Early cumulus cell removal reduces available embryo rate while having no negative effect on live-birth and malformation rate in IVF: a propensity score-matched cohort study

Background Previous studies of the effect of early cumulus cell removal (ECCR) on clinical outcomes remain controversial. Some studies indicated that ECCR combined early rescue ICSI contributed to avoid total fertilization failure, while the other studies demonstrated that ECCR may be detrimental to early embryo development. The aim of this study is to investigate the ecacy and safety of early cumulus cell removal (ECCR) during human IVF. Methods A retrospective analysis was performed between January 2011 and December 2016. The study enrolled 655 couples who underwent IVF treatments with ECCR. After propensity score matching at a 1:2 ratio, 1310 couples who underwent overnight coincubation of gametes were selected. All data were obtained from the Shanghai First Maternity and Infant Hospital IVF patient database. The main outcome measure was the live birth rate and the secondary outcome measures were the normal fertilization rate, polyspermy rate, available embryo rate, clinical pregnancy rate, miscarriage rate and malformation rate. Results No signicant differences were found in the live birth rate (28.55% vs 28.4%; RR of 1.008; 95% CI: 0.869-1.170; p=0.916), clinical pregnancy rate (48.28% vs 45.16%; RR of 1.069; 95% CI: 0.951-1.202; p=0.268), implantation rate (32.67% vs 33%; p=0.896), miscarriage rate (13.33% vs 9.32%; RR of 1.43; 95% CI: 0.916-2.232; p=0.115), neonatal congenital anomalies rate (1.32% vs 1.01%; RR of 1.306; 95% CI: 0.315-5.417; p=0.713) or birthweight between the two groups. The study showed that ECCR was associated with a signicantly lower fertilization rate (73.86% vs 80.12%; p=0.000), normal fertilization rate (2PN)(62.76% vs 69%, p=0.000) and available embryo rate (59.62% vs 62.29%, p=0.001).

Conclusions ECCR tended to confer increased risk of a lower available embryo rate but had no negative effect on the live birth rate or the neonatal malformation rate.

Background
Total fertilization failure (TFF) in in vitro fertilization (IVF) is a desperate event for patients. It is reported that the incidence of TFF after traditional IVF is 4%-16% [1,2]. When TFF happens, medical teams have limited options. The rst option is to cancel the current IVF cycle and to change to intracytoplasmic sperm injection (ICSI) in the subsequent cycle, which may bring a heavy nancial and emotional burden.
The second choice is to perform ICSI in the present cycle. Because of low e cacy [3], late rescue ICSI is not the best option.
Short coincubation of gametes and early removal of cumulus cells combined with early rescue ICSI has been demonstrated to avoid TFF [4] and has been offered to populations at high risk of TFF. It has been indicated that early rescue ICSI not only contributes to avoidance of TFF or lower fertilization [5,6] but also may be bene cial by reducing potential detrimental exposure to free oxygen radicals (ROS) and other metabolic products which come from spermatozoa [7,8]. This protocol was adopted by many IVF centers in China.
However, the e cacy and safety of early removal of cumulus cells remains debatable. The early removal of cumulus cells may be detrimental to embryo development [9] due to breakage of the important crosstalk between oocyte and cumulus cells during early embryos development [10]. In our IVF center, less than 20% patients who underwent early removal of cumulus cells were offered early rescue ICSI (unpublished), and most patients' oocytes were instead continued in culture after removal of the cumulus cells in which two poly bodies were seen, indicating normal fertilizing capabilities of the oocyte and sperm. To avoid TFF, most patients who did not need to perform early rescue ICSI were offered early cumulus cell removal. Whether early cumulus cell removal has negative effects on embryo developmental potential and live birth rate are the main questions we sought to answer in this study, in which we will classify the clinical applicability, effectiveness and safety of early removal of cumulus cells.

Materials And Methods:
Patients This study was reviewed and approved by the Institutional Reviews Boards and Ethics Committee of Shanghai First Maternity and Infant Hospital, China. This study was a retrospective cohort study of patients undergoing IVF-ET treatment between January 2011 and December 2016 at the Reproductive Medicine Center.
The inclusion criteria were as follows: rst IVF treatment, female aged between 20 and 42 years and cleavage-stage embryo transfer. Patients were excluded from the study if they had congenital or acquired uterine malformations, abnormal results on parental karyotyping, uterine broid, adenomyosis, untreated hydrosalpinx and blastocyst-stage embryo transfer. Well controlled diabetes and hypertension were not exclusion criteria. The use of donor semen was allowed.
Couples with unexplained infertility or primary infertility with a longer infertility duration were routinely given short-time fertilization and early cumulus cell removal for a fertilization check (insemination time, 6 h). Unexplained infertility was de ned as no cause for infertility identi ed after a complete infertility evaluation or failure to conceive after intrauterine insemination (IUI) treatment [11,12]. The study comprised 655 patients who received short coincubation and early cumulus cell removal and 1310 patients who undergone traditional IVF with overnight coincubation of gametes after propensity score matching (PSM) based on female age and primary infertility or secondary infertility.

Measures
The primary outcome of interest was the live birth rate per starting cycle. Additional outcomes included the clinical pregnancy rate, normal fertilization rate, polyspermy fertilization rate, implantation rate, miscarriage rate and malformation rate.

Cumulus Cells Removal
In the early cumulus cell removal group, cumulus cells were mechanically removed from cumulus-oocyte complexes (COCs) after 4-6 h of coincubation. When two polar bodies were present, a zygote was considered fertilized. Zygotes with two polar bodies were transferred to another fresh microdroplet without sperm and cultured overnight. In the traditional IVF group, cumulus cells were removed after 18-20 h insemination for a fertilization assessment.

Statistical analysis
We used a PSM that was based on propensity score to construct a weighted cohort of patients who differed with respect to time of fertilization check but were similar with respect to other measured characteristics. A PSM with a 0.05 caliper width was used to match the two groups at a 1:2 ratio.
For univariate analysis of categorical variables, chi-square tests were used. Where the chi-square analysis of a 2×2 table did not meet the assumption that less than 20% of expected values were less than 5, the Fisher's exact test was used. Independent t-tests and rank-sum tests were used for continuous variables. Losses before follow-up were excluded. Statistical analysis was performed using STATA software version 12.0.

Results
A total of 655 cycles were enrolled in the early cumulus cell removal group, and after PSM, 1310 cycles were included in the traditional IVF group. The baseline characteristics of the patients between the two groups are shown in Table 1. There were no signi cant differences in terms of female age, BMI, basal serum level of FSH and proportion of primary infertility and nulliparity between the two groups. The early cumulus cells removal group showed a longer period of infertility compared to the traditional IVF group.
The proportion of unexplained infertility in early cumulus cell removal group was signi cantly higher compared to the traditional IVF group. Similarly, the early cumulus cell removal group showed signi cantly more total doses of gonadotropin and longer durations of stimulation.  Discussion TFF is a frustrating experience for patients that carries heavy nancial and emotional burdens. The occurrence of TFF is still hard to predict, but many IVF centers in China have adopted short coincubation of gametes and early cumulus cell removal combined with early Rescue ICSI (R-ICSI) to avoid TFF.
In our IVF center, we found that only a small portion of patients were offered R-ICSI and that most patients' oocytes were only cleared of cumulus cells after a short time of coincubation of gametes and there was no need to perform R-ICSI (unpublished data). Although brief coincubation of gametes decreases the level of ROS, early removal of cumulus cells may disturb the crosstalk between oocytes and the cumulus-oocyte complex (COC), which plays an important role in early development of embryos [13], which begs the question of whether early removal of cumulus cells affects IVF outcomes. There are controversial results regarding this question. A prospective randomized sibling-oocytes study showed that early cumulus removal after 3 h of insemination had no effect on the normal fertilization rate but was associated with a higher polyspermy rate [14]. Another retrospective study showed that early removal of COC after 6 h of coincubation with gametes has a signi cantly lower polyspermy rate [15]. Further, a prospective cohort study found that removal of COC after 2-4 h insemination had an effect on polyspermy rate [5]. Although there were con icting results regarding polyspermy rate, all of these papers demonstrated that early cumulus cell removal did not affect normal fertilization [5,14,15]. Our study showed that early cumulus cell removal was associated with lower normal fertilization but had no effect on polyspermy. These controversial results may be due to different times of COC removal. At early times after insemination, the oocytes are more vulnerable [9], and the earlier the removal of COC is, the greater susceptibility to damage. Different times of COC removal after insemination maybe pose different effects on fertilization procedures, which needs further study.
Regarding the effect of early cumulus cell removal on the developmental potential of embryos, current data shows similarly controversial results. Some studies have found that early removal of COC was associated with higher available embryos rate [16,17], other studies have shown that early removal of COC did not affect embryonic development [5,14], while Wei D et al. found that early removal of cumulus cells may impact embryonic development [9]. Our study showed available embryos on Day 3 in early removal of COC group was signi cantly lower than traditional IVF group. These con icting results may be due to different times of COC removal after insemination and different patients recruited. Further RCTs are required to classify the effect of early removal of COC on the developmental potential of embryos.
A meta-analysis published in 2013 showed that a brief coincubation was associated with a higher ongoing pregnancy rate and clinical pregnancy rate [18]. Current studies demonstrated that early removal of COC has no signi cant difference in clinical pregnancy rate and live birth rate compared to traditional IVF. Our study also showed that early cumulus cell removal had no negative effect on the clinical pregnancy rate and live birth rate and that early removal of COC did not increase the miscarriage rate.
Similarly, our study did not indicate that early removal of COC elevated the rate of malformations.
Our study has some limitations that should be considered. First, although this study adopted propensity score-match to weigh patients' basal characteristics, our study was a retrospective cohort study. Second, the study was conducted at a single IVF center, so our results may not be applicable to other IVF centers.

Conclusions
In conclusion, our retrospective PSM study showed that early cumulus cell removal results in a lower available embryos rate but has no negative effect on the live birth rate. Further studies are required to con rm our ndings. There is a real need for noninvasive biomarkers to predict TFF in order to decrease the percentage of patients who had early cumulus cell removal performed on their oocytes to avoid TFF. Availability of data and materials The datasets used during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests  Tables   Table 1 Baseline Table 2 Comparison of fertilization and cilinical outcoms between the two group