Dei21(2).book(dei343.fm)

Human Reproduction Vol.21, No.2 pp. 364–369, 2006
Advance Access publication November 10, 2005.
Indomethacin effect on implantation rates
in oocyte recipients
R.Bernabeu1,4, M.Roca2,4, A.Torres3,5 and J.Ten1
1Instituto Bernabeu, Alicante, 2Instituto Bernabeu, Cartagena, 3Department of Health and Social Sciences, Medical School, Murcia University and 4Reproductive Health Chair, Universidad Miguel Hernández (Alicante), Spain 5To whom correspondence should be addressed. E-mail: [email protected] BACKGROUND: Implantation failure is the main limiting factor for success of IVF. Even when transfer techniques
are carried out extremely carefully, embryo transfer may produce an endometrial and cervical reaction that may res-
ult in an inflammatory response and impaired implantation. There are no formal specifications on the use of
indomethacin in reproductive cycles and there are no studies published in the scientific literature on its effect on
embryo implantation. Oocyte donation is the best model to evaluate the determinants of implantation. The aim of this
study is to evaluate the potential benefit of indomethacin administered at embryo transfer. METHODS: A rand-
omized pilot trial of 136 oocyte recipients was carried out. Seventy-two women received standard specifications plus
100mg of indomethacin rectally given as three doses every 12h starting on the night prior to transfer. RESULTS:
Positive HCG (>= 6mUI/ml) occurred in 59.7% of treated women and in 59.4% of women in the control group.
Implantation rates were 27.8% in the indomethacin group and 26.4% in the controls. CONCLUSIONS: The indo-
menthacin group did not show significantly higher implantation rates. A larger study exploring alternative treatment
protocols might be appropriate.
Key words: embryo transfer/endometrial receptivity/implantation/randomized pilot study Introduction
interaction between the endometrium and myometrium, we Implantation failure is the main limiting factor for success in cannot consider the uterus as a passive container. Although, for IVF in reproductive medicine. As an aggressive procedure, obvious reasons, we cannot study the immediate endometrial embryo transfer provokes a uterine response involving response in humans after the embryo transfer, it may be endometrial inflammatory phenomena and increased myome- hypothesized that factors related to a decreased uterine activity trial activity. Factors that may induce a uterine response and inflammatory endometrial response at the moment of include hyperphysiological hormonal levels, direct myometrial embryo transfer might increase the chances of embryo implan- effects of the drugs used in the cycles, local inflammatory tation. Until recently, little attention has been paid to the role of responses due to external particles introduced by manipulation, embryo transfer techniques (Englart et al., 1986; Mansour dynamic responses due to stimulation of the cervix and intrac- et al., 1990; Mansour et al., 2002) and even less to its pharma- avitary canalization, the stress experienced by a woman who is cological management (Kovacs, 1999).
undergoing a reproductive cycle, or unknown causes.
Indomethacin, a non steroidal anti-inflammatory drug Uterine activity is well established (Ijland et al., 1996; Van (NSAID), widely used in clinical practice, has well known Gestel et al., 2003). Since the introduction of IVF, an increased anti-prostaglandin effects that reduce uterus contractility. It uterine activity was also documented, as well as its harmful also has vasodilatory action (Hiemeyer, 1967; Saksena, 1974; effects on embryo attachment (De Vries et al., 1990; Ijland Lau, 1973). Its uterolytic effects in gravid uterus are well et al., 1998, 1999; Bulleti et al., 2000) and a high number of known, as well as its benefits in treating dysmenorrhoea.
embryos ejected (Menezo et al., 1985; Poindexter et al., 1986).
We also know that the production of inflammatory cytokines During mock embryo transfers, a tenaculum applied to the is important for successful implantation, but excessive pro- cervix elicited the release of oxytocin and increased uterine duction may be detrimental (Chaouat et al., 2002). In addition to the exogenous manipulation, introducing extrauterine parti- Therefore, even when transfer techniques are carried out cles (cervical mucus, bacteria, detritus, and so on) could trigger extremely carefully, embryo transfer may produce an inflam- a ‘pro-inflammatory status’ (Chaouat et al., 2002). Nonetheless, matory response and/or increased contractility, which may res- in 4 day pregnant rats, intrauterine indomethacin at moderate ult in implantation failure. Due to the close relationship and or low doses did not show any anti-implantation effects (Gupta 364 The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected] Indomethacin and implantation rates
et al., 1981). Furthermore, indomethacin has been used success- embryo quality, and quality of transfer was determined in all women.
fully in animal models to reduce the anti-implantation effect of To reduce variability in oocyte quality, all donors were young (mean intrauterine devices (Chaudhury, 1975; Hurst et al., 1982).
age 26.2 ±0.4 years) with normal body mass index (BMI) and no ovu- However, there are no studies published on the effect of latory disorders (normal baseline hormonal levels and normal ovarian indomethacin on human embryo implantation, and no formal ecography). All donors received the same induction protocol usingrFSH (Gonal F; Laboratorios Serono, Madrid, Spain) under LHRH specifications for its use in reproductive cycles.
analogue suppression in a short protocol (Procrin; Abbott Laborato- Oocyte donation is the best model to evaluate the determi- ries, Madrid, Spain). This study was approved by the ethical and nants of implantation, for several reasons. First, there is a min- research committee of the Institute. The study design is presented in imal variability in embryo quality, as donors are young women of a similar age with no ovulatory disorders. Secondly, thepreparation of the endometrium is similar, as all recipients Intervention protocol
receive the same hormonal replacement protocol. Finally, as Standard specifications for oocyte recipients included a standard proto- procedures are performed under the same circumstances and col of ethynyl-estradiol transdermal patches with increasing doses by the same medical team, the embryo transfer techniques are starting at 50 mg daily from day 1 to 7, (Dermestril, Laboratorios Rot- similar for all participants. Therefore, in spite of being fully tafharm, Barcelona, Spain), 100 mg daily from day 8 to 11, and allogenic, embryos transferred in oocyte donation programmes 150 mg from day 11 onwards, plus intravaginal micronized progester- have relatively high implantation rates.
one (Utrogestán, Laboratorios Seid, Barcelona, Spain) 200 mg every The objective of the study was to assess whether indometh- 8 h starting on the afternoon of the oocyte pick-up and continuing up to the day of β-HCG measurement 11 days later. If the β-HCG result acin has a positive effect on implantation rates using an oocyte was positive, we maintained the same protocol until the 11th week of recipient model. The specific hypothesis of the study was that pregnancy. The intervention group received standard specifications indomethacin would improve oocyte implantation rates.
plus three doses of 100mg of indomethacin rectally every 12h, startingon the night prior to the transfer. The non-intervention group received Subjects and methods
Selection of the subjects
Embryo transfer procedure
An ongoing randomized clinical trial of 173 first cycle oocyte recipi- All transfers were done on days 2 or 3 following the same protocol, ents following IVF and ICSI was established in June 2003. A ran- with a full bladder, using ultrasound guidance and the same soft cath- dom sequence of 173 treatment and control codes was generated by eter (Embryo transfer catheter Rocket Medical, Washington, UK), and the epidemiologist and written in a table, in which each cell had one performed by the same experienced biologist and medical team. Both, number (from 1 to 100) and one treatment (A) or control (B) code.
the biologist and the medical team were blind to the status (indometh- The gynaecologist specially assigned to attend recipients was acin or control group) of the patients.
instructed to assess inclusion and exclusion criteria and was respons- ible for including women in the study and for writing the medicalhistory number of the women in the table. Once all 100 code num-bers had been allocated, the allocation sequence was restarted atnumber 1.
Women who approached the clinic for IVF or ICSI and were will- ing to collaborate and eligible for inclusion were randomly assigned to either the intervention or control group by their attending gynaecolo- gist on the day of their first visit. Women were informed of the objec- tives of the study and their consent was obtained before proceeding.
The table with the codes and medical history numbers remained withthat gynaecologist throughout the study. Patients who were in the con- trol group did not receive a placebo.
Once reproductive cycles were completed for all women, the data from the medical history were written onto a database by a biologistblind to the treatment or control code of the patient. Women included in the study were later identified using their medical history numbers and their corresponding codes for treatment or control included in thedatabase. Therefore, the clinical staff that performed the transfer andassessed study outcomes were blind to whether the women were in the Inclusion criteria were: (i) first cycles of women candidates to be oocyte recipients (IVF and ICSI); (ii) no known allergic reaction toNSAIDs and; (iii) no neurological or gastrointestinal disease. Exclu-sion criteria for recipients were: (i) recurrent miscarriage; (ii) endome-trial pathology; and (iii) severe endometriosis.
Baseline variables were obtained in all women on their visit to the clinic. Information on possible confounders, age of recipient anddonating women, endometrial thickness, type of endometrial line, Figure 1. Study design of the randomized pilot study on indomethacin
semen quality, oocyte quality, number of embryos transferred, effectiveness for women recipients in an oocytes donation program.
R.Bernabeu et al.
Outcome variables
Table II. Causes of women’s infertility
There were two outcome variables in the study. Biochemical preg-nancy is defined as β-HCG >5IU. Implantation is defined as an embryonic sac visible with ecography at day 11 post-transfer inwomen with a biochemical pregnancy. The implantation rate the main outcome of the study is defined as the ratio between the number of embryonic sacs visible with ecography at day 11 post-transfer in women with a biochemical pregnancy and the number of embryos transferred. Biochemical analyses and ecographies were performed by professionals blind to the indomethacin status of the women, and Sample size, data management and analysis
Due to the slow recruiting process involved in identifying suitableoocyte recipients, the study was designed as a concurrent and still The main reason why women were included in the donation ongoing study. We present here the results of the first 136 women who have completed the cycles from the 173 women who were Clinical outcomes are presented in Table III. It should be recruited for the study. Data entry was done in SPSS 12 by a member noted that no adverse effects due to the use of indomethacin of the staff blind to the objectives and clinical outcomes of the study.
were reported. Out of the 136 women who completed the pro- In the analyses, we compared the percentages of biochemical preg- tocol, 59.7% of those who received indomethacin had positive nancies and implantation in the two groups. Adjustments were made β-HCG values, while β-HCG was positive in 59.4% of the 64 for possible confounders. Crude and adjusted analyses were con-ducted using SPSS 12.
women in the control group [relative risk (RR)= 1.0; 95% con-fidence interval (CI) 0.5–2.0]. Implantation occurred in 50% of In total, 187 embryos were transferred in the indomethacin In total, 173 women were randomly assigned to either group.
group of which 52 implanted, resulting in an implantation rate In 37 women (13 women in the indomethacin group and 24 in of 27.8%. In the control group, 166 embryos were transferred, the control group) who were recruited into the study, the cycles and the implantation rate was 26.4%. Adjustment for possible have not been completed. Consequently, embryo transfers were confounders did not alter the results. The differences between not performed in those women and analyses were conducted on the two groups were not statistically significant.
the women in whom embryo transfer were performed. Baselinevariables and possible confounders were similarly distributed in Discussion
both groups (Table I). The reasons why women were includedin the oocyte donation programme are presented in Table II.
Several strategies have been proposed in order to improveuterus receptivity at the time the embryo reaches the endome-trial cavity and to minimize the uterine activity. Reducing cer-vical stimulation by a careful technique, non-traumatic pass of Table I. Comparison of baseline characteristics in the intervention and control
groups in women who completed the protocol (n = 136)
the catheter through the uterine cavity, or ecographicallyguided transfer have all shown a beneficial effect. However, apart from the use of progesterone (Fanchin et al., 2001) or ritodrine (Pinheiro et al., 2003), a pharmacological approach toembryo transfer has not been considered.
We chose indomethacin for its well known action on the gonadal axis inhibiting the release of LH in the hypophysis, delaying or suppressing ovulation, decreasing the number of oxytocin receptors in the endometrium, and finally decreasing the myometrial activity throughout the cycle. Therefore, two possible sites of action could be hypothesized: in the endometrium by decreasing the inflammatory response due to mechanical manipulation and introduction of foreign particles; and in the myometrium by decreasing its activity.
Myometrial contractions progressively increase in frequency, amplitude and direction of propagation toward the uterine fundus throughout the follicular phase, to reach a maximum during the mid-cycle (Abramoivicz, 1990; De Vries, 1990; Lyons, 1991; Ijland, 1996; Bulleti, 2000). The prostaglandin E concentration measured in the myometrium was found to increase progres-sively, rising to a peak at the end of the follicular and ovulatory *Quality of transfer evaluates the presence of blood or mucous in cervix, number of attempts to transfer, and whether endometrium was or not touched.
periods, and dropping suddenly after ovulation (Vijayakumar, Indomethacin and implantation rates
Table III. Outcome variables in the 136 women of the indomethacin and control groups in whom embryo transfers were performed
embryonic sacs per woman with implantationImplantation rates Differences were not statistically significant.
aTotal number of embryos transferred.
1981). Uterine contractions are known to affect embryo should be noted, though, that 720 women would have been implantation in animals (Adams, 1980; Liedholm, 1980) and in needed in order to achieve a power of 80% and level of signifi- humans (Fanchin et al., 2001). Recently, Maslow and Lyons cance of 95% with an estimated implantation rate of 27% in (2004) have reported a clear inhibitory action of ibuprofen on untreated women and an effect associated to indomethacin use mid-cycle myometrial contractions. Indomethacin had also been of 10%. Therefore, small positive effects of indomethacin can- used successfully to reduce uterine contractility (Lenz, 1991).
Physiological implantation is described by apposition, The oocyte donor–recipient model allowed us to minimize adhesion and trophoblast invasion phases. Apposition other possible confounders related to oocyte quality, causes of requires an inflammatory-type reaction followed by an infertility and male factors. The main reasons why women anti-inflammatory-type reaction. Our results suggest that entered the oocyte donation programme as recipients were indomethacin did not affect the initial inflammation-type reac- ovarian causes. We may expect that their uterine receptivity tion which is essential for the implantation to occur. Studies status would be quite homogeneous. However, women with using animal models had already indicated that indomethacin other causes (endometriosis, unexplained or others) may have did not have deleterious effect on this phase (Gupta et al., unequal uterine receptivity. In fact, when we compared the 1981), and could even reduce the anti-implantation effect of intervention and control groups within women with ovarian intrauterine devices (Chaudhury, 1975; Hurst et al., 1982).
causes of infertility only, indomethacin seems to improve However, to our knowledge, this is the first study that shows reproductive outcomes by 10% or more (see Table IV), while no adverse effect of indomethacin on human implantation.
in the remaining group of women with causes of infertility In our study, the overall pregnancy rate, plasma β-HCG, other than ovarian, the placebo group was far better off (see implantation rates and the number of sacs were similar in both Table V). Although the differences are not statistically signific- groups, suggesting that there was no better uterine compliance ant, these results are compatible with indomethacin having a in the indomethacin as compared with the control groups. The positive effect in women with ovarian causes of infertility, small differences found were not statistically significant. It while having no effect (or even a deleterious effect) in women Table IV. Comparison of outcome variables in both groups of women recipients with underlying ovarian causes (n = 84) as a reason for entering the oocyte
donation programme
embryonic sacs per woman with implantationImplantation rates Differences were not statistically significant.
R.Bernabeu et al.
Table V. Comparison of outcome variables in both groups in women recipients with other that ovarian causes (n = 52) as a reason for entering the oocyte donation
programme
woman with implantationImplantation rates Differences were not statistically significant.
with other causes of infertility. Our study design does not Acknowledgements
allow us to draw conclusions on the basis of within-group A previous version of this work was presented as an oral communica- comparisons, but it does raise these intriguing hypotheses for Some limitations of our study should be noted. The study References
was not blind for the patient or the physician conducting the Abramowicz JS and Archer (DF) 1990 Uterine endometrial peristalsis—a first evaluation and the recruitment of the patients into the transvaginal ultrasound study. Fertil Steril 54,451–454.
study. Selection bias might, theoretically, have occurred if the Adams CE (1980) Retention and development of eggs transferred to the uterus attending gynaecologist had selectively assigned patients to at various times after ovulation in the rabbit. J Reprod Fertil 60,309–315.
one group or another, or excluded them from the study on the Bulletti C, de Ziegler D, Polli V, Diotallevi L, Del Ferro E and Flamigni C (2000) Uterine contractility during the menstrual cycle. Hum Reprod 15 basis of criteria other than those established in the research protocol. As the research protocol was strictly followed in the Chaouat G, Zourbas S, Ostojic S, Lappree-Delage G, Dubanchet S, Ledee N enrolment and assignment phases, we do not believe that selec- and Martal J (2002) A brief review of recent data on some cytokine expres-sions at the materno-fœtal interface which might challenge the classical tion bias might have had a significant role in this study. Moreo- Th1/Th2 dichotomy. J Reprod Immunol 53,241–256.
ver, as the team conducting the laboratory procedures, clinical Chaudhury MR (1975) The effect of drugs on implantation in rats with an transfers and evaluation of the results of the transfer were blind intrauterine device in one horn of the uterus. J Reprod Fertil 42,571–573.
to the status of the patients, we feel that significant information De Vries K, Lyons EA, Ballard G, Levi CS and Lindsay D (1990) Contractions of the inner third of myometrium. Am J Obstet Gynecol 162,679–682.
Englert Y, Puissant F, Carmus M, Van Hoeck J and Leroy F (1986) Clinical A final consideration is that indomethacin may be useful but study on embryo transfer after human in vitro fertilization. J In Vitro Fertil at dosages or regimen protocols different from the one we fol- Fanchin R, Ayoubi JM and Righini C (2001) Uterine contractility decreases at lowed. The protocol and/or the dosages we used for adminis- the time of blastocyst transfer. Hum Reprod 16,1115–1119.
tering indomethacin may not have been adequate. It should be Gupta U, Malhotra N, Varma SK and Chaudhury RR (1981) Effect of intrau- noted that there are no formal specifications on the use of terine administration of antiprostaglandin drugs on implantation in the rat.
indomethacin, and higher doses or longer administrations may Hiemeyer V and Feyen H (1967) Studies on the antiphlogistic action of be needed to obtain beneficial effects. Additionally, we did not indomethacin using the piromen skin test. Med Klin 62,105–107.
obtain information prior to transfer on whether indomethacin Hurst PR, Peplow PV and Gamble VM (1982) Intrauterine delivery of had or not been used. Having done that would have disclosed indomethacin from a silastic device: effect on the preimplantation embryoand implantation. Contraception 6,97–107.
their status to the team conducting embryo transfer. However, Ijland MM, Evers JLH, Dunselman GAJ, Katwijk CV, Lo CR and Hoogland the disadvantage is that we do not know the degree of compli- HJ (1996) Endometrial wavelike movements during the menstrual cycle.
ance in the intervention group. Although our clinical experi- ence is that women seeking IVF-ICSI are highly compliant Ijland MM, Evers JL, Dunselman GA and Hoogland HJ (1998) Endometrial wavelike activity, endometrial thickness, and ultrasound texture in con- with medical protocols, low indomethacin use may not be trolled ovarian hyperstimulation cycles. Fertil Steril 70,279–283.
Ijland MM, Hoogland HJ, Dunselman GA and Evers JL (1999) Endometrial wave A favourable finding of our study is that no adverse effects direction switch and the outcome of in vitro fertilization. Fertil Steril 71,476–481.
or reduced implantation rates were found in the indomethacin Kovacs GT (1999) What factors are important for successful embryo transfer after in-vitro fertilization? Hum Reprod 14,590–592.
group, suggesting that indomethacin use does not disrupt Lau IF, Saksena SK and Chang MC (1973) Pregnancy blockade by indometh- acin, an inhibitor of prostaglandin synthesis: its reversal by prostaglandins An alternative explanation for our findings is that contractility and progesterone in mice. Prostaglandins 4,795–803.
Lenz S, Lindenberg S, Sundberg K, Hamberger L and Sjogren A (1991) Intra- and/or the inflammatory reaction following embryo transfer uterine capsules for incubation of gametes and subsequent release of may not be an important determinant of implantation failure. A embryos. J In Vitro Fertil Embryo Transf 8,265–271.
larger study exploring alternative treatment protocols and pos- Lesny P, Killick SR, Robinson J, Raven G and Maguiness SD (1999) Junc- sible differential effects in patient subsets might be appropriate tional zone contractions and embryo transfer: is it safe to use a tenaculum?Hum Reprod 14,2367–2370.
before completely ruling out any positive effects of indometh- Liedholm P, Sundstrom P and Wramsby H (1980) A model for experimental studies on human egg transfer. Arch Androl 5,92.
Indomethacin and implantation rates
Lyons EA, Taylor PJ, Zheng XH, Ballard G, Levi CS and Kredentser JV (1991) or pregnancy rates in intracytoplasmic sperm injection (ICSI) cycles. J Assist Characterization of subendometrial myometrial contractions throughout the menstrual cycle in normal fertile women. Fertil Steril 55,771–774.
Poindexter AN, Thompson DJ, Gibbons WE, Findley WE, Dodson MG and Mansour RT and Aboulghar MA (2002) Optimizing the embryo transfer tech- Young RL (1986) Residual embryos in failed embryo transfer. Fertil Steril Mansour R, Aboulghar M and Serour G (1990) Dummy embryo transfer: a Saksena SK and Harper MJ (1974) Prostaglandin-mediated action of intrauter- technique that minimizes the problems of embryo transfer and improves the ine devices: F-prostaglandins in the uterine horns of pregnant rabbits with pregnancy rates in human in vitro fertilization. Fertil Steril 54,678–681.
unilateral intrauterine devices. Fertil Steril 25,121–126.
Maslow KD and Lyons EA (2004) Effect of prostaglandin and antiprostaglan- Van Gestel I, Ijland MM, Hoogland HJ and Evers JLH (2003) Endometrial wave- din on midcycle myometrial contractions. Fertil Steril 82,511–513.
like activity in the non-pregnant uterus. Hum Reprod Update 9,131–138.
Menezo Y, Anker D and Salat-Baroux J (1985) Conception and realization of Vijayakumar R and Walters W (1981) Myometrial prostanglandins during the artificial dyed embryos for training in vitro fertilization and embryo transfer human menstrual cycle. Am J Obstet Gynecol 141,313–318.
(IVF and ET). Acta Eur Fertil 16,55–58.
Pinheiro OL, Cavagna M, Baruffi RL, Mauri AL, Peterson C and Franco JG Jr Submitted on February 9, 2005; resubmitted on September 6, 2005; accepted (2003) Administration of beta2-adrenergic agonists during the peri-implantation

Source: http://www.institutobernabeu.com/upload/ficheros/enlaces/indomethacin_effect_on_implantation_rates_in_oocyte_recipients.pdf