Results of 6139 artificial insemination cycles with donor spermatozoa

A. Botchan,1, R. Hauser, R. Gamzu, L. Yogev, G. Paz and H. Yavetz

Institute for the Study of Fertility, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, Sackler Faculty of Medicine, Tel Aviv University, Israel


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Artificial insemination by donor spermatozoa (AID) can prove a valuable treatment for a number of male factor disorders, although its success rate is variable. METHODS: Retrospective analysis of the results of 6139 cycles performed in 1001 women during an 18 year period is presented. Pregnancy rates per cycle are presented as a function of: female fertility history, treatment modalities, medication used for induction of ovulation, female age, year of treatment, consecutive cycle effect and the use of fresh versus frozen–thawed spermatozoa. RESULTS: Overall pregnancy rate of 12.6% and cumulative pregnancy rate after 12 months of treatment of 75% were achieved. Age was found to be the most important determinant for success rate. CONCLUSIONS: Since the establishment of AID treatments, the mean age of the population of women receiving treatment has increased each year. Consequently, success rate did not improve, even with the use of more sophisticated medical modalities.


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Artificial insemination with donor spermatozoa (AID) is used worldwide to treat couples with azoospermia, or severe male factor infertility, in families carrying genetic diseases which may be transmitted by husband spermatozoa, or in single women. Anonymous donor sperm banking has been a fundamental aspect of reproductive medicine for several decades. In 1987, >170 000 women in the USA were treated for infertility using artificial insemination (Critser, 1998Go). Approximately 3000 couples seek AID treatment annually in the UK, and there were >16 000 treatment cycles in 1992 (Human Fertilisation and Embryology Authority, 1994Go).

Although widely used, the success rate of AID treatments is inconsistent. Frequent problems in evaluating success rates of these treatments in many studies are: small number of treatments performed; differences in treatment protocols; heterogeneity of the population evaluated, and inclusion of patients from different clinics.

Our study was conducted to summarize the results of 1001 women treated by AID in the Institute for the Study of Infertility, Tel Aviv, Israel, during the last 18 years.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The case records of all consecutive patients who were referred for donor insemination at the Institute for the Study of Fertility, Tel Aviv Sourasky Medical Centre between 1980 and 1997, were included in this study. During this period, 6376 donor insemination cycles were performed in 1001 women. Information concerning pregnancy, abortion, termination of pregnancy or delivery was recorded from the patients' files. Information regarding pregnancy was missing from 237 cycles, therefore these cycles were eliminated from the statistical analysis, leaving 6139.

AID was performed according to regulations of the Israeli Ministry of Health, last revised in 1994 (Israeli Ministry of Health, 1994Go). Donor selection, sperm preparation and cryopreservation techniques were performed as previously described (Paz et al., 1991Go; Yavetz et al., 1991Go). Only young university students donated semen. Donors were accepted if sperm concentration was above 40x106/ml; sperm motility (at first hour) >50% motile (most in progressive motility); normal morphology >45% (World Health Organization, 1980Go); and post-thaw motility >35% motile. Sperm quality was assessed before each insemination to assure it corresponded with the expected quality. Frozen–thawed donor spermatozoa were used in 5588 (88%) cycles. Basically, a post-thaw quantity of >=4x106 progressively motile spermatozoa were used per insemination (8x106/ml). Until November 1987, 550 treatment cycles of fresh donor spermatozoa were performed. Usually, a minimum of 40x106/ml spermatozoa were used. In each cycle, age was calculated according to the differences between the year of birth and year of treatment.

Healthy women with no known gynaecological abnormalities and with regular cycles, normal hormonal profile and normal bi-manual and ultrasound findings were enrolled in the programme. Tubal patency was determined by hysterosalpingography (HSG) or laparoscopy, if there was a past history of pelvic inflammatory disease or any other event that may caused mechanical disruption. In the absence of clinical suspicion of pelvic pathology, evaluation was performed when the patient failed to conceive after 3–6 cycles of super-ovulation.

Ovulation time was usually assessed by either baseline body temperature charts, urine LH, which was tested at home, or monitored by oestradiol and progesterone serum concentration, combined with follicular diameter assessed by ultrasound. Ovulating women were usually scheduled to begin treatment during natural cycles. Ovulation augmentation was offered to women who failed to conceive after a few cycles (1–4). Clomiphene citrate (CC), either alone or with human chorionic gonadotrophin (HCG), was the first choice of treatment, later continued by human menopausal gonadotrophin (HMG) combined with HCG. Inseminations were usually performed twice in each cycle (1–4 in each). Donor spermatozoa were introduced to the cervix by a cervical cap, or by intrauterine insemination (IUI) catheter.

Information concerning the end results was usually given by a letter or by phone call at the end of the pregnancy. This was written in the patient file only with reference to: abortions, termination of pregnancies or deliveries. We are lacking information concerning the number of gestational sacs that were present at the beginning of the pregnancy and the number of babies born. The effect of the age of women on the number of insemination cycles performed until conception was calculated. Return for treatment after pregnancy mandated the beginning of new calculations. This explained why several women underwent 30–40 treatment cycles, usually in order to achieve more than one pregnancy.

The average number of cycles performed until the first pregnancy was achieved was defined as ‘first course’. The mean number of cycles performed by the same previously pregnant women until their next pregnancy was defined as ‘second course’.

Statistical analysis
Results are given as mean ± SD. {chi}2-test was used for two categorical variables for comparison between achieving pregnancy and woman's age, or year of treatment. Correlation between treatment year and woman's age was performed by Pearson's correlation. Multivariate logistic regression analysis (WALD) was used, with pregnancy rate (PR) as the dependent variable and year of treatment, woman's age and sperm origin (fresh or frozen–thawed) as the independent variables. Cumulative PR was calculated according to a published model (Cramer et al., 1979Go). These tests were calculated using PC-based SPSS (Statistical Package for Social Science) software.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Overall success rate
Results of 6139 insemination cycles performed during 18 years (1980–1997) at the Institute for the Study of Fertility are presented. During this period, 1001 women were inseminated with donor spermatozoa. Women's age ranged from 18 to 48 years, the average being 33.6 ± 6.1. Each woman had 1–41 treatment cycles (6.4 ± 5.6), usually in order to achieve more than one pregnancy. Almost half of these women (485) conceived and 442 women delivered at least once. Pregnancies were recorded in 773 cycles (12.6% per cycle); of these, 609 resulted in deliveries of at least one child (9.9% per cycle, 78.8% per pregnancy). Spontaneous abortion was recorded in 134 pregnancies (2.2% per cycle, 17.3% per pregnancy). Thirteen pregnancies (0.2% per cycle, 1.7% per pregnancy) were selectively terminated due to diagnosis of severe fetal malformations (three cases of trisomy 21, two with renal anomalies, two non-specified anomalies, and fetuses with Turner syndrome, Klinefelter syndrome, anencephalus and one with cardiac anomaly). In two cases, fetal infections were identified (cytomegalovirus and varicella-zoster virus). Chemical pregnancies were recorded in 17 cycles (0.3% per cycle, 2.2% per pregnancy). The maximum number of pregnancies per woman was eight. This woman received 34 treatment cycles in a 9 year period (two ended in deliveries, one in intrauterine fetal death, four in spontaneous abortion and one was a chemical pregnancy).

The cumulative PR calculated for 3, 6 and 12 months were 36, 53 and 75% respectively. The cumulative PR according to age groups were: (i) women <37 years: 44, 61 and 82%; (ii) women 37–40: 24, 44 and 61%; and (iii) women >40 years: 13, 22 and 50% respectively.

The effect of previous pregnancies on success rates
The PR for those who had never been pregnant previously was 380/2959 (12.8%), for those who had aborted spontaneously or intentionally 111/999 (11.1%), and for those who had delivered in the past it was 278/2173 (12.8%) (not significant).

We calculated the average number of cycles performed until the first pregnancy (‘first course’), and the average number of cycles performed by the same group of women when they returned for further treatments (‘second course’). On average these women performed 4.4 ± 3.7 versus 5.9 ± 5.3 insemination cycles until pregnancy, for ‘first’ versus ‘second’ course respectively (P = 0.023).

Treatment modalities and success rate
Cap insemination cycles resulted in 657 pregnancies in 5095 cycles (12.9%), versus 115 pregnancies in 1041 IUI cycles (11.1%) (not significant).

The effect of different regimes on success rates
Ovulation either followed spontaneously (in 2331 cycles), or was induced by medication. CC was the most popular form of treatment, either alone (1660 cycles) or combined with HCG (1011 cycles), followed by HMG stimulation (912 cycles) or HCG alone (220 cycles). PR per cycle were 13.8, 11.6, 13.6, 11.8 and 11.8% for natural cycles, and cycles stimulated by CC, HCG, CC plus HCG and HMG respectively (not significant).

Age and success rates
Women aged 18–47 years were inseminated by donor spermatozoa. The oldest pregnant woman was 44 years. A statistically significant, age-related decline in PR was found (P < 0.0001) (Table IGo). These results were more pronounced when dividing the female population into three groups: aged <37, 37–40 and >40 years. In the <37 year age group, 620 pregnancies out of 4056 cycles (15.3%) occurred, declining to 103 pregnancies out of 1190 cycles (8.7%) between ages 37 and 40 years, and to 5.5% only (49/892) in older women (P < 0.0001). This trend was more pronounced for the delivery rate per cycle. Up to 37 years, there were 513 deliveries (12.6%), at 37–40 years only 75 women delivered (6.3% per cycle) and after 40 years of age only 2.2% of the cycles ended in delivery (20/892). An opposite effect was demonstrated for abortion rate, which was found to be 14.5, 22.3 and 42.9% for women aged <37, 37–40 and >40 years respectively (P < 0.0001).


View this table:
[in this window]
[in a new window]
 
Table I. Pregnancy rate according to woman's age
 
The average number of insemination cycles performed until achieving each of the 773 pregnancies was counted according to age. The average number of cycles needed for achieving pregnancy did not increase with increasing age (r = 0.177, not significant) (Figure 1Go).



View larger version (69K):
[in this window]
[in a new window]
 
Figure 1. The average number of cycles until pregnancy, according to age. Values are mean ± SD.

 
The connection between the year of treatment and success rate
PR was evaluated per year of treatment. A reduced PR was found during an 18 year period (P = 0.005) (Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Pregnancy rate according to the year of treatment
 
During the years of the study period, the mean age of our patient population has increased, as can be seen by extrapolating age versus the year of treatment (Figure 2Go). Pearson correlation test for age and treatment year, revealed that age is the most important factor for the occurrence of pregnancy (r = 0.327, P < 0.0001). Thus, it was assumed that the higher female age was responsible for the reduction in PR with advancing years. However, when we divided the female population into two groups (group I, women <=40 years; and group II, >40 years), we could not prove that the reduction in PR resulted from increased female age. In group I, we performed 5246 treatment cycles during the years of the study period, which resulted in 723 pregnancies (13.8%). In this group, the same statistically significant decline in PR with year of treatment, as in the all study groups, was found (P = 0.024). However, in the older women (group II), we had only 48 pregnancies out of 889 cycles (5.4%), but PR did not change throughout.



View larger version (25K):
[in this window]
[in a new window]
 
Figure 2. Female age according to treatment year.

 
Consecutive cycle effect on PR
Of the 1001 women who started treatments, 132 were only treated once. However, until the 18th cycle there were >=10 women per cycle. From the 18th cycle onward, there were only a few women per cycle. PR was found to decrease with consecutive cycles, but it remained >10% even in the 15th cycle (P = 0.0001), (Table IIIGo).


View this table:
[in this window]
[in a new window]
 
Table III. Consecutive cycles effect on pregnancy rate (PR)
 
The effect of freezing and thawing on success rates
Fresh donor spermatozoa were used until November 1987. Thereafter, in accordance with the Israeli regulations (Israeli Ministry of Health, 1994Go), the use of frozen–thawed semen became routine for patients requiring AID, due to awareness of the syndrome caused by HIV. Overall, 550 fresh donor sperm insemination cycles were performed up to November 1987, which gave rise to 98 pregnancies (17.8%), whereas the other 674 pregnancies were conceived by using frozen–thawed spermatozoa in 5588 cycles (12.1%) (P < 0.0001).

By logistic regression analysis, age was found to be the major determinant affecting the occurrence of pregnancy (P < 0.00001), while the advantage of using fresh versus frozen–thawed spermatozoa entered as the second most important variable (P = 0.0008), and the year of treatment as the least important (P = 0.0026). Stepwise regression showed that the effect of age countered the effect of the treatment year (P =0.0001), as was found for the effect of fresh spermatozoa over the year of treatment (P =0.0001).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Artificial insemination by donor is a common procedure all over the world. It is usually practised in different groups of women, i.e. single women or in couples with severe male factor infertility. Today AID is usually offered to infertile couples only after failure of different testicular sperm retrieval procedures, or after fertilization failure by IVF–intracytoplasmic sperm injection, or due to genetic reasons.

The success rate of AID depends on a wide variety of factors. Some are unchangeable, such as the age of the female, while others may be under control of good medical care, such as the quality of the spermatozoa used for insemination, insemination timing, ovulation-induction protocols used, or choice between cervical insemination and IUI. Evaluation of the importance of some of these variables is within the scope of our work. However, this is not a retrospective work aiming to evaluate the different effect of each variable on success rate, but rather a retrospective analysis of continuous work performed during 18 years period, at the same clinic, with almost unchanged protocols throughout this long period.

During the 18 year period, we had information concerning 6139 treatment cycles, which resulted in 773 pregnancies, yielding PR per cycle of 12.6%. Similar success rates are published in the literature, and was not changed much during the last few decades, remaining at ~10% per treated cycle (Table IVGo). The cumulative PR was found to be high (36, 53 and 75% for 3, 6 and 12 months respectively). This higher cumulative rate is superior to the 31–37% cumulative probability achieved by different ovulation detection methods and ICI performed by Brook et al. (Brook et al., 1994Go). This is even higher than the 21, 40 and 62% probability for 3, 6 and 12 months in women <30 years of age who participated in the AID programme of the Middlesex Hospital, London (Shenfield et al., 1993Go).


View this table:
[in this window]
[in a new window]
 
Table IV. Success rate with artificial insemination by donor in different clinics
 
Patients were referred to our institute only for the purpose of performing donor sperm insemination treatments. After conceiving they returned back to their physicians. Information concerning multiple pregnancy results was missing. Since the issue of children born from sperm donors is very sensitive, we did not find it suitable to make phone calls to all these families and interview them on this sensitive issue. We are fully aware of this weakness in our data analysis, but are now unable to change it.

Spontaneous abortion rates were 17.3% per pregnancy (134 women aborted), four extrauterine pregnancies (0.5%), and 13 pregnancies were selectively terminated due to severe fetal malformations (1.7%). These spontaneous abortion and extrauterine PR do not differ from those of the normal population. Similar findings have been obtained from the results of 21 597 AID pregnancies gathered from a collaborative study of the French Centre for Cryopreservation of Eggs and Spermatozoa (CECOS Federation) (Lansac et al., 1997Go). They found 18% fetal loss, 0.9% tubal pregnancy, and 1.9% fetal malformation rate. Recently, considering only pregnancies of >=20 weeks duration, Hoy et al. showed that the overall prevalence of birth defects did not differ much between AID and control births (Hoy et al., 1999Go). However, they found a higher chromosomal abnormality rate in the AID group, compared with controls [seven of 1552 pregnancies (0.4%) compared with 13 of 7717 pregnancies (0.2%) respectively].

The effect of previous pregnancies on success rates
In the present study, past fertility and delivery history were not found to be important for present fertility potential. Intuitively, it was assumed that those women with secondary infertility would perform better than women with primary infertility. The classification we used (never been pregnant, those who had abortion and those who delivered in the past), is not comparable with ‘primary and secondary infertility’, since most women who had never been pregnant did not try to do so, and thus their fertility potential had never been tested. Moreover, those who had been pregnant previously may have conceived in their youth, and despite the indication of past fertility and probable tubal patency, this could not serve as a predictor for present fertility potential. Similar findings were reported for nulligravidae versus multigravidae (Shenfield et al., 1993Go).

In contrast to our expectations, women performed better on ‘first versus second course’, indicating that success is not a positive predictor for the next trial. This contradicts previous findings in the literature (Shenfield et al., 1993Go). However, two factors may contribute to these findings: first, those who had previously been treated by fresh spermatozoa returned later for treatment using frozen–thawed spermatozoa. A second possible explanation might have been the older age of those who returned for treatment several years after their last pregnancy.

Treatment modalities and success rate
Different treatment modalities, i.e. cap insemination versus IUI, had no effect on PR. Nevertheless, we routinely started treatment with cap insemination due to its physiological nature. Only after several failures was treatment modified to IUI. Thus, a lower fertility potential, indicated by failure to conceive with cap insemination, may contribute to lower success rate with IUI treatments. Most authors have found that IUI is more advantageous than intracervical insemination (ICI) (Hurd et al., 1993Go; Williams et al., 1995Go; Matorras et al., 1996Go). Ford et al. published an analysis of 11 studies involving frozen–thawed semen, comparing the use of IUI with ICI (Ford et al., 1997Go). They revealed that in comparison with timed ICI, IUI significantly increased the PR, both with and without gonadotropin stimulation of the female partner [odds ratio (95% confidence interval) 1.92 (1.02–3.61) and 2.63 (1.52–4.54) respectively]. Tur et al. published interesting findings of a higher incidence of multiple pregnancy after IUI (32%) than that of ICI (21%) with donor spermatozoa (Tur et al., 1997Go). Although IUI is probably superior to ICI, its disadvantages should also be considered, i.e. more invasive procedure, its use obligates sperm processing, it requires on average more donor semen, and for all these reasons the procedure is far more expensive.

The effect of different regimes on success rates
We showed that neither treating women on natural cycles, nor using different ovulation induction protocols, had any net effect on success rate (P = 0.314). Caution should be used in interpretation of these results, since the working chart of each patient demanded no or minimal intervention (no medical treatment). Failure to conceive usually prompted initiation of CC alone, or in combination with HCG, and only if pregnancy was not yet achieved were patients then treated with gonadotrophins. Therefore, this was not a controlled study comparing different regimes, but only evaluation of results achieved during this time frame.

Studies comparing the efficacy of different ovulation induction protocols usually involved IUI with fresh husband spermatozoa, and usually favour the use of HMG plus IUI over any other alternative (Ombelet et al., 1995Go). In contrast, studies involving the use of donor semen usually suffered from the same limitations as in our work (Matilsky et al., 1998Go). However, Depypere et al., who used frozen donor semen, found HMG plus IUI to be more advantageous than natural or CC-stimulated cycles (PR of 23.9, 12.5 and 13.1% respectively, P < 0.03) (Depypere et al., 1994Go).

Age and success rates
Advanced age was found to have a negative effect on PR (P < 0.0001) and on the delivery rate, and a positive effect on abortion rate. These findings are not surprising, since the importance of age in every aspect of natural and artificial reproduction is common knowledge. This includes almost every study involving AID (Shenfield et al., 1993Go; Kang and Wu, 1996Go). The decline in fertility rate with age probably results from a combination of progressive follicular depletion, decline in granulosa function and poor oocyte quality (Hughes et al., 1990Go; Navot et al., 1991Go) and reduced endometrial receptivity (Yaron et al., 1993Go).

Surprisingly, however, in those who conceived, increased age was not found to affect the number of insemination cycles needed to achieve the pregnancy (Figure 1, rGo = 0.177). This can be explained by assuming that those women who are fertile will remain so at every age, and that their fertility potential will remain less affected with advancing age.

The connection between the year of treatment and success rate
Donor sperm banking and AID treatment started more than two decades ago. It was expected that PR would increase with time, because of the advanced new technologies in ovulation induction and in timing of insemination. Nevertheless, In our hands, this did not occur, and in fact PR significantly decreased during the years of the study period (P = 0.005). This could theoretically be explained by the fact that our patients' population has grown older (Figure 2Go), and therefore PR declined. The same decline in PR was found for women <40 years of age (P = 0.024), and not for women >40 years of age (P = 0.369).

The fact that PR declined in women aged <40 years may hint at another explanation. Until November 1987, fresh donor spermatozoa were used for treatment, yielding better results (PR = 17.8%). Thereafter, frozen–thawed spermatozoa were used, with a lower PR (12.1%). If only cycles performed with frozen–thawed spermatozoa are considered, i.e. 5589 insemination cycles resulted in 675 pregnancies, an unchanged PR with advancing years is found (P < 0.05).

Consecutive cycle effect on PR
PR deteriorates with consecutive cycles (Table IIIGo). This again is a common phenomenon in the literature. However, pregnancies are recorded even after the 18th cycle, thus for a woman who desired to conceive by as simple an intervention as possible, or could not afford a more sophisticated medical procedure, such as IVF, treatment continuation is a relevant option.

The effect of freezing and thawing on success rates
PR were higher when using fresh versus frozen–thawed semen (P < 0.0001). Most authors agree that there is a reduction in PR with the use of frozen–thawed versus fresh semen (Richter et al., 1984Go; Subak et al., 1992Go). PR after freezing and thawing was found to be in positive correlation with post-thaw motility (Johnston et al., 1994Go; Clarke et al., 1997Go). However, reduced post-thaw motility is a common phenomenon after freezing, probably connected with loss of membrane integrity or the generation of reactive oxygen species (Aitken et al., 1989Go). In this regard, we previously reported that the freezing–thawing process does not impair the capability of recovered spermatozoa to bind to the zona pellucida (Gamzu et al., 1992Go). Moreover, fertilization in vitro and implantation rates seem to be similar for fresh and frozen–thawed spermatozoa (Mahadevan et al., 1982Go). These findings may hint at the major contribution of the use of higher total motile spermatozoa in fresh donor samples (>40x106/ml), probably responsible for a parallel increased in the number of cells with normal cell function.

In summary, the major merit of this paper is a description of the success rates of artificial insemination with donor spermatozoa over an extended period of time. Information regarding results of 6139 insemination cycles with donor spermatozoa performed in our institution during an 18 year period was analysed. Overall PR per cycle was 12.6%. Cumulative PR for 3, 6 and 12 months of treatment were 36, 53 and 75%, respectively. No effect was found for the woman's past fertility history. No superiority was found for different modes of treatment, or medication used for induction of ovulation. As was expected, age was found to be a major determinant for success rates. Due to the increased age of the female population attending our clinics over the years, no improvements in PR were demonstrated in our results, although progress in medical surveillance and knowledge would have been expected to have had certain positive effects. The mandatory change from the use of fresh donor spermatozoa to the use of frozen–thawed spermatozoa at the end of the 1980s had a seriously detrimental effect on PR.


    Notes
 
1 To whom correspondence should be addressed at: Institute for the Study of Fertility, Lis Maternity Hospital, 6 Weizman Street,Tel Aviv 64239, Israel. E-mail: amnonbb{at}inter.net.il Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Aitken, R.J., Clarkson, J.S. and Fishel, S. (1989) Generation of reactive oxygen species, lipid peroxidation, and human sperm function. Biol. Reprod., 41, 183–197.[Abstract]

Brook, P.F., Barratt, C.L. and Cooke, I.D. (1994) The more accurate timing of insemination with regard to ovulation does not create a significant improvement in pregnancy rates in a donor insemination program. Fertil. Steril., 61, 308–313.[ISI][Medline]

Clarke, G.N., Bourne, H., Hill, P. et al. (1997) Artificial insemination and in-vitro fertilization using donor spermatozoa: a report on 15 years of experience. Hum. Reprod., 12, 722–726.[Abstract]

Coulson, C., McLaughlin, E.A., Harris, S. et al. (1996) Randomized controlled trial of cervical cap with intracervical reservoir versus standard intracervical reservoir injection to inseminate cryopreserved donor semen. Hum. Reprod., 11, 84–87.[Abstract]

Cramer, D.W., Walker, A.M. and Schiff, I. (1979) Statistical models in evaluating the outcome of infertility therapy. Fertil. Steril., 32, 80–86.[ISI][Medline]

Critser, J.K. (1998) Current status of semen banking in the USA. Hum. Reprod., 13 (Suppl. 2), 55–67, discussion 68–69.[Abstract]

David, G., Czyglik, F., Mayaux, M.J. et al. (1980) Artificial insemination with frozen sperm: protocol, method of analysis and results for 1188 women. Br. J. Obstet. Gynaecol., 87, 1022–1028.[ISI][Medline]

Depypere, H.T., Gordts, S., Campo, R. et al. (1994) Methods to increase the success rate of artificial insemination with donor semen. Hum. Reprod., 9, 661–663.[Abstract]

Ford, W.C., Mathur, R.S. and Hull, M.G. (1997) Intrauterine insemination: is it an effective treatment for male factor infertility? Baillières Clin. Obstet. Gynaecol., 11, 691–710.[ISI][Medline]

Friedman, S. (1977) Artificial donor insemination with frozen human semen. Fertil. Steril., 28, 1230–1233.[ISI][Medline]

Gamzu, R., Yogev, L., Yavetz, H. et al. (1992) Fresh and frozen–thawed human sperm bind in a similar pattern to the zona pellucida in the hemizona assay. Fertil. Steril., 58, 1254–1256.[ISI][Medline]

Hoy, J., Venn, A., Halliday, J. et al. (1999) Perinatal and obstetric outcomes of donor insemination using cryopreserved semen in Victoria, Australia. Hum. Reprod., 14, 1760–1764.[Abstract/Free Full Text]

Hughes, E.G., Robertson, D.M., Handelsman, D.J. et al. (1990) Inhibin and estradiol responses to ovarian hyperstimulation: effects of age and predictive value for in vitro fertilization outcome. J. Clin. Endocrinol. Metab., 70, 358–364.[Abstract]

Human Fertilisation and Embryology Authority (1994) 3rd Annual Report. Human Fertilisation and Embryology Authority, London.

Hurd, W.W., Randolph, J.F., Ansbacher, R. et al. (1993) Comparison of intracervical, intrauterine, and intratubal techniques for donor insemination. Fertil. Steril., 59, 339–342.[ISI][Medline]

Israeli Ministry of Health (1994) Israeli regulations for sperm donation and sperm bank. Ministry of Justice publication.

Johnston, R.C., Kovacs, G.T., Lording, D.H. et al. (1994) Correlation of semen variables and pregnancy rates for donor insemination: a 15-year retrospective. Fertil. Steril., 61, 355–359.[ISI][Medline]

Kang, B.M. and Wu, T.C. (1996) Effect of age on intrauterine insemination with frozen donor semen. Obstet. Gynecol., 88, 93–98.[Abstract/Free Full Text]

Lansac, J., Thepot, F., Mayaux, M.J. et al. (1997) Pregnancy outcome after artificial insemination or IVF with frozen semen donor: a collaborative study of the French CECOS Federation on 21,597 pregnancies. Eur. J. Obstet. Gynecol. Reprod. Biol., 74, 223–228.[ISI][Medline]

Le Lannou, D., Gastard, E., Guivarch, A. et al. (1995) Strategies in frozen donor semen procreation. Hum. Reprod., 10, 1765–1774.[Abstract]

Lincoln, S.R., Long, C.A. and Cowan, B.D. (1995) One artificial insemination per cycle with donor sperm is as efficacious as two inseminations. J. Assist. Reprod. Genet., 12, 67–69.[ISI]

Mahadevan, M.M., Trounson, A.O., Milne, B.J. and Leeton, J.F. (1982) Influence of semen and donor factors on the success rate of artificial insemination with frozen semen. Clin. Reprod. Fertil., 1, 185–193.[Medline]

Matilsky, M., Geslevich, Y., Ben-Ami, M. et al. (1998) Two-day IUI treatment cycles are more successful than one-day IUI cycles when using frozen–thawed donor sperm. J. Androl., 19, 603–607.[Abstract/Free Full Text]

Matorras, R., Gorostiaga, A., Diez, J. et al. (1996) Intrauterine insemination with frozen sperm increases pregnancy rates in donor insemination cycles under gonadotropin stimulation. Fertil. Steril., 65, 620–625.[ISI][Medline]

Navot, D., Bergh, P.A., Williams, M.A. et al. (1991) Poor oocyte quality rather than implantation failure as a cause of age-related decline in female fertility. Lancet, 337, 1375–1377.[ISI][Medline]

Ombelet, W., Puttemans, P. and Bosmans, E. (1995) Intrauterine insemination: a first-step procedure in the algorithm of male subfertility treatment. Hum. Reprod., 10 (Suppl. 1), 90–102.[ISI][Medline]

Paz, G., Yogev, L., Gottreich, A. et al. (1991) The use of an electric freezer in human semen banking. Eur. J. Obstet. Gynecol. Reprod. Biol., 38, 141–144.[ISI][Medline]

Richter, M.A., Haning, Jr, R.V. and Shapiro, S.S. (1984) Artificial donor insemination: fresh versus frozen semen; the patient as her own control. Fertil. Steril., 41, 277–280.[ISI][Medline]

Shenfield, F., Doyle, P., Valentine, A. et al. (1993) Effects of age, gravidity and male infertility status on cumulative conception rates following artificial insemination with cryopreserved donor semen: analysis of 2998 cycles of treatment in one centre over 10 years. Hum. Reprod., 8, 60–64.[Abstract]

Subak, L.L., Adamson, G.D. and Boltz, N.L. (1992) Therapeutic donor insemination: a prospective randomized trial of fresh versus frozen sperm. Am. J. Obstet. Gynecol., 166, 1597–1604.[ISI][Medline]

Tur, R., Buxaderas, C., Martenez, F. et al. (1997) Comparison of the role of cervical and intrauterine insemination techniques on the incidence of multiple pregnancy after artificial insemination with donor sperm. J. Assist. Reprod. Genet., 14, 250–253.[ISI][Medline]

Williams, D.B., Moley, K.H., Cholewa, C. et al. (1995) Does intrauterine insemination offer an advantage to cervical cap insemination in a donor insemination program? Fertil. Steril., 63, 295–298.[ISI][Medline]

World Health Organization (1980) Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interaction, 1st edn, Cambridge University Press, Cambridge.

Yaron, Y., Botchan, A., Amit, A. et al. (1993) Endometrial receptivity: the age-related decline in pregnancy rates and the effect of ovarian function. Fertil. Steril., 60, 314–318.[ISI][Medline]

Yavetz, H., Yogev, L., Homonnai, Z. and Paz, G. (1991) Prerequisites for successful human sperm cryobanking: sperm quality and prefreezing holding time. Fertil. Steril., 55, 812–816.[ISI][Medline]

Submitted on January 1, 2001; accepted on July 10, 2001.