Use of Hyaluronan in the Selection of Sperm for Intracytoplasmic Sperm Injection (ICSI): A Systematic Review and Meta-Analysis

Background : Studies on the effect of intracytoplasmic injection of hyaluronan-bound spermatozoa (HA-ICSI) on infertility are insufficient, and its use in treating patients remains controversial. Therefore, we aimed to determine the effectiveness of HA-ICSI in couples with infertility. Methods : A systematic literature review and meta-analysis were conducted to explore the effect of HA-ICSI on couples with infertility. All studies were examined using relative risks (RR) with 95% confidence intervals (95% CI). Results : A total of 1174 publications were retrieved, of which 16 (10 randomized controlled trials (RCTs), five cohort trials, and one publication, including an RCT and a cohort trial) were considered eligible for inclusion. Meta-analysis of the cohort studies indicated a significant advantage for HA-ICSI in terms of live birth rate (LBR), clinical pregnancy rate (CPR), biochemical pregnancy rate (BPR), implantation rate (IR), fertilization rate (FR), and good-quality embryo rate. No difference in spontaneous abortion rate (SAR) or cleavage rate between the HA-ICSI and conventional intracytoplasmic sperm injection (ICSI) groups was observed. Based on the pooled results of all available studies and RCTs, SAR was significantly reduced in the HA-ICSI group than in the conventional ICSI group. The benefits of CPR, IR, and FR were recognized in the pooled results of all available studies; however, RCT analysis did not demonstrate these benefits. Conclusions : The cohort studies indicated a significant advantage of HA-ICSI in terms of LBR, CPR, BPR, IR, FR, and good-quality embryo rates. In RCTs, HA-ICSI significantly reduced the SAR compared to conventional ICSI. Further RCTs with larger sample sizes are required to confirm the beneficial effects of HA-ICSI.


Introduction
Intracytoplasmic sperm injection (ICSI) improves fertilization rate (FR) in couples with male factor infertility.ICSI is a significant achievement in assisted reproductive technology (ART) and has become probably the most important therapy for male infertility in recent years.Globally, ICSI use is increasing, mainly because of increased ICSI use in cycles carried out for non-male factors, such as frozen oocytes, in vitro maturation of human immature oocytes, preimplantation genetic testing, and infertility with previous fertilization failure.ICSI use has been observed in 70-80% of fresh cycles, with increasing applications encountered [1].ICSI is considered the most "revolutionary" in vitro insemination technique since an embryologist artificially selects a single spermatozoon for injection, which can fertilize an oocyte notwithstanding its morphology or motility.Additionally, some natural fertilization processes, such as sperm-cumulus interaction, sperm-zone penetration, and acrosome reaction, are circumvented in the ICSI process [2].During conventional ICSI, sperm are selected based on motility and morphology, which may not reflect sperm quality.Theoretically, the prominent use of ICSI may increase the possibility of injecting spermatozoa that are defective in centrosome integrity, genetic constitution, phospholipase C zeta content, or DNA methylation [3][4][5].Embryo quality is influenced by the quality of gametes, oocytes, and spermatozoa.Natural barriers to fertilization are circumvented in ICSI; therefore, embryo quality is affected.Therefore, ICSI treatments may be optimized by selecting the ideal spermatozoa before injection.However, the shapes of individual spermatozoa do not indicate chromatin integrity and chromosomal aberrations.Visual shape assessment, which selects the best-looking sperm during ICSI, is unreliable and may result in abnormal chromosomes in subsequent developmental stages [6,7].Concerns regarding the potential adverse effects of ICSI on subsequent fertilization, embryo development, and offspring health owing to deviations from natural selection exist.Therefore, a practical test for selecting healthy sperm for ICSI is required.
Polyvinylpyrrolidone (PVP) is used in conventional ICSI to reduce sperm velocity and facilitate the smooth injection of oocytes.However, PVP causes submicroscopic changes in sperm structure and damages sperm membrane integrity and sperm nucleus [8,9].Moreover, human oocytes are unable to degrade PVP, subsequently affecting pregnancy rates [10][11][12].
Recently, several sperm-selection techniques have been developed.These techniques are categorized into methods based on sperm density, morphology, motility, membrane integrity, surface charge, and nuclear integrity.However, these techniques have not demonstrated enhanced clinical outcomes that would support routine clinical application despite their ability to improve sample quality.
Hyaluronan (hyaluronic acid, HA) is a linear anionic polysaccharide containing N-acetyl-D-glucosamine and Dglucuronic acid bound with β-1,3 and β-1,4 glycosidic bonds [13].This macromolecule is a constituent of the cumulus cell matrix in human oocytes.HA is a physiological selector, and only mature spermatozoa with low chromosomal aneuploidy and low fragmentation can bind to it [14].The head region of mature spermatozoa contains HA receptors that facilitate passage through the extracellular matrix to the oocyte [15].Sperms without HA receptors may be deselected, resulting in a lower possibility of reaching oocytes.Immature sperms lacking HA-binding sites are also associated with an increased frequency of chromosomal aberrations [14].However, sperm without HA receptors can be selected and injected into oocytes during conventional ICSI.Sperm morphology is not associated with the selection of haploid spermatozoa [16]; however, the frequency of chromosomal diploidy in HA-selected spermatozoa is reduced by 4-to 6-fold compared to that in semen sperm [17].This selection process is expected to facilitate the selection of a single mature spermatozoon with desirable morphological characteristics and without cytoplasmic retention, persistent histones, chromosomal aneuploidy, or DNA fragmentation [17,18].
Theoretically, the selection of HA-bound sperm for ICSI facilitates fertilization and produces high-quality embryos, leading to better clinical outcomes.Studies on intracytoplasmic injection of hyaluronan-bound spermatozoa (HA-ICSI) are insufficient, and the application of HA-ICSI to treat infertility remains controversial, although it has been used for over a decade.HA-ICSI includes HAcoated dishes (PICSI dishes) and HA-containing media (SpermSlow TM and SpermCatch TM ).Several studies have documented the practical benefits of HA-ICSI in terms of fertilization, embryo quality, and pregnancy outcomes [19][20][21][22].However, some studies have challenged the application of HA-ICSI [23,24].Therefore, we performed a metaanalysis and systematic review to evaluate the effectiveness of HA-ICSI in couples with infertility.

Literature Search and Study Selection
This systematic review and meta-analysis was previously registered with PROSPERO (CRD42024540998) and followed PRISMA guidelines.A comprehensive PubMed, Embase, and Cochrane search was performed to identify studies aimed at comparing the clinical outcomes of HA-ICSI with those of conventional ICSI in couples with infertility.The search was restricted to papers fully published in English on 13 October, 2023, using the following keywords: ("sperm" or "spermatozoa" or "spermatozoon" or "IVF" or "in vitro fertilization" or "ICSI" or "intracytoplasmic sperm injection") and ("hyaluronic acid" or "hyaluronan").Our study's title and abstract were used to screen all items retrieved from the primary search.Irrelevant items were removed, and the articles of potential interest were further investigated for citations that met the inclusion criteria.We also manually reviewed the bibliographies of original and reviewed articles.

Eligibility Criteria
All studies investigating the effect of HA-ICSI met the following inclusion criteria: (1) randomized controlled trials (RCTs) or cohort trials, (2) had outcomes, such as live birth rate (LBR), clinical pregnancy rate (CPR), spontaneous abortion rate (SAR), biochemical pregnancy rate (BPR), implantation rate (IR), fertilization rate (FR), cleavage rate, and good-quality embryo rate, and (3) published in English.
Exclusion criteria were as follows: (1) no original data for retrieval, (2) duplicate publications, (3) no full texts, and (4) review articles, case reports, or comments from editors.Two authors independently identified relevant studies, and any discrepancies were discussed.

Data Extraction and Quality Assessment
Two authors independently extracted data from each included article.The extracted data included the first author's name, year of publication, country, study design and period, HA-binding methods, control group, number of cycles or patients in the HA and control groups, and inclusion and exclusion criteria.Disagreements between the two authors were resolved through discussion.A third reviewer's input was required for unresolved disagreements.
The Cochrane Collaboration tool was used to evaluate the methods of random allocation, the presence and quality of allocation concealment and blinding, and the existence of incomplete outcome data and selective outcome reporting for all included RCTs [25].The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of individual cohort studies.The NOS uses a star system with a maximum of nine stars to assess a study in three domains: selection of study groups, comparability of groups, and ascertainment of interest outcome.We determined studies that received a score of nine stars to be of a low risk of bias.Studies scoring seven or eight stars were of a moderate risk of bias, and those scoring ≤six were classified as high risk.Quality assessment was performed independently by two authors, and disagreements were resolved through discussion.

Statistical Analysis
The effect of HA-ICSI was examined using relative risk (RR) with 95% confidence intervals (95% CI).The Q (significance level of p < 0.1) and I 2 statistics were used to determine the statistical heterogeneity between studies.Absence of, moderate, high, and extreme heterogeneities were indicated by I 2 <25%, 25% ≤ I 2 < 50%, 50% ≤ I 2 < 75%, and I 2 ≥75%, respectively.The random-effects model (DerSimonian and Laird method) was employed if the p-value of the Q test was <0.1; otherwise, the fixedeffects model (Mantel-Haenszel method) was used.When ≥10 studies were included in a single analysis, Begg's funnel plot and Egger's linear regression were used to assess the possibility of publication bias.A subgroup analysis was performed based on the type of study design (RCTs versus cohort studies) to explore potential sources of heterogeneity.When three or more studies were included in an analysis, a sensitivity analysis was performed by excluding one study from each round and evaluating each study's influence on the overall effect size.Data were analyzed using the STATA software (StataCorp, College Station, TX, USA).Statistical significance was set at p < 0.05.

Literature Search
A total of 1174 results were obtained from the PubMed, Embase, and Cochrane databases and other sources.A total of 1122 articles were excluded after reading the titles and abstracts.Two authors read the full texts of the 52 articles, and 16 articles were finally analyzed (Fig. 1).

Meta-Analysis of Live Birth Rate
We evaluated the LBR between the HA-ICSI and conventional ICSI groups in six studies (five RCTs and one cohort study).The statistical analysis performed using the random effects model revealed no significant difference in the LBR between the two groups (RR = 1.25, 95% CI: 0.94-1.65);however, a high heterogeneity was observed (I 2 = 49.3%).A subgroup analysis was performed based on the included study designs.The pooled outcomes from the five RCTs also showed no significant difference (RR = 1.09, 95% CI: 0.98-1.21).Additionally, no heterogeneity between the two groups was observed (I 2 = 0%).However, a significantly higher LBR was observed in the HA-ICSI group than in the conventional ICSI group from the single cohort study (RR = 3.20, 95% CI: 1.59-6.42).The pooled LBR results did not vary with the removal of any study from the overall included studies or RCTs (Supplementary Material), demonstrating no differences in LBR between the two groups.

Meta-Analysis of Clinical Pregnancy Rate
A total of eight RCTs and four cohort studies reported the CPR between the HA-ICSI and conventional ICSI groups.The pooled results suggested that CPR was significantly higher in the HA-ICSI group than in the conventional ICSI group (RR = 1.21, 95% CI: 1.01-1.43),with a high heterogeneity observed (I 2 = 53.5%).However, the pooled outcomes from the eight RCTs showed no significant difference (RR = 1.00, 95% CI: 0.92-1.09),and no significant heterogeneity between the two groups was observed (I 2 = 0%).The pooled outcomes from the four cohort studies revealed a significantly higher CPR in the HA-ICSI group than in the conventional ICSI group (RR = 1.86, 95% CI: 1.40-2.48),with no heterogeneity observed (I 2 = 0%).In contrast to the pooled overall results, removing the studies by Scaruffi P, et al. [19], Rezaei M, et al. [21], Erberelli RF, et al. [27], Parmegiani L, et al. [31], or Troya J and  A star system with a maximum of nine stars was used to assess a study in three domains.
Zorrilla I [24] led to a different sensitivity analysis result (Supplementary Material), demonstrating no difference in CPR between the two groups.However, for the RCTs, the pooled CPR results did not vary with the removal of any study (Supplementary Material), demonstrating no differences in CPR between the two groups.For the cohort studies, the pooled CPR results did not vary with the removal of any study; however, a significantly higher CPR was observed in the HA-ICSI group than in the conventional ICSI group.

Meta-Analysis of Spontaneous Abortion Rate
A total of six RCTs and three cohort studies that provided SAR data were included in our meta-analysis.A significantly lower SAR was observed in the HA-ICSI group than in the conventional ICSI group (RR = 0.65, 95% CI: 0.50-0.84),with no heterogeneity observed (I 2 = 0%).The pooled outcomes from the six RCTs also showed a significantly lower SAR in the HA-ICSI group than in the conventional ICSI group (RR = 0.63, 95% CI: 0.48-0.85),with no heterogeneity observed (I 2 = 0%).However, no significant difference was observed between the two groups from the three cohort studies (RR = 0.72, 95% CI: 0.39-1.36);however, a moderate heterogeneity was observed (I 2 = 38.7%).In contrast to previous pooled overall studies and RCTs' results, removing the study conducted by Miller, et al. [23] led to a different sensitivity analysis result (Supplementary Material), demonstrating no differ-ence in SAR between the two groups.For the cohort studies, the pooled SAR results did not vary with the removal of any study (Supplementary Material), demonstrating no differences in SAR between the two groups.

Meta-Analysis of Biochemical Pregnancy Rate
We compared the BPR between the HA-ICSI and conventional ICSI groups in seven studies (five RCTs and two cohort studies).The meta-analysis revealed no significant difference between the two groups (RR = 1.16, 95% CI: 0.93-1.46);however, a moderate heterogeneity was observed (I 2 = 48.8%).The pooled outcomes from the five RCTs also showed no significant difference between the two groups (RR = 1.00, 95% CI: 0.92-1.10),and no heterogeneity between the two groups was observed (I 2 = 0%).The pooled outcomes from the two cohort studies revealed a significantly higher BPR in the HA-ICSI group than in the conventional ICSI group (RR = 1.94, 95% CI: 1.30-2.91),with no heterogeneity observed (I 2 = 0%).The pooled BPR results did not vary with the removal of any study from the overall studies or RCTs (Supplementary Material), demonstrating no differences in BPR between the two groups.

Meta-Analysis of Implantation Rate
Pooled results from three RCTs and two cohort studies indicated a significantly higher IR in the HA-ICSI group than in the conventional ICSI group (RR = 1.43, 95% CI: 1.01-2.04),with high heterogeneity observed (I 2 = 64.6%).
The meta-analysis of the three RCTs revealed no difference in IR between the HA-ICSI and conventional ICSI groups (RR = 1.11, 95% CI: 0.86-1.45),and no heterogeneity was observed (I 2 = 0%).The pooled outcomes from the two cohort studies revealed a significantly higher IR in the HA-ICSI group than in the conventional ICSI group (RR = 2.03, 95% CI: 1.45-2.83),with high heterogeneity observed (I 2 = 61.7%)(Table 4).In contrast to previous pooled overall results, removing the studies by Scaruffi P, et al. [19], Majumdar G and Majumdar A [28], Parmegiani L, et al. [31], or Parmegiani L, et al. [32] led to a different sensitivity analysis result (Supplementary Material), demonstrating no differences in IR between the two groups.However, for RCTs, the pooled IR results did not vary with the removal of any study (Supplementary Material), demonstrating no differences in IR between the two groups.

Meta-Analysis of Fertilization Rate
A total of seven RCTs and four cohort studies provided FR data and were included in the meta-analysis.The overall results showed a significantly higher FR in the HA-ICSI group than in the conventional ICSI group (RR = 1.05, 95% CI: 1.003-1.09),with high heterogeneity observed (I 2 = 65.8%).The pooled results from the seven RCTs revealed no difference in FR between the HA-ICSI and conventional ICSI groups (RR = 1.02, 95% CI: 0.97-1.07);however, moderate heterogeneity was observed (I 2 = 45.0%).The pooled outcomes from the four cohort studies revealed a significantly higher FR in the HA-ICSI group than in the conventional ICSI group (RR = 1.10, 95% CI: 1.06-1.14),with high heterogeneity observed (I 2 = 51.8%).In contrast to the pooled overall results, removing the studies conducted by Kim et al. [22], Liu, et al. [26], Erberelli, et al. [27], Parmegiani, et al. [31], Parmegiani, et al. [32], and Van Den Bergh, et al. [33] led to a different sensitivity analysis result (Supplementary Material), demonstrating no differences in FR between the two groups.For RCTs, the pooled FR results did not vary with the removal of any study (Supplementary Material), demonstrating no differences in FR between the two groups.For the cohort studies, the pooled FR results did not vary with the removal of any study (Supplementary Material); however, a significantly higher FR in the HA-ICSI group than in the conventional ICSI group was observed.

Meta-Analysis of Cleavage Rate
We evaluated the cleavage rate between the HA-ICSI and conventional ICSI groups in six studies (three RCTs and three cohort studies).The meta-analysis revealed a significantly lower cleavage rate in the HA-ICSI group than in the traditional ICSI group (RR = 0.97, 95% CI: 0.94-0.999),with high heterogeneity observed (I 2 = 74.3%).However, the pooled outcomes from the three RCTs showed no significant difference between the two groups (RR = 0.98, 95% CI: 0.94-1.03),with high heterogeneity observed (I 2 = 64.2%).A similar result was observed from the pooled results of the three cohort studies (RR = 0.96, 95% CI: 0.92-1.01),with a high heterogeneity observed (I 2 = 78.5%).In contrast to the previous pooled overall results, removing the studies conducted by Kim SJ, et al. [22], Parmegiani L, et al. [31], Parmegiani L, et al. [32], or Ciray HN, et al. [34] led to a different sensitivity analysis result (Supplementary Material), demonstrating no differences in cleavage rates between the two groups.For the RCTs and cohort studies, the pooled cleavage rate results did not vary with the removal of any study (Supplementary Material), demonstrating no differences in cleavage rate between the two groups.

Meta-Analysis of Good-Quality Embryo Rate
We evaluated the good-quality embryo rate between the HA-ICSI and conventional ICSI groups in seven studies (five RCTs and two cohort studies).No significant difference between the two groups was observed (RR = 1.19, 95% CI: 0.97-1.46),with high heterogeneity observed (I 2 = 83.8%).The pooled outcomes from the five RCTs revealed no difference in good-quality embryo rate between the HA-ICSI and conventional ICSI groups (RR = 1.04, 95% CI: 0.90-1.21),with a high heterogeneity observed (I 2 = 51.9%).The pooled outcomes from the two cohort studies revealed a significantly higher good-quality embryo rate in the HA-ICSI group than in the traditional ICSI group (RR = 1.56, 95% CI: 1.36-1.79);however, no heterogeneity was observed (I 2 = 0%).For all studies and RCTs, the pooled results of good-quality embryo rates did not vary with the removal of any study (Supplementary Material), demonstrating no differences in good-quality embryo rates between the two groups.

Publication Bias
In our meta-analysis, evidence of a publication bias was observed.Begg's funnel plot and Egger's linear regression for CPR yielded p values of 0.19 and 0.04, respectively.In addition, for studies examining FR, Begg's funnel plot and Egger's linear regression yielded p values of 0.88 and 0.57, respectively.

Discussion
Our meta-analysis included RCTs and cohort studies.We identified whether HA-ICSI was beneficial for patients compared to conventional ICSI in terms of LBR, CPR, SAR, BPR, IR, FR, cleavage rate, and good-quality embryo rate.We screened 16 publications (10 RCTs, five cohort studies, and one publication, including an RCT and a cohort trial).Our meta-analysis of the cohort studies indicated a significant advantage for HA-ICSI in terms of LBR, CPR, BPR, IR, FR, and good-quality embryo rate, and no differences in SAR and cleavage rate between the HA-ICSI and control groups.Compared with that in the conventional ICSI group, the SAR in the HA-ICSI group significantly decreased in a meta-analysis of all available studies and RCTs.Similar to the RCT results, a meta-analysis of all studies revealed no difference between the HA-ICSI and control groups in terms of LBR, BPR, and good-quality embryo rates.The advantages of CPR, IR, and FR have been recognized in a meta-analysis of all available studies.However, a meta-analysis of RCTs did not demonstrate these advantages.HA-ICSI significantly decreased the cleavage rate in the meta-analysis of all available studies; however, for RCTs, no difference in the cleavage rate between the HA-ICSI and conventional ICSI groups was observed.
Our conclusions differ from those of the previous studies.Based on data from four RCTs, Lepine S, et al. [36] discovered that HA-ICSI reduced miscarriage but did not improve LBR or CPR.When our meta-analysis was limited to data sets from RCTs, we obtained similar results.RCTs have superior study designs owing to their internal validity; however, they may lack external validity [37][38][39].Therefore, it is necessary to include cohort studies.Beck-Fruchter R, et al. [40] reviewed six RCTs and two cohort studies.It was concluded that using the HA-ICSI technique yielded no improvement in fertilization and pregnancy rates.A meta-analysis of all available studies showed an improvement in embryo quality and IR, and a metaanalysis of RCTs only revealed an improvement in embryo quality.We conducted a meta-analysis of all cohort studies and recent RCTs, resulting in an extensive dataset for analysis and different results.Our meta-analysis and systematic review, including RCTs and cohort studies, are the most comprehensive analyses of the effects of HA binding on sperm selection in ICSI.In contrast to previous reviews, we thoroughly searched for relevant articles and updated some recent studies.
I 2 was relatively high in most analyses, indicating significant heterogeneity among these analyses.Therefore, a random-effects model was employed to minimize heterogeneity.An absence of heterogeneity in the RCTs or cohort studies in the meta-analysis of LBR, CPR, BPR, IR, and good-quality embryo rates was observed after a subgroup analysis was performed based on the type of study design, suggesting that differences in design and methods affect heterogeneity.
For the cohort studies, none of the outcomes varied significantly with the removal of any study, demonstrating that the results were not highly influenced by any study, suggesting the stability and reliability of the findings.For RCTs, all outcomes, except the SAR, did not vary with the removal of any study, demonstrating that the meta-analysis was thorough and that none of the studies greatly influenced the results.However, in our meta-analysis, the SAR outcome in RCTs was unreliable.Six RCTs reported SARs, excluding the study conducted by Miller D, et al. [23], which led to a different result of no significant difference in the SAR between the two groups, similar to the pooled results of cohort studies.Therefore, we suggest that the pooled SAR result in RCTs was greatly influenced by the study by Miller D, et al. [23], in which 2772 couples undergoing ICSI were randomly assigned to receive either HA-ICSI or conventional ICSI, and a lower miscarriage rate in HA-ICSI was observed.Additionally, the results of most of the included studies were not significant.
Our study had several limitations.First, the significant problem was that the results showed a high degree of statistical heterogeneity in some analyses.We conducted a subgroup analysis and discovered that the design and methodological differences between the RCTs and cohort studies may have contributed to the high heterogeneity observed.However, we were unable to eliminate the effects of other factors, such as the etiology of infertility, ovarian stimulation protocols, and exclusion criteria, which were not available for subgroup analysis owing to insufficient data.Moreover, the use of Embryoscope TM for embryo culture in one study [26] is a significant factor that affects heterogeneity.Second, some of the included articles were of low quality, and the number of included articles was insufficient.Third, a potential publication bias was observed.FR and Begg's funnel plot showed no apparent publication bias for CPR; however, Egger's linear regression results showed evidence of publication bias for CPR.The language of the included articles was restricted to English; therefore, the presence of another publication bias is possible.Additionally, our search was limited to published articles; as a result, unpublished articles with a possibility of meeting our inclusion criteria might have been excluded.Fourth, our metaanalysis included studies that met different inclusion and exclusion criteria.The enrollment of couples undergoing ICSI from the general in vitro fertilization (IVF) population, unselected for specific indications, might underestimate the potential advantages of HA-ICSI.Hence, the final results should be interpreted carefully and further evaluated using large samples, multiple centers, and high-quality studies.
Notwithstanding these limitations, our analysis included recently published articles, which provided greater detail on ICSI outcomes and focused on subgroup analysis according to the included study designs, leading to an increased significance of our meta-analysis.Additionally, most of the meta-analyzed studies demonstrated improved clinical outcomes with HA-ICSI.No study reported that HA-ICSI adversely affected ICSI outcomes.If a comprehensive multicenter prospective randomized study confirms the positive effects of HA-ICSI, HA-ICSI is a potential first-line infertility treatment for 'physiological' sperm selection before ICSI because of its ability to reduce genetic complications and its non-toxic state.

Conclusions
In summary, the cohort studies indicated a statistically significant advantage of HA-ICSI in terms of LBR, CPR, BPR, IR, FR, and good-quality embryo rate.All included studies and RCTs revealed that HA-ICSI significantly decreased SAR compared to conventional ICSI.The advantages of CPR, IR, cleavage rate, and FR have been recognized in a meta-analysis of all available studies.However, a meta-analysis of RCTs did not demonstrate these advantages.HA-ICSI did not have detrimental effects on ICSI outcome parameters.Extensive multicenter prospective randomized studies are required to confirm the beneficial effects of HA-ICSI.Additionally, HA-ICSI should be considered for physiological sperm selection before ICSI.

Fig. 1 .
Fig. 1.Flowchart of the selection of studies for inclusion in this meta-analysis.RCTs, randomized controlled trials.

Table 1 . Characteristics of the studies included in the systematic review.
≥1 × 10 6 /mL on the day of oocyte collection; metaphase II (MII) oocytes ≥4.Motile sperm count <1 × 10 6 /mL on the day of oocyte collection.Miller D, et al. 2019 [23] UK RCT/from February 1, 2014 to August 31, 2016 PICSI Standard ICSI 1381 patients 1371 patients Women were 18-43 years old; women's BMI was 19-35 kg/m 2 ; a FSH concentration was 3-20 mIU/mL or, an AMH concentration ≥1.5 pmol/L.Men were 18-55 years old; abstinence for at least 3 days.Donor or frozen gametes or undergoing split IVF-ICSI.Men had a vasovasostomy or were treated for cancer in the 24 months before recruitment.Unexplained infertile patients with normal semen parameters in according to WHO 2010 criterion for first cycle.Age >38; presence of uterine anomalies, hydrosalpinx, or morderate and severe en-dometriosis; oocytes ≤3.