You are on page 1of 8

REVIEW

CURRENT
OPINION Endometrial ‘scratching’: what the data show
Xavier Santamaria a, Nora Katzorke a,b, and Carlos Simón a,c

Purpose of review
Since its first description in 2003, the endometrial scratching procedure has been the topic of over 1000
studies. This procedure, used to improve endometrial receptivity for assisted reproduction, is accessible –
any gynecologist can easily perform it – and has been adapted into clinical routine by some reproductive
units. However, the available data are controversial, and no biological plausibility exists to support the use
of this intervention. This study aims to critically review the existing data, focusing on the last 2 years,
regarding the efficiency of endometrial scratching.
Recent findings
A total of five randomized controlled studies, one meta-analysis, and a systematic review related to
endometrial scratching/injury were published in 2014 and 2015. Considerable heterogeneity exists
among these studies regarding the selected population, type of treatment, and even timing and devices
used to perform the endometrial injury. Importantly, none of these studies reported improved reproductive
outcomes in terms of live birth rates following endometrial scratching.
Summary
Overall, data from properly designed and powered randomized controlled studies demonstrate no
beneficial effect of this intervention that is based on unknown biological effects. Endometrial scratching
produces pain, costs money, and the side-effects of systematic scratching in the production of Asherman
syndrome remain to be seen. Think before scratching.
Keywords
endometrial receptivity, endometrial scratching, meta-analyses, randomized controlled trials, recurrent
implantation failure

INTRODUCTION this topic, no consensus exists about what scratch-


Endometrial scratching or injury of the endometrial ing is, how to perform it, for how many times, when,
lining, performed to improve assisted reproduction or to whom.
outcomes, has attracted increasing attention since a The type of endometrial manipulation applied
2003 study reported that this procedure improves to produce scratching includes endometrial manip-
implantation rates [1]. Accordingly, 15 randomized ulation during hysteroscopy alone [23] or in
&&
controlled trials (RCTs) [2–4,5 ,6–16] (Table 1) and addition to endometrial biopsy [2,15] or curettage
&&
&&
five meta-analyses [17,18 ,19–21] have since cov- [12], as well as endometrial biopsy alone [3,4,5 ,6–
ered the procedure. In an earlier review, data pub- 11,13,14,16]. During endometrial biopsy, different
lished through 2013 were thoroughly analyzed types of catheters have been used – Pipelle de Cor-
&&
&
[22 ], prompting the conclusion that endometrial nier (Laboratoire CCD, Paris, France) [3,5 ,6,
scratching before an IVF cycle did not improve 8–11,13,15,16], Novak curette (CooperSurgical,
implantation rates or live birth rates (LBR). In the
present work, we critically analyze RCTs and meta-
a
analyses published on this topic during the last Department of Obstetrics and Gynecology, Universidad de Valencia,
Instituto Universitario IVI, INCLIVA, Valencia, Spain, bDepartment of
2 years, to enable an evidence-based medical
Obstetrics and Gynecology, Heinrich Heine University Medical Center,
opinion about the utility of this intervention. Duesseldorf, Germany and cDepartment of Obstetrics and Gynecology,
Stanford University, Stanford, California, USA
Correspondence to Carlos Simón, Parc Cientific, Universidad de Valén-
WHAT IS ENDOMETRIAL SCRATCHING? cia, C/Catedrático Agustin Escardino No. 9, Paterna 46980, Spain.
Endometrial scratching applies intentional damage Tel: +34639125447; e-mail: carlos.simon@ivi.es
to the endometrial lining to improve endometrial Curr Opin Obstet Gynecol 2016, 28:242–249
receptivity. However, in the published articles about DOI:10.1097/GCO.0000000000000279

www.co-obgyn.com Volume 28  Number 4  August 2016

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


Endometrial ‘scratching’: what the data show Santamaria et al.

hormonally regulated, independent of any mech-


KEY POINTS anical stimulus, and the presence or absence of
 Overall, data from properly designed and powered conception. Second, it is assumed that local endo-
RCTs demonstrate no beneficial effect of endometrial metrial injury facilitates implantation via aseptic
scratching that is based on unknown biological effects. inflammation [25]. A wound healing process may,
therefore, be triggered by the secretion of inflam-
 Meta-analyses of ‘everything’ create confusion that can
matory factors, such as cytokines, growth factors,
promote spurious associations by the use of inadequate
or insufficient data to compute incorrect or incomplete interleukins, macrophages, and dendritic cells,
conclusions. which subsequently support embryo implantation.
No peer-reviewed publications have yet supported
 Endometrial scratching is an intervention that produces this hypothesis. Finally, endometrial scratching
pain, costs money, and the side-effects of systematic
may retard endometrial maturation, which is abnor-
scratching in the production of Asherman syndrome
remain to be seen. mally advanced during the controlled ovarian
stimulation performed for assisted reproductive
techniques (ART). Again, however, no data have
substantiated this hypothesis.
Trumbull, USA) [12,14], Tao Brush (Cook Incorpor-
ated, Bloomington, USA) [7], Karman’s cannula
(Narang Medical Limited, New Delhi, India) [4], or RANDOMIZED CONTROLLED TRIALS AND
forceps [2] – but there has been no indication of META-ANALYSES OF ENDOMETRIAL
where in the uterine cavity the procedure was per- SCRATCHING, 2014–2015
formed. Further, endometrial tissue obtained from &&
Since 2003, 15 RCTs [2–4,5 ,6–16] and five meta-
the curettages has never been analyzed to determine &&
analyses [17,18 ,19–21] have been published on
how deep the ‘intervention’ reached. The number of endometrial scratching. Our assessment of the 10
&&
scratchings performed varied – one [2,4,5 ,6–10, RCTs and three meta-analyses published until 2013
12–14,16] or two [3,11,15] – as did the timing of the concluded that endometrial scratching had no
procedure within the menstrual cycle – during the beneficial effect in reproductive outcome when
follicular phase [2,6,7,10,12], during the luteal &
applied in patients [22 ]. In the present work, we
&&
phase [3,4,5 ,8,9,13,14,16], or during both phases analyze RCTs and meta-analyses published on this
of the same cycle [11,15]. Additionally, variation topic during 2014 and 2015.
exists in the length of time from endometrial The 2015 updated Cochrane systematic review
scratching to embryo transfer – it has been per- on endometrial injury suggests a potential positive
&&
formed in the prior cycle [2–4,5 ,6–13,15,16] or effect of endometrial scratching on LBR for women
in the same cycle [7,14]. The study populations have with RIF undergoing assisted reproduction [17]. In
included complete heterogeneous infertile popu- contrast, another recent systematic literature review
lations undergoing different treatments, ranging &&
by Panagiotopoulou et al. [18 ] revealed that a
from programmed intercourse [8,10] and intrauter- meta-analysis could not be performed because of
&&
ine insemination (IUI) [2,6,7], to first IVF cycle [5 ], significant differences in study quality, population,
patients with at least one IVF failure [3,13,14], and and intervention characteristics among the studies
patients with more than one IVF failure [4,15], considered; the authors concluded that insufficient
or with recurrent implantation failures (RIF) evidence is currently available to support the use of
[11,12,16]. Despite this great number of variables, endometrial scratching in women with RIF under-
endometrial scratching is considered a single inter- going ART. How can the same data lead to such
vention that has undergone five meta-analyses. This contradictory conclusions? Let’s go to the source.
fact deserves a reflection by the reader. The most recent RCT analyzed the effect of
endometrial scratching in 387 normoresponders
under age 40 with at least one previous IVF failure
THE BIOLOGICAL PLAUSIBILITY OF [3]. An endometrial biopsy was obtained twice, on
ENDOMETRIAL SCRATCHING day 21 and on day 23–24 of the cycle, using a Pipelle
The biological responses induced by endometrial de Cornier; IVF, in which either one blastocyst or
scratching remain uncertain and unproven. Several two day 3 embryos were transferred, was performed
hypotheses are proposed. First, endometrial scratch- in the following cycle. In the control group, an
ing might induce stromal cell decidualization, intracervical manipulation with the introduction
which increases the probability of embryo implan- of a probe through the cervix to the internal ovarian
tation [24]. However, unlike in rodents and stimulation was performed as a sham control.
other mammals, decidualization in humans is Additionally, if any difficulty was noticed during

1040-872X Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-obgyn.com 243

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


244
Table 1. Summary of all randomized clinical trials assessing the efficiency of endometrial scratching, including a rating regarding the study design

Timing of
Sample Inclusion Device endometrial Outcome because of Study
Journal, IF size criteria used scratching ART endometrial scratching design

Gibreel et al. 2015 Endometrial scratching for Gynecol n ¼ 387 Normoresponders Pipelle de Cornier Two interventions: day IVF No statistically significant
women with previous IVF Endocrinol 1.33 <40 years 21 and day 23–24 increase in LBR or CPR
failure undergoing IVF 1 failed IVF cycle of menstrual cycle
treatment prior to IVF
Singh et al. 2015 Does endometrial injury JHRS 0 n ¼ 60 Normoresponders Karman’s cannula Day 14–21 of ICSI Implantation rates
Reproductive endocrinology

www.co-obgyn.com
enhances implantation in <35 years menstrual cycle prior significantly higher
recurrent in-vitro fertilization >1 failed IVF cycle to IVF LBR not significantly higher
failures? A prospective No significant difference in
randomized control study OPR, AR, and
from tertiary care center miscarriage rate
El-Khayat et al. 2015 Comparing the effect of office EJOG 1.695 n ¼ 332 Normoresponders Hysteroscopy and Day 4–7 of menstrual IUI No significant difference in
hysteroscopy with <35 years scratching with forceps cycle prior to IUI CPR, AR, MPR, and LBR
endometrial scratch versus Regular menstrual
office hysteroscopy on cycles
intrauterine insemination BMI < 32 kg/m2
outcome: a RCT Mild male factor
Unexplained
infertility
Zarei et al. 2014 The effects of endometrial Iran J Reprod n ¼ 144 18–40 years Pipelle de Cornier Day 6–8 of menstrual IUI Comparable pregnancy
injury on intrauterine Med 0.188 Mild male factor cycle prior to IUI rate, AR, OPR, and
insemination outcome: a Unexplained endometrial thickness
RCT infertility E2 significantly lower in
Mild endometriosis endometrial biopsy
group
Yeung et al. 2014 The effect of endometrial injury Hum n ¼ 300 Unselected Pipelle de Cornier Day 21 (7 days after IVF or ICSI No significant difference in
on ongoing pregnancy rate Reprod 4.569 subfertile women LH surge in OPR, implantation rates,
in unselected subfertile Normal uterine ovulatory women) of CPR, miscarriage rate,
women undergoing in vitro cavity menstrual cycle prior and MPR
fertilization: a RCT demonstrated by to IVF In multivariate regression
saline infusion analysis, women’s age
sonogram or was the only factor
hysteroscopy significantly affecting
OPR in women
undergoing first IVF

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


cycle
In multivariate regression
analysis, the number of
top quality embryos
transferred was the only
significant factor in
women having repeated
cycles of IVF

Volume 28  Number 4  August 2016


Table 1 (Continued)

Timing of
Sample Inclusion Device endometrial Outcome because of Study
Journal, IF size criteria used scratching ART endometrial scratching design

Nastri et al. 2013 Endometrial scratching Ultrasound Obstet n ¼ 158 General population Pipelle de Cornier Day 14–21 of IVF LBR, CPR, pain score,
performed in the nontransfer Gynecol 3.853 of infertile menstrual cycle prior endometrial
cycle and outcome of women to IVF vascularization index,
assisted reproduction: a RCT <38 years and VFI were
significantly higher
No significant difference
regarding miscarriage
rate, MPR, endometrial
thickness, endometrial
volume, and flow index
Parsanezhad et al. Pregnancy rate after Iran J Reprod n ¼ 217 Normoresponders Pipelle de Cornier Day 13 (the day of timed intercourse Significant difference
2013 endometrial injury in Med 0.188 Infertility duration detecting urinary LH regarding pregnancy
couples with unexplained of 2–5 years surge) of menstrual rate and OPR
infertility: a RCT 23–35 years cycle prior to No significant difference
BMI of 18–25 kg/ intercourse regarding AR
m2
Gibreel et al. 2013 Endometrial scratching to J Obstet Gynaecol n ¼ 105 Unexplained Pipelle de Cornier Days 21–26 of None BPR and CPR were
improve pregnancy rate in Res 1.072 infertility >1 menstrual cycle prior significantly higher
couples with unexplained year to conceiving
subfertility: a RCT 20–39 years

1040-872X Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved.
Cycles of 22–34
days
Abdelhamid et al. The success rate of pregnancy Arch Gynecol n ¼ 150 Unexplained Tao Brush Days 8–9 of the IUI Pregnancy percentages
2013 in IUI cycles following Obstet 1.364 infertility menstrual cycle prior were higher
endometrial sampling. A Normoresponders to IUI or days 8–9
RCT 22–35 years of the same IUI cycle
Baum et al. 2012 Does local injury to the Gynecol n ¼ 36 Normoresponders Pipelle de Cornier Two interventions: days IVF or ICSI Implantation rates and LBR
endometrium before IVF Endocrinol 1.33 18–41 years 9–12 and days 21– were not statistically
cycle really affect treatment RIF (3 embryo 24 of the menstrual different
outcome? Results of a transfer) cycle prior to IVF CPR was significantly
randomized placebo Fresh embryo lower
controlled trial transfer in the
following cycle
Shohayeb et al. Does a single endometrial EJOG 1.695 n ¼ 210 RIF (>2 failed IVF/ Hysteroscopy and Days 4–7 of menstrual ICSI Statistically higher
2012 biopsy regimen (S-EBR) ICSI cycles scratching with Novak cycle prior to ICSI implantation rates, CPR,
improve ICSI outcome in <39 years curette and LBR

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


patients with repeated Endometrium <5 AR not significantly
implantation failure? A RCT mm on day 4 of different

www.co-obgyn.com
menstrual cycle
Endometrial ‘scratching’: what the data show Santamaria et al.

245
246
Table 1 (Continued)

Timing of
Sample Inclusion Device endometrial Outcome because of Study
Journal, IF size criteria used scratching ART endometrial scratching design

Safdarian et al. 2011 Local injury to the endometrium Iran J Reprod n ¼ 100 Normoresponders Pipelle de Cornier Day 21 of the IVF No significant differences
does not improve the Med 0.188 20–39 years menstrual cycle prior in implantation rates,
implantation rate in good FSH < 11 to IVF CPR, and BPR
responder patients Women with PCO
undergoing IVF
Reproductive endocrinology

www.co-obgyn.com
Karimzade et al. 2010 Local injury to the endometrium Arch Gynecol n ¼ 156 First IVF cycle Novak Day 17 of same IVF IVF or ICSI Implantation rates, CPR,
on the day of oocyte Obstet 1.07 <38 years cycle and OPR were
retrieval has a negative BMI > 19/<30 significantly lower
impact on implantation in kg/m2
assisted reproductive cycles:
a RCT
Narvekar et al. 2010 Does local endometrial injury J Hum Reprod n ¼ 100 Normoresponders Hysteroscopy and Two interventions: days IVF or ICSI Implantation rates, CPR,
in the nontransfer cycle Sci 0 1 failed IVF cycle scratching with Pipelle 7–10 plus and LBR were
improve the IVF-embryo <37 years de Cornier hysteroscopy and significantly higher
transfer outcome in the days 24–25 of the
subsequent cycle in patients menstrual cycle prior
with previous unsuccessful to IVF
IVF? A randomized
controlled pilot study.
Karimzadeh et al. 2009 Endometrial local injury Aust N Z J Obstet n ¼ 115 20–40 years Pipelle de Cornier Days 21–26 of the IVF Implantation rates and CPR
improves the pregnancy rate Gynaecol 1.31 RIF (2–6 failed menstrual cycle prior were significantly higher
among recurrent ART) cycles to IVF No difference in
implantation failure patients No history of blood miscarriage rate
undergoing in vitro diseases
fertilisation/intra
cytoplasmic sperm injection:
a randomised clinical trial

A ‘green happy smiley’ ( ) indicates a well conducted study with a proper study design, clearly defined inclusion criteria, given common primary and secondary study outcomes, and an a priori sample size
calculation. Studies marked with a ‘red sad smiley’ ( ) do not take all these requirements into account.
AR, abortion rate; ART, artificial reproductive techniques; BPR, biochemical pregnancy rate; COS, controlled ovarian stimulation; CPR, clinical pregnancy rate; E2, estradiol; FSH, follicle stimulating hormone; ICSI,
intracytoplasmic sperm injection; IF, impact factor; IUI, intrauterine insemination; LBR, live birth rate; LH, luteinizing hormone; MPR, multiple pregnancy rate; OCP, oral contraceptive pill; OPR, ongoing pregnancy rate;
PCO, polycystic ovary; RIF, repeated implantation failure.

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


Volume 28  Number 4  August 2016
Endometrial ‘scratching’: what the data show Santamaria et al.

the first intervention, an office hysteroscopy was Patients were randomized by using random allo-
performed at the time scheduled for the second cation software to endometrial scratching or not
scratching in 7.8% of patients in the intervention in the mid-luteal phase of the menstrual cycle
group and 6.2% in the control group. The results preceding IVF. No significant differences were found
indicate no statistically significant differences in in implantation rate (32.8 vs. 29.7%, P ¼ 0.120),
LBR or clinical pregnancy rate (CPR) between endo- CPR (34 vs. 38%, P ¼ 0.548), OPR (26.7 vs. 32.0%,
metrial scratching group and controls; however, P ¼ 0.375), miscarriage rate (30.3 vs. 18.6%,
subgroup analysis revealed a favorable effect of this P ¼ 0.150), or multiple pregnancy rate (31.3 vs.
procedure in 64 women with two or more previous 19.3%, P ¼ 0.325). Furthermore, subgroup analyses
IVF failures. Implantation rate was not provided. revealed a significantly lower OPR and LBR for
Another RCT was published in 2015 in a journal women undergoing repeated cycles in the scratch-
without impact factor [4]. In this study, scratching ing group. Upon multivariate logistic regression
was performed once between days 14 and 21 of the analysis, women’s age was the only significant factor
menstrual cycle prior to the IVF cycle in 30 women affecting the OPR in women undergoing first cycle
with at least one previous IVF failure compared with of IVF, whereas the number of top quality embryos
30 women in the control group. Endometrial injury transferred was the only significant factor for suc-
was performed by scratching the anterior and cess.
posterior wall, using a 4-mm disposable Karman’s The 2015 updated Cochrane review identified
cannula. The authors found a higher implantation 535 references from which they selected 11 pub-
&&
rate in the scratching group vs. controls (19.4 vs. lished RCTs [3,5 ,9,11–16,27,28], two preliminary
8.1%, P ¼ 0.028), although no statistically signifi- results obtained from ongoing trials from Clinical-
cant differences were observed in LBR (3.3 vs. Trials.gov [29,30], and one conference abstract [31]
10%, P ¼ 0.612) and ongoing pregnancy rate for assessing the effectiveness and safety of endo-
(OPR) (16.7 vs. 0.0%, P ¼ 0.052). Notably, the con- metrial scratching performed before embryo trans-
trol group in this study had an implantation rate of fer in women undergoing ART. Out of these 11
8.1% and a pregnancy rate of 0%, transferring a studies, eight were deemed highly biased in at least
mean of three embryos. In contrast, according to one domain. A serious methodological problem
European Society of Human Reproduction and common to these RCTs is that their inclusion
Embryology data for 2011, the mean implantation criteria vary regarding population and previous
rate per embryo transfer was 33.2% after IVF and treatments. A second problem is that different inter-
31.6% after intracytoplasmic sperm injection, with ventions occurring at different times during the
higher rates in patients below age of 35 years [26]. menstrual cycle were pooled together. As a result,
A third RCT published in 2015, not yet included four studies favored endometrial scratching during
in any meta-analysis, showed no significant differ- the luteal phase, whereas the others concluded that
ences in CPR in 166 patients undergoing office endometrial scratching at the time of oocyte
hysteroscopy with endometrial injury performed retrieval in the same cycle [14] was deleterious for
using grasping forceps with teeth, compared with endometrial receptivity.
&&
166 control women who underwent office hystero- The latest systematic literature review [18 ]
scopy alone in the early proliferative phase of the identified 1115 publications, of which only four
cycle before IUI [2]. studies met the inclusion criteria [9,11,12,16]. Only
In 2014, a study by Zarei et al. investigated the RCTs comparing endometrial scratching in the cycle
effect of endometrial scratching on the outcome of preceding ART with placebo or no treatment in
IUI cycles in women with unexplained infertility, women with at least two or more previous implan-
mild male factor, and mild endometriosis. No sig- tation failures were selected. In addition, only RCTs
nificant differences were found in pregnancy rate that reported at least one of the following outcomes
per patient (23.6 vs. 19.4%, P ¼ 0.686), pregnancy of interest were included – CPR, LBR, implantation
rate per cycle (13.5 vs. 13.3%, P ¼ 0.389), miscarriage rates, miscarriage rate, or procedure-related compli-
rate (6.9 vs. 9.7%, P ¼ 0.764), or OPR (16.7 vs. 9.7%, cations. The authors concluded that a meta-analysis
P ¼ 0.325) between the two groups. Further, the could not be properly performed because of
authors presented important limitations of the significant differences and bias in study quality,
study, such as low compliance among patients fol- population investigated, and intervention charac-
lowing the protocol and lack of uniformity in the teristics. The strength of this systematic review is
number of IUI cycles among patients. that it only included RCTs in which the intervention
A properly powered RCT at the University of by endometrial scratching was performed in women
Hong Kong, published in 2014, included 300 unse- with two or more implantation failures in the
&&
lected infertile women undergoing IVF [5 ]. cycle preceding ART. However, the timing of the

1040-872X Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-obgyn.com 247

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


Reproductive endocrinology

endometrial biopsy differed across the studies; in REFERENCES AND RECOMMENDED


one study, it was obtained in the proliferative phase READING
Papers of particular interest, published within the annual period of review, have
[12], in two others in the luteal phase [9,16], and in been highlighted as:
one study in both phases [11]. In addition to com- & of special interest
&& of outstanding interest
paring endometrial biopsy with no intervention
or sham procedure, the study by Nastri et al. [9] 1. Barash A, Dekel N, Fieldust S, et al. Local injury to the endometrium doubles
the incidence of successful pregnancies in patients undergoing in vitro
performed diagnostic hysteroscopy in the control fertilization. Fertil Steril 2003; 79:1317–1322.
group. 2. El-Khayat W, Elsadek M, Saber W. Comparing the effect of office hystero-
scopy with endometrial scratch versus office hysteroscopy on intrauterine
insemination outcome: a randomized controlled trial. Euro J Obstet Gynecol
CONCLUSION Reprod Biol 2015; 194:96–100.
3. Gibreel A, El-Adawi N, Elgindy E, et al. Endometrial scratching for women with
Modern evidence-based medicine is based on the previous IVF failure undergoing IVF treatment. Gynecolog Endocrinol 2015;
31:313–316.
scientific method, which is the most appropriate 4. Singh N, Toshyan V, Kumar S, et al. Does endometrial injury enhances
way to understand various phenomena. Simplified, implantation in recurrent in-vitro fertilization failures? A prospective randomized
control study from tertiary care center. J Hum Reprod Sci 2015; 8:218–223.
this approach requires that any tentative descrip- 5. Yeung TW, Chai J, Li RH, et al. The effect of endometrial injury on ongoing
tion, called a hypothesis, must be supported or && pregnancy rate in unselected subfertile women undergoing in vitro fertilization:
a randomized controlled trial. Human Reprod 2014; 29:2474–2481.
refuted by evidence. When evidence is consistent, A properly powered RCT including 300 unselected infertile women undergoing
the hypothesis becomes a theory that provides a IVF demonstrates no significant differences in implantation rates, CPR, OPR, and
miscarriage rate between endometrial scratching and control groups
coherent set of principles to explain a class of 6. Zarei A, Alborzi S, Dadras N, Azadi G. The effects of endometrial injury on
phenomena; these principles can then be applied intrauterine insemination outcome: a randomized clinical trial. Iran J Reprod
Med 2014; 12:649–652.
clinically. 7. Abdelhamid AM. The success rate of pregnancy in IUI cycles following
In our case, endometrial scratching is an inter- endometrial sampling. A randomized controlled study: endometrial sampling
and pregnancy rates. Archiv Gynecol Obstet 2013; 288:673–678.
vention that produces pain [9], costs money, and the 8. Gibreel A, Badawy A, El-Refai W, El-Adawi N. Endometrial scratching to
side-effects of systematic scratching in the pro- improve pregnancy rate in couples with unexplained subfertility: a randomized
controlled trial. J Obstet Gynecol Res 2013; 39:680–684.
duction of Asherman syndrome remain to be seen. 9. Nastri CO, Ferriani RA, Raine-Fenning N, Martins WP. Endometrial scratching
Overall, data from properly designed and powered performed in the nontransfer cycle and outcome of assisted reproduction: a
randomized controlled trial. Ultrasound Obstet Gynecol 2013; 42:375–382.
RCTs demonstrate no beneficial effect of this inter- 10. Parsanezhad ME, Dadras N, Maharlouei N, et al. Pregnancy rate after
vention that is based on unknown biological effects. endometrial injury in couples with unexplained infertility: a randomized clinical
trial. Iran J Reprod Med 2013; 11:869–874.
A common criticism of meta-analyses is that ‘they 11. Baum M, Yerushalmi GM, Maman E, et al. Does local injury to the endome-
combine apples with oranges’ when comparing clin- trium before IVF cycle really affect treatment outcome? Results of a rando-
mized placebo controlled trial. Gynecol Endocrinol 2012; 28:933–936.
ically diverse studies with different methodology 12. Shohayeb A, El-Khayat W. Does a single endometrial biopsy regimen (S-EBR)
and heterogeneous populations. As a result, when improve ICSI outcome in patients with repeated implantation failure? A
randomised controlled trial. Eur J Obstet Gynecol Reprod Biol 2012;
bias is present in each or some of the individual 164:176–179.
studies included, the meta-analysis may become a 13. Safdarian L, Movahedi S, Aleyasine A, et al. Local injury to the endometrium
does not improve the implantation rate in good responder patients undergoing
compound of errors and may result in an inaccurate in-vitro fertilization. Iran J Reprod Med 2011; 9:285–288.
conclusion that could be mistakenly interpreted as 14. Karimzade MA, Oskouian H, Ahmadi S, Oskouian L. Local injury to the
endometrium on the day of oocyte retrieval has a negative impact on
correct. Therefore, meta-analyses such as the 2015 implantation in assisted reproductive cycles: a randomized controlled trial.
updated Cochrane systematic review on endo- Archiv Gynecol Obstet 2010; 281:499–503.
15. Narvekar SA, Gupta N, Shetty N, et al. Does local endometrial injury in the
metrial injury [17], which consists of inadequate nontransfer cycle improve the IVF-ET outcome in the subsequent cycle in
and insufficient data, compute incorrect or incom- patients with previous unsuccessful IVF? A randomized controlled pilot study.
J Hum Reprod Sci 2010; 3:15–19.
plete conclusions. 16. Karimzadeh MA, Ayazi Rozbahani M, Tabibnejad N. Endometrial local injury
Would clinicians and patients accept the per- improves the pregnancy rate among recurrent implantation failure patients
undergoing in vitro fertilisation/intra cytoplasmic sperm injection: a rando-
formance of an intervention named ‘retinal scratch- mised clinical trial. Aust N Z J Obstet Gynaecol 2009; 49:677–680.
ing’ without any proven benefit for vision? Clearly, 17. Nastri CO, Lensen SF, Gibreel A, et al. Endometrial injury in women under-
going assisted reproductive techniques. Cochrane Database Syst Rev 2015;
the answer is no. Thus, we must consider the import- 3:CD009517.
ance of information with respect to endometrial 18. Panagiotopoulou N, Karavolos S, Choudhary M. Endometrial injury prior to
assisted reproductive techniques for recurrent implantation failure: a sys-
scratching. It is easier to scratch than to think, &&

tematic literature review. Euro J Obstet Gynecol Reprod Biol 2015; 193:27–
but, please, think before you scratch. 33.
Systematic review that identifies 1115 publications, of which only four studies met
the inclusion criteria. The authors concluded that a meta-analysis could not be
Acknowledgements properly performed because of significant differences and bias in study quality,
None. population investigated, and intervention characteristics.
19. Potdar N, Gelbaya T, Nardo LG. Endometrial injury to overcome recurrent
embryo implantation failure: a systematic review and meta-analysis. Reprod
Financial support and sponsorship Biomed Online 2012; 25:561–571.
20. El-Toukhy T, Sunkara S, Khalaf Y. Local endometrial injury and IVF outcome: a
None. systematic review and meta-analysis. Reprod Biomed Online 2012; 25:345–
354.
Conflicts of interest 21. Nastri CO, Gibreel A, Raine-Fenning N, et al. Endometrial injury in women
undergoing assisted reproductive techniques. Cochrane Database Syst Rev
There are no conflicts of interest. 2012; 7:CD009517.

248 www.co-obgyn.com Volume 28  Number 4  August 2016

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.


Endometrial ‘scratching’: what the data show Santamaria et al.

22. Simon C, Bellver J. Scratching beneath ‘The Scratching Case’: systematic 27. Guven S, Kart C, Unsal M, et al. Endometrial injury may increase the clinical
& reviews and meta-analyses, the back door for evidence-based medicine. pregnancy rate in normoresponders undergoing long agonist protocol ICSI
Human Reprod 2014; 29:1618–1621. cycles with single embryo transfer. Eur J Obstet Gynecol Reprod Biol 2014;
Opinion study assessing 10 RCTs and three meta-analyses published until 2013 173:58–62.
concluded that endometrial scratching had no beneficial effect in reproductive 28. Inal Z, Görkemli H, Hasan Ali I. The effect of local injury to the endometrium
outcome when applied in patients. for implantation and pregnancy rates in ICSI-ET cycles with implantation
23. Pundir J, Pundir V, Omanwa K, et al. Hysteroscopy prior to the first IVF cycle: a failure: a randomised controlled study. Eur J Gen Med 2012; 9:223–
systematic review and meta-analysis. Reprod Biomed Online 2014; 28:151– 229.
161.
29. Aleyamma TK, Acharya M. Local endometrial injury in women undergoing
24. Li R, Hao G. Local injury to the endometrium: its effect on implantation. Curr
repeat in-vitro fertilization /intracytoplasmic sperm injection (IVF/ICSI)-
Opin Obstet Gynecol 2009; 21:236–239.
embryo transfer (ET). Clinical Trials Registry - India 2013. [CTRI/2013/
25. Gnainsky Y, Granot I, Aldo P, et al. Local injury of the endometrium induces an
04/003564]
inflammatory response that promotes successful implantation. Fertil Steril
2010; 94:2020–2026. 30. Polanski LT, Raine-Fenning N. A study to assess if scratching the lining of the
26. European IVF-Monitoring Consortium (EIM), European Society of Human womb prior to IVF treatment increases the chances of pregnancy. Clinical-
Reproduction and Embryology (ESHRE). Assisted reproductive technology Trials gov 2014. [NCT01882842]
and intrauterine inseminations in Europe, 2011: results generated from 31. Karim Zadeh Meybodi M, Ayazi M, Tabibnejad N. Effect of endometrium local
European registers by ESHRE. MS K, editor. Munich, Germany: Annual injury on pregnancy outcome in patients with IVF/ICSI. Human Reprod 2008;
Meeting of ESHRE; 2014. 23 (Suppl 1):i126.

1040-872X Copyright ß 2016 Wolters Kluwer Health, Inc. All rights reserved. www.co-obgyn.com 249

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.