Infertility affects approximately 13%-15% of couples. The term unexplained infertility describes cases where no abnormality is detected in sperm analysis, ovulation tests and pathological evaluation of the fallopian tubes [1]. The prevalence of unexplained infertility varies between 22% and 28%. Tests used in the investigation of routine fertility are insufficient in the evaluation of the reproductive system. The etiology of unexplained infertility has primarily been ascribed to immunological, genetic and endocrinological factors [2]. According to recent literature, oxidative stress is associated with conditions affecting the female reproductive system, including unexplained infertility, endometriosis, tubal infertility, Polycystic Ovary Syndrome (PCOS), hydatidiform molar pregnancy and embryopathies [3].

Oxidative stress causes the emergence of reactive oxygen species (ROS), substances that occur naturally within cells during any aerobic process. An increase of these substances in the body has been reported to cause DNA damage, lipid peroxidization and protein damage [4].

Oxygen used in aerobic metabolism plays an important role in the energy requirements of human gametes and in the physiological processes of free radicals in the ovaries. Several studies have shown that ROS found in the follicular fluid environment could affect the folliculogenesis and steroidogenesis cycle. Oocytes obtained from weakly vascularized follicles with reduced development potential have been shown to have low intrafollicular oxygen; lipid peroxidization is elevated in these conditions [5].

Oxidative stress contributes to the development of impaired embryos, halting embryo growth in assistive reproductive treatment that in turn increases apoptosis and DNA damage related to induced cell membrane damage. As a result of apoptosis, fragmented embryos, which are a reason for limited implantation, decrease live birth rates [6, 7].

Increased ROS during ischemia decreases the cobalt binding capacity of albumin; this new chemically changed albumin is known as ischemia-modified albumin (IMA). IMA measurement has recently been proposed as a sensitive marker for the diagnosis of myocardial ischemia and acute coronary syndrome [8]. In addition, IMA is used in the diagnosis of pulmonary thromboembolism, lower limb ischemia, and cerebrovascular ischemic diseases [9]. IMA is currently regarded as a marker of oxidative stress related to ischemia reperfusion in any organ because it is elevated in various clinical entities associated with oxidative stress, such as systemic sclerosis, type-2 diabetes, and PCOS [10-13].

To the best of our knowledge, there are no studies focusing on the relationship of IMA levels and unexplained infertility, and the effect on IVF results. Accordingly, we set out to measure the oxidative stress index (OSI) and IMA levels in both serum and follicular fluid in patients applied with IVF in order to study the effect on embryo development and pregnancy.

Patients included in the study received IVF treatment following a diagnosis of unexplained infertility at the Kahramanmaras Sutcu Imam University IVF Centre between May 2018 and May 2019. Informed consent was obtained from all the study participants. Patients ranged 23-38 years old, showing no uterine or tubal pathology, nor diagnosis of male factor and ovulation disorder. All patients were undergoing treatment with antagonist protocol in the first cycle of IVF after not achieving pregnancy with at least two intrauterine inseminations. OSI and IMA values in the blood and follicular fluid were analyzed. Age, duration of infertility, and gravida were recorded. Follicle stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), and anti-Müllerian hormone (AMH) hormone profiles were recorded at the initial visit. Patients outside the age range, unwilling to participate, or with any chronic disease, history of cancer, endocrine disorder, as well as those from whom eggs had not been taken or transferred, were excluded from the study.

The antagonist protocol (rFSH, Gonal F, Merck Sereno, Modugno, Italy) was applied for follicle stimulation starting on the second or third day of each patient’s menstrual cycle. To prevent early ovulation, 250 mcg Cetrorelix (Cetrotide, Pharm Int., Idron, France) was started when the dominant follicle size reached 13 mm. Data were recorded of the total stimulation days, total FSH dose administered, human chorionic gonadotropin (HCG) day E2 and endometrium thickness. When follicles exceeded 17 mm, 250 mcg/0.5 mL HCG (Ovitrelle, Merck Sereno, Modugno, Italy) was administered, followed by oocyte pick-up (OPU) after 36 hours.

Immediately prior to OPU, blood was measured for IMA level and OSI. Serum was separated from blood samples and stored at -80 ℃ until assay. The follicular fluid collected during OPU was stored in an incubator until application of intracytoplasmic sperm injection (ICSI). Although follicular fluid collected with oocytes is ordinarily destroyed, it was retained and stored at -80 ℃ in order to measure IMA levels. Upon conclusion of the study, samples were treated and the IMA levels were calculated using the rapid, calorimetric method as described in detail by Bar-Or D et al. [14]. A Spekol^® 1300 spectrophotometer was used for the measurements.

Total Oxidant Status (TOS) and Total Antioxidant Status (TAS) levels were measured using commercially available kits (Rel Assay Kit Diagnostics, Turkey), with their ratio accepted as the OSI. In the calculation, the resulting unit of TAS was converted to μmol/L, and the OSI value was calculated according to the following formula: OSI (arbitrary unit) = TOS (μmol H₂O₂ equivalent/L) / TAS (μmol Trolox equivalent/L), as described in detail by Kosecik M et al. [15].

For each patient, a record was made of the following: total number of oocytes collected in OPU; mature oocytes (MII); metaphase I (MI); oocytes with germinal vesicle (GV); degenerated oocytes (DG) as well as MII oocytes applied with the micro-injection process; fertilization rates at 12-16 hours after micro-injection; embryo quality (grades) on the second, third and fifth days; transfer day; pregnancy rates; live births; and abortus. For luteal support following transfer, patients were administered 90 mg vaginal progesterone (Crinone^®; Merck Serono, Istanbul, Turkey) and subcutaneous progesterone 25 mg (Prolutex, IBSA, Lamon, Switzerland) for two weeks.

The data of 80 patients were analyzed. When the correlation between beta HCG positivity and live birth rates was examined with the serum and follicular fluid OSI values, no significant correlation was determined between HCG positivity and serum OSI and live birth rates (p = 0.775, r = 0.033 and r = -0.053, p = 0.639). A statistically significant negative correlation was determined between follicular fluid OSI values and beta HCG positivity and live birth rates (r = -0.407, p < 0.001, and r = -0.312, p = 0.005). Comparable analysis for IMA values likewise found no significant correlation between HCG positivity and serum IMA and live birth rates (r = -0.030, p = 0.790, and r = -0.045, p = 0.695), along with statistically significant negative correlation between follicular fluid IMA values and beta HCG positivity and live birth rates (r = -0.615, p < 0.001, and r = -0.412, p = 0.005).

In the correlation analysis applied, a significant relationship was determined between beta HCG positivity and live birth rates; through ROC analysis of follicular fluid OSI and IMA values, a follicular fluid IMA value < 0.509 predicted live birth with 78.6% sensitivity and 50.8% specificity. When ROC analysis was applied to the relationship between live births and the follicular fluid OSI value, we found that a cutoff value of < 11.15 predicted live births at 61.5% sensitivity and 70.4% specificity. In the ROC analyses of the IMA and OSI values in the follicular fluid, the same cutoff value of 0.509 for IMA in the follicular fluid was found to predict beta HCG positivity with sensitivity of 76.5% and specificity of 45.3%, while a cutoff value of 11.15 for follicular fluid OSI predicted beta HCG positivity with sensitivity of 58.8% and specificity of 76.1%.

Patients were grouped above and below the specified cutoff values for follicular fluid OSI and IMA. Comparative analyses of the two groups found no statistically significant difference between the demographic data of the patients and the follicular-embryonal values. In the follicular fluid OSI ≤ 11.15 group (Group 1, n = 32), beta HCG positivity was seen in 65.6% of patients; live births were recorded in 50%. In the OSI > 11.15 group (Group 2, n = 48), beta HCG positivity was seen in 33.3% of patients; live births were recorded in 20.8%. As for follicular fluid OSI values, the difference in terms of both beta HCG positivity and live birth rates was seen to be statistically significant (p = 0.003, p = 0.007).

In the follicular fluid IMA ≤ 0.509 group (Group 1, n = 24), beta HCG positivity was seen in 66.6% of patients, while live births were recorded in 50%. In the follicular fluid IMA > 0.509 group (Group 2, n = 56), beta HCG positivity was seen in 30.3% of patients; live births were recorded in 25%. In respect of the demographic and embryonal development parameters of the cases, the BMI value was significantly high in the IMA > 0.509 group (p < 0.05). No statistical difference was observed between the other parameters compared (Tables 1 and 2).

To the best of our knowledge, this is the first study to have investigated the relationship between follicular-embryonal development, pregnancy success, and IMA levels in follicular fluid in patients undergoing IVF treatment for unexplained fertility. While the results of the study found no correlation between serum OSI and IMA values and the follicular and embryonal development parameters, a significant negative correlation was determined between the implantation and live birth rates and increased OSI and IMA values in the follicular fluid of the oocyte micro-environment. When patients were grouped according to the specified OSI and IMA cutoff points, a high BMI value was noticeable in the high IMA group.

In the current study, the implantation and live birth rates were found to be lower in the groups with high follicular fluid IMA and OSI values, likely due to increased oxidative stress. Reportedly, increased ROS decrease female fertility by modulating different reproductive functions including ovarian steroidogenesis, oocyte maturation, ovulation, formation of blastocysts [16], sperm capacity and sperm-oocyte interaction [17], fertilization, implantation and early embryo development [18].

Using the thermo-chemiluminescence assay method, Wiener-Megnazi et al. found that the oxidative status in follicular fluid increased oxidative stress in follicular fluid following fertilization in processes such as cleavage and pregnancy. The authors indicate that a certain threshold of oxidative stress in the mature follicle may be necessary for good embryo quality, although high levels may be harmful [19].

Several studies show that oxidative stress could have harmful effects on embryo development through different mechanisms. The authors associated such mechanisms variously with mitochondrial alterations, embryo cell block caused by lipid peroxidation, adenosine triphosphate (ATP) depletion, and apoptosis [20-25].

In contrast, a recent study investigated the relationship between IVF success and culture media ROS levels, reporting no significant relationship between ROS levels in a culture medium and embryo quality, blastocyst formation or embryonal arrest [26].

Studies showing the relationship between IMA, a relatively new oxidative stress marker, and infertility and reproductive functions, have focused on PCOS, insulin resistance and hyperandrogenemia. In the first study on this subject, Guven et al. revealed a relationship between high serum IMA levels in patients with PCOS and testosterone levels, BMI, and Ferriman-Gallwey scores [13]. In support of that study, Çağlar et al. showed that in weak PCOS cases with high IMA levels, increased chronic hypoxia and associated secondary oxidative stress could be responsible for metabolic problems in these patients [27].

In a similar study by Beyazit et al., IMA levels were found to be high in PCOS cases; according to the authors, oxidative stress could play a role in the etiology of PCOS. They also determined a correlation between IMA and testosterone levels based on the negative effects of the follicular micro-environment on follicle and oocyte development [28]. In contrast, a study of PCOS patients by Çakir et al. found that TOS and TAS levels were similar to those of the control group, and while the IMA levels were high in PCOS patients, the elevation did not reach a level of statistical significance [29]. When our patients were examined and compared in two groups according to high or low IMA levels, increased BMI values were noticeable in the high IMA group. This is consistent with a study by Piva et al., in which a relationship was found between high IMA and obesity [30].

The primary limitation of this study is that testosterone and insulin resistance levels were not examined. As most studies in literature have investigated high IMA and increased oxidative stress in PCOS patients, it has been suggested that the mechanism causing reduced fertility could be increased androgens and insulin resistance [13, 28]. Another limitation of the study was that there was no PCOS group. If a PCOS group had been included, a clearer conclusion could have been made as to whether oxidative stress was only a factor in unexplained fertility or also in PCOS.

In conclusion, in cases of unexplained infertility with high BMI, increased IMA but not OSI in follicular fluid during IVF treatment could be related to decreased implantation and live birth rates. Follicular fluid IMA levels below the cutoff value may predict pregnancy success.

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Kahramanmaras Sutcu Imam University (session number: 2019/10, approval number: 15).

Thanks to all the peer reviewers and editors for their opinions and suggestions.

The authors declare no conflict of interest.