The use of propofol for procedural sedation in emergency departments

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Analysis 1.1

Comparison 1 Adverse effects, Outcome 1 Desaturation.

Analysis 1.2

Comparison 1 Adverse effects, Outcome 2 Recovery agitation.

Analysis 1.3

Comparison 1 Adverse effects, Outcome 3 Pain with injection.

Analysis 1.4

Comparison 1 Adverse effects, Outcome 4 Oversedation.

Analysis 1.5

Comparison 1 Adverse effects, Outcome 5 Agitation.

Analysis 1.6

Comparison 1 Adverse effects, Outcome 6 Post‐discharge nausea/vomiting.

Analysis 1.7

Comparison 1 Adverse effects, Outcome 7 Post‐discharge persistent sedation.

Analysis 1.8

Comparison 1 Adverse effects, Outcome 8 Post‐discharge fever.

Analysis 1.9

Comparison 1 Adverse effects, Outcome 9 Post‐discharge recall.

Analysis 1.10

Comparison 1 Adverse effects, Outcome 10 Agitation.

Analysis 1.11

Comparison 1 Adverse effects, Outcome 11 Laryngospasm.

Analysis 1.12

Comparison 1 Adverse effects, Outcome 12 Moaning.

Analysis 1.13

Comparison 1 Adverse effects, Outcome 13 Partial airway obstruction.

Analysis 1.14

Comparison 1 Adverse effects, Outcome 14 Vomiting.

Analysis 1.15

Comparison 1 Adverse effects, Outcome 15 Apnoea.


Study	Propofol (n=109)	Etomidate (n=105)
Propofol vs. etomidate
Miner 2007	10.3 mm (95% CI 7.0 to 13.6)	9.8 mm (95% CI 6.1 to 13.6)

Analysis 2.1

Comparison 2 Participant satisfaction, Outcome 1 Participant satisfaction using a visual analogue scale.


Study	Propofol (n=50)	Ketamine (n=47)
Propofol vs. ketamine
Miner 2010	100% of patients reporting satisfaction with the procedure	100% of patients reporting satisfaction with the procedure

Analysis 2.2

Comparison 2 Participant satisfaction, Outcome 2 Participant satisfaction by asking if satisfied with treatment received.


Study	Propofol (n=11)	Midazolam (n=12)
Propofol vs. midazolam (aged < 65 years)
Parlak 2006	All 11 patients satisfied with procedure	All 12 patients satisfied with procedure
Propofol vs. midazolam (aged ≥ 65 years)
Parlak 2006	20 patients satisfied with procedure; 2 not sure	20 patients satisfied with procedure; 2 not sure

Analysis 2.3

Comparison 2 Participant satisfaction, Outcome 3 Participant satisfaction by using a Likert‐type questionnaire.


Study	Propofol (n=9)	Etomidate (n=9)	Midazolam (n=8)	Midazolam with flumazenil (n=6)
Propofol vs. etomidate vs. midazolam (with or without flumazenil)
Coll‐Vinent 2003	7 patients were "very satisfied"; 2 patients were "satisfied"	7 patients were "very satisfied"; 2 patients were "satisfied"	4 patients were "very satisfied"; 4 patients were "satisfied"	2 patients were "very satisfied"; 4 patients were "satisfied"

Analysis 2.4

Comparison 2 Participant satisfaction, Outcome 4 Participant satisfaction using an ordinal scale.

Summary of findings for the main comparison. Intravenous propofol compared with alternative intravenous sedative or hypnotic for emergency department procedural sedation

Intravenous propofol compared with alternative intravenous sedative or hypnotic for emergency department procedural sedation
Patient or population: emergency department procedural sedation Settings: emergency departments Intervention: intravenous propofol Comparison: alternative intravenous sedative or hypnotic
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Alternative intravenous sedative or hypnotic	Intravenous propofol
Adverse effects (as defined by study authors)	Study population		Not estimable	527 (7 RCTs)	⊕⊝⊝⊝ Very low ¹	Clinical heterogeneity prevented a summary statistic
	0 per 1000	0 per 1000 (0 to 0)
Participant satisfaction (as defined by study authors)	Study population		Not estimable	413 (4 RCTs)	⊕⊝⊝⊝ Very low ²	Clinical heterogeneity prevented a summary statistic
	0 per 1000	0 per 1000 (0 to 0)
Pain with injection	Study population		Not estimable	193 (3 RCTs)	⊕⊝⊝⊝ Very low ³	Clinical heterogeneity prevented a summary statistic
	0 per 1000	0 per 1000 (0 to 0)
The basis for the assumed risk* (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomized controlled trial.
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.
¹The 7 included studies all employed different comparator interventions (Coll‐Vinent 2003; Dunn 2011; Godambe 2003; Havel 1999; Miner 2010; Parlak 2006; Taylor 2005). The quality of evidence was rated down 3 levels for risk of bias because of very serious concerns about inadequate blinding. Coll‐Vinent 2003 reported that the physician responsible for observing time intervals and recovery time was not blinded to the agent used. Dunn 2011 employed no blinding. Godambe 2003 reported blinding of participants, parents and reviewers of the recorded procedure. Havel 1999 reported that blinding during sedation was achieved by shielding medications, infusion tubing and intravenous site from everyone but the study investigator. Miner 2010 reported that neither participants nor staff were blinded to the study drug being administered. Parlak 2006 reported that only the researcher collecting data was blinded to the study drug. Taylor 2005 reported that the doctor performing the procedure was blinded but not the sedation doctor. The quality of evidence was rated down 3 levels for imprecision because the reported CIs around the estimates of treatment effect were very wide. The quality of evidence was rated down 3 levels for inconsistency due to significant clinical heterogeneity in the studies reporting this outcome measure. The quality of evidence was rated down 1 level for indirectness because in 1 study the setting included the coronary care unit (Parlak 2006). The quality of evidence was rated down 3 levels for publication bias because the available evidence comes from small studies (6 of the 7 studies reporting this outcome measure employed fewer than 100 participants). ²The 4 included studies all employed different comparator interventions (Coll‐Vinent 2003; Miner 2007; Miner 2010; Parlak 2006). The quality of evidence was rated down three levels for risk of bias because of very serious concerns about inadequate blinding. Coll‐Vinent 2003 reported that the physician responsible for observing time intervals and recovery time was not blinded to the agent used. Miner 2010 and Miner 2007 reported that neither participants nor staff were blinded to the study drug being administered. Parlak 2006 reported that only the researcher collecting data was blinded to the study drug. The quality of evidence was rated down 3 levels for imprecision because the only reported CI around the estimates of treatment effect was very wide (Miner 2007). The quality of evidence was rated down 3 levels for inconsistency due to significant clinical heterogeneity in the studies reporting this outcome measure. The quality of evidence was rated down 1 level for indirectness because in 1 study the setting included the coronary care unit (Parlak 2006). The quality of evidence was rated down 3 levels for publication bias because the available evidence comes from small studies (3 of the 4 studies reporting this outcome measure employed fewer than 100 participants). ³The 3 included studies all employed different comparator interventions (Coll‐Vinent 2003; Havel 1999; Taylor 2005). The quality of evidence was rated down 3 levels for risk of bias because of very serious concerns about inadequate blinding. Coll‐Vinent 2003 reported that the physician responsible for observing time intervals and recovery time was not blinded to the agent used. Havel 1999 reported that blinding during sedation was achieved by shielding medications, infusion tubing and intravenous site from everyone but the study investigator. Taylor 2005 reported that the doctor performing the procedure was blinded but not the sedation doctor. The quality of evidence was rated down 3 levels for imprecision because the reported CIs around the estimates of treatment effect were very wide (Havel 1999; Taylor 2005). The quality of evidence was rated down 3 levels for inconsistency due to significant clinical heterogeneity in the studies reporting this outcome measure. The quality of evidence was not rated down for indirectness because all the studies reporting this outcome measure, which is important to participants, where applied to the emergency department participant population. The quality of evidence was rated down 3 levels for publication bias because the available evidence comes from small studies (all the studies reporting this outcome measure employed fewer than 100 participants).

Summary of findings for the main comparison. Intravenous propofol compared with alternative intravenous sedative or hypnotic for emergency department procedural sedation

Table 1. Comparator interventions of included studies

Study	Total number of participants randomized	Intervention 1	Intervention 2	Intervention 3
Ab‐Rahman 2010	40	Propofol/fentanyl	Midazolam/fentanyl	None
Coll‐Vinent 2003	32	Propofol	Etomidate	Midazolam
Dunn 2011	40	Propofol/remifentanyl	Midazolam/fentanyl	None
Godambe 2003	113	Propofol/fentanyl	Ketamine/midazolam	None
Havel 1999	89	Propofol	Midazolam	None
Holger 2005	32	Propofol	Midazolam	None
Miner 2007	214	Propofol	Etomidate	None
Miner 2010	97	Propofol	Ketamine	None
Parlak 2006	70	Propofol/fentanyl	Midazolam/fentanyl	None
Taylor 2005	86	Propofol	Midazolam/fentanyl	None

Table 1. Comparator interventions of included studies

Table 2. Adverse effects reported

Study	Adverse effects reported
Coll‐Vinent 2003	Myoclonus, bronchospasm, pain at injection site, re‐sedation
Dunn 2011	Oversedation
Godambe 2003	Agitation, emesis, laryngospasm, apnoea, delayed adverse reactions (nightmares and behavioural change)
Havel 1999	Pain with injection, oversedation, post‐discharge complications (nausea/vomiting, persistent sedation, fever and recall)
Miner 2010	Recovery agitation
Parlak 2006	Desaturation, apnoea
Taylor 2005	Moaning, partial airway obstruction, pain at intravenous site, vomiting

Table 2. Adverse effects reported

Table 3. Methods used to assess participant satisfaction

Study	Method used to assess participant satisfaction
Coll‐Vinent 2003	Ordinal scale (not satisfied, moderately satisfied, satisfied, very satisfied)
Miner 2007	100‐mm satisfaction visual analogue scale consisting of the question, How satisfied are you with the treatment you received during this procedure? With the words 'completely satisfied' and 'not satisfied at all' on either side of the 100‐mm line
Miner 2010	Quote: "after the patients returned to their baseline mental status, they were asked if they felt any pain during the procedure or were able to recall any of the procedure (yes/no). They were also asked if they were satisfied with the treatment they received during the procedure"
Parlak 2006	Quote: "patient satisfaction subsequently was evaluated with a questionnaire including Likert‐type questions"

Table 3. Methods used to assess participant satisfaction

Table 4. Propofol versus ketamine: secondary outcome measures

Secondary outcome measure	Odds ratio (95% confidence interval)
Procedural recall	0.93 (0.28 to 3.1)
Incidence of hypoxia	1.11 (0.34 to 3.59)
Incidence of hypotension	0.94 (0.13 to 6.94)

Table 4. Propofol versus ketamine: secondary outcome measures

Table 5. Propofol versus etomidate: secondary outcome measures

Secondary outcome measure	Summary statistic (95% confidence interval)
Procedural recall	SMD ‐0.24 (‐0.51 to 0.03)
BIS nadir	MD 1.6 (‐4.1 to 6.2)
Incidence of hypoxia	OR 0.96 (0.38 to 2.41)
Need for ventilation	OR 1.21 (0.32 to 4.65)
Decrease in SBP from baseline	MD ‐4.1 (‐6.4% to 1.7%)
MD: mean difference; OR: odds ratio; SMD: standardized mean difference.

Table 5. Propofol versus etomidate: secondary outcome measures

Table 6. Propofol/fentanyl versus ketamine/midazolam: secondary outcome measures

Secondary outcome measure	P value
Physician satisfaction score	0.245
Recall	1.0
Hypoxia	0.002
Hypotension	1.0

Table 6. Propofol/fentanyl versus ketamine/midazolam: secondary outcome measures

Table 7. Propofol versus midazolam/fentanyl

Secondary outcome measure	Mean difference (95% confidence interval)	P value
Time to first awakening	4.6 (0.7 to 8.6)	0.097
Recall	6.3% (‐6.1% to 18.6%)	0.309
Hypoxia	3.1% (‐9.9% to 16%)	0.69
Hypotension	2.6% (‐4.8% to 10.1%)	0.442

Table 7. Propofol versus midazolam/fentanyl

Comparison 1. Adverse effects

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Desaturation Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

1.1 Propofol vs. midazolam (aged < 65 years)	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
1.2 Propofol vs. midazolam (aged ≥ 65 years)	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
2 Recovery agitation Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

3 Pain with injection Show forest plot	3		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

3.1 Propofol vs. midazolam	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
3.2 Propofol vs. midazolam/fentanyl	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
3.3 Propofol vs. etomidate	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
4 Oversedation Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

5 Agitation Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

6 Post‐discharge nausea/vomiting Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

7 Post‐discharge persistent sedation Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

8 Post‐discharge fever Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

9 Post‐discharge recall Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

10 Agitation Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

11 Laryngospasm Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

12 Moaning Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

13 Partial airway obstruction Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

14 Vomiting Show forest plot	2		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

14.1 Propofol/fentanyl vs. ketamine/midazolam	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
14.2 Propofol vs. midazolam/fentanyl	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
15 Apnoea Show forest plot	1		Odds Ratio (M‐H, Fixed, 95% CI)	Totals not selected

15.1 Propofol vs. midazolam (aged < 65 years)	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
15.2 Propofol vs. midazolam (aged ≥ 65 years)	1		Odds Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]

Comparison 1. Adverse effects

Comparison 2. Participant satisfaction

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Participant satisfaction using a visual analogue scale Show forest plot			Other data	No numeric data

1.1 Propofol vs. etomidate			Other data	No numeric data
2 Participant satisfaction by asking if satisfied with treatment received Show forest plot			Other data	No numeric data

2.1 Propofol vs. ketamine			Other data	No numeric data
3 Participant satisfaction by using a Likert‐type questionnaire Show forest plot			Other data	No numeric data

3.1 Propofol vs. midazolam (aged < 65 years)			Other data	No numeric data
3.2 Propofol vs. midazolam (aged ≥ 65 years)			Other data	No numeric data
4 Participant satisfaction using an ordinal scale Show forest plot			Other data	No numeric data

4.1 Propofol vs. etomidate vs. midazolam (with or without flumazenil)			Other data	No numeric data

Comparison 2. Participant satisfaction