Comparative effectiveness of physiological intracytoplasmic sperm injection and intracytoplasmic sperm injection in couples affected by abnormal sperm morphology
Article information
Abstract
Objective
This study compared outcomes after physiological intracytoplasmic sperm injection (PICSI) versus conventional intracytoplasmic sperm injection (ICSI) in couples affected by abnormal sperm morphology.
Methods
This retrospective cohort study included 246 couples in whom the male partner had abnormal sperm morphology, defined as normal morphology <4% according to World Health Organization 2021 criteria, and who underwent assisted reproductive treatment at the in vitro fertilization and Tissue Engineering Center, Hanoi Medical University Hospital, from January 2022 to January 2024. Couples underwent either PICSI or conventional ICSI according to routine clinical practice. To reduce confounding, 1:1 matching based on female age, anti-Müllerian hormone level, and antral follicle count was performed. Embryological and clinical outcomes were compared between groups. Clinical outcomes were assessed only in first frozen embryo transfer cycles using day-5 blastocysts.
Results
No significant differences were observed between the PICSI and ICSI groups in embryological outcomes, including the number of metaphase II oocytes, fertilization rate, day-5 blastocyst formation rate, and high-quality blastocyst rate. Clinical outcomes were also comparable, including the rates of positive serum beta-human chorionic gonadotropin, clinical pregnancy, implantation, ongoing pregnancy, pregnancy with multiple gestational sacs, and live birth. Subgroup analyses stratified by female and male age likewise showed no significant differences in outcomes.
Conclusion
In this real-world Vietnamese cohort of couples affected by abnormal sperm morphology, PICSI demonstrated no significant advantage over conventional ICSI. These findings support individualized selection of sperm injection techniques rather than routine PICSI use. Future prospective studies incorporating sperm DNA fragmentation testing and sibling-oocyte designs are warranted.
Introduction
Male infertility is a major contributor to infertility worldwide, accounting for approximately 30% to 50% of infertility cases and substantially affecting reproductive outcomes among couples [1]. Abnormal sperm morphology is associated with impaired natural conception and assisted reproductive technology (ART) outcomes [2]. Intracytoplasmic sperm injection (ICSI) is widely used for male factor infertility, particularly in cases of impaired sperm quality [3].
Conventional ICSI bypasses several physiological mechanisms of sperm selection. During ICSI, sperm selection is based primarily on motility and morphology assessed under light microscopy and therefore depends on the embryologist’s experience [4]. This approach cannot directly assess sperm nuclear maturity or DNA integrity. Accumulating evidence suggests that sperm DNA fragmentation may adversely affect embryo development and pregnancy outcomes after ART [5].
Physiological intracytoplasmic sperm injection (PICSI) was developed to improve sperm selection by leveraging the ability of mature spermatozoa to bind hyaluronan (HA), a component of the cumulus-oocyte matrix [6,7]. In some studies, HA-bound spermatozoa have shown greater maturity, better chromatin integrity, and lower DNA fragmentation, providing a biological rationale for the use of PICSI in selected cases of male factor infertility [7].
However, the clinical effectiveness of PICSI compared with conventional ICSI remains a topic of debate. Several randomized and observational studies have reported no consistent improvements in fertilization, embryo quality, clinical pregnancy, or live birth outcomes [8-10]. Most available evidence comes from populations outside Southeast Asia, and real-world data from Vietnam remain limited. Differences in patient characteristics, laboratory protocols, access to fertility services, and clinical decision-making may influence ART outcomes and the generalizability of published findings.
Although large-scale prospective controlled trials remain necessary to define the optimal indications for PICSI, retrospective real-world studies from underrepresented settings are also valuable. Therefore, this study aimed to provide matched, single-center evidence from a Vietnamese population of couples affected by abnormal sperm morphology, thereby informing local clinical practice and contributing Southeast Asian data to the broader literature.
Methods
1. Study design
This retrospective observational matched cohort study was conducted at the in vitro fertilization (IVF) and Tissue Engineering Center, Hanoi Medical University Hospital, Vietnam. Medical records from IVF/ICSI cycles performed between January 2022 and January 2024 were reviewed. Patients were not randomly assigned to treatment groups. The choice of sperm injection technique, PICSI or conventional ICSI, was determined according to routine clinical practice at the time of treatment.
2. Study population
The inclusion criteria were couples in whom the female partner had at least three ovarian follicles measuring ≥14 mm after controlled ovarian stimulation and the male partner had a sperm concentration >15 million/mL with normal morphology <4% according to World Health Organization 2021 criteria. Couples underwent either PICSI or conventional ICSI. Female patients were excluded if they had stimulation failure, used donor oocytes, had endometriosis, or had endocrine or metabolic disorders such as polycystic ovary syndrome, thyroid dysfunction, or hyperprolactinemia. Male partners were excluded if spermatozoa were surgically retrieved, cryopreserved, or donor-derived. These criteria were applied to improve homogeneity and reduce major clinical confounding.
3. Matching strategy and methodological justification
To improve comparability between groups, 1:1 direct matching was performed using female age, anti-Müllerian hormone (AMH) level, and antral follicle count (AFC). These variables were selected because female age and ovarian reserve are major determinants of oocyte yield, embryo competence, and ART outcomes and therefore could confound comparisons between sperm injection techniques. Each PICSI cycle was matched to an ICSI cycle with a female age difference ≤2 years and AMH and AFC values within a 20% range.
These thresholds were chosen as a pragmatic clinical matching strategy. A 2-year age window was considered sufficiently narrow to reduce age-related differences while preserving an adequate matched sample. A 20% range for AMH and AFC was used to maintain similarity in ovarian reserve while avoiding excessive loss of eligible cycles in this retrospective single-center dataset. Propensity score matching was not applied because of the relatively small sample size and the availability of clinically relevant variables that enabled direct matching.
4. Study flow
A total of 464 medical records were initially retrieved. After eligibility criteria and direct matching were applied, 246 cycles were included in the final matched cohort, with 123 cycles in the PICSI group and 123 cycles in the ICSI group. For the clinical outcome analysis, only first eligible frozen embryo transfer cycles using day-5 blastocysts were included. Accordingly, 89 of 123 PICSI cycles and 92 of 123 ICSI cycles met the criteria for first eligible day-5 frozen blastocyst transfer. The remaining 34 PICSI cycles and 31 ICSI cycles were excluded from the clinical analysis because they either had day-5 blastocysts vitrified but had not undergone a first frozen embryo transfer within the available study follow-up period or had incomplete transfer outcome records. None of these cycles were excluded because only day-6 blastocysts were available. Day-6 blastocyst outcomes were not included because they were not consistently captured in the retrospective database. The study flow is shown in Figure 1.
5. Procedural overview
1) Oocyte retrieval protocol
Patients underwent ovarian stimulation with recombinant follicle-stimulating hormone (rFSH), including follitropin alfa (Gonal-F; Merck Serono), follitropin beta (Puregon; Organon), or Follitrope (LG Chem), starting on days 2–4 of the menstrual cycle. The rFSH dose was determined based on ovarian reserve markers, including age, AMH level, and AFC. When the leading follicles reached ≥12 mm, patients received either a gonadotropin-releasing hormone (GnRH) antagonist—cetrorelix acetate (Cetrotide) or ganirelix (Orgalutran)—at 0.25 mg/day or dydrogesterone (Duphaston 30 mg/day) according to the progestin-primed ovarian stimulation protocol.
Final oocyte maturation was triggered with human chorionic gonadotropin (β-hCG; 5,000–10,000 IU of Ovitrelle or IVF-C), a GnRH agonist (triptorelin 0.2 mg), or a dual trigger when ≥3 follicles measuring ≥17 mm were present. Oocyte retrieval was performed 34 to 36 hours after the trigger under transvaginal ultrasound guidance.
Follicular fluid was transferred to the IVF laboratory, where embryologists used a stereomicroscope to identify and collect cumulus-oocyte complexes (COCs) in a sterile workspace. The COCs were transferred to washing medium (GMOPS; Vitrolife) and cultured in a tri-gas incubator at 37 °C with 5% O₂ and 6% CO₂. The COCs were denuded using G-IVF medium (Vitrolife) containing 10% hyaluronidase.
2) Sperm preparation protocol
Semen samples were prepared using double-density gradient centrifugation with SpermGrad (Vitrolife). Briefly, 1 mL of semen was layered over 1 mL of 45% gradient and 1 mL of 90% gradient in a 15-mL Falcon tube. Samples were centrifuged at 250 × g for 8 minutes at room temperature. The supernatant was discarded, and the sperm pellet was washed with SpermRinse (Vitrolife) and centrifuged again at 250 × g for 5 minutes. The final pellet was used for sperm selection. For conventional ICSI, spermatozoa were loaded into ICSI medium (Vitrolife) 30 minutes before injection, and morphologically normal spermatozoa were selected for injection into oocytes.
For PICSI, PICSI dishes (CooperSurgical) were prepared 30 minutes before injection by hydrating the HA microdots with 10 μL of GMOPS Plus (Vitrolife). Drops of 10% polyvinylpyrrolidone ICSI medium (Vitrolife) were placed elsewhere in the dish for sperm immobilization. The dishes were covered with 3–4 mL of light mineral oil (Irvine Scientific) and incubated for 20 minutes at room temperature to optimize sperm binding. Spermatozoa that bound effectively to HA on the PICSI dish were selected and injected into the oocyte cytoplasm.
3) Embryo culture, blastocyst assessment, vitrification, warming, and embryo transfer
Injected oocytes were cultured in CSCM-C medium (Irvine Scientific) in a benchtop incubator. Fertilization was assessed 16 to 18 hours after injection based on the presence of two pronuclei. Embryos were cultured to the blastocyst stage and assessed using the Gardner classification system. High-quality blastocysts were defined as blastocysts with an expansion grade ≥3 and inner cell mass and trophectoderm grades of A or B (≥3BB).
For embryological analysis, day-5 blastocyst formation and high-quality day-5 blastocyst outcomes were evaluated because these variables were consistently recorded in the laboratory database. Day-6 blastocyst data were not included in the primary embryological comparison because their retrospective capture was incomplete. For clinical outcome analysis, only first frozen embryo transfer cycles using day-5 blastocysts were included; day-6 transfer cycles were excluded from the clinical dataset.
All blastocysts were vitrified using Vitrification Kit 101 (Cryotech) according to the manufacturer’s protocol. Briefly, blastocysts were equilibrated in equilibration solution until shrinkage and re-expansion were observed, exposed to vitrification solution for approximately 60 seconds, loaded onto the carrier in minimal volume, and immediately plunged into liquid nitrogen.
Warming was performed using Warming Solution 205 (Cryotech) according to the manufacturer’s protocol, including sequential exposure to warming, dilution, and washing solutions at controlled temperatures and times. After warming, embryos were cultured in CSCM medium (Irvine Scientific) for approximately 2 hours before intrauterine transfer.
4) Endometrial preparation and frozen embryo transfer protocol
For hormone replacement therapy, the endometrium was prepared using oral estradiol (Valiera 2 mg) at a total dose of 8 to 12 mg/day. Luteal-phase support began when the endometrial thickness reached 8 to 14 mm and consisted of vaginal progesterone (Utrogestan 800 mg/day) and oral Duphaston (40 mg/day).
6. Outcome measures and definitions
Embryological outcomes included the number of oocytes retrieved, the number of metaphase II (MII) oocytes, fertilization rate, day-5 blastocyst formation rate, and high-quality day-5 blastocyst rate. The fertilization rate was defined as the number of two-pronuclear zygotes divided by the number of injected MII oocytes. The day-5 blastocyst formation rate was calculated from normally fertilized embryos according to routine laboratory records. The high-quality day-5 blastocyst rate was calculated using the number of blastocysts graded ≥3BB.
Clinical outcomes included the rates of serum β-hCG positivity, clinical pregnancy, implantation, ongoing pregnancy, live birth, and pregnancy with multiple gestational sacs. Positive β-hCG was defined as a positive serum β-hCG test after embryo transfer. Clinical pregnancy was defined as visualization of at least one gestational sac on ultrasound. The implantation rate was defined as the number of gestational sacs observed divided by the total number of embryos transferred. Ongoing pregnancy was defined as a viable pregnancy beyond 12 gestational weeks. Live birth was defined as the delivery of at least one live-born infant.
7. Statistical analysis
Continuous variables are reported as mean±standard deviation, and categorical variables are reported as frequencies and percentages, as appropriate. Numerical variables were compared between study groups using the Student’s t-test for independent samples. Categorical variables were compared using the chi-square (χ²) test, and the Fisher’s exact test was used when expected cell counts were less than 5. A p-value of less than 0.05 was considered to indicate statistical significance. Numerical variables were compared across male age groups using one-way analysis of variance. All statistical analyses were performed using SPSS version 20.0 (IBM Corp.). Subgroup analyses stratified by male age (≤35 years vs. >35 years) and female age (<30 years vs. ≥30 years) were conducted to explore potential effect modification.
8. Ethical approval
The study was approved by the Institutional Review Board of Hanoi Medical University (IRB No. IRB.VN01.001/RB00003121/FWA00004148; approval ID: HMUIRB2538). Because of the retrospective design and use of anonymized data, the requirement for informed consent was waived. Trial registration was not applicable for this retrospective observational study.
Results
1. Characteristics of the study population
From January 2022 to January 2024, 246 couples met the inclusion criteria and were included in two matched groups: the PICSI group (n=123) and the ICSI group (n=123). The baseline characteristics of the study participants are presented in Table 1.
No measured baseline variable differed significantly between the PICSI and ICSI groups. Mean female age was 35.40±4.44 years in the PICSI group and 35.15±4.09 years in the ICSI group (p=0.646). Ovarian reserve markers were also similar between groups: mean AMH level was 3.18±2.77 ng/mL in the PICSI group and 3.13±2.14 ng/mL in the ICSI group (p=0.874), and mean AFC was 11.86±4.60 and 11.77±4.87, respectively (p=0.882). Stimulation duration ranged from 7 to 12 days in both groups, with no statistically significant between-group difference (p=0.529). Among male partners, age, sperm concentration, motility, and morphology did not differ significantly between groups (all p>0.05). Mean sperm concentration was 67.71±35.13 million/mL in the PICSI group and 83.78±112.57 million/mL in the ICSI group, and sperm motility and morphology were also comparable. These findings indicate that the matching process achieved comparable baseline characteristics between groups, supporting subsequent between-group outcome comparisons (Table 1).
2. Embryological outcomes
The mean number of oocytes retrieved was 12.36±7.00 in the PICSI group and 12.23±6.88 in the ICSI group (p=0.883). The mean number of mature MII oocytes was also similar between groups, at 8.96±5.54 and 9.14±5.02, respectively (p=0.789). Fertilization rates were 88.59%±15.99% after PICSI and 91.39%±13.15% after ICSI (p=0.134). The day-5 blastocyst formation rate was 63.07%±26.68% in the PICSI group and 66.95%±29.45% in the ICSI group (p=0.279). The high-quality day-5 blastocyst rate was 44.94%±25.29% after PICSI and 40.68%±23.21% after ICSI (p=0.170). Overall, no statistically significant differences in embryological outcomes were observed between groups, indicating comparable embryo-generation efficiency (Table 2).
3. Clinical outcomes after first day-5 frozen blastocyst transfer
Among the 246 matched cycles included in the embryological analysis, 181 cycles underwent the first eligible day-5 frozen blastocyst transfer, including 89 cycles in the PICSI group and 92 cycles in the ICSI group. No day-6 embryo transfer cycles were included in this clinical outcome analysis. Clinical outcomes after the first day-5 frozen blastocyst transfer were comparable between the PICSI and ICSI groups. No statistically significant differences were observed in the rates of β-hCG positivity, clinical pregnancy, implantation, ongoing pregnancy, live birth, or pregnancy with multiple gestational sacs. Overall, these findings suggest that PICSI did not improve clinical outcomes compared with conventional ICSI in first day-5 frozen blastocyst transfer cycles (Table 3).
4. Subgroup analyses by female and male age
Subgroup analyses were performed according to female age (<30 years vs. ≥30 years) and male age (≤35 years vs. >35 years). No statistically significant differences were observed between PICSI and ICSI in either female age subgroup. Among women aged ≥30 years, the high-quality day-5 blastocyst rate was numerically higher in the PICSI group, whereas this pattern was not observed among women aged <30 years. Clinical outcomes, including β-hCG positivity, clinical pregnancy, implantation, ongoing pregnancy, and live birth, remained comparable between groups. These exploratory findings suggest no clear advantage of PICSI based on female age alone (Table 4).
Similarly, no statistically significant differences were observed between PICSI and ICSI for any measured outcome in either male age subgroup (≤35 or >35 years; all p>0.05). Fertilization rates, day-5 blastocyst rates, high-quality day-5 blastocyst rates, β-hCG positivity, clinical pregnancy, implantation, ongoing pregnancy, and live birth were comparable between groups. These findings indicate that PICSI was not associated with superior outcomes compared with ICSI, including among couples with male partners aged >35 years (Table 5).
Discussion
In this matched retrospective cohort, PICSI was not associated with a statistically significant advantage over conventional ICSI among couples in whom the male partner had abnormal sperm morphology. Embryological outcomes, including fertilization, day-5 blastocyst formation, and high-quality day-5 blastocyst rates, were comparable between groups. Similarly, among cycles involving first eligible day-5 frozen blastocyst transfer, PICSI did not significantly improve clinical outcomes. These findings are consistent with previous studies reporting no clear benefit of PICSI over ICSI for embryo quality, clinical pregnancy, or live birth outcomes [8,10,11].
One possible explanation is that abnormal sperm morphology alone may be insufficient to identify patients most likely to benefit from HA-based sperm selection. Although HA-bound spermatozoa have been associated with greater maturity and lower DNA fragmentation, routine morphology assessment does not directly reflect sperm DNA integrity [7,12]. Therefore, patients with abnormal sperm morphology may represent a heterogeneous population that includes men both with and without clinically relevant sperm DNA damage. This heterogeneity may partly explain why PICSI did not produce a measurable overall benefit in the present study.
Despite the absence of statistically significant differences, some subgroup patterns were observed. In the female age subgroup of ≥30 years and the male age subgroup of >35 years, certain embryo-related outcomes were numerically more favorable with PICSI than with ICSI. These findings should be interpreted cautiously because the subgroup analyses were exploratory and underpowered. Nevertheless, these patterns may suggest that age-related changes in sperm DNA integrity or oocyte repair capacity influence the potential benefit of PICSI in selected subgroups. This interpretation is biologically plausible because sperm DNA fragmentation and oocyte DNA repair capacity have been associated with embryo development and ART outcomes [5,13]. Larger prospective studies are needed to determine whether these patterns represent a clinically relevant effect.
The main strength of this study is that it provides real-world data from a Vietnamese ART population, in which evidence regarding PICSI remains limited. In addition, direct 1:1 matching based on female age, AMH level, and AFC improved baseline comparability between the PICSI and ICSI groups, because these variables are important determinants of ovarian response and embryo yield. The clinical outcome analysis was also restricted to the first eligible day-5 frozen blastocyst transfer, reducing potential bias from repeated transfer cycles and variation between day-5 and day-6 embryo transfers.
Several limitations should be acknowledged. First, the retrospective, single-center design limits causal inference and may introduce selection bias. Second, the sperm DNA fragmentation index (DFI) was not assessed. This is an important limitation because PICSI is primarily intended to select mature spermatozoa with better DNA integrity, and the absence of DFI data makes it difficult to determine whether the included patients represented the optimal target population for PICSI. Abnormal sperm morphology alone is not equivalent to elevated sperm DNA fragmentation; therefore, patients selected based on morphology may represent a heterogeneous population with variable levels of DNA damage, potentially diluting any benefit of PICSI. Future prospective studies should incorporate standardized DFI assessment, for example using the sperm chromatin structure assay or terminal deoxynucleotidyl transferase deoxyuridine triphosphate (dUTP) nick-end labeling, to identify patients most likely to benefit from physiological sperm selection. Third, although matching was performed for major ovarian reserve factors, other potential confounders, such as body mass index, infertility duration, lifestyle factors, detailed infertility causes, and embryo transfer strategy, could not be fully controlled. Finally, the clinical analysis was limited to first eligible day-5 frozen blastocyst transfers; therefore, the results may not be generalizable to day-6 transfers, fresh transfers, or subsequent frozen embryo transfer cycles. In addition, although pregnancies with multiple gestational sacs were identified from first-ultrasound data, the twin birth rate at delivery could not be reliably assessed because postnatal follow-up records were incomplete in this retrospective dataset. This limitation precludes a full evaluation of multiple-birth risk, an important safety outcome when multiple embryos are transferred.
In this real-world Vietnamese matched cohort, PICSI did not demonstrate superior effectiveness compared with conventional ICSI in couples affected by abnormal sperm morphology. No statistically significant differences were observed in embryological or clinical outcomes. However, non-significant numerical patterns in selected older age subgroups suggest that PICSI may warrant further investigation in selected populations. Further prospective, multicenter studies incorporating sperm DNA fragmentation testing, sibling-oocyte designs, and standardized single-blastocyst transfer protocols are needed to better identify couples who may benefit from physiological sperm selection.
Notes
Conflict of interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
The authors thank the staff of the IVF and Tissue Engineering Center, Hanoi Medical University Hospital, for their support with data collection and laboratory procedures. The authors also acknowledge the patients who participated in this study.
Author contributions
Conceptualization: HPN, DNL. Methodology: HPN, DNL. Formal analysis: HPN, DNL, HTN, PTTD. Data curation: DNL, HTL. Project administration: SKN. Visualization: PTTD. Software: PTTD. Validation: HPN, DNL, PTTD. Investigation: DNL, HTL. Supervision: SKN. Writing-original draft: HPN, DNL, HTN, THD. Writing-review & editing: HPN, DNL, HTN, THD, HTL, PTTD, SKN. Approval of final manuscript: HPN, DNL, HTN, THD, HTL, PTTD, SKN.
