**4. Results**

### **4.1 Studies examining biological factors playing a role to ectopic pregnancy**

A total of 42 abstracts were retrieved and further screened. Only studies that performed basic investigations were included in this part. Out of 20 included studies, 3 were performed in USA (15%), 2 in Germany (10%), Israel, Sweden and UK, and from one (5%) in Brazil, Canada, China, Croatia, Denmark, Finland, Hungary, India and Poland. 8 (40%) of them considered themselves clinical, 8 (40%) experimental, 1 (5%) prospective, 1 (5%) as pilot study, 1 (1%) as preliminary report, 1 (5%) as hypothesis testing. The type of study per country initiated can be seen in Fig 1.

All studies selected used human tissue. Five studies used fallopian tubes (25%). From the other studies 1 (5%) used fallopian tube and peripheral blood, 1 used deciduas, placenta, primary first trimester trophoblast cells and peripheral blood, 1 used decidual tissue, 2 endometrial tissue, 1 epithelial tissue and 1 mucosal tissue, 1 cervical specimen and

Medline searches (up to March 2011) were performed using various combinations of terms: ectopic pregnancy, heterotopic pregnancy, *In Vitro* Fertilization, Intrauterine Insemination, pelvic inflammatory disease, Chlamydia trachomatis, heterotopic pregnancy, cervical

The search was complemented with perusal of the bibliographies of retrieved papers and review articles. We included studies that evaluated the presence of an ectopic pregnancy after IVF in case reports, although in other chapters information was obtained from studies

Number of tested samples was not an exclusion criterion. Only studies including human

For each study, information was obtained on authors, journal, year of publication, country and years of study enrollment, study design and study target, number of tested samples, tissue and disease tested, searched molecules and pathways involved, clinical outcome and

Data extraction was performed independently by two investigators, and conflicts were

Frequencies of all important parameters were performed. Statistical analyses were performed in using the Statistical Package for Social Sciences (SPSS) version 12.0 (SPSS,

A total of 42 abstracts were retrieved and further screened. Only studies that performed basic investigations were included in this part. Out of 20 included studies, 3 were performed in USA (15%), 2 in Germany (10%), Israel, Sweden and UK, and from one (5%) in Brazil, Canada, China, Croatia, Denmark, Finland, Hungary, India and Poland. 8 (40%) of them considered themselves clinical, 8 (40%) experimental, 1 (5%) prospective, 1 (5%) as pilot study, 1 (1%) as preliminary report, 1 (5%) as hypothesis testing. The type of study per

All studies selected used human tissue. Five studies used fallopian tubes (25%). From the other studies 1 (5%) used fallopian tube and peripheral blood, 1 used deciduas, placenta, primary first trimester trophoblast cells and peripheral blood, 1 used decidual tissue, 2 endometrial tissue, 1 epithelial tissue and 1 mucosal tissue, 1 cervical specimen and

No sponsor was involved in the study design, report writing, or paper submission.

**4.1 Studies examining biological factors playing a role to ectopic pregnancy** 

**2. Methods** 

pregnancy, cost-effectiveness.

in the general population.

subjects were included.

resolved after discussion.

**3. Main outcomes 3.1 Statistical methods** 

Chicago, IL, USA).

**4. Results** 

**3.2 Founding source** 

country initiated can be seen in Fig 1.

**2.2 Data extraction** 

**2.1 Identification and eligibility of relevant studies** 

whether biopsy was performed with the site.

Fig. 2. Type of study per tissue used when examining biological factors in ectopic pregnancy

fallopian tube samples, 1 used human endometrium and fallopian tube, 2 used human placental tissue, 1 used ovarian, prostate, endometrial, tubal and semen, 1 used trophoblast, 2 used serum samples, 1 used transervical specimens and one used stimulated cervical mononuclear cell supernatants. The type of study per tissue used can be seen in Fig 2.

4 (20%) of the studies were published in American Journal of Reproductive Immunology, 2 (10%) in Human Reproduction and from 1 (5%) in Molecular Human Reproduction, Reproduction, Reproductive Biology & Endocrinology, Reproductive Sciences, Biology of Reproduction, Cellular Microbiology, Clinical and Vaccine Immunology, European Journal of Obstetrics & Gynecology, Histochemistry & Cellular Biology, Infection and Immunity, The Journal of Immunology, The Journal of Infectious Diseases ,The Journal of Clinical Endocrinology & Metabolism, The Medical Hypothesis journal. The distribution of type of study per Journal is seen in Fig 3. Eight studies (40%) did not mentioned their controls, four studies (20%) used normal pregnant patients and intrauterine pregnancy, and from one study (5%) used women with no infection and without infertility problem, normal desidual tissue, normal endometrium and normal Fallopian tube, normal pregnant patient peripheral blood, spontaneous abortion,, tissue from women undergoing tubal ligation with segmental resection and women with viable and non-viable intrauterine pregnancy.

Samples size examined ranged from 3 (in each group) to 144. Disease distribution examined presented as: ectopic pregnancy (8/40%), spontaneous abortion and ectopic pregnancy (2/10%), Chlamydia infection (2/10%), Chlamydia infection in patients with no infertility compared with women with Chlamydia and tubal damage, ectopic pregnancy and decidualized endometrium, ectopic pregnancy and blighted ovum, ectopic pregnancy and Chlamydia infection, pelvic inflammatory disease and ectopic pregnancy, post IVF ectopic pregnancy, viable ectopic pregnancy while 1 (5%) did not mentioned disease. Funding source of each study per Journal published is seen on Fig 4.

Ectopic Pregnancy and Assisted Reproductive Technologies: A Systematic Review 49

All studies except two (3/15%) mentioned the molecule studied. Two studies (10%) mentioned C Trachomatis DNA, one C Trachomatis serum antibodies, while from one study mentioned CD14B7H4,CHSP-60,E-Cadherin,estrogen receptor, IgG antibodies, IL-1, IL-8, Ki-67, MMP-2, PIGF, SLIT/ROBO proteins, Svcam-1, TAG-72, Treg. Ten studies (50%) did not mentioned a second molecule while the other studies mentioned CT antibodies (1/5%) chlamydial sarkosyl-soluble 57-kDa protein, cHSP 60, Cytokeratin 7, estrogen receptor, hCG, IL-1a, MMP-9, P38 and progesterone. 11 studies (55%) did not mentioned third outcome while the other ones mentioned Chsp 60 (2/10%), bhCG, cHSP 10 (1/5%), ERK, Fibronectin, FOXP 3, IL-10 and MMP-14. Thirteen studies (13/65%) did not mentioned a fourth outcome, while the other mentioned activin (1/5%) IFN-γ, Laminin, MAPK, neurophilin, p38 inhibitor and TIMP-1. Fifth molecule was mentioned in only 6 studies. Only four studies mentioned pathways involved: invasive pathway, p38 MAP-kinasses pathway, ERK-MAPK pathway and SLIT/ROBO pathway. Only two studies (2/5%)

In terms of bio-analytic techniques, PCR techniques and immunohistochemistry was mainly used. For immunohistochemistry techniques, four studies (20%) did not mentioned it, seven studies (7%) mentioned as immunohistochemistry, 23 (15%) as immunofluorence immunohistochemistry, and from one as electron microscopic immunohistochemistry, immunoblotting, immunoperoxidase staining (IP), immunosorbent assay, microimmunofluoresence, microimmunofluorence-immunoblot. For PCR techniques, eleven studies (55%) did not mentioned PCR technique, 4 (20%) studies mention it as quantitative PCR, 2 (10%)

When ectopic pregnancy exists, Arias-Stella reaction is observed in endometrium. At that time, B7H4 positive macrophages is significantly lower when compared with secretory

Chlamydia trachomatis are highly associated with ectopic pregnancy (Brunham et al., 1992). Chlamydia trachomatis antigens, exist in asymptomatic, culture negative men and women with chronic infection and may act as immunostimulants and re-activate Chlamydia (Toth et al., 2000). After first episode of ectopic pregnancy, antibody response to conserved epitope of cHSP-60 (Chlamydia heat shock protein) is associated with increased probability of adverse pregnancy outcome (Sziller et al., 2007). So this biomarker could be used for counseling women with first episode of ectopic pregnancy: if sensitized to this epitope, *in vitro* fertilization should be offered. Another indication for the significance of this biomarker is that in infertile women, when Chlamydia infected tissue is exposed to Chlamydia heat shock proteins (cHSP-60 and cHSP-10) increased release of IFN-gamma, IL-10 and TNFalpha (Srivastava et al., 2008) affect mucosal immune function. From the other side, Ct-IgG and c-hsp6 antigens, were not found as an independent predictor of ectopic pregnancy

In the fallopian tube of serologically positive patients for Chlamydia trachomatis that had ectopic pregnancy, there is an increase in the expression of activin βΑ subunit, type II

In infected tubes from Chlamydia trachomatis infection, interleukin -1 production from epithelial cells initiates tissue destruction. By blocking IL-1 with IL-1RA receptor antagonist and/or IL-10, tissue destruction is eliminated. Chlamydia infected cells, also, produce IL-8

studies mention it as PCR, two (10%) as RT-PCR, and 1 (5%) as N-PCR.

**4.2 Outcomes** 

mentioned intervention medication.

endometrium (Wicherek et al., 2009)

(Bjartling et al., 2006).

receptors, follistatin and iNOS (Refaat et al., 2009).

by ERK MAPK pathway (Buchholz et al., 2007).

Fig. 3. The distribution of type of study per Journal when examining biological factors in ectopic pregnancy

Fig. 4. Funding source of each study per Journal published

#### **4.2 Outcomes**

48 Ectopic Pregnancy – Modern Diagnosis and Management

Fig. 3. The distribution of type of study per Journal when examining biological factors in

Fig. 4. Funding source of each study per Journal published

ectopic pregnancy

All studies except two (3/15%) mentioned the molecule studied. Two studies (10%) mentioned C Trachomatis DNA, one C Trachomatis serum antibodies, while from one study mentioned CD14B7H4,CHSP-60,E-Cadherin,estrogen receptor, IgG antibodies, IL-1, IL-8, Ki-67, MMP-2, PIGF, SLIT/ROBO proteins, Svcam-1, TAG-72, Treg. Ten studies (50%) did not mentioned a second molecule while the other studies mentioned CT antibodies (1/5%) chlamydial sarkosyl-soluble 57-kDa protein, cHSP 60, Cytokeratin 7, estrogen receptor, hCG, IL-1a, MMP-9, P38 and progesterone. 11 studies (55%) did not mentioned third outcome while the other ones mentioned Chsp 60 (2/10%), bhCG, cHSP 10 (1/5%), ERK, Fibronectin, FOXP 3, IL-10 and MMP-14. Thirteen studies (13/65%) did not mentioned a fourth outcome, while the other mentioned activin (1/5%) IFN-γ, Laminin, MAPK, neurophilin, p38 inhibitor and TIMP-1. Fifth molecule was mentioned in only 6 studies. Only four studies mentioned pathways involved: invasive pathway, p38 MAP-kinasses pathway, ERK-MAPK pathway and SLIT/ROBO pathway. Only two studies (2/5%) mentioned intervention medication.

In terms of bio-analytic techniques, PCR techniques and immunohistochemistry was mainly used. For immunohistochemistry techniques, four studies (20%) did not mentioned it, seven studies (7%) mentioned as immunohistochemistry, 23 (15%) as immunofluorence immunohistochemistry, and from one as electron microscopic immunohistochemistry, immunoblotting, immunoperoxidase staining (IP), immunosorbent assay, microimmunofluoresence, microimmunofluorence-immunoblot. For PCR techniques, eleven studies (55%) did not mentioned PCR technique, 4 (20%) studies mention it as quantitative PCR, 2 (10%) studies mention it as PCR, two (10%) as RT-PCR, and 1 (5%) as N-PCR.

When ectopic pregnancy exists, Arias-Stella reaction is observed in endometrium. At that time, B7H4 positive macrophages is significantly lower when compared with secretory endometrium (Wicherek et al., 2009)

Chlamydia trachomatis are highly associated with ectopic pregnancy (Brunham et al., 1992). Chlamydia trachomatis antigens, exist in asymptomatic, culture negative men and women with chronic infection and may act as immunostimulants and re-activate Chlamydia (Toth et al., 2000). After first episode of ectopic pregnancy, antibody response to conserved epitope of cHSP-60 (Chlamydia heat shock protein) is associated with increased probability of adverse pregnancy outcome (Sziller et al., 2007). So this biomarker could be used for counseling women with first episode of ectopic pregnancy: if sensitized to this epitope, *in vitro* fertilization should be offered. Another indication for the significance of this biomarker is that in infertile women, when Chlamydia infected tissue is exposed to Chlamydia heat shock proteins (cHSP-60 and cHSP-10) increased release of IFN-gamma, IL-10 and TNFalpha (Srivastava et al., 2008) affect mucosal immune function. From the other side, Ct-IgG and c-hsp6 antigens, were not found as an independent predictor of ectopic pregnancy (Bjartling et al., 2006).

In the fallopian tube of serologically positive patients for Chlamydia trachomatis that had ectopic pregnancy, there is an increase in the expression of activin βΑ subunit, type II receptors, follistatin and iNOS (Refaat et al., 2009).

In infected tubes from Chlamydia trachomatis infection, interleukin -1 production from epithelial cells initiates tissue destruction. By blocking IL-1 with IL-1RA receptor antagonist and/or IL-10, tissue destruction is eliminated. Chlamydia infected cells, also, produce IL-8 by ERK MAPK pathway (Buchholz et al., 2007).

Ectopic Pregnancy and Assisted Reproductive Technologies: A Systematic Review 51

protein expression (Duncan et al., 2010). Also serum VCAM-1 was found comparable between the three pregnancy types, normal, ectopic and failed thus making this marker not

Also a role for ectopic pregnancy formation might exist from ovulation induction regimens. Clomiphene citrate, which used for anovulatory infertility, may indirectly contribute to ectopic pregnancy creation. Chronic treatment with clomiphene activates estrogen receptors, particularly in cilia, and inducing tubal apoptosis of isthmus epithelial cells while slowing oocyte cumulus complex passage from the fallopian tube (Shao et al., 2009). This may

A total of 56 abstracts were retrieved and further screened. Out of 40 included studies, 10 (25%) were performed in USA, 3 (7.5%) in England, India and in Netherlands, 2 (5%) in France, Greece, Israel, Italy and Nepal, while 1 (2.5%) in Australia, Canada, Germany, Japan, Jordan, Korea, Norway, Singapore, Taiwan, Thailand and between USA/Sweden. 17 (42.5%) of them considered themselves retrospective, 11 (27.5%) as case report, and from 1 (2.56%) as cohort study, prospective randomized double blinded cross over study, prospective cohort study, population based cost-effectiveness study, epidemiological study, economical analysis, cost effectiveness analysis, cost benefit analysis, cost analysis, while 3 (7.5%) do not mentioned the

useful for the diagnosis of ectopic pregnancy (Daniel et al., 2000).

type of study. The type of study per country initiated can be seen in Fig 5.

Fig. 5. The type of study per country initiated examining IVF and ectopic pregnancy

All studies selected used human tissue. Fourteen studies mentioned embryo (35%) as tissue examined plus 4 (10%) mentioned especially the blastocyst, 4(10%) used fallopian tube and

contribute to ectopic pregnancy formation.

**5.1 Ectopic pregnancy and IVF** 

**5. Section 2** 

Regulators T cells (Treg) express LH/CG receptor on their surface during pregnancy and are present at the fetal-maternal interface, attracted by high levels of human chorionic gonadotrophin. In ectopic pregnancies, regulatory T cells are not attracted to the same degree by the lower levels of hCG (Schumacher et al., 2009).

Another possible biomarker for ectopic pregnancy after IVF is E-cadherin, because it is highly expressed in cytotrophoblast cells of chorionic villi from these pregnancies, when compared with spontaneous ectopic pregnancy (Revel et al., 2008).

From micro-array studies, Savaris et al, by using the model of ectopic pregnancy, found that the transcriptome of the decidua is influenced by trophoblast products, in endocrine fashion (Savaris et al., 2008).

In close proximity of the tubal implantation site, MUC-1 and TAG-72 are present in the epithelial cells and might contribute to the deeper trophoblast invasion in the tubal wall (von Rango et al., 2003) while a significant reduction in NK-cell numbers at the tubal implantation site, could be seen, induced by local antigen-presenting cells in the presence of mucins (Laskarin et al., 2010).

Viable tubal pregnancies implant at the mesosalpingeal side of the tubal wall show a massively increased invasion of extravillous trophoblast cells (EVT) into the tubal wall, and the proliferation of trophoblast cells extends deeply into the invasive zone in the invasive pathway (Kemp et al., 1999). Tubal pregnancies that will undergo tubal abortion implant at the antimesometrial side which show shallow invasion and poor trophoblast proliferation (Kemp et al., 1999). In extrauterine pregnancy, expression of integrin subunit α3 is nearly exclusively restricted to the basal plasmalemma of the first layer of trophoblast cells while only the first proximal layer of EVT (in direct contact to the basement membrane) expresses integrin α6. The switch to the integrin subunits αv and α5 takes place already in the second layer of trophoblast cells, as soon as the latter detach from the basement membrane (Kemp et al., 2002).

In tubal pregnancies, MMP-9 and TIMP-1, -2 and -3 are produced by all types of extravillous cytotrophoblast (EVCT) cells, while MMP-2 and -14 mainly exist in distal column cytotrophoblast (CCT) cells and invasive EVCT cells (Bai et al., 2005). In parallel, MMP-14 and TIMP-1 and -2 are increased along the invasive pathway toward maternal interstitium. MMP-2, -9 and -14 and TIMP-1, -2 and -3 were all detected in the villous CT (VCT) cells (Bai et al., 2005).

Another promising technique for early detection of an ectopic pregnancy is that trophoblast cells can be reliably obtained and identified among cervical cells in the first trimester and labeled with antibody to HLA-G. The number of trophoblast cells per total cervical cells (trophoblast frequency) is significant lower in ectopic pregnancy when compared to intrauterine pregnancy but not to blighted ovum (Imudia et al., 2009). Using ROC curves the positive predictive value for abnormal pregnancy was 97% and the negative predictive value was 87%. From intrauterine tissue and sera, Horne et al, examined PIGF, localized to the cytotrophoblast cells (Horne et al, 2010). Expression of PIGF mRNA was significantly reduced in trophoblast cells, isolated from women with ectopic pregnancy compared with intrauterine pregnancies. Serum PIGF was undetectable in women with tubal ectopic pregnancies and reduced, or undetectable, in miscarriage compared with viable intrauterine pregnancies.

From the other side, it seems that SLT/ROBO pathway and protein expression in endometrium and fallopian tube, is not implicated in ectopic pregnancy because known factors that contribute to EP (e.g. smoking/cotinine or chlamydial infection) do not alter protein expression (Duncan et al., 2010). Also serum VCAM-1 was found comparable between the three pregnancy types, normal, ectopic and failed thus making this marker not useful for the diagnosis of ectopic pregnancy (Daniel et al., 2000).

Also a role for ectopic pregnancy formation might exist from ovulation induction regimens. Clomiphene citrate, which used for anovulatory infertility, may indirectly contribute to ectopic pregnancy creation. Chronic treatment with clomiphene activates estrogen receptors, particularly in cilia, and inducing tubal apoptosis of isthmus epithelial cells while slowing oocyte cumulus complex passage from the fallopian tube (Shao et al., 2009). This may contribute to ectopic pregnancy formation.
