ICSI improves fertilization in isolated teratozoospermic men: a study with strictly controlled external factors and WHO-5 standard

Abstract

A clear clinical management pathway (conventional in vitro fertilization, IVF or intracytoplasmic sperm injection, ICSI) for treating patients with teratozoospermia is lacking. Here we conducted a retrospective study of fertility indices in 2,178 IVF/ICSI cycles in order to reevaluate clinical management of couples with isolated teratozoospermia (< 4% morphologically normal sperms and normal sperm concentration and motility with the standard of WHO-5).We strictly controlled external factors that could affect oocyte quality or endometrial receptivity to minimize the impact of confounders. Fertilization, total fertilization failure, embryo quality, blastocyst formation rate, and pregnancy rate were studied. Retrospectively, in conventional IVF cycles a significantly lower fertilization rate and higher total fertilization failure rate were observed in couples with isolated teratozoospermia as compared to couples with a normal semen profile. Furthermore, when ICSI was used to treat these teratozoospermic couples, improvement in fertilization was noted. However, the embryo quality, blastocyst formation rate, and pregnancy of couples with isolated teratozoospermia were not enhanced by ICSI. Multiple variable analysis showed that many factors including percentage of morphologically normal sperm are statistically correlated with fertilization rate and total fertilization failure in conventional IVF cycles. In addition the insemination method was correlated with fertilization rate in cases with isolated teratozoospermia. Further studies are warranted to compare outcomes of conventional IVF and ICSI in cases of isolated teratozoospermia, where less than 5 oocytes are retrieved.

Keywords:

• fertilization
• fertilization failure
• intracytoplasmic sperm injection (ICSI)
• in vitro fertilization (IVF)
• isolated teratozoospermia
• sperm morphology
• WHO-5 standard

Introduction

Consensus towards sperm morphology evaluation began with the WHO-1 (WHO Laboratory Manual for the Examination of Human Semen and Semen-Cervical Mucus Interaction) in 1980 [ WHO 1980] with the most recent revision of the WHO-5 in 2010 [WHO 2010]. Generally two approaches are considered. The liberal approach [Comhaire et al.1994], as illustrated by MacLeod and Gold in 1952 [MacLeod et al. 1952] and the strict (Tygerberg) approach, as introduced by Menkveld in 1987 [Menkveld 1987] and described in detail by Menkveld et al. in 1990 [Menkveld et al. 1990]. The WHO-1 and the WHO-2 [WHO 1987] followed the liberal approach, in which no specific criteria were put forward for a morphologically normal spermatozoon. Subsequently the WHO-3 [WHO 1992], WHO-4 [WHO 1999], and WHO-5 [WHO 2010] adopted the strict (Tygerberg) approach in which, the definition for a morphologically normal spermatozoon is precise, such that the ‘borderline’ or ‘slightly abnormal’ head forms are regarded as abnormal.

The diagnostic criteria of teratozoospermia has evolved with each reiteration of the WHO manual. The threshold value of the normal sperm morphology rate has been changed from 80.5% in WHO-1 to 50% in WHO-2, and 30% in WHO-3 [WHO 1980; 1987; 1992]. In the WHO-4, the cut-off value was not defined. Nevertheless, it did state that with conventional insemination, fertilization can be impaired. The risk of fertilization failure rises when less than 5% of the sperms are morphologically normal. In the WHO-5, the lower reference limit for normal forms is 4% (5^th percentile, 95% CI 3.0–4.0) [WHO 2010].

We have observed a drastic decrease in normal sperm morphology cut-off values over the years. This is highlighted by the current WHO-5, cut-off value for sperm morphology of 4%.There are many reasons for the decline. For example using the strict approach the evaluation of sperm morphology was, to some extent, overcritical with regard to normality. Second, because of negative environmental factors, semen parameters including normal sperm morphology have declined. These factors have yet to be considered for their impact on the genetic component [Lewis et al. 2008] although this is beginning to be assessed through surrogate markers [Platts et al. 2007; Anton and Krawetz 2012]. It is widely acknowledged that when the rate of morphologically normal sperms drops to below 5%, the fertilization rate is significantly impaired and the risk of total fertilization failure is increased. The question is, whether conventional IVF or ICSI should be adopted for treating couples with isolated teratozoopermia as defined by the WHO-5 manual that has proven of utility in cases of oligospermia [Hashimoto et al. 2010]. To date, there has been a body of literature supporting the notion that morphology is a significant sperm parameter to predict fertilization success in IVF [Grow et al. 1994; Guzick et al. 2001; van der Merwe et al. 2005]. Several studies have shown that in IVF cycles with teratozoospermia, ICSI improves fertilization rate over that of conventional IVF [Kihaile et al. 2003; Pisarska et al. 1999; Plachot et al. 2002]. However, this issue still remains controversial. On one hand, some institutes recommend ICSI for the treatment of isolated teratozoospermia [Plachot et al. 2002; Tournaye et al. 2002]. On the other hand, there are of series of studies indicating that fertilization rate and/or pregnancy outcome are not improved by ICSI for these couples [Hall et al. 1995; Hotaling et al. 2011; Keegan et al. 2007; Mansour et al. 1995; Robinson et al.1994].

The stark contrasts among these studies probably stem from the heterogeneity of the studied populations of several factors including the duration of infertility, treatment, and comorbidities that could affect oocyte quality, or endometrial receptivity. These factors may carry more weight on the outcome of IVF/ICSI cycles than isolated teratozoospermia itself which confounds the results when poorly controlled. Besides, the inter-observer bias in documenting teratozoospermia and inter-embryologist bias in selecting individual sperm that appear morphologically ‘‘normal” should also be considered.

The aim of this study was to compare conventional IVF with ICSI in males with isolated teratozoospermia as defined by the WHO-5 manual. We conducted a retrospective study of fertility indices in 2,178 IVF/ICSI cycles and tried to minimize the impact of confounders to obtain more reliable results.

Results

The analysis reviewed 2,178 IVF/ICSI cycles. No statistically significant difference was noted in the demographic parameters. Female age, number of retrieved oocytes, number of MII oocytes, and year of infertility were not significantly different between groups (p > 0.05). For group 1 and group 2 (conventional IVF cycles), the MII oocytes were graded on the second day of oocyte retrieval when the cumulus cells were removed for evaluation of fertilization. For group 3 and group 4 (ICSI cycles), the MII oocytes were assessed 2 hours after retrieval when the cumulus cells were removed for performing ICSI. In our study, fertilization rates for both conventional insemination and ICSI cycles were obtained by dividing the number of fertilized oocytes with the number of oocytes retrieved in each cycle. This reflected the challenge to assess oocyte maturity at the time of retrieval due to the large amount of cumulus cells surrounding the nuclei. The fertilization rate of group 2 (52.2 ± 26.6%) was lower than group 1 (58.0 ± 23.6%, p < 0.05). In comparison the fertilization rate of group 4 (61.8 ± 15.8%) was higher than group 2 (p < 0.05) and the difference between the fertilization rates of group 1 and group 3 were not statistically significant (p > 0.05). As shown in Table 1, no statistically significant difference was noted in the other fertility indices (e.g., rates of high quality embryos, blastocyst formation, and pregnancy) between groups. However, the total fertilization failure rate of group 2 (11.1%) was significantly higher than group1 (5.4%, p < 0.05).

Table 1. Decrease in fertilization rates of couples with isolated teratozoospermia were rescued by ICSI.

Group	1	2	3	4
Normal morphology (%)	≥ 4	< 4	≥ 4	< 4
Insemination method	Conventional IVF	Conventional IVF	ICSI	ICSI
Number of cycles	1,971	153	27	27
Fertilization rate (%)	58.0 ± 23.6	52.2 ± 26.6^*	56.8 ± 24.6	61.8 ± 15.8#
High quality embryo rate (%)			21.6 ± 17.9
Blastocyst formation rate (%)	24.0 ± 17.1	20.4 ± 18.0	14.3 ± 10.4	23.3 ± 14.8
Failed fertilization rate (%)		18.0 ± 27.3	0	14.6 ± 26.0
Positive HCG rate (%)	16.3 ± 25.8	11.1(17/153)^*	63.6(14/22)	0
Clinical pregnancy rate (%)			54.5(12/22)	50.0(12/24)
Mean female age in years	5.4(106/1971)	59.4(76/128)	29.5 ± 3.3	41.7(10/24)
Mean eggs retrieved per cycle	58.6(993/1694)	55.5(71/128)	14.7 ± 7.2	29.4 ± 2.8
Mean MIIoocytes per cycle	53.8(911/1694)	29.6 ± 3.3	11.3 ± 5.2	14.2 ± 6.1
	29.6 ± 3.4	16.1 ± 8.6		11.2 ± 5.4
	16.4 ± 8.0	13.2 ± 7.5
	13.7 ± 6.8
Mean infertility years	4.7 ± 3.1	4.6 ± 3.2	5.4 ± 3.0	5.2 ± 2.9

*p < 0.05 (compare the fertilization rate of group 2 to one of the group 1)

^#p < 0.05 (compare the fertilization rate of group 4 to one of the group 2)

*p < 0.05 (compare the fertilization failure rate of group 2 to one of the group 1)

To determine what factors affect fertilization rate and total fertilization failure, multi factor analysis in conventional IVF cycles (group 1 and group 2) were considered. Fertilization rate and total fertilization failure were considered respective dependent variables. Independent female variables included, age, years of infertility, oviduct obstruction, and the levels of FSH, LH, T, E2, and PRL on the third day of menstruation. Independent male variables included, primary or secondary infertility, percentage of morphologically normal sperm, semen volume, sperm concentration, and progressive motility. The specific values of female age, years of infertility, levels of FSH, LH, T, E2, and PRL on the third day of menstruation, percentage of morphologically normal sperm, semen volume, sperm concentration, and progressive motility were used in the model. Primary infertility was assigned the rank of 0, secondary infertility 1, without oviduct obstruction 0, with oviduct obstruction 1, total fertilization failure 0, and fertilization 1. The factors that showed a significant positive correlation with total fertilization failure were the percentage of morphologically normal sperm, female primary or secondary infertility, with or without oviduct obstruction, male primary or secondary infertility, and sperm concentration. There were significant positive correlations between fertilization rate with male primary or secondary infertility, with or without oviduct obstruction, percentage of morphologically normal sperm, sperm concentration, and progressive motility (Table 2). As summarized in Table 3, the years of female infertility showed a significant negative correlation with total fertilization failure.

Table 2. Factors that affect fertilization rate in conventional IVF cycles.

Independent factors	Fertilization rate
	r	p value
Female age(years)	−0.011	0.626
Infertility years	−0.019	0.400
Female primary or secondary infertility	0.031	0.158
Male primary or secondary infertility*	0.129	0.000
With or without oviduct obstruction*	0.089	0.000
Morphologically normal sperm(%)*	0.057	0.010
Semen volume(ml)	0.004	0.841
Sperm concentration(×10^6/ml)*	0.046	0.038
Progressive motility(%)*	0.074	0.001
D3 FSH(mIU/mL)	0.013	0.545
D3 LH(mIU/mL)	−0.002	0.937
D3 PRL(ng/ml)	0.014	0.541
D3 E2(pg/mL)	−0.014	0.514
D3 T(ng/dL)	−0.034	0.119

*p 0.05 The D3 FSH, LH, PRL, E2 and T mean the value of FSH, LH, PRL, E2 and T on the third day of menstruation.

Table 3. Factors correlated with fertilization failure in conventional IVF cycles.

	Fertilization failure
			95% CI for EXP(B)
Independent factors	p value	OR	Lower	Upper
Infertility years	0.003	0.915	0.863	0.971
Female primary or secondary infertility	0.027	1.971	1.081	3.595
Male primary or secondary infertility	0.003	2.693	1.402	5.174
With or without oviduct obstruction	0.002	2.303	1.372	3.865
Morphologically normal sperm (%)	0.000	1.073	1.035	1.112
Sperm concentration (×10^6/ml)	0.000	1.028	1.015	1.041

The D3 FSH, LH, PRL, E2 and T mean the value of FSH, LH, PRL, E2 and T on the third day of menstruation.

As summarized in Table 1, using conventional IVF cycles the fertilization rate for patients with isolated teratozoospermia was lower than for patients with normal semen profiles. Furthermore, ICSI was shown to improve fertilization for isolated teratozoospermia. To determine what factors may play a role, a multivariate analysis in cases with isolated teratozoospermia (group 2 & group 4) was undertaken. The fertilization rate was considered the dependent variable. The independent female variables included, the insemination method, age, years of infertility, oviduct obstruction, and the levels of FSH, LH, T, E2, and PRL on the third day of menstruation. Independent male variables included, primary or secondary infertility, percentage of morphologically normal sperm, semen volume, sperm concentration, and progressive motility. The specific values of female age, years of infertility, levels of FSH, LH, T, E2, and PRL on the third day of menstruation percentage of morphologically normal sperm, semen volume, sperm concentration, and progressive motility were used in the model. Use of conventional IVF insemination was assigned a rank of 1, ICSI insemination 2, primary infertility 0, secondary infertility 1, without oviduct obstruction 0, oviduct obstruction 1, total fertilization failure 0, and fertilization 1. As summarized in Table 4, significant positive correlations between fertilization rate with insemination method, progressive motility, with or without oviduct obstruction in cases with isolated teratozoospermia, were noted.

Table 4. Factors that affect fertilization rate in cases with isolated teratozoospermia.

	Fertilization rate
Independent factors	r	p value
Insemination method*	0.328	0.000
Female age(years)	−0.033	0.669
Infertility years	−0.091	0.236
Female primary or secondary infertility	0.105	0.172
Male primary or secondary infertility	0.000	0.994
With or without oviduct obstruction*	0.177	0.020
Semen volume(ml)	0.048	0.533
Sperm concentration(×10^6/ml)	0.087	0.258
Progressive motility(%)*	0.233	0.002
D3 FSH(mIU/mL)	0.061	0.424
D3 LH(mIU/mL)	0.036	0.635
D3 PRL(ng/ml)	−0.038	0.623
D3 E2(pg/mL)	−0.033	0.669
D3 T(ng/dL)	−0.002	0.976

*p 0.05 The D3 FSH, LH, PRL, E2 and T mean the value of FSH, LH, PRL, E2 and T on the third day of menstruation.

Discussion

In this retrospective study, strict inclusion criteria were adopted to minimize the impact of confounders on the outcomes. Confounding effects of low motile sperm were prevented by including only those couples with normal semen parameters. Female causes of infertility like senior age and comorbidities that impair oocyte quality as well as endometrium receptivity were employed as one of the exclusion criteria. Subjects were also excluded based on poor response to ovarian stimulation, which was diagnosed with a harvest of less than five metaphase II oocytes. Only chromosomally normal couples were included to preclude confounding effects of obvious genetic factors. Also, only those cases that were undergoing their first cycle of IVF/ICSI were recruited to improve homogeneity of the subjects. Fertilization rates for both conventional insemination and ICSI cycles were obtained by dividing the number of fertilized oocytes with the number of oocytes retrieved in each cycle. This decision was based on the difficulty to estimate oocyte maturity at the time of retrieval due to the large amount of cumulus cells surrounding the nuclei, thus obscuring maturity status.

Morphology has been recognized as one of the most important predictors of fertilization in IVF [Grow et al. 1994; Guzick et al. 2001; van der Merwe et al. 2005]. As summarized above, the fertilization rate of group 2 (52.2 ± 26.6%) was lower than group 1(58.0 ± 23.6%, p < 0.05). Group 4 (61.8 ± 15.8%) exhibited a higher fertilization rate than group 2 (p < 0.05). No statistically significant difference was noted in the other fertility indices (e.g., rates of high quality embryos, blastocyst formation and pregnancy) between groups (Table 1). The total fertilization failure rate shown in Table 1 of group 2 (11.1%) was significantly higher than group 1 (5.4%, p < 0.05). Total fertilization failure was not observed in groups 3 and 4. Multiple variable analysis revealed that many factors including percentage of morphologically normal sperm were statistically correlated with fertilization rate and total fertilization failure in conventional IVF cycles. Several factors including the insemination method were correlated with fertilization rate in cases with isolated teratozoospermia (Tables 2 – 4). As expected the fertilization rates in IVF for cases with isolated teratozoospermia are impaired in comparison with those where semen profiles are normal. However, ICSI was shown to improve the rate of fertilization for isolated teratozoospermia. The most probable explanation lies in that all the recruited cases had normal ovarian reserve, which somehow safeguards the number of eggs retrieved and hence number of high quality embryos regardless of the method of insemination applied. Therefore, whether ICSI improves fertilization and thus optimizes pregnancy outcome still warrants further research in the future. Hence, one may argue that ICSI is advantageous over conventional IVF in terms of fertilization. Moreover, this advantage may not only be applied to cases with teratozoospermia, but also be applicable when semen morphology indices are normal. One may speculate that this difference may also exist in males with normal sperm morphology. However, as presented above the fertilization rates of group 1 and group 3 were comparable, indicating that ICSI does not improve fertilization when the semen profile is normal.

Others have compared fertilization rates of IVF or ICSI in cases with isolated severe teratozoospermia using oocytes from siblings [Pisarska et al. 1999] to resolve the confounding effects from heterogeneity of female subjects. In contrast to the above a statistically significant difference in fertilization rate, total fertilization failure, pregnancy outcome, or live birth rates among patients with isolated teratozoospermia (morphologically normal sperms < 5% of total sperm count) and normal morphological semen profile in the first or second IVF cycle, and ICSI did not seem to improve the outcome in cases of isolated teratozoospermia [Keegan et al. 2007]. However, the confounding factors such as duration of infertility, history of previous treatments, and comorbidities that may affect quality of oocytes or endometrium receptivity were not well-controlled. As we have shown these factors can influence the outcome of IVF/ICSI cycles and are likely significant in teratozoospermia.

ICSI may offer several advantages over IVF in cases with isolated teratozoospermia. On one hand, morphologically ‘normal’ sperms can be selected for insemination by the embryologists, even when the number of such ‘normal’ sperms are few. In this case, fertilization is no longer random, and the outcome may not genuinely represent the original status of the semen sample. On the other hand, in conventional IVF, when the proportion of morphologically competent sperms is too low (< 4%), the probability that oocytes can be fertilized by normal sperms drops sharply. As shown above the fertilization rates comparable to that of IVF for couples with a morphologically normal semen profile can be achieved when ICSI is used to rescue cases with isolated teratozoospermia.

There are still several limitations in the study presented. First, it is retrospective in nature, predisposed to selection bias. Personal preference towards IVF or ICSI of the individual embryologist and physician is also a source of bias. However, the large sample size and the participation of eight physicians and six embryologists should minimize the effect. Second, although the total sample size was large, the number of subjects in groups 2, 3, and 4 were relatively small. Third, complete data on miscarriage and live birth could not be included since many patients included in the study had not delivered. Therefore, studies with larger sample size and longer duration of follow-up are indispensible to accurately compare the impact of IVF and ICSI on clinical outcomes of couples with isolated teratozoospermia. Finally, we included only patients from whom ≥ 5 eggs were retrieved. Comparisons of IVF and ICSI outcomes should also be carried out on the patients from whom less than 5 oocytes are retrieved, so that fertilization and clinical outcomes of IVF and ICSI can be compared in this subgroup.

In conclusion, the present study shows that isolated teratozoospermia can significantly compromise total fertilization rate and has a higher risk of fertilization failure in conventional IVF cycles that can be rescued to some extent with ICSI. However, in these cases, neither embryo quality, blastocyst development, or pregnancy rates were improved with ICSI. Further studies are warranted to compare outcomes of conventional IVF and ICSI in cases of isolated teratozoospermia, where less than 5 oocytes are retrieved.

Materials and Methods

Subjects

A retrospective study was performed in infertile couples undergoing IVF/ICSI treatments at Women and Children Hospital of Jiangxi Province, China from September 2010 – March 2012. Patients were enrolled when they meet the following criteria: 1) sperm concentration ≥ 15 × 10⁶ spermatozoa per ml and total motility (progressive motility + non-progressive motility), ≥ 40% or progressive motility, ≥ 32% with or without less than 4% normal morphologic forms; 2) female patient's age was no more than 35 years; 3) couples were undergoing their first cycle of IVF/ICSI; 4) Besides simple oviduct obstruction in some cases, there were no other causes for infertility attributable to the female patients; 5) at least five metaphase II eggs were retrieved; and 6) no chromosomal abnormalities were discovered in either one of the couples.

The study involving the use of human tissue specimens was approved by the Reproductive Ethics Committees of the Women and Children Hospital of Jiangxi Province, China and written informed consent was obtained from each of the patients for the patient samples used in this study.

Included cycles were divided into 4 groups: group 1 (n = 1,971): normal sperm morphology (≥ 4% morphologically normal sperm at semen analysis with normal sperm concentration and motility) with conventional insemination; group 2 (n = 153): isolated teratozoospermia (< 4% morphogogically normal sperms at semen analysis with normal sperm concentration and motility) with conventional insemination; group 3 (n = 27): normal sperm morphology (≥ 4% morphologically normal sperm at semen analysis with normal sperm concentration and motility) with ICSI; and group 4 (n = 27): isolated teratozoospermia (< 4% morphogogically normal sperms at semen analysis with normal sperm concentration and motility) with ICSI.

Sperm/Semen Analysis and Manipulation

Semen samples were collected by masturbation for semen analysis or in vitro insemination after an abstinence of 3–5 d. In both occasions semen volume, concentration, and motility were determined according to WHO-5 [WHO 2010] criteria. Semen samples were analyzed after a standing of at least 30 min so that they can completely liquefy. The ejaculate volume was determined by pipetting, and sperm concentration and motility were determined by placing a 5-uL drop of the sample on a Makler chamber (MidAtlantic Diagnostics, Inc., Mount Laurel, NJ, USA) and observed under a light microscope. In this standardized procedure, at least 200 sperm were counted on two separate chambers and average values were recorded. Semen for IVF were processed with isolate gradient (Irvine Scientific, Santa Ana, CA, USA) and swim-up protocols to enrich motile sperm, and the concentration before as well as after processing were both recorded. Each Makler chamber was cleaned and examined daily for physical damage. Control sperm specimen in low (< 15 million/mL) and normal (> 15 million/mL) concentrations were used to test and proofread the chamber daily before clinical use.

Sperm morphology was assessed in the initial semen analysis, and was repeated on the day of ovum retrieval using the Kruger/Tygerberg Strict Criteria as outlined by the WHO-5 [WHO 2010] manual. Briefly, 5–10 uL of semen (depending on the sperm concentration) was placed on a previously cleaned slide and stained using the Papanicolaou staining protocol. Two technicians, who have attained the national docimaster qualification and have been trained for documenting teratozoospermia with the standard of WHO-5, analyzed at least 200 sperm cells independently under an oil immersion microscope with 1,000 X magnification. Mean value of their reports were recorded. Quality control for morphology assessment included a weekly calculation of inter-observer coefficient of variation as obtained by their concurrent evaluation of the same discarded semen sample. An inter-observer variation of < 10% was considered to be acceptable. The dyes were checked daily for cross-contamination and were changed weekly.

Ovulation induction, oocyte retrieval, embryo culture, and transfer

Patients received purified and/or recombinant gonadotropin preparations in controlled ovarian stimulation protocols. GnRH agonists and/or antagonists were used to suppress endogenous gonadotropin secretion. Patients who had three or more follicles that were at least 17 mm in diameter received 10,000 IU of hCG injections as ovulation triggers. In patients with exceedingly high serum oestradiol levels and multiple mature follicles, dosage of hCG was reduced to 5,000 IU to reduce risks of ovarian hyperstimulation syndrome. Ultrasound-guided transvaginal follicular aspiration was performed 36 h after hCG injection. In the IVF cycles, the oocytes were inseminated with a sperm concentration of 2 × 10⁵/ml 4–5 h after oocyte aspiration. In ICSI cycles, cumulus cells and the corona radiate of the oocytes were removed by brief exposure (10 s) to Gamete-20 containing 40 IU/ml hyaluronidase 2 h after retrieval, and ICSI was performed on metaphase II (MII) oocytes with grossly normal spermatozoa as observed under an inverted microscope (Hoffman modulation, 400 x). Fertilization of the oocytes was assessed at 16 − 18 h after insemination. Fertilization was judged upon the observation of two clearly distinct pronuclei. Embryo cleavage and quality were further assessed 24 and 48 h later. Embryos with acceptable developmental potential and quality were transferred on the third day of in vitro culture. Pregnancy was first confirmed by serum human chorionic gonadotrophin assay 13 d after embryo transfer. Clinical pregnancy was defined as the presence of intrauterine gestational sac(s) with fetal heartbeat 28–35 d post retrieval.

Outcome parameters

Fertilization, embryo quality, blastocyst formation rate, and pregnancy outcomes were compared after conventional in vitro fertilization versus ICSI among the 4 groups. Fertilization rate was defined as the rate of zygotes with two pronuclei and two polar bodies 16–18 h after insemination among all the retrieved oocytes. A total fertilization failure was defined as complete absence of fertilized oocytes. A clinical pregnancy was assessed by the presence of a gestational sac with fetal heartbeat on ultrasound scans approximately 5 w after the embryo transfer procedure.

Statistical analysis

The rates of total fertilization failure, positive HCG, and clinical pregnancy were compared between groups using chi-square analyses. Mean fertilization rate, high quality embryo rate, blastocyst formation rate, and mean number of eggs retrieved per cycle, mean female age, and duration of infertility were compared between groups using Student's t test. Multiple factors analysis was used to determine if any correlation existed between morphologically normal sperm rate and insemination method with fertilization. P values of < .05 were considered to be statistically significant.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Author contributions: Designed and wrote this manuscript: YZ; Collected all the data and statistically analyzed the data: X-J Z, YZ; Studied the clinical part of the research: G-xL, GL, Q-fW, YZ; In charge of the study of the embryos and the research of the andrology part of the experiment: ClX, Z-hH; Provided administrative support for the study: Q-fW, C-lX; Revised the paper: GL; Approved the final manuscript: G-xL.

Abbreviations:

ICSI:	=	intracytoplasmic sperm injection
IVF:	=	in vitro fertilization
WHO manual:	=	WHO Laboratory Manual for Examination and Processing of Human Semen