Optimizing sensitivity of the human sperm motility assay for embryo toxicity testing

Oswaldien E. Claassens1, J.Bernhard Wehr and Keith L. Harrison

Queensland Fertility Group, 225 Wickham Terrace, Brisbane 4000, Australia


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The human sperm motility assay was used as a measure of quality control in the IVF laboratory. The effects of albumin supplementation and incubation time on the sensitivity of the human sperm motility assay were investigated. The assay was also compared with mouse embryo development. The human sperm motility assay and mouse embryo development assays were performed on 25 items commonly used in IVF laboratories. Sperm motility assay was conducted after 2, 4, 6, 8, 24 and 48 h incubation intervals under standard embryology conditions. A calculated sperm motility index value <0.75 was used to indicate sperm toxicity. It was found that optimum sensitivity (P < 0.01) of the human sperm motility assay was attained in the absence of albumin after 4, 8 and 48 h incubation periods. Items identified to be sperm toxic within 8 h by the human sperm motility assay were considered to be of clinical significance due to the close concordance of these results with mouse embryo development.

Key words: albumin/IVF quality control/mouse embryo development/sperm motility assay


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Many components used for IVF can impair the growth of human embryos. These components, be they from media, disposable items or changes made to the environment, have to be identified and therefore toxicity testing is an important quality control measure contributing to a successful IVF programme. The suitability of various reagents and materials for use in human IVF can be tested using a range of bioassays. Bioassays are a reliable and objective quality control measure independent of clinical factors. These bioassays include the culture of 1-cell (Quinn et al., 1984Go) and 2-cell (Ackerman et al., 1984Go; Parinaud et al., 1986) mouse embryos, zona-free mouse embryos (Fleming et al., 1987Go), mouse hybridoma cells (Bertheussen et al., 1989Go), hamster ovary cells and human fibroblasts (Ray et al., 1987Go), hamster sperm motility (Bavister and Andrews, 1988Go) and human sperm motility assays (Edwards et al., 1980Go).

Developmental rates of mouse 1-cell and 2-cell embryos are the standard bioassay used for quality control in clinical laboratories. However, sperm motility has also been used in several studies as a tool for quality control in the IVF laboratory (Bavister and Andrews, 1988Go; Critchlow et al., 1989Go) and as a predictor of fertilization (Stovall et al., 1994Go) and pregnancy outcome (Alvarez et al., 1996Go).

Human sperm motility bioassays have been performed over varying time periods and conditions, e.g. 4 h at 40°C (Alvarez et al., 1996Go), 24 h at 37°C (Franco et al., 1993Go) and 96 h at room temperature (Critchlow et al., 1989Go). Addition of a protein source to culture medium has been shown to maintain sperm motility during prolonged incubation in vitro (Makler et al., 1984Go). In a study with hamster spermatozoa (Bavister and Andrews, 1988Go), it was shown that the sensitivity of the bioassay is greatly increased by the absence of albumin in the medium. However, the use of human sperm motility has been discouraged because human spermatozoa are very robust, probably making the bioassay less sensitive (Bavister and Andrews, 1988Go).

In the present study, the human sperm motility bioassay was used for quality control in our laboratory. In particular, the effect of albumin supplementation, semen variation, sperm concentration and incubation time on the sensitivity of the human sperm motility assay was investigated. In addition, the human sperm motility assay was compared with the mouse embryo development assay to evaluate the validity of the human sperm motility assay to predict embryo toxicity.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Items tested
The human sperm motility assay and mouse embryo development assays were performed on 25 items commonly used in IVF laboratories. Items included blood collection systems, culture tubes, media/serum filters, pipettes and syringes (full list given in Tables I and IIGoGo). The manufacturers of the items were as follows: items 1, 18–19, 21, 23 and 25: Sarstedt, Rommelsdorfer Germany; items 2 and 8: Gelman Sciences, Ann Arbor, MI, USA; items 3 and 11: Sartorius, Goettingen, Germany; item 4: Codan, Rodby, Denmark; item 5: B. Braun, Melsungen, Germany; item 6: Terumo, Elkton, MD, USA; item 7: Kimble, Vineland, NJ, USA; items 9 and 10: Millipore, Bedford, MA, USA; items 12–16 and 20: Becton Dickinson, Franklin Lakes, NJ, USA; item 17: Terumo Corporation, Leuven, Belgium; item 22: Corning Samco, Scientific Inc., San Fernando, CA, USA; and item 24: Bio-Service, Melbourne, Australia.


View this table:
[in this window]
[in a new window]
 
Table I. Sperm motility index (SMI) values after 2–48 h exposure to items 1–15, compared to the mouse embryo development (MED) assay
 

View this table:
[in this window]
[in a new window]
 
Table II. Sperm motility index (SMI) values after 2–48 h exposure to items 16–25, compared to the mouse embryo development (MED) assay
 
Precision analysis
The precision of the sperm motility assay was assessed by testing items 12 and 20 in triplicate on a day and replicated on 3 different days with different samples of the same items and different batches of culture medium to determine the inter- and intra-assay variation in the presence and absence of albumin. The effect of sperm quality (source variation) on the assay with and without albumin was evaluated using spermatozoa obtained from four donors with normal sperm parameters according to World Health Organization criteria (WHO, 1992Go) and respective exposure to item 20. The influence of sperm concentration on the precision of the assay with and without albumin was determined on item 20 by dilution of spermatozoa with human tubal fluid (HTF) culture medium (Quinn et al., 1985Go) to a final concentration of 1.25, 2.5 and 5x106/ml.

Experimental design
Fifteen of the 25 items were tested for toxicity by the sperm motility assay in the presence and absence of albumin and replicated on 2 different days. Ten of the 25 items were tested for toxicity by the sperm motility assay in the absence of albumin only, and replicated on 2 different days. The toxicity of the 25 items was also tested by mouse embryo development assay in duplicate and replicated on 2 different days. Separate investigators conducted the sperm motility assay and mouse embryo development assay and results were compared after completion of the study.

Sperm motility assay: sperm preparation
Three semen samples from a single, fertile donor were used. Specimens were produced by masturbation in a plastic container and prepared within 1 h of ejaculation. An aliquot of 1 ml semen was layered on a PureSperm gradient (prepared as described below). Two gradients were prepared simultaneously in order to obtain adequate sperm numbers.

PureSperm gradients were prepared in 15 ml plastic conical centrifuge tubes (Falcon 2095; Becton Dickinson) and warmed in a 5% CO2 incubator at 37°C for at least 3 h before use. The bottom fraction of the PureSperm gradient (90%) was prepared by mixing 27 ml of commercially supplied PureSperm solution (Nidacon Laboratories AB, Goteborg, Sweden) with 3 ml of HTF medium. The upper fraction (45%) was obtained by dilution of the 90% solution with an equal volume of HTF culture medium and stored at 4°C until used. To prepare the gradient for sperm purification, 1ml of 90% PureSperm was pipetted into the bottom of the centrifuge tube and 1 ml of 45% PureSperm was carefully layered over the bottom fraction.

The tubes were centrifuged at 200 g for 20 min. The resulting pellets were aspirated, pooled and diluted in 2 ml HTF. The pellet solution was divided into two equal aliquots and washed by centrifugation (200 g for 10 min) and resuspended with (i) HTF alone and (ii) HTF supplemented with 2 mg/ml human albumin (Albumex 4, CSL) respectively. The final concentration was adjusted to 5x106 motile spermatozoa/ml and 0.5 ml aliquots used for sperm motility studies.

Sperm motility assay: toxin assay
The tubes (items 1, 12–21 and 23) were tested by placing sperm aliquots into the tubes for up to 48 h. Filters (items 2, 3, 8–11) were tested by flushing HTF through the filters and collecting the 1st and 10th ml of filtrate. Then, a sperm aliquot was mixed (1:1) with the 1st and 10th ml of filtrate respectively, and the motility monitored. Pipettes (items 7, 22, 24 and 25) and syringes (items 4, 5 and 6) were tested by exposing spermatozoa for 5 min and continuing the incubation in culture tubes (Falcon 2003) for the motility studies. Control tubes (Falcon 2003) were set up simultaneously, containing sperm aliquots resuspended in (i) HTF alone and (ii) HTF supplemented with 2 mg/ml human albumin respectively. The culture tubes were placed in a 5% CO2-incubator at 37°C and samples taken after 2, 4, 6, 8, 24 and 48 h incubation intervals.

Sperm motility assay: evaluation
A wet preparation was prepared by placing 1 drop of sperm sample on top of a 2 mg/ml albumin-supplemented drop of medium to prevent spermatozoa from sticking to the glass for samples incubated in HTF alone. The slide was covered with a coverslip and evaluated in a blind fashion under high-power magnification (x400) within 1 min. The proportion of spermatozoa with progressive motility was estimated to the nearest 5 or 10%. For the test to be regarded as valid, >=70% of spermatozoa in the control tube had to have progressive motility (Critchlow et al., 1989Go). Sperm motility index (SMI) values were calculated by dividing the percentage progressive motility of the test by the percentage progressive motility of the control at the specified times. SMI values <0.75 are considered predictive of poor assisted reproduction technology outcome (Alvarez et al., 1996Go) and this cut-off value was used to indicate sperm toxicity in the present study.

Mouse embryo development
Items 1–25 were tested in duplicate on a day and replicated on 2 different days with approximately 30 F1 hybrid CBA/C57 2-cell mouse embryos. The assay was read as the percentage mouse embryos growing to the blastocyst stage over 72 h under standard embryology culture conditions. Results were pooled and when duplicates varied by >10%, the experiment was repeated. The laboratory standard for acceptability is >=70% development to the blastocyst stage. The outcome of the mouse embryo development test was compared with the sperm motility assay.

Statistical analysis
Results were analysed by the Wilcoxon Signed Ranks Test to determine the effect of albumin supplementation on the sperm motility assay. The average relative deviations of replicate determinations were calculated as follows:

where a and b represents the two SMI values (replicates) determined for each item and (ab) denotes the mean.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Precision analysis
The inter- and intra-assay precision of the sperm motility assay indicated that both items gave comparable SMI values when determined on the same day or on different days (Table IIIGo) and the omission of albumin had no negative effect on the assay variation (data not shown). The effect of semen source (quality) on the sperm motility assay was minimal and semen obtained from the four donors showed a similar decrease in SMI values over 2–8 h in the presence and absence of albumin (data not shown). The sperm concentration slightly influenced the sperm motility index in the absence of albumin. No effect of sperm concentration on the SMI was discernible in the presence of albumin (Figure 1Go).


View this table:
[in this window]
[in a new window]
 
Table III. Precision of the sperm motility assay conducted in the absence of albumin. Values are the mean SMI ± SD of three determinations conducted over 2–48 h incubation time
 


View larger version (14K):
[in this window]
[in a new window]
 
Figure 1. Effect of sperm concentration and the presence (•) or absence ({circ}) of albumin on the mean sperm motility index (SMI) values after 8 h incubation. Error bars, if not obscured by the symbols, represent the SD of three determinations.

 
Human sperm motility assay
In the presence of albumin, only one out of 15 items (i.e. item 1, Sarstedt Monovette Serum Gel Tube) was identified to be sperm toxic (i.e. SMI value <0.75) after 2 h (Table IGo). Items 2 and 3 (1st ml of filtrates from Gelman Acrocap Filter and Sartorius Sartolab P Filter respectively) manifested sperm toxicity after 48 h in the presence of albumin (Table IGo). Twelve items (items 4–15) had no sperm toxic effect in the presence of albumin. Mouse embryo development showed two of the 15 items, i.e. 1 and 3 (Sarstedt Monovette Serum Gel Tube and 1st ml of filtrate from Sartorius Sartolab P Filter respectively) to be toxic (Table IGo).

In the absence of albumin supplementation, item 1 was also identified to be sperm toxic after 2 h incubation (Table IGo). Items 2 and 3 (1st ml of filtrates from Gelman Acrocap Filter and Sartorius Sartolab P Filter respectively) revealed sperm toxicity after 8 h incubation. Items 4, 5 and 6 (Primo, Braun and Terumo syringes respectively) showed sperm toxicity after 24–48 h of incubation. In the absence of albumin, nine items (items 7–15) showed no sperm toxic effect at any time interval.

The addition of albumin to culture medium had a statistically significant effect (P < 0.01) on the sperm motility assay after 4, 8 and 48 h incubation periods. The average relative deviation of replicate determinations varied between 0 and 15.2%.

Since albumin was lowering the sensitivity of the assay, an additional 10 items (items 16–25) were tested by the human sperm motility assay in the absence of albumin (Table IIGo). Four of the items were sperm toxic with SMI values <0.75 (items 16–19) after 2 h. Two additional items (items 20 and 21) had SMI values <0.75 after 6 h. Items 22, 23 and items 24, 25 had SMI values <0.75 after 24 and 48 h incubation respectively. The average relative deviation of replicate determinations varied between 5.7 and 21.6%. Six of the 10 items (items 16–20 and 25) were toxic to mouse embryos.

Pooling the results of items 1–25 (Tables I and IIGoGo) tested in the absence of albumin only, the human sperm motility assay revealed a total of 5, 7, 9, 12 and 16 of the items to be sperm toxic after 2–4, 6, 8, 24 and 48 h respectively. Eight of the 25 items were toxic to mouse embryo development. Therefore, a strong agreement was found between the outcome of the sperm motility assay at 8 h and mouse embryo development.

Evaluating the human sperm motility assay after 8 h in the absence of albumin, revealed that the blood collection tubes (items 1, 16–21) and the 1st ml of filtrates from two filter types (items 2 and 3) were sperm toxic and the culture tubes (items 12–15), pipettes (items 7, 22, 24, 25), syringes (items 4–6), one filter type (item 9) and 10th ml of filtrates from all three filter types tested (items 8, 10 and 11) were not toxic. This was in agreement with the outcome of the mouse embryo development assay.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Precision analysis
In the present study, sperm motility immediately after preparation was consistently >90% and the control samples maintained a level of >=70% motility after 48 h, which is in agreement with the life span of human sperm motility. Following the preparation and isolation of motile spermatozoa, it has been recorded that 25% of these spermatozoa lose their motility after 2.9 days (Mansour et al., 1995Go). SMI values slightly greater than 1.0 (Tables I and IIGoGo) can be attributed to the randomized blind fashion of motility evaluation. The average relative deviation between replicate SMI values (Tables I and IIGoGo) indicates good repeatability of the proposed method.

The inter- and intra-assay variation gave coefficients of variation for SMI values ranging between 0–14% (Table IIIGo) for item 12 over the initial 24 h of incubation, indicating an acceptable precision of the method. For item 20, the coefficient of variation was larger, since the SMI values decreased rapidly over time due to the highly sperm toxic nature of the item. The rate of change over the dynamic region (0–8 h) gave values ranging from –0.1 to –0.12 SMI units per hour for within day and between days analyses in the presence and absence of albumin, again indicating good precision. In addition, the fast decrease in SMI values allowed us rapidly to distinguish (within 2–3 h) sperm toxic from non-toxic items.

The minimal effect of semen source on the outcome of the sperm motility assay seems to indicate that a prior standardization of semen samples is not an absolute requirement. Therefore, semen samples with normal sperm parameters can be successfully employed for toxicity testing. Since it is likely that cryopreserved semen samples may respond differently in the sperm motility assay compared with fresh semen samples, the effect of semen quality may become more pronounced when using cryopreserved semen.

The effect of sperm concentration on the sensitivity of the sperm motility assay in the absence of albumin was expected. Since spermatozoa can bind toxic components, the number of toxin molecules bound per spermatozoon is higher with lower sperm concentration, resulting in decreased SMI values in the absence of albumin. In the presence of albumin, the toxic compounds may be preferentially adsorbed by albumin, thereby delaying or preventing expression of sperm toxicity and reducing sensitivity of the assay. The results of this part of the study indicate that the sperm concentration needs to be standardized to obtain comparable results.

Human sperm motility assay
Subjecting items 1–15 to the sperm motility assay in the presence or absence of albumin indicated that albumin supplementation of culture medium had a statistically significant masking effect (P < 0.01) on the ability of the human sperm motility assay to detect toxic items after 4, 8 and 48 h incubation periods. This is in agreement with the finding (Bavister and Andrews, 1988Go) that protein supplementation of the culture medium greatly decreased the sensitivity of the hamster sperm motility bioassay. It could be reasoned that a stressed sperm population is needed to detect toxins. Albumin may possibly bind the toxic component, delaying expression of toxicity. In a study aimed to evaluate the efficiency of a chemically defined protein free medium (Parinaud et al., 1998Go), no difference was found in sperm motility after 30 min and 4 h, but motility was significantly reduced in protein-free medium after 24 h compared to albumin-supplemented medium. Since the human albumin molecule contains 17 disulphide bridges that may have free-radical quenching abilities, lipid peroxidation of the sperm cell membrane may be minimized in the presence of albumin. Loss of motility by spermatozoa in vitro is influenced by the rate of endogenous lipid peroxidation (Alvarez and Storey, 1985Go). Albumin may also adsorb or bind the toxic substance thereby reducing any sperm toxic effect. Therefore, it is proposed that the sperm motility assay be conducted in the absence of albumin to attain maximum sensitivity of the assay. It was also recently demonstrated that the sensitivity of a mouse bioassay system to detect embryology contaminants is improved if the culture medium is not supplemented with albumin (Van den Abbeel et al., 1999Go) but this observation is not supported by other findings (Quinn et al., 1998Go).

The present study showed that toxic items could be identified after 2 h by the human sperm motility assay and, in the absence of albumin, all these items also manifested toxicity to mouse embryos. Items identified to be sperm toxic within 8 h by the human sperm motility assay were considered to be of clinical significance due to its close agreement with the outcome of the mouse embryo development test. With increasing incubation time more items were identified to be sperm toxic than embryo toxic. Very low toxin levels, manifesting themselves only after a prolonged sperm incubation period could explain the gradual increase of sperm toxic items with increasing exposure time.

The blood collection tubes found to be toxic could be attributed to rubber components, blood clotting agents and tube coating, e.g. silicon, lithium and heparin. In our laboratory, we have observed that tissue culture tubes without any additives (Falcon 2037 or Falcon 2095) are equally effective as blood collection systems when using autologous serum as medium supplement.

The human sperm motility assay, in the absence of albumin, clearly indicated that the toxic component in the filters (e.g. surfactants and remnants from ethylene oxide sterilization) was sufficiently removed after the 10th ml of medium was passed through the filters (Table IGo). Consequently, the filters were considered safe for use in IVF. The importance of discarding the initial filtrate of filters has been emphasized (Harrison et al., 1990Go). Mouse embryo development failed to confirm the sperm toxic nature of one filter type (item 2) in the present study and this could be attributed to a different mode of action of toxins. Indeed, it has been concluded (Gorrill et al., 1991Go) that the mouse bioassay is appropriate to determine grossly embryotoxic conditions (comparison of 100% versus 0% development) whereas more subtle and intermediate effects are detectable by the sperm motility assay.

All three syringes (items 4, 5 and 6) were sperm toxic after 48 h incubation in the absence of albumin. This result was considered insignificant due to the delayed sperm toxic expression and because mouse embryo development was not inhibited after exposure to the syringes. The observed sperm toxicity of syringes could be explained by contact of spermatozoa with the rubber of the syringe plunger during the experimental set-up. It has been shown that spermatozoa are extremely sensitive to rubber components (Critchlow et al., 1989Go), and one commonly available disposable plastic syringe was found to be sperm and embryo toxic (De Ziegler et al., 1987Go).

The use of human instead of hamster spermatozoa for IVF quality control is advantageous, as no laboratory animals are required, which is especially convenient for those laboratories where these animals are not permitted (e.g. hamsters in Australia). The sperm motility assay is less time consuming (results obtainable within 1 working day), requiring less technical skill and equipment compared to mouse embryo development, especially for IVF laboratories where the mouse model is difficult to employ. However, simultaneous employment of the human sperm motility and mouse embryo development assays for IVF quality control would give a good indication on the suitability of any new items/media to be used since these two assays seem to both complement and confirm each other.

In conclusion, optimum sensitivity of the human sperm motility assay is obtained in the absence of albumin. The human sperm motility assay can also serve as an accurate predictor of embryo toxicity interpreted as soon as 2 h after preparation. The human sperm motility assay therefore provides an effective and convenient measure of IVF quality control, also minimizing the killing of laboratory animals.


    Acknowledgments
 
The authors would like to thank Ms J.Priest, School of Land and Food, The University of Queensland, for assistance with statistical analyses and Dr R.Menkveld, Reproductive Biology Unit, Tygerberg Hospital, South Africa for valuable opinions.


    Notes
 
1 To whom correspondence should be addressed at: Queensland Fertility Group, 225 Wickham Terrace, Brisbane 4000, Australia Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ackerman, S., Swanson, R.J., Stokes, G. and Veeck, L. (1984) Culture of preimplantation mouse embryos as a quality control assay for human in vitro fertilization. Gamete Res., 9, 145–152.[ISI]

Alvarez, J.G. and Storey, B.T. (1985) Spontaneous lipid peroxidation in rabbit and mouse epididymal spermatozoa: dependence of rate on temperature and oxygen concentration. Biol. Reprod., 32, 342–351.[Abstract]

Alvarez, J.G., Minaretzis, D., Barrett, C.B. et al. (1996) The sperm stress test: a novel test that predicts pregnancy in assisted reproductive technologies. Fertil. Steril., 65, 400–405.[ISI][Medline]

Bavister, B. and Andrews, J.C. (1988) A rapid sperm motility bioassay procedure for quality control testing of water and culture media. J. In Vitro Fert. Embryo Transf., 5, 67–75.[ISI][Medline]

Bertheussen, K., Holst, N., Forsdahl, F. et al. (1989) A new cell culture assay for quality control in IVF. Hum. Reprod., 4, 531–535.[Abstract]

Critchlow, J.D., Matson, P.L., Newman, M.C. et al. (1989) Quality control in an in-vitro fertilization laboratory: use of human sperm survival studies. Hum. Reprod., 4, 545–549.[Abstract]

De Ziegler, D., Cedars, M.I., Hamilton, F. et al. (1987) Factors influencing maintenance of sperm motility during in vitro processing. Fertil. Steril., 48, 816–820.[ISI][Medline]

Edwards, R.G., Steptoe, P.C. and Purdy, J.M. (1980) Establishing full-term human pregnancies using cleaving embryos grown in vitro. Br. J. Obstet. Gynaecol., 87, 737–756.[ISI][Medline]

Fleming, T.P., Pratt, H.P.M. and Braude, P.R. (1987) The use of mouse preimplantation embryos for quality control of culture reagents in human in vitro fertilization programs: a cautionary note. Fertil. Steril., 47, 858–860.[ISI][Medline]

Franco, J.G., Mauri, A.L., Petersen, C.G. et al. (1993) Efficacy of the sperm survival test for the prediction of oocyte fertilization in culture. Hum. Reprod., 8, 916–918.[Abstract]

Gorrill, M.J., Rinehart, J.S., Tamhane, A.C. et al. (1991) Comparison of the hamster sperm motility assay to the mouse one-cell and two-cell embryo bioassays as quality control tests for in vitro fertilization. Fertil. Steril., 55, 345–354.[ISI][Medline]

Harrison, K.L., Sherrin, D.A., Hawthorne, T.A. et al. (1990) Embryotoxicity of micropore filters used in liquid sterilization. J. In Vitro Fert. Embryo Transf., 7, 347–350.[ISI][Medline]

Makler, A., Fisher, M., Murillo, O. et al. (1984) Factors affecting sperm motility. IX. Survival of spermatozoa in various biological media and under different gaseous compositions. Fertil. Steril., 41, 428–432.[ISI][Medline]

Mansour, R.T., Aboulghar, M.A., Serour, G.I. et al. (1995) The life span of sperm motility and pattern in cumulus coculture. Fertil. Steril., 63, 660–662.[ISI][Medline]

Parinaud, J., Reme, J-M., Monrozies, X. et al. (1987) Mouse system quality control is necessary before the use of new material for in vitro fertilization and embryo transfer. J. In Vitro Fert. Embryo Transf., 4, 56–58.[ISI][Medline]

Parinaud, J., Milhet, P., Vieitez, G. et al. (1998) Human sperm capacitation and in-vitro fertilization in a chemically defined and protein-free medium SMART1. Hum. Reprod., 13, 2579–2582.[Abstract]

Quinn, P., Warnes, G.M., Kerin, J.F. et al. (1984) Culture factors in relation to the success of human in vitro fertilization and embryo transfer. Fertil. Steril., 41, 202.[ISI][Medline]

Quinn, P., Kerin, J.F. and Warnes, G.M. (1985) Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubal fluid. Fertil. Steril., 44, 493.[ISI][Medline]

Quinn, P., Keel, B.A., Serafy, N.T. et al. (1998) Results of the American Association of Bioanalysts (AAB) embryology proficiency testing (PT) program. Fertil. Steril., 70, S100 (O-267).

Ray, B.D., McDermott, A. Wardle, P.G. et al. (1987) In vitro fertilization: fertilization failure due to toxic catheters. J. In Vitro Fert. Embryo Transf., 4, 58–61.[ISI][Medline]

Stovall, D.W., Guzick, D.S., Berga, S.L. et al. (1994) Sperm recovery and survival: two tests that predict in vitro fertilization outcome. Fertil. Steril., 62, 1244–1249.[ISI][Medline]

Van den Abbeel, E., Vitrier, S., Lebrun, F. et al. (1999) Optimized mouse bioassays for the detection of embryology contaminants. Hum. Reprod., 14, 114 (O-205).[Abstract/Free Full Text]

World Health Organization (1992) Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interaction, 3rd edn. Cambridge University Press, Cambridge.

Submitted on December 16, 1999; accepted on March 31, 2000.