FERTILITY AND STERILITY Vol. 64, No.4, October 199 Copyright (0 199 American Society for Reproductive Medicine Printed on acid-free paper in U. S. A Fertilization rates using intracytoplasmic sperm injection are greater than subzonal insemination but are dependent on prior treatment of sperm James Catt, Ph.D.* John Ryan, Ph.D. Ian Pike, Ph.D. Chris O'Neill, Ph.D. Human Reproduction Unit, Royal North Shore Hospital, St. Leonards, New South Wales, Australia Objective: To determine a suitable method of sperm preparation for use in micromanipulation. To compare the fertilization rates of sibling oocytes inseminated by intracytoplasmic sperm injection (ICSI) and subzonal insemination (SUZI). Design: Two methods of gamete micromanipulation to overcome male factor infertility were compared. Within this study, three trials were conducted to determine the most suitable method of sperm preparation. This method then was used to complete the study. Setting: Procedures were performed in a teaching hospital research environment. Patients: Ninety-six patients were recruited for this study, undergoing 99 stimulation cycles. Interventions: Sibling oocytes were inseminated by sub zonal sperm or intracytoplasmic injection. Main Outcome Measures: Fertilization, zygote development, and pregnancy rates. Results: In the absence of manipulative pretreatment of sperm (trial 1), there was no difference in normal fertilization rates between ICSI and SUZI (19% and 2%, respectively). In the second trial there again was no sperm pretreatment for SUZI but, for ICSI, a polyvinylpyrrolidone (PVP) solution was used to reduce sperm velocity and the sperm tails were incised before injection. The fertilization rates were significantly different between ICSI (44%) and SUZI (17%). In the third trial, PVP was added to the sperm used for both types of insemination and the sperm tails also were incised for the ICSI insemination. Fertilization was again significantly different (16% for SUZI and 44% for ICSI). The trial 2 method of sperm preparation then was used to complete the study (trial 4) and confirmed the results of trial 2 (21% for SUZI and 42% for ICSI). The results suggest that ICSI can give improved fertilization compared with SUZI provided the sperm are treated before injection. No significant differences were found in the development rates of zygotes suitable for transfer or cryopreservation between the two micromanipulation methods. Fertil Steril 199;64:764-9 Key Words: Male infertility, IVF, fertilization, micromanipulation, intracytoplasmic sperm injection (ICSI), subzonal insemination (SUZI) Micromanipulation of sperm for treatment of male factor infertility has been used for several years and has been adopted as a clinically useful procedure. The most widespread method to date has been the subzonal sperm insertion method (SUZI). There are, however, several drawbacks to this procedure. The Received August 4, 1994; revised and accepted April 20, 199. * Reprint requests and present address: James Catt, Ph.D., North Shore ART, Clinic 20, Royal North Shore Hospital, St. Leonards, New South Wales 206, Australia (FAX: 61-2-906-1872). overall fertilization rates have been 30% to 40% but a considerable proportion (20% to 30%) of observed fertilized oocytes have been polyspermic. In addition, approximately 30% to 40% of patients do not achieve fertilization of at least one oocyte (1-4). A technique for injecting a sperm directly into the cytoplasm (intracytoplasmic sperm injection [ICSm has been developed in an attempt to overcome the disadvantages of SUZI. Intracytoplasmic sperm injection has been suggested as a means of obtaining fertilization for a number of years (,6) and recently the technique has been shown to be clinically useful 764 Catt et al. Improved fertilization with ICSI Fertility and Sterility
for treatment of male factor infertility (7-10). The present study compares the fertilization rates of sibling oocytes when inseminated using either the SUZI or ICSI technique and assesses the necessity of micromanipulative pretreatment of sperm for successful fertilization. Patient Selection MATERIALS AND METHODS Patients were invited to participate in this study if they had been accepted for the micromanipulation program at the Royal North Shore Hospital (4). The patients could be separated into three groups. Group A patients had a history of failed fertilization with conventional IVF (with a minimum of 10 oocytes) and group B patients had a history of low fertilization with conventional IVF ( < 1% with a minimum of six oocytes). Group C patients had sperm parameters «1 X 10 6 motile, morphologically normal, by the World Health Organization criteria [11]), which our clinical experience has shown to have a poor prognosis for IVF. There were 22 treatment cycles in group A, 14 in group B, and 60 in group C and they were distributed evenly throughout the course of the trial. Stimulation and Recovery All patients were given individualized stimulation protocols and follicular growth monitored as reported previously (4). Cumulus-oocyte complexes were recovered by transvaginal aspiration and kept in human tubal fluid (HTF) medium (12) supplemented with 10% (vol/vol) homologous serum until required. The recovered cumulus-oocyte complexes were treated with 300 IU/mL hyaluronidase (type VIII; Sigma Chemical Co., St. Louis, MO) and aspirated through narrow pipettes to remove the cumulus cells. The denuded oocytes were kept in HTF + 10% serum in an incubator (% CO 2 in air) until manipulated. A random number list was generated and patients, as they commenced oocyte retrieval, were assigned the next number on the list. If the assigned number was even then the first oocyte recovered was treated by ICSI, the second by SUZI and so on, in sequential order. If the random number was odd then the first oocyte recovered was treated by SUZI and the second by ICSI and so on. For this study, immature oocytes (at the time of injection) also were included and injected. Immature oocytes were defined as those either having a germinal vesicle or not having an observable first polar body (presumptive metaphase I oocytes). The immature oocytes were injected after injection of the metaphase II oocytes. Vol. 64, No.4, October 199 Sperm Preparation After liquefaction, sperm were separated on a discontinuous Percoll density gradient. Sperm were harvested from the 83% Percoll fraction and washed twice by centrifugation. The final pellet was resuspended in a minimum volume of HTF and underlayed beneath a further 0.2 ml of HTF. The sperm were allowed to swim-up for 2 hours (4). If the sperm concentration after this swim-up was < x 10 /ml, then the supernatant was centrifuged (3,000 X g for 1 minute) and the pellet was resuspended in 10 I1L ofhtf. Polyvinylpyrrolidone (PVP; mol wt 360,000; Sigma Chemical Co.) was used to retard sperm motility (13). A 20% (wt/vol) PVP solution was dialyzed extensively against water and the osmolarity was adjusted to 28 ± mosm/kg using a lox concentrated stock of HTF and the concentration of PVP adjusted to 8% (wtlvol) with HTF. The sperm suspension was diluted 1:1 with this PVP. Sperm for the four trials received the following pretreatments. Trial 1, no sperm pretreatment for both SUZI and ICSI; trial 2, no sperm pretreatment for SUZI and for ICSI the sperm was resuspended in PVP and individual sperm immobilized by manipulation immediately before injection; trial 3, sperm were resuspended in PVP for SUZI and the same treatment as trial 2 was used for ICSI; trial 4 treatments were the same as for trial 2. Micromanipulation Micromanipulation was conducted using essentially the same equipment as reported previously (4). A Nikon Diaphot inverted microscope (Nikon Instrument Co., Tokyo, Japan) equipped with Hoffman Modulation Contrast Optics (Modulation Optics Inc., Greenvale, NY) was used for all manipulations. Intracytoplasmic sperm injection pipettes were made from thin walled capillary glass, without a microfilament (1 mm outer diameter X 0.78 mm inner diameter; cat. no. GC100T - 1; Clark Electromedical Instruments, Pangbourne, Surrey, United Kingdom) using a Sutter P-87 (Sutter Instrument Co., Novoto, CA) pipette puller. The two-stage program pulled shallow taper pipettes and these were cut to the correct diameter (7 to 8 mm outer diameter, to 6 mm inner diameter) using a scalpel blade under a dissecting microscope. The pipettes were bevelled to 3 to 40 using a wet grinding system (N arishige EG- 9; Narishige Instrument Co., Tokyo, Japan) and the tips were washed by repeated aspiration in boiling water. A 4 bend was introduced to enable manipulation. Holding pipettes were made as reported previously (4). The pipettes were dry heat sterilized (minimum of 4 hours, 10 C) before use. Catt et al. Improved fertilization with ICSI 76
Both the holding and injection pipettes were attached to Hamilton 0.-mL screw threaded plunger syringes (Hamilton, Reno, NV) via vinyl tubing that had been filled previously with light paraffin oil to provide the necessary manipulating pressures. Both types of manipulation were carried out in droplets (10 to 20 ml) of HTF under paraffin oil (heavy grade; BDH Ltd., Poole, Dorset, United Kingdom). The paraffin oil had been pre-equilibrated for ~ days by incubating 40 ml over 10 ml ofhtf, in the CO 2 incubator. For the SUZI technique, 1 to 20 sperm were aspirated into the injection pipette and the microtools were moved to another droplet containing the oocyte. The sperm were introduced into the perivitelline space, adjacent to the polar body, the oocyte having been orientated so that the polar body was between the 9 and 12 o'clock position. For the ICSI procedure, a spermatozoon was incised, aspirated, and moved to the tip of the pipette. The incision of the sperm was carried out by lowering the injection pipette onto the sperm tail and swiping across it, causing it to bend at right angles to the longitudinal axis. This invariably immobilized the sperm, enabling its aspiration into the injection pipette. The oocyte was held using the holding pipette so that the polar body was orientated at 12 o'clock or on the great circle from 12 o'clock. This was designed so that the introduction of the pipette would be removed sufficiently from the metaphase II spindle so that the chances of affecting that spindle would be minimal. To ensure cytoplasmic injection, back pressure was used to aspirate the oocyte cytoplasm into the injection pipette until the cytoplasm moved freely past the sperm. This was found to be fundamental for successful injection. Intact cytoplasm could often be drawn > 100 11m into the pipette before rupturing the plasmalemma. A SUZI insertion usually took <2 minutes and an ICSI took < 1 minute. Up to four oocytes were treated at the same time and then were returned to culture. Fertilization Assessment and Zygote Culture Injected oocytes were examined for second polar body abstriction and pronuclear formation using the inverted microscope described above. Oocytes were examined at 16 to 18 hours after insemination and again at 24 to 26 hours. Zygotes arising from the two types of insemination were kept separate. After a further 24 hours of culture in HTF supplemented with platelet activating factor (14), cleaved embryos either were replaced by uterine transfer or were cryopreserved. A maximum of two embryos were replaced in anyone transfer, unless the patient had a 766 Catt et a1. Improved fertilization with ICSI 0 3112! b 0/113 71177 Trial! (26 cycles) Trial 2 (24 cycles) Trial 3 (20 cycles) Trial 4 (29 cycles) Figure 1 Fertilization results of sibling oocytes inseminated by SUZI (D) or ICSI ([j ). Columns with the same superscripts are not significantly different. history of implantation failure with at least three embryo transfers or the patient was older than 3 years. In these cases a maximum of three embryos were transferred. For anyone transfer, only embryos arising from one micromanipulation treatment type were transferred. The first oocyte recovered was treated with ICSI or SUZI according to the random number list, as described above. This generated a list of treated oocytes that were examined for fertilization in the same sequence. The first oocyte that was observed to be fertilized determined the type of embryos to be transferred, i.e., if the first oocyte fertilized was derived from ICSI then the embryos for transfer were all ICSI derived. Alternatively, if the first oocyte to be fertilized was derived from SUZI then all the embryos for transfer were from SUZI. This ensured the randomness of the procedure, thus allowing comparison of unbiased ET data. Results were compared by paired X 2 test. Patient Monitoring This was essentially as described previously (4). In the absence of menses 18 days after ET, a urinary.8-hcg pregnancy test was conducted. An ultrasound examination was conducted 3 weeks after a positive.8-hcg test, to confirm the presence and number of fetal hearts. RESULTS The fertilization data for the four trials are summarized in Figure 1. All oocytes treated were included and no account was taken of maturity or damage during the procedures. Sperm that had no pretreatment (trial 1) gave similar fertilization rates for both ICSI and SUZI (19% and 2%, P > 0.0, respectively). For the ICSI procedure in trial 2 the sperm was treated with PVP to reduce motility and Fertility and Sterility
Table 1 Outcome of Inseminated Sibling Oocytes Table 2 Outcome of Immature Oocyte Injection SUZI Intracytoplasmic sperm injection SUZI Intracytoplasmic sperm injection No. of patients No. of patient cycles No. of oocytes recovered and inseminated No. of mature oocytes No. of total oocytes degenerated No. of mature oocytes degenerated No. of oocytes normally fertilized No. of abnormally fertilized Haploid Multinucleate No. of embryos transferred or frozen 96 99 4 481 (88)* 34 (6) 24 () 110 (20) 67 99 (90) * Values in parentheses are percentages. 96 99 60 08 (91) 136 (24) 112 (22) 206 (37) 1 8 19 (9) the tail was incised before injection. This gave a significant increase in fertilization for lcsl compared with SUZl (44% and 17%, P < 0.01). Treatment of sperm with PVP before SUZl (trial 3) did not increase the fertilization rate compared with either lcsl of the sibling oocytes (16% and 44%, respectively) or from the SUZl results of trials 1 and 2. On the basis of these results we adopted the sperm pretreatment performed in trial 2 to complete the rest of the study (trial 4). From the 29 cycles in trial 4, the fertilization rate for SUZl was 21% and 42% for lcs!. The results of trial 2 and 4 were not different for both SUZl and lcsl (P > 0.0), giving the combined fertilization rate of 19% (4/278) for SUZl and 43% (128/298) for lcs!. Table 1 shows the outcome of oocytes inseminated by the SUZl and lcsl techniques for the complete study. Even though these results include those from trial 1 in which the fertilization rate was not different between SUZl and lcsl, there were overall twice as many embryos generated by the lcsl procedure (P < 0.01). The number of multinucleate zygotes was greater (P < 0.01) for SUZl (12%, 67/4) than for lcsl (1.4%, 8/60). Oocytes assessed as having one pronucleus were less prevalent (P < 0.0) with SUZl (1 %, /4) than with lcsl (3%, 1/60). Table 2 shows the outcome ofinseminatingimmature oocytes. Some presumptive metaphase I oocytes were fertilized by both procedures, but the low numbers involved precluded any meaningful analysis. No fertilization was achieved inseminating germinal vesicle stage oocytes. The lcsl procedure caused a higher degeneration rate than SUZl (P < 0.01,47% and 17%, respectively) with these immature oocytes. Degeneration of presumptive mature metaphase II oocytes after micromanipulation (overt lysis, cytoplasmic tanning, or shrinkage) was monitored immediately after insemination, at the time offertiliza- Vol. 64, No.4, October 199 Germinal vesicle No. injected No. degenerate No. fertilized Metaphase I No. injected No. degenerate No. fertilized 2 9 (17)* o 12 1 (8) 2 (17) * Values in parentheses are percentages. 47 22 (47) o 2 (40) 2 (40) tion assessment and at approximately 48 hours postinsemination. Degeneration was found to be more prevalent (P < 0.01) with lcsl (22%, 112/08) than with SUZl (%, 24/481). Both insemination techniques had a minimal effect on early cleavage (>90% of fertilized ova cleaved) to embryos suitable for transfer or cryopreservation). The results of fresh and cryopreserved ET procedures are shown in Table 3. Two pregnancies resulted from 24 ET procedures with 39 fresh SUZlderived embryos (8% per ET; % fetal hearts per embryo transferred). Nine pregnancies resulted from 3 ETs conducted with 89 fresh lcsl-derived embryos (17% per ET; 10% fetal hearts per embryo transferred). Three pregnancies have arisen from 22 frozen ET cycles with 4 SUZl-derived embryos (14% per ET; % fetal hearts per embryo transferred). One pregnancy resulted from 36 frozen ET cycles with 71 lcsl-derived embryos (3% per ET; 1 % fetal hearts per embryo transferred). The numbers of pregnancies and fetal hearts were not sufficient to give a meaningful statistical analysis. All pregnancies were singleton and all have gone to term. There were eight female and seven male babies born, one of the female babies had a pansystolic heart murmur the rest are normal. ' Table 3 Embryo Transfers Fresh No.ofETs No. of embryos transferred No. of pregnancies Fetal hearts per embryo transferred (%) Frozen No.ofETs No. of embryos transferred No. of pregnancies Fetal hearts per embryo transferred (%) SUZI 24 39 2 (8)* 22 4 3 (14) * Values in parentheses are percent per ET. Intracytoplasmic sperm injection 3 89 9 (17) 10 36 71 1 (3) Catt et al. Improved fertilization with ICSI 767 1
; I The lower incidence of achieving ET for SUZI compared with lesl was a reflection of the lower overall fertilization rate with this procedure (110/4, 20% for SUZI and 206/08, 37% for lesi) and the lower incidence of at least one oocyte being fertilized by the procedure (47/99 patient cycles for SUZI and 701 99 patient cycles for lesi). DISCUSSION The results demonstrate fundamental differences between the two types of manipulation. Subzonal insemination is prone to polyspermia because multiple sperm have to be placed under the zona pellucida to give an acceptable fertilization rate (4) and this allows the possibility of multiple sperm fusion. Only one sperm is injected with the lesl procedure and so this eliminates polyspermia, although some zygotes were seen as multinucleate. These multinucleate zygotes probably were derived either from second polar body retention or decondensing chromatin from a disrupted metaphase spindle. Intracytoplasmic sperm injection is a far more invasive technique than SUZI and consequently the number of damaged oocytes was greater. This damage can take the form of activation resulting in haploid development or degeneration of the oocyte either immediately or over a period of time. Intracytoplasmic sperm injection resulted in a greater proportion of metaphase II oocytes degenerating than the SUZI procedure. However, the increase in fertilization with lesl more than compensates for the increased damage. As techniques and skill with lesl improve one would expect the damage rate to decrease. The results show that germinal vesicle oocytes are not suitable for microinjection. Attempting to fertilize immature oocytes is not without precedent as germinal vesicle stage oocytes can be penetrated by sperm in conventional IVF (1). Pronucleus formation and subsequent cleavage is delayed until the oocyte matures to metaphase II. In our hands the plasma membrane of germinal vesicle stage oocytes was fragile, easily disrupting on application of back pressure and the oocyte often lysed immediately on withdrawal of the pipette. Subzonal insemination, although less damaging to the oocyte, also gave no fertilization of germinal vesicle oocytes. The results of this study confirm previous observations (7-10) that lesl may result in higher fertilization rates than SUZI. It also shows for the first time that this improved success rate is dependent upon the treatment that sperm receive before injection. This prior treatment involved the use of a high viscosity media (containing PVP), which had the technical advantage of slowing the swimming speed of sperm; and the manipulation of sperm by mechani- cal stroking of the tail with a glass microinjection pipette, rendering the sperm immotile. In the absence of these procedures, lesl was not more successful than SUZI, whereas treatment with PVP of sperm destined for SUZI was of no extra benefit. The combined results of the trials using both PVP and tail incision showed that lesl was a suitable treatment for a range of male factor etiologies. Good fertilization rates were achieved with few couples failing to achieve fertilization of at least one oocyte. Some of the resulting zygotes were developmentally competent. More studies are needed to confirm whether the zygotes are competent and able to survive cryopreservation as those generated by other procedures. In this study, sperm for the lesl procedure were not treated with either PVP or incision separately, thus it was not possible to distinguish between the potential benefits of these treatments. The manner by which these pretreatments improved the fertilizing ability of sperm after lesl is not clear. The reduced motility due to the high viscosity PVP medium may allow "better quality" sperm to be selected, although it is not clear what this means. Palermo (8) found no correlation between various sperm parameters and the fertilization rate with lesi, although a suggestion of a negative correlation of poor sperm with implantation rate was made. The incising of the sperm by the injection pipette induces immotility. This may have the effect of allowing the sperm to be injected into the cytoplasm with a consistently lower volume of carrier media than may occur with more motile sperm. Alternatively, motile sperm may disrupt the internal architecture of the oocyte, even though motility generally ceases soon after injection. A further possibility may be that the manipulation of the sperm causes physical damage to the sperm plasma membrane. This in turn may facilitate the exposure of the sperm nucleus or other internal components to the oocyte cytoplasm, prompting decondensation and the other events of normal fertilization. These propositions are speculative and require experimental investigation. The ethical and logistical limitations of experimentation with human gametes and zygotes suggest that an appropriate animal model will facilitate future development. Successful fertilization has been achieved with murine lesi, although the oocytes appear to have a high susceptibility to lysis at injection (16) and the size of the sperm relative to the oocyte is large compared with human or domestic species gametes. A recent report (17) showed success in achieving fertilization with domestic species oocytes by lesi, suggesting its use as an animal model. The use of these models will allow a more fundamental understanding of the fate of sperm after injection, 768 Catt et al. Improved fertilization with ICSI Fertility and Sterility
the mechanism by which the activation of the normal developmental program is achieved and the causes offailed fertilization. This information will allow the new techniques to be developed to their full clinical potential. REFERENCES 1. Ng S-C, Bongso A, Ratnam SS. Microinjection of human 00- cytes: a technique for severe oligoasthenoteratozoospermia. Fertil Steril 1991;6:1117-23. 2. Fishel S, Antinori S, Jackson P, Johnson J, Rinaldi L. Presentation of six pregnancies established by sub zonal insemination (SUZI). Hum Reprod 1991;6:124-30. 3. Cohen J, Malter HE, Talansky BE, Grifo J, editors. Micromanipulation of human gametes and embryos. New York: Raven Press, 1991. 4. Catt JW, Krzyminska U, Tilia L, Csehi E, Ryan J, Pike I, et al. Subzonal insertion of multiple sperm is a treatment for male factor infertility. Fertil Steril1993;61:118-24.. Lanzendorf SE, Maloney MK, Veeck LL, Slusser J, Hodgen GD, Rosenwaks Z. A preclinical evaluation of pronuclear formation by microinjection of human spermatozoa into human oocytes. Fertil Steril 1988;49:83-42. 6. Sathananthan AH, Ng S-C, Trounson A, Bongso A, Laws King A, Ratnam SS. Human micro-insemination by injection of single or multiple sperm: ultrastructure. Hum Reprod 1989; 4:74-83. 7. Palermo G, Joris H, Devroey P, Van Steirteghem A. Pregnancies after intra-cytoplasmic injection of single spermatozoon into an oocyte. Lancet 1992;340:17-8. 8. Palermo G, Joris H, Derde M-P, Camus M, Devroey P, Van Steirteghem AC. Sperm characteristics and outcome of human assisted fertilization by subzonal insemination and intracytoplasmic sperm injection. Fertil Steril1993; 9:826-3. 9. Van Steirteghem AC, Liu J, Joris H, Nagy Z, Janssenwillen C, Tournaye H, et al. Higher success rate by intracytoplasmic sperm injection than that by subzonal insemination. Report of a second series of 300 consecutive treatment cycles. Hum Reprod 1993;8:10-60. 10. Van Steirteghem AC, Nagy Z, Joris H, Liu J, Staessen C, Smitz J, et al. High fertilization and implantation after intracytoplasmic sperm injection. Hum Reprod 1993;8:1061-6. 11. World Health Organization. WHO laboratory manual for examination of human semen and semen-cervical mucus interaction. 2nd ed. Cambridge: The Press Syndicate of the University of Cambridge, 1987:-20. 12. Quinn P, Warnes GM, Kerin JF, Kirby C. Culture factors affecting the success rate of IVF and embryo transfer. Ann NY Acad Sci 198;442:19-9. 13. Thadani VM. A study of hetero-specific sperm-egg interactions in the rat, mouse and deer mouse using in vitro fertilization and sperm injection. J Exp Zool 1980;212:43-3. 14. O'Neill C, Collier M, Ammit AJ, Ryan JP, Saunders DM, Pike IL. Supplementation of in vitro fertilization culture medium with platelet activating factor. Lancet 1989;2:769-72. 1. Van Blerkom J, Davis PW, Merriam J. The developmental ability of human oocytes penetrated at the germinal vesicle stage after insemination in vitro. Hum Reprod 1994;9:697-708. 16. Markert CL. Fertilization of mammalian eggs by sperm injection. J Exp Zool 1983;228:19-201. 17. Catt JW, Rhodes SL. Comparative intracytoplasmic sperm injection in human and domestic species. Reprod Fertil Dev 199;7:161-7. Vol. 64, No.4, October 199 Catt et at. Improved fertilization with ICSI 769