Study design and participants
41 couples undergoing ICSI cycles at the Demetra ART Center of Florence (Italy) from November 2016 to August 2018 were enrolled (Table 1 reports the baseline characteristics of the participating couples).
Women were included in the study according to the following criteria.
Age ≤ 43 years
Normal body mass index (BMI) from 18 to 26 kg/m2
Basal Follicle stimulating hormone (FSH) < 10mUI/ml; Antimüllerian hormone (AMH) > 1.5 ng/ml; Antral Follicle Count (AFC) > 6
Tubal and/or male factor or idiopathic infertility
The exclusion criteria were:
Polycystic ovary syndrome (PCOS) according to Rotterdam criteria24
Poor ovarian response according ESHRE 201026
A team of gynecologists and embryologists coordinated all the steps of the study, ensuring that oocyte pick up procedures, culture protocols and embryo assessment were standardized.
Ovarian stimulation protocols
All the patients were treated according to the short standard ovarian stimulation protocol consisting of gonadotropin stimulation from day 2 of the cycle, combined with a flexible antagonist protocol (cetrorelix 0.25 mg/day Cetrotide, MERCK SERONO, Germany or ganirelix 0.25 mg, Orgalutran, MSD, Italy). In 13 cases follicle stimulation was performed with r–FSH monotherapy (Gonal F, MERKSERONO, Germany or Puregon, MSD, Italy). In 28 women with normal ovarian reserve parameters, but previous unexpected suboptimal or hypo-response to ovarian stimulation (Poseidon Group 1–27,8,9), r-LH 75–150 IU (Luveris, MERCK SERONO, Germany) was added from day 1 of stimulation. This subgroup of women belongs to the novel stratification proposed by POSEIDON (Patient-Oriented Strategies Encompassing IndividualizeD Oocyte Number) group7,8,27 and is characterized by low IVF prognosis in terms of cumulative live birth rate compared to the former group treated with r-FSH monotherapy28. As proposed by POSEIDON Group9,29 we opted for FSH and LH co-treatment in this study group. The starting dose of r-FSH ranged from 150 to 225 IU, according to age and response to previous ovarian stimulation. The starting dosage was then modified according to patient response assessed by serum estradiol levels and ultrasound evaluation at 2-day intervals, until at least two follicles reached or exceeded a mean diameter of 17 mm. Finally, oocyte maturation was induced by injection of 5,000 IU of u-hCG (Gonasi, IBSA FARMACEUTICI, Italy) or 250 µg r-hCG (Ovitrelle, MERCK SERONO, Germany). Gonadotropin stimulation and GnRH-antagonist was continued until the day of hCG triggering. Each individual oocyte and the individual corresponding FF were collected separately about 35 h later using a sonographically guided puncture of each follicle, under sedation and local anesthesia. Luteal support was given to all patients, administered as intravaginal micronized progesterone (Progeffik, 200 mg three times daily, EFFIK, Italy), from the day after oocyte pick up until 12 days after embryo transfer, when serum βhCG was measured. In the event of positive βhCG levels, clinical pregnancy was verified by ultrasonographic visualization of the gestational sac about 15 days later and, consequently, classified as pregnant or non-pregnant.
A schematic representation of sample distribution and numbers of mature/immature and fertilized oocytes, embryo development, embryo transfer and clinical pregnancy in the two groups of ovarian stimulation is shown in Fig. 1.
Collection and handling of follicular fluids
Oocyte pick-up was performed under vaginal ultrasound guidance, each follicle was aspirated separately and placed in a different dish.
The ovum aspiration was performed using double lumen needles (COOK, Australia) which were washed with flushing medium to avoid FF contamination between the different aspirated follicles. Each FF collected and the corresponding retrieved oocyte were recorded with a code so that the resulting embryo could be clearly identified with the same code. The volume and the appearance of each FF sample was recorded. Only aspirates of follicles which were uncontaminated with blood and contained an oocyte were included in the analysis. Each collected FF was centrifuged, divided into aliquots of 200 µl, stored in sterile tubes at − 20 °C and individually analyzed for steroid content. After excluding blood contaminated FF and those retrieved from oocyte-lacking follicles, a total of 351 FF and their corresponding oocytes were included in the study. Of these, 35 were subsequently excluded because the oocytes were degenerated (Fig. 1). Each oocyte was individually followed during the entire IVF cycle to track its development after fertilization.
Oocyte maturity, ICSI procedure and outcomes
All collected oocytes were observed under microscopy and classified into two groups according to their maturity degree: mature (oocytes in metaphase II) and immature (oocytes in metaphase I or germinal vesicles). For ICSI, Nikon Eclipse TE2000-S microscope equipped with Narishige IM-9B Microinjector (NIKON, Japan) was used. After 17 ± 1 h from microinjection, oocytes were observed under microscopy for the presence of two pronuclei to assess the occurrence of correct fertilization and, consequently, classified as fertilized and non-fertilized oocytes. Cultures of each embryo were extended to blastocyst stage.
Continuous Single Culture Complete medium (IRVINE SCIENTIFIC, Santa Ana, CA) was used for embryo culture. After 44 ± 1 and 68 ± 1 h, pace of division, degree of fragmentation, size and symmetry of the blastomeres were evaluated by microscopy. Blastocysts were incubated in a MIRI incubator (ESCO MEDICAL GROUP, Denmark) at 37 °C, 6% CO2 and 5% O2 and were scored according to the Gardner criteria30. The degree of blastocyst development is represented by a numerical value between 1 and 6 and the overall quality of the trophectoderm and inner cell mass indicated by a letter grade between A and C. Surplus transferable embryos were cryopreserved. Based on the criteria described above, embryos were divided into three groups: high quality (HQ), intermediate quality (IQ) and low quality (LQ) according to Supplemental Table 1.
After 5 days post oocyte retrieval, a single blastocyst, chosen on the basis of the morphologic criteria described above, was transferred into the uterus of each woman. The remaining blastocysts were cryopreserved. In 14 women for whom the fresh cycle did not result in a clinical pregnancy, a single frozen blastocyst was subsequently transferred.
High performance liquid chromatography coupled with tandem mass spectrometry (HPLC–MS/MS) for evaluation of steroid hormones
Steroid levels were evaluated in 316 FF deriving from mature and immature oocytes (Fig. 1). P,17-OH-P, A, T, E2 and E1 and their respective internal standards (Progesterone-2H3, 17α-Hydroxyprogesterone-13C3, Androstenedione-13C3, Testosterone-2H3, Estradiol-2H3 and Estrone-2H4) were purchased from SIGMA-ALDRICH (USA) and stocked at 1,000 µg/ml in methanol. Water and Methanol were ULC-MS grade by BIOSOLVE BV while Formic Acid, Ammonium Formate and Ammonium Fluoride were provided by SIGMA-ALDRICH (USA).
P, 17-OH-P, A and T were analysed in positive ion mode while E2 and E1 in negative mode in two different runs. The MS conditions and the acquired transitions for each steroid are reported in Supplemental Table 2. Briefly, the internal standards in 100 µl of methanol were added to follicular fluids (5 µl), samples were centrifuged, the supernatant diluted to 1000 µl with water and 10 µl were injected.
Cleanup and concentration was performed in isocratic conditions (Water/Methanol 95/5) on a Phenomenex Luna 5 µm C18 20 × 2 mm column whereas the analytical separation was carried out in gradient mode using a Phenomenex Luna 3 µm C18 50 × 2 mm for the positive Ion mode acquisition (androgens analysis), or the Phenomenex Luna 3 µm PFP(2) 50 × 2 mm for the negative ion mode acquisition (estrogens analysis). Both columns were maintained at 40 °C.
For the analysis in positive ion mode, the mobile phase consisted in water and methanol containing 10 mM of formic acid and 5 mM ammonium acetate buffer. The separations in negative ion mode were performed using water added with 0.2 mM ammonium fluoride and methanol.
Data were analyzed with SPSS (STATISTICAL PACKAGE FOR THE SOCIAL SCIENCES, USA), version 25.0 for Windows. All the continuous variables displayed a non-normal distribution (as evaluated by Kolmogorov–Smirnov test) and therefore were expressed as median (interquartile range-IQR) values. Because the variability of measurements among individual follicles of a patient was comparable to that among the different women (Fig. 2), each FF (and the corresponding oocyte) was analyzed as a single event. Considering that statistically significant differences were observed between the two ovarian stimulation protocols for some outcomes (Table 1), all the statistical analyses were conducted in the two groups (FSH and FSH + LH) separately. For the categorical variables, Chi-square test was used to evaluate differences between the two groups. Inter-subject variability was expressed as coefficient of variation (CV) = (SD/mean of the determinations) × 100. Correlations among steroid hormone levels were assessed using Spearman’s method. Mann–Whitney U test was used for comparisons between the following groups: mature vs. immature oocytes, fertilized vs. non-fertilized oocytes, high vs. intermediate vs. low embryo quality, pregnant vs. non-pregnant. Logistic binary regressions were applied for multivariate analyses to adjust data for male factor31 and female age32, two relevant factors impacting ART success. Female factor was not considered because only women with tubal factor (which is overcome by in vitro fertilization) were included. Male factor is defined, according to the WHO criteria33, as the presence of at least one of these two parameters: sperm concentration ≤ 15 × 106/ml and/or sperm progressive motility ≤ 32%. We used receiver operating characteristic (ROC) curve analysis to test the accuracy (area under the curve, AUC) with 95% confidence interval, the sensitivity and the specificity in predicting oocyte maturity and achievement of fertilization. All statistical tests were 2-sided, and P values of ≤ 0.05 were considered statistically significant.
The experimental protocol has been approved by the internal ethical committee of Demetra ART Center of Florence (Italy). All recruited couples were informed of the aim of the research and signed an informed consent for participating in the prospective study, allowing the collection of follicular fluids (usually discarded after oocyte retrieval during ART procedure) and accepting the transfer of a single blastocyst. The research was conducted in accordance with the Declaration of Helsinki.