Biologie de la Reproduction, Département de Gynécologie Obstétrique et Reproduction Humaine, CHU Bretonneau, 37044 Tours, France
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: blastomere survival/cleavage/cryopreservation/embryo transfer/freezing-thawing
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
IVF, ICSI procedures and decisions regarding embryo freezing
Briefly, the ovarian stimulation protocol involved down-regulation with triptorelin (0.1 mg s.c. Decapeptyl®; Ipsen-Biotech Laboratories, Paris, France), a gonadotrophin-releasing hormone agonist. Once down-regulation was confirmed, treatment with highly purified FSH (Metrodin HP®; Serono Laboratories, Geneva, Switzerland) or recombinant follicle stimulating hormone (rFSH) (Follitropin , Gonal F®; Serono Laboratories or Follitropin ß, Puregon®; Organon Pharmaceuticals, Saint-Denis, France) was initiated. Monitoring involved measurement of serum estradiol levels and ultrasound evaluation by vaginal probe. HCG (10 000 IU, i.m. Gonadotrophine Chorionique Endo®; Organon) was administered when at least three follicles reached a mean diameter of 18 mm. Transvaginal ultrasound-guided oocyte retrieval was performed 3436 h following HCG injection. IVF and ICSI were performed according to routine protocols. Fertilization was checked 1618 h after insemination or microinjection (day 1). Embryonic development was assessed 24 h later (day 2) prior to transfer, according to the developmental stage and degree of cytoplasmic fragmentation. The two or three morphologically best embryos were selected for fresh transfer and supernumerary grade 1 embryos (4-cell stage with <10% fragmentation) were cryopreserved on day 2. As a consequence of these very strict criteria, very few embryos were suitable for cryopreservation on day 2 in our programme (
10% of all supernumerary embryos). Provided they had <50% fragmentation, supernumerary embryos that remained at day 2 (80% of total) were kept in culture to be frozen if they reached the (full or expanded) blastocyst stage.
Freezing protocol
Cryopreservation was carried out in a cell freezer (Minicool LC 40; Air Liquide, Paris, France) following a slow freeze protocol using 1,2-propanediol (PROH) (Sigma-Aldrich, St Quentin Fallavier, France) as cryoprotectant (Lassalle et al., 1985). Embryos were frozen in plastic straws (Cryo Bio System; L'Aigle, France). Before freezing, embryos were placed in 1.5 mol/l PROH added to 20% human serum albumin (HSA) (LFB; Courtaboeuf, France) for 20 min at room temperature. They were then placed in the same medium with 0.1 mol/l sucrose (Prolabo, Paris, France) for 5 min at room temperature and aspirated into the straws (two embryos per straw). The straws were cooled from room temperature down to 7°C at a rate of 2°C/min, followed by touching the straws with cold forceps to manually induce ice nucleation. After seeding, the temperature was first lowered to 30°C at the rate of 0.3°C/min, then straws were rapidly frozen to 135°C at the rate of 35°C/min and finally plunged and stored in liquid nitrogen.
Thawing protocol
Cryopreserved embryos were thawed 1 day before embryo transfer, according to the same protocol (Lassalle et al., 1985) with some modifications. Briefly, the straws were quickly removed from liquid nitrogen and immediately placed in 5-min baths at room temperature with decreasing PROH concentrations (1.0 mol/l, then 0.5 mol/l and finally 0.0 mol/l) while sucrose concentration was kept constant (0.2 mol/l) to remove the cryoprotectant. After transfer to a sucrose-free culture medium at 37°C, each embryo was first carefully evaluated for the number of surviving blastomeres and then individually cultured for 20 h. A second evaluation was performed prior to transfer in order to record the resumption of mitosis and the total number of blastomeres for each thawed embryo. Resumption of mitosis was defined as the cleavage of at least one blastomere during post-thaw culture.
Replacement of frozenthawed embryos
Two different regimens were used to prepare recipients for frozen embryo transfers. Thawed embryos were transferred during the course of stimulated (ovulatory patients) or substitutive (an- or dysovulatory patients) treatments. For stimulated cycles, rFSH was administered (50 to 100 IU daily) from day 2 to day 9. Monitoring involved measurement of serum estradiol levels and ultrasound evaluation by vaginal probe. HCG was administered (5000 IU) according to follicle growth, estradiol levels and LH surge. Frozen embryo transfer was performed on day 4 or 5 after HCG administration, depending on the progesterone level. For hormone replacement therapy, patients were treated with GnRH agonist (triptorelin, 3.75 mg i.m. Decapeptyl LP®; Ipsen-Biotech Laboratories, Paris, France) starting on the first day of their menstrual cycle. Once down-regulation was confirmed 15 days later, endometrial development was achieved by transdermal administration of estradiol (Estreva gel®; Theramex, Monaco) and oral micronized estradiol (Estreva®; Theramex) at a gradually increasing dose to mimic the natural cycle. When estradiol levels and endometrial thickness were suitable, this phase was complemented by vaginal administration of 200 mg progesterone, twice a day (Utrogestan®; Besins Iscovesco, Paris, France). Embryo transfer was performed on the fourth day of progesterone administration. Long-acting progesterone (500 mg i.m. Progesterone LP®; Schering SA, Lys Lez Lannoy, France) was also administered i.m. on the day of transfer, then every 10 days. Serum HCG concentration was determined 12 days after embryo replacement and then 48 h and one week later. If pregnancy was initiated, steroid supplementation was maintained until week 12 of gestation. Clinical pregnancy was defined as the presence of a gestational sac with fetal heartbeat on ultrasound examination at 7 weeks of pregnancy. Embryo implantation rate was defined as the number of gestational sacs per number of transferred embryos [implantation rate (IR) per transferred embryo] or as the number of gestational sacs per number of thawed embryos (IR per thawed embryo).
Analysis of parameters
Embryos were considered to be surviving when at least half of the initial number of blastomeres were intact. Blastomeres were considered to be damaged when they were lysed, degenerated or dark. Three types of cryopreserved transfer were evaluated to investigate the influence of the presence of damaged blastomeres in cryopreserved embryos: i.e. transfers with only fully intact embryos (100% blastomere survival); mixed transfers with both intact and damaged embryos and transfers with only damaged embryos (50 or 75% blastomere survival). Three types of cryopreserved transfer were also evaluated to assess the influence of the resumption of mitosis in cryopreserved embryos after culture (20 h): i.e. transfers with only cleaved embryos; mixed transfers with both cleaved and uncleaved embryos and transfers with only uncleaved embryos. Four homogeneous groups were defined for transferred embryos to assess the respective weighting of each of these parameters (blastomere survival and resumption of mitosis): i.e. all damaged embryos without any cleavage; all damaged embryos with resumption of mitosis; all intact embryos without any cleavage and all intact embryos with resumption of mitosis. Mixed transfers were excluded for the last evaluation combining blastomere survival and resumption of mitosis.
Statistical analysis
Statistical analysis was performed using Staview 4.1® (Abacus). ANOVA followed by post-hoc comparisons or contingency tables were used depending on the parameter being evaluated. Multivariate analysis was used to measure the weighting of each parameter in implantation prognosis. Differences were considered significant if P < 0.05.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Outcome of transferred embryos in terms of blastomere survival
The 316 transfer cycles comprised 102 transfer cycles with fully intact embryos (FT), 100 mixed transfer cycles (MT) and 114 transfer cycles with only damaged embryos (DT). The mean age of the women did not differ significantly between groups (32.6 ± 4.1, 33.1 ± 4.1 and 32.9 ± 4.3 years for FT, MT and DT groups respectively). General data concerning the defined groups revealed differences between FT and DT groups in terms of original assisted reproduction procedure (IVF: 64 versus 79%; P < 0.05 for FT and DT groups respectively) and treatment for embryo replacement (stimulated cycle: 65 versus 90%; P < 0.05, for FT and DT groups respectively).
The clinical results for the three groups are shown in Table I. A total of 654 embryos was transferred during 316 transfer procedures, including 195 fully intact embryos (30%). Significantly higher IR per transferred embryo was observed when transferred embryos were fully intact compared with damaged embryos (22.0 versus 7.2% respectively; P < 0.0001). This result was confirmed when the birth rate per transfer was calculated (30% for intact embryos versus 9% for damaged embryos; P < 0.0001). In contrast the birth rate per transfer (26%) was fairly similar for mixed transfers compared with the intact embryo group. However a higher number of embryos was transferred in the mixed group compared with intact group. Moreover, the IR per transferred embryo was significantly lower in the mixed group compared with the intact group (12.7 versus 22.0% respectively; P < 0.05).
|
The clinical results for the three groups are summarized in Table II. A total of 654 embryos were transferred in the 316 transfer cycles, including all 299 cleaved embryos (46%). A significantly higher IR per transferred embryo was observed when all transferred embryos were cleaved compared with uncleaved transferred embryos (19.7 versus 3.0% respectively; P < 0.0001). This result was confirmed by the birth rate per transfer (28% for all cleaved embryos versus 4% for uncleaved embryos; P < 0.0001). With mixed transfers, the birth rate (25%) was very similar to all cleaved embryos but with a higher number of transferred embryos (2.5 ± 0.5 versus 2.0 ± 0.7 respectively; P < 0.0001). The IR per transferred embryo was in fact significantly lower in the mixed group compared with the intact group (11.9 versus 19.7% respectively; P < 0.05).
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
From the initial part of the study it was clear that complete survival of an embryo after cryopreservation was followed by the same potential to implant as usually occurs with fresh embryo transfer. Our results are in agreement with a previous study (Edgar et al., 2000) in which intact thawed embryos were reported to have the same developmental potential as equivalent fresh embryos. The reason for the lower viability of partially damaged human embryos after cryopreservation that we observed remains controversial. Some previous studies reported that damaged embryos had the same capacity to produce pregnancies as fully intact embryos and children were born after transfer of partially damaged cryopreserved embryos, although the low number of embryos investigated in these studies should be borne in mind (Mohr et al., 1985
; Veiga et al., 1987
; Hartshorne et al., 1990
; Testart et al., 1990
). On the other hand, more recent studies using higher numbers of embryos have reported a lower development potential for damaged embryos compared to embryos with all blastomeres intact (Van den Abbeel et al., 1997
; Burns et al., 1999
; Edgar et al., 2000
). Although our results showed that the transfer of damaged embryos did not preclude any chance to achieve pregnancy, they nonetheless confirmed a lower implantation rate for such embryos. In the present study, the IR was about three times lower when transferred embryos were damaged compared with fully intact embryos, and even four times lower when IR was expressed per thawed embryo. Several studies have highlighted the negative influence of the presence of damaged blastomeres on further embryo development. Complete in-vivo development of damaged embryos may occur in the mouse but the rate is almost half that of intact embryos (Rülicke and Autenried, 1995
). A possible toxic effect from damaged blastomeres on the other blastomeres has been evoked. The destruction of blastomeres by micromanipulation allowed mouse embryos to develop until the blastocyst stage but dramatically reduced the hatching rate and implantation potential, which were restored when degenerating material was removed microsurgically (Alikani et al., 1993
). Another hypothesis is that complete blastomere survival might reflect a higher intrinsic embryonic quality of the intact embryos before cryopreservation, allowing them to tolerate the freezingthawing procedure better. Various studies have reported a relationship between the outcome of thawed embryos and embryo morphology before cryopreservation (Schalkoff et al., 1993
; Kondo et al., 1996
; Lightman et al., 1997
). In the present study, initial embryo morphology was similar for all cryopreserved embryos (drastic selection criteria). In spite of apparently similar morphology, not all thawed embryos tolerated cryopreservation equally. Bearing in mind that two embryos were preserved per straw, we cannot exclude the possibility that physical conditions might differ for some embryos from different straws during freezing or thawing procedures. It is also possible that, apart from morphological appearance or blastomere damage, another intrinsic factor which occurs randomly during the freezingthawing procedure might influence embryo development.
Another result from this study was that embryos that had further cleaved during the post-thaw culture period had a significantly higher chance of implantation. The overall clinical PR for frozenthawed embryos in our centre was 25% per transfer; however it reached 34% when all transferred embryos had resumed mitosis. Similarly the IR per transferred embryo was 14% overall but reached 19.7% when all transferred embryos had resumed mitosis. Moreover, when comparing uncleaved and cleaved embryos, the IR per transferred embryo was six times lower and IR per thawed embryo was seven times lower for uncleaved embryos. Thus failure to resume mitosis seems to give poor implantation potential for thawed embryos. These results are in agreement with previous reports showing cleavage ability as a good indicator of embryo viability (Van der Elst et al., 1997; Ziebe et al., 1998
). One study investigated the incidence of numerical abnormalities for chromosomes X, Y and 1 in blastomeres of human preimplantation embryos that survived cryopreservation but did not cleave further after thawing (Laverge et al., 1998
). A large proportion of uncleaved frozenthawed embryos were reported to carry chromosomal aberrations. The embryos studied were day 2 or day 3 frozen embryos with equally or unequally sized blastomeres and <20% fragmentation. These findings are in agreement with the lower IR observed by others and ourselves when uncleaved embryos were transferred.
The inclusion criteria for embryo freezing in many laboratories are not as strict as in our centre, thus allowing freezing of a greater embryo cohort. In our centre, the percentage of cycles leading to embryo freezing remains 10% whereas it usually reaches at least 20% for most of laboratories. It should be noted that the proportion of supernumerary embryos reaching the blastocyst stage during the study period was low (14%), which suggests a low development potential for embryos not selected for freezing. Additionally, indirect evidence of selection criteria was given by the pregnancy rate observed after fresh embryo transfer (33 versus 24% for cycles with embryo freezing compared to cycles without embryo freezing, respectively; data not shown). Thus, although a more permissive strategy allows for more embryos to be thawed, it raises the crucial problem of selecting the best embryos for transfer. Culturing embryos for an extra day might be beneficial for the selection of embryos in terms of the embryo's ability to resume cleavage after thawing. The association of both parameters allowed us to distinguish three main situations with gradually increasing IR per transferred or thawed embryo.
Firstly, when whole transferred embryos not only completely survived but also cleaved after culture, the IR per transferred embryo (27.4%) reached the IR observed after transfer of fresh embryos. Secondly when only one criterion was fulfilled (blastomere survival or cleavage = intermediate situations), IR per transferred embryo (between 1113%) were similar for both criteria. The fairly acceptable implantation rate observed when embryos failed to resume mitosis, despite optimal survival, needs to be interpreted with caution as the sample size was small. However it cannot be excluded that in such cases the increased delay in resumption of mitosis might be related to the freezing or thawing procedure. Thirdly, no pregnancy was initiated in our study when neither criterion was fulfilled for transferred embryos.
It thus seems likely that the most important criterion might be the total number of blastomeres for transferred embryos however they are obtained (total blastomere survival and/or resumption of mitosis). This was confirmed in our study by multivariate analysis, which provided a clearly increased OR for pregnancy rate 6-cell embryos compared with <6-cell embryos. This may have practical implications for units where a large number of embryos are frozen, raising selection problems. Indeed individual follow-up of each thawed embryo for survival and further cleavage may make it possible to modify the transfer strategy. Our results therefore led us to conclude that blastomere survival and resumption of mitosis are both valuable and complementary for the selection of which, and how many, frozenthawed embryos to transfer. As survival and cleavage on day 3 may not be interpreted as strict criteria of developmental competence, the benefit of further in-vitro development to the blastocyst stage before transfer remains to be assessed.
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Burns, W.N., Gaudet, T.W., Martin, M.B., Leal, Y., Schoen, H., Eddy, C.A. and Schenken, R.S. (1999) Survival of cryopreservation and thawing with all blastomeres intact identifies multicell embryos with superior frozen embryo transfer outcome. Fertil. Steril., 72, 527532.[ISI][Medline]
Camus, M., Van den Abbeel, E., Van Waesberghe, L., Wisanto, A., Devroey, P. and Van Steirteghem, A.C. (1989) Human embryo viability after freezing with dimethylsulfoxide as a cryoprotectant. Fertil. Steril., 51, 460465.[ISI][Medline]
Cohen, J., De Vane, G.W., Elsner, C.W., Fehilly, C.B., Kort, H.I., Massey, J.B. and Turner, T.G. Jr (1988) Cryopreservation of zygotes and early cleaved human embryos. Fertil. Steril., 49, 283289.[ISI][Medline]
Edgar, D.H., Bourne, H., Speirs, A.L. and Mc Bain, J.C. (2000) A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos. Hum. Reprod., 15, 175179.
Freeman, L., Trounson, A. and Kirby, C. (1986) Cryopreservation of human embryos : progress on the clinical use of the technique in human in vitro fertilization. J. Assist. Reprod. Genet., 3, 5361.
Hartshorne, G.M., Wick, K., Elder, K. and Dyson, H. (1990) Effect of cell number at freezing upon survival and viability of cleaving embryos generated from stimulated IVF cycles. Hum. Reprod., 5, 857861.[Abstract]
Kondo, I., Suganuma, N., Ando, T., Asada, Y., Furuhashi, M. and Tomoda, Y. (1996) Clinical factors for successful cryopreserved-thawed embryo transfer. J. Assist. Reprod. Genet., 13, 201206.[ISI][Medline]
Lassalle, B., Testart, J. and Renard, J.P. (1985) Human embryo features that influence the success of cryopreservation with the use of, 1,2 propanediol. Fertil. Steril., 44, 645651.[ISI][Medline]
Laverge, H., Van der Elst, J., De Sutter, P., Verschraegen-Spae, M.R., De Paepe, A. and Dhont, M. (1998) Fluorescent in-situ hybridization on human embryos showing cleavage arrest after freezing and thawing. Hum. Reprod., 13, 425429.[ISI][Medline]
Lightman, A., Kol, S., Wayner, V., Vertman, D., Manor, D. and Itskovitz-Eldor, J. (1997) The presence of a sponsoring embryo in a batch of poor quality thawed embryos significantly increases pregnancy and implantation rate. Fertil. Steril., 67, 711716.[ISI][Medline]
Lin, Y.P., Cassidenti, D.L., Chacon, R.R., Soubra, S.S., Rosen, G.F. and Yee, B. (1995) Successful implantation of frozen sibling embryos is influenced by the outcome of the cycle from which they were derived. Fertil. Steril., 63, 262267.[ISI][Medline]
Mandelbaum, J., Belaisch-Allart, J., Junca, A., Antoine, J.M., Plachot, M., Alvarez, S., Alnot, M.O. and Salat-Baroux, J. (1998) Cryopreservation in human assisted reproduction is now routine for embryos but remains a research procedure for oocytes. Hum. Reprod., 13 (Suppl. 3), 161177.[Abstract]
Mohr, L.R., Trounson, A. and Freeman, L. (1985) Deep-freezing and transfer of human embryos. J. In Vitro Fertil. Embryo Transfer, 2, 110.[Medline]
Rülicke, T. and Autenried, P. (1995) Potential of two-cell mouse embryos to develop to term despite partial damage after cryopreservation. Lab. Anim., 29, 320326.[ISI][Medline]
Schalkoff, M.E., Oskowitz, S.P. and Powers, R.D. (1993) A multifactorial analysis of the pregnancy outcome in a successful embryo cryopreservation program. Fertil. Steril., 59, 10701074.[ISI][Medline]
Scott, R.T., Hoffmann, G.E., Veeck, L.L., Jones, H.W. Jr and Muasher, S.J. (1991) Embryo quality and pregnancy rates in patients attempting pregnancy through in vitro fertilization. Fertil. Steril., 55, 426428.[ISI][Medline]
Staessen, C., Camus, M., Khan, I., Smitz, J., Van-Waesberghe, L., Wisanto, A., Devroey, P. and Van Steirteghem, A.C. (1989) An 18-month survey of infertility treatment by in vitro fertilization, gamete and zygote intrafallopian transfer and replacement of frozen-thawed embryos. J. In Vitro Fertil. Embryo Transfer, 6, 2229.[ISI][Medline]
Staessen, C., Van den Abbeel, E., Carle, M., Khan, I., Devroey, P. and Van Steirteghem, A.C. (1990) Comparison between human serum and Albuminar-20 (TM) supplement for in-vitro fertilization. Hum. Reprod., 5, 336341.[Abstract]
Testart, J., Lassalle, B., Belaisch-Allart, J., Hazout, A., Forman, R., Rainhorn, J.D. and Frydman, R. (1986) High pregnancy rate after early human embryo freezing. Fertil. Steril., 46, 268272.[ISI][Medline]
Testart, J., Lassalle, B., Forman, R., Gazengel, A., Belaisch-Allart, J., Hazout, A., Rainhorn, J.D. and Frydman, R. (1987) Factors influencing the success rate of human embryo freezing in an in vitro fertilization and embryo transfer program. Fertil. Steril., 48, 107112.[ISI][Medline]
Testart, J., Volante, M. and Lassalle, B. (1990) Comparison of cryoprotectants and stages of human embryo. In Mashiah, S. (ed.), Advances in Assisted Reproduction Technologies. Plenum, New York, p. 573.
Toner, J.P., Brzyski, R.G., Oehninger, S., Veeck, L.L., Simonetti, S. and Muasher, S.J. (1991) Combined impact of the number of pre-ovulatory oocytes and cryopreservation on IVF outcome. Hum. Reprod., 6, 284289.[Abstract]
Trounson, A. and Mohr, L. (1983) Human pregnancy following cryopreservation, thawing and transfer of an eight-cell embryo. Nature, 305, 707709.[ISI][Medline]
Van den Abbeel, E., Van der Elst, J., Van Waesberghe, L., Camus, M., Devroey, P., Khan, I., Smitz, J., Staessen, C., Wisanto, A. and Van Steirteghem, A. (1988) Hyperstimulation: the need for cryopreservation of embryos. Hum. Reprod., 3 (Suppl. 2), 5357.[Medline]
Van den Abbeel, E., Camus, M., Van Waesberghe, L., Devroey, P. and Van Steirteghem, A.C. (1997) Viability of partially damaged human embryos after cryopreservation. Hum. Reprod., 12, 20062010.[Abstract]
Van der Auwera, I., Meuleman, C. and Koninckx, P.R. (1994) Human menopausal gonadotrophin increases pregnancy rate in comparison with clomiphene citrate during replacement cycles of frozen/thawed pronucleate ova. Hum. Reprod., 9, 15561560.[Abstract]
Van der Elst, J., Camus, M., Van den Abbeel, E., Maes, R., Devroey, P. and Van Steirteghem, A.C. (1995) Prospective randomized study on the cryopreservation of human embryos with dimethylsulfoxide or, 1,2-propanediol protocols. Fertil. Steril., 63, 92100.[ISI][Medline]
Van der Elst, J., Van den Abbeel, E. Camus, M., Smitz, J., Devroey, P. and Van Steirteghem, A.C. (1996) Long-term evaluation of implantation of fresh and cryopreserved human embryos following ovarian stimulation with buserelin acetate-human menopausal gonadotrophin (HMG) or clomiphene citrate-HMG. Hum. Reprod., 11, 20972106.[Abstract]
Van der Elst, J., Van den Abbeel, E., Vitrier, S., Camus, M., Devroey, P. and Van Steirteghem, A.C. (1997) Selective transfer of cryopreserved human embryos with further cleavage after thawing increases delivery and implantation rates. Hum. Reprod., 12, 15131521.[Abstract]
Van Steirteghem, A., Van den Abbeel, E., Camus, M., Van Waesberghe, L., Braeckmans, P., Khan, I., Nijs, M., Smitz, J., Staessen, C., Wisanto, A. et al. (1987) Cryopreservation of human embryos obtained after gamete intra-Fallopian transfer and/or in-vitro fertilization. Hum. Reprod., 2, 593598.[Abstract]
Veiga, A., Calderon, G., Barri, P.N. and Coroleu, B. (1987) Pregnancy after the replacement of a frozen-thawed embryo with <50% intact blastomeres. Hum. Reprod., 2, 321323.[Abstract]
Ziebe, S., Bech, B., Petersen, K., Mikkelsen, A.L., Gabrielsen, A. and Nyboe Andersen, A. (1998) Resumption of mitosis during post-thaw culture: a key parameter in selecting the right embryos for transfer. Hum. Reprod., 13, 178181.[Abstract]
Submitted on August 17, 2001; resubmitted on October 22, 2001; accepted on January 20, 2002.