Advertisement

Intracytoplasmic sperm injection increased fertilization and good-quality embryo formation in patients with non–male factor indications for in vitro fertilization: a prospective randomized study1

      Abstract

      Objective: To compare the fertilization rate and formation of good-quality embryos with conventional IVF and ICSI in patients with non–male factor infertility.
      Design: Prospective controlled study.
      Setting: Infertility clinic.
      Patient(s): Thirty-five patients with non–male factor infertility.
      Intervention(s): Retrieved sibling oocytes were randomly assigned to conventional IVF or ICSI. Of sibling oocytes assigned to ICSI, only metaphase II oocytes were injected with sperm.
      Main Outcome Measure(s): Fertilization rate and formation of good-quality embryos per retrieved oocyte.
      Result(s): Per retrieved oocyte, ICSI resulted in better fertilization rate compared with conventional IVF (71.3% [134 of 188] vs. 57.2% [107 of 187]). Per retrieved oocyte, ICSI also resulted in better formation of good-quality embryos at 48 hours after retrieval compared with conventional IVF (64.4% [121 of 188] vs. 47.1% [88 of 187]).
      Conclusion(s): In IVF patients with non–male factor infertility, subjecting some sibling oocytes to ICSI increased the fertilization rate and formation of good-quality embryos per retrieved oocyte. It also avoided the problem of total fertilization failure in almost all cases.

      Keywords

      In couples with male factor infertility, the woman usually undergoes oocyte retrieval followed by injection of good-quality sperm into the cytoplasm of the oocyte. An acceptable pregnancy rate can now be achieved in such couples through IVF-ICSI. A review of this topic was published in 1998 (
      • Khamsi F.
      • Lacanna I.
      • Endman M.
      • Wong J.
      Recent advances in assisted reproductive technologies.
      ).
      Another group of infertile couples have benefited from this technology. Ziebe et al. (
      • Ziebe S.
      • Andersen A.N.
      • Andersen A.G.
      • Mikkelsen A.L.
      • Lindenberg S.
      Results of intracytoplasmic sperm injection in relation to indication.
      ) treated couples with non–male factor infertility who had no or low fertilization in a previous IVF cycle. When IVF-ICSI was used in these couples, much improved fertilization was noted and a pregnancy rate of 29% per embryo transfer was obtained. Aboulghar et al. (
      • Aboulghar M.A.
      • Mansour R.T.
      • Serour G.I.
      • Sattar M.A.
      • Amin Y.M.
      Intracytoplasmic sperm injection and conventional in vitro fertilization for sibling oocytes in cases of unexplained infertility and borderline semen.
      ) and Ruiz et al. (
      • Ruiz A.
      • Remohi J.
      • Minguez Y.
      • Guanes P.P.
      • Simon C.
      • Pellicer A.
      The role of in vitro fertilization and intracytoplasmic sperm injection in couples with unexplained infertility after failed intrauterine insemination.
      ) obtained similar results when they performed ICSI on some of the oocytes retrieved from couples with infertility of undetermined cause. They randomly subjected half of the retrieved oocytes to ICSI and the remaining half to conventional IVF insemination. They reported that 11.4% and 23%, respectively, of their patients had no fertilization of their oocytes with standard IVF insemination, whereas the sibling oocytes were fertilized by ICSI.
      Both groups of investigators believed that patients in IVF programs should be informed that a certain percentage of these patients will have no fertilization with standard IVF, whereas performance of ICSI on some of the retrieved oocytes will ensure some fertilization in all patients with infertility of undetermined cause.
      The reports reviewed above evaluated the effectiveness of ICSI in patients with infertility of undetermined cause or in patients with no or low fertilization in a previous IVF cycle. With such information available, we thought it timely to conduct a randomized prospective study in all IVF patients, regardless of cause of infertility. We subjected half of their sibling oocytes to ICSI and the remaining half to standard IVF insemination. The exception was couples with male-factor infertility, for whom all retrieved oocytes would undergo ICSI. We sought to determine whether ICSI would result in better fertilization rate and formation of good-quality day-2 embryos than standard IVF in patients with infertility of various causes.

      Materials and methods

       Patients

      Thirty-five women who were enrolled in our IVF program agreed to have half of their retrieved sibling oocytes undergo conventional IVF insemination and the remaining half to ICSI in a randomized manner. Their ages ranged from 21 to 44 years, with a mean (±SD) of 33.9 ± 5.46 and a median of 33. Causes of infertility were tubal (n = 18), unknown (n = 12), and endometriosis (n = 2). One patient had oocyte donation, and four patients had therapeutic donor insemination. The duration of infertility ranged from 6 months to 18 years, with an average of 5.2 ± 4.0 years. Twenty-nine patients had primary infertility and six had secondary infertility. The husbands’ sperm had a volume of 3.10 ± 1.54 mL, motility of 52.50% ± 12.87%, and count of 78.40 ± 37.48 × 106/mL; they had 129.0 ± 103.4 × 106 total motile sperm. Patients with poor sperm progression were offered full ICSI and were not included in this group. We also excluded patients with male factor infertility.
      The study was approved by our institutional review board, and all participants gave written informed consent. The possibility of performing ICSI on half of the retrieved oocytes was discussed with infertile couples, and some who wished to undergo this procedure are included in this study. Our patients also underwent counseling with one of the authors (M.E.).

       Semen analysis

      At the original consultation, semen analysis was performed to rule out male factor infertility. Analysis was done according to recommendations of the World Health Organization (WHO) (

      WHO laboratory manual for examination of human semen and sperm. 3rd ed. New York: Cambridge University Press, 1992:44.

      ) for count, motility (including forward progression), and morphology. If any of these variables were below the recommended cut-off value, patients were considered to have male factor infertility.
      On the day of IVF, count and motility were assessed again. The values obtained at this assessment may have varied from those in the original assessment, but the new values were not used as the basis for diagnosis of male factor infertility. The latter data are shown in Table 1.
      TABLE 1Patient profiles, semen characteristics, and pregnancy results: HALF ICI VS. IVF STUDY 1999 (first or second half of wells).
      TDI = therapeutic donor insemination.
      ,
      Khamsi. Intracytoplasmic sperm injection. Fertil Steril 2001.
      PatientAge (y)Cause of infertilityFertilization rate (%)Good embryos (%)SpermPregnancy
      IVF (per assigned oocyte)ICSI (per assigned oocyte)ICSI (per injected oocyte)IVF (per assigned oocyte)ICSI (per assigned oocyte)Motility (%)Count (M/mL)Total with motility (M)ChemicalNo. of gestational sacsNo. of heart beatsOngoing
      129Unknown4/4 (100)4/4 (100)4/4 (100)4/4 (100)4/4 (100)62128277.76Yes11Yes
      230Unknown9/12 (75)4/9 (44)4/9 (44)6/12 (50)4/9 (44)4051.357.46Yes11Yes
      328Unknown3/4 (75)4/4 (100)4/4 (100)2/4 (50)4/4 (100)4030.548.80No
      435Tubal5/6 (83)5/6 (83)5/6 (83)3/6 (60)4/6 (67)5571.5235.95No
      532Tubal2/13 (15)4/6 (67)4/4 (100)2/13 (15)4/6 (67)356344.10No
      632Unknown6/11 (55)7/11 (64)7/9 (78)2/11 (18)2/11 (18)50108216.00No
      731Tubal7/8 (88)7/7 (100)7/7 (100)6/8 (75)7/7 (100)4510633.39No
      833Tubal3/4 (75)3/5 (60)3/3 (100)3/4 (75)3/5 (60)5092101.20Yes11Yes
      933Unknown3/3 (100)5/6 (83)5/5 (100)3/3 (100)4/6 (67)5039.535.55Yes33Yes
      1036Tubal0/6 (0)4/8 (50)4/5 (80)0/6 (0)4/8 (50)6539177.45Yes11Yes
      1138Unknown1/3 (33)4/4 (100)4/4 (100)1/3 (33)4/4 (100)5075112.50No
      1234Tubal1/3 (33)2/3 (67)2/3 (67)1/3 (33)2/3 (67)555590.75No
      1329Tubal6/7 (86)6/8 (75)6/6 (100)6/7 (86)6/8 (75)7075.6238.14No
      1444Unknown2/2 (100)3/3 (100)3/3 (100)2/2 (100)3/3 (100)88209367.84No
      1532Tubal4/4 (100)2/4 (50)2/4 (50)4/4 (100)2/4 (50)5585.5141.08No
      1621Endomtrial4/5 (80)2/4 (50)2/2 (100)4/5 (80)2/4 (50)6089240.30No
      1736Unknown0/5 (0)5/8 (63)5/8 (63)0/5 (0)5/8 (63)552781.68No
      1826Tubal6/7 (86)5/7 (71)5/5 (100)3/7 (43)3/7 (43)703598.00No
      1936Tubal4/6 (67)5/5 (100)5/5 (100)2/6 (33)5/5 (100)55136149.60Yes10No
      2040Unknown2/4 (50)3/4 (75)3/4 (75)1/4 (25)3/4 (75)4538.560.64No
      2138Tubal2/3 (67)2/3 (67)2/3 (67)2/3 (67)2/3 (67)48103123.60No
      2242Tubal6/8 (75)6/8 (75)6/8 (75)5/8 (63)5/8 (63)6072.587.00No
      2332Tubal4/4 (100)2/4 (50)2/4 (50)4/4 (100)2/4 (50)303825.08Yes22Yes
      2438Unknown1/16 (17)5/6 (83)5/5 (100)1/6 (17)5/6 (83)338872.60No
      2542Ovarian failure, oocyte donation0/4 (0)3/4 (75)3/3 (100)0/4 (0)3/4 (75)4561.555.35No
      2639Unknown, TDI5/5 (100)4/5 (80)4/4 (100)5/5 (100)2/5 (40)307112.78No
      2740Tubal0/2 (0)0/1 (0)0/1 (0)0/2 (0)0/1 (0)457349.28No transfer
      2831TDI1/6 (17)5/6 (83)5/6 (83)1/6 (17)4/6 (67)3011019.80No
      2931Tubal4/4 (100)2/4 (50)2/2 (100)4/4 (100)2/4 (50)506548.75No
      3043Tubal2/3 (67)1/3 (33)1/2 (50)2/3 (67)1/3 (33)7081141.75No
      3132Endometrial1/1 (100)1/2 (50)1/2 (50)1/1 (100)1/2 (50)4010584.00No
      3224Tubal, TDI1/10 (10)8/12 (67)8/12 (67)0/10 (0)8/12 (67)409154.6Yes11Yes
      3330TDI2/6 (33)5/6 (83)5/5 (100)2/6 (33)5/6 (83)385734.2No
      3440Tubal0/1 (0)1/1 (100)1/1 (100)0/1 (0)1/1 (100)70103468.65No
      3531Unknown (1 OVX)6/7 (86)5/7 (71)5/6 (83)6/7 (86)5/7 (71)408769.60No
      legend TDI = therapeutic donor insemination.
      legend Khamsi. Intracytoplasmic sperm injection. Fertil Steril 2001.

       Preparation of sperm for insemination of retrieved oocytes

       Pure sperm gradient technique

      We frequently used the pure sperm gradient technique. The semen was allowed to liquefy at room temperature for 30–60 minutes. The specimen was mixed by using a sterile disposable Pasteur pipette. A sample of the semen was removed and subjected to semen analysis. In a tube, two layers of pure sperm (NidaCon Laboratories ABC, Sweden) at 90% and 45% concentration were prepared at room temperature. One mL of liquefied sperm was slowly added to the side wall of the tube and layered on top of pure sperm. The tube was centrifuged at a gradient of 1,300 rpm for 20 minutes at room temperature. The sperm pellet was removed and was placed in a sterile centrifuge tube. Sperm wash medium (human tubal fluid plus 0.5% human serum albumin) was added to the tube at a volume of 4–6 mL; the sperm was dispersed in this medium. The tube was centrifuged at 1,800 rpm for 6 minutes. The pellet was removed and transferred to another sterile tube, and sperm wash medium was added until appropriate volume for insemination was obtained.

       Swim-up technique

      We used the swim-up technique infrequently. The semen was liquefied at room temperature for 30–60 minutes. The specimen was mixed by using a sterile disposable Pasteur pipette. A sample was removed for sperm analysis. The sperm was diluted in a ratio of 1 sperm volume:3 sperm wash medium volume; the medium consisted of human tubal fluid plus 0.5% human serum albumin. The fluid was thoroughly mixed by using a sterile Pasteur disposable pipette. The mixture was centrifuged at 1,300 rpm for 10 minutes at room temperature. The supernatant was discarded, and 3–5 mL of sperm wash medium was added according to the size of the pellet and mixed well. This mixture was centrifuged at 1,300 rpm for 5 minutes. The supernatant fluid was discarded, and the tube was tapped to loosen the pellet. The pellet was layered over with 1 mL of sperm wash medium and incubated at 45° angle in an incubator for 30–60 minutes. This permitted the motile spermatozoa to swim up to the surface of the medium. At the end of incubation, 0.5 mL of upper-phase fluid was removed and placed into a tube for counting.

       Insemination of oocytes

      Cultured oocytes were inseminated 5–8 hours after retrieval. Each oocyte was inseminated with 50,000 motile washed spermatozoa per culture drop of 0.1 cc. By using a sterile Eppendorf pipette, the sperm was injected slowly around the cumulus-oocyte complex. Test-yolk buffer was not used.

       Ovarian stimulation and oocyte retrieval

      A standard ovarian suppression protocol was used. Buserelin treatment was given from day 20 of the menstrual cycle for 10 days or more until the estradiol value was <184 pmol/L. Patients were then given 3–6 ampoules per day of hMG, 75 IU, until two follicles reached an average diameter of 1.8 cm. At this time, hCG (10,000 IU) was given in the evening, and oocyte retrieval was performed 35 hours later as described elsewhere (
      • Khamsi F.
      • Roberge S.
      • Lacanna I.C.
      • Wong J.
      • Yavas Y.
      Effect of granulosa cells, cumulus cells, and oocyte density on in vitro fertilization in women.
      ).

       Oocyte culture, IVF, ICSI, and embryo transfer

      The retrieved oocyte-cumulus complexes were washed two or three times in culture medium (IVF Science Scandinavia, Gothenburg, Sweden), were transferred individually into 100-μL microdrops of culture medium overlaid by oil, and were put in an incubator (37°C, 5% CO2). On retrieval, diameters of oocyte-cumulus complexes were measured under a dissecting microscope by using a micrometer. Oocyte-cumulus complexes were graded as follows: 0, no cumulus at all; 1, diameter of 75–100 μm; 2, diameter of 125–150 μm; and 3, diameter of 200–225 μm (

      WHO laboratory manual for examination of human semen and sperm. 3rd ed. New York: Cambridge University Press, 1992:44.

      ).
      Oocytes were then randomly assigned in approximately equal numbers within patients to standard IVF or ICSI. In 16 patients, randomization was done by allotting the first or second half of the oocytes to ICSI. In the remaining 21 patients, a coin was tossed; if it was “heads,” oocytes in odd-numbered microdrops were allotted to ICSI, and if it was “tails,” oocytes in even-numbered microdrops were allotted to ICSI.
      Standard IVF insemination and ICSI protocols were performed as described elsewhere (
      • Khamsi F.
      • Roberge S.
      • Lacanna I.C.
      • Wong J.
      • Yavas Y.
      Effect of granulosa cells, cumulus cells, and oocyte density on in vitro fertilization in women.
      ). The ICSI procedure was performed by the same scientist (Y.Y.) throughout the study. Of the oocytes allotted to ICSI, only those at metaphase II stage were injected with sperm.
      Fertilization was confirmed by observation of two pronuclei 14 to 18 hours after IVF insemination or ICSI. Two days after retrieval, embryos were assessed and graded as good quality (equal-size blastomeres with no or minor fragmentation [that is, 5%–20% of blastocytes were fragmented]) or poor quality (unequal-size blastomeres or major fragmentation [that is, >20% of blastomeres were fragmented]).
      The rates of fertilization and good-quality embryo formation are the proportions of fertilized oocytes and the proportions of good-quality embryos, respectively, of the number of oocytes allotted to respective treatments. Good-quality embryos were transferred on day 3 (day 0 was the day of oocyte retrieval) in all patients except three who had embryo transfer on day 2.

       Statistical analysis

      Grades of oocyte-cumulus complexes assigned to IVF and those assigned to ICSI were analyzed with the Student t-test by using SAS software (

      SAS/STAT User’s Guide, Release 6.03 ed. Cary (NC): SAS, Inc., 1996.

      ). Proportions of oocytes that fertilized and proportions of oocytes that resulted in good-quality embryo formation were analyzed by treatment group with χ2 analysis, using SAS software (

      SAS/STAT User’s Guide, Release 6.03 ed. Cary (NC): SAS, Inc., 1996.

      ). Statistical significance was set at P<.05.

      Results

      Patient characteristics and data are shown in Table 1. The oocyte-cumulus complex grades did not differ (P=.59) between oocyte-cumulus complexes assigned to IVF and those assigned to ICSI (1.4 μm ± 0.45 vs. 1.5μm ± 0.44). (Oocyte score was based on the size of the oocyte-cumulus complex). Rates of fertilization and formation of good-quality embryos are shown in Table 2. Of 187 oocytes subjected to standard IVF insemination, 107 oocytes fertilized, for a fertilization rate of 57.2%. Of the other 188 sibling oocytes assigned to ICSI, 164 had attained metaphase II and were injected with sperm; of these, 134 oocytes became fertilized.
      TABLE 2Fertilization rate and good-quality embryo formation in sibling oocytes allotted to IVF or ICSI in 35 women.
      NS = not significant.
      ,
      Khamsi. Intracytoplasmic sperm injection. Fertil Steril 2001.
      VariableTreatmentP value
      IVFICSI
      Fertilization rate per assigned oocyte (%)57.271.3.005
      (107/187)(134/188)
      The fertilization rate was 81.7% (134/164) when it was based on the number of metaphase II oocytes injected with sperm.
      Good-quality embryos per assigned oocyte (%)47.164.4.001
      (88/187)(121/188)
      Good-quality embryos per fertilized oocyte (%)82 (88/107)90 (121/134)NS
      legend NS = not significant.
      a The fertilization rate was 81.7% (134/164) when it was based on the number of metaphase II oocytes injected with sperm.
      legend Khamsi. Intracytoplasmic sperm injection. Fertil Steril 2001.
      On the basis of the number of oocytes assigned to ICSI regardless of whether they were subsequently injected with sperm, the fertilization rate was 71.3% (134 of 188 oocytes). The difference in fertilization rate between standard IVF insemination (57.2%) and ICSI (71.3%) was highly significant (P=.005). On the basis of the number of metaphase II oocytes injected with sperm, fertilization rate in the ICSI treatment was 81.7% (134 of 164 oocytes).
      Four patients had no fertilization and seven patients had low fertilization (10%–33%) with standard IVF insemination, but they had fairly normal fertilization (50%–100%) among their sibling oocytes subjected to ICSI. In one patient with two oocytes (one of which was assigned to standard IVF insemination and one to ICSI), neither oocyte fertilized.
      Of 187 oocytes subjected to standard IVF insemination, 88 developed to good-quality embryos (47.1%). Of the remaining 188 sibling oocytes assigned to ICSI, 121 formed good-quality embryos (64.4%). The difference was highly significant (P=.001).
      The rate of formation of good-quality embryos per fertilized oocyte was 82% (88 of 107) with IVF and 90% (121 of 134) with ICSI; this difference was not statistically significant. This finding indicates that ICSI does not improve or diminish the formation of good embryos; rather, the difference between the two groups in good embryo formation per oocyte was related to the better fertilization rate obtained with ICSI.
      On the basis of observation of at least one fetal heart beat, the pregnancy rate per retrieval was 20.0% (7 of 35 patients) overall and 29.2% (7 of 24) in patients younger than 38 years of age.

      Discussion

      Many IVF centers have extended ICSI treatment to patients other than those with male factor infertility. These centers usually perform ICSI on some but not all of the retrieved oocytes because, based on their clinical experience, they may avoid cases of total failure of fertilization. This point has been well proven by Aboulghar et al. (
      • Aboulghar M.A.
      • Mansour R.T.
      • Serour G.I.
      • Sattar M.A.
      • Amin Y.M.
      Intracytoplasmic sperm injection and conventional in vitro fertilization for sibling oocytes in cases of unexplained infertility and borderline semen.
      ) and Ruiz et al. (
      • Ruiz A.
      • Remohi J.
      • Minguez Y.
      • Guanes P.P.
      • Simon C.
      • Pellicer A.
      The role of in vitro fertilization and intracytoplasmic sperm injection in couples with unexplained infertility after failed intrauterine insemination.
      ) in their prospective randomized studies.
      In our study, the rate of total fertilization failure with conventional IVF was 14.3% (5 of 35 patients); Jun et al. (

      Jun JH, Lim CK, Kim JW, Son IP, Koong MK, Song IO, et al. Comparison of fertilization and embryonic development between conventional insemination and ICSI treatment in the sibling oocytes of non-male factor infertility [abstract]. In: Program supplement of the Annual Meeting of American Society for Reproductive Medicine. 1999;72(Suppl 1):S6.

      ) found a similar rate of 6.8% in their study. In similar studies, Aboulghar et al. (
      • Aboulghar M.A.
      • Mansour R.T.
      • Serour G.I.
      • Sattar M.A.
      • Amin Y.M.
      Intracytoplasmic sperm injection and conventional in vitro fertilization for sibling oocytes in cases of unexplained infertility and borderline semen.
      ) and Ruiz et al. (
      • Ruiz A.
      • Remohi J.
      • Minguez Y.
      • Guanes P.P.
      • Simon C.
      • Pellicer A.
      The role of in vitro fertilization and intracytoplasmic sperm injection in couples with unexplained infertility after failed intrauterine insemination.
      ) reported rates of 11.4% and 23%, respectively. No total fertilization failure occurred in sibling oocytes in any of the three aforementioned studies; in our study, total fertilization failure occurred in one patient who had only two oocytes assigned to conventional IVF and one assigned to ICSI.
      Staessen et al. (
      • Staessen C.
      • Camus M.
      • Clasen K.
      • DeVos A.
      • Van Steirteghem A.
      Conventional in vitro fertilization versus intracytoplasmic sperm injection in sibling oocytes from couples with tubal infertility and normozoospermic semen.
      ) recently conducted a study using the same design as the previous reports and ours. They also found a much lower rate of fertilization failure in the ICSI group. Although ICSI provided a better fertilization rate, it did not reach statistical significance (
      • Staessen C.
      • Camus M.
      • Clasen K.
      • DeVos A.
      • Van Steirteghem A.
      Conventional in vitro fertilization versus intracytoplasmic sperm injection in sibling oocytes from couples with tubal infertility and normozoospermic semen.
      ). We convincingly show that regardless of the cause of infertility (excluding male factor infertility), success of IVF was more certain if half of the oocytes were subjected to ICSI.
      Simultaneously and independent of us, Jun et al. (

      Jun JH, Lim CK, Kim JW, Son IP, Koong MK, Song IO, et al. Comparison of fertilization and embryonic development between conventional insemination and ICSI treatment in the sibling oocytes of non-male factor infertility [abstract]. In: Program supplement of the Annual Meeting of American Society for Reproductive Medicine. 1999;72(Suppl 1):S6.

      ) conducted a study with a similar protocol. They randomly assigned sibling oocytes to standard IVF or ICSI. They studied 103 cycles; causes of infertility were tubal, endometriosis, and unexplained. They reported a significantly higher overall fertilization rate with ICSI (65.6% than with standard IVF insemination (52.5%); respective values in our study were 71.5% and 56.4%. They also reported better formation of good-quality and fair-quality embryos with ICSI (59.2%) than for standard IVF insemination (49.2%) (respective values in our study, 64% and 46.7%). We demonstrated that more good-quality embryos per retrieved oocyte are obtained with ICSI than with conventional IVF.
      Although allotment of oocytes was done in a strict random manner, we analyzed our data to ensure that the oocytes in the two subgroups were of equal morphology. We previously reported that the greater the diameter of the-oocyte cumulus complex, the better the chance of fertilization and development (
      • Khamsi F.
      • Roberge S.
      • Lacanna I.C.
      • Wong J.
      • Yavas Y.
      Effect of granulosa cells, cumulus cells, and oocyte density on in vitro fertilization in women.
      ). This morphological criterion was the same in the two groups, providing assurance that allocation of oocytes was random.
      Of note, in three of four patients who had therapeutic donor insemination, better results were obtained with ICSI. What is the specific biological explanation of the beneficial effect of ICSI in these patients, and all the others, who benefited from ICSI? Is the problem with the sperm, the oocytes, or both? There are couples in whom passage of sperm through cumulus, zona pellucida, or ooplasm is deficient. Are we dealing with one simple defect in all these cases, or may there be multiple defects? As a matter of speculation, some problems must be entirely related to the sperm, and a routine sperm count may not give us the answer. To some extent, this phenomenon is reflected in the patient profiles in Table 1; 16 patients had sperm motility <50%. Among these patients, the fertilization rate was 48% for IVF and 79% for ICSI. In contrast, the other 19 patients with sperm motility ≥50%, the fertilization rate was 66% for IVF and 85% for ICSI. Therefore, ICSI was more beneficial in the group with lower motility, which indicates that male factor was at least in part a cause of the lower fertilization rate with IVF vs. ICSI in our patient population.
      At the time of the original consultation with the couple, we performed semen analysis according to WHO criteria (

      WHO laboratory manual for examination of human semen and sperm. 3rd ed. New York: Cambridge University Press, 1992:44.

      ) for abnormal semen (count, motility, and morphology); we based the designation of male factor infertility on these criteria. We did not include patients who had any of these abnormalities at the original consultation. However, the semen studies reported in this article are those obtained on the day of IVF, and some patients showed abnormalities at that time. We did not perform sperm penetration assay; if we had, we may have considered some of our enrolled patients to have had male factor infertility. This is particularly relevant, as a negative result on sperm penetration assay in patients with non–male factor infertility correlated very highly with failed fertilization in IVF patients (
      • Margalioth E.J.
      • Navot D.
      • Laufer N.
      • Lewin A.
      • Rabinowitz R.
      • Schenker J.G.
      Correlation between the zona-free hamster egg sperm penetration assay and human in vitro fertilization.
      ).
      Routine testing for antisperm antibodies was not performed, but it may have been relevant to fertilization with ICSI. Some of the patients whom we designated as having non–male factor infertility may have categorized as having it if we had adhered to more strict criteria for motility and morphology on the day of IVF. This would have had a bearing on fertilization, as has been shown previously (
      • Chen S.H.
      • Wallach E.E.
      Five decades of progress in management of the infertile couple.
      ). Similarly, if we had routinely performed sperm penetration assay and assay for sperm antibodies, we may have classified some of our patients as having male factor infertility. Therefore, it is possible that other centers that use different protocols may not find routine ICSI as beneficial as we did. The fertilization rate may also be influenced by methods of sperm preparation and insemination.
      Hardening of the zona pellucida at different stages may also be an important factor in failure of fertilization despite healthy spermatozoa. Finally, there may be problems in oocytes related to the type of ovarian stimulation and the culture conditions rather than an intrinsic abnormality of the ovaries. We are undertaking a systematic study of these factors, using treatment success to find the underlying disease mechanism.
      In conclusion, our results and those of Jun et al. (

      Jun JH, Lim CK, Kim JW, Son IP, Koong MK, Song IO, et al. Comparison of fertilization and embryonic development between conventional insemination and ICSI treatment in the sibling oocytes of non-male factor infertility [abstract]. In: Program supplement of the Annual Meeting of American Society for Reproductive Medicine. 1999;72(Suppl 1):S6.

      ) indicate that patients undergoing IVF should be counseled about the benefit of ICSI in some oocytes.

      References

        • Khamsi F.
        • Lacanna I.
        • Endman M.
        • Wong J.
        Recent advances in assisted reproductive technologies.
        Endocrine. 1998; 9: 15-25
        • Ziebe S.
        • Andersen A.N.
        • Andersen A.G.
        • Mikkelsen A.L.
        • Lindenberg S.
        Results of intracytoplasmic sperm injection in relation to indication.
        Acta Obstet Gynecol Scand. 1997; 76: 335-339
        • Aboulghar M.A.
        • Mansour R.T.
        • Serour G.I.
        • Sattar M.A.
        • Amin Y.M.
        Intracytoplasmic sperm injection and conventional in vitro fertilization for sibling oocytes in cases of unexplained infertility and borderline semen.
        J Assist Reprod Genet. 1996; 13: 38-42
        • Ruiz A.
        • Remohi J.
        • Minguez Y.
        • Guanes P.P.
        • Simon C.
        • Pellicer A.
        The role of in vitro fertilization and intracytoplasmic sperm injection in couples with unexplained infertility after failed intrauterine insemination.
        Fertil Steril. 1997; 68: 171-173
      1. WHO laboratory manual for examination of human semen and sperm. 3rd ed. New York: Cambridge University Press, 1992:44.

        • Khamsi F.
        • Roberge S.
        • Lacanna I.C.
        • Wong J.
        • Yavas Y.
        Effect of granulosa cells, cumulus cells, and oocyte density on in vitro fertilization in women.
        Endocrine. 1999; 10: 161-166
      2. SAS/STAT User’s Guide, Release 6.03 ed. Cary (NC): SAS, Inc., 1996.

      3. Jun JH, Lim CK, Kim JW, Son IP, Koong MK, Song IO, et al. Comparison of fertilization and embryonic development between conventional insemination and ICSI treatment in the sibling oocytes of non-male factor infertility [abstract]. In: Program supplement of the Annual Meeting of American Society for Reproductive Medicine. 1999;72(Suppl 1):S6.

        • Staessen C.
        • Camus M.
        • Clasen K.
        • DeVos A.
        • Van Steirteghem A.
        Conventional in vitro fertilization versus intracytoplasmic sperm injection in sibling oocytes from couples with tubal infertility and normozoospermic semen.
        Hum Reprod. 1999; 10: 2474-2479
        • Margalioth E.J.
        • Navot D.
        • Laufer N.
        • Lewin A.
        • Rabinowitz R.
        • Schenker J.G.
        Correlation between the zona-free hamster egg sperm penetration assay and human in vitro fertilization.
        Fertil Steril. 1986; 45: 665-670
        • Chen S.H.
        • Wallach E.E.
        Five decades of progress in management of the infertile couple.
        Fertil Steril. 1994; 62: 665-685