Confirmation of the beneficial effectsof brief coincubation of gametes inhuman in vitro fertilization

Patrick Quinn, Ph.D., Michael L. Lydic, M.D., Minh Ho, M.D., Martin Bastuba, M.D.,Fadi Hendee, M.D., and Steven A. Brody, M.D.

IVF Laboratory, Alvarado Hospital Medical Center, San Diego, California

Objective: To confirm whether brief exposure of human oocytes to spermatozoa in vitro results in equivalent fertilizationrates and possibly better quality embryos than overnight coincubation and to determine if there was a difference in outcomewith regard to the type of culture medium used.

Design: Prospective distribution of gametes between treatments in sequential patients.

Setting: Assisted reproductive technology program in private hospital.

Patient(s): Consecutively treated subfertile couples entering an infertility program.

Intervention(s): Assisted reproductive technology treatment for infertility involving oocyte retrieval and in vitro fertili-zation.

Main Outcome Measure(s): When possible, the outcome of fertilization and embryo quality were compared whengametes were coincubated for 1 hour or overnight. Two different formulations of human tubal fluid were compared in somecases.

Result(s): There was no statistically significant difference in fertilization rates between a brief or overnight coincubationof gametes or between the two treatment groups with regard to the type of culture medium used. The quality of the embryoswas significantly better in the 1-hour exposure group. The embryos in Basal XI human tubal fluid medium were ofsignificantly better morphological quality than their siblings in D31 human tubal fluid medium.

Conclusion(s): Coincubation of oocytes and spermatozoa for a shorter period produced embryos of superior morphologicalquality than the generally accepted overnight protocol. A simple glucose and phosphate-free human tubal fluid mediumresulted in early cleavage embryos of better morphological quality than a medium supplemented with glucose, taurine, andglutathione. (Fertil Sterilt 1998;69:399–402. ©1998 by American Society for Reproductive Medicine.)

Key Words: In vitro fertilization, gametes, coincubation, reactive oxygen species

The exposure of human oocytes for severalhours to the standard concentration of sperma-tozoa traditionally used for in vitro fertilizationhas been recommended and has been reportedto result in equivalent or better fertilizationrates and to result in embryos with increasedviability than when overnight gamete coincu-bation is used (1, 2). Although there was somecriticism of the lack of statistical significancein the initial report (3, 4), the subsequent reportand other brief communications on this topic(5, 6) have substantiated the fact that a reducedcoincubation time of gametes in human IVF isbeneficial or at least not detrimental to fertili-zation and subsequent development.

The objective of this study was to confirmwhether a shortened gamete coincubation pro-

tocol would be of benefit to the outcome ofIVF. Two modifications of human tubal fluid(HTF) medium (7) were also investigated todetermine if medium composition had an influ-ence on outcome.

MATERIALS AND METHODS

A series of fifteen consecutively seen pa-tients entering the IVF program at AlvaradoHospital Medical Center were eligible for par-ticipation in this study. The institutional reviewboard of the hospital approved the project, andeach couple signed an informed consent docu-ment after the objectives of the study werefully explained to them and they had agreed toparticipate. The cause of the couples’ infertility

Received May 21, 1997;revised and acceptedOctober 21, 1997.Presented at the 45thannual meeting of thePacific Coast FertilitySociety, Indian Wells,California, April 10–13,1997.Reprint requests: PatrickQuinn, Ph.D., HCLD, 1605Starling Court, Carlsbad,California 92009-5019(FAX: 760-930-9331).

PACIFIC COAST FERTILITY SOCIETYFERTILITY AND STERILITY tVOL. 69, NO. 3, MARCH 1998Copyright ©1998 American Society for Reproductive MedicinePublished by Elsevier Science Inc.Printed on acid-free paper in U.S.A.

0015-0282/98/$19.00PII S0015-0282(97)00576-1

399

was as follows: four couples had endometriosis, three hadtubal disease, one had a chromosomal abnormality, two hadovarian failure, one had male factor infertility, one hadunexplained infertility, and three had two or more causes ofinfertility. All couples had been unable to conceive for atleast one year before treatment.

The mean age (6SD) of the female patients from whomoocytes were retrieved was 31.96 2.3 years (range, 21–41years). All patients underwent controlled ovarian hyper-stimulation for anticipated oocyte retrieval (8). Only somecouples participated in the study of coincubation time ofgametes. The distribution of patients to various assistedreproductive technology (ART) protocols was as follows:one patient had GIFT, and IVF and was excluded from thecoincubation time study; three couples had intracytoplasmicsperm injection (ICSI) only and were excluded; two coupleshad their gametes coincubated for the standard overnightperiod, and one couple had their gametes coincubated for theshortened period only (all three couples were excluded). Theremaining eight couples participated in the coincubationtime study. One of these couples had ICSI, three had donatedoocytes, and the remaining four couples underwent standardIVF.

After oocyte retrieval, standard IVF procedures (previ-ously reported in [7]) were performed. In brief, collectedoocytes were washed in culture medium and placed in 30-mLdrops of fresh equilibrated medium in tissue culture dishes(Falcon #3002, Becton Dickinson Co., Franklin Lakes, NJ)under paraffin oil (BDH, catalog no. 19104). Washed sper-matozoa (25,000–40,000 motile) were added to the drops ofmedium containing oocytes 39–42 hours after the womanhad received her hCG injection before oocyte retrieval.

One hour after insemination, approximately one-half ofthe oocytes were randomly selected (every other oocyte wassequentially collected), gently washed through two dishescontaining 3 mL of equilibrated medium under paraffin oil asdescribed by Gianaroli et al. (1) and then grouped together ina fresh 30-mL drop of equilibrated culture medium. Thewashing procedure usually removed most of the accompa-nying cumulus oophorus and left the oocyte surrounded byits corona radiata only. Strenuous efforts were not made toremove cumulus cells if they could not be detached by gentleaspiration in and out of a 51⁄40 Pasteur pipette with use of thewashing medium. The other one-half of the sibling oocyteswere left in the standard insemination drop overnight (aver-age length of time, 18 hours; range, 16–20 hours).

The following morning, the oocytes were microscopicallychecked for evidence of fertilization. Two pronuclear zy-gotes that had been produced from the overnight gametecoincubation were left in insemination drops for coculture(9). The zygotes derived from the 1-hour gamete coincuba-tion protocol were left in the culture drops containing thecorona radiata cells and any cumulus cells that had accom-panied the washed oocytes. Culture of the embryos was

continued until day 2 or 3 of development, at which timethey were assessed and transferred back to the patient’suterus. The embryos replaced in patients were of the bestmorphological grade available and were selected irrespectiveof their treatment category. Therefore, most patients had amixture of embryos from the various treatment groups re-placed.

One to two hours before replacement, the morphology ofall developing embryos was assessed under a stereoscopicdissecting microscope (Model SMZ-U, Nikon, Melville,NY) and given a morphology rating (10). The embryos wereallocated to one of four categories depending on their ap-pearance. The value 4 was given to the best-quality embryoscontaining even-sized, symmetrical blastomeres with no ob-vious fragments, a value of 3 was given if they had blas-tomeres of uneven size or if the total cytoplasmic masscontained,10% fragmentation, a value of 2 was given if10%–50% of their cytoplasm was fragmented, and a value of1 was given if they showed.50% cytoplasmic fragmenta-tion.

A score was then calculated for each embryo by multi-plying the morphological value described above by the num-ber of blastomeres observed to be present in each embryo.Thus, for example, a four-cell embryo assessed as havingperfect morphology would be given an embryo score of 434 5 16, and an eight-cell embryo with,10% fragmentationwould have an embryo score of 83 3 5 24. The averageembryo score for all embryos in a particular treatment groupwas calculated and compared with the other treatmentgroups from each patient.

The two culture media used in this study were modificationsof the original HTF formulation. One medium, referred to asBasal XI HTF (7), did not have glucose and phosphate but, inaddition to the remaining components in HTF, contained eth-ylenediamine tetraacetic acid (EDTA) and glutamine. The othermedium was named D31 HTF (11) and, in addition to thecomponents in Basal XI HTF, contained 0.5 mM of glucose,1.0 mM of taurine, and 1.0 mM of glutathione. The exogenousprotein source in all media was a dialyzed form of human serumalbumin used at 5 mg/mL. All media and the human serumalbumin were obtained from Advanced Reproductive Technol-ogies, Inc. (San Clemente, CA).

The differences among average embryo scores betweenmatched groups of embryos comparing length of coincuba-tion time or the two culture media used were assessed forsignificance with use of the Wilcoxon matched-pairs signed-rank test (12). Differences in fertilization rates betweengroups of oocytes were analyzed byx2 with use of Bartlett’scorrection for continuity. In all analyses, a P value of#0.05was defined as statistically significant.

RESULTSThe fertilization rates of oocytes in the various treatment

groups are shown in Table 1. There were no statistically

400 Quinn et al. Pacific Coast Fertility Society Vol. 69, No. 3, March 1998

significant differences between the proportion of fertilizedoocytes in either the 1-hour or overnight coincubation treat-ments or the two culture media groups. The overall fertili-zation rate of all oocytes inseminated was 72% and of these,7.8% were polypronuclear.

The average morphological score of all embryos derivedfrom the short- or long-term gamete coincubation treatmentgroups on either the second or third day of development isgiven in Table 2. Overall, in ten matched groups of embryos,those derived from the 1-hour gamete coincubation treat-ment had a superior embryo morphology on day 2 or day 3of development than their siblings in the overnight coincu-bation treatment, whereas the opposite was true in three ofthe matched groups. This distribution showed a statisticallysignificant superiority (P,0.05; one-tailed test) of the 1-hourgamete coincubation treatment.

The distribution of the superiority in morphological em-bryo score after culture of embryos in Basal XI or D31 HTFmedium is shown in Table 3. Eleven matched groups ofembryos showed better morphology in Basal XI HTF, andthree groups were better in D31 HTF. This distribution wasstatistically significant (P,0.01, one-tailed test;P,0.02,

two-tailed test). Comparisons were insufficient to show anysignificant interaction between the length of gamete coincu-bation and the type of culture medium used.

Of the fifteen consecutive patients involved in this study,twelve had clinical pregnancies (80% per transfer) with animplantation rate of 0.39 (26 embryonic sacs from 67 em-bryos and oocytes transferred). Eleven of the pregnanciesdelivered. Seven of the patients had mixed transfers consist-ing of embryos derived from gametes that had been com-mingled for 1 hour or overnight, five had transfers of em-bryos from 1-hour coincubation (or ICSI) only, and theremaining three patients had embryos transferred from theovernight gamete coincubation treatment only. Three of thepregnant patients had implantation sacs that were derivedfrom embryos that had originated from the brief 1-hourgamete coincubation treatment only.

DISCUSSION

The results reported herein confirm previous observations(1, 2, 5, 6) that brief exposure of human oocytes to the rel-atively large number of spermatozoa used in a standard IVFprocedure produces the same fertilization rate as overnightgamete coincubation. In addition, the brief gamete commin-gling results in embryos of superior morphological quality,as was observed by Gianaroli et al. (1, 2). The brief 1-hourcoincubation of gametes before washing of the oocyte and itstransfer to fresh medium with only the accompanying sper-matozoa that have lodged within the corona radiata andremaining cumulus cells must be sufficient to permit theultimate fertilizing spermatozoon to firmly adhere to theoocyte and its remaining cellular investments.

This supposition is supported by the microscopic evalu-ation of Gianaroli et al. (1) that showed that the number ofspermatozoa located within the cumulus mass reaches aplateau of 15 within 15 minutes of insemination. Such a timecourse of events is also compatible with our reported successrates of the direct ovum-sperm transfer (DOST) procedure(13) and would suggest that consistent results with theDOST procedure may be achieved if the gametes are left to

T A B L E 1

Fertilization rates of oocytes coincubated with spermatozoafor 1 hour or overnight in either Basal XI HTF or D31 HTFmedium.

Length of gametecoincubation

Culture medium

TotalBasal XI HTF D31 HTF

1 hour 52/75 (69%) 27/39 (69%) 79/114 (69)(4 3 3PN)* (23 3PN)*

Overnight 50/68 (74) 25/33 (76) 75/101 (74)(5 3 3PN)* (13 3PN)*

Total 102/143 (71) 52/72 (72) 154/215 (72)

Note: All values represent no. of oocytes fertilized/total no. inseminated(%).* No. of fertilized oocytes with three pronuclei (PN).

T A B L E 2

Average morphological score of embryos on day 2 or day3 of development in matched groups of sibling oocytescoincubated with spermatozoa for 1 hour or overnight.

Day of development

Superiority of coincubation time

1 hour. overnight Overnight. 1 hour

2 16 versus 9 (1)* 8 versus 4 (1)3 22 versus 14 (9) 36 versus 21 (2)

Total (10)† (3)†

* No. of observations.† P 5 0.035 by one-tailed Wilcoxon matched-pairs signed-rank test.

T A B L E 3

Average morphological score of embryos after culture for 2or 3 days as matched groups in Basal XI HTF or D31 HTFmedium.

Superiority of

Basal XI . D31 D31 . Basal XI

22.7 versus 11.6 (11)*† 15.7 versus 10.9 (3)

* No. of observations.† P 5 0.006 by one-tailed and 0.013 by two-tailed Wilcoxon matched-pairssigned-rank test.

FERTILITY & STERILITY t 401

commingle in vitro for at least 1 hour before replacement inthe uterus. Indeed, an infertility clinic undertaking the DOSTprocedure that has adopted a 1-hour gamete coincubation invitro has obtained a 28% clinical pregnancy rate (H. Dun-away, personal communication).

The inferior embryo quality obtained from the overnightexposure of oocytes to large numbers of spermatozoa is mostlikely due to the suboptimal culture conditions created by theexcessive generation of reactive oxygen species and otherdeleterious products from sperm metabolism (1, 2). The useof culture medium supplemented with compounds that mightbe expected to ameliorate these injurious conditions (i.e.,taurine and glutathione) had no apparent beneficial effects onoutcome and, in fact, the quality of the embryos was superiorwhen the culture was continued in a simple formulation ofHTF medium that did not have glucose and phosphate butthat did contain EDTA and glutamine. This observationcorroborates previous data that showed a significantly im-proved outcome in IVF using this medium (7, 14).

The carry over of corona radiata and remaining cumuluscells to a drop of fresh medium and the continued culture ofresulting embryos in this same medium drop is a slightlymodified version of the cumulus cell coculture procedurepreviously described (9, 15). The procedure reported hereinis also somewhat like the technique described by Mansour etal. (16) in which excess spermatozoa are removed by aspi-ration of most of the medium from the insemination drop themorning after gamete coincubation and replaced with freshmedium, thus removing large numbers of spermatozoa whileleaving attached cumulus cells within the droplet with theembryo. This strategy does not, however, have the benefitsof the brief coincubation of oocytes with the spermatozoa.

We believe that the procedure detailed and used herein issuperior to both the strategies described previously (15, 16)in that it incorporates the benefits of both cumulus andcorona cell coculture and brief gamete commingling. Thecoculture strategy in conjunction with use of the simpleBasal XI HTF medium may have obscured any beneficialeffect that the more fortified D31 HTF medium may havehad on the outcome. As is generally accepted, the ultimateaim of coculture is to define and identify the beneficialcomponents to develop fully synthetic culture medium con-taining all components needed for the optimal developmentof embryos in vitro (11). The results reported herein indicatethat brief gamete coincubation may play an important role in

culture strategies that should be used with superior culturemedia for optimized success rates.

Acknowledgments:The authors thank Ms. Alma Lastrella for efficienttechnical assistance, which was essential for the completion of this study.

References1. Gianaroli L, Magli MC, Ferraretti AP, Fiorentino A, Tosti E, Panzella

S, Dale B. Reducing the time of sperm-oocyte interaction in humanin-vitro fertilization improves the implantation rate. Hum Reprod 1996;11:166–71.

2. Gianaroli L, Fiorentino A, Magli MC, Ferraretti AP, Montanaro N.Prolonged sperm-oocyte exposure and high sperm concentration affecthuman embryo viability and pregnancy rate. Hum Reprod 1996;11:2507–11.

3. Bergh C, Waldenstrom U, Hamberger L. Reducing the time of sper-matozoa-oocyte interaction in human IVF improves implatation rate.[letter]. Hum Reprod 1996;11:1809.

4. Walters DE. Reducing the time of spermatozoa-oocyte interaction inhuman IVF improves implatation rate. [letter]. Hum Reprod 1996;11:1810–1.

5. Waldenstrom U, Hamberger L, Nilsson L, Ryding M. Short-timesperm-egg exposure to prevent zona hardening. J Assist Reprod Genet1993;10:190.

6. Dirnfeld M, Goldman S, Koifman M, Gonen Y, Calderon I, AbramoviciH. Very short exposure of oocytes to normal sperm in vitro improvesimplantation rates: a prospective randomized study [abstract 0-154]. In:Proceedings of the 51st Annual Meeting of the American Society forReproductive Medicine. Seattle, WA: American Society for Reproduc-tive Medicine, 1995:S76.

7. Quinn P. Enhanced results in mouse and human embryo culture usinga modified human tubal fluid medium lacking glucose and phosphate. JAssisted Reprod Genet 1995;12:97–105.

8. Edwards RG, Brody SA. Principles and practice of assisted humanreproduction. Philadelphia: Saunders, 1995:233–84.

9. Quinn P, Margalit R. Beneficial effects of coculture with cumulus cellson blastocyst formation in a prospective trial with supernumeraryhuman embryos. J Assisted Reprod Genet 1996;13:9–14.

10. Steer CV, Mills CL, Tan SL, Campbell S, Edwards RG. The cumulativeembryo score: a predictive embryo scoring technique to select theoptimal number of embryos to transfer in an in-vitro fertilization andembryo transfer programme. Hum Reprod 1992;7:117–9.

11. Quinn P, Horstman FC. Co-culture or conditioned media—can it beimproved? In: Gomel V, Leung PCK, editors. Proceedings of the 10thWorld Congress on In Vitro Fertilization and Assisted Reproduction.Bologna: Monduzzi Editore, 1997:155–62.

12. Siegel S. Nonparametric statistics for the behavioral sciences. Tokyo:McGraw-Hill Kogakusha, 1956.

13. Quinn P, Steinberg J, Schein S, Greenberg P. Direct ovum-spermtransfer (DOST) to the uterus as a mode of treatment following failedIUI but preceding IVF or GIFT. Proceedings of the 44th AnnualMeeting of the Pacific Coast Fertility Soc 1996:P-32.

14. Quinn P, Moinipanah R, Steinberg JM, Weathersbee PS. Successfulhuman in vitro fertilization using a modified human tubal fluid mediumlacking glucose and phosphate ions. Fertil Steril 1995;63:922–4.

15. Quinn P. Use of coculture with cumulus cells in insemination mediumin human in vitro fertilization (IVF). J Assisted Reprod Genet 1994;11:270–7.

16. Mansour RT, Aboulghar MA, Serour GI, Abbass AM. Co-culture ofhuman pronucleate oocytes with their cumulus cells. Hum Reprod1994;9:1727–9.

402 Quinn et al. Pacific Coast Fertility Society Vol. 69, No. 3, March 1998