The use of propofol for procedural sedation in emergency departments
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To identify and evaluate all randomized controlled trials comparing propofol to alternative drugs (benzodiazepines, barbiturates, etomidate, and ketamine) used in the ED setting for PS.
Background
Many patients presenting to emergency departments are in an anxious state because they are in distressing pain (for example, due to a joint dislocation). Their anxiety is further heightened by some of the painful procedures required for the management of the underlying clinical condition (for example, reduction of a dislocated joint). Procedural sedation (PS) may be required for sedation, hypnosis and relaxation for painful procedures. It may also be required to provide adequate operating conditions by minimizing movement or by inducing amnesia for unpleasant procedures (for example, wound closure by suturing in children). When analgesia cannot be guaranteed in adults, PS may also be required.
Accordingly, the management of sedation and analgesia is an important component of comprehensive emergency medical care (Godwin 2005). Patients undergoing painful procedures in the emergency department (ED) may, therefore, require a potent sedative in addition to narcotic analgesia to proactively address pain and anxiety. This proactive strategy may improve quality of care and patient satisfaction by facilitating interventional procedures and minimizing patient suffering (Godwin 2005).
Since its introduction in 1977, propofol (2,6‐diisopropylphenol) has gained popularity with anaesthetists for sedation in the operating room, and for PS in many settings (DOH 2002; Frazee 2005; Ruiz 2000). It is an intravenous hypnotic agent commonly used in general anaesthesia for the induction and maintenance of anaesthesia and sedation. Propofol has many properties that make it an attractive agent for PS in the ED setting (Burton 2006; Frazee 2005; Symington 2006). Propofol has a rapid onset of action, its clinical effect is essentially immediate after administration ('one arm‐brain circulatory time'), and it produces hypnosis usually within 40 seconds from the time of injection. Peak effect occurs at 92 seconds (Diprivan 2002; Reves 2000). It has an ultrashort half‐life (distribution t½ 2 to 4 minutes) with extremely short recovery times after sedation, typically between 5 to 15 minutes (Reves 2000). Its marked potency reliably produces effective procedural sedation and analgesia, even for very painful procedures (Green 2003). Propofol use is rarely associated with emesis (Green 2003). The disadvantages of propofol include respiratory and haemodynamic depression, a narrow therapeutic window, lack of a direct analgesic effect, and lack of a reversal agent (Burton 2006; Diprivan 2002; Reves 2000).
There is increasing evidence that propofol may be an appropriate agent for use as part of ED procedural sedation (Burton 2006; Green 2003; Symington 2006). In the United States of America, the Joint Commission on Accreditation of Health Care Organizations (JCHAO) permits the use of propofol by emergency physicians, depending on the policy of their individual hospitals (Bahn 2005; Green 2003; JCHAO 2005). However, the JCHAO emphasises that the person administering propofol for sedation must be qualified to manage the patient at whatever level of sedation or anaesthesia is achieved, either intentionally or unintentionally (JCHAO 2011). In addition, the JCHAO emphasises that the person administering the sedation must be qualified to monitor the patient, even if there is need for additional monitoring personnel (JCHAO 2011). The aim of this systematic review is to provide an objective evidence base to inform future clinical practice guidelines regarding the efficacy and safety profile of propofol when used in the ED setting for procedural sedation.
Objectives
To identify and evaluate all randomized controlled trials comparing propofol to alternative drugs (benzodiazepines, barbiturates, etomidate, and ketamine) used in the ED setting for PS.
Methods
Criteria for considering studies for this review
Types of studies
We will include randomized controlled trials (RCTs) in all languages.
We define a randomized controlled trial as a study in which patients are allocated to treatment groups on the basis of a random or quasi‐random method (for example, using random number tables, hospital number, date of birth). We will exclude studies with a cross‐over design.
Types of participants
We will include patients of all ages undergoing procedural sedation by ED staff in the ED setting.
Types of interventions
The target intervention will be administration of intravenous propofol, with or without the use of adjunctive narcotic or non‐narcotic analgesic agents, compared to another intravenous sedative or hypnotic also administered with or without the use of adjunctive narcotic or non‐narcotic analgesic agents to provide procedural sedation in the ED setting.
Inclusion criteria
Any study in which propofol with or without the use of adjunctive narcotic or non‐narcotic analgesic agents is compared against another sedative or hypnotic with or without the use of adjunctive narcotic or non‐narcotic analgesic agents for PS in the ED setting.
Exclusion criteria
Any study in which two or more sedatives are used in either study arm.
Types of outcome measures
Primary outcomes
1. Adverse effects (as defined by the study authors)
2. Patient satisfaction (as defined by the study authors).
Secondary outcomes
-
Physician satisfaction (as defined by the study authors).
-
Awakening time (as defined by the study authors).
-
Procedural recall by patient.
-
Bispectral index (BIS) score during PS.
-
Incidence of hypoxia.
-
Need for ventilation.
-
Incidence of hypotension.
-
Mortality.
-
Cost.
Search methods for identification of studies
Electronic searches
We will perform a computer‐assisted search of the latest issue of the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library) (Appendix 1) for randomized controlled trials of propofol (Diprivan; Fresofol; Pofol; Propofol; Recofol) in the ED setting. The terms for the search strategy are outlined in the appendices. We will add other terms, as appropriate.
We will adapt this strategy for the databases MEDLINE (WebSPIRS Ovid SP) (1966 to present) (Appendix 2), and EMBASE (Ovid SP) (1980 to present) (Appendix 3) using appropriate RCT filters for MEDLINE (Higgins 2006) and EMBASE (Ovid SP) (Appendix 3) (Lefebvre 1996). Additionally, we will search EBSCO CINAHL (1982 to present) (Appendix 4) and ISI Web of Science (Appendix 5).
Searching other resources
We will also search the reference lists of review articles, relevant trials, textbooks, and abstracts of scientific meetings to identify further randomized controlled trials. We will review the titles and abstracts to identify all potential randomized controlled trials. We will obtain the full text versions of these articles. We will make additional efforts to identify potential randomized controlled trials relevant to the topic from the following data sources:
-
foreign language literature;
-
grey literature (theses, internal reports, non‐peer reviewed journals);
-
references (and references of references) cited in primary sources;
-
other unpublished sources known to experts in the speciality (to be sought by personal communication);
-
raw data from published trials (to be sought by personal communication);
-
contacting pharmaceutical companies.
We will not impose a language restriction.
Data collection and analysis
Selection of studies
Two authors (Abel Wakai [AW] and Ronan O'Sullivan [ROS]) will independently decide on the inclusion of studies, having read the methods section of each study and applied the stated criteria. We will resolve differences by consensus.
Data extraction and management
Two authors (AW and ROS) will independently extract data using a standardized data collection form that includes information regarding the name of the first author, year of publication, study design, study population and study setting. In addition to information pertaining to patient characteristics, study inclusion and exclusion criteria, details of the interventions compared and study outcomes, we will extract information regarding study methodology. This will include the method of randomization, allocation concealment, frequency and handling of withdrawals and adherence to the intention‐to‐treat principle. We will attempt to contact the trialists to obtain missing data or to clarify study design features, where necessary. We will resolve disagreements through discussion and in consultation with a third review author (Fergal Cummins [FC]) as required. We will not be blinded to the names of the study authors, investigators, institutions, nor the results.
Assessment of risk of bias in included studies
Two authors (AW and ROS) will independently assess and rate the methodological quality of each trial using the Cochrane Collaboration tool for assessing risk of bias (Higgins 2011). We will judge the quality of the studies by evaluating the studies for the six domains found in Appendix 6. The six domains are as follows:
-
Random sequence generation.
-
Allocation concealment.
-
Blinding of participants and personnel.
-
Blinding of outcome assessment.
-
Blinding of outcome assessment.
-
Incomplete outcome data.
We will evaluate each study and assess separately for these domains. We will judge each explicitly as follows.
-
Low risk of bias.
-
High risk of bias.
-
Unclear risk (lack of information or uncertainty over the potential for bias).
We will enter the data on what was reported to have happened in the study in the 'risk of bias' table in Review Manager 5 (RevMan 5.1). We will contact trialists to obtain further information if clarity is required regarding an included study. We will present a summary figure of the 'risk of bias in included studies' in the review. This will provide a context for discussing the reliability of the results of this review. We will resolve any disagreement by referring to a third author (FC) to reach a consensus.
Measures of treatment effect
We will calculate summary estimates of treatment effect with 95% confidence intervals [CI] for each comparison. For continuous data, we will use the mean differences (MD) whenever outcomes are measured in a standard way across studies. This has the advantage of summarizing results in natural units that are easily understood. When it is desirable to summarize results across studies with outcomes that are conceptually the same but measured in different ways (for example, different pain scores), standardized mean differences will be used. For dichotomous (or binary) data, we will describe results both as a relative measure (relative risk) and an absolute measure (risk difference). Relative measures (such as relative risks) can be used to combine studies but absolute measures (such as the risk difference) are particularly useful when considering trade‐offs between likely benefits and likely harms of an intervention (Deeks 2008).
Dealing with missing data
No simple solution exists for the problem of missing data. We will handle this problem by contacting the investigators, whenever possible, to ensure that no data are missing for their study. In addition, we will make explicit the assumptions of whatever method is used to cope with missing data.
Assessment of heterogeneity
We will evaluate clinical heterogeneity (differences between studies in key characteristics of the participants, interventions, or outcome measures). In the absence of clinical heterogeneity, we will use the I2 statistic to describe the percentage of total variation across studies that is due to heterogeneity rather than chance (Higgins 2003). An I2 > 50% will be considered as significant statistical heterogeneity. We will also use visual inspection of the graphic representation of studies with their 95% confidence intervals (CI) to assess heterogeneity. We will generate tables and graphs using the analysis module included in RevMan 5.1. We will represent pooled odds ratios pictorially as a 'forest plot' to permit visual examination of the degree of heterogeneity between studies.
Assessment of reporting biases
Detecting publication bias is difficult; it is better to avoid it (Glasziou 2001). We will avoid publication bias by comprehensive literature searching and use of study registries (Glasziou 2001). We will use a graphical display (funnel plot) of the size of the treatment effect against the precision of the trial (1/standard error) to investigate publication bias, by examining for signs of asymmetry. Publication bias is associated with asymmetry (Light 1984). If there is asymmetry, reasons other than publication bias will also be sought; for example, poor methodological quality of smaller studies, true heterogeneity, artefact or chance (Egger 1997).
Data synthesis
The results will concentrate on the objectives and comparisons specified in the protocol of the review. Post hoc analyses will be identified as such. We will analyse results using both fixed‐effect and random‐effects models, because for each model there are situations where the result is counterintuitive. If there is significant statistical heterogeneity (I2 > 50%) and the differences in the results are of practical importance, more emphasis will be given to the random‐effects model. The random‐effects model takes into account between study variability as well as within study variability. We will also use a fixed‐effect model to test the robustness of the analysis and for outliers. We will consider the appropriateness of meta‐analysis in the presence of significant clinical or statistical heterogeneity. We will perform meta‐analyses using RevMan software (RevMan 5.1).
Subgroup analysis and investigation of heterogeneity
We will investigate heterogeneity by performing three subgroup analyses based on intuitive reasons. Firstly, we would perform subset analyses on the elderly (patients aged 65 and older) because these patients have relatively limited cardiorespiratory reserve with an increased likelihood to suffer adverse effects during PS. Secondly, we will carry out subgroup analyses on single sedationist or operator versus separate operator and sedationist. Thirdly, we will perform a subgroup analyses of the use of propofol in trauma versus non‐trauma emergencies.
Sensitivity analysis
Finally, we will perform sensitivity analyses to test how sensitive the results are to reasonable changes in the assumptions that are made and in the methods for combining the data (Lau 1998). We will perform sensitivity analysis regarding randomized versus quasi‐randomized studies and eventually good quality studies versus poor quality studies.
Summary of findings
We will use the principles of the GRADE system (Guyatt 2008) to assess the quality of the body of evidence in our review associated with the primary outcome measure of this review and construct a 'Summary of findings' (SoF) table using the GRADE software. The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The quality of a body of evidence considers within study risk of bias (methodologic quality), the directness of the evidence, heterogeneity of the data, precision of effect estimates and risk of publication bias.
