear risk Low risk *Study authors mentioned that these were single-blinded trials, suggesting that participants were blinded but caregivers (physicians and nurses) were not. kg with subsequent doses at the sedating physician s discretion), 2) K-P (one-time dose of ketamine at 0.3 mg/ kg followed by propofol at an initial dose of 1 mg/kg, subsequent boluses of propofol at the sedating physician s discretion), and 3) ketofol (a mixture of ketamine and propofol in the same syringe, 1:1 with an initial bolus of 0.3 mg/kg of each medication, and subsequent boluses at the sedating physician s discretion). The remaining trials were all two-arm studies comparing combination therapy with propofol alone. 1,19,22 Five studies reported the frequency of adverse respiratory events (Table 1). 1,18 20,22 Andolfatto et al. 19 reported this outcome according to the Quebec Criteria, which provide standardized, intervention-based definitions for adverse event reporting in PSA research. Definitions for adverse respiratory events as reported by each included study are outlined in Table 1. Additional adverse events including hypotension, nausea and vomiting, rash, emergence phenomena, and agitation were reported in the same five studies. 1,18 20,22 Four of the study authors were contacted by e-mail or phone for additional information regarding missing data and study methodology (e.g., details on randomization methods, blinding). Two authors provided additional information and clarification. The authors of the studies who reported median values for sedation, procedure times, and recovery times were contacted to retrieve original data so means could be calculated and potentially pooled. However, these outcomes could not be pooled due to insufficient data and have been reported descriptively. Risk of Bias Table 3 shows the data for the risk assessment. Data Synthesis Data on the frequency of adverse respiratory events were available for five included trials, involving a total of 875 patients: 393 in the propofol group and 482 in the K-P group. 1,18 20,22 The pooled estimate showed a statistically significant difference in adverse respiratory events when K-P (29.0%) was compared to propofol alone (35.4%) for patients undergoing procedural sedation in the ED (RR = 0.82; 95% CI = 0.68 to 0.99), with very low statistical heterogeneity (I 2 = 0%) between the studies (Figure 2). Five of the included trials reported overall adverse events (Figure 3). 1,18 20,22 The pooled estimate showed Figure 2. The frequency of adverse respiratory events in patients undergoing procedural sedation in the emergency department with ketamine-propofol compared to propofol alone.

ACADEMIC EMERGENCY MEDICINE September 2015, Vol. 22, No. 9 www.aemj.org 1009 Figure 3. The frequency of overall adverse events in patients undergoing procedural sedation in the ED with ketamine-propofol compared to propofol alone. no significant difference in overall adverse events when K-P (38.8%) was compared to propofol (42.5%; RR = 0.88; 95% CI = 0.75 to 1.04), with low heterogeneity (I 2 = 11%) between the included trials. A summary of findings is presented in Table 4. Although sedation time, procedure time, and recovery time were reported in three of the included studies, these outcomes could not be pooled in our study due to the consistent reporting of medians, not means (Table 5). As reported by Del Pizzo et al., 21 sedation time, procedure time, and recovery time did not differ among children in each group for closed forearm fracture reduction. Likewise, the study by Andolfatto et al. 19 reported no differences in these outcomes. However, the median sedation time reported by Miner et al. 18 was found to be statistically different in the 4:1 K-P group compared to propofol monotherapy (D 2.2 minutes; 95% CI = 0.3 to 4.1 minutes) in their randomized three-arm clinical trial for procedural sedation in the ED. Additionally, Miner et al. reported longer recovery time in the 1:1 K-P group compared to propofol alone (D 4.0 minutes; 95% CI = 2.0 to 6.0 minutes). DISCUSSION The principle of combining ketamine with propofol is based on the many synergies that theoretically exist between these two agents and the benefits this may provide. Findings from our systematic review and metaanalysis suggest the use of K-P may reduce the frequency of adverse respiratory events in patients undergoing PSA in the ED compared to propofol alone. Our results differ from those of previously reported studies on this topic. A systematic review conducted by Slavik and Zed 13 examined the efficacy and safety of K- P for procedural sedation. Eight clinical trials were included in the review and the results did not show clinical benefit when K-P was compared to propofol alone. However, there were considerable limitations to the studies that were selected for inclusion. The authors acknowledged that there were significant sources of methodologic and statistical heterogeneity due to varied patient populations, procedures being performed, and study settings; thus, pooling of data for meta-analysis was not possible. Furthermore, all of the trials included in the review were small; the largest sample size was 66 patients (33 in each group). Finally, all of the included studies were conducted for elective procedures in non- ED settings, limiting the generalizability of the findings for patients in the ED where K-P is used frequently for procedural sedation. Another systematic review by Loh and Dalen 12 attempted to evaluate the efficacy and safety of low-dose ketamine in addition to propofol for procedural sedation. In this review, the authors included 11 trials, most of which had small sample sizes and were conducted in non-ed settings. The combination of ketamine and propofol did not demonstrate any difference with respect to time to discharge, need for fluid or vasopressor administration, supplemental oxygen, or assisted ventilation when compared with propofol alone. However, patients in the K-P group who received higher doses of ketamine reported an increased incidence of nausea, vomiting, and emergence reactions following the procedure. The authors concluded there was insufficient clinical evidence to recommend the routine use of low-dose ketamine with propofol for procedural sedation in the ED setting. Recently, a qualitative review by Black et al. 23 examined the safety and efficacy of propofol for procedural sedation in the ED setting. The authors included 13 RCTs and 20 observational studies and reported no statistically significant difference in the occurrence of respiratory depression with propofol compared to alternative agents. The review concluded that the use of opioids in addition to propofol may not provide added benefit with regard to pain recall and patient satisfaction and may contribute to increased rates of adverse events. Furthermore, a trial by Messenger et al. 24 sought to compare the safety and efficacy of subdissociative-dose ketamine versus fentanyl as adjunct analgesics to propofol for procedural sedation in the ED. The authors enrolled 63 patients aged 14 to 65 years who required procedural sedation for orthopedic reduction or abscess drainage and randomized participants to receive either 0.3 mg/kg ketamine or 1.5 lg/kg fentanyl IV prior to propofol. While it is still common ED

1010 Yan et al. KETAMINE-PROPOFOL VS. PROPOFOL FOR PROCEDURAL SEDATION IN THE ED Table 4 GRADE Evidence Profile for Ketamine-Propofol Compared to Propofol Alone for Procedural Sedation and Analgesia in the ED Quality Assessment No. of Patients Effect GRADE No. of Studies Design Risk of Bias Inconsistency Indirectness Imprecision Ketamine- Propofol Propofol Relative (95% CI) Absolute Quality Importance Adverse Respiratory Events 5 Randomized No serious trials risk Overall Adverse Events 5 Randomized trials No serious risk No serious inconsistency No serious inconsistency No serious indirectness No serious indirectness Serious* 140/482 (29%) 139/393 (35.4%) Serious* 187/482 (38.8%) 167/393 (42.5%) RR = 0.82 (0.68 0.99) 64 fewer per 1000 (from 4 fewer to 113 fewer) 32.2% 58 fewer per 1000 (from 3 fewer to 103 fewer) RR = 0.88 (0.75 1.04) 51 fewer per 1000 (from 106 fewer to 17 more) 43.3% 52 fewer per 1000 (from 108 fewer to 17 more) Ο MODERATE Ο MODERATE IMPORTANT IMPORTANT GRADE Working Group grades of evidence: high quality further research is very unlikely to change our confidence in the estimate of effect; moderate quality further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate; low quality further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate; very low quality we are very uncertain about the estimate. RR = risk ratio. *Some studies included relatively few patients and had few events and thus have wide confidence intervals around the estimate of the effect.

ACADEMIC EMERGENCY MEDICINE September 2015, Vol. 22, No. 9 www.aemj.org 1011 Table 5 Sedation Time, Procedure Time, and Recovery Time as Reported by the Included Trials Trial Andolfatto (2012) 19 Del Pizzo (2011) 21 Miner (2015) 18 Sedation Time Procedure Time Recovery Time K-P P K-P P K-P P 7 min 7 min 4 min 5 min 8 min 6 min IQR = 4 9 IQR = 4 9 IQR = 2 7 IQR = 2 7 IQR = 7 10 IQR 2 8 18 min K+P 20 min 14 min K+P 18 min 9 min K+P 11 min Range = 12 56 Range 10 55 Range = 6 25 Range = 9 41 Range = 2 53 Range = 4 21 19 min K-P 14 min K-P 12 min K-P range = 4 69 Range = 5 33 Range = 1 49 16 min 1:1 K-P 14 min 7 min 1:1 K-P 7 min 10 min 1:1 K-P 6 min IQR = 10 23 IQR = 5 25 IQR = 5 10 IQR = 5 9.5 IQR = 5 17 IQR = 4 10 15 min 4:1 K-P 7 min 4:1 K-P 8 min 4:1 K-P IQR = 11 18.5 IQR = 6 11 IQR = 4 15 IQR = interquartile range; K-P = ketamine-propofol; K+P = ketamine + propofol; P = propofol. practice to administer opioids with propofol when providing sedation, the results of this study suggest that the addition of opioids significantly increases the risk of adverse cardiorespiratory events and necessity for interventions when compared to the addition of ketamine. The authors concluded that subdissociative-dose ketamine was safer than fentanyl for procedural sedation in the ED and appeared to have similar efficacy. However, this trial was terminated prematurely based on the results from interim analyses. Truncated trials have been shown to systematically overestimate treatment effects, especially in cases where the number of accrued outcome events is small. 25 LIMITATIONS As with all systematic reviews and meta-analyses, the results from this study are limited by the quality of trials that were included. Two studies were abstracts that were unlikely to be peer reviewed to the same level of scrutiny as the full-text journal articles. Furthermore, these abstracts were limited with respect to the amount of information that could be included, resulting in high or unclear risks in study methodology, varying or unclear definitions of outcomes, and how they were measured. Although four study authors were contacted for further information, only two replied, with only one providing additional data for review. Although the search strategy used to identify potentially relevant studies was comprehensive, only Englishlanguage articles were included in this review. It is possible that some studies may have been missed if they were published in other languages or conducted in other countries. The primary outcome of the occurrence of an adverse respiratory event was defined somewhat differently in each trial. While most studies considered events such as a change in end-tidal CO 2, drop in oxygen saturation, or airway maneuvers as adverse events, others did not report some of these outcomes. The administration of preprocedural supplemental oxygen was inconsistently reported in the included trials. It is unknown if any of the adverse respiratory events could have been prevented by the administration of oxygen. Additionally, the definitions of sedation and recovery time were varied among the studies. Miner et al. 18 defined sedation time as the time from first study medication until return to baseline mental status and recovery time as the time from procedure completion until return to baseline mental status. Andolfatto et al. 19 defined sedation time as the time from first medication administration to time of procedure completion and recovery time as the interval from the last dose of medication administered until discharge criteria were met. Del Pizzo et al. 21 defined sedation time as the first administration of study drug to first purposeful verbal response and recovery time as the time of last administration of study drug to first purposeful verbal response. Most of the trials used an initial dose of approximately 1 mg/kg of the study drug in the propofol group, but comparators varied with respect to the initial dose, ratios, and administration (i.e., ketamine and propofol combined in one syringe or given separately in two syringes) of the study drug in the K-P group. Furthermore, in the trial by David and Shipp, 20 each patient received a single IV dose of fentanyl 0.5 to 1.0 lg/kg 5 minutes before sedation. Because all enrolled patients received fentanyl, the comparison in this study was K-P versus propofol. However, we cannot rule out a potential interactive effect between fentanyl and the given study drug, even though it was not given as part of the sedation protocol. While our systematic review did not demonstrate significant heterogeneity across the studies included in the analysis, there is clinical heterogeneity warranting further discussion. Trials differed with respect to the ages of patients that were enrolled, and although the majority of patients were undergoing procedural sedation for orthopedic manipulation, a variety of other painful procedures were also included in each study. Some procedures may require deeper sedation or take longer to complete than others. Many of the studies did not use standardized, quantifiable endpoints to define sedation and recovery. The included trials used varied definitions of time to event outcomes, so we were unable to provide conclusive comments on time to adequate sedation, total sedation time, and ED length of stay. Additionally,

1012 Yan et al. KETAMINE-PROPOFOL VS. PROPOFOL FOR PROCEDURAL SEDATION IN THE ED we were unable to comment on patient recall or patient/ physician satisfaction with either regimen, as these outcomes were uncommonly reported. A priori, we planned to perform subgroup analyses to account for the heterogeneity between trials. However, either we were unable to obtain detailed information on the number of patients in each of the clinical subgroups or subgroup sizes were too small to generate appropriate and meaningful analyses. Although the results of our meta-analysis suggest K-P reduces adverse respiratory events compared to propofol alone, we believe that our results should be interpreted with caution. Of the included studies, the only individual positive trial was an abstract, which if excluded on this basis would negate any overall positivity to the pooled estimate. Further research may be necessary to definitively determine if in fact there is a benefit to combination therapy with K-P and whether or not this would include a clinically important reduction in adverse respiratory events. CONCLUSIONS These results suggest the use of combination of ketamine and propofol reduces the frequency of adverse respiratory events, but not overall adverse events, in patients undergoing procedural sedation and analgesia in the ED compared to propofol alone. Additionally, the combination of ketamine and propofol does not appear more dangerous than the use of propofol alone and likely has applications for clinical practice. References 1. Phillips W, Anderson A, Rosengreen M, Johnson J, Halpin J. Propofol versus propofol/ ketamine for brief painful procedures in the emergency department: clinical and bispectral index scale comparison. J Pain Palliat Care Pharmacother 2010;24:349 55. 2. Miner JR, Burton JH. Clinical practice advisory: emergency department procedural sedation with propofol. Ann Emerg Med 2007;50:182 7. 3. Shah A, Mosdossy G, McLeod S, Lehnhardt K, Peddle M, Rieder M. A blinded, randomized controlled trial to evaluate ketamine/propofol versus ketamine alone for procedural sedation in children. Ann Emerg Med 2011;57:425 33. 4. Andolfatto G, Willman EV. A prospective case series of pediatric procedural sedation and analgesia in the emergency department using single-syringe ketamine-propofol combination (ketofol). Acad Emerg Med 2010;17:194 201. 5. Willman EV, Andolfatto G. A prospective evaluation of ketofol (ketamine/propofol combination) for procedural sedation and analgesia in the emergency department. Ann Emerg Med 2007;49:23 30. 6. Borgeat A, Wilder-Smith OH, Saiah M, Rifat K. Subhypnotic doses of propofol possess direct antiemetic properties. Anesth Analg 1992;74:539 41. 7. Smith I, Monk TG, White PF, Ding Y. Propofol infusion during regional anesthesia: sedative, amnestic, and anxiolytic properties. Anesth Analg 1994;79: 313 9. 8. Friedberg BL. Propofol-ketamine technique: dissociative anesthesia for office surgery (a 5-year review of 1264 cases). Aesthetic Plast Surg 1999;23:70 5. 9. Picard P, Tramer MR. Prevention of pain on injection with propofol: a quantitative systematic review. Anesth Analg 2000;90:963 9. 10. Koo SW, Cho SJ, Kim YK, Ham KD, Hwang JH. Small-dose ketamine reduces the pain of propofol injection. Anesth Analg 2006;103:1444 47. 11. Barbi E, Marchetti F, Gerarduzzi T, et al. Pretreatment with intravenous ketamine reduces propofol injection pain. Paediatr Anaesth 2003;13:764 8. 12. Loh G, Dalen D. Low-dose ketamine in addition to propofol for procedural sedation and analgesia in the emergency department. Ann Pharmacother 2007;41:485 92. 13. Slavik VC, Zed PJ. Combination ketamine and propofol for procedural sedation and analgesia. Pharmacotherapy 2007;27:1588 98. 14. Higgins JP, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0. The Cochrane Collaboration (2011). Available from: www.cochrane-handbook.org. 15. Saklad M. Grading of patients for surgical procedures. Anesthesiology 1941;2:281 84. 16. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490 94. 17. Guyatt GH, Gutterman D, Baumann MH, et al. Grading strength of recommendations and quality of evidence in clinical guidelines: report from an American College of Chest Physicians task force. Chest 2006;129:174 81. 18. Miner JR, Moore J, Austad EJ, Plummer D, Hubbard L, Gray RO. Randomized, double-blinded, clinical trial of propofol, 1:1 propofol/ketamine, and 4:1 propofol/ketamine for deep procedural sedation in the emergency department. Ann Emerg Med 2015;65:479 88. 19. Andolfatto G, Abu-Laban RB, Zed PJ, et al. Ketamine-propofol combination (ketofol) versus propofol alone for emergency department procedural sedation and analgesia: a randomized doubleblind trial. Ann Emerg Med 2012;59:504 12. 20. David H, Shipp J. A randomized controlled trial of ketamine/propofol versus propofol alone for emergency department procedural sedation. Ann Emerg Med 2011;57:435 41. 21. Del Pizzo J, Agha BS, Downes K, Mularoni P. Efficiency in sedation for forearm fracture reduction in children: propofol vs. ketamine-propofol vs. ketofol [abstract]. Pediatric Emergency Care Conference: American Academy of Pediatrics, AAP Section on Emergency Medicine National Conference and Exhibition. 2011;27:10. 22. Sawas A, Youngquist ST, Madsen TE, Davis VW. Combined ketamine and propofol sedation vs propofol sedation for emergency department procedures: a prospective randomized trial [abstract]. Acad Emerg Med 2013;62:S1. 23. Black E, Campbell SG, Magee K, Zed PJ. Propofol for procedural sedation in the emergency department: a

ACADEMIC EMERGENCY MEDICINE September 2015, Vol. 22, No. 9 www.aemj.org 1013 qualitative systematic review. Ann Pharmacother 2013;47:856 68. 24. Messenger DW, Murray HE, Dungey PE, van Vlymen J, Sivilotti ML. Subdissociative-dose ketamine versus fentanyl for analgesia during propofol procedural sedation: a randomized clinical trial. Acad Emerg Med 2008;15:877 86. 25. Guyatt GH, Briel M, Glasziou P, Bassler D, Montori VM. Problems of stopping trials early. BMJ 2012;344:e3863. Supporting Information The following supporting information is available in the online version of this paper: Data Supplement S1. Search Strategy -Ovid MED- LINE(R) 1946 to September, 2014.

本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 (www.xuebalib.com) 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具