Cryopreservation—the ability to freeze and thaw with retention of viability—provides flexibility in human infertility therapy when gametes or embryos are handled in vitro because frozen tissue can be stored indefinitely in liquid nitrogen at -196°C. The freezing of human sperm is an established procedure that has resulted in the birth of thousands of progeny, as many as 30,000 per year. Not only can partner or donor sperm be frozen for therapeutic insemination (TI) at a subsequent date, but sperm banking provides the assisted reproductive technologies (ART) couple with a backup option if a sample cannot be collected on demand or if sample quality is poor on the day eggs are available. In the case of donor insemination, samples are quarantined for 6 months before use, thereby minimizing the risk of infectious disease transmission. In other words, the donor is screened for sexually transmitted diseases at the time of collection, and the frozen sample is released for use 6 months later only after the donor passes rescreening. Pregnancies have been established with sperm stored at low temperatures for more than 10 years; however, it is generally recognized that fecundity using cryopreserved sperm is lower than that obtained with nonfrozen sperm.

The cryopreservation of human ova would be advantageous to the patient with ovarian disease or any other condition that limits egg production. Moreover, freezing oocytes circumvents the ethical issues associated with embryo freezing. Unfortunately, the technology has not yet been perfected for egg freezing on a routine basis, despite the reports of successful pregnancies.^1,2,3 In contrast, embryo cryopreservation is an established procedure that has been employed successfully for several years. Embryo banking was originally designed to provide alternatives for the ART couple with an embarrassment of riches. Because the risk of multiple pregnancy restricts the number of embryos that can be transferred during the treatment cycle, cryopreservation was begun to store surplus embryos for transfer at a later date to either the egg donor or to other women in the context of an embryo donor program. Additionally, embryo banking may be appropriate when embryos cannot be transferred during the egg pickup cycle. Another advantage to embryo freezing is that embryo thaw and transfer can be conducted in a natural or regulated cycle when the patient has not been subjected to ovarian stimulation with exogenous hormones.

Freezing is detrimental to most cells, with the plasma membrane as the principal target of injury and intracellular ice crystal formation as the primary mechanism. Additionally, cell exposure to high solute concentrations and osmotic shock during freezing and thawing can induce damage. Hence, the danger during cryopreservation is not low-temperature storage per se, but rather transitions through the temperature zone where ice crystal formation is most likely to occur. Based on the pioneering studies of Mazur,⁴ we now recognize that cells behave uniquely during cooling and warming because of differences in their ability to respond to osmotic challenges. This ability is, in turn, determined by the cell's permeability coefficient, its surface area, and the osmotic gradient between outside and inside. These relations have been formalized to the extent that the response of many cells can be predicted.

Cryoprotectants are used to enhance cryosurvival. Although their mechanism of action is incompletely understood, cryoprotectants are involved in cellular dehydration, and their presence results in a lowering of the freezing point of the solution. The resultant osmotically driven dehydration is manifested by cell shrinkage. In 1 M dimethylsulfoxide, intracellular ice crystal formation may not occur until the cell is cooled to -40°C.⁵ Additional osmotically driven dehydration results below the seeding temperature (the temperature where extracellular ice crystal formation is induced; usually -6°C for human embryos in glycerol, propanediol, or dimethylsulfoxide) because the extracellular osmotic pressure is increased as pure water freezes, leaving the cell in increasingly concentrated unfrozen solute. The use of slow, controlled rates of cooling and warming is important when large hydrated embryos are frozen; however, spermatozoa, which contain limited amounts of cytoplasm and are permeable to water, tolerate relatively rapid rates. The cryosurvival of embryos, which is high compared with that of unfertilized eggs, has been attributed to differences in permeability coefficients (increasing from the unfertilized one-cell egg to the blastocyst⁶) and adverse temperature-dependent effects on the egg's meiotic spindle. Other approaches to lowtemperature storage have been described involving ultrarapid cooling or vitrification.^7,8 In these cases, the extremely rapid rates of cooling (-2500°C/min) used in the presence of high concentrations of cryoprotectant preclude intracellular ice crystal formation; residual water is transformed to a highly viscous, glass-like solid state or vitrified. Embryo vitrification is an area of active investigation.

In summary, parameters of concern in cryobiology include cell type, cryoprotectant selection, cryoprotectant addition, seeding, cooling rate, final cooling temperature, hydration state of the cells, warming rate, and cryoprotectant removal.

The successful cryopreservation of sperm is based on the serendipitous discovery of the cryoprotective properties of glycerol.⁹ After initial application to the storage of human sperm by Sherman,¹⁰ a number of different cryoprotective media and techniques have been devised. All rely on glycerol as the permeable cryoprotectant; some include extenders such as citrated TES-Tris or HEPES-buffered egg yolk. The use of programmable freezers and controlled cooling or warming rates has not proven especially useful with sperm, where, unlike egg or embryo freezing, large numbers of cells are commonly available. A cryosurvival rate of 20% might be acceptable when the starting population is 300 million strong, but with eggs or embryos, survival rates in excess of 80% are highly desirable, if not essential. Although some mammalian sperm are injured by exposure to low temperature alone (so-called chilling injury or cold shock), this problem does not plague human sperm. Moreover, our species and a few others are unique among mammals in that sperm cryosurvival is high.

In 1972, Whittingham and associates¹¹ and Wilmut¹² succeeded in cryopreserving eight-cell mouse embryos. Since that time, thousands of embryos from a number of mammalian species have been frozen and thawed successfully, as manifested by term births after embryo transfer. This approach has become routine and cost-effective for embryo banking in cattle and laboratory rodents, and it provided the prerequisite experience base for human application, which was first reported in the context of ART in 1983 by Trounson and Mohr.¹³ Protocols are available for the cryopreservation of human fertilized eggs (zygotes) and cleaving embryos at both early and late stages of preimplantation development. All involve the use of programmable freezers and relatively slow rates of cooling (0.3° to 0.5°C/min). The cryoprotectants in widespread use include propanediol and dimethylsulfoxide in the presence of sucrose for zygotes or early-stage embryos and glycerol for uterine-stage embryos (blastocysts). The routine cryopreservation of uterine-stage human embryos is dependent on improved success in the culture of embryos produced by in vitro fertilization (IVF). Thus, although the implantation efficiency of blastocysts transferred to the uterus is greater than that for oviductal-stage embryos, until just a few years ago only a small percentage of IVF-produced embryos developed to the blastocyst stage in vitro. This deficiency may reflect autocrine or paracrine regulation of embryonic growth that is subnormal under the in vitro conditions currently used. Dramatic improvements in culturing efficiency have been reported on coculturing embryos on monolayers of monkey-derived Vero cells,¹⁴ on the addition of exogenous growth factors,¹⁵ or on the more recent development and use of specialized media.¹⁶

The freezing of sperm is relatively simple and rapid: it does not necessarily involve seeding or the use of slow, controlled rates of cooling. Many sperm banks do not use programmable freezers but simply expose samples diluted in cryoprotectant, with or without precooling, to liquid nitrogen vapors (-130°C), followed by a plunge into liquid nitrogen (-196°C). After sample collection and liquefaction, cryoprotectant is added at ambient temperature, usually in a 1:1 ratio, to give a final glycerol concentration of 5% to 10%. Cryoprotectant solutions may contain extenders such as egg yolk and pH buffers such as TES and Tris and are available commercially (Irving Scientific, Inc., Santa Ana, CA). Diluted semen samples are aliquoted (about 1 mL) during transfer to straws or vials for freezing. A labeling and recording system must be in place to prevent sample loss or misidentification.

The application of conventional cryopreservation to the treatment of male-factor infertility by relatively low-technology approaches such as TI is limited in scope because sperm in semen with poor characteristics do not usually survive well. This expectation largely eliminates the theoretical possibility of treating oligozoospermia, asthenozoospermia, or teratozoospermia by accumulating several frozen ejaculates and conducting TI with pooled samples. In the context of an ART program with intracytoplasmic sperm injection (ICSI) capabilities, however, sperm cryopreservation takes on greater importance. Because only a few viable sperm are required to support ICSI (one per oocyte available for injection), cryopreservation of very poor-quality semen, ejaculated sperm, or even testicular sperm or tissue can be considered. Of course, for non-male-factor patients, cryopreservation allows the storage of a backup sample for in vitro fertilization (IVF) or gamete intrafallopian transfer (GIFT). Fertilization levels should not be reduced when cryostored sperm are used from donors or patients with normal semen parameters. Sperm banking is useful for the patient anticipating a vasectomy, cancer chemotherapy, or any other procedure that might jeopardize testicular function and sperm production. TI with frozen husband sperm may be an alternative for couples who cannot conveniently time intercourse with ovulation.

HIV can be transmitted heterosexually. The Centers for Disease Control & Prevention reports that more than one fourth of US women with the disease acquired it through heterosexual relations. Fresh semen for TI, even that obtained from screened, seronegative donors, carries an unacceptable risk of transmission because a window of 120 days or longer exists between the time of infection and seroconversion and, hence, detection by current serum screening methodology. In 1988, the American Fertility Society (now the American Society for Reproductive Medicine) published revised guidelines¹⁷ mandating the exclusive use of frozen, quarantined (6 months) semen for TI. In addition to HIV testing, donors are subjected to a battery of other sexually transmitted disease screens, along with a threegeneration medical history, physical examination, semen analysis, and cryosurvival tests.

An assessment of cryodamage, with the objective of improving cryosurvival, should include characterization of each step in the process—sample dilution, cryoprotectant addition, cooling to the freezing point, and freezing and thawing. However, on a routine basis, post-thaw motility scores, in the presence of cryoprotectant, are most commonly used. A minimum standard is that 50% of the motile sperm originally present survive, and at least 20 million motile cells are available for insemination. Additionally, a stress test can be applied in which sperm are first washed free of cryoprotectant and then incubated in an appropriate medium at 37°C for up to 24 hours. Quantitative motility scores using computer-assisted video imaging techniques can be obtained. However, the latter will become advantageous only if correlations with fertility potential can be drawn. Preliminary studies suggest that the size of the hyperactivated population (sperm with high curvolinear velocity and low linearity) of sperm may be predictive of fecundity.

To evaluate the efficacy of TI with frozen semen, it is useful first to contrast fresh versus frozen cycle fecundities by intracervical insemination. Typically, ranges of 10% to 15% versus 4% to 9% are associated with the use of fresh and frozen sperm, respectively. With the transition to mandated, exclusive use of quarantined semen, efforts to improve TI outcome have centered on increasing the number of motile cells used and on the placement of sperm closer to the site of fertilization by intrauterine insemination (IUI). This strategy has succeeded, because cycle fecundities with IUI and cryopreserved sperm approach or perhaps even exceed those associated with the prior use of fresh sperm in cervical inseminations (Table 1). More recently, cryoprotectant combinations have been developed for the successful freezing of washed or IUI-ready sperm.²² The availability of IUI-ready donor sperm will be a major convenience for infertility treatment in the physician's office setting: samples can simply be thawed and used directly without additional processing.

TABLE 1. Therapeutic Insemination by Donor-Influence of the Route of Insemination on Success with Quarantined, Cryopreserved Semen

The successful cryostorage of oocytes could circumvent many of the ethical objections to storing embryos and could provide alternatives to patients whose reproductive capacity is threatened by health concerns. Unfortunately, significant deleterious effects of cryopreservation on oocyte survival and function have been noted, including disruption of the microtubular spindle, leading to aneuploidy after fertilization. Efforts to overcome these limitations by freezing immature oocytes have met with limited success.²³

Thus, routine cryopreservation has focused on embryos. The developmental stage of the embryo(s) for freezing dictates the protocol and cryoprotectant used. Glycerol is often used for cryopreservation at the blastocyst stage. Dimethylsulfoxide, although popular as a cryoprotectant for early cleavage-stage embryos in the 1980s, has been largely replaced by propanediol.^24,25 Although this cryoprotectant, in the presence of nonpermeating sucrose, can be used for the one-cell zygote or pronuclear stage, as well as for two- and four-cell-stage embryos, freezing at the one-cell stage has become common. Success in the cryopreservation of the one-cell fertilized zygote is based on high cryosurvival rates (70% to 80%) and enhanced post-thaw screening, because the zygote must complete fertilization and cleave in culture before it is considered for transfer. Also, the use of embryo quality as a major determinant in allocating embryos (freezing versus immediate transfer) is minimized with zygotes, because their quality is highly uniform. A decision can be predicated simply on the total number of embryos available and the maximal number desired for immediate transfer. If, for instance, 12 zygotes are available and the maximum number that the couple, in consultation with their physician, will accept for transfer during the egg pickup cycle is 5, then 4 or 5 zygotes could be held for culture, with the expectation that on average 3 or 4 would cleave to the 2- to 4-cell stage for transfer. The balance would then be cryopreserved.

Individual steps in the freezing process include cryoprotectant addition at ambient temperature, sample transfer to vials or straws, cooling to the seeding temperature (at -2°C/min), seeding (at -6° to -7°C), cooling to the plunging temperature (-30° or -80°C, at -0.3° to -0.5°C/min), and finally plunging into liquid nitrogen for storage at -196°C. The cryoprotectant is added and removed stepwise, and sucrose is present (as a nonpermeating cryoprotectant) during freezing and thawing to minimize osmotic shock, which can cause cell lysis, and to assist in cellular dehydration. The final concentration of propanediol is 1.5 M in phosphate-buffered saline (Dulbecco's) containing protein in the form of maternal serum or albumin. Embryos can be stored indefinitely in liquid nitrogen, but a maximum storage time is often specified in the freezing agreement or consent form (e.g., for the reproductive lifetime of the egg donor, or for 3 to 4 years).

For thawing, solutions are prepared and the desired vial or straw is identified and removed from liquid nitrogen storage. With vials, the cap should be released and replaced to prevent trapped nitrogen from expanding during warming, which can result in vial rupture. The sample is warmed by vigorous agitation in a 37°C water bath until the ice disappears. The contents are decanted or removed with a pipet, and the embryo is recovered and transferred sequentially through Dulbecco's phosphate-buffered saline containing 1.0 M propanediol-0.2 M sucrose, 0.5 M propanediol-0.2 M sucrose, no propanediol-0.2 M sucrose, HTF-HEPES with albumin, and finally fresh growth medium (HTF-bicarbonate-15% serum) equilibrated at 37°C in a 5% CO₂ in air incubator. Although not critical, thawing can also be conducted at controlled rates in a programmable freezer.

Survival of 1-cell zygotes can be evaluated conveniently by culturing for 24 hours, at which time further embryonic development should have produced 2- or 4-cell-stage embryos. In contrast, when 2- or 4-cell-stage embryos are frozen, post-thaw survival is more difficult to evaluate definitively, because the basis is the number and appearance of intact blastomeres (at least 50% of the original number required). The post-thaw culture of cleaving embryos is normally confined to such a short time interval that most embryos do not divide further.

Embryo Thaw and Transfer

Embryo thaw and transfer is conducted during natural cycles or in patients with managed cycles (e.g., with exogenous clomiphene citrate or estrogen replacement regimens and/or human chorionic gonadotropin). The timing of transfer during the cycle is critical and most often based on urinary levels of luteinizing hormone, as determined by home kit testing. Follicular ultrasound and serum progesterone levels in the periovulatory period are also important monitors for accurate timing. With 1-cell zygotes, thawing occurs 2 days after the luteinizing hormone surge, after progesterone levels have increased over baseline.

The transfer is conducted about 24 hours later with the 2- to 4-cell-stage embryo, in exact synchrony with the endometrial environment. The technique most often used for embryo transfer involves a nonsurgical, transcervical approach similar to that used in the transfer of nonfrozen embryos in the egg pickup cycle. An important concept in any embryo transfer is that higher implantation rates are expected when embryos are transferred at or near exact synchrony with the developing endometrium, and when oviductal-stage embryos are transferred to the oviduct and uterine-stage embryos to the uterus. Because of this, increased attention has been focused on the cryopreservation of blastula, because this would allow the synchronous transfer of uterine-stage embryos to the uterus.

Embryo Thaw and Transfer Results

To evaluate the contribution that cryopreservation makes to an ART program, a reference to success rates with nonfrozen embryos is useful. According to published summary data representing IVF activity in the United States in 1988²⁶ and in the United States and Canada²⁷ in 1994, 16% and 29.1% of embryo transfer cycles, respectively, resulted in a clinical pregnancy. Slightly more than 3% and 35%, respectively, of these pregnancies resulted in multiple births. In contrast, 10% and 19% of transfer cycles in 1988 and 1994, respectively, involving cryopreserved embryos resulted in clinical pregnancies, with a multiple pregnancy rate of less than 1% in 1988 (Table 2). Comparable rates for cryopreserved embryos were reported in a multicenter European survey²⁸ (11.5%) and in another US survey²⁹ (13.4%). These surveys represent results obtained after cryopreservation of principally two- to four-cell embryos in dimethylsulfoxide or propanediol. Pregnancy rates associated with the transfer of embryos (usually 2 or 3) cryopreserved at the pronuclear or 1-cell stage may be somewhat higher, at 13% to 30% (Table 3). These latter results are based on the data of individual programs as opposed to the multicenter survey reports cited above. In either event, cumulative pregnancy rates for couples in whom cryopreserved embryos are available can readily exceed 50%

TABLE 2. Embryo Thaw and Transfer Survey Results*

TABLE 3. Cryosurvival and Pregnancy Rates From Human, One-called Zygotes

As mentioned previously, cryopreservation is invariably associated with some degree of cell damage. Consequently, improvements in cryobiology and in our understanding of cryodamage and ways to minimize it should lead to improved results. This is desirable in the cryopreservation of sperm from patients with male-factor infertility, where pooled samples from multiple ejaculates used for TI might provide an effective treatment modality, although ICSI now provides a viable alternative for these patients and places relatively little demand on the efficiency of cryostorage. Another example where improved cryopreservation protocols are necessary involves the unfertilized egg. If adequate protocols were available, eggs could be stored indefinitely at low temperatures for patients with impaired or threatened ovarian function.

Although embryo cryopreservation has become a routine procedure and, hence, an integral component of an ART program, any improvement in existing protocols that translates into higher pregnancy rates would be valuable. Advantages may eventually accrue to the use of ultrarapid or vitrification approaches as opposed to conventional cryopreservation. If the relatively high rates associated with the transfer of embryos cryopreserved at the one-cell stage can be maintained or slightly improved, the cryopreservation of all embryos may become routine. This would eliminate embryo transfers during egg pickup cycles and would allow optimization of the cycle for the maximal number of mature eggs without concern for the quality of the luteal phase. Indeed, given that cryodamage is minimal, the transfer of cryopreserved embryos during natural menstrual cycles should provide the highest pregnancy rates.

Finally, another advantage to embryo cryopreservation is the ability to conduct genetic screening on embryos before their transfer. Preliminary successes have already been reported, and this capability will undoubtedly become routine in the next several years.³¹

Appreciation is expressed to Patsy Kimzey for secretarial assistance and to NIH grant RR00163 for support. ORPRC Publication No. 1828.

1. Al-Hasani S, Diedrich K, van der Ven H et al: Cryopreservation of human oocytes. Hum Reprod 2: 695, 1987

2. van Uem JFHM, Siebzehnrübl ER, Schuh B et al: Birth after cryopreservation of unfertilised oocytes. Lancet 1: 752, 1987

3. Chen C: Pregnancies after human oocyte cryopreservation. Ann NY Acad Sci 541: 541, 1988

4. Mazur P: Freezing of living cells: Mechanisms and implications. Am J Physiol 247: C125, 1984

5. Whittingham DG: Principles of embryo preservation. In: Ashwood-Smith MJ, Farrant J (eds): Low Temperature Preservation in Medicine and Biology, p 65. Tunbridge Wells, England, Pitman Medical, 1980

6. Jackowski S, Leibo SP, Mazur P: Glycerol permeabilities of fertilized and unfertilized mouse ova. J Exp Zool 212: 329, 1980

7. Rall WF, Fahy GM: Ice-free cryopreservation of mouse embryos at -196°C by vitrification. Nature 313: 573, 1985

8. Kuzan FB, Quinn P: Cryopreservation of mammalian embryos. In: Wolf DP (ed): In Vitro Fertilization and Embryo Transfer: A Manual of Basic Techniques, p 301. New York, Plenum Press, 1988

9. Polge C, Smith AU, Parkes AS: Revival of spermatozoa after vitrification and dehydration at low temperatures. Nature 164: 666, 1949

10. Sherman JK: Synopsis of the use of frozen human semen since 1964: state of the art of human semen banking. Fertil Steril 24: 397, 1973

11. Whittingham DG, Leibo SP, Mazur P: Survival of mouse embryos frozen to -196° and -269°C. Science 178: 411, 1972

12. Wilmut I: The effect of cooling rate, warming rate cryoprotective agent and stage of development on survival of mouse embryos during freezing and thawing. Life Sci 2: 1071, 1972

13. Trounson A, Mohr L: Human pregnancy following cryopreservation, thawing and transfer of an eight-cell embryo. Nature 305: 707, 1983

14. Menezo, YJR, Guerin JF, Czyba JC: Improvement of human early embryo development in vitro by coculture on monolayers of Vero cells. Biol Reprod 42: 301, 1990

15. Paria BC, Dey SK: Preimplantation embryo development in vitro: Cooperative interactions among embryos and role of growth factors. Proc Natl Acad Sci USA 87: 4756, 1990

16. Desai N, Kinzer D, Loeb A, Goldfarb J: Use of synthetic serum substitute and α-minimum essential medium for the extended culture of human embryos to the blastocyst stage. Hum Reprod 12: 328, 1997

17. The American Fertility Society: Revised new guidelines for the use of semen-donor insemination. Fertil Steril 49:211, 1988

18. Byrd W, Bradshaw K, Carr B et al: A prospective randomized study of pregnancy rates following intrauterine and intracervical insemination using frozen donor sperm. Fertil Steril 53: 521, 1990

19. Barratt CLR, Cooke S, Chauhan M, Cooke ID: A prospective randomized controlled trial comparing urinary luteinizing hormone dipsticks and basal body temperature charts with time donor insemination. Fertil Steril 52: 394, 1989

20. Silva PD, Meisch J, Schauberger CW: Intrauterine insemination of cryopreserved donor semen. Fertil Steril 52: 243, 1989

21. Patton PE, Burry KA, Thurmond A et al: Intrauterine insemination outperforms intracervical insemination in a randomized, controlled study with frozen, donor semen. Fertil Steril 57: 559, 1992

22. Larson JM, McKinney KA, Mixon BA et al: An intrauterine insemination-ready cryopreservation method compared with sperm recovery after conventional freezing and post-thaw processing. Fertil Steril 68: 143, 1997

23. Son WY, Park SE, Lee KA et al: Effects of 1,2-propanediol and freezing-thawing on the in vitro developmental capacity of human immature oocytes. Fertil Steril 66: 995, 1996

24. Lassalle B, Testart J, Renard JP: Human embryo features that influence the success of cryopreservation with the use of 1,2-propanediol. Fertil Steril 44: 645, 1985

25. Testart J, Lassalle B, Belaisch-Allart J et al: High pregnancy rate after early human embryo freezing. Fertil Steril 46: 268, 1986

26. Medical Research International and the Society for Assisted Reproductive Technology, The American Fertility Society: In vitro fertilization-embryo transfer in the United States: 1988 results from the IVF-ET registry. Fertil Steril 53:13, 1990

27. Assisted reproductive technology in the United States and Canada: 1994 results generalized from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil Steril 66:697, 1996

28. Van Steirteghem AC, Van den Abbeel E: World results of human oocyte and embryo cryopreservation. VI World Congress on In Vitro Fertilization and Alternate Assisted Reproduction Program and Abstracts, p 21. Jerusalem, April 2–7, 1989

29. Fugger EF: Clinical status of human embryo cryopreservation in the United States of America. Fertil Steril 52: 986, 1989

30. Quinn P: Success of oocyte and embryo freezing and its effect on outcome with in vitro fertilization. Semin Reprod Endocrinol 8: 272, 1990

31. Verlinsky Y, Munne S, Simpson JL et al: Current status of preimplantation diagnosis. J Assist Reprod Genet 14: 72, 1997