Hum. Reprod. Advance Access originally published online on June 21, 2007
Human Reproduction 2007 22(8):2243-2248; doi:10.1093/humrep/dem165
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Microbial contamination of embryo cultures in an ART laboratory: sources and management
1 Department of Reproductive Medicine, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands 2 Department of Medical Microbiology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands
3 Correspondence address. Tel: +31 30 2507524; Fax: +31 30 2505433; E-mail: p.kastrop{at}umcutrecht.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
BACKGROUND: Although rare, microbial contamination of culture dishes occasionally occurs in our IVF/ICSI programme. Despite stringent culture conditions and the use of medium containing penicillin and streptomycin, an increasing number of infections was observed once they were routinely recorded. In this study, 95 cases of contaminated culture dishes were examined, in an attempt to identify possible causes.
METHODS: Relevant data of the IVF/ICSI treatment cycles and the micro-organisms isolated from the infected culture dishes were evaluated retrospectively.
RESULTS: Infections were observed only in IVF culture dishes and never after applying intra-cytoplasmic sperm injection. Identification of the contaminating micro-organisms showed that infections were mainly caused by Escherichia coli (n = 56; 58.9%) and Candida species (n = 24; 25.3%). Of the E. coli strains isolated, 41 (73.2%) appeared to be resistant to both antibiotics used in the culture medium and 13 (23.2%) appeared to resist either penicillin or streptomycin. Of all bacterial strains isolated, the resistances were 61.4% to both and 30% to one of the antibiotics used.
CONCLUSIONS: Applying the ICSI procedure prevents colonization of the culture dishes by micro-organisms. Infections in IVF culture dishes are mainly caused by bacterial strains insensitive to the antibiotics used or due to yeast colonization by Candida species which frequently reside in the vagina.
Key words: embryo culture/infections/IVF/micro-organisms
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
In the ART laboratory, the culture of gametes and embryos requires stringent culture conditions. A high standard of hygiene, cleaning and waste disposal must be observed in order to avoid infection of staff and patients and contamination of the culture dishes and equipment. Every step in the laboratory procedures and manipulations must be carried out with a rigorous discipline of aseptic technique (Elder et al., 2005
). However, neither the vagina nor an ejaculate can be considered as sterile environments. Therefore, extreme care should be taken to minimize the risk of transferring micro-organisms when carrying out clinical or laboratory procedures. The majority of ART laboratories use culture media containing antibiotics to minimize the risks of microbial growth. Nevertheless, occasionally micro-organisms colonize culture dishes of oocytes and embryos. The exact frequency of these microbial contaminations or infections is unknown. The very limited number of publications and case reports dealing with this subject (Ng et al., 1987; Ben-Chetrit et al., 1996; Cottell et al., 1996; Burrello et al., 2004) might suggest that infections in the culture dishes form a negligible risk. Ben-Chetrit et al. (1996) reported five dishes with yeast colonization in 729 cycles (0.69%), and Cottel et al. (1996) reported six cases where microbial contamination of the embryo culture medium was observed after 1691 oocyte collections (0.35%). Extrapolating this frequency to the number of IVF cycles performed in Europe (Nygren and Nyboe Andersen, 2002) indicates that in Europe alone, many hundreds of IVF culture dish contamination occur every year. In our experience, over a period of 8 years, 95 infections of IVF culture dishes were observed after 13 977 laboratory cycles (0.68%). Since 1997, microbial contamination of culture dishes has been routinely registered due to the implementation of a quality management system. In the following years, an increasing number of infections were observed. In an attempt to ascertain the possible causes of the microbial contamination, strains isolated from the culture dishes were identified and evaluated together with all relevant data. As far as we are aware, this is the first study to address the long-term evaluation of microbial contamination of culture dishes in an ART laboratory.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Subjects
Over an 8-year period (January 1997—December 2004), 95 cases were recorded in which some or all culture dishes were contaminated with micro-organisms. All infections were proven by microbiologic investigations. Relevant details of both the fertility treatment cycles and the isolated and identified microbial strains were investigated.
Semen preparation
Freshly ejaculated semen samples were collected in sterile pots. After liquefaction, semen concentration and motility were assessed microscopically. The semen samples were diluted 1:1 with human tubal fluid (HTF) medium (Cambrex Bio Science, Verviers, Belgium), containing 31.5 IU/ml penicillin, 10 µg/ml streptomycin and supplemented with 10% of a pasteurized human plasma-protein solution (Red Cross Central Blood Transfusion Laboratory, Amsterdam, The Netherlands). After centrifugation over a discontinuous gradient of two layers of PureSperm (Nidacon, Gothenburg, Sweden), sperm pellets were washed twice by resuspension and centrifugation. If possible, spermatozoa were allowed to swim up after the first washing procedure. Sperm concentration and motility of washed sperm suspensions or swim-up fractions were assessed by means of a Makler chamber.
Frozen-thawed semen samples were directly centrifuged over the PureSperm gradient. Pellets were always washed twice and assessed as described for fresh semen samples.
Remaining fractions of semen suspensions used for either insemination in IVF treatment cycles or injection in ICSI treatment cycles were stored at room temperature for 48 h.
Follicular stimulation
Routinely, controlled ovarian stimulation consisted of a long gonadotrophin-releasing hormone analogue (GnRH-a) (Lucrin; Abbott, The Netherlands) protocol, starting with pituitary desensitization in the midluteal phase. After menstruation, the follicular stimulation was started with pure follicle-stimulating hormone (FSH) (Metrodin; Serono Benelux BV, The Netherlands) during the first couple of years, or follitropin alpha (rFSH) (Gonal-F; Serono Benelux BV) subsequently. The stimulation was monitored by transvaginal ultrasound and blood samples for serum estradiol measurement. Ovulation was induced by the administration of human chorionic gonadotrophin (hCG; 10 000 IU) (Profasi; Serono Benelux BV).
Collection and processing of oocytes
Oocytes were recovered by transvaginal ultrasound-guided follicle aspiration, 34–36 h after hCG administration. Before retrieval, the vagina was swabbed with sterile gauze soaked in sterile saline. Follicular aspirates were collected in sterile tubes containing 1 ml of HEPES-buffered HTF medium (Cambrex Bio Science), containing the same antibiotic concentrations and supplemented with 20 µ/ml Heparin. Cumulus–oocyte complexes isolated from the follicular aspirates were washed in HEPES-buffered HTF medium supplemented with 10% of a pasteurized human plasma-protein as well as HTF medium supplemented with 10% of a pasteurized human plasma-protein (HTF culture medium). Oocytes retrieved were incubated in 1 ml HTF culture medium with a maximum of five oocytes per dish. After maturation in vitro for 
4 h, oocytes were inseminated (IVF cycles) or injected (ICSI cycles) between 40 and 41 h post-hCG administration. Generally, in the IVF cycles an insemination concentration between 2 x 105 and 1 x 106 motile spermatozoa per ml were used. In IVF cycles with donor semen, the insemination concentration was 1 x 105 motile spermatozoa per ml. The ICSI procedure has been described in detail previously (Kastrop et al., 1999). About 16–18 h after insemination or microinjection, oocytes were examined for the presence of pronuclei and polar bodies. Fertilization was considered normal when two clearly distinct pronuclei were present. If a single pronucleus was observed, a second evaluation was carried out 4–6 h later. Normally fertilized oocytes were pooled and cultured in 1 ml HTF culture medium. Cleavage to embryos was assessed after a 48 h culture period.
Culture condition and assessment of infection
Gametes and embryos were cultured in HTF culture medium, i.e. HTF medium, derived from Quinn et al. (1985) and manufactured under orders by Cambrex Bio Science, containing 31.5 IU/ml penicillin, 10 µg/ml streptomycin and supplemented with 10% of a pasteurized human plasma-protein solution (Red Cross Central Blood Transfusion Laboratory).
Whenever a culture dish appeared turbid and was suspected of microbial contamination, the HTF culture medium was collected and sent to the Department of Medical Microbiology for microbiologic examination. If infections were discovered, semen suspensions used for the insemination of the oocytes were also subjected to microbiologic examination, if they were available.
Regardless of embryo quality, embryos descended from contaminated dishes were never selected for transfer to the patients' uterus.
Microbiologic examination
IVF tissue cultures suspected microscopically of bacterial or fungal contamination were sampled and sent to the microbiology laboratory for further investigation. These samples were cultured according to standard procedures. Samples were inoculated on blood agar plates and MacConkey agar plates for detection of Enterobacteriaceae, and on mout agar plates to detect yeast or fungal infections. Plates were incubated aerobically at 37°C and read within 24 h. Identification and antibiotic susceptibility testing were performed using Vitek with AMS R09.1 software (bioMérieux, Marcy l'Etoile, France) and CHROM agar plates (bioMérieux). If still available, semen, which was used in the IVF procedure, was cultured by the same method.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
During an 8-year period, from 1997 till 2004, 13 977 IVF and ICSI cycles were performed. In 95 cycles (0.68%), microbial contamination was observed in some or all culture dishes (Table 1). However, in 2926 ICSI cycles, no infection was ever observed. All 95 cases concerned IVF cycles with 85 couples involved, giving a mean frequency of microbial contamination of 0.86% in 11 051 IVF cycles (Table 1). After a steady increase of this frequency during 6 years up to 1.3% in 2002, an inexplicable drop to 0.4% was observed in the following year.
|
Identification of the contaminating micro-organisms showed that infections were mainly caused by Escherichia coli (n = 56; 58.9%) and Candida species (n = 24; 25.3%) (Table 2). One microbial contamination was caused by the mould Aspergilles terreus, probably due to contaminated oil used for the culture overlays. The remaining 14 infections were caused by six different bacterial strains. Remarkably, eight of these bacterial contaminations were observed in IVF cycles using donor semen. However, in only three cases, it was proved that the contaminations originated from the donor semen used. Overall, the incidence of microbial contamination in IVF cycles with donor semen was 1.32% (13 infections in 986 cycles), which was not significantly different (
2-test; P < 0.2) from the incidence of 0.81% (82 infections in 10 065 cycles) in IVF cycles using the partner's semen.
|
There were 39 (41.1%) infections detected during the pronuclear examination at day 1, whereas 56 (58.9%) were observed at day 2–4, at the time of embryo selection or embryo replacement (Table 3). Frequency distribution of infections caused by E. coli, Candida species or other micro-organisms differ when day 1 is compared to day 2–4. The majority (60.7%) of infections caused by E. coli were detected on day 1, whereas Candida species were mainly (87.5%) observed at day 2–4.
|
In 50 cases (37 at day 1 and 13 at day 2–4), the semen suspension used for fertilization was still available for microbiologic examination. In 16 (32%) cases (13 at day 1 and 3 at day 2–4), the bacterial strain was also detected in the semen.
Microbiologic examination revealed that 70 cases (73.7%) of the microbial contamination observed were caused by a single bacterial strain. In 56 cases (80%) of bacterial infection, E. coli was isolated, of which 41 (73.2%) appeared to be resistant to both antibiotics used in the culture medium and 13 (23.2%) were resistant to either penicillin or streptomycin. Of all bacterial strains isolated, the resistances were 61.4% to both and 30.0% to one of the antibiotics used (Table 4).
|
In order to avoid recurrence of infections in a subsequent cycle, the couples involved received antibiotic prophylaxis up to the day of the ovum pick-up. However, eight (9.4%) couples showed recurrent infection in one (six couples) or two (two couples) subsequent treatment cycles. In total, the 85 couples involved underwent 330 treatment cycles. Overall, 45 (52.9%) couples conceived, of which 37 (43.5%) were in a cycle following the cycle with microbial contamination in some or all culture dishes.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Ever since the implementation of a quality management system in our daily practice, every adverse event, incident, mistake, deviation, deficiency and complaint has been registered and analysed. The occurrence of microbial contamination of culture dishes is considered an adverse event, which not only should be registered in order to detect trends, but also analysed in order to maximize the possibility of taking active measures aiming to prevent such events. As part of this policy, all culture dishes suspected of microbial contamination are examined by standard microbiologic examination. If available, the processed semen sample used for fertilization of the oocytes should also be investigated. However, these samples are kept for only 48 h after use, and consequently they were no longer available when infections were detected during embryo assessment or embryo replacement on day 2–4.
Another aspect of our policy is to avoid replacing embryos derived from infected culture dishes. This is based on the ignorance and anxiety concerning the possibility that micro-organisms could be carried into the oocyte by spermatozoa during fertilization, and the danger of introducing micro-organisms into the uterus during embryo transfer and causing infections. We experienced that in those cases where embryo culture dishes were obviously contaminated with bacteria, the quality of the developing embryos was poor. On the other hand, when culture dishes were colonized by yeast, embryo quality appeared not to be compromised. This substantiates earlier observations made by Ben-Chetrit et al. (1996) who proposed that yeast has no detrimental effect on human embryos, possibly due to a co-culture effect of the yeast. However, substantial data sets on the transfer of embryos cultured in the presence of yeast are currently absent and therefore we refrain from their transfer.
After starting to register and evaluate infections, an increasing number of infections was observed in the following years. Together with a hospital hygienist, several investigations were performed during the ovum pick-up as well as in the laboratory in order to identify possible causes. However, no explanation could be found regarding whether this was due to bad practice during ovum pick-up or laboratory procedures. Despite intensified measures, e.g. to exclude the perforation of the cover of the ultrasound probe used during oocyte retrieval and improper disinfection of the probe as possible sources of contamination, the number of infections increased in subsequent years. Therefore, an increasing resistance of bacterial strains against the antibiotics used, prolonged embryo culture or just coincidence remained as possible explanations. In order to evaluate this phenomenon, literature was studied. However, the very limited number of publications dealing with this subject, reporting incidences of 0.35% (Cottell et al., 1996) and 0. 69% (Ben-Chetrit et al., 1996), might suggest that infections in the culture dishes form a negligible risk. Over the 8-year period evaluated in this study, 95 microbial contaminations of culture dishes were observed in 13 977 IVF and ICSI laboratory cycles, indicating an incidence of 0.68%. However, this incidence increases to 0.86% when taking into account the fact that all infections were observed in IVF cycles. No infections have ever been observed in ICSI culture dishes, which implies a significantly (2-test; P < 0.001) reduced risk of colonization of the culture dishes by micro-organisms, when the ICSI procedure has been applied. This may be due to the isolation of single motile spermatozoa from the PVP solution during the ICSI procedure. Although both sperm samples for IVF and ICSI were prepared by discontinuous gradient centrifugation, which removes most of the micro-organisms in the final sperm sample, the selection of a single sperm for injection may reduce the risk of contamination additionally.
On the other hand, the reduced risk of infections during the ICSI procedure could also be due to the removal of cumulus and corona cells, the use of hyaluronidase or the extensive washing steps during the denudation procedure. In a veterinary embryo production in vitro programme, hyaluronidase treatment has been applied for virus decontamination of embryos (Bureau et al., 2005). In an attempt to test the anti-microbial effect of hyaluronidase, strains of E. coli and Candida species, previously isolated from infected culture dishes, were inoculated and cultured on hyaluronidase impregnated agar plates. However, growth was not affected compared with control plates (data not shown), indicating that the use of hyaluronidase appeared to have no preventative effect on microbial contamination.
In order to prevent recurrent infection, oocytes isolated in a next treatment cycle were washed additionally and the majority of cumulus cells were removed. However, contaminated culture dishes were observed in a subsequent treatment cycle of eight couples. In five cases, the infections were always observed at day 1 and sperm samples used for insemination were available for examination. As no micro-organisms were detected in any those sperm samples, the efficacy of the measures applied can be queried.
Although not statistically significant, the incidence of infection in IVF cycles using donor semen appears to be increased. Moreover, 8 of the 14 infections not caused by E. coli or Candida species as well as the Aspergillus infection were observed when donor semen was used. However, only three infections could be traced back to contaminated donor semen. In two cases, no microbial contamination of the donor semen could be detected, and in the remaining three cases, no semen sample was available due to the fact that the infections were observed during the embryo assessment at day 3. When two Stenotrophomonas infections could be traced back to donor semen of a specific external sperm bank within a short period, we contacted this sperm bank in order to allow them to take active measures.
Microbial contaminations were predominantly (58.9%) detected at the time of embryo selection on day 2–4. However, there is a striking difference between the different strains. Most E. coli infections (60.7%) were observed at pronucleus inspection on day 1, whereas most Candida species (87.5%) and other micro-organisms (86.7%) manifested themselves at day 2–4. In order to determine the contribution of the semen used for fertilization in causing infections, we have to focus on the findings with day 1 infections. Of the 39 infections observed on day 1, 37 semen suspensions were available. In two cases, no semen suspension was left, as the whole sample was used to obtain the required insemination concentration in the fertilization dishes. In 13 cases, the same micro-organism, as based on the antibiotic susceptibility tests, was isolated both from the culture dishes and the semen suspension. These findings indicate that about one-third (13/37) of the microbial contaminations of the culture dishes were caused by the semen sample used for fertilization. The presence of micro-organisms in semen has been subject of many publications. Although micro-organisms are commonly found in semen, the clinical significance remains controversial. Several approaches have been described for decontamination of semen. As early as 1985, Hewitt et al. (1985) proposed routine culture of semen samples before a couple starts a treatment cycle. If micro-organisms were detected repeatedly, men should be treated with appropriate antibiotics. Other groups examined the effectiveness of processing semen samples in antibiotic-rich medium (Huyser et al., 1991; Cottel et al., 1997). However, the effectiveness of these approaches was never substantiated further and neither one is routinely applied in our centre. To prevent a recurrent contamination in the next cycle, both partners are treated with appropriate prophylactic antibiotics. Nevertheless, eight couples (9.4%) experienced recurrent infection in one or two subsequent treatment cycles, indicating that the use of antibiotic prophylaxis does not necessarily prevent recurrence of infections. Based on the fact that no infections have ever been observed in ICSI treatment cycles, couples with repeated infection of the IVF culture dishes are now routinely treated with ICSI.
Overall, 91.4% of the bacterial strains were insensitive to one or both antibiotics supplemented. The few contaminations with bacterial strains susceptible for the antibiotics used can probably be explained either by a high bacterial load in the culture dishes or the inaccuracy of the applied antibiotic susceptibility test. During the 8-year period comprising this study, an increasing number of Penicillin and Streptomycin resistant bacterial strains were noticed, particularly E. coli strains. Therefore, the effectiveness of Penicillin and Streptomycin supplementation of culture media has become more and more questionable. The use of Penicillin in particular has been superseded, mainly because of its very short half-life of only a couple of hours at 37°C. Penicillin and Streptomycin containing media have been increasingly abandoned and exchanged for culture media supplemented with Gentamycin. In 2005, we also decided to switch to Gentamycin containing culture medium, and since then colonization of culture dishes only by Candida species has been observed (unpublished data). A retrospective evaluation of the antibiotic susceptibility of all 70 bacterial strains isolated from contaminated culture dishes demonstrated Gentamycin sensitivity for all bacterial strains. This indicates that many if not all bacterial contaminations observed could have been prevented if Gentamycin containing medium had been used.
This study is the first systematic evaluation of the occurrence of microbial contaminations of IVF culture dishes. It could be deduced that the incidence of infections in our ART programme is considerable or even unacceptably high. As only very few publication of studies and case reports are available, a thorough comparison of data cannot be performed. In addition, the elevated incidence detected may be due to the scrupulous and thorough manner in which culture dishes are inspected in order to detect infection. With regard to the huge number of IVF cycles performed worldwide every year, microbial contamination should not be underestimated as a complication of an ART treatment cycle. Personal communication with several colleagues has revealed that every laboratory occasionally deals with infections. The incidence appears to be independent of whether gametes and embryos are handled on the bench in a low traffic area, or handled only in laminar airflow cabinets. Moreover, the present data show that despite extensive handling of cumulus–oocyte complexes and sperm cell suspensions required for ICSI, this is certainly not associated with an increased risk of bacterial contamination. Culture dish infection in ART should be evaluated in more detail, in order to learn more about the source of the micro-organism and to evolve adequate measures to prevent microbial contamination. Such data would also be very helpful in the discussion about air quality requirements evoked by the EU Directive 2004/23 and both annexes with technical specifications (European Union, 2004). In the second annex with technical specification, EU Directive 2006/86, it is stated that whenever tissues or cells, including reproductive cells and embryos, are exposed to the environment during processing, an air quality with particle counts and microbial colony count equivalent to those of Grade A is required (European Union, 2006). Long before this directive was officially adopted, a heated debate had already arisen regarding the feasibility and effectiveness of applying clean room air quality standards to IVF laboratories (Hartshorne, 2005; Mortimer, 2005). However, this second annex of the EU Tissues and Cells Directive includes some additional clauses, specifying exceptions to the stringent air quality requirement. It stated that a less stringent environment may be acceptable where it is demonstrated that the mode and route of application of the tissue or cell to the recipient implies a significantly lower risk of transmitting bacterial or fungal infection to the recipient than with cell or tissue transplantation, or where it is not technically possible to carry out the required process in a Grade A environment. However, it must be demonstrated and documented that the chosen environment achieves the quality and safety required. Therefore, more detail about the incidence, the source and prevention of microbial contamination of culture dishes is necessary to support and justify the choice of working in an environment with less stringent air quality.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
The authors would like to thank the staff of the IVF laboratory and microbiology laboratory for technical assistance, especially Cis Sneek-Rowaan, and Annet Troelstra for sharing their microbiological expertise. We also like to thank Nick Macklon and Kay Elder for critical reading of the manuscript before its submission.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Ben-Chetrit A, Shen O, Haran E, Brooks B, Geva-Eldar T, Margalioth EJ. Transfer of embryos from yeast-colonized dishes. Fertil Steril (1996) 66:335–337.[Web of Science][Medline]
Bureau M, Dea S, Sirard MA. Evaluation of virus decontamination techniques for porcine embryos produced in vitro. Theriogenology (2005) 63:2343–2355.[Web of Science][Medline]
Burrello N, Calogero AE, Perdichizzi A, Salmeri M, D'Agata R, Vicari E. Inhibition of oocyte fertilization by assisted reproductive techniques and increased sperm DNA fragmentation in the presence of Candida albicans a case report. RBM Online (2004) 8:569–573.[Medline]
Cottell E, McMorrow J, Lennon B, Fawsy M, Cafferkey M, Harrison RF. Microbial contamination in an in vitro fertilization-embryo transfer system. Fertil Steril (1996) 66:776–780.[Web of Science][Medline]
Cottell E, Lennon B, McMorrow J, Barry-Kinsella C, Harrison RF. Processing of semen in an antibiotic-rich medium to minimize microbial presence during in vitro fertilization. Fertil Steril (1997) 67:98–103.[CrossRef][Web of Science][Medline]
Elder K, Baker D, Ribes J. Infections, infertility and assisted reproduction. (2005) Cambridge, UK:: Cambridge University Press.
European Union. Directive 2004/23/EC of the European Parliament, of the Council of 31 March 2004 on setting standards of quality and safety for the donation, procurement, testing, processing, preservation, storage, and distribution of human tissues and cells. Official Journal of the European Union (2004) L 102/48.
European Union. Commission Directive 2006/86/EC of 24 October 2006 implementing Directive 2004/23/EC of the European Parliament and of the Council as regards traceability requirements, notification of serious adverse reactions and events and certain technical requirements for the coding, processing, preservation, storage and distribution of human tissues and cells. Official Journal of the European Union (2006) L 294/32.
Hartshorne GM. Challenges of the EU ‘tissues and cells’ directive. RBM Online (2005) 11:404–407.[Medline]
Hewitt J, Cohen J, Fehilly CB, Rowland G, Steptoe P, Webster J, Edwards RG, Fishel SB. Seminal bacterial pathogens and in vitro fertilization. J In Vitro Fertil Embryo Transf (1985) 2:105–107.[CrossRef][Medline]
Huyser C, Fourie FL, Oosthuizen M, Nethling A. Microbial flora in semen during in vitro fertilisation. J In Vitro Fertil Embryo Transf (1991) 8:260–264.[CrossRef][Web of Science][Medline]
Kastrop PMM, Weima SM, Van Kooij RJ, Te Velde ER. Comparison between intracytoplasmic sperm injection an in-vitro fertilization (IVF) with high insemination concentration after total fertilization failure in a previous IVF attempt. Hum Reprod (1999) 14:65–69.
Mortimer D. A critical assessment of the impact of the European Union Tissues and Cells directive (2004) on laboratory practices in assisted conception. RBM Online (2005) 11:162–176.[Medline]
Ng SC, Edirisinghe WR, Sathananthan AH, Ratnam SS. Bacterial infection of human oocytes during in vitro fertilization. Int J Fertil (1987) 32:298–301.[Web of Science][Medline]
Nygren KG, Nyboe Andersen A. Assisted reproductive technology in Europe, 1999. Results generated from European registers by ESHRE. Hum Reprod (2002) 17:3260–3274.
Quinn P, Kerin JF, Warnes GM. Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubal fluid. Fertil Steril (1985) 44:493–498.[Web of Science][Medline]
Submitted on March 16, 2007; resubmitted on May 10, 2007; accepted on May 15, 2007.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||