1 IVF Unit, Department of Obstetrics and Gynaecology, Sapir Medical Centre, Meir Hospital, Kfar Saba, 2 Male Fertility Laboratory, Faculty of Life Sciences, Bar-Ilan University, Ramat Gan, 3 Bar Ilan University, Center for Infrastructures and IS, Ramat Gan and 4 Fertility Medical Centre, Rishon-Le-Zion, Israel
5 To whom correspondence should be addressed at: Male Fertility Laboratory, Faculty of Life Sciences, Bar-Ilan University, Ramat Gan 52900, Israel. Email: bartoob{at}mail.biu.ac.il
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: intracytoplasmic morphologically selected sperm injection (IMSI)/male fertility/sperm nucleus
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In a recent, prospective, feasibility study we demonstrated that fine morphological integrity of human sperm nuclei is positively associated with fertilization and pregnancy rates following IVF-ICSI. These results were obtained by the application of a new method of unstained, real-time, high magnification motile sperm organellar morphology examination (MSOME) to the leftover sperm fraction selected for microinjection in 100 random couples referred for ICSI (Bartoov et al., 2002). Accordingly, we introduced a modified IVF procedureintracytoplasmic morphologically selected sperm injection (IMSI)based on microinjection into retrieved oocytes of selected spermatozoa with strictly defined morphologically normal nuclei. The modified IMSI treatment did, in fact, result in significantly higher pregnancy rates, compared with conventional IVF-ICSI (Bartoov et al., 2001
, 2003
).
In addition to the strict morphological sperm selection, the IMSI method adapted some minor modifications in sperm preparation which were not conducted in the conventional IVF-ICSI process, i.e. use of a density gradient (Sil Select) in the routine preparation prior to selection, use of polyvinyl pyrrolidine (PVP), low temperature and a glass-bottomed dish during selection, prolonged sperm manipulation post-separation from the seminal fluid and sperm storage prior to microinjection.
Thus, one may claim that the advantage of IMSI is caused by these modified techniques, and is not necessarily related to the nuclear normality of the injected sperm cells. The aim of this study was to determine whether the increased pregnancy outcome was, indeed, attributable to the preferred nuclear morphology of the selected spermatozoa and not to the special sperm preparation technique adapted for IMSI. For this purpose, two matched IMSI groups with identical sperm preparation but different selection criteria were compared.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In 38 (12%) of these cycles, we were unable to find any spermatozoa with strictly defined morphologically normal nuclei. Subsequently, all the oocytes within one cycle were injected by second best morphologically evaluated sperm cells, selected according to compromised criteria (see MSOME criteria for spermatozoa selected for IMSI). In 126 (41%) of the cycles, the number of retrieved spermatozoa with preferred nuclear morphology was insufficient for all oocytes retrieved within one cycle. Thus, in these 126 cases, the ova were microinjected with spermatozoa with morphologically normal nuclei or with spermatozoa selected under compromised criteria (mixed microinjection). The fertilization rate and the percentage of top embryos at strict microinjection were significantly higher than the parallel variables at compromised microinjection (71.5±23.6 versus 56.6±36.0% and 30.14±24.2 versus 18.7±26.3, respectively, t=4.0, P0.01 and t=2.7, P
0.02, respectively). The mixed microinjection group was subdivided further into two subgroups: in 98 cycles, embryos derived from both kinds of microinjection were transferred in parallel (mixed transfer), while in the remaining 28 cases only embryos obtained from microinjection into the ova of spermatozoa with strictly defined morphologically normal nuclei were transferred. In 145 cases (47%), all the oocytes were injected with spermatozoa exhibiting the preferred nuclear morphology. Subsequently, all the transferred embryos in these cases were also obtained from microinjection of spermatozoa with strictly defined morphologically normal nuclei.
The pregnancy and abortion rates related to the mixed embryo transfer (n=98) were 26.5 and 30.7%, respectively. The pregnancy and abortion rates related to embryo transfer following the strict sperm selection (n=173) were 46.2 and 8.8%, respectively.
It should be noted that the above IMSI outcome data refer to the overall patient group from which the study group was derived and are not analysed further.
Negative group
The 38 cases where the transferred embryos were obtained from microinjection into the oocytes of the second best morphologically evaluated sperm cells, selected according to compromised criteria (see MSOME criteria for spermatozoa selected for IMSI), were designated as the negative group. All the participants in this group were informed that only second best morphologically evaluated sperm cells were retrievable for microinjection. Nevertheless, these patients wished to continue with the IMSI procedure. A written, informed, consent was obtained from each couple.
Positive group
A comparative study between the IVF-IMSI cycles, involving transfer of embryos derived following compromised versus strict sperm selection was conducted by matching the 38 negative cases with 38 IVF-IMSI cycles, where only embryos derived from microinjection of spermatozoa with strictly defined morphologically normal nuclei were transferred. The matching criteria were female's age and the number of previous failed ICSI attempts. The time period between the negative and positive matched cases was up to 4 months. The 38 positive matching cases and the remaining 135 cases, where embryos derived from injection of spermatozoa with strictly defined morphologically normal nuclei, were statistically similar in all examined demographic and IMSI outcome parameters.
The examined demographic parameters, i.e. male's and female's age, pregnancy expectation (the period between the first attempt to conceive and the IMSI cycle included in this study) and the numbers of previous failed ICSI and IMSI trials were statistically similar in the negative and positive study groups (Table I). Neither of the study groups included any previously reported data.
|
Sperm preparation for individual retrieval based on MSOME
About 4000 motile high density sperm cells, obtained from the above fraction, were transferred to a 4 µl observation microdroplet of SPERM medium (Medi-Cult, Jyllinge, Denmark) containing PVP solution (ICSITM-100, Vitrolife, Kungsbacka, Sweden).
In order to estimate the morphological state of the sperm nucleus, the technician has to follow the motile sperm cell by moving the microscopic stage in the x, y and z directions for at least 20 s. Therefore, the PVP solution was necessary to slow the sperm speed and, thereby, to prevent the highly motile spermatozoa from disappearing from the monitor screen during measurement. To reduce PVP toxicity, the concentration of the PVP was adjusted to a minimum (range 08%), which still enables image analysis. The observation microdroplet was placed in a sterile, glass-bottomed dish (Willco wells BV, Amsterdam, The Netherlands) under sterile paraffin oil (OVOIL-100; Vitrolife, Kungsbacka, Sweden). The sperm cells, suspended in the observation microdroplet, were used for individual retrieval by MSOME. For this purpose, the sterile glass-bottomed dish containing the microdroplet was placed on a microscopic stage over the top of a Uplan Apo x100 oil/1.35 objective lens previously covered by a droplet of immersion oil. In this way, the motile sperm cells, suspended in the observation droplet, could be examined at high magnification by the inverted microscope (Olympus IX 70, Tokyo, Japan) equipped with Nomarski differential interference contrast optics.
The success of the MSOME is associated with four image properties. (i) Optical resolution, which depends on microscope optics and on the light source of the microscope: theoretical resolution is light wavelength divided by numerical aperture of objective (in our case 1.5). This resolution is
200 nm for blue light and 300350 nm for red light. When used with the usual white-yellow light, the resolution is
300 nm. (ii) Image contrast, which is increased by Nomarski optics (special phase contrasting light processing). (iii) Maximal optical magnification
150, which is defined by the following factors: (a) objective magnification (100x); (b) magnification selector (1.5x); and (c) video coupler magnification 0.99 (UPMTV x0.3, PE x3.3). (iv) Magnification of the video system, which could be calculated from the following data: (a) CCD chip diagonal dimension 8 mm; and (b) TV monitor diagonal dimension 356 mm.
Thus, the calculated video magnification (b/a) is 44.5 (356/8). It is, however, dependent on the size of the display. In our case, the calculated total magnification is
150 x 44.5 = 6675x. This number is a ratio between the size of the object on display and the real size. The actual digitally enhanced magnification, as determined by a 0.01 mm Olympus objective micrometer, was 6300x.
The digital image processing provides additional image enhancement (special filters) for detail recognition by the operator and tools for morphological measurements. This video enhancement is very important for the MSOME analysis, which requires morphological definition of the size, counter and content of the sperm nucleus.
MSOME criteria for spermatozoa selected for IMSI
As described in our previous studies, the strict descriptive criteria for normally shaped nuclei were based on those defined by scanning electron microscopy, i.e. smooth, symmetric and oval configuration (Bartoov et al., 1981, 2002
, 2003
). For MSOME, the average length and width of this configuration were estimated in 100 spermatozoa with an obviously normal nuclear shape and were found to be 4.75±0.28 and 3.28±0.20 µm, respectively. During sperm selection, a fixed, transparent, celluloid form of a sperm nucleus fitting the normal criteria was superimposed on the examined cell; any sperm cell with a nuclear shape was excluded from selection if it varied in length or width by 2 SDs from the normal mean axes values. According to transmission electron microscopy estimations, the nuclear chromatin content was considered normal if it contained no more than one vacuole, which occupies <4% of the nuclear area (Bartoov et al., 1981
). To exclude from selection spermatozoa with abnormal chromatin content, a fixed, transparent, celluloid form of a vacuole with a borderline diameter of 0.78±0.18 µm was superimposed on the examined cell. In the cases with few small vacuoles, the estimation was made by eye. We must emphasize that spermatozoa with doubtful determination were excluded from selection.
When no spermatozoa with normal nuclei were available for microinjection, the second best morphologically evaluated sperm cells with minimally impaired nuclei were retrieved according to the hierarchy presented in Table II and Figure 1. Sperm cells exhibiting an abnormal but oval nuclear shape and a morphologically normal nuclear content were considered for the first choice selection (small or large oval nuclear forms, n=5 and n=2, respectively, Figure 1b and c, respectively). Spermatozoa with non-oval, abnormal nuclear shapes, but with normal nuclear content, were retrieved as a second choice (wide or narrow forms, n=2 and n=7, respectively, Figure 1d and e, respectively) and sperm cells with regional shape disorder (an extrusion or invagination of the nuclear mass, n=2, Figure 1f), as a third choice. Sperm cells exhibiting large nuclear vacuoles but normal oval nuclear shape (n=18, Figure 1g), were considered preferable for selection to those with a non-specific combined nuclear malformation (narrow forms + large nuclear vacuoles, n=2, Figure 1h). It should be noted that, just as in the conventional ICSI trial, sperm cells with pin, amorphous, tapered, round or multinucleated head shapes were totally excluded from retrieval.
|
|
As mentioned above, in the positive group only, spermatozoa with strictly defined morphologically normal nuclei were retrieved and injected into the oocytes. In the negative group, we also tried to inject all the oocytes within one cycle with spermatozoa exhibiting exactly the same nuclear aberration, i.e. all the oocytes of each specific couple were injected with similar sperm. This condition could be accomplished since each semen sample usually exhibited typical nuclear malformations. The distribution of the seven specific aberrations for each patient is demonstrated in Table II.
Sperm storage prior to microinjection
Post-selection, the sterile glass-bottomed dish containing the recipient selection droplet was transferred from the Male Fertility Laboratory in Bar-Ilan University to the IVF centre (30 min away) in an insulated box at
25°C. The average time between completion of sperm selection and microinjection was
2 h (range 14). About half an hour before microinjection, the dish was incubated at 37°C with an atmosphere of 5% CO2.
Preparation for IVF-IMSI
The preparation for the IVF-IMSI procedure was conducted according to the method described by Van Steirteghem et al. (1993).
Microinjection
The transferred, retrieved, cumulus-free ova were placed into drops of SPERM medium prepared in the same glass dish with the recipient droplet. As mentioned above, the latter contained the sperm cells morphologically selected for IMSI. The majority of these cells located in the recipient droplet were easily found by the embryologist at the usual magnification of 200400x. No shortage of selected spermatozoa was reported in any of the IMSI cycles included in this study. Microinjection of the selected spermatozoa into the oocytes was conducted in SPERM medium. Each microinjected oocyte was immediately transferred to a 4-well dish (Nunc), incubated in 0.5 ml of IVF or ISM 1 medium (Medi-Cult, New York, NY), covered with 0.5 ml of mineral oil (Medi-Cult) at 37°C with an atmosphere of 5% CO2.
The definitions and calculations of the fertilization rate, percentage of top embryos and implantation rate were described in detail earlier (Bartoov et al., 2003). The pregnancy rate was calculated per transfer. The abortion rate was calculated per clinical pregnancy occurrence.
Statistical analysis
All statistical analyses were performed using SPSS for Windows Version 11.0 (SPSS Inc., Chicago, IL). Data were presented as means±SD for continuous variables, including the numbers of retrieved and injected oocytes, fertilization rate, percentage of top embryos, number of transferred embryos and implantation rate. Comparisons between the positive and negative matched groups in the above variables were performed using univariate analysis of variance (ANOVA). The data for each cycle have been averaged prior to calculating the means. The discrete variables, including IMSI pregnancy and abortion rates, were presented as percentages. Comparison between the matched groups in the above variables was made by 2 tests.
Non-parametric statistical anslyses involving MannWhitney tests were performed in parallel to the described statistical analyses in order to avoid possible artefacts due to the high standard deviations of some sperm parameters. The non-parametric methods revealed results similar to those reported here.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Of the 38 IVF-IMSI cycles with compromised sperm selection included in the negative group, seven (18.4%) pregnancies were achieved (six singletons and one twin). Three of these pregnancies resulted in normal deliveries while four others resulted in a first trimester missed abortion (Table II). Of the three deliveries, two were achieved following ovum microinjection of sperm cells with large oval nuclei (choice 1, Table II) and the third occurred in a case of narrow formed sperm heads (choice 2). Three missed abortion cases occurred when ovum microinjection was conducted with spermatozoa exhibiting large nuclear vacuoles (choice 4, Table II). In another abortion case, the sperm cells exhibited a combined malformation: large vacuoles associated with narrow formed head shape (choice 5, Table II).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Indeed, the significantly higher fertilization rate, percentage of top embryos, implantation and pregnancy rates, and the lower abortion rate obtained following strict sperm selection, compared with the compromised selection, fit our previous findings. The fact that in the excluded mixed microinjection group (see Patients), the fertility and top embryo ratios obtained following injection of strictly selected spermatozoa were significantly higher than those obtained in cases of compromised selection has only strengthened these findings. The present results also confirm the results of studies by Tesarik et al. (2002) and De Vos et al. (2003)
who reported that sperm morphological quality influences fertilization, implantation and pregnancy rates following ICSI.
Our decision to use compromised selection criteria in cases where spermatozoa with strict morphologically normal nuclei could not be found was based on the following assumptions. First, it seems logical to suppose that spermatozoa with multiple combined (non-specific) nuclear malformations reflect extra-chromosomal damage compared with those with single specific abnormalities. Secondly, we assumed that within the category of specific morphological malformations, the existence of large vacuoles in the sperm nuclei indicates more damage to the nuclear DNA content and organization than nuclear shape or size impairment. This assumption is based on preliminary unpublished data, obtained in our laboratory, which indicate a significant negative relationship between the size of the nuclear vacuoles and chromatin stability assessed by the sperm chromatin structure assay (SCSA). Since, according to Larson et al. (2000), chromatin stability appears to influence embryonic development, we preferred, where possible, to retrieve sperm cells with a single shape/size nuclear deviation rather than those with a normal nuclear configuration, but with an impaired content. Of the specific nuclear defects with normal nuclear content, we chose to retrieve sperm cells with oval nuclear configuration (either large or small) rather than non-oval sperm cells. The results of Lee et al. (1996)
, who demonstrated that no increase in chromosome aberrations was found in spermatozoa with large or small heads, support our approach.
Indeed, all three pregnancies in the negative study group which resulted in normal deliveries were achieved when spermatozoa with a single specific nuclear shape malformation were injected into the ova (choices 1 and 2); two of them with large oval heads (choice 1). On the other hand, the four pregnancy cases in the negative study group which terminated in early missed abortions were achieved by microinjection of spermatozoa with large nuclear vacuoles.
The pregnancy rate of 18.4% obtained in the negative study group raises the question of sensitivity regarding the strict sperm selection criteria adapted for IMSI. However, because of the small number of IMSI cycles conducted with spermatozoa retrieved under compromised criteria and the low success rate in the negative group, we were still unable to draw any conclusions concerning the relationship between specific morphological malformations of the sperm nucleus and IMSI pregnancy outcome. Further investigation using compromised sperm selection may prove that spermatozoa with certain so-called morphological nuclear abnormalities are still capable of inducing pregnancy following IVF-IMSI. If this is the case, then, in order to reduce the false-negative value of the sperm selection, the preliminary criteria which we have adopted in our laboratory should be re-evaluated. It seems, therefore, that at this stage of the research, the elimination process used in the compromised sperm selection does not infer that one choice is inherently better than another.
Although the advantage of the modified IMSI treatment compared with the classic ICSI approach in pregnancy outcome is quite clear, some important issues have to be considered prior to introducing this method into a routine IVF laboratory: (i) a significant increase in the time taken to complete a case; (ii) the need for additional highly trained man power; and (iii) the additional cost of upgrading for the routine inverted microscope equipment, including high resolution Nomarsky optics (15 000) and a micromanipulator with two memories in space (
12 000). Taking into account these issues, a busy IVF centre may prefer to obtain sperm selection services from a separate andrological laboratory. Establishment of special andrological units which will serve several IVF centres seems to us the most practical way to introduce IMSI.
In conclusion, implantation and pregnancy by ICSI is associated with morphological nuclear normalcy of sperm. Sperm with a morphologically abnormal nucleus usually have low fertility potential, but some with certain nuclear abnormalities may still be able to produce pregnancy following ICSI.
The association between specific morphological nuclear abnormalities of the sperm microinjected into the oocyte and IMSI outcome should be investigated further.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bartoov B, Eltes F, Pansky M, Lederman H, Caspi E and Soffer Y (1993) Estimating fertility potential via semen analysis data. Hum Reprod 8, 6570.[Abstract]
Bartoov B, Eltes F, Pansky M, Langzam J, Reichart M and Soffer Y (1994) Improved diagnosis of male fertility potential via a combination of quantitative ultramorphology and routine semen analyses. Hum Reprod 9, 20692075.[Abstract]
Bartoov B, Berkovitz A and Eltes F (2001) Selection of spermatozoa with normal nuclei to improve the pregnancy rate with intracytoplasmic sperm injection. N Engl J Med 345, 10671068.
Bartoov B, Berkovitz A, Eltes F, Kogosowski A, Menezo Y and Barak Y (2002) Real-time fine morphology of motile human sperm cells is associated with IVF-ICSI outcome. J Androl 23, 18.
Bartoov B, Berkovitz A, Eltes F, Kogosovsky A, Yagoda A, Lederman H, Artzi S, Gross M and Barak Y (2003) Pregnancy rates are higher with intracytoplasmic morphologically selected sperm injection than with conventional intracytoplasmic injection. Fertil Steril 80, 14131419.[CrossRef][ISI][Medline]
Berkovitz A, Eltes F, Soffer Y, Zabludovsky N, Beyth Y, Farhi J, Levran D and Bartoov B (1999) ART success and in vivo sperm cell selection depend on the ultramorphological status of spermatozoa. Andrologia 31, 18.[CrossRef][Medline]
Chemes EH and Rawe YV (2003) Sperm pathology: a step beyond descriptive morphology. Origin, characterization and fertility potential of abnormal sperm phenotypes in infertile men. Hum Reprod Update 9, 405428.
De Vos A, Van De Velde H, Joris H, Verheyen G, Devroey P and Van Steirteghem A (2003) Influence of individual sperm morphology on fertilisation, embryo morphology, and pregnancy outcome of intracytoplasmic sperm injection. Fertil Steril 79, 4248.[CrossRef][ISI][Medline]
Kruger TF, Acosta AA, Simmons KF, Swanson JR, Matta JF and Oehninger S (1988) Predictive value of sperm morphology in in vitro fertilisation. Fertil Steril 49, 112121.[ISI][Medline]
Larson KL, DeJonge CJ, Barnes AM, Jost LK and Evenson DP (2000) Sperm chromatin structure assay parameters as predictors of failed pregnancy following assisted reproductive techniques. Hum Reprod 15, 17171722.
Lee JD, Kamiguchi Y and Yanagimachi R (1996) Analysis of chromosome constitution of human spermatozoa with normal and aberrant head morphologies after injection into mouse oocytes. Hum Reprod 11, 19421946.[Abstract]
Liu DY and Baker HW (1992) Tests of human sperm function and fertilization in vitro. Fertil Steril 58, 465483.[ISI][Medline]
Mashiach R, Fisch B, Eltes F, Tadir Y, Ovadia J and Bartoov B (1992) The relationship between sperm ultrastructural features and fertilisation capacity in vitro. Fertil Steril 57, 10521057.[ISI][Medline]
Tesarik J, Mendoza C and Greco E (2002) Paternal effects acting during the first cell cycle of human preimplantation development after ICSI. Hum Reprod 17, 184189.
Van Steirteghem AC, Nagy Z, Joris H, Liu J, Staessen C, Smitz J, Wisanto A and Devroey P (1993) High fertilisation and implantation rates after intracytoplasmic sperm injection. Hum Reprod 8, 10611066.[Abstract]
Submitted on February 5, 2004; resubmitted on June 24, 2004; accepted on September 10, 2004.
|