Evidence-Based Management of Infertile Couples with Repeated Implantation Failure Following IVF

Home , Recurrent miscarriage

200 Current Women’s Health Reviews, 2010, 6, 200-218 Evidence-Based Management of Infertile Couples with Repeated Implantation Failure Following IVF

Kim D. Ly1, Nabil Aziz2, Joelle Safi1 and Ashok Agarwal1,*

1Center for Reproductive Medicine, Cleveland Clinic, Cleveland, OH 44195, USA; 2Liverpool Women’s Hospital and The University of Liverpool, Liverpool, UK

Abstract: Embryo implantation depends on both embryo quality and the endometrial environment. Implantation failure has a complex, variable pathophysiology and is detrimental to the outcome of in vitro fertilization (IVF). Thus, patients with multiple implantation failure require an individualized approach to diagnosing and managing treatment options for future IVF cycles. These options should be based on concrete, unambiguous, consistent scientific evidence with randomized, controlled trials. We review and discuss 14 treatment options: (i) blastocyst transfer, (ii) assisted hatching, (iii) co-culture, (iv) preimplantation genetic screening, (v) hysteroscopy, (vi) sildenafil, (vii) salpingectomy for tubal disease, (viii) oocyte donation, (ix) transfer of six or more embryos, (x), intratubal embryo transfer, (xi) natural cycle IVF, (xii) antiphospholipid antibodies (APA) testing and treatment, (xiii) allogenic lymphocyte therapy, and (xiv) IV immunoglobin therapy. The approaches were evaluated based on available information from studies, expert opinions, consensus, etc. We conclude that blastocyst transfer, assisted hatching, salpingectomy for tubal disease, and hysteroscopy in IVF procedures are clinically effective. This review serves as a summary of current treatment options for clinicians to counsel patients and manage their expectations based on strong and reliable evidence. Keywords: Repeat implantation failure, in vitro fertilization, endometrial receptivity, blastocyst transfer, assisted hatching, hysteroscopy, salpingectomy.

INTRODUCTION the stroma of the uterine lining, forming a vascular connec- tion to the mother. The world's first baby conceived by in vitro fertilization (IVF) was Louise Joy Brown in 1978. Today, more than 3 The etiological causes of implantation failure include million children worldwide have been born as a result of embryo quality; endometrial receptivity; immunological fac- IVF. The numbers continue to increase as science discovers tors; uterine, tubal and peritoneal factors; and culture media ways to overcome barriers for various subgroups of infertile [3]. Poor response to superovulation and chromosomal ane- patients. However, despite the improvements in IVF tech- uploidy due to advanced maternal age negatively affect em- nology and methods, many couples experience multiple IVF bryo quality; suboptimal embryos are less likely to implant. failures. After each failed IVF attempt, pregnancy rates in The disruption in prostaglandin synthesis is one of many subsequent attempts decrease by as much as 57% with the factors that decrease endometrial receptivity in some patients most remarkable decrease after the third attempt [1, 2]. The prone to RIF [4]. Other cellular and adhesion pathways af- main causes of multiple IVF failures include: (i) poor fected by abnormal gene expression in the endometrium response to ovarian stimulation, (ii) repeated fertilization have been observed to be linked to RIF [5]. The cultured failure, (iii) repeated difficult transfers, and (iv) repeated endometrial cells of RIF patients were found to have differ- implantation failures (RIF). ent gene expressions than those found in the cultured endometrial cells of women who miscarried or had an ongoing Implantation failure is a major limiting step for IVF [3]. pregnancy [6]. From this observation, the differential gene The process of embryo implantation is described as having expression of RIF patients is assumed to negatively affect three phases: 1. Apposition: “unstable adhesion” of the trans- critical signaling pathways important for the development of ferred embryo to the surface of the uterine lining. 2. Attach- adhesion molecules in the embryo-endometrium bond and ment (adhesion): “stable adhesion,” believed to involve sig- may be linked to implantation failures. Immunological fac- naling back and forth between the embryo and the lining. 3. tors such as antiphospholipid antibodies (APA), abnormal Penetration (invasion): invasion of the trophectoderm cells expression of endometrial natural killer cells, cytokines [3], from the embryo through the surface of the lining deeper into local and systemic immune factors, anti-sperm antibodies, and anti-thyroid antibodies [7] have been found in significant

amounts among RIF patients and have been reported to af- *Address correspondence to this author at the Center for Reproductive fect implantation. APA interferes with the normal function Medicine Cleveland Clinic 9500 Euclid Avenue, Desk A19.1 Cleveland, of blood vessels by either causing narrowing/irregularity of OH 44195 USA; Tel: 216-444-9485; Fax: 216-445-6049; E-mail: [email protected] the blood vessels (vasculopathy) or by causing the develop-

1573-4048/10 $55.00+.00 © 2010 Bentham Science Publishers Ltd. Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 201 ment of blood clots in the blood vessels (thrombosis). In the and these advances have enabled researchers to culture zy- majority of cases, failed implantation appears to be related to gotes to the blastocyst stage prior to implantation in the the quality of embryos transferred rather than to the endo- uterus. The original intent of blastocyst transfer was to select metrial receptivity. Part of the evidence stems from the sig- the healthiest embryos for transfer. Thus, a single or the few- nificantly higher implantation rates found in egg donation est possible embryos may be transferred per cycle to reduce programs, even in couples that have failed IVF repeatedly high order pregnancies, which are associated with very real, using their own eggs. serious risks to mother and baby. Several advantages of culture and transfer of blastocysts make this option appealing In this paper, we focus on the options supported by clini- for RIF patients. Self-selection of blastocysts increases en- cal evidence to improve implantation and pregnancy rates for dometrial receptivity. The trophectoderm cells of the blasto- couples with multiple IVF failures. Evidence-based medicine has three levels of recommendation. Level A recommenda- cyst cross-talk with the endometrium and develop the ability to attach to the lining of the endometrial lining of the uterus, tions are based on good and consistent scientific evidence both of which are likely to improve implantation success. with randomized, controlled trials. At level B, recommenda- However, blastocyst culture and transfer usually is recom- tions are based on limited or inconsistent scientific evidence mended for patients with a good prognosis--those who are with clinical controlled trials, cohort, etc. Level C includes younger in age with > 6 oocytes available from an unevent- recommendations based primarily on consensus and expert opinion. Clinicians would benefit from knowing to which ful ovarian stimulation. Sequential media, a two-part culture medium, commences with the embryo initially grown in me- category a treatment option belongs to be able to counsel dia rich with pyruvate to Day 3. The embryo is then trans- patients and manage their expectations for future IVF cycles. ferred to the second media, rich with glucose, on Day 3 to We will examine 14 approaches to repeated implantation support embryo growth from the eight-cell stage to the blas- failure in IVF: (i) blastocyst transfer, (ii) assisted hatching, tocyst stage. (iii) co-culture, (iv) preimplantation genetic screening, (v) hysteroscopy, (vi) sildenafil, (vii) salpingectomy for tubal A study by Weissman et al. observed a significant im- disease, (viii) oocyte donation, (ix) transfer of six or more provement in pregnancy rates from blastocyst culture and embryos, (x), intratubal embryo transfer, (xi) natural cycle transfer to patients who were young, did not have multiple IVF, (xii) APA testing and treatment, (xiii) allogenic lym- IVF failures, who produced multiple oocytes, and whose phocyte therapy, and (xiv) IV immunoglobin therapy. (See zygotes developed into good quality cleavage-stage embryos also Table 1). [1]. However, for patients in the general population with a single failed IVF attempt, pregnancy and implantation rates Table 1. Fourteen Possible Approaches to the Management of were observed to significantly decrease with subsequent IVF Infertile Couples with Repeat Implantation Failure attempts employing blastocyst culture and transfer strategy [12]. In patients with multiple implantation failures, Cruz Blastocyst transfer et al. reported a positive correlation between blastocysts and improved implantation and pregnancy rates for RIF patients Assisted hatching (AH) in a non-randomized population [13]. However, a prospec- Co-culture tive, randomized control study by Levitas et al. that specifi- Preimplantation genetic screening (PGS) cally examined the effects of blastocyst transfer for RIF pa- Hysteroscopy tients compared with Day 2 embryo transfer found that blastocyst transfer was beneficial only in patients with an ac- Sildenafil ceptable response to ovarian stimulation [14]. However, Salpingectomy for tubal disease there was no difference in the incidence of multiple pregnan- Oocyte donation cies between the groups. Blastocyst transfer on Day 5 for RIF patients permits the selectivity of a higher quality em- Transfer of six or more embryos bryo after embryonic genomic activation has occurred [15]. Intra-tubal embryo transfer It may also decrease the rate of ectopic pregnancy because of Natural cycle IVF the larger diameter size of the Day 5 embryo; because the Antiphospholipid antibodies (APA) testing and treatment blastocyst normally resides in the uterus, an improvement to receptivity is also expected. Allogenic lymphocyte therapy I.V. immunoglobin therapy ASSISTED HATCHING (AH) Differences in embryo development during IVF treat- BLASTOCYST TRANSFER ment compared with in vivo are well-documented. One ob- Embryo transfer occurs either during the cleavage stage served difference is the hardening of the zona pellucida (ZP), on Day 2/Day 3 or during the blastocyst stage (the embryo which prevents the blastocyst from escaping the ZP during has inner cell mass, trophectoderm layer, and blastocoele) on hatching. The culture medium, which differs in its available Day 5/Day 6. Several studies have demonstrated higher rates nutritive source from the in vivo environment, and the cryo- of implantation and pregnancy with blastocyst culture and preservation method have been implicated as possible causes transfer compared with Day 3 cleavage stages [8-11]. of implantation failure [16]. Recent advances in culture technology have been made Assisted hatching was developed as a workaround to the due to a better understanding of early embryo metabolism, hardened ZP of in vitro embryos. It involves thinning or cre- 202 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al. ating a hole in the ZP, either of which can be performed by a cells were associated with a higher percentage of early mechanical (glass pipette), chemical (acidic Tyrodes’s or embryos developing to the eight-cell stage. enzymes), or laser method (contact and non-contact) with In some countries, including the United States, non- equal efficacy. The industry standard today includes the use human cell types have been banned by the Food and Drug of laser-assisted hatching for rapid and precise drilling. AH Administration due to concerns regarding disease transmis- is not a benign procedure since the time between the zona sion from the cell types to the developing embryo or mother. drilling and embryo transfer is a vulnerable period for the Consequently, the use of autologous endometrial cells has blastocyst as it is exposed to foreign elements, undefined gained in popularity [39]. Several different human cell types antibodies, or surveillance cells in the endometrial environ- are now available, including cumulus cells, luteinized granu- ment. However, blastocysts are believed to be more prepared losa cells, fallopian ampullary epithelial cells (FAEC), and for exposure to uterine lymphocytes and immune cells at this endometrial epithelial cells [40]. No randomized controlled stage [17]. In addition, an increased incidence of monozy- studies are available to compare and determine which of the gotic twinning may occur during the ablation procedure, human cell types provides maximum benefit. which in some reports has caused herniation and division of the inner cell mass into two [18]. This may result from (i) a In a study by Weichselbaum et al., very poor quality em- too-narrow opening of the ZP, trapping the blastomere in a bryos, in which more than 50% of embryo volume had frag- figure-eight formation, thus causing a subdivision of the mentation and the blastomeres were of unequal size with blastocyst to form twins or (ii) a premature hatching of the grainy to dark cytoplasm, were rescued from cleavage arrest blastocyst from a too-large opening combined with loosen- and degeneration when co-cultured with fallopian ampullary ing of the tight junctions between blastomeres to cause divi- epithelial cells. In addition, blastocyst formation increased sion of the inner cell mass to form twins [18]. by 56%, suggesting that co-culture may help embryos overcome developmental incompetence [40]. Eyheremendy et al. A Cochrane review of AH that included 28 randomized used the best embryos from women with multiple implanta- control trials from 24 publications or abstracts shows an in- tion failure to co-culture with monolayers of autologous en- crease in clinical pregnancy rates for women with RIF but dometrial cells [41]. Of the 68 patients who failed to become suggests additional research is needed [19]. There is no evi- pregnant after multiple IVF treatments, 39 become pregnant, dence supporting AH after one failed IVF attempt, but 19 attained a live birth, and 10 remained pregnant beyond 12 women with multiple implantation failures benefited most weeks. The study also suggested that an endometrial biopsy from AH [19]. The review’s validity, coupled with multiple should be performed about 7 days after ovulation and that good quality, randomized, controlled studies and strong rec- the co-culture medium contain both stromal and glandular ommendations, suggest that AH does improve implantation cells in the monolayer to improve implantation rates [41]. In rates for women with multiple implantation failures (3+) due 2008, Desai et al. reported for the first time the success of a to hardening of the ZP, occurring mainly from cultured IVF human endometrial culture system in a clinical environment, treatment or methods of embryo cryopreservation. thus supporting the two previously mentioned studies [42]. CO-CULTURE Interestingly, although the benefits of co-culture media have been demonstrated over the years, its prevalence is not Co-culture refers to the placement of human and non- widespread, perhaps because it remains a labor-intensive and human live cell types (feeder cells) alongside the embryo unproven technique [41]. Several limitations must be over- during in vitro culture. It has been suggested that co-culture come to realize its benefit. The most serious limitation is that improves embryo growth and development by improving in supplementation of human or non-human live cells with the vitro parameters--it removes toxic substances such as heavy culture media results in an uncontrolled and undefined al- metals and ammonium and free radicals [20]. teration to the conditions of the media with unknown growth Endometrial (Simon, Mercader et al. 1999, Weichselbaum, factors in unknown concentrations [43]. The introduction of Paltieli et al. 2002) and granulose [21-23] cells are com- sequential media in the past decade has slowly replaced not monly used in co-culture. Studies have demonstrated conclu- only non-human but also human cell types because of its sively the existence of cross-talk between endometrial cells safety and ease of use [44]. Additional research is needed in and embryonic cells, resulting in a paracrine release of mole- the area of co-culture, with special attention to the type of cules believed to improve implantation [20]. The presence cells and type of culture media (conventional or sequential of uterine epithelial cells, the first cells in contact between media), to support its use prior to its widespread adoption in maternal and fetal cells, may initiate a signaling cascade clinical settings. for cells from the blastocyst to become depolarized in pre- paration for endometrial attachment, thus improving embryo PREIMPLANTATION GENETIC SCREENING competence for implantation [24]. Chromosomal abnormalities have been widely reported The heterogeneous co-culture cell lines that have been to be a major cause of early spontaneous abortions in as utilized include human and bovine oviductal cells [25-30], much as 60% of the general population [45]. In couples with bovine uterine epithelial cells [31, 32], African Green Mon- multiple implantation failures undergoing fertility treatment, key kidney cells (Vero cells) [28, 33-35], ovarian cancer the frequency of chromosomal abnormalities appears to be cells [36], buffalo rat liver cells [37], and human skin fibro- higher regardless of maternal age [46-49]. Specifically, em- blasts [38]. One randomized study using conventional media bryonic aneuploidy in patients with multiple implantation compared the various types of co-culture cell lines and found failures was 54-57% compared with 35% in a control group that granulosa cells and bovine oviductal uterine epithelial Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 203

[50, 51]. Thus, preimplantation genetic screening (PGS) should focus on understanding the embryonic role in RIF by could be utilized as an appropriate means for selecting sampling the blastocyst stage where maximum embryonic normal embryos to improve implantation and pregnancy gene expression occurs [59]. rates and, ultimately, live birth rates in couples with RIF. The occurrence and pathology underlying complex However, recent studies present conflicting data supporting chromosomal abnormality (three or more whole chromo- the use of PGS for couples with RIF. some imbalance) was another characteristic abnormality The European Society of Human Reproduction and found in patients with RIF in a recent study using array Embryology (ESHRE) PGD Consortium Steering Commit- comparative genomic hybridization (CGH) analysis of tee reports the use of PGS as having little impact on improv- chromosomal material in Day 3 cleavage-stage embryos. ing the pregnancy rate for women with RIF [52]. In addition, Chromosomal breakage and failure of mitotic cell cycle Gianaroli et al. showed that PGS did not increase implanta- checkpoints to detect abnormalities has been suggested as tion or pregnancy rates per embryo transfer in RIF patients the cause of mosaic complex chromosomal abnormalities in [53]. Other data from ESHRE’s review, ESHRE PGD nearly 30% of RIF patients [57, 60]. This study was further Consortium IX, demonstrated that 57 IVF centers performed supported by the use of CGH analysis of blastocysts [48]. 748 IVF cycles with oocyte retrieval for RIF, which resulted Abnormal replication and segregation of chromosomes dur- in a 27% clinical pregnancy rate, 24% implantation rate, and ing early embryo development of RIF patients is likely an 11% delivery rate between January and December 2006 caused by maternal cytoplasmic factors or mutation in the [54]. cell cycle control genes [61]. Thus, the use of PGS in identifying abnormal embryos in advance to improve implantation For comparison, the 27% clinical pregnancy rate for PGS is restricted. in RIF patients is better than the 18% clinical pregnancy rate in the total patient population undergoing PGS for various Fluorescent in situ hybridization (FISH) is the most reasons. This latest data collection from 2006 highlighted commonly used method today to detect chromosomal ane- RIF as a main indication for PGS in fertility centers despite uploidy. However, FISH techniques can only analyze a small the ongoing debate regarding its efficacy [54]. In a retrospec- subset of the chromosomes, usually the most commonly in- tive study of 121 first PGS for RIF, multivariate logistic re- volved in aneuploidy. Embryonic mosaicism (multiple germ gression analysis was utilized to generate a predictive model cell lines in one embryo) further complicates analysis of test [55]. The model demonstrated that to have a 90% probability results in terms of reliability since only one blastomere, rep- of having an embryo transfer after PGS, the patient should resenting the entire embryo, is removed for testing. Accord- have at least 10 mature oocytes, eight normally fertilized ing to one study, 38% of embryos were incorrectly graded as oocytes, and six embryos for biopsy on Day 3. normal using a five-probe set on blastomeres [57]. In another study using a nine-probe set, 25% of embryos were misdiag- The cause of aneuploidy among RIF patients differs from nosed [62]. In the most recent study, which used a 12-probe other cases of random failed implantation. One main charac- set, 19% of embryos were incorrectly graded as normal [48]. teristic of chromosomal abnormality found in preimplanta- The acceptable error rates from the studies suggest FISH is tion embryos of couples with multiple implantation failures able to reliably detect aneuploidy in mosaic embryos and is the low probability of meiotic errors resulting in chromo- further implies that mosaic embryos have a sufficiently high somal abnormalities [56, 57]. Two studies by Mantzouratou ratio of abnormal-to-normal blastomeres for cleavage-stage et al. and Voullaire et al. have shown that meiotic errors are biopsy to serve as clinically useful. Despite several recent an unlikely cause in this group of patients and that chromo- advances in diagnostic methods, including whole genome somal abnormalities are reflective of an inefficiency in mi- amplification with comparative genomic hybridization and totic division due to abnormal cell cycle regulation by the the use of microarrays to overcome the limitations in FISH, embryo. This may explain the lower rate of success in the identifying complex chromosomal abnormalities has limited management of fertility for couples with RIF in studies using success, is labor-intensive, and costly [65]. polar body analysis. Currently, convincing evidence for the wide use of PGS In 2009, Fragouli et al. released a study comparing re- in RIF is insufficient. The technique did not improve implan- sults of PGS between polar body I, polar body II and blasto- tation rates for RIF patients [53, 63, 64]. Moreover, some cyst stages in which the aneuploidy rates were 36.5%, normal embryos might be lost due to the error rate. Further- 45.8%, and 45.2% after meiosis I, meiosis II, and mitosis, more, with the advent of less invasive methods for predicting respectively [48]. Though the numbers appear to indicate better quality embryos such as metabolomics and pro- that aneuploidy rates are higher after meiosis II, errors from teomics, PGS may become a less popular option. It is not meiosis I carried into meiosis II should be considered. The likely to happen anytime soon because no studies to date higher aneuploidy rate at the blastocyst stage may be ex- have identified the significant metabolites or proteins in- plained by the fact that the embryo is more vulnerable to volved in early development that are unique to embryos with developmental arrest between the four-cell and eight-cell a low implantation success rate. The role of PGS in the man- stage as it switches from maternal to embryo gene expres- agement of RIF patients remains unresolved. sion [58]. Maximum embryonic gene expression has been shown not to occur until the blastocyst stage. Therefore, the HYSTEROSCOPY disturbed immature cell cycle regulation increases the likelihood of chromosomal abnormalities, which persists to the Hysteroscopy is an invasive diagnostic procedure that is blastocyst stage, thus reducing the likelihood of successful used to visualize uterine pathologies, including submucous implantation. These observations suggest that future studies fibroids, polyps, intrauterine adhesions, and uterine malfor- 204 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al. mations (i.e. septate uterus) that may be associated with in- plasminogen activator inhibitor 1(PAI-1), and (iii) vascular fertility. Studies have shown the procedure to be more sensi- endothelial growth factor (VEGF) – 1154. The study found a tive, specific and accurate than pelvic ultrasound in evaluat- link between one or more of the genes and patients with re- ing uterine pathology for patients with RIF [65]. Currently, peat implantation failure. A significantly higher correlation hysteroscopy has not been adopted as part of the routine existed for RIF patients than the control group. More studies workup for infertility but rather as a secondary investigative are needed to confirm the findings, which could facilitate the measure to evaluate patients after three or more implantation clinician's ability to identify and counsel patients prone to failures [66, 67]. implantation failure [72]. Women with RIF have been shown to have a higher inci- Several studies have focused on increasing endometrium dence of uterine pathologies. In the most recent observa- thickness with sildenafil therapy for women with RIF. In tional study of 1475 women with RIF, an abnormality was 2000, Sher and Fisch applied sildenafil vaginally in women found in the uterus of 36.6% of these women (16.7% endo- with RIF and a thin endometrium to increase blood flow for metrial polyps, 12.5% endometrial adhesions, 1.5% endocer- endometrial growth. They hypothesized that a thicker endo- vical adhesions, 4.3% endometritis, 0.9% uterine septa, and metrium (> 8 mm) may improve implantation and pregnancy 0.8% submucous fibroids) [68]. The study reported that rates [73]. The preliminary study showed that three of the 22.2% of the population had a prior ultrasound screening four women had a successful pregnancy outcome after silde- with a false negative result and subsequent hysteroscopy nafil treatment during the proliferative phase. Two years intrauterine pathology (endometrial adhesions in 12.5%, en- later, Sher and Fisch released a follow-up study on 105 pa- dometritis in 4.3%, endometrial polyps in 3.3%, endocervi- tients who were given sildenafil; 70% developed endometrial cal adhesions in 1.5%, uterine septa in 0.5%, and submucuos thickness 9 mm with the remaining patients (30%) being myomas in 0.1%). The same study also compared the use 9 mm. Of the patients who developed endometrial thickness and non-use of hysteroscopy for women with RIF in a new 9 mm, 45% had a significant implantation rate [74]. The IVF attempt and showed a significant increase in the study does not explain why the majority (55%) of women implantation rate and pregnancy rate in the former group treated with sildenafil experienced optimal endometrial [68]. It strongly suggests the use of hysteroscopy for women growth but subsequent IVF failure. However, it highlighted after two failed IVF attempts for two reasons: (i) a the importance of endometrium quality [75]. significant number of uterine pathologies are undetected by In a case report on two patients with Asherman’s ultrasound and (ii) the significant success rate of ongoing syndrome, sildenafil improved endometrial thickness but did pregnancy after hysteroscopy [68]. For women with so by a much smaller margin. The endometrium of one endometrial polyps, a hysteroscopic polypectomy (surgical woman increased from 6.5 mm to 8.9 mm, and another removal of polyps) may be more efficacious than non- woman’s increased from 5.0 mm to 6.6 mm. Despite subop- intervening hysteroscopy [69]. timal endometrial measurements of 9 mm, both women A novel approach to the use hysteroscopy for evaluating had healthy offspring [76]. In a study by Paulus et al., the endometrial cell integrity has been described [70]. The tech- data did not support Sher and Fisch’s findings or those nique involves the application of chemical agents to high- of Zinger et al. In 10 women with at least one IVF failure, light cellular-level or mucosal anomalies of the endometrium sildenafil increased endometrial thickness, but only three (chromohysteroscopy). Preliminary studies on a small pa- of 10 patients had a successful pregnancy. However, the tient population have shown promising results for women heterogeneous and small patient population may have with RIF [70]. affected the results [77]. Natural killer cell activity levels, which have been re- SILDENAFIL (VIAGRA) ported as a predictor for recurrent miscarriages, may also be Sildenafil (Viagra®, Pfizer, New York, N.Y.), most nota- involved in RIF [78-80]. An increase in both peripheral ble for the treatment of male erectile dysfunction, is a vaso- blood natural killer cells and endometrial natural killer cells dilator used to improve vascular supply [71]. Endometrial appears to be associated with lower pregnancy rates in pa- receptivity is believed to improve with increased blood flow tients with recurrent miscarriages. Upon activation with ni- in the uterine arteries of women with thin endometrium to tric oxide, the natural killer cells release cytokines such as increase the thickness of the endometrium, the portion of the tumor necrosis factor-, which has been implicated as a uterus associated with successful implantation. Both the cause of implantation failure [81]. One study extended the quality and quantity of endometrial tissue are important to hypothesis to suggest that repeated implantation failures may consider in a strategy aimed at improving the endometrial be caused by high levels of peripheral natural killer cell ac- environment for a successful pregnancy. tivities. It showed that sildenafil lowered peripheral blood natural killer cell activity, thereby improving the local en- A recent case control study by Goodman et al. investi- dometrial immunological environment in women with multi- gated three gene polymorphisms that have been linked to ple IVF failure. However, more randomized control studies patients with repeat implantation failures in prior individual must be done to confirm these findings. It also will be impor- studies [72]. Successful implantation depends on the blasto- tant to determine the appropriate period to prescribe silde- cyst’s ability to infiltrate the endometrium and develop a nafil therapy [75]. sustaining blood supply, which requires the following genes to produce the necessary proteins for digesting the endo- In a pilot study, the amount of NO released by the em- metrial cellular matrix, regulate cell growth, and induce an- bryo in vitro correlated with implantation success. It was giogenesis: (i) p53 codon 7 tumor suppressor factor, (ii) reported that higher quality embryos producing an elevated Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 205 amount of NO in culture media had a higher success rate hydrosalpinx fluid, resulting in either endometrial alterations than the controls [82]. Thus, the use of sildenafil to block the that are hostile to embryo implantation and development breakdown of cyclic guanosine monophosphate (cGMP) [87], (ii) the mechanical wash out of the embryo [88] before causes an accumulation of NO. As a result of increased NO, it has a chance to implant, (iii) altered endometrial receptiv- radial uterine blood vessels dilate to increase the vasculature ity due to variations in the levels of certain biomarkers such and perpetuate endometrial growth. However, as NO induces as LIF [89] and v3 integrin [90, 91], (iv) release of intrau- natural killer cells to produce cytokines that can cause im- terine cytokines, prostaglandins, leukotrienes, and other in- plantation failure, it has been recommended that sildenafil flammatory compounds directly to the endometrium or via not be used five or more days prior to embryo transfer [74]. the circulatory or lymphatic system [92, 93], or (v) chronic On the other hand, one study using mouse embryos reported endometriosis caused mainly by asymptomatic Chlamydia that higher concentrations of NO in vitro and in vivo resulted trachomatis [94]. Thus, some studies support preventative in lower implantation rates in a dose-dependent manner [81]. salpingectomy for infertile patients meeting two criteria: (i) The same higher concentration of NO (1 mM) inhibited im- large hydrosalpinges visible by ultrasound and (ii) bilateral plantation in vitro as it did for mouse embryos. Cytostatic hydrosalpinges [85, 86]. The emphasis on the severity of the and cytotoxic effects resulting from an extended production tubal disease (see Table 2 for hydrosalpinx scoring system) period of NO in reproductive tissues to protect against infec- for salpingectomy as a treatment option is important since it tion, immunological reactions, and pathological conditions limits or removes chances for a natural conception and is (e.g. endometriosis, reproductive tract infections) is a sug- likely to result in disruptions to ovarian blood flow and nerve gested cause for the lower implantation rates found in the supply and reduce ovarian reserves [95] -- all of which are mouse embryo study [81]. important for follicle production, hormone production, and the number and quality of the ova [84]. An additional crite- The effects of sildenafil therapy on the endometrium and rion for irreversibility of the tubal condition should also be nearby environments should be better understood before it is widely adopted in IVF clinics for RIF patients. Natural killer included so that patients have the option to maintain their fallopian tubes. cells and NO play important roles in the female reproductive tract. Alterations in NO-synthase (an enzyme that converts A study by Dechaud et al. reported significantly higher the nitrogen in L-arginine to NO in the presence of NADPH implantation rates per transfer among women <41 years of and dioxygen) and/or NO production of these tissues could age who had experienced RIF after laparoscopic bilateral directly affect the development of human embryos, espe- hydrosalpingectomy compared with those who did not [96]. cially during the early stages of pregnancy. The pregnancy rate was also higher for the group with salpingectomy than the group without (23.5% and 9.9 %, SALPINGECTOMY FOR TUBAL DISEASE respectively, P value = 0.01). Bilateral salpingectomy also was reported to not only increase implantation rates but also Of all the different tubal pathologies, hydrosalpinx (dis- decrease the time to pregnancy. A group of women who un- tension of the fingered portion of the fallopian tube due to an derwent salpingectomy became pregnant within three IVF abnormal accumulation of fluid or water most likely result- attempts as opposed to a group who did not undergo ing from an inflammatory response) has the most detrimental salpingectomy in which some patients required as many as effect on implantation [83]. Generally, hydrosalpinx is 11 IVF attempts to become pregnant [96]. Furthermore, a caused by abortion, pelvic inflammatory disease, endometri- recent Cochrane study for the surgical treatment of tubal osis, previous operations, a history of tuberculosis and peri- diseases performed a meta-analysis on five randomized trials tonitis, or an unknown reason [84]. It usually occurs on both and found a significant increase in clinical pregnancy rates sides but can occur exclusively on one side. Interestingly, a and ongoing pregnancy rates for patients with hy- one-sided hydrosalpinx will usually correspond with an ab- drosalpinges who underwent salpingectomy and IVF (for normal fallopian tube on the opposite side. For women with the first time) as compared with those who did not undergo infertility problems, removing the fallopian tubes is a major surgical intervention [97]; thus, salpingectomy should decision. Nonetheless, hydrosalpingectomy has been re- be considered for all patients with ultrasound-visible ported as a promising option for a subgroup of RIF women hydrosalpinges. with severe tubal factor infertility [85, 86]. The main theo- retical reasons for its use as a treatment option are (i) the In contrast, there is an argument in preference of tubal embryotoxic effect of the fluid in the region by leakage of microsurgery, which is a less invasive procedure that pre-

Table 2. Hydrosalpinx Scoring System (Puttemans and Brosens 1996)

I. Simple hydrosalpinx represented by a thin-walled translucent hydrosalpinx with flattened and separated mucosal folds in a single lumen but without mucosal adhesions. II. Hydrosalpinx foliculans represented by a thin-walled hydrosalpinx with mucosal adhesions which can be focal or extensive. As a result the tubal lumen may become divided into locules by agglutinated folds forming compartments or pseudoglandular spaces. III. Thick-walled hydrosalpinx represented by an ampullary wall >2 mm thick and absent mucosal folds or some fibrotic fold remnants at most; tubal distension is less marked and the amount of invisible intralumina fluid is less abundant.

Source: Puttemans PJ, Brosens IA. Salpingectomy improves in-vitro fertilization outcome in patients with a hydrosalpinx: blind victimization of the fallopian tube?. Hum Reprod 1996; 11(10): 2079-81. 206 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al. serves the fallopian tubes, for women with less severe tubal from oocyte donation [100-102] to overcome factors that pathology [84]. The higher cost and greater time commit- have been implicated in implantation failure such as geno- ment make it a less popular option among clinicians. Also, type and age of the oocyte. [100]. Two committees, the no studies have been done to compare efficacy, recovery, American Society for Reproductive Medicine (ASRM) and psychological factors, risks, invasiveness of tubal microsur- Society for Assisted Reproductive Technology (SART) pub- gery (a surgical procedure to repair and open the fallopian lished the 2008 Guideline for Gamete and Embryo Donation, tubes that is sometimes referred to as tuboplasty) versus hy- which provides the most recent recommendations and infor- drosalpingectomy [98]. Perhaps a randomized controlled mation from the U.S. Center for Disease Control (CDC), study to compare salpingectomy and tubal microsurgery for U.S. Food and Drug Administration (FDA), and American RIF patients with various severity of hydrosalpinx could Association of Tissue Banks (AATB) for optimal screening determine the optimal procedure in the appropriate patient and testing of oocyte donors and recipients (see Table 3). population. Additional randomized controlled studies are The guideline specifically identifies women with “multiple needed to clarify whether unilateral or bilateral salpingec- previous failed attempts to conceive via ART” as an indica- tomy is more appropriate and necessary than tubal surgery to tion for oocyte donation [99]. minimize unnecessary removal of healthy fallopian tubes More recently, the advancement of cryopreservation [97]. methods and technologies for freezing and storing of oocytes has increased the number of available unused oocytes [103]. OOCYTE DONATION Patients undergoing controlled ovarian stimulation to cryo- preserve their oocytes prior to treatment for illnesses that In the past two decades, oocyte donation has become a pose a serious threat to their future fertility (e.g., cancer) valuable and effective therapeutic option for infertile patients have the option to donate their extra oocytes for research or a who choose assisted reproductive technology (ART) and has donor bank once they no longer need the oocytes. Accord- become an acceptable indication for patients with multiple ingly, RIF patients have a choice between using fresh and implantation failures [99]. Egg donors are prescreened for medical and physical history of infectious disease, genetic cryopreserved oocytes. Although the latter option is avail- disorders, polycystic ovarian syndrome, serious malforma- able and supported by the ASRM as a fertility preservation tions (resulting in severe functional or cosmetic handicap strategy, there is not enough evidence to support its safety such as spina bifida or heart malformations), mental health, and efficacy at this time [104]. For the first time, a recent and uterine pathologies (for accessibility to oocyte retrieval) prospective, randomized study comparing fertilization and and then selected based on age and phenotypic match to the embryo development rates and ongoing pregnancy rates recipient [99, 100]. The screening process also extends to the found no difference between fresh oocytes and vitrified oo- oocyte recipients (medical, physical, psychological, and so- cytes fertilized via intracytoplasmic sperm injection and then cial) and their partners (for infectious disease and paternal developed in vitro [105]. Further clinical studies are needed factors) to identify other possible sources that may adversely to clarify the long-term safety concerns to support the use of affect the outcome. cryopreserved oocytes.

Three groups of patients, those with (i) repeat implanta- TRANSFER OF SIX OR MORE EMBRYOS tion failure and no other indications, (ii) RIF in combination with advanced age, or (iii) RIF in patients with balanced A definitive answer to the question of what constitutes translocations in homologous chromosomes, may benefit the proper number of embryos for transfer in IVF has eluded

Table 3. ASRM and SART Guidelines for Evaluating the Oocyte Recipient

I. Provide psychological counseling by a mental health professional and further psychological consultation if necessary prior to consent. II. Obtain medical physical examination and reproductive history to detect reproductive abnormalities. Provide treatment as appropriate prior to use of donor oocytes. III. Complete a general physical exam and pelvic exam. IV. Assess the uterine cavity with hysterosalpingography or similar device to detect any significant uterine abnormality. V. Other recommended tests including: a. Blood type, RH factor, and antibody screen b. Rubella and varicella titers c. HIV-1 and HIV-2 d. Serologic test for syphilis e. Hepatitis B surface antigen f. Hepatitis B core antibody (IgG and IgM) g. Hepatitis C antibody h. Cervical cultures or similar tests for Neisseria gonorrhoeae and Chlamydia trachomatis Source: Guidelines for gamete and embryo donation: a Practice Committee report. Fertil Steril (2008); 90(5 Suppl): S30-44. Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 207 clinicians and scientists in the field for the last decade. strictions by considering the transfer of more embryos than In 1995, Azem et al. reported a significant increase in recommended only “in exceptional cases when women pregnancy rates with the transfer of six or more embryos in with poor prognoses have had multiple failed fresh IVF-ET comparison with five in women with repeated implantation cycles” with a level of evidence/recommendation III-C failure (at least four prior failed IVF-ET attempts) [106]. No [111]. The most recent Cochrane systematic review on the other published evidence has demonstrated improved preg- number of embryos to transfer in IVF compares pregnancy nancy or live birth rates after the transfer of more embryos in rates and chances of multiple pregnancies following single subsequent IVF-ET cycles than the number transferred in versus double, three and four embryo transfer in fresh IVF previous failed cycles. Despite this lack of evidence, the treatments with various results, reflecting the experience of transfer of greater numbers of embryos than the recom- selected young women in a single fresh cycle of IVF/ICSI. mended guidelines in women with multiple fresh IVF-ET As such, data concerning older women and women with failures is commonly performed in practice. This has also previous multiple failed IVF attempts have yet to be assessed been extended to women predicted to have a poor conception [112]. prognosis based on indicators such as embryo quality, also IVF practices will most likely be modified as scientific with little supportive evidence regarding efficacy [107]. advances allow a more accurate assessment of the implanta- On the other hand, women undergoing IVF with multiple tion potential of a given embryo, which will most likely embryo transfer face an increased risk of higher-order multi- include the determination of the embryonic genome and ple pregnancies (HOMP) with their known medical, social, the metabolic and proteomic fingerprints of viable versus and economic consequences. nonviable embryos using microarray technology. The first guidelines on the number of embryos to transfer were issued by the ASRM in 1996 [108]. These guidelines INTRATUBAL EMBRYO TRANSFER were revised four times since then, with a subsequent reduc- Embryo transfer is a vital step of IVF treatments. There- tion in HOMP. As recently as 2003, the triplet rate for IVF in fore, in patients with repeated implantation failure, clinicians women younger than 40 was approximately 6%. In 2007, it frequently focus on the transfer procedures to enhance em- was less than 2%, and this decrease is directly related to the bryonic implantation following IVF [113, 114]. Tubal trans- decrease in the number of embryos transferred per cycle fer of embryos or zygotes has been widely utilized as part of [109]. the arsenal in the treatment of difficult cases of repeated IVF In the latest ASRM guidelines, issued in November 2009 failure [115, 116] although it has never been proven to be [110] (see Table 4) the maximum recommended number of beneficial beyond any doubt. embryos to transfer is five, and this number only applies to Zygote intrafallopian transfer (ZIFT) has been hypothe- women aged 41-42. For all patients with one or two previous sized to have many advantages over transcervical embryo failed fresh IVF cycles, the guidelines also recommend transfer (ET), mainly that it provides a “natural” growth mi- transfer of one supplementary embryo (in comparison with lieu for zygotes under physiological regulation with numer- the standard recommended number according to age group, ous growth factors and cytokines from the tubal fluid, which see Table 4), after proper counseling regarding the risk of helps these zygotes attach to the uterus with greater synchro- HOMP and justification in the patient’s medical records nization, thus enhancing implantation potential [117]. After [110]. This only brings the number of embryos to transfer to all, natural Day 1/Day 2 embryos belong in the fallopian six in the 41-42 age group if previous failed IVF attempts are tubes and not the uterus. Environment or in vitro culture sys- documented. For all other age groups, the numbers are even tems play a crucial role in the early development stages of more limited (as low as 1-2 transferred embryos for women the embryo, and a suboptimal environment may have ad- younger than 35). The Society of Obstetricians and Gyne- verse effects. The ZIFT technique also prevents spillage of cologists of Canada (SOGC) goes even further in their re- embryos after transcervical ET and solves the problem of

Table 4. Recommended Limits on the Numbers of Embryos to Transfer

Age

Prognosis < 37 yrs 35-37 yrs 38-40 yrs 42-42 yrs

Cleavage-stage embryos Favorable* 1-2 2 3 5 All others 2 3 4 5 Blastocysts Favorable* 1 2 2 3 All others 2 2 3 3

*Favorable = first cycle of IVF, good embryo quality, excess embryos available for cryopreservation, or previous successful IVF cycle. 208 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al. technically difficult ET in patients with cervical stenosis dence, ZIFT continues to be proposed in clinical practice [61]. Nowadays, the environmental advantage of tubal trans- exclusively to RIF patients [124]. fer seems limited since laboratory conditions have improved along with the composition of culture media over the last NATURAL CYCLE IVF two decades. Despite the fact that the first IVF/ET baby was born in Although initial retrospective reports of ZIFT showed 1978 after a natural unstimulated cycle, this technique soon higher pregnancy rates than with intrauterine ET [118-120], was practically abandoned, mainly because of the very high other investigators have found that the main value of ZIFT is cancellation rates. Controlled pharmacological ovarian hy- limited to RIF cases [115]. ZIFT was reported to be a valid perstimulation became the standard treatment in IVF cycles alternative to standard ET in most subgroups of patients with of high- and normo-responder patients [126]. However, natu- either first or multiple failed attempts at IVF [114, 115]. It ral cycles have regained attention in poor-responder patients, was demonstrated that in patients with tubal factor and a where only very few follicles can be recruited and very few confirmed patency of one fallopian tube, ZIFT can be ap- oocytes can be retrieved after controlled ovarian hyperstimu- plied successfully as a treatment for RIF; the pregnancy rates lation. In these patients, natural IVF cycles may offer com- and implantation rates for all ZIFT cycles in RIF patients parable outcomes (comparable number of follicles) [127- were 35.1% and 11.1% - significantly higher PRs and IRs as 128], reduced side effects such as multiple pregnancy and compared to standard transcervical ET [121]. This study ovarian hyperstimulation syndrome, and may represent a concluded that ZIFT should be recommended to IVF patients more cost-effective and patient-friendly alternative [129]. with a mild form of tubal factor and proved patency of one A recent retrospective study of 500 consecutive natural tube, whereas in severe forms, salpingectomy remains the cycles demonstrated that IVF in natural cycles is an afford- recommended treatment. able and valid alternative in poor-responder patients [129], in The interest in intratubal transfer was greatly diminished accordance with the data reported in an earlier meta-analysis after the results of a meta-analysis and a randomized, con- [130]. Controversial data exist as to the efficacy of cycles trolled trial that failed to demonstrate any advantage for with minimal stimulation(i.e. GnRH antagonist plus mild ZIFT over standard IVF/ET [122]. Furthermore, a study by gonadotropin stimulation), and there are currently no studies Aslan et al. including 229 patients with RIF showed compa- in the literature comparing natural versus minimal stimula- rable outcomes following ZIFT and transcervical ET [123] . tion cycles [129]. However, these studies are not readily comparable due to Moreover, to address the specific issue of implantation different methods and selection criteria. It is likely that a failure, a better understanding of the key determinants of patient’s age, number of previous IVF attempts and etiology successful implantation is warranted, one of which is endo- of infertility may influence the results [123]. A number of metrial receptivity. The endometrium is receptive to blasto- drawbacks to the use of ZIFT exist, such as the need for gen- cyst implantation during a limited spatio-temporal window, eral anesthesia, laparoscopy, and heavy medical equipment, called “the implantation window” [131]. In humans, this which increases the medical risks to the patients, as well as period begins 6–10 days after the LH surge and lasts for 48 h the expenses related to the need for operative conditions [132, 133]. Successful implantation depends on synchroniza- [123]. An association between tubal embryo transfer and the tion between the developmental stages of the embryo itself increased risk of ectopic gestation has been inconsistently and the complex endocrine environment [134, 135]. There is reported [122, 123]. evidence in the literature suggesting that controlled ovarian hyperstimulation (COH) can alter endometrial receptivity ZIFT is normally performed in the pronuclear stage, one [136, 137]. Supraphysiologic doses of hormones can cause day after egg retrieval. For practical reasons, ZIFT is some- asynchronism between the embryo and endometrium and times deferred by one day, and cleavage-stage embryos are altered concentrations of growth and adhesion factors, caus- transferred two days after egg retrieval--a procedure termed ing implantation failure [138, 139]. “embryo intra-Fallopian transfer,” or EIFT [114]. In a recent retrospective study by Weissman et al., ZIFT and EIFT Ledee-Bataille et al. conducted a study in which endo- transfers had comparable outcomes in regards to implanta- metrial CD56 bright cells (uterine natural killers or uNK) tion and pregnancy rates as well as in all other study parame- were immunostained. They found elevated numbers of endometrial NK cells in RIF patients after COH cycles and ters such as miscarriage rates, ectopic pregnancy rates, and significantly lowered numbers after natural cycles [140]. multiple pregnancy rates, which were found to be unaccepta- Data suggest that uNK may be directly or indirectly involved bly high [124]. in controlling the early steps of the implantation process, in In conclusion, RIF patients are a heterogeneous group part because of their role in vascular remodeling, specifically with unclear and various pathophysiological diagnoses. The spiral uterine arteries [141]. Furthermore, on the molecular potentially high efficiency of ZIFT in RIF patients is only biology level, alterations in endometrial gene expression partially understood, with a commonly accepted interpreta- have been reported with the use of gonadotropins in stimu- tion being that ZIFT embryos possibly are protected from lated cycles [142]. More recently, Haouzi et al., using DNA expulsion from the uterine cavity by junctional zone contrac- microarrays of endometrial biopsies, identified for the first tions as well as from the introduction of cervical micro- time five genes that are up-regulated during the implantation organisms into the uterine cavity by the transfer catheter window and proposed them as new biomarkers for explora- [116, 125]. Meanwhile, and despite the scant available evi- tion of endometrial receptiveness, including during a natural Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 209 cycle. This novel strategy could prove useful in patients with more comprehensive range of APA [156]. Coulam et al. poor implantation after IVF or ICSI [143]. found a 22% prevalence of seven APAs in women experiencing implantation failure after IVF/ET. Only 4% of In conclusion, in light of the many potential advantages women with positive antibodies would have been detected if of natural cycle IVF, and with the many improvements in laboratory conditions and fertilization techniques such as only ACL were tested and only 14% if APS were added to ACA [156]. Unfortunately, with the exception of ACL, no ICSI, it seems worthwhile to re-evaluate the place of natural universally accepted standard for the determination of APA cycle IVF in the arsenal of fertility treatments, especially in concentrations exists, which adds to the confusion. Actually, RIF patients. A randomized, controlled trial, comparing the most broadly used clinical assays for these antibodies test natural cycle IVF with current standard practice, is justified. for ACL antibodies using enzyme linked immunosorbent assay (ELISA) and lupus anticoagulant (LA) [156]. It is ANTIPHOSPHOLIPID ANTIBODY (APA) TESTING noteworthy here to emphasize the fact that APA can be AND TREATMENT found in low concentrations in as many as 16% of “normal” Antiphospholipid antibodies (APAs) [144] are acquired controls, i.e. healthy fertile women [153]. immunoglobulins or monoclonal antibodies (IgG, IgM, The association of antiphospholipids with RIF has been and/or IgA) directed against negatively charged membrane shown in some early studies [145, 153, 154, 156, 158], but phospholipids, which were characterized as thrombophilic large prospective studies failed to reveal an association [155, factors because of their association with slow progressive 157, 159, 160]. A meta-analysis considered the effect of thrombosis and infarction in the placenta, leading to utero- APAs on the likelihood of IVF success and concluded that placental insufficiency [145]. With regard to implantation testing for APAs was unjustified in patients undergoing IVF and pregnancy, it is more appropriate to classify APAs as [161]. However, these results did not close the debate be- autoimmune factors because of the complex nature of their cause of the heterogeneity of the cohort studies, the popula- interactions [7]. APAs have been shown experimentally to tions included, because the RIF group of patients was not block in vitro trophoblast migration, invasion, and syn- addressed specifically [162]. A strong association was dem- cytialization, to reduce trophoblast production of the vital onstrated between antibodies to the cofactor 2 glycoprotein hormone human chorionic gonadotrophin [144] and activate 1 and IVF implantation failure [153]. Antibodies to annexin- complement on the trophoblast surface inducing an inflam- V, which acts as an inhibitor of phospholipid-dependent co- matory response [146]. Increased concentration of APA in agulation and may be necessary for trophoblast differentia- the follicular fluid, which was once thought to be an expla- tion, were found to have a significantly greater incidence nation for the direct effect of APA on the implanting embryo (8.3%) in women with RIF than in controls (1.1%) [163]. [147], is still debatable. A recent study demonstrated that Other findings by Geva et al. suggest that although APAs this increased concentration had no adverse effect on the may be important in recurrent fetal loss and spontaneous reproductive outcome of women undergoing IVF [148]; an- abortions, neither the serum concentration nor the number of other study found a significant relationship between follicu- positive APAs appear to have significance in recurrent im- lar fluid APA concentrations and fertilization rates in IVF plantation failure, cumulative pregnancy, or live birth rates failure patients [149]. [164]. According to the ASRM (2008), no association is pre- APAs associated with reproductive failure are lupus anti- sent between APA abnormalities and IVF success, there is coagulant (LCA), anti-cardiolipin (ACL), anti-phosphatidyl no indication for the assessment of APA in couples undergo- serine (APS), anti-phosphatidyl inositol (API), antiphos- ing IVF, and therapy is not justified on the basis of existing phatidyl glycerol (APG), anti-phosphatidyl ethanolamine data [165]. (APE), and phosphatidic acid (PA) [150]. Not only do APAs Despite the uncertainty concerning the pathophysiology bind to their direct antigens, they also bind to plasma-bound of APAs in reproductive failure, their presumed thrombotic proteins and co-factors such as 2 glycoprotein I (2GPI), effects have led to the widespread use of heparin and aspirin the most studied anti-phospholipid protein, and prothrombin, for women with RIF [145]. Heparin is thought to protect the which is also important in this role [151]. There is evidence trophoblast from injury by inhibiting the binding of phos- that the presence of 2GPI on trophoblast and decidual cell pholipids with antibodies, thus promoting implantation and membranes might explain the clinical association between placentation [166]. Only a very few randomized, placebo- recurrent fetal loss and 2GPI-dependent APA and the controlled studies evaluating the benefits of heparin and as- pathogenic role for these antibodies at the same time [152]. pirin for APA-positive women with RIF have been under- Antinuclear antibodies (ANA) also may be associated with taken. While some evidence exists that treatment with reproductive failure, but they were shown to be positive in unfractionated heparin and low-dose aspirin can improve 9% of normal fertile women and appear to lack specificity in live birth rates [167], other studies have shown that neither low titre [153]. implantation nor pregnancy rates are improved with heparin Even after 20 years of investigation, the role of APA re- and aspirin [168, 169]. A recent randomized, placebo- mains elusive when it comes to assisted reproduction, mainly controlled cross-over study in patients with strictly defined because most of the groups reported an increased prevalence RIF did not show any benefit of heparin and low-dose aspi- of APA in infertile patients [154, 155], but the evidence is rin in patients seropositive for at least one antiphospholipid much less definite with respect to IVF outcome [156, 157]. (APA), antinuclear (ANA), or beta(2) glycoprotein I autoan- An explanation for the conflicting evidence might be related tibody, when the outcome measured was ongoing pregnancy to differences in antibodies tested. Some groups solely tested or implantation rates [169]. for ACL, LCA, or ANA [154], while others evaluated a 210 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al.

The most recent Cochrane review by Empson et al. ex- NATURAL KILLER CELLS amined the outcomes of all treatments to maintain pregnancy A difference is thought to exist between the uterine and in women with prior miscarriages and positive APA. The peripheral natural killer (NK) cells of women with recurrent results found that only unfractionated heparin combined with pregnancy loss (RPL) compared with controls. Higher num- aspirin appeared to reduce pregnancy losses (by 54%) when bers of uterine NK cells have been found in the preimplanta- compared with aspirin alone. However, these results were tion endometrium of women with RIF [178]. However, this only based on two small trials, one of which lacked satisfac- abnormality was only part of a composite range of immune tory allocation concealment. Low molecular weight heparin and vascular abnormalities found in the endometrium of RIF (LMWH) combined with aspirin had no statistically signifi- patients [178, 179]. The uNK from nonpregnant RPL patients cant effect when compared with aspirin or intravenous im- who exhibit lower CD56 expression (classified as CD16+ munoglobulin (IVIg) [170]. Aspirin alone had no significant CD56dim) are more frequent than CD16+CD56bright, as effect on any of the outcomes examined; corticosteroids did opposed to fertile controls [180]. Non-pregnant RPL patients not show any benefit but demonstrated increased adverse also show evidence of increased numbers of activated NK outcomes [170]. IVIg did not significantly differ from pred- cells in peripheral blood mononuclear cells [181]. Kwak et nisone or aspirin in outcomes [171] and was shown to have al. also observed the up-regulated expression of CD56＋, lower live births rates than LMWH plus aspirin [172, 173]. CD56+-CD16＋, and CD19＋cells in peripheral blood lym- The beneficial effect of IVIg has only been proved only in phocytes in pregnant women with RPL [182]. Moreover, uncontrolled studies [174]. Aoki et al. also reported that high preconceptional NK cell To date, the combined use of low-dose aspirin and hepa- activity was associated with higher abortion rates in the next rin is considered standard therapy for women seropositive pregnancy [80]. Studies of immune factors investigated at for APAs, despite the lack of adequate, prospective, random- the time of miscarriage showed that deficiency of CD56 ized, placebo-controlled studies addressing the RIF group of bright natural killer cells in the decidua and high natural kil- patients specifically. Caution must be exercised in recom- ler cell cytotoxicity in peripheral blood monocyte cells mending any given treatment. (PBMCs) increase the risk of euploid miscarriage [183-185]. In conclusion, in women with RIF, no consensus exists CYTOKINES regarding testing for APA, assays to be used, auto-antibodies to test, definitions of patient groups or therapy for seroposi- Strong evidence exists that locally secreted cytokines tive patients [175]. Patients must, therefore, be counseled control the implantation process and can cause implantation failure [186, 187]. The Th1 cytokines include IL-2, IFN and prior to starting any treatment that no clear evidence of TNF, and the Th2 cytokines include IL-4, IL-5 and IL-10. benefit for anticoagulation exists [116]. Evidence suggests that the mean of the Th1:Th2 ratio in patients with RPL and in patients with multiple implantation ROLE OF IMMUNE MECHANISMS IN RECURRENT failure after IVF-embryo transfer [188] is significantly PREGNANCY LOSS AND RIF higher than in normal fertile women. This predominance of Background Th1 cytokines was demonstrated to exist in endometrial cells as well as peripheral blood mononuclear cells before preg- The immune system of a patient with a successful pregnancy [187, 189, 190] and at the time of miscarriage in de- nancy has been considered a paradox since Medawar in 1953 cidual cells [191]. However, there are significant discrepan- described the embryo as a semi-allogen, but one that is pro- cies in the results of the different studies, as some suggest tected from allogenic recognition by antigenic immaturity, that Th1 cytokines production was higher in normal women possibly explained by non-classical class I HLA molecules. than in RPL patients in early pregnancy [192], and others Since then, numerous studies have supported the theory of even found that the production of Th1 and Th2 cytokines alloimmune causes as an explanation for miscarriages and was similar in RPL patients who subsequently had successful implantation failure. The hypothesis is that the absence of or failed pregnancies [193]. Th1 dominance may well be a such alloimmunoprotective mechanisms would result in allo- result of the miscarriage rather than a cause, and much more immune-mediated miscarriage [176]. Moreover, Wegmann basic knowledge is needed about the complex cytokine net- et al. suggested that successful pregnancy might result from works in pregnancy and the correlation between cytokine the predominance of T helper 2 (Th2) cytokines over T production in peripheral mononuclear cells and decidual helper (Th1) cytokines [177]. In spite of the central role at- lymphocytes [194] before tests measuring cytokines can be tributed to immunology in reproductive failure and the in- introduced in clinical practice. With further research and tense debates on its scope, no appropriate diagnostic strategy newly discovered cytokines, it is now clear that acceptance has been established to date [176]. Genetic and immunologi- of the Thl:Th2 paradigm as a single explanation for implan- cal factors interact with each other in a complex network of tation failure would be an overly simplistic approach to a antibodies, adhesion molecules, metalloproteinases, natural very complex mechanism [195]. Other cytokines, particu- killer cells, and cytokines [150]. Other factors influencing larly leukemia inhibitory factor (LIF), recently have been reproduction and implantation include human leukocyte an- shown to play a role in women with RIF [196]. Mannose- tigen expression, antisperm antibodies, integrins, leukemia binding lectin, a constituent of the innate immune system inhibitory factor, cytokines, antiphospholipid antibodies, that modulates cytokine production by monocytes [197], was endometrial adhesion factors, mucin-l, and uterine natural shown to have significantly lower levels in women with RPL killers [116, 150]. [198-200]. Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 211

TREATMENT WITH INTRAVENOUS IMMUNO- no-treatment group [214]. However, the most current GLOBULINS (IVIg) Cochrane review shows no significant increase in the overall pregnancy success rate over placebo or no treatment [215]. The mechanism of action of IVIg, a fractionated blood product, is multifactorial [201]. It is involved in a number of In conclusion, IVIg, an expensive treatment with possible processes, including modulation of T cells, B cells, NK cells, side effects, is widely used off-label in the treatment of early monocytes, and macrophages; down-regulation of antibody reproductive failure. Systematic reviews have generated in- production; inhibition of antibody function; and modulation conclusive results so evidence concerning its efficacy is con- of complement activation [202]. Immunoglobulins develop troversial. Rigorous randomized, controlled trials studying their suppressive activity in vitro through the CD200 toler- the efficacy of the treatment on various subgroups of RIF ance-signaling molecule, which is released from the surface patients and additional measurements of CD200-dependent of subsets of blood mononuclear leukocytes and may bind to IVIg effects should be undertaken to solve the current con- IVIg [203]. CD200 is known to promote generation of regu- troversy [211]. latory T cells in mice [204]. A more recent report suggests IVIg suppresses NK activity, specifically the CD56 bright ALLOGENIC LYMPHOCYTE THERAPY subset of NK cells found at the feto-maternal surface [205]. During pregnancy, the paternal human leukocyte antigen Additionally, one underlying mechanism may be the restora- (HLA) is recognized by the maternal immune system, which tion of Th1/Th2 balance with dominant Th2 [201]. High- induces production of several alloantibodies. These alloanti- dose IVIgs nearly always have been combined with corticos- bodies include anti-paternal cytotoxic antibodies (APCA), teroids or anti-thrombotics, so that their precise efficacy anti-idiotypic antibodies (Ab2), and mixed lymphocyte reac- cannot be readily estimated and is practically hard to assess tion blocking antibodies (MLR-Bf). Once expressed, they [201]. may coat the trophoblast to render it undetectable by the ma- A review of the literature concerning IVIg treatments ternal immune response system [216]. A reduction in or the yields conflicting results. A meta-analysis of six trials by absence of these alloantibodies during pregnancy may cause Daya et al. demonstrated a lack of clinical efficacy of IVIg fetal loss [201, 217, 218]. As has been shown repeatedly, on live birth rates [206], and a prospective, randomized, increased sharing of HLA may prohibit the mother from double-blinded, and placebo-controlled study by Stephenson producing these alloantibodies, leading to an increased ten- et al. observed no differences between the IVIg-treated and dency toward repeated fetal loss [201]. Maternal immuno- the placebo groups [207]. A prospective, randomized, dou- modulation via transfusion of paternal leukocytes (lympho- ble-blinded, and placebo-controlled study by Coulam et al. cytes) prior to conception has been proposed as a solution to demonstrated the efficacy of IVIg treatment in increasing the RIF [219-221], while the use of third party donor white cells percentage of live births among women experiencing unex- or trophoblast membranes transfusions have been largely plained RIF [208]. In a randomized study by Triolo et al., abandoned because of doubts about efficacy [201]. Allogenic IVIg was less efficacious than low-dose aspirin and low mo- lymphocyte isoimmunization (ALT) also has been proposed lecular weight heparin in increasing live births rate [172]. On to solve the Th1/Th2 paradigm by shifting the balance to- the other hand, a randomized controlled trial by Christiansen wards Th2 cytokines, thus enhancing the implantation proc- et al. showed no improvement in live birth rates with IVIg ess [181, 188]. compared with placebo, but thye suggested a possible bene- The first randomized, controlled study using ALT with ficial role of IVIg in women with secondary recurrent mis- paternal PBMCs showed a 24% increase in successful preg- carriage [209]. This effect was confirmed by the review of nancy rates [222]. However, subsequent trials provided con- Hutton et al., although the data concerning primary recurrent flicting results due to variations in cell numbers, number of miscarriage was inconclusive [210]. A meta-analysis of ran- injections, routes of administration, and types of placebo. domized and cohort controlled trials of IVIg in RIF patients These differences make comparisons and meta-analyses dif- showed a significant increase in the live birth rate per ficult to realize [209]. A subsequent intention-to-treat meta- woman (p=0.012) and number needed to treat for one addi- analysis by the Recurrent Miscarriage Immunotherapy Trial- tional live birth = 6) [211]. Relevant variables appeared to ists Group showed an increased live birth rate of approxi- be selection of patients with abnormal immune test results mately 9–10% [223]. The number needed to treat for an ad- and properties of IVIg preparations, as different biological ditional live birth was limited to three to four women with preparations vary significantly in their ability to suppress NK primary recurrent pregnancy losses who were seronegative activity in vitro. Another variable is the scheduling of IVIg for autoantibodies (ANA and ACL) [219]. Since then, the treatment, as it is argued that pre-conception treatment is beneficial effects of ALT in RIF patients has been demon- better in both primary recurrent miscarriage patients and in strated primarily by non-randomized studies. In a retrospec- IVF failure patients [211, 212]. An observational pilot study tive, non-randomized review of 686 couples referred for found that elevated numbers of NK cytotoxic CD16+ CD56+ ALT, Kling et al. found a temporary beneficial effect lasting cells are independent predictors of treatment success, and for six months after immunization; this effect seemed to be that IVIg ameliorated the numbers of these NK cells in RIF most pronounced in couples who failed three or more cycles [213]. Another observational study by Winger et al. studied of IVF-embryo transfer [224]. Despite the lack of random- for the first time a subset of IVF patients with elevated ized trials for RIF, a large randomized trial in RSA patients Th1/Th2 cytokines and showed an improvement of implanta- failed to show any beneficial effect of ALT [225]. Pooling tion and live birth rates for the groups treated with IVIg and the results of the randomized and non-randomized studies, both IVIg and TNF-alpha inhibitors as compared with the Pandey et al. showed that 67% of RSA patients who re- 212 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al. ceived paternal lymphocyte immunotherapy had successful EXPERT COMMENTARY pregnancy outcomes in comparison to 36% success in women with RSA in the control group who received either Having to endure not one but two or more failed rounds autologous lymphocytes or no therapy [220]. of IVF is painfully frustrating to the patient as well as to the clinicians and technicians involved. We discuss 14 current In a double-blind, placebo-controlled trial in women with options that are supported by varying degrees of scientific unexplained RSA, immunization with paternal lymphocytes evidence. Clinicians would benefit from knowing which proved to be beneficial over autologous (maternal) lympho- treatment options are proven in order to counsel patients cyte therapy [221]. On the other hand, a Cochrane meta- effectively and manage their expectations for future IVF analysis of relevant trials [215] concluded that paternal and cycles. Unfortunately, even the more promising ones third-party ALT provide no significant beneficial effect over are successful only to a certain subgroup of patients with placebo in preventing miscarriages. The results of this meta- specific conditions. Our understanding of the mechanism analysis are extremely controversial because it included a behind embryo implantation remains poorly understood, large negative trial using immunization with paternal lym- which hampers our ability to find a solution. To date, strong phocytes stored overnight [225]. This factor impairs the pro- evidence supports the use of blastocyst transfer, assisted tective anti-abortive effect of the procedure and causes loss hatching, salpingectomy for tubal disease, and hysteroscopy in surface CD200, at least in mice [226]. The results of the in the management of couples with previous implantation Ober study, despite the debate over its design, decisively failures and clearly dismisses the use of aspirin and heparin influenced the issues of the most current Cochrane review with IVIg. Other treatment options such as allogenic cited above [215], as well as the US FDA regulation, [227] lymphocyte therapy, ZIFT/EIFT, co-cultures, sildenafil, which states that administration of such cells as allogenic use of donor oocytes, transfer of six embryos, natural IVF lymphocytes or cellular products in humans should only be and PGS are controversial, and their efficacy remains to be performed as part of a clinical research project and then only elucidated. if an Investigational New Drug application is in effect. Moreover, concerns over possible adverse effects of LIT have been raised. These include transfusion-related prob- FIVE-YEAR VIEW lems, autoimmune disorders, graft-versus-host reaction, and Endometrial receptivity is a vital component for embryo transmission of infection such as hepatitis B virus or HIV implantation. A better understanding of the critical success [176], or even cancer and gestational pathology [228]. Ad- factors required for successful implantation is recognized. verse neonatal outcomes are rare, but a case of neonatal allo- What is the optimal condition of the embryo? What is the immune thrombocytopenia and intracranial hemorrhage in an optimal condition of the endometrium? Recent genetic re- infant whose mother received immunizations of paternal search focused on cultured endometrial cells from RIF mononuclear cells has been reported [229]. However, a pro- women has demonstrated a difference in transcriptional ac- spective study by Kling et al., with follow-up after 2-3 years, tivity during the implantation window. A possible disruption showed that the acute side effects of intradermal ALT were in genetic expression of specific genes that regulate the cell comparable to those reported after intradermal vaccination cycle has been implicated. Further research is needed to fully for infectious diseases and that specific risks for anaphylaxis, understand the genes involved in the defunct pathway of autoimmune, or graft- versus- host disease were not signifi- endometrial cells during the implantation window, thus re- cant [228]. sulting in implantation failure. In conclusion, proposing ALT to RIF patients in the absence of standard and broadly applicable diagnostic tests of KEY POINTS immune-mediated pregnancy losses, of reliable methods for Implantation failure and embryo quality are major limit- judging the immunization effects, and of unified protocols of • immunization should await further randomized, controlled ing steps for IVF treatment. trials based on adequate patient selection and more complete • Etiological sources of implantation failure include em- knowledge of the underlying pathophysiology of the as- bryo quality; endometrial receptivity; immunological fac- sumed alloimmune causes of recurrent miscarriage. tors; uterine, tubal, and peritoneal factors; and culture media. The treatment options discussed in the article are CONCLUSIONS intended to address the particular causes and improve Randomized, controlled trials have shown that blastocyst implantation rates. transfer, assisted hatching, salpingectomy for tubal disease, • Differences in the local environment of the endometrium and hysteroscopy in IVF procedures are beneficial for im- of RIF patients compared with other infertile patients proving treatment outcome in patients with repeated implan- have been found. These differences (i.e. gene expression) tation failure. Studies also demonstrate that treatment with possibly affect cross-talks between the embryo and the aspirin and heparin with IVIg does not have a clear impact endometrium and thus implantation. on treatment outcome. Allogenic lymphocyte therapy, ZIFT/EIFT, co-cultures, sildenafil, use of donor oocytes, • Blastocyst transfer, assisted hatching, salpingectomy for transfer of six embryos, natural IVF, and PGS await further tubal disease, and hysteroscopy are highly recommended clinical assessment. The management of RIF should be indi- treatment options based on good, consistent scientific vidualized because the pathophysiology is so variable and evidence with randomized, controlled studies. often complex. Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 213

REFERENCES [22] Fabbri R, Porcu E, Marsella T, et al. Human embryo development and pregnancies in an homologous granulosa cell coculture system. [1] Weissman A, Biran G, Nahum H, Glezerman M, Levran D. J Assist Reprod Genet 2000; 17(1): 1-12. Blastocyst culture and transfer: lessons from an unselected, [23] Dirnfeld M, Goldman S, Gonen Y, Koifman M, Calderon I, difficult IVF population. Reprod Biomed Online 2008; 17(2): 220- Abramovici H. A simplified coculture system with luteinized 8. granulosa cells improves embryo quality and implantation rates: a [2] Martin-Johnston MK, Uhler ML, Grotjan HE, Lifchez AS, Nani controlled study. Fertil Steril 1997; 67(1): 120-2. JM, Beltsos AN. Lower chance of pregnancy with repeated cycles [24] Escribá MJ, Escobedo-Lucea C, Mercader A, de los Santos MJ, with in vitro fertilization. J Reprod Med 2009; 54(2): 67-72. Pellicer A, Remohí J.Ultrastructure of preimplantation genetic [3] Ola, Bolarinde; Li, Tin-Chiu Implantation failure following in-vitro diagnosis-derived human blastocysts grown in a coculture system fertilization. Curr Opin Obstet Gynecol 2006; 18(4): 440-5. after vitrification. Fertil Steril 2006; 86(3): 664-71. [4] Achache H, Tsafrir A, Prus D, Reich R, Revel A. Defective endo- [25] Bongso A, Ng SC, Ratnam S. Co-cultures: their relevance to as- metrial prostaglandin synthesis identified in patients with repeated sisted reproduction. Hum Reprod 1990; 5(8): 893-900. implantation failure undergoing in vitro fertilization. Fertil Steril [26] Bongso A, Soon-Chye N, Sathananthan H, Lian NP, Rauff M, 2009. [Epub ahead of print] Ratnam S. alImproved quality of human embryos when co-cultured [5] Koler M, Achache H, Tsafrir A, Smith Y, Revel A, Reich R. Dis- with human ampullary cells. Hum Reprod 1989; 4(6): 706-13. rupted gene pattern in patients with repeated in vitro fertilization [27] Morgan K, Wiemer K, Steuerwald N, Hoffman D, Maxson W, (IVF) failure. Hum Reprod 2009; 24(10): 2541-8. Godke R.Use of videocinematography to assess morphological [6] Bersinger NA, Wunder DM, Birkhäuser MH, Mueller MD. qualities of conventionally cultured and cocultured embryos. Hum Gene expression in cultured endometrium from women with differ- Reprod 1995; 10(9): 2371-6. ent outcomes following IVF. Mol Hum Reprod 2008; 14(8): 475- [28] Magli MC, Gianaroli L, Ferraretti AP, Fortini D, Fiorentino A, 84. D'Errico A. Human embryo co-culture: results of a randomized [7] Cline AM, Kutteh WH.Is there a role of autoimmunity in implanta- prospective study. Int J Fertil Menopausal Stud 1995; 40(5): 254-9. tion failure after in-vitro fertilization? Curr Opin Obstet Gynecol [29] Tucker MJ, Morton PC, Wright G, et al. Enhancement of outcome 2009; 21(3): 291-5. from intracytoplasmic sperm injection: does co-culture or assisted [8] Milki AA, Hinckley MD, Fisch JD, Dasig D, Behr B. Comparison hatching improve implantation rates? Hum Reprod 1996; 11(11): of blastocyst transfer with day 3 embryo transfer in similar patient 2434-7. populations. Fertil Steril 2000; 73(1): 126-9. [30] Wiemer KE, Garrisi J, Steuerwald N, et al. Beneficial aspects of [9] Wilson M, Hartke K, Kiehl M, Rodgers J, Brabec C, Lyles R. Inte- co-culture with assisted hatching when applied to multiple-failure gration of blastocyst transfer for all patients. Fertil Steril 2002; in-vitro fertilization patients. Hum Reprod 1996; 11(11): 2429-33. 77(4): 693-6. [31] Wiemer KE, Cohen J, Amborski GF, et al. In-vitro development [10] Levron J, Shulman A, Bider D, Seidman D, Levin T, Dor J. A and implantation of human embryos following culture on fetal bo- prospective randomized study comparing day 3 with blastocyst- vine uterine fibroblast cells. Hum Reprod 1989; 4(5): 595-600. stage embryo transfer. Fertil Steril 2002; 77(6): 1300-1. [32] Wiemer KE, Cohen J, Wiker SR, Malter HE, Wright G, Godke RA. [11] Papanikolaou EG, D'haeseleer E, Verheyen G, et al. Live birth rate Coculture of human zygotes on fetal bovine uterine fibroblasts: is significantly higher after blastocyst transfer than after cleavage- embryonic morphology and implantation. Fertil Steril 1989; 52(3): stage embryo transfer when at least four embryos are available on 503-8. day 3 of embryo culture. A randomized prospective study. Hum [33] Van Blerkom J. Development of human embryos to the hatched Reprod 2005; 20(11): 3198-203. blastocyst stage in the presence or absence of a monolayer of Vero [12] Shapiro BS, Richter KS, Harris DC, Daneshmand ST. Dramatic cells. Hum Reprod 1993; 8(9): 1525-39. declines in implantation and pregnancy rates in patients who un- [34] Sakkas D, Jaquenoud N, Leppens G, Campana A. Comparison of dergo repeated cycles of in vitro fertilization with blastocyst trans- results after in vitro fertilized human embryos are cultured in rou- fer after one or more failed attempts. Fertil Steril 2001; 76(3): 538- tine medium and in coculture on Vero cells: a randomized study. 42. Fertil Steril 1994; 61(3): 521-5. [13] Cruz JR, Dubey AK, Patel J, Peak D, Hartog B, Gindoff PR.Is [35] Turner K, Lenton EA. The influence of Vero cell culture on human blastocyst transfer useful as an alternative treatment for patients embryo development and chorionic gonadotrophin production in with multiple in vitro fertilization failures? Fertil Steril 1999; vitro. Hum Reprod 1996; 11(9): 1966-74. 72(2): 218-20. [36] Ben-Chetrit A, Jurisicova A, Casper RF. Coculture with ovarian [14] Levitas E, Lunenfeld E, Har-Vardi I, et al. Blastocyst-stage embryo cancer cell enhances human blastocyst formation in vitro. Fertil transfer in patients who failed to conceive in three or more day 2-3 Steril 1996; 65(3): 664-6. embryo transfer cycles: a prospective, randomized study. Fertil [37] Hu Y, Maxson W, Hoffman D, et al. Co-culture with assisted Steril 2004; 81(3): 567-71. hatching of human embryos using Buffalo rat liver cells. Hum Re- [15] Barrenetxea G, López de Larruzea A, Ganzabal T, Jiménez R, prod 1998; 13(1): 165-8. Carbonero K, Mandiola M.Blastocyst culture after repeated failure [38] Wetzels AM, Bastiaans BA, Hendriks JC, et al. The effects of co- of cleavage-stage embryo transfers: a comparison of day 5 and day culture with human fibroblasts on human embryo development in 6 transfers. Fertil Steril 2005; 83(1): 49-53. vitro and implantation. Hum Reprod 1998; 13(5): 1325-30. [16] Gabrielsen A, Agerholm I, Toft B, et al. Assisted hatching im- [39] Kattal N, Cohen J, Barmat LI. Role of coculture in human in vitro proves implantation rates on cryopreserved-thawed embryos. A fertilization: a meta-analysis. Fertil Steril 2008; 90(4): 1069-76. randomized prospective study. Hum Reprod 2004; 19(10): 2258- [40] Weichselbaum A, Paltieli Y, Philosoph R, et al. Improved devel- 62. opment of very-poor-quality human preembryos by coculture with [17] Germond M, Primi MP, Senn A. Hatching: how to select the clini- human fallopian ampullary cells. J Assist Reprod Genet 2002; cal indications. Ann N Y Acad Sci 2004; 1034: 145-51. 19(1): 7-13. [18] Sheen TC, Chen SR, Au HK, Chien YY, Wu KY, Tzeng CR. [41] Eyheremendy V, Raffo FG, Papayannis M, Barnes J, Granados C, Herniated blastomere following chemically assisted hatching may Blaquier J. Beneficial effect of autologous endometrial cell cocul- result in monozygotic twins. Fertil Steril 2001; 75(2): 442-4. ture in patients with repeated implantation failure. Fertil Steril [19] Das S, Blake D, Farquhar C, Seif MM. Assisted hatching on 2010; 93(3): 769-73 assisted conception (IVF and ICSI). Cochrane Database Syst Rev [42] Desai N, Abdelhafez F, Bedaiwy MA, Goldfarb J. Live births in 2009; (2): CD001894. poor prognosis IVF patients using a novel non-contact human en- [20] Simón C, Mercader A, Garcia-Velasco J, et al. Coculture of human dometrial co-culture system. Reprod Biomed Online 2008; 16(6): embryos with autologous human endometrial epithelial cells in 869-74. patients with implantation failure. J Clin Endocrinol Metab 1999; [43] Richter KS. The importance of growth factors for preimplantation 84(8): 2638-46. embryo development and in-vitro culture. Curr Opin Obstet Gyne- [21] Carrell DT, Peterson CM, Jones KP, et al. A simplified coculture col 2008; 20(3): 292-304. system using homologous, attached cumulus tissue results in im- [44] d'Estaing SG, Lornage J, Hadj S, Boulieu D, Salle B, Guérin JF. proved human embryo morphology and pregnancy rates during in Comparison of two blastocyst culture systems: coculture on Vero vitro fertilization. J Assist Reprod Genet 1999; 16(7): 344-9. cells and sequential media. Fertil Steril 2001; 76(5): 1032-5. 214 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al.

[45] Griffin DK. The incidence, origin, and etiology of aneuploidy. Int [67] Oliveira FG, Abdelmassih VG, Diamond MP, et al. Uterine cavity Rev Cytol 1996; 167: 263-96. findings and hysteroscopic interventions in patients undergoing in [46] Maroulis GB, Koutlaki N. Preimplantation genetic diagnosis. Ann vitro fertilization-embryo transfer who repeatedly cannot conceive. N Y Acad Sci 2006; 1092: 279-84. Fertil Steril 2003; 80(6): 1371-5. [47] Farfalli VI, Magli MC, Ferraretti AP, Gianaroli L.Role of ane- [68] Makrakis E, Hassiakos D, Stathis D, Vaxevanoglou T, Orfanoudaki uploidy on embryo implantation. Gynecol Obstet Invest 2007; E, Pantos K. Hysteroscopy in women with implantation failures af- 64(3): 161-5. ter in vitro fertilization: findings and effect on subsequent preg- [48] Fragouli E, Katz-Jaffe M, Alfarawati S, et al. Comprehensive nancy rates. J Minim Invasive Gynecol 2009; 16(2): 181-7. chromosome screening of polar bodies and blastocysts from [69] Madani T, Ghaffari F, Kiani K, Hosseini F. Hysteroscopic polypec- couples experiencing repeated implantation failure. Fertil Steril tomy without cycle cancellation in IVF cycles. Reprod Biomed On- 2009. [Epub ahead of print] line 2009; 18(3):412-5. [49] Raziel A, Friedler S, Schachter M, Kasterstein E, Strassburger D, [70] Küçük T, Safali M. "Chromohysteroscopy" for evaluation of en- Ron-El R. Increased frequency of female partner chromosomal dometrium in recurrent in vitro fertilization failure. J Assist Reprod abnormalities in patients with high-order implantation failure after Genet 2008; 25(2-3): 79-82. in vitro fertilization. Fertil Steril 2002; 78(3): 515-9. [71] Boolell M, Allen MJ, Ballard SA, et al. Sildenafil: an orally active [50] Gianaroli L, Magli MC, Munné S, Fiorentino A, Montanaro N, type 5 cyclic GMP-specific phosphodiesterase inhibitor for the Ferraretti AP. Will preimplantation genetic diagnosis assist patients treatment of penile erectile dysfunction. Int J Impot Res 1996; 8(2): with a poor prognosis to achieve pregnancy? Hum Reprod 1997; 47-52. 12(8): 1762-7. [72] Goodman C, Jeyendran RS, Coulam CB. Coulam, P53 tumor sup- [51] Pehlivan T, Rubio C, Rodrigo L, et al. Impact of preimplantation pressor factor, plasminogen activator inhibitor, and vascular endo- genetic diagnosis on IVF outcome in implantation failure patients. thelial growth factor gene polymorphisms and recurrent implanta- Reprod Biomed Online 2003; 6(2): 232-7. tion failure. Fertil Steril 2009; 92(2): 494-8. [52] ESHRE PGD Consortium Steering Committee. ESHRE Preimplan- [73] Sher G, Fisch JD. Vaginal sildenafil (Viagra): a preliminary report tation Genetic Diagnosis Consortium: data collection III (May of a novel method to improve uterine artery blood flow and endo- 2001). Hum Reprod 2002; 17(1): 233-46. metrial development in patients undergoing IVF. Hum Reprod [53] Gianaroli L, Magli MC, Ferraretti AP, Munné S. Preimplantation 2000; 15(4): 806-9. diagnosis for aneuploidies in patients undergoing in vitro fertiliza- [74] Sher G, Fisch JD. Effect of vaginal sildenafil on the outcome of in tion with a poor prognosis: identification of the categories for vitro fertilization (IVF) after multiple IVF failures attributed to which it should be proposed. Fertil Steril 1999; 72(5): 837-44. poor endometrial development. Fertil Steril 2002; 78(5): 1073-6. [54] Goossens V, Harton G, Moutou C, Traeger-Synodinos J, Van Rij [75] Check JH, Graziano V. Multiple confounders--measured with M, Harper JC. ESHRE PGD Consortium data collection IX: cycles error? Fertil Steril 2003; 79(5): 1256-7; author reply 1257-8. from January to December 2006 with pregnancy follow-up to Oc- [76] Zinger M, Liu JH, Thomas MA. Successful use of vaginal silde- tober 2007. Hum Reprod 2009; 24(8): 1786-810. nafil citrate in two infertility patients with Asherman's syndrome. J [55] Platteau P, Staessen C, Michiels A, Van Steirteghem A, Liebaers I, Womens Health (Larchmt) 2006; 15(4): 442-4. Devroey P. Which patients with recurrent implantation failure after [77] Paulus WE, Strehler E, Zhang M, Jelinkova L, El-Danasouri I, IVF benefit from PGD for aneuploidy screening? Reprod Biomed Sterzik K. Benefit of vaginal sildenafil citrate in assisted reproduc- Online 2006; 12(3): 334-9. tion therapy. Fertil Steril 2002; 77(4): 846-7. [56] Mantzouratou A, Mania A, Fragouli E, et al. Variable aneuploidy [78] Quenby S, Bates M, Doig T, et al. Pre-implantation endometrial mechanisms in embryos from couples with poor reproductive histo- leukocytes in women with recurrent miscarriage. Hum Reprod ries undergoing preimplantation genetic screening. Hum Reprod 1999; 14(9): 2386-91. 2007; 22(7): p. 1844-53. [79] Gilman-Sachs A, DuChateau BK, Aslakson CJ, et al. Natural killer [57] Voullaire L, Wilton L, McBain J, Callaghan T, Williamson R. (NK) cell subsets and NK cell cytotoxicity in women with histories Chromosome abnormalities identified by comparative genomic hy- of recurrent spontaneous abortions. Am J Reprod Immunol 1999; bridization in embryos from women with repeated implantation 41(1): 99-105. failure. Mol Hum Reprod 2002; 8(11): 1035-41. [80] Aoki K, Kajiura S, Matsumoto Y, et al. Preconceptional natural- [58] Braude P, Bolton V, Moore S. Human gene expression first occurs killer-cell activity as a predictor of miscarriage. Lancet 1995; between the four- and eight-cell stages of preimplantation devel- 345(8961): 1340-2. opment. Nature 1988; 332(6163): 459-61. [81] Strandell A, Lindhard A, Waldenström U, Thorburn J. Nitric oxide [59] Wells D, Bermudez MG, Steuerwald N, et al. Expression of genes inhibits development of embryos and implantation in mice. Mol regulating chromosome segregation, the cell cycle and apoptosis Hum Reprod 1998; 4(5): 503-7. during human preimplantation development. Hum Reprod 2005; [82] Battaglia C, Ciotti P, Notarangelo L, Fratto R, Facchinetti F, de 20(5): 1339-48. Aloysio D. Embryonic production of nitric oxide and its role in im- [60] Voullaire L, Collins V, Callaghan T, McBain J, Williamson R, plantation: a pilot study. J Assist Reprod Genet 2003; 20(11): 449- Wilton L. High incidence of complex chromosome abnormality in 54. cleavage embryos from patients with repeated implantation failure. [83] Strandell A, Waldenström U, Nilsson L, Hamberger L. Hydrosal- Fertil Steril 2007; 87(5): 1053-8. pinx reduces in-vitro fertilization/embryo transfer pregnancy rates. [61] Margalioth EJ, Ben-Chetrit A, Gal M, Eldar-Geva T. Investigation Hum Reprod 1994; 9(5): 861-3. and treatment of repeated implantation failure following IVF-ET. [84] Bontis JN, Dinas KD. Management of hydrosalpinx: reconstructive Hum Reprod 2006; 21(12): 3036-43. surgery or IVF? Ann N Y Acad Sci 2000; 900: 260-71. [62] Munné S, Magli C, Bahçe M, et al. Preimplantation diagnosis of [85] Strandell A, Lindhard A, Waldenström U, Thorburn J. Hydrosal- the aneuploidies most commonly found in spontaneous abortions pinx and IVF outcome: cumulative results after salpingectomy in a and live births: XY, 13, 14, 15, 16, 18, 21, 22. Prenat Diagn randomized controlled trial. Hum Reprod 2001; 16(11): 2403-10. 1998;18(13): 1459-66. [86] Mansour R, Aboulghar M, Serour GI. Controversies in the surgical [63] Kahraman S, Bahçe M, Samli H, et al. Healthy births and ongoing management of hydrosalpinx. Curr Opin Obstet Gynecol 2000; pregnancies obtained by preimplantation genetic diagnosis in pa- 12(4): 297-301. tients with advanced maternal age and recurrent implantation fail- [87] Granot I, Dekel N, Segal I, Fieldust S, Shoham Z, Barash A. Is ure. Hum Reprod 2000; 15(9): 2003-7. hydrosalpinx fluid cytotoxic? Hum Reprod 1998; 13(6): 1620-4. [64] Munné S, Sandalinas M, Escudero T, et al. Improved implantation [88] Sharara FI. The role of hydrosalpinx in IVF: simply mechanical? after preimplantation genetic diagnosis of aneuploidy. Reprod Bio- Hum Reprod 1999; 14(3): 577-8. med Online 2003; 7(1): 91-7. [89] Seli E, Kayisli UA, Cakmak H, et al. Removal of hydrosalpinges [65] Arefi S, Soltanghoraei H, Sadeghi MR, Tabaei AS, Zeraati H, increases endometrial leukaemia inhibitory factor (LIF) expression Zarnani H. Findings on hysteroscopy in patients with in vitro fer- at the time of the implantation window. Hum Reprod 2005; 20(11): tilization by intra cytoplasmic single sperm injection and embryo 3012-7. transfer failures. Saudi Med J 2008; 29(8): 1201-2. [90] Bildirici I, Bukulmez O, Ensari A, Yarali H, Gurgan T. A prospec- [66] Lorusso F, Ceci O, Bettocchi S, et al. Office hysteroscopy in an in tive evaluation of the effect of salpingectomy on endometrial re- vitro fertilization program. Gynecol Endocrinol 2008; 24(8): 465-9. Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 215

ceptivity in cases of women with communicating hydrosalpinges. options that have not been proven to benefit the couple. Reprod Hum Reprod 2001; 16(11): 2422-6. Biomed Online 2005; 11(3): 382-91. [91] Meyer WR, Lessey BA, Castelbaum AJ.Hydrosalpinx and altered [117] Jansen RP. Endocrine response in the fallopian tube. Endocr Rev uterine receptivity. Fertil Steril 1997;68(5): 944-5. 1984; 5(4): 525-51. [92] Grifo JA, Jeremias J, Ledger WJ, Witkin SS.Interferon-gamma in [118] Devroey P, Staessen C, Camus M, De Grauwe E, Wisanto A, Van the diagnosis and pathogenesis of pelvic inflammatory disease. Am Steirteghem AC.Zygote intrafallopian transfer as a successful J Obstet Gynecol 1989; 160(1): 26-31. treatment for unexplained infertility. Fertil Steril 1989; 52(2): 246- [93] Ben-Rafael Z, Orvieto R. Cytokines--involvement in reproduction. 9. Fertil Steril 1992; 58(6): 1093-9. [119] Hammitt DG, Syrop CH, Hahn SJ, Walker DL, Butkowski CR, [94] Jones RB, Mammel JB, Shepard MK, Fisher RR. Recovery of Donovan JF.Comparison of concurrent pregnancy rates for in-vitro Chlamydia trachomatis from the endometrium of women at risk for fertilization--embryo transfer, pronuclear stage embryo transfer and chlamydial infection. Am J Obstet Gynecol 1986; 155(1): 35-9. gamete intra-fallopian transfer. Hum Reprod 1990; 5(8): 947-54. [95] Lass A. What effect does hydrosalpinx have on assisted reproduc- [120] Asch RH.Uterine versus tubal embryo transfer in the human. Com- tion? What is the preferred treatment for hydrosalpines? The parative analysis of implantation, pregnancy, and live-birth rates. ovary's perspective. Hum Reprod 1999; 14(7): 1674-7. Ann N Y Acad Sci 1991; 626: 461-6. [96] Dechaud H, Anahory T, Aligier N, Arnal F, Humeau H, Hedon B. [121] Farhi J, Weissman A, Nahum H, Levran D.Zygote intrafallopian Salpingectomy for repeated embryo nonimplantation after in vitro transfer in patients with tubal factor infertility after repeated failure fertilization in patients with severe tubal factor infertility. J Assist of implantation with in vitro fertilization-embryo transfer. Fertil Reprod Genet 2000; 17(4): 200-6. Steril 2000; 74(2): 390-3. [97] Johnson NP, Mak W, Sowter MC. Surgical treatment for tubal [122] Habana AE, Palter SF. Is tubal embryo transfer of any value? A disease in women due to undergo in vitro fertilisation. Cochrane meta-analysis and comparison with the Society for Assisted Repro- Database Syst Rev 2004; (3): CD002125. ductive Technology database. Fertil Steril 2001; 76(2): 286-93. [98] Johnson NP, Mak W, Sowter MC. Laparoscopic salpingectomy for [123] Aslan D, Elizur SE, Levron J, et al. Comparison of zygote intrafal- women with hydrosalpinges enhances the success of IVF: a Coch- lopian tube transfer and transcervical uterine embryo transfer in pa- rane review. Hum Reprod 2002; 17(3): 543-8. tients with repeated implantation failure. Eur J Obstet Gynecol Re- [99] Practice Committee of American Society for Reproductive Medi- prod Biol 2005; 122(2): 191-4. cine; Practice Committee of Society for Assisted Reproductive [124] Weissman A, Eldar I, Ravhon A, et al. Timing intra-Fallopian Technology. Fertil Steril 2008; 90(5 Suppl): S30-44. transfer procedures. Reprod Biomed Online 2007; 15(4): 445-50. [100] Klein J, Sauer MV.Oocyte donation. Best Pract Res Clin Obstet [125] Weissman A, F.J., Levran D. Zygote intrafallopian transfer. Gynaecol 2002; 16(3): 277-91. Reproductive Techniques: Laboratory and Clinical Perspectives, [101] Sauer MV, Kavic SM.Oocyte and embryo donation 2006: review- 2nd ed. W.A. Gardner DK, Howies CM, Shoham Z, Eds. London ing two decades of innovation and controversy. Reprod Biomed Taylor and Francis 2004; pp. 735-48. Online 2006; 12(2): 153-62. [126] Ubaldi F, Rienzi L, Ferrero S, et al. Natural in vitro fertilization [102] Lee RM, Silver RM. Recurrent pregnancy loss: summary and clini- cycles. Ann N Y Acad Sci 2004; 1034: 245-51. cal recommendations. Semin Reprod Med 2000; 18(4): 433-40. [127] Morgia F, Sbracia M, Schimberni M, et al. A controlled trial of [103] Lee RM, Silver RM. The efficacy and safety of human oocyte natural cycle versus microdose gonadotropin-releasing hormone vitrification. Semin Reprod Med 2009; 27(6): 450-5. analog flare cycles in poor responders undergoing in vitro fertiliza- [104] ASRM Practice Committee response to Rybak and Lieman: elec- tion. Fertil Steril 2004; 81(6): 1542-7. tive self-donation of oocytes. Fertil Steril 2009; 92(5): 1513-4. [128] Bassil S, Godin PA, Donnez J. Outcome of in-vitro fertilization [105] Rienzi L, Romano S, Albricci L, et al. Embryo development of through natural cycles in poor responders. Hum Reprod 1999; fresh 'versus' vitrified metaphase II oocytes after ICSI: a prospec- 14(5): 1262-5. tive randomized sibling-oocyte study. Hum Reprod 2010; 25(1): [129] Schimberni M, Morgia F, Colabianchi J, et al. Natural-cycle in 66-73. vitro fertilization in poor responder patients: a survey of 500 con- [106] Azem F, Yaron Y, Amit A, et al. Transfer of six or more embryos secutive cycles. Fertil Steril 2009; 92(4): 1297-301. improves success rates in patients with repeated in vitro fertiliza- [130] Pelinck MJ, Hoek A, Simons AH, Heineman MJ. Efficacy of natu- tion failures. Fertil Steril 1995; 63(5): 1043-6. ral cycle IVF: a review of the literature. Hum Reprod Update 2002; [107] Hu Y, Maxson WS, Hoffman DI, et al. Maximizing pregnancy 8(2): 129-39. rates and limiting higher-order multiple conceptions by determin- [131] Paria BC, Lim H, Das SK, Reese J, Dey SK. Molecular signaling in ing the optimal number of embryos to transfer based on quality. uterine receptivity for implantation. Semin Cell Dev Biol 2000; Fertil Steril 1998; 69(4): 650-7. 11(2): 67-76. [108] Stern JE, Cedars MI, Jain T, et al. Assisted reproductive technol- [132] Paria BC, Lim H, Das SK, Reese J, Dey SK. Time of implantation ogy practice patterns and the impact of embryo transfer guidelines of the conceptus and loss of pregnancy. N Engl J Med 1999; in the United States. Fertil Steril 2007; 88(2): 275-82. 340(23): 1796-9. [109] Campbell B, Erickson L. The number of embryos to transfer: cur- [133] Martín J, Domínguez F, Avila S, et al. Human endometrial recep- rent practices in Minnesota. Minn Med 2009; 92(4): 40-1. tivity: gene regulation. J Reprod Immunol 2002; 55(1-2): 131-9. [110] Guidelines on number of embryos transferred. Fertil Steril 2009; [134] Davis GE, Senger DR. Endothelial extracellular matrix: biosynthe- 92(5): 1518-9. sis, remodeling, and functions during vascular morphogenesis and [111] Guidelines for the number of embryos to transfer following in vitro neovessel stabilization. Circ Res 2005; 97(11): 1093-107. fertilization No. 182, September 2006. Int J Gynaecol Obstet 2008; [135] Bischof P, Campana A. Molecular mediators of implantation. 102(2): 203-16. Baillieres Best Pract Res Clin Obstet Gynaecol 2000; 14(5): 801- [112] Pandian Z, Bhattacharya S, Ozturk O, Serour G, Templeton A. 14. Number of embryos for transfer following in-vitro fertilisation or [136] Basir GS, O WS, Ng EH, Ho PC. Morphometric analysis of peri- intra-cytoplasmic sperm injection. Cochrane Database Syst Rev implantation endometrium in patients having excessively high 2009(2): CD003416. oestradiol concentrations after ovarian stimulation. Hum Reprod [113] Schoolcraft WB, Surrey ES, Gardner DK. Embryo transfer: tech- 2001; 16(3): 435-40. niques and variables affecting success. Fertil Steril 2001; 76(5): [137] Devroey P, Bourgain C, Macklon NS, Fauser BC. Reproductive 863-70. biology and IVF: ovarian stimulation and endometrial receptivity. [114] Levran D, Farhi J, Nahum H, Royburt M, Glezerman M, Weissman Trends Endocrinol Metab 2004; 15(2): 84-90. A. Prospective evaluation of blastocyst stage transfer vs. zygote in- [138] Tavaniotou A, Albano C, Smitz J, Devroey P. Impact of ovarian trafallopian tube transfer in patients with repeated implantation stimulation on corpus luteum function and embryonic implantation. failure. Fertil Steril 2002; 77(5): 971-7. J Reprod Immunol 2002; 55(1-2): 123-30. [115] Levran D, Mashiach S, Dor J, Levron J, Farhi J. Zygote intrafallo- [139] Basille C, Fay S, Hesters L, Frydman N, Frydman R. In vitro fer- pian transfer may improve pregnancy rate in patients with repeated tilization (IVF): why doing it in unstimulated cycles?. Gynecol Ob- failure of implantation. Fertil Steril 1998; 69(1): 26-30. stet Fertil 2007; 35(9): 877-80. [116] Urman B, Yakin K, Balaban B. Recurrent implantation failure in [140] Lédée-Bataille N, Dubanchet S, Kadoch J, Castelo-Branco A, assisted reproduction: how to counsel and manage. B. Treatment Frydman R, Chaouat G. Controlled natural in vitro fertilization 216 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al.

may be an alternative for patients with repeated unexplained im- [163] Matsubayashi H, Arai T, Izumi S, Sugi T, McIntyre JA, Makino T. plantation failure and a high uterine natural killer cell count. Fertil Anti-annexin V antibodies in patients with early pregnancy loss or Steril 2004; 82(1): 234-6. implantation failures. Fertil Steril 2001; 76(4): 694-9. [141] Croy BA, Ashkar AA, Minhas K, Greenwood JD.Can murine uter- [164] Eldar-Geva T, Wood C, Lolatgis N, et al. Cumulative pregnancy ine natural killer cells give insights into the pathogenesis of and live birth rates in women with antiphospholipid antibodies un- preeclampsia? J Soc Gynecol Investig 2000; 7(1): 12-20. dergoing assisted reproduction. Hum Reprod 1999; 14(6): 1461-6. [142] Horcajadas JA, Riesewijk A, Polman J, et al. Effect of controlled [165] Practice Committee of American Society for Reproductive Medi- ovarian hyperstimulation in IVF on endometrial gene expression cine. Anti-phospholipid antibodies do not affect IVF success. Fertil profiles. Mol Hum Reprod 2005; 11(3): 195-205. Steril 2008; 90(5 Suppl): S172-3. [143] Haouzi D, Mahmoud K, Fourar M, et al. Identification of new [166] Ermel LD, Marshburn PB, Kutteh WH. Interaction of heparin with biomarkers of human endometrial receptivity in the natural cycle. antiphospholipid antibodies (APA) from the sera of women with Hum Reprod 2009; 24(1): 198-205. recurrent pregnancy loss (RPL). Am J Reprod Immunol 1995; [144] Di Simone N, Meroni PL, de Papa N, et al. Antiphospholipid anti- 33(1): 14-20. bodies affect trophoblast gonadotropin secretion and invasiveness [167] Kutteh WH. Antiphospholipid antibody-associated recurrent preg- by binding directly and through adhered beta2-glycoprotein I. Ar- nancy loss: treatment with heparin and low-dose aspirin is superior thritis Rheum 2000; 43(1): 140-50. to low-dose aspirin alone. Am J Obstet Gynecol 1996; 174(5): [145] Kutteh WH.Antiphospholipid antibodies and reproduction. J Re- 1584-9. prod Immunol 1997; 35(2): 151-71. [168] Kutteh WH, Yetman DL, Chantilis SJ, Crain J. Effect of antiphos- [146] Girardi G, Yarilin D, Thurman JM, Holers VM, Salmon pholipid antibodies in women undergoing in-vitro fertilization: role JE.Complement activation induces dysregulation of angiogenic fac- of heparin and aspirin. Hum Reprod 1997; 12(6): 1171-5. tors and causes fetal rejection and growth restriction. J Exp Med [169] Stern C, Chamley L, Norris H, Hale L, Baker HW. A randomized, 2006; 203(9): 2165-75. double-blind, placebo-controlled trial of heparin and aspirin for [147] el-Roeiy A, Gleicher N, Friberg J, Confino E, Dudkiewicz A. Cor- women with in vitro fertilization implantation failure and antiphos- relation between peripheral blood and follicular fluid autoantibod- pholipid or antinuclear antibodies. Fertil Steril 2003; 80(2): 376-83. ies and impact on in vitro fertilization. Obstet Gynecol 1987; 70(2): [170] Empson M, Lassere M, Craig J, Scott J. Prevention of recurrent 163-70. miscarriage for women with antiphospholipid antibody or lupus an- [148] Buckingham KL, Stone PR, Smith JF, Chamley LW. Antiphos- ticoagulant. Cochrane Database Syst Rev 2005; (2): CD002859. pholipid antibodies in serum and follicular fluid--is there a correla- [171] Vaquero E, Lazzarin N, Valensise Het, et al. Pregnancy outcome in tion with IVF implantation failure? Hum Reprod 2006; 21(3): 728- recurrent spontaneous abortion associated with antiphospholipid 34. antibodies: a comparative study of intravenous immunoglobulin [149] Matsubayashi H, Sugi T, Arai T, et al. IgG-antiphospholipid anti- versus prednisone plus low-dose aspirin. Am J Reprod Immunol bodies in follicular fluid of IVF-ET patients are related to low fer- 2001; 45(3): 174-9. tilization rate of their oocytes. Am J Reprod Immunol 2006; 55(5): [172] Triolo G, Ferrante A, Ciccia F, et al. Randomized study of subcu- 341-8. taneous low molecular weight heparin plus aspirin versus intrave- [150] Choudhury SR, Knapp LA.Human reproductive failure I: immu- nous immunoglobulin in the treatment of recurrent fetal loss asso- nological factors. Hum Reprod Update 2001; 7(2): 113-34. ciated with antiphospholipid antibodies. Arthritis Rheum 2003; [151] Galli M, Comfurius P, Maassen C, et al. Anticardiolipin antibodies 48(3): 728-31. (ACA) directed not to cardiolipin but to a plasma protein cofactor. [173] Dendrinos S, Sakkas E, Makrakis E. Low-molecular-weight hepa- Lancet 1990; 335(8705): 1544-7. rin versus intravenous immunoglobulin for recurrent abortion asso- [152] Meroni PL, Gerosa M, Raschi E, Scurati S, Grossi C, Borghi ciated with antiphospholipid antibody syndrome. Int J Gynaecol MO.Updating on the pathogenic mechanisms 5 of the antiphos- Obstet 2009; 104(3): 223-5. pholipid antibodies-associated pregnancy loss. Clin Rev Allergy [174] Clark AL, Branch DW, Silver RM, Harris EN, Pierangeli S, Spin- Immunol 2008; 34(3): 332-7. nato JA.Pregnancy complicated by the antiphospholipid syndrome: [153] Stern C, Chamley L, Hale L, Kloss M, Speirs A, Baker HW. Anti- outcomes with intravenous immunoglobulin therapy. Obstet Gyne- bodies to beta2 glycoprotein I are associated with in vitro fertiliza- col 1999; 93(3): 437-41. tion implantation failure as well as recurrent miscarriage: results of [175] Wong RC, Favaloro EJ. Clinical features, diagnosis, and manage- a prevalence study. Fertil Steril 1998; 70(5): 938-44. ment of the antiphospholipid syndrome. Semin Thromb Hemost [154] Birkenfeld A, Mukaida T, Minichiello L, Jackson M, Kase NG, 2008; 34(3): 295-304. Yemini M. Incidence of autoimmune antibodies in failed embryo [176] Takeshita T. Diagnosis and treatment of recurrent miscarriage transfer cycles. Am J Reprod Immunol 1994; 31(2-3): 65-8. associated with immunologic disorders: Is paternal lymphocyte [155] Gleicher N, Liu HC, Dudkiewicz A, et al. Autoantibody profiles immunization a relic of the past? J Nippon Med Sch 2004; 71(5): and immunoglobulin levels as predictors of in vitro fertilization 308-13. success. Am J Obstet Gynecol 1994; 170(4): 1145-9. [177] Wegmann TG, Lin H, Guilbert L, Mosmann TR. Bidirectional [156] Coulam CB, Kaider BD, Kaider AS, Janowicz P, Roussev cytokine interactions in the maternal-fetal relationship: is success- RG. Antiphospholipid antibodies associated with implantation ful pregnancy a TH2 phenomenon? Immunol Today 1993; 14(7): failure after IVF/ET. J Assist Reprod Genet 1997; 14(10): 603- 353-6. 8. [178] Lédée-Bataille N, Bonnet-Chea K, Hosny G, Dubanchet S, [157] Denis AL, Guido M, Adler RD, Bergh PA, Brenner C, Scott RT Jr. Frydman R, Chaouat G. Role of the endometrial tripod interleukin- Antiphospholipid antibodies and pregnancy rates and outcome in in 18, -15, and -12 in inadequate uterine receptivity in patients with a vitro fertilization patients. Fertil Steril 1997; 67(6): 1084-90. history of repeated in vitro fertilization-embryo transfer failure. [158] Geva E, Yaron Y, Lessing JB, et al. Circulating autoimmune Fertil Steril 2005; 83(3): 598-605. antibodies may be responsible for implantation failure in in vitro [179] Quenby S, Farquharson R. Uterine natural killer cells, implantation fertilization. Fertil Steril 1994; 62(4): 802-6. failure and recurrent miscarriage. Reprod Biomed Online 2006; [159] Kowalik A, Vichnin M, Liu HC, Branch W, Berkeley AS. Midfol- 13(1): 24-8. licular anticardiolipin and antiphosphatidylserine antibody titers do [180] Lachapelle MH, Miron P, Hemmings R, Roy DC. Endometrial T, not correlate with in vitro fertilization outcome. Fertil Steril 1997; B, and NK cells in patients with recurrent spontaneous abortion. 68(2): 298-304. Altered profile and pregnancy outcome. J Immunol 1996; 156(10): [160] Hill JA, Choi BC. Maternal immunological aspects of pregnancy 4027-34. success and failure. J Reprod Fertil Suppl 2000; 55: 91-7. [181] Ntrivalas EI, Kwak-Kim JY, Gilman-Sachs A, et al. Status of pe- [161] Hornstein MD, Davis OK, Massey JB, Paulson RJ, Collins JA. ripheral blood natural killer cells in women with recurrent sponta- Antiphospholipid antibodies and in vitro fertilization success: a neous abortions and infertility of unknown aetiology. Hum Reprod meta-analysis. Fertil Steril 2000; 73(2): 330-3. 2001; 16(5): 855-61. [162] Gleicher N, Vidali A, Karande V. The immunological "Wars of the [182] Kwak JY, Beaman KD, Gilman-Sachs A, Ruiz JE, Schewitz D, Roses": disagreements amongst reproductive immunologists. Hum Beer AE. Up-regulated expression of CD56+, CD56+/CD16+, and Reprod 2002; 17(3): 539-42. CD19+ cells in peripheral blood lymphocytes in pregnant women Evidence-Based Management of Infertile Couples Current Women’s Health Reviews, 2010, Vol. 6, No. 3 217

with recurrent pregnancy losses. Am J Reprod Immunol 1995; spontaneous abortions of unexplained etiology: down-regulation of 34(2): 93-9. NK cell activity and subsets. Am J Reprod Immunol 2001; 46(6): [183] Yamada H, Kato EH, Kobashi G, et al. High NK cell activity in 399-404. early pregnancy correlates with subsequent abortion with normal [203] Clark DA, Chaouat G. Loss of surface CD200 on stored allogeneic chromosomes in women with recurrent abortion. Am J Reprod leukocytes may impair anti-abortive effect in vivo. Am J Reprod Immunol 2001; 46(2): 132-6. Immunol 2005; 53(1): 13-20. [184] Yamamoto T, Takahashi Y, Kase N, Mori H. Role of decidual [204] Gorczynski RM. Thymocyte/splenocyte-derived CD4+CD25+Treg natural killer (NK) cells in patients with missed abortion: differ- stimulated by anti-CD200R2 derived dendritic cells suppress mixed ences between cases with normal and abnormal chromosome. Clin leukocyte cultures and skin graft rejection. Transplantation 2006; Exp Immunol 1999; 116(3): 449-52. 81(7): 1027-34. [185] Quack KC, Vassiliadou N, Pudney J, Anderson DJ, Hill JA. Leu- [205] Wright GJ, Cherwinski H, Foster-Cuevas M, et al. Characterization kocyte activation in the decidua of chromosomally normal and ab- of the CD200 receptor family in mice and humans and their inter- normal fetuses from women with recurrent abortion. Hum Reprod actions with CD200. J Immunol 2003; 171(6): 3034-46. 2001; 16(5): 949-55. [206] Daya S, Gunby J, Porter F, Scott J, Clark DA. Critical analysis of [186] Lédée-Bataille N, Laprée-Delage G, Taupin JL, Dubanchet S, intravenous immunoglobulin therapy for recurrent miscarriage. Frydman R, Chaouat G. Concentration of leukaemia inhibitory fac- Hum Reprod Update 1999; 5(5): 475-82. tor (LIF) in uterine flushing fluid is highly predictive of embryo [207] Stephenson MD, Dreher K, Houlihan E, Wu V. Prevention of un- implantation. Hum Reprod 2002; 17(1):213-8. explained recurrent spontaneous abortion using intravenous immu- [187] Kwak-Kim JY, Chung-Bang HS, Ng SC, et al. Increased T helper 1 noglobulin: a prospective, randomized, double-blinded, placebo- cytokine responses by circulating T cells are present in women with controlled trial. Am J Reprod Immunol 1998; 39(2): 82-8. recurrent pregnancy losses and in infertile women with multiple [208] Coulam CB, Krysa L, Stern JJ, Bustillo M.Intravenous immuno- implantation failures after IVF. Hum Reprod 2003; 18(4): 767-73. globulin for treatment of recurrent pregnancy loss. Am J Reprod [188] Ng SC, Gilman-Sachs A, Thaker P, Beaman KD, Beer AE, Kwak- Immunol 1995; 34(6): 333-7. Kim J. Expression of intracellular Th1 and Th2 cytokines in [209] Christiansen OB, Pedersen B, Rosgaard A, Husth M.A randomized, women with recurrent spontaneous abortion, implantation failures double-blind, placebo-controlled trial of intravenous immunoglobu- after IVF/ET or normal pregnancy. Am J Reprod Immunol 2002; lin in the prevention of recurrent miscarriage: evidence for a thera- 48(2): 77-86. peutic effect in women with secondary recurrent miscarriage. Hum [189] Lim KJ, Odukoya OA, Ajjan RA, Li TC, Weetman AP, Cooke ID. Reprod 2002; 17(3): 809-16. The role of T-helper cytokines in human reproduction. Fertil Steril [210] Hutton B, Sharma R, Fergusson D, et al. Use of intravenous immu- 2000; 73(1): 136-42. noglobulin for treatment of recurrent miscarriage: a systematic re- [190] Hill JA, Polgar K, Anderson DJ. T-helper 1-type immunity to tro- view. BJOG 2007; 114(2): 134-42. phoblast in women with recurrent spontaneous abortion. JAMA [211] Clark DA, Coulam CB, Stricker RB. Is intravenous immunoglobu- 1995; 273(24): 1933-6. lins (IVIG) efficacious in early pregnancy failure? A critical review [191] Piccinni MP, Beloni L, Livi C, Maggi E, Scarselli G, Romagnani S. and meta-analysis for patients who fail in vitro fertilization and Defective production of both leukemia inhibitory factor and type 2 embryo transfer (IVF). J Assist Reprod Genet 2006, 23(1): 1-13. T-helper cytokines by decidual T cells in unexplained recurrent [212] Carp HJ. Intravenous immunoglobulin: effect on infertility and abortions. Nat Med 1998; 4(9): 1020-4. recurrent pregnancy loss. Isr Med Assoc J 2007; 9(12): 877-80. [192] Bates MD, Quenby S, Takakuwa K, Johnson PM, Vince GS. Aber- [213] van den Heuvel MJ, Peralta CG, Hatta K, Han VK, Clark DA. rant cytokine production by peripheral blood mononuclear cells in Decline in number of elevated blood CD3(+) CD56(+) NKT cells recurrent pregnancy loss? Hum Reprod 2002; 17(9): 2439-44. in response to intravenous immunoglobulin treatment correlates [193] Kruse C, Varming K, Christiansen OB. Prospective, serial investi- with successful pregnancy. Am J Reprod Immunol 2007; 58(5): gations of in-vitro lymphocyte cytokine production, CD62L ex- 447-59. pression and proliferative response to microbial antigens in women [214] Winger EE, Reed JL, Ashoush S, Ahuja S, El-Toukhy T, Taranissi with recurrent miscarriage. Hum Reprod 2003; 18(11): 2465-72. M. Treatment with adalimumab (Humira) and intravenous immu- [194] Borzychowski AM, Croy BA, Chan WL, Redman CW, Sargent IL. noglobulin improves pregnancy rates in women undergoing IVF. Changes in systemic type 1 and type 2 immunity in normal preg- Am J Reprod Immunol 2009; 61(2): 113-20. nancy and pre-eclampsia may be mediated by natural killer cells. [215] Porter TF, LaCoursiere Y, Scott JR. Immunotherapy for recurrent Eur J Immunol 2005; 35(10): 3054-63. miscarriage. Cochrane Database Syst Rev 2006; (2): CD000112. [195] Chaouat G, Ledée-Bataille N, Dubanchet S, Zourbas S, Sandra O, [216] Pandey MK, Rani R, Agrawal S. An update in recurrent spontane- Martal J. TH1/TH2 paradigm in pregnancy: paradigm lost? Cytoki- ous abortion. Arch Gynecol Obstet 2005; 272(2): 95-108. nes in pregnancy/early abortion: reexamining the TH1/TH2 para- [217] McIntyre JA, Coulam CB, Faulk WP. Recurrent spontaneous abor- digm. Int Arch Allergy Immunol 2004; 134(2): 93-119. tion. Am J Reprod Immunol 1989; 21(3-4): 100-4. [196] Steck T, Giess R, Suetterlin MW, et al. Leukaemia inhibitory factor [218] Tamura M, Takakuwa K, Arakawa M, Yasuda M, Kazama Y, (LIF) gene mutations in women with unexplained infertility and re- Tanaka K. Relationship between MLR blocking antibodies and the current failure of implantation after IVF and embryo transfer. Eur J outcome of the third pregnancy in patients with two consecutive Obstet Gynecol Reprod Biol 2004; 112(1): 69-73. spontaneous abortions. J Perinat Med 1998; 26(1): 49-53. [197] Jack DL, Read RC, Tenner AJ, Frosch M, Turner MW, Klein NJ. [219] Daya S, Gunby J. The effectiveness of allogeneic leukocyte immu- Mannose-binding lectin regulates the inflammatory response of nization in unexplained primary recurrent spontaneous abortion. human professional phagocytes to Neisseria meningitidis serogroup Recurrent Miscarriage Immunotherapy Trialists Group. Am J Re- B. J Infect Dis 2001; 184(9): 1152-62. prod Immunol 1994; 32(4): 294-302. [198] Kilpatrick DC, Bevan BH, Liston WA. Association between man- [220] Pandey MK, Thakur S, Agrawal S. Lymphocyte immunotherapy nan binding protein deficiency and recurrent miscarriage. Hum Re- and its probable mechanism in the maintenance of pregnancy in prod 1995; 10(9): 2501-5. women with recurrent spontaneous abortion. Arch Gynecol Obstet [199] Dahl M, Tybjaerg-Hansen A, Schnohr P, Nordestgaard BG. Mor- 2004; 269(3): 161-72. bidity and mortality in persons with mannose-binding lectin defi- [221] Gatenby PA, Cameron K, Simes RJ, et al. Treatment of recurrent ciency. The Osterbro study. Ugeskr Laeger 2004; 166(50): 4606-9. spontaneous abortion by immunization with paternal lymphocytes: [200] Kruse C, Rosgaard A, Steffensen R, Varming K, Jensenius JC, results of a controlled trial. Am J Reprod Immunol 1993; 29(2): 88- Christiansen OB. Low serum level of mannan-binding lectin is a 94. determinant for pregnancy outcome in women with recurrent spon- [222] Mowbray JF, Gibbings C, Liddell H, et al. Controlled trial of taneous abortion. Am J Obstet Gynecol 2002; 187(5): 1313-20. treatment of recurrent spontaneous abortion by immunisation with [201] Shetty S, Ghosh K. Anti-phospholipid antibodies and other immu- paternal cells. Lancet 1985; 1(8435): 941-3. nological causes of recurrent foetal loss--a review of literature of [223] Worldwide collaborative observational study and meta-analysis on various therapeutic protocols. Am J Reprod Immunol 2009; 62(1): allogenic leukocyte immunotherapy for recurrent spontaneous 9-24. abortion. Recurrent Miscarriage Immunotherapy Trialists Group. [202] Morikawa M, Yamada H, Kato EH, et al. Massive intravenous Am J Reprod Immunol 1994; 32(2): 55-72. immunoglobulin treatment in women with four or more recurrent 218 Current Women’s Health Reviews, 2010, Vol. 6, No. 3 Ly et al.

[224] Kling C, Steinmann J, Westphal E, Magez J,Kabelitz D. Experi- [227] Research, C.f.B.E.a., Lymphocyte Immune Therapy. 2001. ence with allogeneic leukocyte immunization for implantation [228] Kling C, Steinmann J, Westphal E, Magez J, Kabelitz D. Adverse failure in the in-vitro fertilization program. Am J Reprod Immunol effects of intradermal allogeneic lymphocyte immunotherapy: acute 2002; 48: 147. reactions and role of autoimmunity. Hum Reprod 2006; 21(2): 429- [225] Ober C, Karrison T, Odem RR, et al. Mononuclear-cell immunisa- 35. tion in prevention of recurrent miscarriages: a randomised trial. [229] Pearlman SA, Meek RS, Cowchock FS, Smith JB, McFarland J, Lancet 1999; 354(9176): 365-9. Aster RH. Neonatal alloimmune thrombocytopenia after maternal [226] Clark DA, Chaouat G. Immunomodulation by CD200 as a basis immunization with paternal mononuclear cells: successful treat- for prevention of cytokine-dependent fetal loss syndrome. Am J ment with intravenous gamma globulin. Am J Perinatol 992; 9(5- Reprod Immunol 2003; 49: 350-1. 6): 448-51.

Received: January 17, 2010 Revised: March 29, 2010 Accepted: June 15, 2010