IMSI does not improve pregnancy outcomes and live birth rate in severe teratozoospermia patients: a double-blind randomized controlled trial

A total of 166 infertile couples were assessed for eligibility to enter the study. After excluding 66 ineligible infertile couples, 100 who fulfilled the inclusion criteria were recruited. The couples were randomly divided into ICSI and IMSI groups (50 per group). In six couples, there was no embryo to transfer (three from each group), and they were excluded from the study. Also, six couples were not willing to transfer any embryos during the study period and were excluded from the research. Finally, statistical analysis was performed on 43 infertile couples in the ICSI group and 45 in the IMSI group, as presented in the CONSORT diagram (Fig.  1 ). Table  1 presents the demographic and baseline clinical characteristics of the participants. Fig. 1 Flowchart of participant’s allocation, treatment and analysis. Table 1 Comparative evaluation of demographic and clinical characteristics of infertile couples between IMSI and ICSI groups at baseline. Basic characteristics ICSI group (n = 43) IMSI group (n = 45) P - value Female Age (year) 31.65 ± 3.98 32.04 ± 5.11 0.689* BMI 26.03 ± 4.73 26.52 ± 4.87 0.637* Education Illiterate Diploma Academic 26 (60.5) 2 (4.7) 15 (34.9) 27 (60) 4 (8.9) 14 (31.1) 0.714 # Infertility year 7.38 ± 4.05 7.27 ± 4.03 0.903* Infertility type Primary Secondary 32 (74.4) 11 (25.6) 26 (57.8) 19 (42.2) 0.10 # Clinical Characteristics AMH 2.41 ± 0.97 2.21 ± 1.02 0.351* Estradiol 2593.81 ± 1334.79 2874.71 ± 1583.15 0.372* Progesterone 1.34 ± 0.99 1.42 ± 1.005 0.693* Mean number of retrieved COCs 10.00 ± 7.00 11.00 ± 6.50 0.151 @ Mean number of MII oocytes 8.00 ± 7.00 9.00 ± 6.00 0.251 @ Male Age (year) 36.06 ± 7.62 37.20 ± 4.95 0.410* BMI 26.15 ± 3.45 26.49 ± 5.78 0.745* Education Illiterate Diploma Academic 21(48.8) 11(25.6) 11 (25.6) 23 (51.1) 12 (26.7) 10 (22.2) 0.934 # Therapy time (month) 2.41 ± 3.20 1.47 ± 2.14 0.108* Clinical Characteristics Sperm concentration (× 10 6 ) 25.00 ± 20.00 30.00 ± 37.00 0.366 @ Progressive motility (%) 25.00 ± 15.00 28.00 ± 12.00 0.077 @ Non progressive motility (%) 10.00 ± 2.00 10.00 ± 3.00 0.316 @ Immotile (%) 65.00 ± 12.00 62.00 ± 11.00 0.403 @ Male factor subgroups: Teratozoospermia Asthenoteratozoospermia Oligoasthenoteratozoospermia 7 (16.3%) 29 (67.4%) 7 (16.3%) 10 (22.2%) 25 (55.6%) 10 (22.2%) 0.519 # Data were presented as * mean ± SD, statistical analysis was performed based on independent sample t-test. Data were presented as # number (%), statistical analysis was performed based on Chi-Square. Data were presented as @ median ± IQ, statistical analysis was performed based on Mann–Whitney U test. IQ = Percentile 75 – Percentile 25. Flowchart of participant’s allocation, treatment and analysis. Comparative evaluation of demographic and clinical characteristics of infertile couples between IMSI and ICSI groups at baseline. Education Illiterate Diploma Academic 26 (60.5) 2 (4.7) 15 (34.9) 27 (60) 4 (8.9) 14 (31.1) Infertility type Primary Secondary 32 (74.4) 11 (25.6) 26 (57.8) 19 (42.2) Education Illiterate Diploma Academic 21(48.8) 11(25.6) 11 (25.6) 23 (51.1) 12 (26.7) 10 (22.2) Male factor subgroups: Teratozoospermia Asthenoteratozoospermia Oligoasthenoteratozoospermia 7 (16.3%) 29 (67.4%) 7 (16.3%) 10 (22.2%) 25 (55.6%) 10 (22.2%) Data were presented as * mean ± SD, statistical analysis was performed based on independent sample t-test. Data were presented as # number (%), statistical analysis was performed based on Chi-Square. Data were presented as @ median ± IQ, statistical analysis was performed based on Mann–Whitney U test. IQ = Percentile 75 – Percentile 25. There was no significant difference between the ICSI and IMSI groups in terms of the mean number of fertilized oocytes (5.88 ± 3.24 vs 6.31 ± 4.41, P -value = 0.608). In the ICSI group, out of the 452 cumulus-oocyte complex (COCs) collected, 375 were MII and were therefore microinjected. A total of 253 (67.46%) oocytes were normally fertilized and 230 (90.90%) embryos developed. In the IMSI group, a total of 553 oocytes were collected, of which 446 denuded ones were MII and were therefore microinjected. A total of 284 (63.67%) oocytes fertilized normally and 260 (94.01%) embryos developed. There was no significant difference in fertilization rate between the ICSI and IMSI groups ( P  = 0.649) (Fig.  2 ). Fig. 2 Comparative evaluation of fertilization rate between two groups ( P  = 0.649). Comparative evaluation of fertilization rate between two groups ( P  = 0.649). The results of this study did not show any superiority of IMSI over ICSI in pregnancy outcomes, LBR and miscarriage rate (Fig.  3 ). It should be noticed that there was a positive trend toward implantation rate and LBR in maternal age ˃ 35 and SDF ≥ 30. These findings support the hypothesis that oocytes retrieved from maternal age ≤ 38, are able to repair the DNA damage of sperm from semen samples with a high SDF (Table 2 ). Fig. 3 Comparative evaluation of pregnancy outcomes between ICSI and IMSI groups. Implantation rate P -value = 0.171, Chemical pregnancy rate P -value = 0.194, Clinical pregnancy rate P -value = 0.456, Ongoing pregnancy rate P -value = 0.122, Live birth rate P -value = 0.257, Miscarriage rate P -value = 0.332, Induced abortion P -value = 1.000, Intrauterine fetal death (IUFD) P -value = 1.000, Ectopic pregnancy rate P -value = 1.000. Table 2 Comparative evaluation of pregnancy outcomes between high and low sperm DNA fragmentation stratified by maternal age. Age ≤ 35 Age ˃35 SDF < 30 (n = 42) SDF ≥ 30 (n = 23) P -value SDF < 30 (n = 15) SDF ≥ 30 (n = 8) P -value Implantation rate (%) 33.33% 21.7% 0.384 6.7 37.5 0.063 * Chemical pregnancy rate (%) 33.3 21.7 0.326 26.7 50 0.263 Clinical pregnancy rate (%) 28.6 17.4 0.317 13.3 37.5 0.181 Ongoing pregnancy rate (%) 26.2 17.4 0.421 13.3 37.5 0.181 Live birth rate (%) 28.6 17.4 0.612 6.7 37.5 0.063 * Abortion rate (%) 4.8 0 0.288 6.7 12/5 0.636 Statistical analysis was performed based on Chi -square test. Comparative evaluation of pregnancy outcomes between ICSI and IMSI groups. Implantation rate P -value = 0.171, Chemical pregnancy rate P -value = 0.194, Clinical pregnancy rate P -value = 0.456, Ongoing pregnancy rate P -value = 0.122, Live birth rate P -value = 0.257, Miscarriage rate P -value = 0.332, Induced abortion P -value = 1.000, Intrauterine fetal death (IUFD) P -value = 1.000, Ectopic pregnancy rate P -value = 1.000. Comparative evaluation of pregnancy outcomes between high and low sperm DNA fragmentation stratified by maternal age. Statistical analysis was performed based on Chi -square test. Although there was no significant difference between the two groups in SDF at the raw semen, processed, and embryologist-selected spermatozoa stages, repeated-measures ANOVA with a Huynh–Feldt correction showed that the mean SDF differed significantly after the two procedures, density gradient centrifugation (DGC) and single sperm selection by the embryologist, within each group (Table  3 ). However, no relationship was observed between SDF at any of the three stages and pregnancy outcomes in either the ICSI or IMSI group ( P ˃ 0.05, Data are shown in supplementary Tables 1 and 2 ). Table 3 Comparative evaluation of three stages of SDF between and within ICSI and IMSI groups. ICSI group (n = 43) IMSI group (n = 45) P -value Procedure effect Procedure*Group effect SDF of semen 28.97 ± 5.04 28.35 ± 4.69 0.551* F = 202.532 P  = 0.000 # F = 0.523 P  = 0.515 SDF of prepared sample 24.97 ± 4.06 24.44 ± 3.87 0.531 * SDF of single selection 22.53 ± 3.98 21.93 ± 3.76 0.469 * Data were presented as mean ± SD. *Statistical analysis was performed based on independent sample t- test. # Statistical analysis was performed based on Repeated measurement test. Comparative evaluation of three stages of SDF between and within ICSI and IMSI groups. F = 202.532 P  = 0.000 # F = 0.523 P  = 0.515 Data were presented as mean ± SD. *Statistical analysis was performed based on independent sample t- test. # Statistical analysis was performed based on Repeated measurement test. The r-STL of selected spermatozoa after semen processing and single selection by the embryologist did not differ between the ICSI and IMSI groups (Table  4 ). There was significant evidence ( P  = 0.009) supporting an association between the r-STL of spermatozoa selected by the embryologist and chemical pregnancy rate. Pearson’s correlation coefficient was 0.774, indicating a strong positive relationship between r-STL of spermatozoa selected at conventional magnification and chemical pregnancy rate (Supplementary Table 3 ). Table 4 Comparative evaluation of three stages of STL between and within ICSI and IMSI groups. ICSI group (n = 10) IMSI group (n = 13) P -value Procedure effect Procedure*Group effect Median ± IQ Median ± IQ STL of semen 0.095 ± 0.042 0.908 ± 2.127 0.072* F = 2.155 P  = 0.141 # F = 0.507 P  = 0.565 # STL of prepared sample 0.113 ± 0.625 1.059 ± 2.042 0.137 * STL of single selection 0.452 ± 2.143 0.891 ± 2.812 0.577 * * Statistical analysis was done based on Mann–Whitney U test. # Statistical analysis was done based on Repeated measurement test. Comparative evaluation of three stages of STL between and within ICSI and IMSI groups. F = 2.155 P  = 0.141 # F = 0.507 P  = 0.565 # * Statistical analysis was done based on Mann–Whitney U test. # Statistical analysis was done based on Repeated measurement test.

This prospective double-blind randomized clinical trial was conducted on infertile couples who were referred to Yazd infertility Center between September 2022 and October 2024. This trial was approved by the Ethics Committee of Yazd University of Medical Sciences (IR.SSU.MEDICINE.REC.1400.308). All methods were performed in accordance with relevant guidelines and regulations, including the Declaration of Helsinki. After being informed about the research method and objectives, written consent was obtained from all participants. Also, this trial was registered in the Iranian Registry of Clinical Trials (IRCTID: IRCT20211211053355N1). The study population comprised 100 infertile couples with male factor infertility characterized by severe teratozoospermia. The inclusion criteria for the male participants were i: fresh ejaculated spermatozoa, ii: semen sample identified as severe teratozoospermia characterized by morphology ≤ 2 14 and iii: having indication for ICSI due to male infertility. Female partners had to i: aged ≤ 38 years old, ii: have at least four antral follicles in the vaginal sonography, and iii: have day 3 serum FSH < 15 IU/lit. Infertile couples with no embryo to transfer were excluded from the study. Couples with any identified female factors including: endometriosis, PCOS, tubal factor, or ovarian reserve disorders were not entered the study to isolate the effect of male factor on pregnancy outcomes. The expected sample size was 43 infertile couples in each arm, considering an 80% test power, 95% confidence interval, 30% clinical pregnancy in the ICSI group and 60% in the IMSI group and 30% difference range between the two groups. The final sample size was estimated as 50 infertile couples in each arm to account for possible attrition. The sample size of this study was calculated based on a previous study entitled “Can intracytoplasmic Morphologically Selected Spermatozoa Injection be Used as First Choice of Treatment for Severe Male Factor Infertility Patients?” 7 . A computer-generated block randomization list was used to design the study in double-blind format. Permuted block randomization with blocks of eight was used for dividing the participants in this study. A total of 13 blocks were needed in order to select 100 participants with male factor infertility characterized by severe teratozoospermia. The blocks were arranged as follows: ABABAABB. By this method, A and B were dedicated to ICSI and IMSI groups, respectively. During the study, the embryologist who performed sperm selection for the two groups knew the random allocation sequences and also which group each participant belonged to. The participants and embryologists who performed the microinjection, however, were blinded to the study groups. During the study, sperm selection was performed at × 400 (Group A) and × 6300 (Group B) magnification by one embryologist. Sperm selection in the IMSI group was performed according to the Cassuto scoring scale. In this scale, spermatozoa with the normal head shape (2 points), symmetrical base (1 point), and without any vacuoles (3 points) are scored 6. Spermatozoa with a 4–6 score are categorized into class I, and spermatozoa with a score of 1–3 as class II or medium quality. Spermatozoa without any normalcy in the three compartments are scored 0 and categorized in class III 5 . The embryologist who performed sperm selection tried to select spermatozoa in classes I and II, but she had to select spermatozoa in class III in some cases with a morphology index ˂1. Sperm selection in ICSI group was performed according to WHO guidelines (2021), sperm head with regularly contoured, normal base, midpiece and tail . Also, the embryologist considered to select spermatozoa with small acrosomal vacuoles 15 . In specimens with morphology index 0 or 1, sperm selection was limited to the choice between sub-optimal and poor quality spermatozoa cells. For the procedure, three droplets of 7% polyvinylpyrrolidone (PVP) solution (Cooper Surgical/USA, Ref: ART-4005-A) were placed in a plastic tissue culture dish. The number of spermatozoa selected from the first droplet was twice the number of oocytes intended for microinjection, allowing the embryologist to choose the most suitable sperm. At least 25 spermatozoa cells selected from the second droplet were used for DNA integrity assessment by sperm chromatin dispersion (SCD) test, while those from the third droplet were allocated for relative telomere length measurement using qPCR. For microinjection, mature oocytes were placed in 5 μL droplets of HEPES-buffered medium in a plastic tissue culture dish (Falcon/USA, Ref: 351,006) and overlaid with mineral oil. Microinjection was performed by on-call embryologist on the same day. The injected oocytes were washed twice and cultured in droplets of one-step medium, then incubated in an incubator (Origio/Denmark, Ref: 67020010A) for subsequent analysis. Sperm DNA integrity was assessed for all participants using the SCD test using with the Halo sperm kit (SDFA kit, Tehran, Iran) according to the manufacturer’s instructions. The assessment was conducted in three stages: on the raw semen sample, the processed sample, and the spermatozoa selected by the embryologist for each participant. 200 spermatozoa cells were evaluated in raw semen and processed samples, and at least 25 selected spermatozoa cells were evaluated in the ICSI and IMSI groups. For each test, 25 µl of the sample was mixed with prewarmed agarose gel, placed on a precoated glass slide, and covered with a 22 × 22 mm cover-slip. The slide was then kept at 4 °C for 5 min. After removing the cover-slip, the slide was sequentially submerged in solutions A and B for 7 and 15 min, respectively. It was then rinsed with distilled water for 2 min, followed by dehydration through a graded ethanol series. Slide staining was carried out using the kit’s reagents C, D, and E for 4, 3, and 2 min, respectively. The DNA fragmentation index (DFI) was calculated as the percentage of spermatozoa with small halo + no halos relative to the total number of sperm analyzed 16 . To evaluate the DFI of the selected spermatozoa, the embryologist aspirated all of the selected spermatozoa into the injection needle and then transferred them into 4 µL of agarose gel prewarmed to 41 °C. The glass slide was covered with a small piece of cover-slip and kept at 4 °C for 5 min. All the next steps were similar to the aforementioned steps. The raw semen samples, processed samples, and selected spermatozoa were washed with phosphate-buffered saline (PBS) and kept at − 80 °C, throughout the study. The assessment of r-STL was performed in pilot format (ten participants in the ICSI and 13 in IMSI groups). Genomic extraction was performed using the DNJia micro kit (ROJE Technologies, Tehran, Iran) according to the manufacturer’s instructions for the raw semen samples, processed samples, selected spermatozoa, and peripheral leukocyte cells. STL was assessed with quantitative polymerase chain reaction (qPCR) to determine the telomere repeat copy number and the single copy gene copy number. Relative STL (r- STL) was measured by qPCR according to Joglekar’s protocol 17 , with some modification. For this purpose, the 36B4 gene was used as the single copy gene. The forward and reverse primer sequences for the two genes were as follows: Telomere forward: 5′- GGTTTTTGAGGGTGAGGGTGAGGGTGAGGGTGAGGGT- 3′ and reverse: 5′- TCCCGACTATCCCTATCCCTATCCCTATCCCTATCCCTA—3′. 36B4- forward: 5′- CAGCAAGTGGGAAGGTGTAATCC- 3′ and reverse: 5′- CCCATTCTATCATCAACGGGTACAA- 3′ 18 . All samples were run in triplicate, and each PCR was performed using 5 ng of DNA per µL in a final reaction volume of 10 µL. The average cycle threshold (Ct) and standard deviation of each sample, no template control (NTC), and positive controls (normal DNA in blood leukocytes) were included in each PCR series. ∆Ct for each sample was determined as the mean Ct of 36B4 – the mean Ct of telomere. ∆∆Ct was defined as: ∆Ct of each sample—∆Ct control. This value was presented as the relative telomere length (RTL) 18 . The primary outcomes of this study included chemical pregnancy (positive ß-hCG two weeks after embryo transfer), implantation (visible gestational sacs on the ultrasound examination), clinical pregnancy (visible gestational sac on transvaginal ultrasound examination five weeks after embryo transfer), ongoing pregnancy (fetal heartbeat on ultrasound at 12 weeks of gestation), LBR (delivery of a live infant after 28 weeks of gestation), and miscarriage rate (pregnancy loss before 12 weeks of gestation). The secondary outcomes were the comparative evaluation of SDF and STL between the ICSI and IMSI groups at baseline and after each sperm selection. Statistical analysis was performed using the Statistical Package for Social Sciences version 21.0 (SPSS Inc., Chicago, IL, USA). The normality distribution of the data was tested by the Kolmogorov–Smirnov test. Normal numeric data were presented as mean ± standard deviation and tested by the independent sample t-test. Non-parametric data were expressed as median ± interquartile range and tested by the Mann–Whitney U test. The categorical data were presented as number (percentage %) and tested using Chi-square and Fisher’s exact tests. The repeated- measures ANOVA was used to evaluate the mean score of DFI and r-STL within each group. The degree of correlation between DFI and STL and pregnancy outcomes, LBR, and miscarriage rate were calculated using Spearman’s correlation coefficient. P -value ˂0.05 was considered statistically significant. The graphs were designed using GraphPad Prism 8.

This randomized controlled clinical trial did not confirm the superiority of IMSI over ICSI concerning pregnancy outcomes in patients with severe teratozoospermia. However, preselection of spermatozoa at either × 400 or ≥  × 6000 magnification may assist embryologists in selecting spermatozoa with lower DNA fragmentation and higher r-STL in cases of severe teratozoospermia.

This double-blind, randomized, controlled, clinical trial was conducted to evaluate the benefit of the MSOME technique in sperm selection compared to conventional ICSI in patients with severe teratozoospermia. The results did not show any superiority of IMSI compared to ICSI in terms of fertilization rate, pregnancy outcomes, live birth rate, and miscarriage rate. Although sperm selection by the embryologist led to harvesting spermatozoa with lower DFI and higher STL compared to the raw semen samples and the processed samples within each group, we did not find any significant differences between the spermatozoa selected at ≥  × 6000 magnification by MSOME and × 400 magnification with respect to DFI and STL. The results of our study showed a strong positive correlation between the spermatozoa selected at conventional magnification and chemical pregnancy rate. MSOME with a resolution of 0.32 µm at ≥  × 6000 magnification provides embryologists with a three-dimensional view to select better spermatozoa cells with respect to morphological parameters 5 , 19 . Among the clinical studies conducted on this subject, Boedino et al. and Mangoli et al. confirmed that morphokinetic parameters and pregnancy outcomes of embryos derived from sperm selected with MSOME were better than those from sperm selection at conventional magnifications 9 , 20 . Mangoli et al. and Balaban et al. 8 , 9 also suggested that this noninvasive technique can improve clinical outcomes by increasing LBR and decreasing miscarriage rate in patients with male factor infertility. It should be noted that Mangoli et al. reported a trend toward improvement, although the differences in pregnancy outcomes were not statistically significant. Although Mangoli et al. included infertile couples with male factor infertility in their study, patients with severe male factor, such as OAT, were excluded 9 . Azad et al. also found that the sperm of infertile men with OAT had lower concentrations and altered distribution patterns of phospholipase C zeta (PLCζ) compared to fertile men 21 , which is associated with lower fertilization rates in this subgroup of male infertility. This may explain the differences between our results and those of other studies. Boedino et al.’s trial was designed as a clinical trial without randomization. Therefore, close to one-fourth of patients allocated to the ICSI group had semen samples with OAT parameters. In the IMSI group, however, one-tenth of patients were characterized as OAT. Therefore, to control for early paternal effects, future studies should be conducted using a randomization design 20 . The results of our study were in accordance with the results of a prospective, multicentric, randomized, controlled trial designed by Leandri et al. (2013). According to Leandri et al., male factor infertility was characterized by fresh ejaculated spermatozoa with less than one million motile spermatozoa after DGC regardless of sperm morphology. IMSI did not find any significant differences in clinical outcomes compared to ICSI, neither for implantation (24% vs. 23%), nor for the clinical pregnancy rate (31% vs. 33%), nor the live birth rate (27% vs. 30%) 11 . Similar to our study, sperm selection at conventional magnification was performed at × 400 magnification in Leandri’s study. Leandri et al. 11 suggested that the ineffectiveness of IMSI could be attributed to how spermatozoon was chosen carefully and sperm cells with large vacuoles were omitted from the microinjection procedure 11 . In the same direction, Lukaszuk et al. 11 showed that inverted microscopy with an optical resolution of 0.5 µm enables embryologists to detect the morphological features of sperm such as angularity, lateral symmetry and some vacuoles in the sperm head with regularly contoured, normal base, midpiece and tail. A standard ICSI system consists of an optical resolution of 0.41 µm for × 40 objectives. This conventional magnification offers adequate resolution for embryologists to distinguish subtle defects of the sperm head 19 . The Different findings across these studies may be attributed to variation in sperm selection operators as well as the transfer of two best-quality embryos per infertile couple. The results of our study suggest that oocytes retrieved from women aged ≤ 38 years, may play an important role in repairing SDF ≥ 30% in patients with severe teratozoospermia. Therefore, the oocyte’s DNA repair capacity might account for the lack of difference between the two groups in terms of clinical outcomes. Ammar et al. (2020) provided clear evidence of an association between teratozoospermia and sperm DNA defects. Therefore, sperm selection based on WHO guidelines at different magnifications may help embryologists select spermatozoa with intact DNA 22 . In our study, reduction in DFI after DGC and sperm selection using an inverted microscope was compared between the ICSI (13.21% and 9.78%, respectively) and IMSI (13.8% and 10.28%, respectively) groups. Our findings were inconsistent with the result of the study by Mangoli et al. (2020), who reported favorable DFI outcomes for spermatozoa selected at ≥  × 6000 magnification after semen processing 9 . Meanwhile, they did not report baseline DFI values the prior to processing, leaving open the possibility that the DFI difference might have existed at baseline as well. Telomere shortening is considered a contributing factor to impaired oocyte performance in decondensing the telomeric region, which may lead to pronuclear formation failure 23 . Yuan et al. (2023) designed a systematic review and reported STL as a new diagnostic biomarker of male infertility that could affect embryo development 3 . In our study, despite similar fertilization rates between the two groups, it is possible that there were no significant differences in the telomere length of spermatozoa selected at × 400 versus ≥  × 6000 magnification. Furthermore, our results showed that, sperm selection at higher magnification (≥ 6000) did not result in significant inter group differences in r-STL or in pregnancy outcomes, LBR and miscarriage rate compared to × 400 magnification. Nevertheless, within each group, the spermatozoa selected by the embryologist exhibited longer telomere lengths than those in the semen samples and processed samples. Long telomere length of spermatozoa at the time of conception has been associated with improved immune function and enhanced maintenance capacity in offspring 24 . This suggests that appropriate sperm selection by an embryologist may increase embryo resilience to environmental stressors and improve pregnancy outcomes and LBR, particularly in patients with severe teratozoospermia. Semen samples characterized by teratozoospermia showed decreased genomic sperm methylation content 25 . Such methylation variation is associated with poor motility and viability of spermatozoa 26 , 27 . Also, exposure of semen samples to PVP more than 15 min has detrimental effects on sperm viability, morphology, and chromatin structure 28 . Therefore, spermatozoa from these poor- quality semen samples are susceptible to cell death, leading to compromised embryo quality and an increased miscarriage rate. These factors may explain the higher miscarriage rate in the IMSI group compared to the ICSI group. The main strength of this study is that it was designed as a double-blind randomized clinical trial. Sperm selection was performed by one operator; therefore, we tried to decrease the impact of operator-related variability. Also, for better evaluation of the beneficial role of the MSOME technique, changes in DNA integrity and r-STL were traced in semen and two stages of sperm selection. Our findings have limitations. Due to time constraints, comparative evaluation between ICSI and IMSI performance was done after the first embryo transfer, whereas for better evaluation of MSOME performance, there is a need to evaluate cumulative pregnancy outcomes, LBR and miscarriage rate in patients with severe teratozoospermia.

Synchronized structural defects in sperm along with abnormal DNA condensation during spermiogenesis, are commonly observed in men with teratozoospermia, who are at an increased risk of nuclear DNA damage 1 . This damage affects not only coding regions of DNA but also the non-coding telomeric sequences, leading to telomere shortening. Such genomic instability in spermatozoa is associated with poor embryo development 2 , 3 , poor pregnancy outcomes 3 and a high miscarriage rate 2 . The concurrent identification of detailed structural abnormalities in the sperm head and compromised DNA integrity prompted the development of the Motile Sperm Organelle Morphology Examination (MSOME) technique, aimed at selecting spermatozoa with optimal morphology 4 . MSOME with magnification ≥  × 6300 has enabled embryologists to evaluate the fine real morphology of each spermatozoon and select spermatozoa with normal morphology 5 . Compared to conventional Intracytoplasmic sperm injection (ICSI), embryos derived from intracytoplasmic morphologically selected sperm injection (IMSI) have shown a tendency to have significantly better trophectoderm, which is followed by higher implantation rates 6 – 8 . The preselection of spermatozoa at high magnification in oligoasthenoteratozoospermia (OAT) patients has shown remarkably higher pregnancy rates compared to previous ICSI attempts among the same patients 7 . Also, Mangoli et al. (2020) detected improvement in live birth rate (LBR) and miscarriage rate, and recommended using this non-invasive technique to assist couples with male factor infertility 9 . Nevertheless, the results of several studies have failed to confirm the superiority of clinical outcomes in patients for whom sperm selection has been carried out by MSOME compared to ICSI 10 – 12 . Also, the results of a Cochrane review (2020) did not show any certain superiority of sperm selection by MSOME over conventional magnification in terms of LBR and miscarriage rate 13 . These studies were considered low-quality evidence, because they were at risk of selection and publication bias. Therefore, the present study was designed as a double -blind randomized clinical trial to evaluate the effect of high magnification on improvement of pregnancy outcomes, LBR, and miscarriage rate in male factor patients with severe teratozoospermia.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1