{"paper_id":"92c57d36-358f-4310-a596-dfe90fc2f9c3","body_text":"A good proliferative modification of endometrium is a key factor to nourish a\nblastocyst in assisted reproduction cycles. Endometrium thickness with or without\nthe endometrial pattern is the outstanding sonographic parameter that have been\nwidely used to obtain maximal endometrial receptivity ( Kasius  et al ., 2014 ;  Arce  et al ., 2015 ;  Yoeli  et al ., 2004 ;  Zhao  et al ., 2012 ;  Hancke  et al ., 2012 ;  Kovacs  et al ., 2003 ). A variety of endometrial\nmaturation regimens has been investigated, to improve endometrium receptivity.\nArtificial cycle regimen is the most common protocol for hormone replacement therapy\nfor endometrium preparation prior to blastocyst transfer. In this routine hormone\nreplacement therapy protocols, the endometrium is induced by exogenously\nadministered estradiol and progesterone ( Veleva\n et al ., 2013 ;  Lathi\n et al ., 2015 ;  Morozov  et al ., 2007 ;  Wright  et al ., 2006 ). Estrogen is used until the\nendometrial thickness on ultrasound meets approximately 0.8cm ( Singh  et al ., 2011 ), then, progesterone\ninitiates according to the stage of blastocyst development. Progesterone stimulation\nfor a specific number of days will induce endometrial receptivity ( Paulson, 2011 ).\nTo date, one of the other common methods for endometrium preparation is adding GnRH\nto hormone supplements, or with follicle-stimulating drugs such as Clomiphene\nCitrate ( Gelbaya  et al .,\n2006 ;  El-Toukhy  et al .,\n2004 ;  Arefi  et al .,\n2016 ;  Peeraer  et al .,\n2015 ). Modification of physiologic endometrium concentration, the concern\nof hormonal exposure in the uterus, greater drug doses required, hormone\ncomplications and high treatment cost were considered as disadvantages of hormonally\nmanipulated protocols. Natural approaches for endometrium preparation, as a\npatient-friendly option, have been initiated in recent years. The lack of above\ndisadvantages was the fundamental reason that led to an increasing trend toward this\ntherapeutic approach. However, the natural cycle regimen has a number of\ncontroversies regarding its use. Current disadvantages contain a frequent ultrasonic\nassessment of the follicles, unexpected ovulation and lack of synchronizing\ndevelopment of the endometrium with the dominant follicles ( Von Wolff  et al ., 2014 ;  Polyzos  et al ., 2016 ;  Allersma  et al ., 2013 ;  González-Foruria  et al ., 2016 ;  Gordon  et al ., 2013 ).\nThereupon, the evidence is still sparse as to which endometrium preparation protocol\nis preferred. Some authors recommend considering the patient's preferences,\ncost-effectiveness, and safety for mother and child ( Pennings & Ombelet, 2007 ;  Groenewoud\n et al. , 2012 ).\nIn view of the above, we address the pregnancy outcome in modified natural cycles\nusing the HCG regimen versus artificial and mild hormonally stimulated protocols in\npatients undergoing vitrified thawed blastocyst transfer.\n\nThis control-randomized trial was carried out at the Fatemezahra Infertility Research\nCenter, affiliated with the Babol University of Medical Science. The study was\napproved by the Research Ethics committee of Babol University of Medical Science and\nwas registered with the number of 201408021760N36 in the Iran clinical trial\nregistry (IRCT).\nA total of 131 patients submitted to vitrified thawed blastocyst transfer in our\nIVF laboratory were invited from March 2015 to January 2016. Women undergoing\nvitrification thawed blastocyst transfer (VTBT) were eligible for the study when\nthey were normo-ovulatory women, between 20 to 40 years of age, with 19<BMI\n<30.\nThe exclusion criteria included women with PCOs, basal FSH>10 IU/ml and basal\nE 2  <70 pg/ml, those with untreated thyroid disorders, severe\nendometriosis, recurrent implantation failure, uterine pathology, recurrent\nabortion, repeated implantation failure, smokers, athletes and patients who had\nused any medication in the two previous months that could interfere with the\nnormal function of the hypothalamic-pituitary-gonadal axis.\nThe 123 women submitted to VTBT, who met the inclusion criteria and provided\nwritten informed consent for the study were included. To apply a\npatient-friendly method, we chose the natural modified cycle with HCG trigger\ninstead of the true natural cycle as the study group. For sample size\ncalculation, we used a confidence interval of 95%, power of 80 in the pregnancy\nrate between the three groups to choose a sample size of 120 patients - an\nadequate number in each group to achieve an 80% power of detection at a\nsignificant level of 0.05 in a ratio of 1:1:2. The randomization was done at the\nstart of the cycle using sequential numbering based on a computer-generated list\nthat had been prepared at the Statistics Center of the Babol University of\nMedical Science and sent to us. Then, the participants were randomly assigned to\neither modified natural cycle with HCG (n=31), mildly hormonally stimulated\ncycle (n=30) or artificial regimen (n=62). The participant and the infertility\nexpert were not blinded for treatment allocation.\nThe sonographer was not changed during the procedure. Laboratory and transfer\ntechniques were the same during the procedure.\nAt first, the patients were assessed by transvaginal ultrasound (TVS) on the\nthird day of men struation (7.5 MHz vaginal probe; Mylab40, Esaote, Italy) to\nremove the patients with ovarian cysts. Then, a serial TVS measured the\nendometrial thickness and follicle diameter consistently.\nOur primary outcome was the pregnancy rate in the modified natural cycle using\nthe HCG protocol versus the mild hormonally stimulated and artificial protocols\nof endometrium preparation following vitrified blastocyst transfer. As\nadditional outcome variables, we evaluated the endometrial characteristics in\nthe modified natural cycle versus hormonally stimulated and artificial cycle\nregimens at the day of vitrified blastocyst transfer.\nWe used the natural cycle with HCG for the patients in this group; no medication\nwas administered during the endometrial preparation. The follicles were\nmonitored by TVS until the dominant follicles reached a diameter of 18-20 mm and\nendometrium thickness >8 mm. Then, 10,000 IU of Human Chorionic Gonadotropin\n(CG, Daroupakhsh, Iran) was administered for ovulation. Vitrified Blastocysts\nfollowing warming were transferred after ovulation was observed, usually on\n36-38 hours after HCG administration.\nThe natural cycle with HCG reduces the number of LH monitoring visits required to\nschedule the day of VTBT; then, we preferred to use HCG for the detachment of\nthe eggs in terms of cost-effectiveness and patient convenience.\nThe mild hormonally stimulated group with clomiphene citrate (Clomid, Iran\nHormone Company) was administered 50 mg daily from day 3 of the menstrual cycle\nfor 5 days. If during TVS a follicle 18-20 mm was visible, ovulation was deemed\nto have occurred. Then, 10, 000 IU of urinary HCG was administered and the\nblastocyst were transferred 36-38 hour after HCG.\nThe Artificial cycles began on the third day of the menstrual cycle or\nprogesterone withdrawal. The dose of oral estradiol valerate (E2) (Aburaihan\nPharmaceutical Co., Tehran, Iran) was 2mg bid (4mg/day). A higher initial dose\nof estradiol (6mg) was administered if the patient showed inadequate endometrial\nthickness in a previous cycle. TVS was carried out on day 10. If the endometrial\nthickness reached 8 mm and further, 50mg progesterone was given IM for 3 days\n(Aburaihan Pharmaceutical Co., Tehran, Iran) and estradiol was continued as\nwell, then the blastocysts were transferred on the fourth day of progesterone\nadministration. If the endometrial thickness was 8mm or less on day 10, the dose\nof estradiol valerate was increased to 4mg twice/day and the blastocyst were\ntransferred 4-5 days following initiation of progesterone administration if the\nsigns of ovulation were observed upon TVS. If the endometrial thickness did not\nreach 8 mm up to day 20, or the ovulation was not confirmed, the cycle was\ncancelled.\nFor luteal supplementation, vaginal suppository Cyclogest 400mg (Actavis Group,\nIceland) twice/daily was recommended for all groups following the day of\nblastocyst transfer during 14 days.\nSince the clinical outcomes of vitrification/warming are superior to\nslow-freezing/thawing ( Rienzi  et\nal ., 2017 ), we chose the vitrification thawed blastocyst\nmethod instead of a frozen embryo or fresh blastocyst.\nAll participants had blastocyst from their prior cycles, which had been\ncryopreserved by vitrification and warming by the Cryotop methodology, as per\ndescribed by Kuwayama ( Kuwayama,\n2007 ).\nAfter warming, the blastocyst was partially or completely re-expanded to the\ndimensions it had before vitrification. We considered a blastocyst had survived\nafter warming if the following morphologic parameters existed; inner-cell mass\n(ICM) should be equally shaped and sized as before cryopreservation. In addition\nto the number and cohesiveness of ICM and trophectoderm, and blastocele\nexpansion according to Gardner's criteria ( Gardner & Schoolcraft, 1999a ). For sample consistency purposes,\nonly good-quality blastocysts were used for transferring. We defined\ngood-morphology blastocysts as the ones that reached at least grades A or B;\nexcellent, (≥3AA) and good, (3, 4, 5, 6, AB and BA) based on ICM and\ntrophectoderm quality score, according to the criteria proposed by Gardner and\ncolleagues ( Gardner & Schoolcraft,\n1999b ).\nAn embryologist, using the same method, did all laboratory procedures.\nIt is noteworthy that the eligible women could not be randomized and contributed\nmore than one cycle. Each of the patients received only one good quality\nblastocyst and the transfer was not repeated if she did not become pregnant.\nAfter the transfer, the failed patients were drawn out of the study, and were\nsubmitted to another recommended endometrium preparation protocols.\nThe endometrial maturation was evaluated by the endometrial thickness and the\npresence of the triple line endometrial pattern at the day of HCG administration\nin the modified natural and hormonally stimulated cycles and at the day of\ntransferring, and for artificial regimen as well. Endometrial thickness was\ndefined as the maximal distance between the echogenic line of the myometrium and\nthe endometrium that was measured in the midsagittal view by two-dimensional TVS\nat the day of HCG administration in the modified natural and mild hormonally\nstimulated cycles, and at the day of transferring for the women submitted to the\nartificial regimen. Triple-line pattern contains two hypoechoic layers that\nsurrounded a central hyperechoic line.\nThe duration of endometrial preparation was defined as the interval from the day\nof menstruation to the day of HCG administration.\nThe chemical pregnancy test was defined as the serum b-hCG≥30 IU/L, 10\ndays (two consecutive tests at 2-day intervals) following the blastocyst\ntransfer. The implantation rate was determined by the percentage of gestational\nsac per blastocyst transferred. A clinical pregnancy was defined as the\nvisualization of a gestational sac with fetal heart activity on TVS in week five\nof gestation. An ongoing pregnancy was a pregnancy that completed ≥24\nweeks of gestation. An abortion was defined as the inability to see a previously\nconfirmed gestational sac or heartbeat between week 7 and week 20.\nThe statistical analysis was performed with the SPSS (Statistical Package for\nSocial Science, SPSS Inc., Chicago, IL, USA) version 16.00 software. We ran the\nanalysis per protocol and excluded the patients lost to follow up. Therefore, 28\npatients in the modified natural cycle, 29 patients in the hormonally stimulated\ncycle and the 56 patients in the artificial regimen were analyzed.\nKolmogorov-Smirnov was used to test the normality distribution of continuous\nvariables. Owing to normally distributed, the statistical comparison was\nassessed using the ANOVA test for continuous variables, and the Chi-square test\nwas used for categorical variables. Post-hoc test confirms where the differences\noccurred between groups. The findings were presented by means with standard\ndeviations and the categorical variables were given as percentages (%).\n p -values <0.05 represents statistical significance.\n\nOut of 131 patients eligible for the study, 123 women were recruited according to our\nexclusion and inclusion criteria, and 8 patients were excluded. The reasons for\nexcluding the participants are illustrated in  Figure\n1 . Finally, the patients were randomized into three groups; natural cycle\ngroup (n=31), ovulation induced (n=30), and artificial cycles (n=61).\nFigure 1 Randomization of the women who participated in the study.\nRandomization of the women who participated in the study.\nThree women in the modified natural group and two women in the artificial group were\ntaken off because of insufficient endometrium growth. Finally, three women in the\nmodified natural group, one in the mild hormonally induced group and three women in\nthe artificial group were lost to follow up, hence 28 women in the natural, 29 women\nin hormonally stimulated and 59 women in the artificial group received blastocysts\nand entered the study.\nAs presented in  Table 1 , the patients in the\nthree groups had similar demographic characteristics. Mean age and BMI in the\nmodified natural group, stimulated ovulation and the artificial group were similar\n(30.40±4.6 vs. 30.5±5.89 vs. 29.71±3.8,\n p =0.78) (25.82±3.83 vs. 25.36±5.7 vs.\n26.19±3.38,  p =0.77). There were no significant differences\namong the three groups in regards to infertility duration, infertility cause, type\nof infertility (primary/secondary) and baseline serum FSH and LH ( Table 1 ).\nBaseline characteristic of the vitrified blastocyst recipients who\nparticipated in the study.\nNS: Not significant.\n147 vitrified blastocysts were transferred to 116 patients. 36 blastocysts belonged\nto the natural cycle group, 38 to hormonally stimulated group and 73 to the\nartificial groups. No significant difference was seen among the women in regards to\nblastocyst quality in the three groups. All the transferred blastocysts had good\nmorphology, as previously described in the method.\nAs a whole, 18.64% (22) chemical pregnancies were achieved in 116 blastocyst stage\ncycles. The gestational sac was not visualized in three patients of the artificial\ngroup. The implantation rate was established at 16.1% (19). Fetal heartbeat was\ndeclared in 16.1% (19) patients. The total ongoing pregnancy rate was 15.7% (18).\nOne patient in the artificial group had a miscarriage in week 14 of pregnancy.\nThe findings of pregnancy outcome and endometrial preparation are illustrated in\n Table 2 . No statistically significant\ndifferences were found in terms of implantation rate, chemical, clinical, ongoing\npregnancy and miscarriage among the three groups.\nReproductive outcome of the women received vitrified blastocyst.\na  vs. b:  p <0.001, b vs. c:\n p <0.02. d vs. e:  p <0.01, e vs.\nf:  p <0.001. NS: Not significant.\nPost hoc  test demonstrated that the endometrium thickness was\nsignificantly greater in the artificial vs. the natural cycle using hCG and in\nnatural cycle vs. hormonally stimulated groups, respectively (8.34±0.89 vs.\n7.3±1.4,  p <0.001; 8.13±0.95 vs. 7.3±1.4,\n p <0.02).\n\nIn the selected population, we did not find any statistically significant difference\nin the reproductive outcome of the modified natural cycle with the HCG trigger\nprotocol, the mild hormonally induced cycle or the artificial cycle regimens.\nHowever, the results showed a trend towards a slightly higher ongoing pregnancy\n(7-8% higher), implantation and clinical pregnancy (6-7.5% higher) rates in the\nmodified natural group. To the best of our knowledge, there are a few randomized\ntrials that investigated the three mentioned cycle regimens for endometrium\npreparation simultaneously; however, our results are consistent with those of\nprevious studies involving clinical outcomes of naturally endometrial preparation\ncompared either with artificial, or hormonally stimulated cycles ( Hancke  et al ., 2012 ;  Kim  et al ., 2010 ;  Kyrou  et al ., 2010 ).\nAlthough the natural protocol showed non-significant higher implantation rate versus\nhormonally stimulated cycle and artificial protocol (21%  vs.  13%\nand 15%), it was associated with less miscarriage (0 and 0  vs. \n6.8%). It seems that the natural cycle is at least safe and lacks consequences in\ncomparison to the other cycle regimens mentioned; besides, one has to consider\npatient convenience, preference and cost-effectiveness. This outcome is contrary to\nthat of Chang  et al.  who found greater miscarriage rates in the\nnatural cycle regimen using HCG versus artificial cycle. This inconsistency may be\nbecause Chang selected the samples' cycle regimen according to patient convenience\nand cost in their retrospective study, and we recruited the patients randomly ( Chang  et al ., 2011 ).\nImplantation is a multifactorial phenomenon, requiring synchronization between the\ndeveloping embryo and optimal endometrial environment ( Lee  et al ., 2006 ). To improve implantation\nrates, some authors have propounded optimal embryo conditions with the natural\nprotocol. To achieve a better outcome, Chang suggested transferring the vitrified\nblastocyst to a natural endometrium preparation ( Chang  et al ., 2011 ). Xiao proposes that the natural\ncycle is superior to reproductive outcome in comparison to artificial cycle when\nexcellent embryo conditions are met. For sample consistency, we decided to transfer\nexcellent or good quality blastocysts for all the cycle regimens ( Xiao  et al ., 2012 ).\nOur other important statistically relevant finding was that the mean endometrial\nthickness on the day of progesterone initiation or hCG administration was more\nfrequently found in the artificial and natural cycle groups than the mild induced\ncycle group; however, the frequency of triple endometrial patterns in the three\ngroups were comparable. We expected higher endometrial thickness in the artificial\ncycles owing to greater Estradiol levels in such cycles. Chang concluded that this\nalternation also occurs in the natural cycle due to the decidualization influence of\nHCG on the endometrium during the implantation ( Chang\n et al ., 2011 ). In our study, this might be a\npossible explanation for a high endometrium thickness in the natural cycle group.\nNevertheless, we found no association between reproductive the outcome and\nendometrium thickness or triple line pattern in the three groups. As we did not\neliminate some confounding variables, maybe these results need to be interpreted\nwith caution.\nOur findings are in accord with those from  Yoeli\n et al.  (2004)  indicating that no relationship was\nfound between decreased implantation or pregnancy rates and increased endometrial\nthickness in assisted reproduction. The present study raises the possibility that\nother factors, including patient's convenience and request, social status, and\nphysician's preferences may be considered in the choice between these three\nendometrium preparation protocols.\nMaybe, a weakness of our study is the lack of blindness that may cause potential\nbiases. In addition, we compared endometrial features on the day of progesterone or\nHCG administration amongst three cycle regimens. Whereas, the endometrium growth\ncontinues to the day of blastocyst transfer, may be the endometrial characteristics\nin the transfer day is not exhaustive. In addition, the number of patients in each\nof the groups was small; and with small sample size, caution must be applied, as the\nresult might not be a source of entire certainty. Our findings must be elucidated by\nwell-conducted RCTs with large-scale and controlled variables design.\n\nThe patients with normal ovarian function achieved the desired outcome using natural\nwith HCG as well as mild hormonally and artificial cycles. We recommend natural\nprotocol with HCG trigger as a therapeutic alternation for preparation of\nendometrium prior to vitrified thawed blastocyst transfer.","source_license":"CC-BY-4.0","license_restricted":false}