Clinical outcomes of MII oocytes with refractile bodies in patients undergoing ICSI and single frozen embryo transfer

Abstract Purpose This study aimed to analyze whether the presence of refractile bodies (RFs) negatively affects fertilization, embryo development, and/or implantation rates following intracytoplasmic sperm injection (ICSI). Methods This retrospective embryo cohort study involved a total of 272 patients undergoing ICSI treatment of blastocyst cryopreservation. Results In the study, no significant differences were found regarding 2PN formation rates between RF(+) (76.5%) and RF(−) oocytes (77.2%). However, the blastocyst formation rate on Day 5 in RF(+) oocytes was 45.8%, which was significantly lower than that of 52.2% in RF(−) oocytes (aOR 0.74, 95% CI 0.59‐0.93, P = .011). Implantation rates were also significantly lower in RF(+) oocytes (24.2%) as compared to RF(−) oocytes (42.2%) (aOR 0.46, 95% CI 0.26‐0.78, P = .005). Furthermore, the implantation rate of RF(+) oocytes (28.6%), when high‐quality blastocysts were transferred, was significantly lower than that of RF(−) oocytes (46.1%) (aOR 0.50, 95% CI 0.25‐0.96, P = .043). Conclusion Our results suggest that oocytes with the presence of RFs have a lower potential for blastocyst development. Even when they develop into high‐quality blastocysts, the chances of implantation are reduced.


| MATERIAL S AND ME THODS
This retrospective study included 316 RF(+) cycles, which had both RF(+) and RF(-) oocytes, involving 272 patients who underwent ICSI treatment of blastocyst cryopreservation between January 2013 and June 2016. A total of 1190 blastocysts were cryopreserved, and of these, 438 were used for single embryo transfer by December 2016.

| Stimulation protocols
The ovarian stimulation protocols were chosen depending on each patient's age and serum anti-Müllerian hormone (AMH) level (Table 1) Ferring, Ferring Pharmaceuticals) or rFSH (Gonalef, Merck Serono), and continued until the day of hCG administration. The starting dose for the clomiphene citrate (CC) protocol was 50 mg/day of (CC) (Clomid, Fuji Pharma) administered orally once a day, on the third day of the menstrual cycle, and continuing until the day of hCG administration. hMG was administered until the day of hCG administration. The starting dose for the aromatase inhibitor (AI) protocol was 5 mg/day of AI (Letrozole "TEVA," Pharmaceutical Industries Ltd), administered orally on the third day of the menstrual cycle and continuing for 5 days. hMG was administered until the day of hCG administration. When at least two follicles reached a diameter of 18 mm or larger, 10 000 IU hCG was administered for ovulation induction.

| Oocyte retrieval and denuding
Oocyte retrieval was performed 35.5 hours after hCG adminis- The diameter of the (RFs) was measured by means of a scale bar displayed on a monitor connected to an OLYMPUS inverted microscope, at ×400 magnification. When the size of an RF was greater than the length of the scale bar, the RF was defined as RF(+).
Sperm were prepared using a density-gradient centrifugation technique with Isolate (Irvine, Cal., USA) and the swim-up method with Universal IVF. Fertilization was confirmed by the presence of two pronuclei 16-18 hours after ICSI (Day 1). The zygotes were placed into the well-of-the-well (WOW) culture system (LinKID ® micro25, DNP) using a single-step medium, and covered with mineral oil.
Cleaving embryos were evaluated on the third day after oocyte retrieval using Veeck's classification. 5 Blastocyst development was evaluated on the fifth and sixth days after oocyte retrieval by means of Gardner's classification. 6 High-quality blastocysts were defined as having a grade of at least 3BB on Day 5.

| Statistical analyses
The differences in fertilization rates, blastocyst formation rates, and implantation rates were compared between RF(+) and RF(−) oocytes. The difference in fertilization rates was compared by the chi-squared test. The difference in blastocyst formation rates, implantation rates, live birth rates, and pregnancy loss rates was investigated using a mixed-effects logistic regression model (generalized linear mixed model, GLMM). GLMM provides a broad range of models for the analysis of grouped data, allowing the differences between groups to be modeled as random effects. Furthermore, as embryo transfer is repeated more than once for some patients, GLMM makes it possible to eliminate individual differences. The difference in blastocyst formation rates was investigated with a GLMM containing terms for RF(±) and patients' age as fixed effects, and patients as a random effect, taking inter-subject correlation into consideration. Implantation, live birth, and pregnancy loss rates were analyzed with GLMM, containing terms for RF(±), age, and the presence of high-quality blastocyst development (0/1) as fixed effects, and patients as a random effect, taking inter-subject correlation into consideration. In all analyses, two-sided statistical tests were performed, and differences were considered statistically significant when the P-value was <.05. R software (version 3.4.4) was used for all statistical analyses.

| RE SULTS
The average age of participants was 37.8 years. A total of 3085 MⅡ oocytes were retrieved. Of these, 648 (21.0%) were RF(+) oocytes.
There were 241 delayed MⅡ oocytes (7.8%), and the rate of RF(+) found among these oocytes was 16.6%. There was no significant difference between the occurrence of RFs in MⅡ oocytes and occurrence in delayed MⅡ oocytes ( Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; COR, crude odds ratio. a Blastocyst formation rates and high-quality blastocyst formation rates were analyzed using a mixed-effects logistic regression model that contains terms for patients' age and the presence of RFs (0/1) as fixed effects and patient-specific intercept as a random effect. Note: Implantation rates, live birth rates, and pregnancy loss rates were analyzed using a mixed-effects logistic regression model that contains terms for the presence of RFs (0/1), age, and high-grade blastocyst development (0/1) as fixed effects and patient-specific intercept as a random effect. Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; COR, crude odds ratio.

RF(+) oocytes
a The neonatal abnormalities found in the newborn babies were pulmonary valve stenosis, small for gestational age, and cleft palate.
b One stillbirth, occurring at 24 wk and 6 d, has been included in the pregnancy loss rate.
The data were analyzed by GLMM to clarify whether the stimulation protocol affected the outcomes and/or presence of RFs. No significant differences were found regarding 2PN formation rates between RF(+) (76.5%) and RF(−) (77.2%) oocytes. 1PN formation rates were 2.2% in RF(+) oocytes vs 2.4% in RF(−) oocytes, and no statistical differences were detected. In addition, multiple pronuclei formation (>3PN) rates were 2.2% in RF(+) oocytes vs 1.9% in RF(−) oocytes, with no significant differences being detected (Table 3).
However, the blastocyst formation rate on Day 5 in RF(+) oocytes was 45.8%, which was significantly lower than that of 52.2% in RF(−) oocytes (aOR 0.74, 95% CI 0.59-0.93, P = .011; Table 3). The rate of blastocysts that reached blastocysts on Day 6 for RF(+) was 10.5%, while it was 9.6% for RF(−) oocytes. No significant differences were found between the two groups (aOR 1.11; 95% CI 0.79-1.56, P = .536; Table 3).  Table 4). Neither were any significant differences detected in the birth rates (19.2% in RF(+), 29.2% in RF(−) oocytes, aOR 0.60, 95% CI 0.31-1.08, P = .101; Table 4). As this study did not extend to oocytes with single or multiple RFs, it is still unclear whether the size and number of RFs influence embryo development and/or implantation rates. The occurrence mechanisms of RFs and their relationship to oocyte maturation and viability are also not yet fully understood. Further research is required to gain a more complete understanding of these factors and thus improve implantation rates.