Pain control in first trimester surgical abortion

Regina‐Maria Renner; Jeffrey TJ Jensen; Mark DN Nichols; Alison Edelman

doi:10.1002/14651858.CD006712.pub2

. 2009 Apr 15;2009(2):CD006712. doi: 10.1002/14651858.CD006712.pub2

Pain control in first trimester surgical abortion

Regina‐Maria Renner ^1,^✉, Jeffrey TJ Jensen ¹, Mark DN Nichols ¹, Alison Edelman ¹

Editor: Cochrane Fertility Regulation Group

PMCID: PMC7390061 PMID: 19370649

Abstract

Background

First trimester abortions especially cervical dilation and suction aspiration are associated with pain, despite various methods of pain control.

Objectives

Compare different methods of pain control during first trimester surgical abortion.

Search methods

We searched multiple electronic databases with the appropriate key words, as well as reference lists of articles, and contacted professionals to seek other trials.

Selection criteria

Randomized controlled trials comparing methods of pain control in first trimester surgical abortion at less than 14 weeks gestational age using electric or manual suction aspiration. Outcomes included intra‐ and postoperative pain, side effects, recovery measures and satisfaction.

Data collection and analysis

Two reviewers independently extracted data. Meta‐analysis results are expressed as weighted mean difference (WMD) or Peto Odds ratio with 95% confidence interval (CI).

Main results

We included forty studies with 5131 participants. Due to heterogeneity we divided studies into 7 groups:

Local anesthesia: Data was insufficient to show a clear benefit of a paracervical block (PCB) compared to no PCB or a PCB with bacteriostatic saline. Pain scores during dilation and aspiration were improved with deep injection (WMD ‐1.64 95% CI ‐3.21 to ‐0.08; WMD 1.00 95% CI 1.09 to 0.91), and with adding a 4% intrauterine lidocaine infusion (WMD ‐2.0 95% CI ‐3.29 to ‐0.71, WMD ‐2.8 95% CI ‐3.95 to ‐1.65 with dilation and aspiration respectively).

PCB with premedication: Ibuprofen and naproxen resulted in small reduction of intra‐ and post‐operative pain.

Analgesia: Diclofenac‐sodium did not reduce pain.

Conscious sedation: The addition of conscious intravenous sedation using diazepam and fentanyl to PCB decreased procedural pain.

General anesthesia (GA): Conscious sedation increased intraoperative but decreased postoperative pain compared to GA (Peto OR 14.77 95% CI 4.91 to 44.38, and Peto OR 7.47 95% CI 2.2 to 25.36 for dilation and aspiration respectively, and WMD 1.00 95% CI 1.77 to 0.23 postoperatively). Inhalation anesthetics are associated with increased blood loss (p<0.001).

GA with premedication: The COX 2 inhibitor etoricoxib, the non‐selective COX inhibitors lornoxicam, diclofenac and ketorolac IM, and the opioid nalbuphine were improved postoperative pain.

Non‐pharmacological intervention: Listening to music decreased procedural pain.

No major complication was observed.

Authors' conclusions

Conscious sedation, GA and some non‐pharmacological interventions decreased procedural and postoperative pain, while being safe and satisfactory to patients. Data on the widely used PCB is inadequate to support its use, and it needs to be further studied to determine any benefit.

Plain language summary

Pain control in first trimester surgical abortion.

Multiple methods of pain control in first trimester surgical abortion at less than 14 weeks gestational age using electric or manual suction aspiration are available, and appear both safe and effective. Pain control methods can be divided in local anesthesia, conscious sedation, general anesthesia and non‐pharmacological methods. Data to support the benefit of the widely used local aneathetic is inadequate. While general anesthesia achieved complete pain control during the procedure, other forms of anesthesia such as conscious sedation with a paracervical block improved postoperative pain control.

Summary of findings

Background

Elective abortions are among the most common outpatient surgical procedures performed on women with an estimated 46 million performed yearly worldwide (WHO 2003). Nearly 90% are performed in the first trimester before 13 weeks gestation (Strauss 2007). A major complication occurs in less than 1 in 100 women and mortality is around 0.7 in 100,000 (Hakim‐Elahi 1990; Bartlett 2004; Koonin 2000). Although the case‐fatality‐rate has decreased since the 1970s, anesthesia‐related events continue to be the leading cause of morbidity (Lawson 1994).

Anesthesia is important for women undergoing an abortion since most will experience pain with the procedure. Key factors that influence the choice of anesthesia or analgesia include effectiveness, safety, side effects, and costs. Other crucial factors include patient preference, practitioner choice or bias, facility resources and medical indications (Maltzer 1999).

Pain perception is a complex phenomenon comprised of both physical and psychosocial elements and their interaction, and varies considerably between women (Stubblefield 1989). The physical pain women experience with abortion most likely originates from the S2 to S4 parasympathetic fibers (the Frankenhäuser plexus) that innervate the cervix and the lower part of the uterine body (Scott 1976; Smith 1991). In addition, the fundus and lower part of uterine body are innervated by sympathetic (Maltzer 1999) fibers from T10 to L1 via the inferior hypogastric nerve, and the ovarian plexus (Maltzer 1999).

Additionally, psychological (affective, motivational, interpretive), and social (context, support) features play into pain perception (Borgotta 1997). Increased pain with abortion has been associated with young age, nulliparity, less education, anxiety, depression, "moral problems" (with the procedure), a retroverted uterus, and dysmenorrhea (Belanger 1989, Glantz 2001). A history of prior vaginal delivery correlates well with decreased pain (Belanger 1989)). Data on the relationship between pain and gestational age, as well as the amount of cervical dilation performed, has been conflicting (Belanger 1989; Borgotta 1997; Smith 1979).

Due to this complex nature, effective management of abortion‐related pain requires a combination of pharmacological and non pharmacological means (Maltzer 1999). Pharmacological methods include local anesthetic, non‐steroidal anti‐inflammatory medications (NSAID) narcotics, anxiolytics, sedatives, and/or hypnotics. Concerns regarding general anesthesia stem from its association with greater costs and personnel and increased morbidity and mortality based on observational data that includes cases until the mid 1980s (Maltzer 1999; Raeder 1992; Grimes 1979; Lawson 1994). Therefore, it is less frequently used in the United States (Lichtenberg 2001).

Non pharmacological aspects of pain have a considerable impact on pain perception (Maltzer 1999). Active participation in one's own pain management, and control over the life situation have been found to be beneficial (Belanger 1989).

Unfortunately, despite these advances, many patients still find surgical abortion extremely uncomfortable; 78‐97% report at least moderate procedural pain (Stubblefield 1989; Belanger 1989; Smith 1979; Rawling 1998). Therefore, optimizing pain control should be a goal in every procedure. Opinions may vary how much pain reduction is clinically relevant. Strategies designed to reduce abortion‐related pain have great public health importance considering the large numbers of women who undergo first trimester surgical abortions. This review will examine the existing randomized controlled trials to compare the effect of different methods of pain control during first trimester surgical abortion on patient perceived pain, satisfaction, side effects, and safety. The review will investigate preemptive as well as intra operative analgesia, focusing on pharmacological methods administered via mucosal (oral, vaginal, intrauterine, buccal/sublingual), intramuscular, or intravenous routes, but also include non‐pharmacological methods.

Objectives

To compare the effect of different methods of pharmacological and non pharmacological pain control administered prior to or during first trimester surgical abortion (< 14 weeks gestation, electric or manual suction aspiration) on patient perceived pain, satisfaction, side effects, and safety.

Methods

Criteria for considering studies for this review

Types of studies

All randomized controlled trials, including placebo controlled in any language.

Types of participants

Pregnant women undergoing first trimester surgical abortion at less than 14 weeks gestational age using electric or manual suction aspiration.

Types of interventions

Any type of pharmacological pain control administered via mucosal (oral, vaginal, intrauterine, buccal/sublingual), intramuscular, or intravenous routes or non‐pharmacological pain control prior to or during a first trimester surgical abortion at less than 14 weeks gestational age using electric or manual suction aspiration.

Types of outcome measures

The main outcome is patient reported effectiveness of pain control on perceived pain during and immediately post abortion using validated scales, e.g. visual analogue, CAT, and Likert scales, categorical or dichotomous assessment (yes versus no).   Additional outcomes are adverse effects, and side effects (including if the method of pain control causes pain), as well as patient satisfaction.

Search methods for identification of studies

See: Cochrane Fertility Regulation Group search strategy

Electronic searches   We searched the Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, and POPLINE for articles for trials of pain control in first trimester surgical abortion at less than 14 weeks gestational age using electric or manual suction aspiration. Electronic literature search of 'Cochrane Central Register of Controlled Trials (4th quarter, 2007), MEDLINE (1950 to January 2008), EMBASE (1974 to January 2008), and POPLINE (1927 to December 2007) used the following respective search strategies:

The Cochrane Controlled Trials Register:   (surgical abortion or abortion or surgical abortion or manual suction aspiration or electric suction aspiration) and (analges* or epidural or lidocaine or fentanyl or NSAID or general anesthesia or narcot* or sedat* or anxiolyt* or pain control or midazolam or diazepam or ibuprofen or NSAID or vicodin or percocet or morphine or propofol or nitrous oxide) and (first trimester)

MEDLINE:   (analges* OR epidural OR lidocaine OR fentanyl OR NSAID OR general anesthesia OR narcot* OR sedat* OR anxiolyt* OR pain control OR midazolam OR diazepam OR ibuprofen OR NSAID OR vicodin OR percocet OR morphine OR propofol OR nitrous oxide) AND (surgical abortion OR abortion OR surgical abortion OR manual suction aspiration OR electric suction aspiration) AND first trimester AND (randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh] OR random allocation [mh] OR double‐blind method [mh] OR single‐blind method [mh] OR clinical trial [pt] OR clinical trials [mh] OR "clinical trial" [tw] OR ((singl* [tw] OR doubl* [tw] OR trebl* [tw] OR tripl* [tw]) AND (mask* [tw] OR blind* [tw])) OR "latin square" [tw] OR placebos [mh] OR placebo* [tw] OR random* [tw] OR research design [mh:noexp] OR comparative study [mh] OR evaluation studies [mh] OR follow‐up studies [mh] OR prospective studies [mh] OR cross‐over studies [mh] OR control* [tw] OR prospectiv* [tw] OR volunteer* [tw]) NOT (animal [mh] NOT human [mh])

EMBASE:   ((induced abortion! or suction(w)aspiration or surgical(w)abortion) and first(w)trimester) and ((pain(w)control or pain(w)relief) or analgesic agent! or anesthesia! or nonsteroid antiinflammatory agent)

POPLINE:   (abortion/first trimester termination) & (analges*/anesthes*/(pain & (control/relief)))

There was no language preference in the application of the search.

Other sources:   We contacted professionals in the field to seek other trials, including unpublished or ongoing trials we might have missed with the electronic search. Reference lists of articles retrieved were also searched.

Data collection and analysis

Selection of studies:

The primary reviewer evaluated the articles identified from the literature search described above. An additional reviewer evaluated articles in question for inclusion, and in some cases inclusion was based on additional information received from the study's author. Trials under consideration were evaluated for appropriateness for inclusion and methodological quality, including the study design, randomization method, group allocation concealment, and exclusion after randomization without consideration of their results.

A quality score for concealment of allocation has been assigned to each trial using the criteria described in the Cochrane Handbook:

(A) adequate concealment of the allocation

(B) unclear whether adequate concealment of the allocation

(D) allocation concealment not used

Trials scoring A, B, and C were included in this review. The few studies that scored C were included due to their overall importance to the review.

First trimester abortion was defined as an abortion at < 14 weeks gestation. In order for a study to be included, the procedure had to be either electric or manual suction aspiration as opposed to sharp curettage. If the authors did not state the gestational age, but suction aspiration only was performed, we assumed based on common practice that the gestational age was in the first trimester and therefore, included the study. If the authors specified that they performed terminations in the 1st trimester, but did not state that the procedure was performed by suction aspiration versus sharp curettage, and the study was performed in a time period and region where suction aspiration was the predominant procedure method, we assumed that the study used suction aspiration as well, and therefore, included the study. If a sharp curettage check was preformed after suction aspiration, we included the study and noted this step in the individual included study table. If the authors did not state gestational age and did not specify that suction aspiration was performed, or if the procedure was a sharp curettage only, we did not include the study.

Studies were not included if the authors did not measure pain; the primary outcome of this review. Studies were also not included if the only pain‐related outcome reported was need for postoperative analgesics, but pain itself was not measured.

Data extraction and management

Data extraction was performed independently by two of the reviewers. In case of discrepancies, this was resolved by consensus. Attempts were made to obtain additional information from authors if clarifications were needed. Data was then entered in RevMan 5.

Data in the present review has been based on the analytic method (e.g., intention‐to‐treat, per‐protocol) used in the trial report. The main focus was on procedural‐related pain. Given that randomization usually occurred just prior to the procedure and that follow‐up was only for the immediate postoperative time period, exclusions after randomization and loss‐to‐follow‐up were not a significant problem.

Measures of treatment effect

Data was processed using the RevMan software. Peto odds ratios using a fixed‐effects model with 95% confidence interval (CI) were calculated for all dichotomous outcomes. Weighted mean differences (WMD) using a a random‐effects model with 95% confidence intervals were used for continuous outcomes. The data from 11‐point visual or verbal pain scales was treated as continuous data to allow comparisons to 10 cm scales.

Data synthesis

A few studies reported their results as graphs (Suprapto 1984; Hall 1997; Lindholm 1994; Marc 2007) or medians (Heath 1989; Dahl 2000; Kan 2004; Kan 2006; Wong 2002) only. Thus, we could not extract data for comparisons, and only descriptions of their results were included. In other studies percentages were reported and we calculated the number of patients based on the total number of participants per group (Barneschi 1985; Bonnardot 1987; Raeder 1992). If the mean and the standard error of the mean (SEM) were reported, we calculated the standard deviation (SD) using the formula standard error (SE) x square root of N (Bone 1988; Collins 1985; Hackett 1982). If the mean and the CI were reported we calculated the SD using the formula square root of N x (upper limit ‐ lower limit)/3.92 for 95% CI (Phair 2002). Some studies reported categorical outcomes with 3 groups. If deemed appropriate, the result groups were dichotomized for outcomes including pain, side effects or satisfaction to allow for analysis (Collins 1985; Dahl 2000; Phair 2002; Suprapto 1984; Wong 2002). In case a study compared more than 2 groups, we selected 2 groups at a time for comparison in Revman.

We attempted to contact study authors with missing or unclear data. We found the studies to be included in this review to be extremely heterogeneous and thus, could not perform a single meta‐analysis. However, it was possible to group trials into 7 groups:

Group 1: local anesthesia

Group 2: paracervical block with premedication

Group 3: analgesia per os only

Group 4: conscious sedation

Group 5: general anesthesia

Group 6: general anesthesia with premedication

Group 7: non‐pharmacological interventions

For the purpose of this review, conscious sedation was defined as a drug‐induced depression of consciousness during which patients responded purposefully to verbal commands (spontaneous respiration with no interventions needed to maintain a patent airway). With deep sedation, the ability to independently maintain ventilation may be impaired and patients may require assisted ventilation. General anesthesia was defined as drug‐ induced loss of consciousness. Patients are not arousable, not even by painful stimulus. Frequently patients will require assistance in maintaining an open airway, possibly including positive pressure ventilation. Cardiovascular function may be affected (Steele 2005).

The primary outcome, pain, was assessed at different time points depending on the type of anesthesia used. With local anesthesia, pain was usually assessed during and sometimes after the procedure. In the sedation and general anesthesia groups, pain was assessed postoperatively. Instruments used to assess pain varied; some were dichotomous, others categorical. Some used visual or verbal analogue scales in a continuous way. This further made direct comparison of studies more difficult and increased data heterogeneity.

While many studies used a 11‐point VAS (visual/verbal analog scale), some used a 100mm VAS (Edelman 2004; Edelman 2006; Kan 2004; Kan 2006; Dahl 2000; Li 2003; Liu 2005; Suprapto 1984). In order to facilitate comparability, if possible the 100mm VAS was converted to a 11‐point VAS by dividing the results by 10 (Edelman 2004; Edelman 2006; Li 2003; Liu 2005).

Within groups we ordered studies by anesthesia technique, substance and within these subgroups by no intervention/placebo versus intervention. In most instances we chose to create subcategories within an outcome for doses and route of administration rather than different times of outcome assessment. Side effects were listed as subcategories when deemed appropriate for better overview.

Co‐interventions were heterogeneous as well, and participants were not randomized to them. They are briefly summarized in the results section and the most important are presented in the Summary of findings table 1; Table 2; Table 3. An individual and detailed description can be found in the included study tables. They may have affected the results and may have even introduced bias. However, due to their heterogeneity we did not include them in our data analysis

Summary of findings 2. General anesthesia.

Study

Fentanyl^$

Alfentanil ^$

Midazolam/diazepam/lorazepam (benzodiazepine)*

Propofol*

Ketamine*

Methohexital (barbiturate)*

Thiopental (barbiturate)*

Etomidate^#

Halothane ^#

Enflurane^#

Trichlorethylene^#

N2O2

Barneschi

Bonnardot

Boysen 1989

Boysen 1990

Collins

Hackett

Hall

Jakobsson 1991

X (vs fent vs placebo)

Jakobsson 1993

Jakobsson 1995

Lindholm

X vs fent vs NS

Ogg

Raeder

Rossi

Study	Fentanyl^$	Alfentanil^$	Midazolam (benzodiazepine)*	Propofol*	Thiopental (barbiturate)*	Desflurane^#	Enflurane#	N2O2^#	Opioid	Non‐selective COX	Selective COX
Bone	x				c		c	c	Nalbuphine vs fentanyl
Dahl		c	c	c				c	Paracetamol with codeine		Paracetamol (COX 3) vs placebo
Heath		c		c				c	dihydrocodeine vs placebo
Hein 1999	c			c				c			Paracetamol vs placebo
Hein 2001	c			c				c		Lornoxicam	Paracetamol
Jakobsson 1996		c			c			c		Sodium‐diclofenac vs ketorolac vs potassium‐diclofenac vs NaCl
Liu				c		c		c			Etoricoxib vs placebo (COX 2)

Local anasthetisia				WMD
	Treatment	Control	Notes, co‐treatments	Dilation	Aspiration	Post‐op	Satisfaction
LOCAL ANESTHETICS
Paracervical block (PCB) v placebo/no treatment
Glantz 2001	Chloroprocaine 1%	Bacteriostatic saline (0.9% benzyl alcohol)	14 ml, 2 sites (4, 8)	‐0.5 (pain with PCB) ns	‐1.5 ns	‐1.9
Glantz 2001	Chloroprocaine 1%	Bacteriostatic saline (0.9% benzyl alcohol)	14 ml, 4 sites (3, 5, 7, 9),	‐1.3 (pain with PCB)	‐1.7	‐1.3 ns
Kan 2004	Lignocaine 1%	No treatment	10 ml, 2 sites (4, 8), 2.5 cm deep, Co‐treatment: conscious sedation Small trial Also other active arm. Only medians reported.	ns	ns	ns
Various local anesthetics
Wiebe 1992	Carbonated lidocaine 2%	Plain lidocaine 2%	10 ml, with 2mg atropin/50ml, No delay, 3 to 6 sites (12, 3, 6 or 12, 2, 4, 6, 8, 10 o'clock) ½ in. deep, no waiting. All participants: premedication with 1mg lorazepam sublingual 30 minutes prior to procedure per patient request.	‐0.8		‐0.4
Wiebe 1995	Carbonated lidocaine 1%	Plain lidocaine 1%	20ml, (10ml injected in 4 to 6 sites around the cervix and 5ml each between 3 and 4 o'clock and between 8 and 9 o'clock, 1 inch deep, no waiting. All participants: premedication with lorazepam 0.5‐1mg SL per patient request 30 minutes prior to procedure		‐0.96 ns	‐0.05 ns
Wiebe 1996	Lidocaine 0.5%	Lidocaine 1%	20 ml. Some patients received preoperative laminaria, lorazepam or ibuprofen.		0.2 ns
Wiebe 1995	Lidocaine 1%	Bupivacaine 0.25%	20 ml, as in other groups		‐0.24 ns
Local anesthesia technique
Depth of paracervical block
Cetin 1997	Deep injection (1ml superficially and 3ml 3cm deep at 4, 6, 8, and 10 o'clock position; total of 16ml)	Regular injection (1.5cm deep at same 4 positions)	16ml 1% lidocaine. All participants: 5mg oral diazepam 60 minutes prior to procedure if preprocedural anxiety of 6 or more (rated by physician not performing procedure). After 2 minute wait, cervical dilation. Vacuum aspiration followed by sharp curette.	‐0.8	‐0.9
Wiebe 1992	Superficially to blanch the mucous membrane: 1ml injected at 6 sites (12, 2, 4, 6, 8 and 10 o'clock). Then 3‐4ml injected 1 to 1.5 inches deep at 4 sites (4, 6, 8, and 10 o'clock). Total of 20ml 1% plain lidocaine with 1mg atropin/50ml.	½ inch deep at the reflection of the vagina off the cervix. 3 to 6 sites (12, 3, 6 or 12, 2, 4, 6, 8, 10 o'clock). 10ml 2% plain lidocaine with 2mg atropin/50ml.	No delay All participants: premedication with 1mg lorazepam sublingual 30 minutes prior to procedure per patient request.	‐2.4	‐1.0
Paracervical block 4 sites v 2 sites
Glantz 2001	4 sites bacteriostatic saline (3, 5, 7, 9)	2 sites bacteriostatic saline (4, 8)	14ml, Also chloroprocaine in 2 groups	0.8 (pain with PCB) ns	0.1 ns	‐0.5 ns
Glantz 2001	4 sites 1% chloroprocaine (3, 5, 7, 9)	2 sites 1% chloroprocaine (4, 8)	14ml, Also saline placebo in 2 groups	0 (pain with PCB) ns	‐0.1 ns	0.1 ns
Waiting v no waiting paracervical block
Phair 2002	Waiting 3‐5 mins	No waiting	12ml 1% buffered lidocaine at 12 (superficially, cervix), 4 and 8 o'clock (1‐2cm deep, paracervical). Co‐treatment: fentanyl IV and or diazepam per patient request	‐0.7	‐0.2 ns	‐0.1 ns	1.58 ns
Slow v fast injection paracervical block
Wiebe 1995	Fast 30 secs	Slow 60 secs	Lidocaine 1%, 20 ml, no waiting Factorial design Outcome: pain with injection	0.62 (pain with PCB) ns
Intrauterine infusion
Edelman 2004	Lidocaine 10ml, 1%	Saline placebo 10ml	All participants: premedication with 800mg ibuprofen, and if requested, 5mg diazepam. Paracervical block with 10ml of 1% lidocaine (1ml 1% nonbuffered lidocaine on the anterior and posterior lip of the cerivx and then 4.5ml of 1% lidocaine paracervical at the 4‐ and 8‐o'clock postitions). 100mm VAS	‐0.3 ns	‐0.4 ns	0.7 ns	‐0.1 ns
Edelman 2006	Lidocaine 5 ml, 4%	Saline placebo 5ml	Co treatment: ibuprofen 800 mg, cervical lidocaine 1% 10 ml, 4 sites, diazepam mg if requested 100mm VAS	‐2	‐2.8	‐0.5 ns	0.5 ns
Topical
Li 2006	Lignocaine yelly 2% 3 ml ?applied to cervix?, to dilator and speculum	Placebo gel	Co‐treatment: All subjects: cervical priming with 400micrg misoprostol prior to the procedure (1‐2 hours in multiparous, 3‐5 hours in nulliparous subjects). Premedication with 5mg diazepam po and 1mg/kg pethidine IM 15‐30 minutes prior to the procedure. Rescue pain medication with pethidine repeat dose IM.	‐0.42 ns	‐0.87	‐0.51 marginal sig
Cervical block
Kan 2004	Cervical, 2 sites (4, 8), 2.5 cm deep, Lignocaine 1%, 10ml	No treatment	10ml, Co‐treatment: All patients: 400mcg misoprostol vaginally for cervical priming 3‐6hrs prior to the procedure. Conscious sedation with 2mg midazolam and 25mcg fentanyl IV 5 minutes prior to cervical dilation. Pethidine IM as needed for additional analgesia. Small trial Also other active arm. Only medians reported.	ns	ns	ns
Cervical v paracervical
Kan 2004	Cervical 2.5 cm deep	Paracervical, 2.5 cm deep	Lidocaine 1%, 10ml , 2 sites (4, 8) Co‐treatment: conscious sedation (details see other arm) Small trial Also no treatment arm. Only medians reported.	ns	ns	ns

Author (year)	Randomization	Randomization unclear	Allocation concealment	Allocation concealment unclear	Allocation concealment Inadequate	Blinding
Barneschi	computer		1 envelope			Participants and outcome assessor= double blind
Bone		1		1		Double blind
Bonnardot (1987)	Table of numbers		1 Opaque envelopes			Outcome assessor
Boysen (1989)		1		1
Boysen (1990)		1		1		Outcome assessor
Cetin	Computer			1		Unclear
Collins		1		1		Unclear
Dahl	Random number list			1		Double blind
Edelman 2004	Computer		1 identical study syringes			Double blind
Edelman 2006	computer		1 identical study syringes			Double blind
Glantz	Permuted block technique		1opaque sealed envelopes
Hackett		1		1		Outcome assessor
Hall, G		1	1 closed envelopes			Investigators blinded
Heath		1		1		Double blind
Hein 1999	computer				1 nurse drew envelope	Double blind
Hein 2001		1	1 envelope			Double blind
Jakobsson 1991	computer		1 envelope			Patient and outcome assessor = double blind
Jakobsson 1993	computer		1 envelope			Patient and outcome assessor = double blind
Jakobsson 1995	computer		1 envelope			Patient and outcome assessor = double blind
Jakobsson 1996	computer		1 sealed envelope			Double blind
Kan 2004	computer		1 opaque envelope			Double blind
Kan 2006	computer		1 sealed envelope			Double blind
Li 2003	computer		1 sealed envelope			Double blind
Li 2006	computer		1 sealed envelope			Double blind
Lindholm		1		1 (identical looking ampoules delivered by pharmacy)		Anesthesiologist
Liu		1	1 sealed opaque envelopes			Double blind
Marc	computer		1 sealed opaque envelopes
Ogg		1		1		Outcome assessor
Phair	computer		1 opaque envelopes			NO BLINDING POSSIBLE
Raeder	computer		1 sealed envelopes			Double blind
Rossi		1		1		Unclear
Shapiro		1		1		Unclear
Suprapto		1		1		Double blind
Wells 1989		1		1		nurses, counselors, physicians and technicians
Wells 1992	computer		1 envelopes			Outcome assessor
Wiebe 1992	computer				1 The nurse drew up the syringes	Some phases double blind
Wiebe 1995	computer		1 opaque envelopes			Double blind for some phases
Wiebe 1996		1			1 assistant drew up the syringes	Double blind
Wiebe 2003	computer		1 opaque envelopes			Double blind
Wong 2002	computer		1 opaque envelopes			Double blind
TOTAL	2 4	16	23	14	3

Date	Event	Description
11 January 2009	Amended	Reviewers comments implemeted.
15 October 2008	Amended	Implemented the reviewers comments.
16 June 2008	Amended	Converted to new review format.
28 January 2007	New citation required and conclusions have changed	Substantive amendment.

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Pain with paracervical block or dilation comparing local anesthetics with bacteriostatic normal saline	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
1.1 1% chloroprocaine versus bacteriostatic (benzyl alcohol) normal saline 14ml at 2 sites‐ pain with PCB	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
1.2 1% chloroprocaine versus bacteriostatic (benzyl alcohol) normal saline 14ml at 4 sites‐ pain with PCB	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
1.3 1% chloroprocaine 14ml versus bacteriostatic (benzyl alcohol) normal saline 14ml all sites combined ‐ pain with PCB	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
2 Pain with aspiration comparing local anesthetics with bacteriostatic normal saline	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
2.1 1% chloroprocaine versus bacteriostatic (benzyl alcohol) normal saline 14ml at 2 sites‐ pain with aspiration	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
2.2 1% chloroprocaine versus bacteriostatic (benzyl alcohol) normal saline 14ml at 4 sites‐ pain with aspiration	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
2.3 1% chloroprocaine 14ml versus bacteriostatic (benzyl alcohol) normal saline 14ml all sites combined‐ pain with aspiration	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3 Pain postpoperatively comparing local anesthetics with bacteriostatic normal saline	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
3.1 1% chloroprocaine 14ml versus bacteriostatic (benzyl alcohol) normal saline 14ml at 2 sites‐ pain postop	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3.2 1% chloroprocaine 14ml versus bacteriostatic (benzyl alcohol) normal saline 14ml at 4 sites‐ pain postop	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3.3 1% chloroprocaine 14ml versus bacteriostatic (benzyl alcohol) normal saline 14ml all sites combined‐ pain postop	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4 Pain with dilation comparing 2% buffered lidocaine with 2% plain lidocaine	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
5 Pain with aspiration comparing 1% buffered lidocaine with 1% plain lidocaine 20ml each	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
6 Pain at end of procedure comparing buffered lidocaine with plain lidocaine	2	Mean Difference (IV, Random, 95% CI)	Totals not selected
6.1 2% buffered lidocaine versus 2% lidocaine 10ml each.	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.2 1% buffered lidocaine versus 1% lidocaine 20ml each.	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
7 Pain with aspiration comparing 0.5% lidocaine with 1% lidocaine 20ml each	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
8 Pain with aspiration comparing 1% lidocaine with 0.25% bupivacaine 20ml each	1	Mean Difference (IV, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Pain with dilation comparing a deep paracervical block with a regular injection technique	2	229	Mean Difference (IV, Random, 95% CI)	‐1.64 [‐3.21, ‐0.08]
2 Pain with aspiration comparing a deep paracervical block with a regular injection technique	2	229	Mean Difference (IV, Random, 95% CI)	1.00 [‐1.09, ‐0.91]
3 Pain with PCB placement comparing 4 site (3‐5‐7‐9 o'clock) with 2 site (4‐8 o'clock) PCB	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
3.1 Bacteriostatis saline	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
3.2 1% chloroprocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4 Pain with aspiration comparing 4 site (3‐5‐7‐9 o'clock) with 2 site (4‐8 o'clock) PCB	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
4.1 Bacteriostatic Saline	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
4.2 1% chloroprocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5 Pain postoperatively comparing 4 site (3‐5‐7‐9 o'clock) with 2 site (4‐8 o'clock) PCB	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
5.1 Bacteriostatic saline	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5.2 1% chloroprocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6 Pain with dilation comparing 3‐5 minute wait with no wait after PCB of 12ml 1% buffered lidocaine	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
7 Pain with aspiration comparing 3‐5 minute wait with no wait after PCB of 12ml 1% buffered lidocaine	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
8 Pain postpoperatively comparing 3‐5 minute wait with no wait after PCB of 12ml 1% buffered lidocaine	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
9 Pain with injection comparing fast injection with slow injection of buffered lidocaine	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
10 Pain with dilation comparing intrauterine lidocaine with placebo	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
10.1 1% lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
10.2 4% lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
11 Pain with aspiration comparing intrauterine lidocaine with placebo	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
11.1 1% lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
11.2 4% lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
12 Pain 30 min postoperatively comparing intrauterine lidocaine with placebo	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
12.1 1% lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
12.2 4% lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
13 Pain with dilation comparing 2% lignocaine gel 10ml with KY jelly 10ml	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
14 Pain with aspiration comparing 2% lignocaine gel 10ml with KY jelly 10ml	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
15 Pain postpoperatively comparing 2% lignocaine gel 10ml with KY jelly 10ml	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
16 Satisfaction with pain control comparing lignocaine gel with KY jelly	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
17 Satisfaction with the abortion experience comparing intrauterine lidocaine with placebo	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
17.1 1% intrauterine lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
17.2 4% intrauterine lidocaine	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
18 Satisfaction with the abortion experience comparing deep with regular PCB injection technique	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Pain with aspiration comparing ibuprofen 600mg po with placebo in addition to PCB	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
2 Pain postpoperatively comparing ibuprofen 600mg po with placebo in addition to PCB	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
3 Pain with aspiration comparing 1mg lorazepam po with placebo in addition to PCB	1	Mean Difference (IV, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Pain with aspiration comparing diclofenac sodium 50mg/misoprostol 200mcg po with misoprostol 200cmg po	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
2 Pain postoperatively comparing diclofenac sodium 50mg/misoprostol 200mcg po with misoprostol 200mcg po	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
3 Acceptability of pain control comparing diclofenac sodium/misoprostol po with misoprostol po	1	Mean Difference (IV, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Satisfaction comparing conscious sedation with placebo	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
2 Pain with aspiration comparing PCB and IV sedation with PCB	1		Mean Difference (IV, Random, 95% CI)	Totals not selected

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Postoperative pain comparing halothane and alfentanil	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
2 Postoperative pain comparing thiopental and fentanyl with thiopental and halothane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
3 Postoperative pain comparing thiopental and fentanyl with thiopental and enflurane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
4 Postoperative pain comparing trichlorethylen with total IV (methohexital) anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
5 Blood loss (ml) comparing inhalational anesthetics with opiates	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
5.1 comparing halothane with alfentanil	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
5.2 comparing enflurane with fentanyl	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6 Anesthetic complications comparing halothane and alfentanil	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
7 Side effects comparing enflurane with fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
7.1 Severe anesthetic complications comparing enflurane with fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.2 Nausea and vomiting comparing enflurane with fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
8 Side effects comparing trichloethylene with total IV anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
8.1 Lanryngospasm comparing trichloethylene with total IV anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
8.2 Pain on induction comparing trichloethylene with total IV anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
8.3 Nausea comparing trichloethylene with total IV anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
8.4 Vomiting comparing trichloethylene with total IV anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
8.5 Intraoperative involuntary muscle movement comparing trichloethylene with total IV anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
9 Recovery time (min.) comparing inhalation anesthetics with opiates	2		Mean Difference (IV, Random, 95% CI)	Totals not selected
9.1 Recovery time (min.) comparing halothane and alfentanil	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
9.2 Recovery time (min.) comparing enflurane with fentanyl	1		Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
10 Postoperative pain comparing propofol with etomidate	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
11 Postoperative pain comparing propofol with thiopental	3	350	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.11 [0.61, 2.02]
11.1 Alfentanil	2	140	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.20 [0.52, 2.75]
11.2 Fentanyl	2	210	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.02 [0.43, 2.42]
12 Postoperative pain comparing propofol with methohexital	2	140	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.42 [0.16, 1.12]
12.1 Alfentanil	1	40	Peto Odds Ratio (Peto, Fixed, 95% CI)	7.79 [0.47, 129.11]
12.2 Fentanyl	1	100	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.28 [0.10, 0.80]
13 Postoperative pain comparing propofol and fentanyl with midazolam and fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
14 Postoperative pain comparing propofol and alfentanil with ketamine and midazolam	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
14.1 Ketamine 0.5mg/kg and midazolam 0.25mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
14.2 Ketamine 1mg/kg and midazolam 0.1mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
15 Postoperative pain comparing propofol and ketamine with propofol and fentanyl	2	180	Peto Odds Ratio (Peto, Fixed, 95% CI)	4.66 [2.16, 10.06]
16 Postoperative pain comparing thiopental and fentanyl with ketamine and diazepam	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
17 Postoperative pain comparing propofol and alfentanil with propofol	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
18 Postoperative pain comparing placebo with alfentanil and fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
18.1 Alfentanil	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
18.2 Fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
19 Postoperative pain comparing alfentanil with fentanyl when added to propofol	2	210	Peto Odds Ratio (Peto, Fixed, 95% CI)	1.96 [1.07, 3.60]
20 Postoperative pain comparing alfentanil with fentanyl when added to thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
21 Side effects comparing propofol with other sedative hypnotic agents	6		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
21.1 Pain on injection comparing propofol with etomidate	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.2 Pain on injection comparing propofol with thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.3 Pain on injection comparing etmoidate with thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.4 Pain on injection comparing propofol with methohexital	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.5 Pain on injection comparing propofol and fentanyl with fentanyl and midazolam	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.6 Pain on injection comparing propofol and fentanyl with propofol and ketamine	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.7 Pain with injection comparing propofol and alfentanil with ketamin 1mg/kg and midazolam 0.1mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.8 Pain with injection comparing propofol with ketamine 1mg/kg and midazolam 0.1mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.9 Pain with injection comparing propofol with propofol and alfentanil	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.10 Apnea comparing propofol with etomidate	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.11 Apnea comparing propofol with thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.12 Apnea comparing etomidate with thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.13 Apnea comparing propofol with methohexital	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.14 Muscle movements comparing propofol with etomidate	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.15 Muscle movements comparing propofol with thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.16 Muscle movements comparing etomidate with thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.17 Muscle movements comparing propofol with methohexital	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.18 Intraoperative movement comparing thiopental and fentanyl with ketamine and diazepam	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.19 Intraoperative movement comparing thiopental and fentanyl with thiopental and halothane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.20 Intraoperative movement comparing ketamine and diazepam with thiopental and halothane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.21 Movement with cervical dilation comparing propofol with ketamin 1mg/kg and midazolam 0.1mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.22 Movement with cervical dilation comparing propofol and alfentanil with ketamin 1mg/kg and midazolam 0.1mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.23 Movement with cervical dilation comparing propofol with propofol and alfentanil	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.24 Nausea comparing propofol and fentanyl with thiopental and fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.25 Nausea comparing propofol and alfentanil/fentanyl with methohexital and alfentanil/fentanyl	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.26 Nausea comparing propofol and fentanyl with propofol and ketamine	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.27 Nausea comparing propofol and fentanyl with midazolam and fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.28 Nausea comparing thiopental and fentanyl with ketamine and diazepam	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.29 Nausea comparing thiopental and fentanyl with thiopental and enflurane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.30 Nausea comparing thiopental and fentanyl with thiopental and halothane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.31 Nausea comparing ketamine and diazepam with thiopental and halothane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.32 Nausea comparing ketamine and diazepam with thiopental and enflurane	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.33 Vomiting comparing propofol and fentanyl with propofol and ketamine	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.34 Vomiting comparing propofol and fentanyl with midazolam and fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.35 Vomiting comparing propofol and alfentanil with ketamine 1mg/kg and midazolam 0.1mg/kg	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.36 Dreams comparing propofol and ketamine with propofol and fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.37 Dreams comparing propofol and ketamine with propofol and thiopental	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.38 Dreams comparing propofol and ketamine with propofol and methohexital	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.39 Hallucinations comparing propofol and fentanyl with propofol and ketamine	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
21.40 Hallucinations comparing fentanyl and midazolam with propofol and ketamine	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22 Side effects comparing propofol and placebo with propofol and either alfentanil or fentanyl	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
22.1 Nausea comparing alfentanil with placebo	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.2 Nausea comparing fentanyl with placebo	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.3 Nausea comparing alfentanil with fentanyl	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.4 Vomiting comparing alfentanil with placebo	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.5 Vomiting comparing fentanyl with placebo	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.6 Vomiting comparing alfentanil with fentanyl	2		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.7 No complications comparing alfentanil with placebo	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.8 No complications comparing fentanyl with placebo	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.9 No complications comparing alfentanil with fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.10 Laryngospasm or difficulty ventilating comparing alfentanil with placebo	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.11 Laryngospasm or difficulty ventilating comparing fentanyl with placebo	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
22.12 Laryngospasm or difficulty ventilating comparing alfentanil with fentanyl	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
23 Time to discharge	3		Mean Difference (IV, Random, 95% CI)	Subtotals only
23.1 Time to discharge (min.) comparing propofol with thiopental	2	200	Mean Difference (IV, Random, 95% CI)	‐14.69 [‐24.95, ‐4.43]
23.2 Time to discharge (min.) comparing propofol and placebo with propofol and alfentanil	1	104	Mean Difference (IV, Random, 95% CI)	‐9.0 [‐24.87, 6.87]
23.3 Time to discharge (min.) comparing propofol and placebo with propofol and fentanyl	1	104	Mean Difference (IV, Random, 95% CI)	2.0 [‐16.50, 20.50]
24 Pain with dilation comparing conscious sedation and PCB with general anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
25 Pain with aspiration comparing conscious sedation and PCB with general anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
26 Postoperative pain comparing conscious sedation and PCB with general anesthesia	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
27 Apnea incidence comparing conscious sedation and PCB with genereal anesthesia	1		Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
28 Duration of sleep (min.) comparing conscious sedation and PCB with general anesthesia	1		Mean Difference (IV, Random, 95% CI)	Totals not selected

Pain (mean) with aspiration comparing ralxation versus pleasant imagery versus analgesic imagery versus contro
Study	relaxation	pleasant imagery	analgesic imagery	control
Wells 1989	6.77	5.45	7.36	5.76

Methods	Randomized controlled trial. Hospital in Firenze, Italy.
Participants	60 pregnant women. Age: 17‐44 years. ASA I or II. Elective pregnancy termination. Gestational age 7‐12 weeks.
Interventions	Group 1: fentanyl 1.5mcg/kg IV two minutes prior to induction with thiopentone sodium 5mg/kg IV; maintenance with N2O 60% in O2 via mask and assisted breathing.   Group 2: ketamine 2mg/kg and diazepam 0.1mg/kg IV; maintenance with N2O 60% in O2 via mask and assisted breathing.   Group3: thiopentone sodium 5mg/kg IV; maintenance with N2O60% in O2 and enflurane 2.5% after complete cervical dilation via mask and assisted breathing   Group 4: thiopentone sodium 5mg/kg IV; maintenance with N2O60% in O2 and halotane 2% after complete cervical dilation via mask and assisted breathing   All participants: premedication with atropine 0.02mg/kg IM 45 minutes prior to the procedure. Vacuum aspiration with or without sharp curettage.
Outcomes	Psychomotor recovery time with Zazzo's test of "deux barrages" and the matrix attentive test. Anesthesia time.   Intraoperative side‐effects: introperative movement.   Postoperative side‐effects: strong abdominal pain, headache, nausea, vomiting, agitation at awakening (all yes/no).
Notes	No power analysis done. Randomization method and blinding not further described.   Per e‐mail communication with Dr. Barneschi: Study length 4 months. Randomization with computer tables. Allocation concealment with envelope technique. Blinding of participants and outcome assessor. Vacuum aspiration. No industry sponsorship.   No major complication reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Randomized double‐blind controlled trial. Leicester Royal Infirmary, Leicester, UK
Participants	40 pregnant women. Age: 16‐40 years. ASA I and II. Gestational week up to 12 weeks. Exclusion criteria: history of asthma, drug allergy or sensitivity to opioids.
Interventions	Group 1: nalbuphine 0.25 mg/kg IV.   Group 2: fentanyl 1.5mcg/kg IV.   All participants: No premedication.   One minute after opioid anesthesia was induced in both groups with thiopentone 3‐4mg/kg. Maintenance with Bain breathing system with 66% nitrous oxide in oxygen and supplementation with enflurane as needed. Syntocinon 10units IV at onset of cervical dilation. Postoperative analgesia with paracetamol 1g orally 6 hourly.
Outcomes	Postoperative pain reported on a 10cm horizontal linear analog scale at 1, 2 and 4 hours postoperatively.   Duration of anesthesia, dose of thiopentone, and maximum concentration of enflurane administration. Time until recovery of consciousness (name, date of birth and correct address) and psychomotor function. Patient assessment as asleep, awake, and calm or awake and restless on four different occasions (pre‐operatively, 1, 2, and 4 hours postoperatively). Postoperative nausea. Need for postoperative analgesics.
Notes	No power calculations. Unclear study length, method of randomization and allocation concealment. Unclear if the procedure was vacuum versus sharp curettage. The author could not be succeessfully contacted to clarify unclear information.   Outcome assessor in recovery room was blinded. No major complication reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized controlled trial. Hospital Tenon, Paris, France. January to October 1986.
Participants	200 pregnant women. ASA I or II.
Interventions	Group 1: propofol 2mg/kg.   Group 2: ketamine 0.5mg/kg and midazolam 0.25mg/kg.   Group 3: propofol 2mg/kg and alfentanil 4mcg/kg.   Group 4: ketamine 1mg/kg and midazolam 0.1mg/kg.   2 successive series of 2 groups (1 and 2, and 3 and 4):   All participants premedicated with midazolam 0.25mg/kg orally.
Outcomes	Side effects including postopoperative pain (yes/no) and vomiting (yes/no), pain with injection, visual disturbances and rash were secondary outcomes.   Primary outcomes: clinical revocery with 4 psychomotor and sensory tests (Newman, Bourdon, Horatz and neursensorial test). Vital signs.
Notes	Per e‐mail communication with Dr. Maillart, the co‐author: randomization with table of numbers, allocation concealment with opaque envelopes, blinding of outcomes assessor, gestational age less than 10 weeks, vacuum evacuation.   No power calculations. No major complication reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Postoperative pain comparing paracetamol supp with placebo	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
1.1 30 minutes postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
1.2 1 hour postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
1.3 At discharge	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
2 Postoperative pain comparing paracetamol/codeine supp with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
2.1 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
2.2 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
2.3 at discharge	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
3 Postoperative pain comparing paracetamol po with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
4 Postoperative pain comparing paracetamol po with lornoxicam	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
5 Postoperative pain comparing diclofenac with ketorolac and with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
5.1 Diclofenac po versus NaCL	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.2 Diclofenac im versus NaCL	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.3 Ketorolac im versus NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.4 Diclofenac po versus ketorolac im	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
5.5 Diclofenac im versus ketorolac im	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
6 Postoperative pain comparing etoricoxib with placebo	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
6.1 15 minutes postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.2 30 minutes postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.3 60 minutes postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
6.4 at discharge	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
7 Side effects comparing COX inhibitors with placebo as premedication for general anesthesia	4	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
7.1 Antiemetic requirements comparing paracetamol supp with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.2 Antiemetic requirements comparing paracetamol po with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.3 Antiemetic requirements comparing lornoxicam with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.4 Antiemetic requirements comparing lornoxicam with paracetamol po	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.5 Nausea comparing diclofenac po with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.6 Nausea comparing diclofenac im with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.7 Nausea comparing ketorolac im with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.8 Vomiting comparing diclofenac po with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.9 Vomiting comparing diclofenac im with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.10 Vomiting comparing ketorolac im with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.11 Anxiety during recovery comparing diclofenac po with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.12 Anxiety during recovery comparing diclofenac im with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.13 Anxiety during recovery comparing ketorolac im with NaCl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
7.14 Satisfaction with pain management comparing etoricoxib with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
8 Recovery time (min.) comparing COX inhibitors with placebo as premedication for general anesthesia	4	Mean Difference (IV, Random, 95% CI)	Totals not selected
8.1 Time to street fittness comparing paracetamol supp with placebo	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.2 Time to discharge comparing paracetamol po with placebo	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.3 Time to discharge comparing lornoxicam with placebo	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.4 Time to discharge comparing lornoxicam with paracetamol po	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.5 Time to discharge comparing diclofenac po with NaCl	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.6 Time to discharge comparing diclofenac im with NaCl	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.7 Time to discharge comparing ketorolac im with NaCl	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.8 Time to discharge comparing diclofenac po with ketorolac im	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.9 Time to discharge comparing diclofenac im with ketorolac im	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
8.10 Time to discharge comparing etoricoxib with placebo	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
9 Postoperative pain comparing Nalbuphine with fentanyl	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
9.1 1 hour postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
9.2 2 hours postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
10 Recovery (reaction time (msec.)) comparing nalbuphine with fentanyl	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
10.1 1 hour postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
10.2 2 hours postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
10.3 4 hours postoperatively	1	Mean Difference (IV, Random, 95% CI)	0.0 [0.0, 0.0]
11 Side effects comparing paracetamol/codeine supp with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
11.1 Nausea at 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.2 Nausea at 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.3 Nausea at discharge	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.4 Fully awake at 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.5 Fully awake at 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.6 Sleepy at 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.7 Sleepy at 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.8 Asleep but easily arousable at 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.9 Asleep but easily arousable at 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.10 Heavily asleep at 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
11.11 Heavily asleep at 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
12 Recovery time (discharge ready) comparing paracetamol/codeine supp with placebo	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
12.1 Discharge ready at 30 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
12.2 Discharge ready at 60 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]
12.3 Discharge ready at 90 minutes postoperatively	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Level of comfort during procedure comparing hypnosis with control group	1	Mean Difference (IV, Random, 95% CI)	Totals not selected
2 N2O request comparing hypnosis with a control group	1	Peto Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
3 Pain with aspiration comparing music with methoxyflurane	1	Odds Ratio (Peto, Fixed, 95% CI)	Totals not selected
4 Pain (mean) with aspiration comparing ralxation versus pleasant imagery versus analgesic imagery versus contro		Other data	No numeric data

Methods	Randomized controlled trial. University of Copenhagen, Denmark.
Participants	60 pregnant women. Age: 18‐44 years. Healthy women. Gestational age up to 12 weeks.
Interventions	Group 1: thiopental 4mg/kg   Group 2: etomidate 0.3mg/kg   Group 3: propofol 2.5mg/kg   for induction of anesthesia.   All participants: No premedication. Alfentanil 15mcg/kg after induction. Additional 1/4 of the induction anesthetic as needed. Oxytocic drugs.   All under assisted ventilation with 100% oxygen with a face mask.
Outcomes	Postoperative side ffects including pain (yes/no) and vomiting.   Dosage requirements. Vital signs. Pain on injection of anesthetic. Recovery: Steward score, coin counting test (CCT), continuous auditory reaction time test (CART). Need for postoperative analgesics. Other side effects: apnea, involuntary muscle movements.
Notes	No power calculations. Unclear study length, method of randomization and allocation concealment. Unclear if the procedure was vacuum versus sharp curettage. The author could not be succeessfully contacted to clarify unclear information.   Outcome assessor was blinded.   Apnea occured in 9, 1 and 3 patients of the propofol, etomidate and thiopental group respectively. There was no significant diffference.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized controlled trial. Randomization with computerized, random number generation carried out outside the familiy planning unit. Per power analysis 33 patients per group required.   Cumhuriyet University, School of Medicine, Sivas, Turkey. Study length: May 1996 to November 1996.
Participants	66 pregnant women. Singleton intrauterine pregnancy at 6‐11 weeks of gestational age by LMP confirmed bv transvaginal ultrasound.
Interventions	Group 1: deep injection (superficially 1ml, 3ml 3cm deep at 4, 6, 8, and 10 o'clock position; total of 16ml) of PCB   Group 2: regular injection (1.5cm deep at same 4 positions) of 10ml 1% lidocaine local anesthetic for paracervical block.   All participants: 5mg oral diazepam 60 minutes prior to procedure if preprocedural anxiety of 6 or more (rated by physician not performing procedure). After 2 minute wait, cervical dilation.   Vacuum aspiration followed by sharp curette.
Outcomes	Pain associated with dilation and curettage was rated by patient on a verbal analog scale of 0 to 10 (0 = no pain, 10 = severe pain) after the procedure.   Anxiety score, procedure time, basal cervical dilation and dilation increased obtained. Follow‐up visit 4‐6weeks after the procedure to assess late complications and to perform a gynecological exam.
Notes	Physician not performing procedure rated preprocedural anxiety (scale 1‐10). Otherwise blinding is not commented on. Unclear how many got diazepam and in which group they were.   The author could not be succeessfully contacted to clarify unclear information.   No adverse effects, no major complications.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized controlled trial. St. Thomas Hospital London, UK.
Participants	66 pregnant women. Age: 16‐38 years old. ASA 1 or 2. Gestational age <12weeks
Interventions	Group 1: general anesthesia with 1.5% halothane and 70% nitrous oxide in oxygen for maintenance after methohexitone induction.   Group 2: general anesthesia with 7mcg/kg alfentanil followed by methohexitone induction and maintenance with alfentanil 0.1‐0.2mg increments as needed and 70% nitrous oxyde in oxygen.   All participants: Oxytocin 5 units IV at onset of cervical dilation.   Vacuum aspiration.
Outcomes	Postoperative abdominal pain (none, slight or moderate/severe). Need for postoperative analgesia.   Blood loss as measured by collection of all blood and the aspirate which was then processed and measured in a standardized way as described by Garrioch, Gilbert and Plantevin 1981. Blood volume was determined using the alkaline haematin method of Hallberg and Nielsson (1964).   Ansethetitc morbidity: duration, hiccup, cough, laryngeal stridor, apnea, limb movement.   Recovery time, nausea and vomiting.
Notes	Randomization and Allocation concelament not described in detail.   Statistical analysis using the group sequential design, in groups of 20 patients. In case of moderate or sever nausea the anesthetic was considered a "failure" for a particular patient.   The trial was stopped after 66 patients, since blood loss was significantly higher with halothane use. One patient in the alfentanil and no patient in the halothane group had apnea; defined as no spontaneoeus respirations for more than 30 seconds.   No power analysis described. Study length, method of randomization and allocation concealment as well as details of blinding are not described. Author could not successfully be contacted regarding missing information.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized controlled double‐blind trial. Randomization made by pharmacy from a list of random numbers. Per power analysis done to detect reduction in postoperative rescue analgesics from 40 to 20% 46 patients are needed in each group.   Hospital in Oslo, Norway.
Participants	100 pregnant women. ASA I or II. Exclusion criteria: cardiac, pulmonary, or renal disease, extreme obesity, allergy to opioids, or paracetamol.
Interventions	Group 1: paracetamol/codeine 800/60mg suppository 1hr prior to surgery.   Groups 2: placebo suppository, identically looking and at the same time   All participants: Premedication with midazolam 0.08mg/kg.   Induction of anestesia for all with propofol 1.5mg/kg and alfentanil 15mcg/kg. Maintenance of anesthesia with 60% nitrous oxide in oxygen and incremental doses of propofol as eeded.   Vacuum aspiration.   Rescue analgesic for postoperative pain: ketobemidone 1‐2mg IV.
Outcomes	Pain level measured with visual analog scale (VAS; 0 = no pain, 100 = worst pain ever) and the verbal pain score (VPS; no pain, slight pain, medium pain, strong pain) at 30 and 60 minutes postoperatively as well as before discharge.   Need for rescue analgesics, occurrence of nausea/vomiting, degree of sedation at 30 and 60 minutes postoperatively as well as before discharge.   Vital signs, duration of surgery, total amount propofol needed.
Notes	Unclear study length, method of allocation concealment. Unclear what the gestational age was, but assume that it was first trimester given that vacuum curettage was used.   100 patients recruited, but 10 exlcuded due to violation of the protocol (insertion of an IUD, use of NSAIDS). Results are only reported for the remaining 90.   No major complications reported.   The author could not be succeessfully contacted to clarify unclear information.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized double‐blind placebo‐controlled trial. Computer generated block randomization sequence (block size 20), through an inverstigator not involved with recruitement. Study syringes were identically looking and prepared by the study nurse and coordinator who labeled them with consecutive numbers. Subjects were also enrolled with consecutive numbers. The Investigator was blinded.   Per power analysis 40 women needed per study group.   Planned Parenthood at Columbia Willamette in Portland, Oregon, USA. Study length: July 2002 to February 2003.
Participants	80 pregnant women. Age: 18 years old or older. Good general health. English speaking. Gestational age: Less than 11 weeks by LMP confirmed by ultrasound. Body weight more than 100lbs (approx 45kg)
Interventions	Group 1: intrauterine infusion of 10ml 1% lidocaine (maximum total lidocaine dose of 200mg).   Groups 2: intrauterine infusion of 10ml of sterile saline.   All participants: premedication with 800mg ibuprofen. If requested, 5mg diazepam. Paracervical block with 10ml of 1% lidocaine (1ml 1% nonbuffered lidocaine on the anterior and posterior lip of the cerivx and then 4.5ml of 1% lidocaine at the 4‐ and 8‐o'clock postitions).   Five experienced surgeons performed all abortions with an electric vacuum pump after dilation of the cervix.
Outcomes	Pain rated by subjects using a 100mm visual analog scale (VAS; anchors: 0 = none, 100 mm = worst imaginable) at several points in time: 1) before the procedure started (anticipatory pain); 2) following speculum insertion; 3) after intrauterine infusion; 4) following dilation; 5) after suction aspiration; and 6) 30 minutes later in the recovery room.   Assessment of overall satisfaction level regarding the abortion experience using the VAS, before discharge. Subjects performed all VAS scales concurrently with each step and not from memory.
Notes	Serum lidocaine were drawn in a subset (n=10) of subjects to obtain safety data. No patient developed overt symptoms of lidocaine toxycity.   One protocol violation in lidocaine group with an inadvertant enrollement of a patient who was not premedicated with ibuprofen. Another patient in the lidocaine group received IV narcotics prior to reaspiration. Her 30 minute pain score was conducted after her reaspiraiton. Analysis in accordance with intent to treat.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Randomized controlled trial at St Thomas' Hospital London, UK. Outcome assessor in recovery room was blinded.
Participants	82 pregnant women. ASA group I or II. Gestational age <12 weeks.
Interventions	Group 1 (enflurane): Induction with methohexitone 2mg/kg. Maintenance with 70% nitrous oxide in oxygen and 3% enflurane until cervical dilation completed.   Group 2 (fentanyl): fentanyl 1.5mcg/kg IV followed by methohexitone 2mg/kg. Maintenance with 70% nitrous oxide in oxygen and incremental doses of methohexitone.   All participants: Vacuum aspiration (Karmen curettage).
Outcomes	Blood loss, duration of operation, minor complications such as coughing, salivation, hiccup, laryngeal stridor, and limb movements. Recovery observations: Recovery of consciousness assessed by time of first eye opening in response to their name. Nausea, vomiting, abdominal pain (nil, slight, moderate/severe) and need for medication (yes/no) after operation during the first hour and each subsequent hour until discharge.
Notes	Unclear study length, method of randomization and allocation concealment. Unclear who was blinded other than the recovery room outcome assessor. Group sequential design to determine trial size. Anestethic agent was considered a failure if moderate to severe nausea occured. The upper boundary was crossed after 74 patients indicating a statistical significnt difference between the anesthesia groups at the 5% level using a one‐sided test.   The author states that the 82 patients were included; 43 in the enflurane and 39 in the fentanyl group. However, patients table 2 with abdominal pain add up to 47 and 110% in the enflurane group.   Severe complications, defined as delaying the procedure, occured in 4 patients; respiratory depression in one patient in the fentanyl group, prolonged coughing with salivation in one patient in the enflurane group and 2 patients in the fentanyl group.   The author could not be succeessfully contacted to clarify unclear information.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized trial. Outpatient clinic of the Department of Obstetrics and Gynecology at the Karolinska Hospital in Stockholm, Sweden. No power calculation.
Participants	200 pregnant women. Age: older than 18. Outpatient clinic, legal 1st trimester surgical abortion.
Interventions	Group 1: preoperative vaginal prostaglandin (PG) (PGE1, gemeprost 1mg given vaginally 3hrs preoperatively) for cervical softening, surgery in general anesthesia (GA) Group 2: preoperative PG and surgery in GA and paracervical block (PCB, 10 and 10ml lidocaine 10mg/ml) Group 3: surgery in GA Group 4: GA and PCB. Premedication with morphine 5‐10mg IM 60‐90 minutes preoperatively. GA with propofol 2mg/kg IV, 60% nitrous oxide in oxygen, breathed spontaneously by mask and isoflurane supplementation as needed. Dilation of the cervix as needed followed by vacuum suction curettage and 10IU oxytocin IV. Postoperatively oral medications per patient request: paracetamol 500mg and codeine phosphate 30mg for pain or thiethylperazin 6.5mg for nausea.
Outcomes	Pain preoperatively, and postoperatively (at 1, 2, 3 and 4 hrs); consumption of analgesics postoperatively; nausea; time interval to discharge home. Study instrument: Visual analog scale (VAS; 10cm from "no pain" to "worst pain ever"; no nausea to extreme nausea) for pain and nausea.
Notes	Per e‐mail communication with Dr. Persson, the coauthor, the study length was 4 months. Patients were randomized and allocation concealment was with closed envelopes. All investigators were blinded; not the operating gynecologist and not the nurses .   Women in the two groups receiving PG were significantly younger, of lower gravity and parity than women in the two groups not receiving PG. The discussion does not include this being a possible confounder of the lower pain perception and analgesia need in the latter two groups.   No major complications reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Randomized controlled double blind trial. Nurse otherwise not involved in study drew the randomized envelope.   Karolinska Institute at Danderyds Hospital, Denderyd, Sweden.
Participants	140 pregnant women. ASA I‐II group. Elective termination of pregnancy.
Interventions	Group 1: 1gm paracetamol suppository.   Group 2: placebo suppository.   All participants: no premedication. Induction of anesthesia with 0.1mg fentanyl and propofol. Maintenance with nitrous oxide in oxygen 2:1 and additional small doses of propofol as needed. Spontaneous breathing, assisted ventilation as needed. 5 units oxytocin at the end of the procedure.   All patients: Vacuum aspiration.   Rescue medication for postoperative pain diclofenac 100mg supp.
Outcomes	Postoperative pain measured with a 10cm visual analog scale (0cm=no pain, 10cm=unbearable pain), measured 30 and 60 minutes postoperatively and at discharge. Analgesic and antiemetic requirements. Time until street fittness.
Notes	Per e‐mail communication with Dr. Jakobsson: Gestational age 7‐12 weeks. Computer‐based randomization with envelopes. The study lasted over 2.5 months.   All procedures were uneventful and no complications were noted. No industry sponsorship.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	High risk	C ‐ Inadequate

PERMALINK

Pain control in first trimester surgical abortion

Regina‐Maria Renner

Jeffrey TJ Jensen

Mark DN Nichols

Alison Edelman

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Search methods for identification of studies

Data collection and analysis

Summary of findings 2. General anesthesia.

Summary of findings 3. General anesthesia with premedication.

Results

Description of studies

Risk of bias in included studies

Effects of interventions

Summary of findings for the main comparison. Local anesthesia.

Summary of findings 4. Quality of evidence.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

Methods	Randomized controlled trial. Karolinska Institute at Danderyds Hospital, Denderyd, Sweden.
Participants	165 pregnant women. ASA I group. Elective ambulatory termination of pregnancy. Gestational age: 7‐12 weeks. Patient age: 17‐44.
Interventions	Group 1: alfentanil 0.5mg.   Group 2: fentanyl 0.1mg.   Group 3: placebo (saline).   All participants: anesthesia induced with propofol and thereafter nitrous oxide in oxygen 2:1. Additional boluses of propofol 10‐30mg as needed. Paracetamol and diclofenac for postoperative pain on patient request.
Outcomes	Peroperative: Complications during anesthesia, such as laryngospasm. Surgeons opinion about quality of anesthesia. Amount propofol required.   Postoperative: Complaint of pain to nurse (yes/no). Request for analgesia. Nausea, vomiting.   Patient self ‐assessment: qustionnaire just prior to discharge, asking about postoperative pain (slight versus intense) and emesis.
Notes	Per e‐mail communication with Dr. Jakobsson: Study lasted approximately 3 months. Blinding of patient and outcome assessor. Randomization per computer. Allocation concealment with envelope technique. Gestational age 7‐12 weeks. Power analysis done. All underwent vacuum aspiration followed by sharp curettage check.   One patient in the placebo group was excluded because of pronounced postoperative beeding. Data of 164 patients was analyzed.   Laryngospasm in 2 patients in the placebo and in one patient in the alfentanil group. Regurgitation requiring intubation in one patient in the placebo group.   No industry sponsorship.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Prospective randomized controlled double‐blind trial. Computer generated randomization list. Opaque, sequentially numbered envelopes with allocation information. Per power calculations 40 patients needed in each arm.   Queen Mary Hospital of University of Hong Kong in Hong Kong, China. Study length: July 2002 to April 2003.
Participants	135 pregnant women. Maternal age >16yrs. Normal general and gynecological examination. Gestational age: equal or less than 12 weeks. Size of the uterus on pelvic examination compatible with estated gestational age.   Exclusion criteria: allergy or contraindication to lignocaine or prostaglandins. History of severe respiratory, cardiac or liver disease, myasthenia gravis, and psychiatric conditions requiring medication.
Interventions	Stratification by parity (nulliparous or multiparous). Randomization into 3 groups.   Group 1: paracervical block (PCB) with 10 mL of 1% lignocaine injected at the 4 and 8 o'clock positions of the vaginal vault 2.5cm beneath the mucosa (5ml each).   Group 2: Cervical block with 10 mL of 1% lignocaine injected at the 4 and 8 o'clock positions of the cervix 2.5cm beneath the mucosa (5ml each)   Group 3: no PCB.   All participants: 400mcg misoprostol vaginally for cervical priming 3‐6hrs prior to the procedure. Prophylactic antibiotics. Conscious sedation with 2mg midazolam and 25mcg fentanyl IV 5 minutes prior to cervical dilation. Pethidine IM as needed for additional analgesia.   Vacuum aspiration (Karmen catheter with an electrical vacuum machine).
Outcomes	Pain scores measured by 100‐point visual analog scale at three points in time: (a) before the operation (following insertion of intravenous catheter), (b) just after the operation to rate pain during PCB, cervical dilation and during suction evacuation (=primary outcome), and (c) 1hr after suction evacuation.   Secondary outcome measures included: satisfaction levels rated by patient prior to discharge home, sedation level and difficulty of the operation rated by the surgeon, additional analgesia, and adverse effects of PCB. Pain and anxiety were measured with a 100mm linear visual analog scale (VAS: 0 = none, 100 = most severe/painful). Sedation was measured with a sedation scale proposed by Ramsey et al. (1974) (6 very detailed defined levels with increasing sedation from level 1 to 6). Satisfaction levels were excellent, satisfactory, fair and unsatisfactory.
Notes	One patient excluded from study since the uterus was found to be enlarged to 14 weeks of gestation in the operation theater. 134 remaining patients were analyzed. No major complications reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Double‐blind, randomized controlled trial. Randomization by computer generated randomization list, and allocation by sealed envelope. Per power calculations 40 patients needed in each arm.   University of Hong Kong, at Queen Mary Hospital, Hong Kong, China. Study length: May 2004 to January 2005.
Participants	90 pregnant women. Inclusion criteria: Maternal age >16yrs, normal general and gyecological examination, gestational age up to 12wks, size of the uterus on pelvic examination compatible with the estimated duration of pregnancy.   Exclusion criteria: history of severe respiratory or cardiac disease, severe and recurrent liver disease, myasthenia gravis, psychiatric conditions requiring medication or disorders that constitute contraindications to the use of prostaglandins. History of entotox use, uperr respiratory infection, sinus blockage, recent history of middle ear or inner ear surgery or histpry of bone marrow suppression.
Interventions	Group 1: entonox (50:50 mixture of nitrous oxide in oxygen) via face mask and T‐piece breathing circuit.   Group 2: air.   The the patient was instructed to inhale the study medication. Maintenance by inhalation as needed.   All participants: 400mcg misoprostol vaginally 3‐6hrs prior to procedure. Prophylactic antibiotics. Conscious sedation with 2mg midazolam (another 1mg if sedation inadequate) and 25mcg fentanyl IV.   Vacuum aspiration.
Outcomes	Pain scores during venipuncture, insertion of intravenous cannula and during vaginal examination were assessed before the procedure using a 100mm linear VAS (0 = no pain, 100 = worst possible pain). Pain scores during the procedure, and 1 hr after the procedure were assessed 1 hr after the procedure using a VAS.   Basal anxiety levels assessed using a state anxiety questionnaire and visual analog scale (VAS). Preprocedural anxiety was assessed by a nurse. Sedation measured by doctor using the Ramsay et al (1974) sedation scale. Post‐operative side‐effects (including nausea, vomiting, dreams, parasthesia, dizziness, dry mouth, memory of operation and euphoria) and satisfaction level assessed 1 hr after the procedure using a questionnaire, as were anxiety and satisfaction level.
Notes	No paracervical block given. No major complications reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Double‐blind randomized controlled trial. Computer generated randomization schedule, medication in sealed, numbered opaque envelope and plastic bag. Per power analysis 45 patients needed in each arm.   University of Hong Kong, at Queen Mary Hospital, Hong Kong, China.
Participants	100 pregnant women; 70 multiparous and 30 nulliparous women. Healthy women, gestational age 7‐12wks. Exlusion criteria: significant medical diseases, medical contraindication to the use of NSAIDs or diclofenac sodium, misoprostol or lorazepam.
Interventions	Group 1: arthotec (misoprostol 200mcg and diclofenac sodium 50mg, a non steroidal anti‐inflammatory drug=NSAID)   Group 2: cytotec (misoprostol 200mcg)   both administered 4hrs prior to procedure.   All participants: lorazepam 1mg sublingually 30 min prior to procedure, rescue pain medication with pethidine25mg IV. Vacuum aspiration.
Outcomes	Pain score during procedure was assessed immediately after procedure. Further pain score was assessed 1hr after the procedure. Measurement instrument: visual analogue scale (100mm linear; 0 = no pain, 100 = most severe pain) given by RN blinded to study group assignment.   Further outcomes measured: Preoperative side effects of medication, cervical priming effect, subject's acceptabiliy of the pain control method (assessed 1 hr after the procedure).
Notes	Unclear study length. 70 multiparous women and 30 nulliparous were randomized seperately. Half life of diclofenac sodium is 1‐2hrs. No paracervical block. One patient excluded from analysis since the operation was done under general anaesthesia upon her request. No intention to treat analysis, but not possible given that with general anesthesia there is no intraoperative pain measurable. No serious complications observed.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Prospective double‐blind randomized placebo‐controlled trial. Computer randomization and allocation with sealed envelopes. Per power analysis for which a decrease of pain level by 0.5 standard deviations chosen as being acceptable 64 patients needed in each arm.   Family Planning Association of Hong Kong, China. Study length: April to June 2005.
Participants	140 pregnant women. Inclusion criteria: Maternal age: older than 16 years. Gestational age 7‐10 weeks at day of procedure. Normal general and gynecological exam. Uterine size corresponding to gestational age or ultrasound dating. Being ethnic Chinese and Cantonese speaking. No history of complicated medical or surgical problems for which operation in community‐based day‐care setting wa contraindicated.   Exlusion criteria: Known allergy to lognocaine or prostaglandin or medical contraindication to the use of prostaglandins.
Interventions	Group 1: 2% lignocaine jelly, 3ml applied to cervix, 7ml applied to Hegar dilators and vaginal speculum 1 minute prior to procedure.   Group 2: KY jelly (placebo) applied in same fashion.   All subjects: cervical priming with 400micrg misoprostol prior to the procedure (1‐2 hours in multiparous, 3‐5 hours in nulliparous subjects). Premedication with 5mg diazepam po and 1mg/kg pethidine IM 15‐30 minutes prior to the procedure. Rescue pain medicatioin with pethidine repeat dose IM. Suction evacuation with electrical vacuum machine.
Outcomes	Pain on an 11‐point verbal analog scale from 0 to 10 (0 = no pain at all, 10 = intolerable pain) preoperative with pethidine injection, on arrival in the operating room for preoperative pain (i.e. from misoprostol), with cervical manipulation and/or dilation, immdiately after operation for overall intraoperative pain and 1 hour after operation. Before discharge satisfaction toward pain control was assessed (0 = totally unsatisfied to 5 excellent). In addition the surgeon graded the level of patient sedation preoperatively and intraoperatively according to standard scale from 1 to 6, described by Ramsay et al 1974.
Notes	Subanalysis for nulliparity versus multiparity was performed. Due to missing part of data 6 patients were excluded in the lignocaine and 3 in the KY jelly group. Results for excluded patients not available. Analysis only includes 64 patients for the lignocaine and 67 patients for the KY Jelly group. No major complications reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Randomized controlled trial. The study solutions were prepared by the local hospital pharmacy and delivered in blinded identical looking ampoules. Anesthesiologists were blinded as well (pupil size was not assessed intraoperatively).   Odense University Hospital, Odense, Denmark.
Participants	120 pregnant women. Gestational age <12 weeks. Healthy patients.   Exclusion criteria: allergy to trial drugs, on medications likely to influence the course of anesthesia, more than 20% overweight.
Interventions	Group 1: normal saline 0.03ml/kg   Group 2: fentanyl 1.5mcg/kg   Group 3: alfentanil 15mcg/kg bolus injection.   All participants: Premedication with lorazepam 2mg orally.   2 minutes after study medication induction of anesthesia with propofol 2mg/kg IV with lignocaine to lessen pain of injection. Additional increments of propofol 0.5mg/kg IV as needed for induction and maintenance of anesthesia. Alfentanil 100mcg/kg IV as rescue medication if more than four incremental doses of propofol required. Ergometrine 0.5mg IV during operation.   Postoperative analgesics were paracetamol 1g for moderate pain, pethidine 75mg IM for severe pain.   Vacuum aspiration.
Outcomes	Postoperative pain intensity assessed with 10cm visual analog scale, anchored with "no pain" at 0 and "worst pain imaginable" at 10cm at 30, 120 and 180 minutes. Need for postoperative analgesics.   Duration of induction, vital signs, induction dose and total dose of propofol, need for alfentanil, quality of induction movement of patient with surgery stimulus. Recovery: open eyes on command, give birthdate, cooperation score. Complications and side effects during and after the operation.
Notes	Unclear randomization. No power analysis mentioned.   The author could not be succeessfully contacted to clarify unclear information.   Laryngospasm occured in 1 patient in the fentanyl group. One patient in the control group was difficult to ventilate.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Randomized placebo‐controlled double‐blind trial. Allocation using sealed opaque envelopes. Power calculations done for a power of 0.8 and alpha=0.05 to detect a 20% difference in opioid usage between groups.   KK Women's and Children's Hospital, Singapore.
Participants	40 pregnant women. ASA I and II, elective first trimester termination of pregnancy in an ambulatory surgical center.   Exclusion criteria: asthma, gastritis, coagulation disorder, renal impairment, allergy to NSAIDs or COX‐2 inhibitors, history of long‐tern analgesc use, or use of any agent that may influence the analgesic response.
Interventions	Group 1: etoricoxib 120mg orally with 20ml plain water   Group 2: placebo   both 30‐60 minutes prior to surgery.   All participants: general anesthesia with IV propofol 2.0‐2.5mg‐kg for induction and oxygen nitrous mixture (40:60) and Desfluran (end tidal) 1% at fresh gas flow of 3l/min. Postoperatively per patient request of if verbal analog scale (VAS) for pain >50mm IV fentanyl bolus of 25mcg every 15 minutes until comfortable or VAS<50mm.
Outcomes	Main outcome: post‐operative need for fentanyl. Post‐operative pain assessed through a blinded observer using a 100mm verbal analogue score (VAS; 0=no pain and 100=the worst imaginable pain). Time points were emergence from general anesthesia, 15 minutes, 30 minutes, and 60 minutes post‐operatively and at discharge.   Further outcomes: side effects (nausea, vomiting gastric pain, heart burn, or dizziness), time to first drink, time to step‐down, time between step‐down and discharge, and patient satisfaction scores (100‐point analogue scale; 0 = very bad experience and 100 = excellent experience) assessed by blinded observer. Via telephone pain sores were assessed at 6 hours and 24 hours post‐operatively as well as need for rescue analgesia with acetaminophen 1g every 6 hours.
Notes	Unclear study length, and randomization method. Unclear if vacuum or sharp curettage.   The author could not be succeessfully contacted to clarify unclear information.   No major complication reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Open randomized pilot study. Computer generated list of random numbers in blocks of 6 and 4. Allocation concealment with sealed opaque envelpes. Study period: August to September 2003.   Family Planning Clinic of the Hospital Saint‐Francois d'Assise, University Laval in Quebec City, Canada.
Participants	30 pregnant women. Age 18 yrs or older, Gestational age: between 6 and <14weeks, elective abortion.   Exclusion criteria: non french speaker, referal from hospital outside of Quebec City, medical condition requiring preplanned IV sedation, daily use of any illegal drug.
Interventions	Group 1: Hypnosis   Group 2: standard care.   All participants: paracervical block with total of 12ml 0.5% lidocaine on the cervix at 12 o'clock, and at 4 and 8 o'clock at 1‐2cm depth. Vacuum aspiration followed by sharp curettage. Nitrous oxide in oxygen 1: 1 via nose mask per patient request.   Intracervcal laminaria 4‐12 hours prior to surgery for gestational age of 9 weeks or more, stenosis or surgical history of the cervix, age less than 20 zears, gravidity of 5 or more.
Outcomes	Self reported pain and anxiety measured with a 11‐point verbal numerical scale (0 = no pain/not unpleasant at all/not anxious, 10 = the most intense pain/unpleasantness/anxious possible) preoperatively, with insertion of the speculum, with maximum dilation, suction and end of curettage as well as in recovery. Use of N2O (dichotomous variable).
Notes	47 patients eligible, 17 refuse to participate, 30 are randomized. One patient in the hypnosis group was excluded after randomization and before hypnotic intervention since IV sedation had been already scheduled.   Study may be underpowered to detect difference in anxiety and pain. No double‐blinding due to hypnosis, may introduce bias. Inability to differentiate between specific and non‐specific effect of hypnosis. No major complications reported.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

Methods	Randomized controlled trial with additional 2 non‐randomized control groups Outcome assessor of side effects was blinded. Addenbrooke's Hospital, Cambridge, UK.
Participants	60 participants. 40 were randomized to the 2 trial arms. Healthy women. Age 16‐40. Gestational age <12 weeks. Additional, non‐randomized control groups of 10 women each for memory function: 1) Inpatients awaiting minor gynecological surgery who had received no general anesthesia. 2) Ten female nurses
Interventions	Group 1 (trichlorethylen): fentanyl 1.5mcg/kg IV followed by methohexitone 1% 1.5mg/kg with 5ml lignocaine. Nitrous oxide 2 5l and oxygen 3l per minute. Trichloethylene 0.5‐1%. Spontaneous respiration was maintained via coaxial Bain system.   Group 2 (total intravenous group, fentanyl): fentanyl 1.5 mcg/kg IV followed by methohexitone 1% 1.5mg/kg with lignocaine 0.1%. Supplements of methohexitone 0.25mg/kg as needed. Spontaneous respiration was maintained Mary Catterral mask.
Outcomes	Drugs administered, duration of anesthesia, immediate recovery time. Side effects during anesthesia and for 2 hours postoperatively including pain on induction and postoperatively (yes/no), as well as hiccoughs, laryngospasm, vomiting, involuntary muscle movements during anesthesia and nausea, vomiting, shivering, dizziness, headache, drowsiness and tearfulness in recovery. Heart rate and mean systolic blood pressure. Memory function.
Notes	Unclear study length, method of randomization and allocation concealment. Unclear who was blinded other than assessor of side effects. Unclear if suction curettage or sharp curettage was used for the procedure. Given that at that time suction curettage was predominantly used, we decided to include this study. No power analysis. Small sample size. Pain not measured in detail, only yes versus no, and it is not clear if it was self reported.   The author could not be succeessfully contacted to clarify unclear information.   Laryngospasm in one patient in the trichlorethylene group.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Methods	Prospective, randomized, non‐blinded (cannot blind waiting versus non waiting) trial. Per power calculations sample size of 196 needed to detect 1cm difference in pain.   Columbia Willamette Planned Parenthood, Portland, Oregon, USA. Study length: May 2000 to June 2001
Participants	199 pregnant women. Age 18 yrs or older. Gestational age <11wks. Good general health. English language comprehension.   Exclusion criteria: major psychiatric symptoms as well as PID, intravenous sedation or misoprostol.
Interventions	Group 1: 3‐5 minute wait between injection and dilation of the cervix   Group 2: no wait.   All participants: paracervical block (PCB) with total of 12ml 1% buffered lidocaine at 12 (superficially), 4 and 8 o'clock (1‐2cm deep). Diazepam 5mg po and /or fentanyl 100mcg IV as additional pain medication per patient request.   Vacuum aspiration.
Outcomes	Self reported pain and measured with a 10cm visual analog scale with dilation, aspiration and post procedure as well as anticipated pain. Satisfaaction at the end of the procedure (very satisfied, somewhat satisfied, netrual, somewhat dissatisfied, very dissatisfied).
Notes	Per e‐mail communication: randomization with computer generated random numbers in blocks of 50. Allocation concealment with sequentially numbered opaque envelopes.   194 women completed the study and were analysed. 2 excluded for administration of extra sedative medications and 2 lost to follow‐up. One woman chose to discontinue. No major complications reported. Subanalysis per fentanyl use.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate