Factors related to infertility in Brazil and their relationship with success rates after assisted reproduction treatment: an integrative review

This integrative review evaluated the most commonly diagnosed causes of infertility in men and women in Brazil, as well as the medically assisted reproduction technologies regularly employed in these cases. We searched in four electronic databases (PubMed, including Medline; Scopus; Web of Science and LILACS), and two grey literature (Google Scholar and OpenGrey), guided by the focused question: "What are the main factors responsible for male and female infertility in Brazil, and what are its relationships with success rates after assisted reproduction treatment?". We included interventional or observational studies, without limitation by language or year of publication. Our searches in the electronic indexers recovered 1,119 articles, and after analyzing the inclusion and exclusion criteria, 27 articles composed the body of analysis for this review. We grouped the studies into four themes: factors responsible for male and female infertility, assisted reproductive technologies (ART) used in the infertility treatment, assisted reproduction procedures, and clinical predictors of success rates in ART. Despite the scarcity of studies analyzing the association between infertility and assisted reproductive technologies in Brazil, it was possible to infer that the most prevalent infertility cause in women was endometriosis, while in men it was azoospermia. The most widely assisted reproductive technology applied in the country is the intracytoplasmic injection of spermatozoa (ICSI), ensuring better success rates in the treatment of infertility for men and women.


INTRODUCTION
The World Health Organization (WHO) defines infertility as the inability of a couple to establish a clinical pregnancy after 12 months of regular unprotected sexual intercourse. Although commonly perceived as a female disorder, male factors are equally prevalent when considering the causes of infertility. Thus, it is estimated that female factors account for 35 to 40% of the causes of infertility, male factors for 20 to 40% and factors in which both have dysfunctions represent 20 to 30%, as the other idiopathic causes, causing sterility without apparent cause (Nardelli et al., 2014;WHO, 2019;Zeqiraj et al., 2018).
Considering the treatment of infertile patients, assisted reproduction becomes an important alternative to try to make pregnancy possible to couples that have difficulty in conceiving naturally. Fertilization rates in Brazil show that the country's assisted reproduction services are effective, reaching international standards. Data from the 12 th report of the National System of Embryo Production (SisEmbrio) indicate that the average success rate in fertilization is 76%, which is within the quality standards suggested in the international literature, ranging from 65% to 75% (SART, 2019;SisEmbrio, 2019).
Although intracytoplasmic sperm injection is one of the most widely used treatments, assisted reproductive techniques also include conventional in vitro fertilization, artificial insemination and methods for monitoring reproductive cycles; and the method or technique to be used is chosen according to the patient's clinical condition. However, little is known, about the most prevalent causes of infertility in Brazil and which treatment is the most appropriate for each case. Considering the importance of this knowledge to support decision making during the planning of an assisted reproduction treatment, the objective of this integrative review was to answer the focused question: "What are the main factors responsible for male and female infertility in Brazil, and its relationship with success rates after assisted reproduction treatment?"

MATERIAL AND METHODS
This integrative review covered the following stages: establishing the hypothesis and objectives of the integrative review; establishing the criteria for paper inclusion and exclusion (sample selection); defining the information to extract from the selected papers; evaluation of the studies included; interpretation of the results and synthesis of the studies. We used the PICO strategy to formulate the question addressed in this review, which is an acronym for Patient (or Population), Intervention, Comparison and Outcomes (Higgins et al., 2013).
This strategy allowed the identification of keywords, which helped locate relevant primary studies in the databases. Thus, the question that outlined the study was "What are the main factors responsible for male and female infertility in Brazil, and what is its relationship with the success rates after assisted reproduction treatment?" So, based on this question, the first element of the strategy (P) consists on patients with infertility; the second (I) refers to the assisted reproduction treatments; the third element (C), refers to the comparison of the interventions applied to the patients, and the fourth element (O) deals with the technique's effectiveness.

Inclusion criteria
The inclusion criteria for the paper selection were: primary studies, conducted in Brazil; that cited infertility; which assisted reproduction treatment was employed; and the treatment outcome (whether there was pregnancy or not). There was no restriction regarding the time of publication or language.

Exclusion criteria
We excluded papers based on the following criteria: (1) studies with animals; (2) treatments conducted in other countries; (3) studies that did not mention the infertility condition; (4) studies that did not mention the assisted reproductive technique employed; (5) studies that did not present the outcome of the treatment; (6) literature reviews, summaries, books, chapters of books, letters, opinion article, technical papers and guidelines.

Information sources
We conducted a computerized bibliographic search in four databases: PubMed (including Medline), Scopus, Web of Science and LILACS (Latin American and Caribbean Literature in Health Sciences), and two grey literature: Google Scholar and OpenGrey, for any references that might could have been missed. Additional information on search strategies is provided in Appendix 1. All searches were conducted on April 14, 2019.

Search
We combined the main descriptors related to the subjects we investigated, listed below, using the Boolean operators "AND" and "OR": "Assisted Reproductive Technology" OR "Assisted Reproductive Technologies" OR "Assisted Reproductive Technique" OR "Assisted Reproductive Techniques" OR "Assisted Reproductive Technic" OR "Assisted Reproductive Technics" AND "Brazil" (more details in Appendix 1).
We chose these keywords because they are associated with the object of study, and they belong to the catalog of PubMed descriptors (MESH Terms). We used the same keywords in LILACS, adding in this case their translations into Portuguese and Spanish, according to the Health Sciences Descriptors (DeCS).
We checked the references we obtained through the electronic search, and we removed the duplicates using the EndNote Web software (Thompson Reuters).

Study selection and data collection process
In the first phase, four independent reviewers selected the studies (K.S.M., M.A.G.S.M., E.A.D.A. and G.R.S.), they evaluated the titles and summaries of the articles we identified by the search strategy. At that stage, we sorted the articles according to the eligibility criteria established. We retained for further analysis of the full text those articles that generated questions and were potentially eligible. In the second phase, we fully read the studies selected by the four reviewers, which made it possible to exclude other articles because they did not meet the revision proposal. In the third phase, we extracted the main information of the articles and synthesized in summary tables, so that they could guide the critical analyses of the studies selected. We resolved the disagreements by discussion, consultation and guidance of a fifth reviewer (D.B.M.).

Data items
The data we collected included the year of publication, place of study/treatment, type of study (observational or interventional), infertility condition of the patient (men and women), assisted reproductive techniques employed in each case and the outcome of the treatment (if there was implantation, gestation and live births).

Data synthesis
We calculated the prevalence of each infertility condition dividing the number of times in which each condition was cited in the articles by the total number of conditions presented in this review, multiplying the result by 100. Figure 1 depicts the bibliographic search process and the review selection criteria. The searches in the electronic databases and in the grey literature recovered 1,125 papers. After removing the duplicates, we evaluated 1,119 papers, and we excluded 916, resulting in 203 articles for full text reading. In this second step, we excluded 168 papers, because they did not meet the eligibility criteria. Thus, we included 27 remaining studies in the qualitative and quantitative synthesis. Table 1 shows a summary of the characteristics of the studies included in this review. Through the categorized analysis of the 27 articles that composed the final sample, we found that all were published between 2003 and 2019. Of these, 4 were cross-sectional studies Esteves & Glina, 2005;Cota et al., 2012;Souza et al., 2017), 12 were cohort studies (Pinheiro et al., 2003;Borges et al., 2003;Glina et al., 2005;Pasqualotto et al., 2005;Romão et al., 2010;Semião-Francisco et al., 2010;Cota et al., 2012;Borges et al., 2013;Coelho Neto et al., 2015;Maia-Filho et al., 2015;Scheffer et al., 2017;Bercaire et al., 2018); 6 were case-control studies (Setti et al., 2011;Pasqualotto et al., 2012;Cavagna et al., 2012;Picinato et al., 2014;Donabela et al., 2015;Costa et al., 2016); 3 were case reports (Valle et al., 2012;Taitson et al., 2012;Borges et al., 2016) and 2 were randomized controlled trials (Geber & Sampaio, 2013;Nastri et al., 2013).

DISCUSSION
This integrative review investigated the main conditions of male and female infertility in Brazil, and their relationship with success rates after the treatment of assisted Review JBRA Assist. Reprod. | v.25 | nº1 | Jan-Feb-Mar / 2021 reproduction. By our knowledge, it is the first to address this theme, allowing the compilation about the relations between such conditions and the assisted reproduction treatments employed in the country.

Main factors responsible for female and male infertility
The main factors associated with female infertility included endometriosis, tubal factor, polycystic ovary syndrome, endocrine/anovulatory and advanced age; while for males it included seminal alterations (azoospermia, oligozoospermia, asthenozoospermia and teratozoospermia), varicocele and infertility after vasectomy, and anatomical factors.
The most prevalent factor in women is endometriosis, which consists of having endometrial tissue outside the uterus, inducing a chronic inflammatory reaction and the formation of adhesions. In this pathology, the endometrial tissue is commonly lodged on the peritoneal surface, ovaries and rectovaginal septum. Endometriosis is increasing in women and has an important relationship with female infertility, since endometrial tissue even outside the uterus, continues to be stimulated monthly by the action of the menstrual cycle hormones. This leads to an inflammatory reaction, which consequently influences the hormonal regulation necessary for ovulation, as well as embryo implantation (Crosera et al., 2010;Nácul & Spritzer, 2010).
The peritoneal tube factor is also an important cause of female infertility. Pelvic inflammatory diseases, especially those caused by Chlamydia trachomatis and Neisseria gonorrhoeae, are undoubtedly among the most common causes of functional loss of uterine tubes, besides other causes such as endometriosis -which leads to tubal infertility (Fernandes et al., 2014).
Despite advances in assisted reproduction, poor ovarian response (POR) to gonadotrophin stimulation remains a problem in fertility treatment, especially in IVF, where a large number of oocytes are desirable. Many protocols and adjuvant therapies have already been proposed for the treatment of bad responders; among them ANDRO-IVF, a new protocol that promotes intraovarian androgenization. The idea that testosterone could be useful to improve poor ovarian response is based on studies which suggest that androgens play an important role in the early stages of follicular development. Besides that, the increase in intraovarian concentration of androgens appears to raise the expression of Follicle-Stimulating Hormone (FSH) receptors in granulosa cells, increasing ovary sensitivity to FSH. Transdermal testosterone before the IVF cycle seems to increase the rate of pregnancy and births by decreasing the   Continuation Table 1. required dose of gonadotropin, as well as the number of days of stimulation. Although these studies are still scarce, they have been promising, bringing new possibilities of adjuvant treatment to patients with low response to ovarian stimulation (Setti et al., 2011;Borges et al., 2013). Polycystic ovary syndrome (POS) is the main gynecological endocrinopathy in women of reproductive age, and it is the most common cause of infertility due to anovulation. This endocrinopathy occurs in the ovaries from an imbalance in hormonal levels of androgens and insulin, causing the formation of ovarian cysts, which can interfere with the ovulation process. In this case, the patient does not ovulate properly and, therefore, may present long intervals between menstrual cycles, and may remain without menstruation for months, thus leading to infertility (Santana et al., 2008).
Advanced maternal age is also an important factor associated with female infertility, since natural fecundity and pregnancy rates decline with increasing age. This is due to significant reductions in the number and quality of the oocytes, since women are born with a limited pool of oocytes, which degenerate throughout their lives at each menstrual cycle. However, the main factor in the etiology of age-related female infertility is the decline in oocyte quality, since older women have higher rates of single chromatid abnormalities in oocytes, as well as chromosomal aneuploidies (Scheffer et al., 2017).
Azoospermia was the most prevalent factor in men, which consists of the absence of sperm in the semen. It is classified as obstructive or excretory azoospermia (OA), in cases where production is normal, but there is excretory pathway obstruction (epididymis, deferent or ejaculatory duct); vasectomy is the main example, and non-obstructive or secretory azoospermia (NOA), when there is testicular failure leading to the abolition of sperm production. In combination, there is azoospermia in approximately 10% of men with infertility (Harris & Sandlow, 2008;Vieira et al., 2009).
Seminal alterations, such as azoospermia Glina et al., 2005;Esteves & Glina, 2005;Pasqualotto et al., 2005;Semião-Francisco et al., 2010;Taitson et al., 2012), oligozoospermia (Borges et al., 2016), asthenozoospermia (Borges et al., 2016) and teratozoospermia (Borges et al., 2016) were the most frequently cited infertility conditions in this review, and refer to abnormalities in sperm number, motility and morphology. Azoospermic men do not have sperm in their ejaculate, while in oligozoospermia there is a reduction in the amount of sperm present in the ejaculated fluid. In asthenozoospermia, there is a decreased spermatozoa motility, which can cause the inability or difficulty for sperm to fertilize the oocyte, while in teratozoospermia there is a large percentage of abnormally shaped sperm. There are morphology defects in different parts of the sperm: head, intermediate part and tail, which may render the natural fertilization, process impossible (Zegers-Hochschild et al., 2017).
Varicoceles are also an important cause of male infertility, being a condition that involves dilations and crooked veins of the pampiniform plexus, which drain blood from the testicles. Such alterations cause changes in temperature, oxygenation, nutrition and release of free radicals in testicular cells, which may compromise spermatogenesis and, consequently, cause male infertility (Cocuzza, 2011).
Another condition associated with male infertility is post-vasectomy infertility, a form of obstructive azoospermia, caused by the ligation of the deferent ducts in man. In these cases, men become infertile due to interruptions in sperm transit. However, given the widespread use of vasectomy as a method of contraception, there is a growing demand for vasovasostomy, which consists of surgical reversal, in which the two separate parts of the deferent duct are reconnected. ICSI combined with percutaneous epididymis sperm aspiration (PESA), or testicular sperm aspiration (TESA) is currently a widely used approach for fertility restoration in post-vasectomy cases, where anastomosis has failed. However, there is a relationship between the post-vasectomy period and the reproductive capacity of spermatozoa, that is, the shorter it is, the higher the chances of fertility return by ICSI. The rates of pregnancy and implants decrease significantly when sperm recovery is performed 14 years after surgery .
The articles included in this revision also mentioned infertility caused by anatomic factors, such as the congenital bilateral absence of the deferential duct, an anomaly responsible for approximately 6% of the cases of obstructive azoospermia, and for 1-2% of the cases of infertility in men (Bernardino et al., 2003).

Assisted reproductive technologies most commonly used in infertility treatment
Five techniques stand out today in medically assisted reproduction: intrauterine insemination (IUI), IVF, ICSI and IMSI. IUI is a simple procedure of three steps: follow-up of natural follicular development, semen collection and processing and artificial insemination through a catheter. IVF, in turn, is based on the collection of gametes, which are placed in contact on the same culture plate, and the sperm will then fertilize the oocyte outside the female body. After embryo fertilization and cultivation in the laboratory, it is transferred to the uterus. ICSI is a variant of IVF, in which we select a spermatozoon and inserted it into the oocyte cytoplasm with the help of a micromanipulator, surpassing the radiate crown and the zona pellucida, thus increasing fertilization rates. Finally, IMSI or super ICSI, is a technique that carefully selects sperm based on morphology, using a microscope that enables a magnification higher than 6,000 times, while in ICSI this increase is of 400X, which, in the second case, does not enable the identification of some morphological abnormalities. However, IMSI is not used world widely anymore, because the technology offers no benefit over ICSI for live birth or miscarriage rates; although there is evidence that IMSI improves clinical pregnancy rates, this evidence is of very low quality (Santos, 2010;González-Ortega et al., 2010;Sermondade et al., 2011;Teixeira et al., 2020).
Among the AR technologies used in the country, ICSI is usually the main technique of choice, since it enables to solve the problems of infertility in most cases in which sperm quantity or motility is significantly reduced Glina et al., 2005;Esteves & Glina, 2005;Pasqualotto et al., 2005;Semião-Francisco et al., 2010;Taitson et al., 2012). The greatest advantage compared to other techniques is that fertilization is less affected by concentration, motility and/or sperm morphology, and its disadvantage is the high cost of the procedure. Therefore, ICSI is today the most used technique in cases where there are serious changes in seminal parameters (Nagy et al., 1995;Santos, 2010).
Considering the success rates achieved after the treatments employed, IVF was more effective in cases of infertility caused by advanced or profound endometriosis; in women with absence of uterine tubes or tubal lesions that prevented natural fertilization or artificial insemination, and in cases of idiopathic infertility. On the other hand, ICSI was effective in cases similar to conventional IVF, and in cases of severe male infertility, obstructive or non-obstructive, leading to absence of sperm in the ejaculate; in cases of varicocele, and in cases using cryopreserved samples.

AR treatment procedures and their relationship with success rates
In addition to these high complexity techniques, we employ several methods to obtain and process gametes, through complementary technologies and procedures that help increase gestational rates. We use sperm recovery methods, such as PESA and TESA, in conditions when the patient has azoospermia, either OA or NOA, with the goal of increasing fertilization rates in assisted reproduction treatments. These techniques are widely used in post-vasectomy infertility, when there is obstructive azoospermia, followed by ICSI, to achieve better success rates Semião-Francisco et al., 2010;Taitson et al., 2012).
Sperm capacitation techniques, such as Swim-up (SUP) and density gradient centrifugation (DGC), which recovers fractions of high-quality sperm, are recommended to select spermatozoa with greater motility, by eliminating seminal plasma, which contains motility inhibitors, and immobile spermatozoa, along with immature cells and debris. Both sperm capacitating techniques are advantageous to be performed prior to assisted reproduction treatments, since the increase in spermatic motility favors penetration into the oocyte and, consequently, increases the fertilization rates (Borges et al., 2013).
On the other hand, follicular washing is a good alternative to enable a greater number of oocytes in patients with POR, since it improves the chances of overcoming the retention of oocytes in the follicle during direct aspiration or in the collection system, thus increasing the number of oocytes recovered (Souza et al., 2017).
Endometrial injury is an intentional damage to the endometrium by biopsy or curettage, which can induce decidualisation and increase the likelihood of implantation. This is because it induces a significant increase in the secretion of cytokines, interleukins, growth factors, macrophages and dendritic cells, which may be beneficial for the implantation of embryos and may lead to better synchronization between the endometrium and the embryo transfer, with a consequent increase in the rates of live births and clinical pregnancy. Overall, the results suggest a benefit from endometrial scratching. However, the studies have significant limitations; thus, the results may be biased. Therefore, it is not possible to say with any confidence whether endometrial injury can increase the probability of pregnancy. It is also important to consider potential adverse events of this procedure, including excessive pain and bleeding (Garris & Garris, 2003;Li & Hao, 2009;Lensen et al., 2016;Gnainsky et al., 2010;Nastri et al., 2012).

Clinical predictors of success rates in assisted reproductive technologies
The search for predictors of success in highly complex assisted reproductive treatments, such as IVF and ICSI, is essential, since it can determine the efficiency of the techniques and, consequently, the rates of gestational success. Among the predictors of success in the high-complexity assisted reproduction treatments discussed in this review, those who best determined the effectiveness of the techniques, as well as the rates of gestational success were: age, embryo quality and number of oocytes recovered, followed by two promising predictors: the HLA-G allele diversity profile and the SOD1 gene expression.
Maternal age is an important clinical predictor of success rates, since it is correlated to the number of oocytes collected and embryo quality, factors that influence the outcomes of assisted reproduction treatments. As age progresses, natural fecundity and pregnancy rates decline, which occurs due to a significant reduction in the number and quality of oocytes. However, the main factor in the etiology of age-related female infertility is the decline in oocyte quality, since older women have higher rates of single chromatid abnormalities in oocytes, as well as chromosomal aneuploidies (Scheffer et al., 2017).
We analyzed embryonic quality from studying the morphology of embryos and the way they evolved, until the 5 th or 6 th day after IVF procedures. Some of the parameters we assessed were oocyte shape, cellular fragmentation degree, number of blastomeres and their morphology, the presence of vacuoles and the first polar body (1 st PB). Regarding quality, the following factors yield bad prognosis: embryos with irregular cleavages, anuclear fragments, irregular blastomeres, low number of blastomeres, and absence of 1 st PB, which in turn may indicate that the oocyte is still immature or that it has already become post-mature, both unfit for insemination. The indicators of good prognosis include the occurrence of cleavage at the right time (leading to a suitable number of cells for the day of development; for example, the presence of embryos with 4 cells on day 2 and with 8 cells on day 3); harmonic and balanced cell division (leading to blastomere symmetry); absence of embryonic fragmentation; vacuole-free cytoplasm and signs of compaction (Araújo et al., 2008).
However, even embryos classified as being of good quality will not necessarily result in a clinical pregnancy, since endometrial health will be the determining factor. This is because endometrial receptivity disorders represent a potential source of implantation failure, even with good quality embryos. Embryonic implantation is a complex process that requires a synchronization between a healthy embryo and a receptive endometrium. This integrative capacity depends on several endocrine, paracrine and autocrine factors, which are responsible for endometrial receptivity. This is the period during which the epithelium of the endometrium acquires function, even transient, for blastocyst welcoming and implantation. This period is of extreme importance for implantation, since if the embryo reaches the endometrium or is transferred during IVF in a period outside the implantation window (either before or after), implantation will not occur (Martins, 2007;Tan et al., 2018).
The number of oocytes recovered is also an important prognostic variable, since there is a correlation between this number and treatment success rates. In Brazil, the average number of oocytes produced per woman is about 9.3. However, higher chances of success are obtained from IVF cycles with 15 oocytes recovered, with success rates of 65% of live births in women aging 18-34 years; 50% in patients aging 35-37 years; 47% of live births for women with 38 or 39 years old; and 35% of live births for women aged 40 or over. However, even with the average number of oocytes recovered below the considered "ideal", the average fertilization in Brazil reaches 76% (Sunkara et al., 2011;SisEmbrio, 2019).
The human-G leukocyte antigen (HLA-G) encodes a protein class I (Ib), whose expression is intense in trophoblastic cells, and has therefore been recognized to confer immunological tolerance to the fetus. Some HLA-G alleles have been associated with increased or reduced expression of the levels of this protein, which have been associated with reproductive failure. However, it is still difficult to reach a consensus on the role of the different HLA-G alleles during pregnancy (Costa et al., 2016). However, it can be a promising clinical predictor, since identifying regulatory differences between alleles will not only improve our understanding of pregnancy biology, but it can also help unravel the immunogenic factors associated with infertility.
Although poorly explored, the gene expression evaluation in cumulus cells (CCS) of infertile women with endometriosis has been used to investigate the mechanisms involved in infertility associated with this disease. There is a greater expression of the SOD1 gene in infertile patients with moderate/severe endometriosis, when compared with healthy infertile patients or those with mild endometriosis. These results suggest an attempt to prevent oocyte oxidative damage triggered by the disease. Since only infertile women with moderate/severe endometriosis showed increased SOD1 expression in CCS compared to women with minimal/mild endometriosis and controls with positive interaction between increased expression and occurrence of clinical pregnancy, SOD1 may be a potential biomarker of clinical pregnancy, followed by ICSI (Donabela et al., 2015).
The birefringence analysis of the zona pellucida and visualization of the meiotic spindle under polarization microscopy (PM) are also clinical predictors, since they have correlation with implantation and pregnancy rates. These rates were significantly higher when transferring embryos derived from high refringence oocytes (HR), while miscarriage rates were higher when transferring embryos derived from low refringence oocytes (LR). Moreover, the fertilization rate was significantly lower when the meiotic spindle was not visible. However, a limitation of this technique is that while it may potentially increase these fertilization rates, this strategy may reduce the number of high-quality embryos, since the additional handling of the oocyte required to perform the PM is the main reason for this reduction (Picinato et al., 2014).
Despite the findings described here, this review had some methodological limitations, the main one being the scarcity of studies analyzing the association between infertility and assisted reproductive technologies in Brazil. Another possible limitation is that we did not include studies published on non-conventional and low-diffusion access bases, despite our effort using search strategies in a broad and careful way.

CONCLUSION
The qualitative and quantitative analysis of the studies that associated infertility and assisted reproductive technologies in Brazil showed that the main factors of female infertility were endometriosis, tubal factor, polycystic ovary syndrome, endocrine/anovulatory and advanced age, while males had reductions in sperm count and/or quality, infertility after vasectomy, varicocele and anatomical factors. Regarding post-treatment success rates, they were associated with the assisted reproduction technologies chosen for each infertility condition, as well as the use of procedures such as PESA, TESA, follicular lavage and endometrial injury, which may result in better gestational rates when employed during assisted reproduction cycles.
ICSI has been the predominant technique employed and has proven to be the most promising for the treatment of female and male infertility.