analgesic options for placement

8/21/2019 Analgesic Options for Placement

http://slidepdf.com/reader/full/analgesic-options-for-placement 1/12

Correspondence: Vasileios Pergialiotis, MD, 80, Vasilisis Sofias Avenue, Athens 115 28, Greece. E-mail: [email protected]

I N T R O D U C T I O N

Fear of pain during and after insertion of an intrauterine

contraceptive may prevent many women from selecting

this method either for fertility regulation or as an

alternative to treat certain gynaecological conditions.

The efficacy of various substances in reducing pain

during this procedure has been studied. Allen et al .

evaluated, in a meta-analysis, interventions that seem

to offer analgesia during IUD insertion1 . Since then,

several additional studies have been published addressing

the efficacy of different analgesics during insertion

of an intrauterine device (IUD) or a levonorgestrel

The European Journal of Contraception and Reproductive Health Care, 2014; 19: 149–160

Analgesic options for placement

of an intrauterine contraceptive:

A meta-analysis

Vasileios Pergialiotis, Dimitrios G. Vlachos, Athanasios Protopappas and Georgios D. Vlachos

1st Department of Obstetrics and Gynaecology, Athens University, Medical School, Alexandra Hospital, Athens, Greece

A B S T R A C T Objective Several randomised controlled trials have been published in the last few years

which evaluated the efficacy of various analgesics in reducing visual analogue (VAS) pain

scores during intrauterine device (IUD) placement. Their results seem to be conflicting andinconclusive.

Methods We searched Medline (1966 –2013), Scopus (2004 –2013), Clinicaltrials.org

(1997 –2013), Popline (1973 –2013), Cochrane CENTRAL (1999 –2013) and Google Scholar

(2004 –2013) engines for published randomised controlled trials, as well as the reference lists

from all electronically retrieved studies.

Results Thirteen studies, involving 1353 women, were finally included in the present

meta-analysis. Among the products used, and with respect to their mode of delivery, only

paracervical lidocaine was effective in producing lower VAS pain scores related to tenaculum

placement (mean difference [MD]: 20.54; 95% confidence interval [CI]: 39.92, 1.15)

and IUD insertion (MD: 28.99; 95% CI: 53.14, 4.84). Misoprostol produced higher

VAS pain scores for the immediate post-insertion period (MD: 2.83; 95% CI: 0.79, 6.45)

and it caused various side effects.Conclusion Paracervical administration of lidocaine prior to IUD insertion reduces VAS

pain scores. In view of the small number of studies assessing its efficacy further studies should

confirm our findings.

K E Y W O R D S Analgesia; Intrauterine device; Lidocaine; Misoprostol; Pain; Visual analogue scale (VAS)

© 2014 The European Society of Contracept ion and Reproductive Health

DOI: 10.3109/13625187.2014.903238



Analgesia for IUD/IUS insertion Pergialiotis et al.

150 The European Journal of Contraception and Reproductive Health Care

releasing-intrauterine system (LNG-IUS)2 –11 . The

administration of lidocaine for analgesia was not assessed

in the aforementioned meta-analysis1 because there

were no randomised trials on record. Furthermore, a

sub-group analysis was needed to investigate whetherthe action of the local anaesthetic depended on its route

of administration. Pharmaceutically-induced cervical

dilatation with a prostaglandin E1 analogue (misopros-

tol) has been claimed to contribute to better visual

analogue scale (VAS) pain scores12 –15 . For this reason

we also included studies addressing VAS among patients

receiving misoprostol.

Aim of the study

We assessed published data pertaining to the efficacy

of various substances (analgesics or not) in prevent-ing pain related to the insertion of an intrauterine

contraceptive.

M E T H O D S

Study design

The present review was designed according to the

PRISMA guidelines16 .

Literature search

We identified relevant papers via Medline (1966 –2013),

Scopus (2004 –2013), Clinicaltrials.org (1997 –2013),

Popline (1973 –2013), Cochrane CENTRAL (1999 –

2013) and Google Scholar (2004 –2013) search engines,

and by screening reference lists of selected studies.

Medline was searched via Pubmed using the

Mesh Terms (‘intrauterine devices’[MeSH Terms]

OR (‘intrauterine’[All Fields] AND ‘devices’[All

Fields]) OR ‘intrauterine devices’[All Fields] OR

(‘intrauterine’[All Fields] AND ‘device’[All Fields]) OR

‘intrauterine device’[All Fields]) AND (‘pain’[MeSH

Terms] OR ‘pain’[All Fields]).

Scopus was also searched using the words (intrauterine

device AND pain). We also searched Clinicaltrials.org for

(intrauterine device pain). Cochrane’s CENTRAL wassearched with two different strategies (intrauterine

device and pain) and (IUD and pain).

An extended search through Google Scholar was

carried out using the terms (intrauterine device, iud,

intrauterine system, pain, tenaculum) in order to ren-

der feasible a manual search. All investigators agreed to

restrict the search to ‘humans’. No language or date

restrictions were applied. Search results are shown in

Figure 1.

Figure 1 Flow diagram representing the method used for article retrieval.




The European Journal of Contraception and Reproductive Health Care 151

Type of studies

Only randomised controlled trials (RCTs) or quasi-

randomised controlled trials were eligible for inclusion

in the study. We assessed the quality of randomisation

by means of the JADAD scale17 . No studies other than

those comparing single medication treatments with

placebo treatment or no treatment were included. Sim-

ilarly, only studies with copper-IUDs and LNG-IUSs

were taken into consideration. Studies that assessed

other intrauterine contraceptives, which are not com-

mercially distributed, or which were withdrawn from

the market, were not included.

Outcome definitions

Primary and secondary outcomes were agreed

upon and predetermined by all authors. Pain outcomes

during tenaculum placement, at IUD insertion and

following insertion were defined as primary outcomes.

Secondary outcomes included physician-perceived

ease of placement, and adverse effects.

Data extraction

Two authors (PV and VD) transferred data abstracted

from each selected article to two structured forms

(Tables 1 and 2) and the other authors reviewed the

data independently in order to increase reliability.

Statistical analysis

For the statistical meta-analysis we resorted to the

RevMan 5.1 software (Copenhagen: The Nordic

Cochrane Centre, The Cochrane Collaboration, 2011).

Meta-analysis was performed for indices presented

among at least four studies. Confidence intervals (CIs)

were set at 95%. With the chi-squared test and the

I-squared test we screened for inconsistency in the results

of the studies. A p value smaller than 0.05 was defined

as indicating statistical significance in the analysis of

heterogeneity. Odds ratios (ORs), mean differences

(MDs) and 95% CIs for all primary and secondary out-

comes were calculated, using the DerSimonian-Laird

random effects model (REM). The random effects

model assumes that heterogeneity exists regarding the

underlying effect among included studies18,19 . In our

analysis we seldom used the REM to estimate out-

comes due to heterogeneity of included studies because

data presented below were originated by independent

researchers, and it would be unlikely that all these

studies would be functionally equivalent. Publication

bias was not assessed due to the small number of

studies retained for the present review, which rendered

its interpretation ambiguous

20

.Indices expressed in median values and range were

transformed into mean and variance (standard devia-

tion) using the formula proposed by Hozo et al .21 .

Ten-scale VAS scores were converted to 100-scale

VAS in order to submit the observed outcomes to

meta-analysis.

R E S U L T S

Included studies

Thirteen studies, involving 1353 women, were finallyretained for the meta-analysis2 –4,6,8 –13,15,22,23 . Study

characteristics and demographic data are presented

in Table 1.

Excluded studies

Seven RCTs we had retrieved were excluded from

the present meta-analysis. One study was not taken

into consideration due to insufficient data5 and two

more due to differences in interpreting findings that

rendered tabulation impossible (data were presented as

discrete variables)24,25 . Two RCTs were not retained

because both the treated- and the control groups addi-

tionally received 100 mg diclofenac14,26 . Guney et al .

investigated the efficacy of lidocaine for removal, not

insertion, of IUDs24 . Massey et al . studied the anal-

gesic effect of naproxen during insertion of Dalkon

Shields, IUDs which for several decennia have not

been available7 .

Characteristics of included studies

and potential bias

Alizadeh et al . enrolled both parous and nulliparous

women and used both copper-IUDs and LNG-IUSs.

Allocation to either intrauterine contraceptive was

based on random sequences formed with computer-

generated programmes and concealed in opaque

envelopes. Lidocaine and lubricant were transferred

to identical sterile containers by a member of the

research team who was not involved in the allocation of

participants or data collection. Administration of either





T a b l e

1

D e s c r i p t i v e c h a r a c t e r i s t i c s o f s t u d i e s a n d p a t i e n t s ( p a t i e n t s

r e c e i v i n g a n a l g e s i a v s . c o n t r o l s )

F i r s t a u t h o r ; y e a r

S t u d y

J a d a d

s c o r e §

S u b s t a n c e

D o s

a g e

M o d e o f d e l i v e r y

N u m b e r o f p a t i e n t s

T r e a t e d v s . c o n t r o l s

A g e , y e a r s *

T y p e o f I U C

( C u - I U D s p e r

t o t a l ) *

A l i z a d e h , 2 0 1 0 1 0

D B - R C T

3

L i d o c a i n e

2 % u n

k n o w n

q u a n t i t y

I n t h e c e r v i c a l c a n a l

6 5

3 1 v s 3 4

2 9 . 1 6 . 7 v s

2 6 . 7 6 . 0

N o t s p e c i fi e d

M a g u i r e , 2 0 1 2 6

D B - R C T

5

L i d o c a i n e

2 % u n

k n o w n

q u a n t i t y


2 0 0

1

0 0 v s 1 0 0

2 7 . 1 6 v s

2 7 . 6 6

6 8 / 1 0 0 v s

7 0 / 1 0 0

M c N i c h o l a s , 2 0 1 2 8

D B - R C T

5

L i d o c a i n e

2 % , 2 - 3 m l


1 9 9

9 9 v s 1 0 0

N / A

2 4 / 9 9 v s

2 3 / 1 0 0

M o d y , 2 0 1 2 9

R C T

2

L i d o c a i n e

1 % , 1 0 m l

P a r a c e r v i c a l b l o c k

5 0

2 6 v s 2 4

3 1 . 9 5 . 9 v s

3 3 . 2 6 . 2

5 / 2 6 v s

7 / 2 4

N e l s o n , 2 0 1 3 1 1

D B - R C T

4

L i d o c a i n e

2 % , 1 . 2 m l

I n s i d e t h e u t e r i n e

c a v i t y

4 0

2 0 v s 2 0

3 2 . 0 6 ( 1 9 – 4 6 ) v s

3 2 . 0 5 ( 2 0 – 4 3 )


C i r i k , 2 0 1 3 2

S B - R C T

2

L i d o c a i n e

1 % , 1 0 m l

P a r a c e r v i c a l b l o c k

6 4

3 4 v s 3 0

2 8 . 5 5 . 7 5 v s

2 7 . 0 5 . 7 5


A l l e n , 2 0 1 3 3

D B - R C T

5

L i d o c a i n e

2 % , 6

m l


1 4 5

7 2 v s 7 3

2 5 . 2 5 . 0 v s

2 6 . 2 5 . 3

1 0 / 7 2 v s

1 0 / 7 3

K a r a b a y i r l i , 2 0 1 2 2 2

D B - R C T

2

T r a m a d o l

1 h b e f o r e

5 0 m g

P e r o s

6 9

3 5 v s 3 4

3 6 4 v s

3 7 6

3 5 / 3 5 v s

3 4 / 3 4

K a r a b a y i r l i , 2 0 1 2 2 2

D B - R C T

2

N a p r o x e n

1 h b e f o r e

5 5 0 m

g

P e r o s

6 8

3 4 v s 3 4

3 5 4 v s

3 7 6

3 4 / 3 4 v s

3 4 / 3 4

C h o r , 2 0 1 2 4

D B - R C T

5

I b u p r o f e n

4 5 m i n b e f o r e

8 0 0 m

g

P e r o s

8 7

4 4 v s 3 7

2 4 . 7 5 . 4 v s

2 7 . 9 6 . 5

0 / 4 4 v s

0 / 3 7

B e d n a r e k , 2 0 1 3 2 3

D B - R C T

5

N i t r o p r u s s i d e g e l

1 0 m g

( 1 m l )


2 4

1 3 v s 1 1

2 4 . 2 4 . 4 v s

2 4 . 4 4 . 4

0 / 1 3 v s

0 / 1 1

D i j k h u i z e n , 2 0 1 1 1 2

D B - R C T

5

M i s o p r o s t o l

3 h b e f o r e

4 0 0 µ g

V a g i n a l

1 9 9

1 0 2 v s 9 7

3 1 . 6 8 . 6 v s

3 0 . 7 8 . 4

1 1 / 1 0 2 v s

9 / 9 7

E d e l m a n 2 0 1 1 1 3

D B - R C T

5


9 0 m i n b e f o r e

4 0 0 µ g

P e r o s

3 5

1 7 v s 1 8

N / A

2 / 1 7 v s

2 / 1 8

S w e n s o n ; 2 0 1 2 1 5

D B - R C T

5


3 – 4 h b e f o r e

4 0 0 µ g

P e r o s o r v a g i n a l

1 0 8

5 4 v s 5 1

2 4 . 6 3 . 8 v s

2 4 . 8 4 . 2

1 1 / 5 4 v s

1 6 / 5 1

I U C , i n t r a u t e r i n e c o n t r a c e p t i v

e ; C u - I U D , c o p p e r r e l e a s i n g - i n t r a u t e r i n e d e v i c e ; D B , d o u b l e b l i n d ; S B , s i n g l e b l i n d ; R C T , r a n d o m i s e d c o n t r o l l e d t r

i a l § T

h e J a d a d

s c o r e a s s e s s e s t h e q u a l i t y o f

c l i n i c a l t r i a l s w i t h r e g a r d t o r a n d o m a s

s i g n m e n t , d o u b l e b l i n d i n g , a n d t h e fl o

w o f p a t i e n t s ; t h e r a n g e o f p o s s i b l e s c o r e s i s 0 ( b a d )

t o 5 ( g o o d ) . * S

t u d y g r o u p v s . c o n t r o l s





substance was performed with cotton swabs. Products

to be given were numbered by allocation sequence.

There were no drop-outs. Results were given in

10-scale VAS scores and their means standard devia-

tions (SDs) were analysed

10

.Also Dijkhuisen et al . included parous- as well as

nulliparous women and used both copper-IUDs and

LNG-IUSs. Randomisation was generated by a com-

puter and was concealed from the investigators. The

200 µ g-misoprostol and placebo tablets had an identi-

cal appearance. Participants were instructed to place

two tablets high up in the vagina three hours prior to

insertion of the intrauterine contraceptive. There were

34 drop-outs in the misoprostol group, and 37 in the

placebo group, before insertion of the intrauterine

contraceptives. Results were presented as 100-scale

VAS scores and analysed as mean SD12 .Edelman et al . enrolled only primiparous women

without a history of prior IUD insertion or of prior

procedure on the cervix. Either a copper-IUD or a

LNG-IUS was fitted. Randomisation was done by

phone. Identical tablets containing either 400 µ g of

misoprostol or placebo were removed from sealed

opaque envelopes and administered to patients via the

buccal route 90 min before the procedure. Three with-

drawals before intervention in the misoprostol- and

two in the control group were recorded. Results were

presented in 100-scale VAS scores and analysed as

mean SD13 .Karabayirli et al . enlisted exclusively parous women

in their study and used only copper-IUDs. Randomi-

sation was computer-generated. The study was

described as having been double-blinded, but the

products administered had a different appearance.

Tramadol 50 mg capsules, naproxen sodium 550 mg

tablets and placebo empty capsules were administered

orally by a physician who was blinded to the study,

one hour prior to insertion of the IUD. No withdraw-

als were recorded. Results were presented in 10-scale

VAS scores and analysed as mean SD22 .

Maguire et al . recruited both parous and nulliparous

women and used both copper-IUDs and LNG-

IUSs. Randomisation was computer-generated and

concealed in opaque envelopes. Lidocaine gel or a

placebo identical in appearance was instilled into a

sterile tube by a research assistant. The product was

transferred with a cotton swab into the cervical canal.

One failed insertion of the intrauterine contraceptive

was described in the lidocaine group, and two in the

control group. All women were included in the final

analysis. Results were presented as 100-scale VAS

scores and analysed as mean SD6 .

McNicholas et al . included parous as well as nullipa-

rous women and used both copper-IUDs and LNG-IUSs. Randomisation was achieved by computer-

generated blocks and was concealed in opaque

envelopes. A physician blinded to the nature of the

product applied 0.5 mL of lidocaine or placebo to

the ectocervix and then instilled 2 –3 mL of the gel

(lidocaine or placebo) into the endocervical canal.

Results were presented as 10-scale VAS scores and

analysed as median range8 .

Mody et al . did not accurately describe exclusion

criteria. They enrolled both nulliparous and parous

women. Both copper-IUDs and LNG-IUSs were

offered as options. Block randomisation with strati-fication by parity was performed. The study was

not blinded, as controls received no treatment. Ten

millilitres of a 1% solution of lidocaine were admin-

istered paracervically. One unsuccessful insertion was

observed in the group treated with lidocaine. Results

were presented as 100-scale VAS-scores and analysed

as mean SD9 .

Swenson et al . enlisted parous as well as nulliparous

women and used both copper-IUDs and LNG-IUSs.

Computer-generated randomisation was performed

with regard to allocation to either misoprostol (400µ g)

or placebo tablets, which were identical in appearance,taste and smell. Participants were instructed to ingest

the tablets or to insert these into the vagina. Allocation

concealment was not described. There were two failed

insertions in the misoprostol group, and three in the

control group. These women were not accounted for

in the final analysis. Results were presented as 100-

scale VAS scores and analysed as mean SD15 .

Chor et al . offered only LNG-IUSs. Computer-

generated randomisation and identical tablets

(containing either 800 mg ibuprofen or placebo) were

given orally and the nature of the product adminis-

tered was concealed from providers and patients.

Three insertions failed in each group. The 10-scale

VAS scores were analysed as mean SD4 .

Nelson et al . recruited both parous and nulliparous

women and used copper-IUDs as well as LNG-IUSs.

Randomisation was done in a 1:1 ratio using random-

generated numbers. The group to which each of the

participants would be allocated was mentioned on slips

of paper which were separately inserted in sequentially





numbered opaque envelopes. A research nurse opened

each patient’s randomisation envelope and placed a

small, but not clearly defined, amount of a 2% lido-

caine solution of lidocaine or saline in a semi-opaque

endometrial aspirator. Both the active compound andthe placebo were instilled into the lower-, the middle-

and the upper third of the endometrial cavity. There

were no withdrawals. Results were presented as

10-scale VAS scores and analysed as mean SD11 .

Bednarek et al . enlisted only nulliparous women

in their study and inserted exclusively LNG-IUSs.

Randomisation was computer-generated. Nitroprusside

gel (1 mL containing 10 mg of the active compound)

or placebo were identical in appearance and allocation

concealment was performed by the hospital’s pharmacy.

Either product was administered into the cervical canal

by means of an angiocatheter. One patient allocated tothe nitroprusside group withdrew from the trial and

was not taken into account in the final analysis. Results

were presented as 100-scale VAS scores and analysed

as mean SD23 .

Cirik et al . provided no information regarding the

type of intrauterine contraceptive they used. Neither

did they describe the methods of randomisation and

of blinding of the procedure. Patients were divided

into three groups according to the product which was

administered paracervically prior to insertion of the

IUD: 10 ml of a 1% solution of lidocaine, 10 ml of a0.9% NaCl solution, or nothing at all. Withdrawals

were not reported. Results were presented as 10-scale

VAS scores and analysed as median range2 .

Allen et al . included both nulliparous and parous

women and used LNG-IUSs as well as copper-

IUDs. Randomisation was done in a 1:1 ratio using

computer-generated numbers3 . The lidocaine gel

(6 ml, 2% concentration) and the placebo (K-Y® jelly)

were identical in appearance. Women and providers

were blinded to the nature of the product that was

instilled into the cervical canal by means of an

angiocatheter, as the latter had been prepared by thehospital’s pharmacy. Two withdrawals were reported

in the placebo arm due to non-insertion of the

intrauterine contraceptive (IUC) and three among

women allocated to the lidocaine group, one due

to non-insertion of the IUC and two because of

protocol violation3 .

Figure 2 Forest plot of the effect of lidocaine on tenaculum placement VAS pain scores. Differences between treated

women and controls were statistically significant only in the case of paracervical block (p 0.04). ‘Intracervical’ in

the cervical canal.





Primary outcomes

The difference in VAS scores dur ing tenaculum place-

ment was not statistically significant for women treated

with lidocaine, regardless of the route of administra-

tion, when compared with controls (722 women,

DerSimonian-Laird random effects model [REM],

mean difference [MD]: 8.33, 95% CI: 18.80, 2.13;

data from seven studies 2,3,6,8,9,11,23 ; Figure 2). How-

ever, when we stratified the studies according to the

mode of delivery of lidocaine we found significantly

decreased pain VAS scores among women treated with

paracervical lidocaine (REM, MD: 20.54, 95% CI:

39.92, 1.15).

Similar results were also observed for IUD inser-

tion following administration of lidocaine: the local

anaesthetic did not significantly diminish VAS painscores when analysed irrespective of the mode of

administration (787 women, REM, MD: 10.93, 95%

CI: 23.68, 1.82; data from eight studies2,3,6,8 –11,23 ;

Figure 3). However, once more, when we stratified

studies per mode of administration a significant effect

was observed among women treated with

paracervical lidocaine (REM, MD: 28.99, 95% CI:

53.14, 4.84).

With regard to misoprostol, VAS pain scores during

IUD insertion among treated women did not signifi-

cantly differ from those of controls (339 women,REM, MD: 2.83, 95% CI: 0.79, 6.45; data from

three studies12,13,15 ; Figure 4).

VAS pain scores related to the pain felt after the

IUD insertion significantly differed only in the case of

misoprostol (140 women, REM, MD: 7.15, 95% CI:

3.61, 10.68; data from two studies13,15 ; Figure 5).

Secondary outcomes

Physician-perceived insertion of the IUD seemed to

be significantly easier among treated women (363

women, REM, MD:

2.29, 95% CI:

3.60,

0.98;data from four studies12,13,15,23 ; Figure 6). However, in

subgroup analysis this effect seemed to remain signifi-

cant only in the case of misoprostol (Figure 6).

Adverse effects were underreported in the included

studies and were assessed using different definitions,

hence they were excluded from analysis.

Figure 3 Forest plot of the effect of lidocaine on VAS pain scores related to insertion of the intrauterine contraceptives.

Differences between treated women and controls were statistically significant only in the case of paracervical block

(p 0.02). ‘Intracervical’ in the cervical canal.





D I S C U S S I O N

Fear of pain prevents many women from selecting

intrauterine contraceptives for birth control. Pain is

usually experienced during insertion of the speculum

into the vagina, cervical tenaculum placement, andinsertion of the device. Of these procedures, tenacu-

lum placement and insertion of the IUD seem to hurt

most: they are associated with higher VAS scores than

speculum insertion.

Findings and interpretation

In the present meta-analysis we evaluated the impact of

lidocaine in reducing pain VAS scores during placement

of the tenaculum on the cervix and found that it seems

to be effective only when it is injected paracervically;

this also applied to IUD insertion. On the contrary,

instillation of lidocaine, whether into the cervical canal

or in the endometrial cavity, failed to significantly lower

VAS scores of experienced pain.

Misoprostol is not an analgesic. However, its poten-

tial to reduce pain and facilitate IUD insertion could

be achieved by softening and ripening the cervix. Our

results, however, clearly indicate that it is inefficacious

in that respect.

With regard to lidocaine and nitroprusside, VAS

pain scores related to the pain felt after IUD insertion

of treated women did not significantly differ from

Figure 4 Forest plot of the effect of misoprostol and placebo on VAS pain scores related to insertion of the intrauterine

contraceptives. VAS scores of women treated with misoprostol were significantly higher than those of controls

(p 0.001).

Figure 5 Forest plot of VAS pain scores related to the period immediately following insertion of the intrauterine

contraceptive. The mean VAS score was significantly higher among misoprostol-treated women (p 0.001).





those of controls (Figure 4). However, in the case of

misoprostol, pain scores were significantly higher

among treated women. This could be explained by the

uterine contractions that misoprostol elicits, thus aug-

menting the pain experienced in the post-insertion

period.

We would have liked to perform a meta-analysis

of the actual cervical dilatation that misoprostol pro-

duces, in order to clarify whether misoprostol failed

to reduce the pain scores due to insufficient ripeningof the cervix. However, the studies included in the

present review did not evaluate this parameter, but

rather the actual number of patients who required

additional mechanical dilatation4,9,13,15 (Table 2). Pre-

viously, however, cervical dilatation after treatment

with misoprostol was assessed with 4 mm Hegar dila-

tors by Saav et al . in a RCT comparing patients

(n 39) given misoprostol (400 µ g) and diclofenac

(100 mg) orally and control patients (n 40) who

had received only diclofenac26 . No significant differ-

ences were found between the two groups. Similar

results were also reported recently by Ibrahim et al .

among women who were randomly allocated to

receive either 400 µ g misoprostol and 100 mg

diclofenac sublingually (n 130) or only 100 mg

diclofenac (n 125)14 . The findings of these two

studies are intriguing14,26 ; they are not in line with

the results of a previous meta-analysis comparing

cervical dilatation among women undergoing hyst-

eroscopy who received either misoprostol (via the

oral-, the sublingual- or the vaginal route) (n 264)

or were left untreated (n 272)27 . In this latter meta-

analysis researchers concluded that cervical dilatation

was significantly greater among treated women

(mean difference of 2.47 mm, 95% CI: 1.81 –3.13;

p 0.001)27 .

According to our findings, physician-perceived ease

of placement of IUDs was significantly greater in

women pre-treated with misoprostol. Two more trials,

excluded from the present review, provided controver-sial results regarding the effect of misoprostol in this

respect14,26 .

Strengths and weaknesses of the study

Our study’s main strength was the comparison of

groups that received treatment as monotherapy, thus

avoiding potential bias arising from co-administered

medications. In all included studies subjects allocated

to the control groups received either placebo or no

therapy at all. Furthermore, we were able to stratify

studies assessing the effect of lidocaine according to

the route of its administration. This way we observed

that its analgesic activity did not always achieve the

expected decrease in VAS pain scores.

On the other hand, discrepancies in scaling of

visual analogue scores among included studies

required the conversion of 10-scale VAS to 100-scale,

in order to analyse and interpret results. For the same

reason values of median and range were converted to

Figure 6 Forest plot of physician-perceived ease of placement. Differences between groups were not statistically

significant (p 0.78).





T a b l e

2

V A S s c o r e s a n d c e r v

i c a l d i l a t a t i o n , s t u d y g r o u p v s . c o n t r o l s .

F i r s t a u t h o r ; y e a r

P a i n a

t I U D / I U S i n s e r t i o n

S p e c u l u m p l a c e m e n t

T e n a c u l u m p l a c e m e n t

C e r v i c a l

d i l a t a t i o n

E a s e o f

p l a c e m e n t

P a i n a f t e r i n s e r t i o n

A l i z a d e h , 2 0 1 0 1 0

3 . 4 1 . 9 v s

3 . 4 1 . 7

N / A

N / A

N / A

N / A

N / A

M a g u i r e , 2 0 1 2 6

5 1 . 0 3

1 v s

5 0 . 9 3

2

N / A

3 5 . 4 2

6 v s

3 4 . 3 2

5

N / A

N / A

N / A

M c N i c h o l a s , 2 0 1 2 8

5 ( 0 – 1 0 ) v s

6 ( 0 – 1 0 )

N / A

4 ( 0 – 1 0 ) v s

4 ( 0 – 1 0 )

N / A

N / A

N / A

M o d y , 2 0 1 2 9

3 7 . 3 3

5 v s

5 2 . 5 2

7 . 5

1 3 . 8 1 3 .

1 v s

1 2 . 9 1 2 .

9

1 8 . 8 2

1 . 8 v s

2 9 . 0 1

6 . 2

2 / 2 6 v s

2 / 2 4

N / A

2 0

. 5 2

5 . 7 v s

1 9

. 4 1

8 . 4

N e l s o n , 2 0 1 3 1 1

2 . 9 5 2 . 6 1 v s

3 . 6 8 2 . 7 1

N / A

2 . 2 0 2 . 4 2 v s

2 . 7 0 2 . 0 3

N / A

N / A

2 . 2 v s

1 . 6 3

C i r i k , 2 0 1 3 2

2 0 1

2 . 5 v s

6 0 1

2 . 5

N / A

4 0 1

5 v s

7 0 1

2 . 5

N / A

N / A

1 0 1

0 v s

4 0 1

2 . 5

A l l e n , 2 0 1 3 3

3 5 . 2 2

7 . 7 v s

3 6 . 7 3

0

2 3 . 5 2 3 .

8 v s

2 1 . 5 2 3 .

7

3 7 . 5 2

6 . 2 v s

4 1 . 6 3

1 . 5

N / A

N / A

2 0

. 6 2

4 v s

2 1 . 4 2

5 . 2

K a r a b a y i r l i , 2 0 1 2 2 2

2 . 3 1 0 . 6 0 v s

4 . 8 8 1 . 0

N / A

N / A

N / A

N / A

N / A

K a r a b a y i r l i l , 2 0 1 2 2 2

2 . 9 4 0 . 7 1 v s

4 . 8 8 1 . 0

N / A

N / A

N / A

N / A

N / A

C h o r , 2 0 1 2 4

3 6 . 9 3

4 v s

3 3 . 4 2

7

N / A

3 8 . 6 2

9 v s

3 8 . 1 3

0

2 / 4 4 v s

1 / 3 7

N / A

N / A

B e d n a r e k , 2 0 1 3 2 3

6 1 2

5 v s

7 4 1

8

2 3 2 2

v s

2 6 1 3

3 0 2

7 v s

4 0 2

7

N / A

3 2 . 4 2

2 . 7 v s

2 6 . 5 2

7 . 2

3 7 3

3 v s

5 0 2

1

D i j k h u i z e n , 2 0 1 1 1 2

4 6 2

8 v s

4 0 2

7

N / A

N / A

N / A

2 . 9 2 . 8 v s

2 . 8 2 . 6

N / A

E d e l m a n , 2 0 1 1 1 3

6 5 2

1 v s

5 5 2

1

1 7 2 2

v s

9 8

3 5 2

5 v s

3 6 2

0

0 / 1 7 v s

3 / 1 8

2 4 1

9 v s

2 9 2

1

2 0 2

3 v s

2 0 1

9

S w e n s o n , 2 0 1 2 1 5

5 8 . 4 3 . 3 v s

5 6 . 9 3 . 0

N / A

N / A

5 / 9 v s

5 / 1 0

2 5 3 . 5 v s

2 7 . 4 3 . 5

3 5

. 1 3 . 4 v s

2 7 . 5 2 . 4

I U D , i n t r a u t e r i n e d e v i c e ; I U S ,

i n t r a u t e r i n e s y s t e m ; N / A , n o t a p p l i c a b l e .





mean and variance according to the formula pro-

posed by Hozo et al .21 . Publication bias was precluded

due to the low number of included studies20 . Sam-

pling bias and in particular exclusion bias was observed

among certain studies, affecting the external validityof comparisons, thus precluding the analysis based on

the fixed-effects model. Several factors related to

lidocaine treatment, such as route of administration

(into the cervical canal, intrauterine, paracervical),

mode of administration (cotton swab, angiocatheter)

and quantity of applied lidocaine could also have

contributed to the observed discrepancies between

studies (Table 1). As already mentioned, in the case

of misoprostol, the route of administration (per os or

vaginal) and interval of administration prior to IUD

placement differed (Table 1).

Differences in results in relation

to previous studies

Allen et al . concluded in their previous meta-analysis

that NSAIDS failed to reduce pain experienced dur-

ing IUD insertion1 . Their study, however, had limita-

tions with regard to the interpretation of the data due

to the small number of included studies. In one of

the latter, the IUD that was fitted, the Dalkon Shield,

is no longer available on the market and its size and

cumbersome insertion technique do not allow the

comparison of the pain or discomfort it caused whenplaced, to those observed with the currently utilised

LNG-IUSs or copper IUDs7 . In the study assessing

misoprostol, which Allen et al . took into consider-

ation, both the study and the control groups received

100 mg diclofenac26 whose analgesic action in all

likelihood blurred the interpretation of the facilitat-

ing effect the PGE1 analogue might have had1 . All

things considered, lidocaine seems to be the only

compound eliciting analgesia when administered

before IUD insertion.

Implications for clinicians

We evaluated VAS pain scores among women being

fitted with an intrauterine contraceptive in accordance

with a standardised procedure that included insertion

of a speculum, use of an antiseptic applied on the

cervix, placement of the tenaculum, and insertion of

the device. The closing of the tenaculum – unless slowly

and smoothly done over some four to five seconds

- is painful. Therefore, in many institutions, tenacula

are seldom used when insertion can be accomplishedwithout their assistance. The present meta-analysis

could not determine whether analgesia significantly

reduces VAS pain scores among women in whom

the device can be fitted without grasping the cervix

with an instrument. But our study does show that,

in cases where use was made of a tenaculum, only

paracervically-injected lidocaine effectively reduces

pain during placement of that instrument and

IUD/IUS insertion. However, firm conclusions are pre-

cluded as only two studies in the present meta-analysis

assessed the effect of paracervically-administered

lidocaine on VAS pain scores.

Future research

Later studies might examine the effect of lidocaine

(or other local anaesthetics) administered paracervically

to clarify whether it should become a standard prac-

tice prior to insertion of an intrauterine contraceptive.

One may contemplate assessing whether analgesia is

necessary when an IUD can be placed without the

use of a tenaculum.

C O N C L U S I O N

Of the various substances assessed, only lidocaine,

when injected paracervically, reduces VAS pain

scores related to tenaculum placement and IUD

insertion. However, firm conclusions are precluded

due to the small number of the included studies.

Misoprostol administration is associated with signifi-

cantly higher VAS pain scores after the completion of

the procedure.

Funding: None.

Declaration of interest: The authors report no con-

flicts of interest. The authors alone are responsible for

the content and the writing of the text.





R E F E R E N C E S

Allen RH, Bartz D, Grimes DA,1. et al . Interventions

for pain with intrauterine device insertion. Cochrane

Database Syst Rev 2009;8:CD007373.

Cırık DA, Ta kın EA, Tu lu A,2. et al . Paracervical blockwith 1% lidocaine for pain control during intrauterine

device insertion: A prospective, single-blinded, control-

led study. Int J Reprod Contracept Obstet Gynecol

2013;2:263 –7.

Allen RH, Raker C, Goyal V. Higher dose cervical 2%3.

lidocaine gel for IUD insertion: A randomized control-

led trial. Contraception 2013;88:730 –6.

Chor J, Bregand-White J, Golobof A,4. et al . Ibuprofen

prophylaxis for levonorgestrel-releasing intrauterine

system insertion: A randomized controlled trial. Contra-

ception 2012;85:558 –62.

Hubacher D, Reyes V, Lillo S,5. et al . Pain from copper

intrauterine device insertion: Randomized trial of

prophylactic ibuprofen Am J Obstet Gynecol 2006;

195:1272 –7.

Maguire K, Davis A, Rosario LT,6. et al . Intracervical lido-

caine gel for intrauterine device insertion: A randomized

controlled trial. Contraception 2012;86:214 –9.

Massey SE, Varady JC, Henzl MR. Pain relief with7.

naproxen following insertion of an intrauterine device.

J Reprod Med 1974;13:226 –31.

McNicholas CP, Madden T, Zhao Q,8. et al . Cervical lido-

caine for IUD insertional pain: A randomized controlled

trial. Am J Obstet Gynecol 2012;207:384 e1 –6.

Mody SK, Kiley J, Rademaker A,9.et al

. Pain controlfor intrauterine device insertion: A randomized trial of

1% lidocaine paracervical block. Contraception 2012;

86:704 –9.

Mohammad-Alizadeh-Charandabi S, Seidi S, Kazemi F.10.

Effect of lidocaine gel on pain from copper IUD inser-

tion: A randomized double-blind controlled trial. Indian

J Med Sc 2010;64:349 –55.

Nelson AL, Fong JK. Intrauterine infusion of lidocaine11.

does not reduce pain scores during IUD insertion.

Contraception 2013;88:37 –40.

Dijkhuizen K, Dekkers OM, Holleboom CA,12. et al .

Vaginal misoprostol prior to insertion of an intrauterine

device: An RCT. Hum Reprod 2011;26:323 –9.Edelman AB, Schaefer E, Olson A,13. et al . Effects of

prophylactic misoprostol administration prior to

intrauterine device insertion in nulliparous women.

Contraception 2011;84:234 –9.

Ibrahim ZM, Sayed Ahmed WA. Sublingual misoprostol14.

prior to insertion of a T380A intrauterine device in

women with no previous vaginal delivery. Eur J

Contracept Reprod Health Care 2013;18:300 –8.

Swenson C, Turok DK, Ward K,15. et al . Self-administered

misoprostol or placebo before intrauterine device inser-tion in nulliparous women: A randomized controlled

trial. Obstet Gynecol 2012;120:341 –7.

Liberati A, Altman DG, Tetzlaff J,16. et al . The PRISMA

statement for reporting systematic reviews and

meta-analyses of studies that evaluate healthcare inter-

ventions: Explanation and elaboration. BMJ 2009

21;339:b2700

Jadad AR, Moore RA, Carroll D,17. et al . Assessing the

quality of reports of randomized clinical trials: Is blind-

ing necessary? Control Clin Trials 1996;17:1 –12.

DerSimonian R, Laird N. Meta-analysis in clinical trials.18.

Control Clin Trials 1986;7:177 –88.

DerSimonian R, Kacker R. Random-effects model for19.

meta-analysis of clinical tr ials: An update. Contemp Clin

Trials 2007;28:105 –14.

Ioannidis JP, Trikalinos TA. The appropriateness of20.

asymmetry tests for publication bias in meta-analyses:

A large survey. CMAJ 2007;176:1091 –6.

Hozo SP, Djulbegovic B, Hozo I. Estimating the mean21.

and variance from the median, range, and the size of a

sample. BMC Med Res Methodol 2005;5:13.

Karabayirli S, Ayrim AA, Muslu B. Comparison of the22.

analgesic effects of oral tramadol and naproxen sodium

on pain relief during IUD insertion. J Minim Invasive

Gynecol 2012;19:581 –4.Bednarek PH, Micks EA, Edelman AB,23. et al . The effect

of nitroprusside on IUD insertion experience in

nulliparous women: A pilot study. Contraception 2013;

87:421 –5.

Scavuzzi A, Souza AS, Costa AA, Amorim MM.24.

Misoprostol prior to inserting an intrauterine device in

nulligravidas: A randomized clinical trial. Hum Reprod

2013;28:2118 –25.

Jensen HH, Blaabjerg J, Lyndrup J. [Prophylactic use of25.

prostaglandin synthesis inhibitors in connection with

IUD insertion]. Ugeskr Laeger 1998;160:6958 –61.

Saav I, Aronsson A, Marions L,26. et al . Cervical priming

with sublingual misoprostol prior to insertion of anintrauterine device in nulliparous women: A randomized

controlled trial. Hum Reprod 2007;22:2647 –52.

Polyzos NP, Zavos A, Valachis A,27. et al . Misoprostol prior

to hysteroscopy in premenopausal and post-menopausal

women. A systematic review and meta-analysis. Hum

Reprod Update 2012;18:393 –404.

analgesic options for placement

Documents