Encouraging results despite complexity of multidisciplinary care of HIV-infected women using assisted reproduction techniques

Jeanine Ohl1,4, Marialuisa Partisani2, Christiane Wittemer3, Jean-Marie Lang2, Stéphane Viville3 and Romain Favre1

1 Centre d’AMP de Strasbourg, Service de Gynécologie-Obstétrique, CMCO-SIHCUS, 19 rue Louis Pasteur – BP 120 – 67303 Schiltigheim, 2 CISIH, Clinique Médicale A, Hôpital Civil, 67091 Strasbourg Cedex and 3 Centre d’AMP de Strasbourg, Service de Biologie de la Reproduction, CMCO-SIHCUS, 19 rue Louis Pasteur, 67303 Schiltigheim

4 To whom correspondence should be addressed. E-mail: jeanine.ohl{at}sihcus.fr


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Assisted reproduction technologies can treat infertility for human immunodeficiency virus (HIV) seropositive women. We assessed the efficacy of these techniques in the results and difficulties encountered while conducting our assisted reproduction programme for 49 couples in which at least the woman had HIV infection that was currently under control. METHODS: Treatments included intrauterine insemination (IUI), IVF and ICSI, with ovarian stimulation. Embryos were transferred on day 3 after oocyte retrieval. An elective single transfer was performed, except for patients aged $ 40 years. RESULTS: The median age of the women was 36 years. Ten IUI, nine IVF, 53 ICSI and 10 frozen–thawed embryo transfers have been performed. No pregnancy occurred following the IUI trials but for the couples with IVF and ICSI attempts the clinical pregnancy rate per embryo transfer was 23.9%. Eight babies have been born leading to a 22.2% take home baby rate per treated couple. Contamination was not observed in any newborn. CONCLUSIONS: Assisted reproduction technologies and particularly ICSI can provide HIV seropositive women with a safe means of mothering children. Results are encouraging when considering the age of the patients and a preferential single embryo transfer.

Key words: assisted reproduction techniques/HIV/pregnancy


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Nearly half of the 40 million people infected with the human immunodeficiency virus (HIV) worldwide are women. Access to potent antiretroviral medication has improved the health and life expectancy of HIV+ patients and young couples can now consider having children. In HIV+ women, adequate care during pregnancy and delivery together with treatment of the newborn have decreased the risk of vertical transmission to < 1% (Mandelbrot et al., 1998Go; European Collaborative Study, 2005Go). In these women, reproductive counselling for self-insemination and use of assisted reproductive technologies in cases of infertility aim to avoid the risk of virus transmission to the partner.

The results and difficulties encountered while conducting our assisted reproduction programme for HIV+ women are presented here.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
According to French law, HIV+ women can be offered assisted reproduction treatment under predefined conditions since May 2001 (Journal Officiel de la République Française, May 15, 2001). Couples are referred to assisted reproduction centres specialised in treating patients with viral infections. Couples are followed by a multidisciplinary team including assisted reproduction clinicians and biologists, infectologists, psychologists, obstetricians and pediatricians. So, the patients were offered multidisciplinary expertise to cover their situation. The inclusion criteria are very specific and come in addition to the usual inclusion criteria for assisted reproduction treatment. Couples are to engage in only safe sexual intercourse. Medical follow-up of the HIV infection is essential, and all women are required to take regular medical advice for this purpose. The French legal requirements include stability of the infection defined as a CD4 lymphocyte count > 200/mm3 and a stable HIV viral load, with no increase > 0.5 log, measured at least twice over the 4 months prior to the beginning of assisted reproduction treatment.

When antiretroviral therapy was not indicated before assisted reproduction treatment, the patients were informed that it would be introduced at the latest at 28 weeks in the case of pregnancy to reduce maternal–fetal transmission. For patients already treated, the combination was modified before assisted reproduction treatment if not compatible with pregnancy.

All patients and their partners provided written informed consent for assisted reproduction treatment. An additional consent form, specifically acknowledging comprehension of the risk of viral transmission to the child, had to be signed by both members of the couple.

During the procedure, manipulations of gametes were conducted in a viral risk area geographically separated from the laboratory space dedicated to other couples not infected with HIV or hepatitis B or C.

The assisted reproduction laboratory followed Good Laboratory Practice (World Health Organization, 1999). Specific precautions were added against the risk of HIV, HCV and HBV contamination in a special biosafety cabinet workstation as previously described (Ohl et al., 2003Go).

A vaginal sample was taken for bacteriological testing a few days before the assisted reproduction procedures and any infection by common germs, mycoplasma or chlamydiae was treated. Semen was systematically cultured for the same purpose.

The ovarian stimulation protocol was chosen based on clinical data, the patients’ hormone profile, and the result of any earlier stimulation. For IVF and ICSI, the stimulation was conducted using either a standard long protocol with a GnRH analogue (Decapeptyl®; Ipsen Pharma, Paris, France; or Suprefact®; Aventis Pharma, Frankfurt am Main, Germany) given s.c. each day or a protocol with an antagonist (Cetrotide®; Serono, Boulogne, France; or Orgalutran®; Organon, Puteaux, France, given daily). Stimulation used recombinant FSH (Gonal F®; Serono, Boulogne, France; or Puregon®; Organon). Oocyte maturation was triggered with HCG 5000 IU (Gonadotrophine Chorionique Endo 5000®; Organon).

For intrauterine insemination (IUI) a monofollicular stimulation with recombinant gonadotrophin (Gonal F or Puregon) was achieved.

All patients were treated with micronized progesterone administered vaginally during the luteal phase.

Cycles were monitored by means of serial transvaginal ultrasonography and serum LH and E2 assays at the assisted reproduction centre nearest to the couple’s home. The oocyte retrievals, ICSI, embryo transfers and insemination procedures took place in our centre in the viral risk area. ICSI was performed as classically described (Palermo et al., 1992Go). Supernumerary embryos on day 3 were frozen and thawed according to a standardized technique involving 1,2-propanediol and sucrose as cryoprotectants (Testart et al., 1986Go). For women aged < 40 years elective single embryo transfers were performed whatever the rank of the attempt. Supernumerary embryos were frozen one by one. When available, for women aged > 40 years, two embryos were transferred. After thawing, a single embryo transfer was performed at the first attempt and a double embryo transfer at the following attempts.

In the case of HIV infection of both partners, inclusion criteria had also to be respected by the man. Semen samples to be used for assisted reproduction treatment were tested for the presence of viral nucleic acids and prepared according to previously described techniques (Ohl et al., 2003Go).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
From May 2001 through December 2004, 71 couples in which the woman was HIV+ attended our centre for assisted reproduction counselling (Figure 1). Five couples were determined to be ineligible for the following reasons: malignant hypertension in one patient, instability in the couple in two cases, very poor semen in two HIV+ male partners. In this last situation, the semen could not be validated with reliable virological tests. Assessment of eligibility is still in progress for 16 couples, and 50 couples have definitively been accepted. These include 49 HIV-1-infected women and one HIV-2-infected patient. In three cases, the woman was co-infected with hepatitis C virus (HCV) and in one case with hepatitis B virus (HBV). In six cases, the male partner was HIV+.



View larger version (18K):
[in this window]
[in a new window]
 
Figure 1. Flow chart. FET = frozen embryo transfer.

 

The mean ± SD age of the women was 35.0 ( ± 4.9) and the median age was 36.0 (range 24 – 43) years. All 50 couples had documented infertility problems. These consisted of self-insemination failure: 18 (36%); tubal occlusion: 16 (32%); male infertility: five (10%); amenorrhoea: four (8%); cervical infertility after conization: three (6%); mixed infertility: three (6%). In one case, the indication for assisted reproduction treatment was preimplantation genetic diagnosis (PGD) of haemophilia (2%).

The virological characteristics of the 50 included women are presented in Table I.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of included patients

 

At inclusion, 18 women did not take any antiretroviral therapy and 32 were treated with highly active antiretroviral therapy (HAART) compatible with pregnancy. In three patients, immunovirological condition allowed HAART discontinuation before assisted reproduction treatment. In the 13 patients who became pregnant, the treatment was either continued or introduced at the latest at 28 weeks of pregnancy. Treatment for HCV was not indicated for any patient.

Thirty-six couples were treated through 82 assisted reproduction attempts: 10 IUI, 9 IVF, 53 ICSI and 10 frozen–thawed embryo transfers. For two patients an ICSI procedure was performed after a previous IUI failure and for four others ICSI followed an IVF failure.

Ten ovarian stimulations were discontinued before assisted reproduction treatment in eight patients because of a poor ovarian response. Their mean basal FSH was 6.4 IU/l, their mean age was 34.5 years and the mean starting dose of FSH was 242 IU per day. Regarding IUI, 10 cycles were initiated leading to eight IUI with a mean of 12.3 ± 5.5x106 sperm inseminated. No pregnancy occurred after IUI.

Biological parameters of the IVF and ICSI cycles are resumed in Table II.


View this table:
[in this window]
[in a new window]
 
Table II. IVF/ICSI features

 

Fifty-six oocyte retrievals led to 46 embryo transfers, two empty follicles, one absence of eligible embryo for transfer after PGD, one case of no transferable embryo after IVF and six fertilization failures (four after IVF and two after ICSI). Three couples with fertilization failure after IVF came back and obtained embryos during a subsequent ICSI attempt.

A total of 13 pregnancies occurred. Outcome of the pregnancies is summarized in Table III. Four pregnancies resulted in a spontaneous miscarriage (30.8%). No case of congenital infection with HIV and no congenital abnormalities have been reported in the population of newborns.


View this table:
[in this window]
[in a new window]
 
Table III. Pregnancies

 


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Legitimacy of pregnancy
A major increase in the life expectancy and quality of life of HIV-infected patients allows young HIV-infected couples to consider pregnancy. In addition, the reduction in HIV viral load before delivery has considerably diminished maternal–fetal transmission in the developed world. Among patients who have access to adequate antiretroviral treatment and monitoring and who deliver by Caesarean section, the frequency of maternal–fetal transmission is now < 1% (Kind et al., 1998Go; Mandelbrot et al., 1998Go; European Collaborative Study, 2005Go).Prospective studies have confirmed the efficacy of HAART in preventing mother-to-child HIV transmission (Mandelbrot et al., 2001Go; Cooper et al., 2002Go).

The clinical and biological course of HIV disease during pregnancy is well understood. In asymptomatic women whose infection is well-controlled, pregnancy does not aggravate HIV disease (Weisser et al., 1998Go; Saada et al., 2000Go).

In our series, patients were considered eligible for assisted reproduction treatment only when their clinical, immunological and virological criteria were compatible with a successful pregnancy outcome and a minimal risk of vertical transmission of the virus. Thanks to multidisciplinary care, immunovirologic stability could be achieved in all patients. For some, the wish for pregnancy and assisted reproduction treatment have thus become the opportunity to optimize the care of the disease. Moreover the investment of the patients in their treatment is intensified on account of a future project taking shape.

Legitimacy of assisted reproduction
Couples in which the woman is HIV+ have often attempted self-insemination at home before turning to assisted reproduction treatment (Delfraissy et al., 2002Go). Unlike in the couples with HIV+ men, infertility factors are predominant in this population.

Feminine indications for assisted reproduction treatment are frequent. The frequency of upper genital tract infections and their sequelae is 10-fold higher in HIV+ women than in the general population (Sobel, 2000Go). Physicians are thus often consulted for the treatment of tubal infertility. Cervical infertility is frequent too because of a possible history of cervical lesions leading to laser ablation, loop excision or conization (Soncini and Condemi, 2003Go; Gilles et al., 2005Go). Assisted reproduction treatment also allows treatment of a possible male sterility at the same time as it prevents HIV transmission to the partner at the time of conception. Our population reflected these previously described aetiologies of infertility.

For HIV-positive women who wished to become pregnant, potential infertility criteria were checked as soon as during the first visit. Self-inseminations were proposed when no obvious infertility factor could be found. Assisted reproductive techniques were however rapidly considered, either because of an infertility factor or because self-inseminations had failed or in cases of advanced age. In addition, in some couples the intrusion of a syringe in their sexual intimacy was unacceptable and repetition of the gesture became more and more difficult to accept. They asked for assisted procreation to recover their previous sexuality.

Efficacy of assisted reproduction treatment
Assisted reproductive techniques in HIV-serodifferent couples reported in the literature show excellent results but only describe HIV-infected men (Guibert et al., 2001Go; Semprini et al., 2002Go; Partisani et al., 2005Go). As infertility factors are predominant in our series of HIV-positive women, results are disappointing when compared to our previously described 48.8% pregnancy rate per transferreported in HIV+ men (Ohl et al., 2003Go). However, in this population, the women were younger and usually free of any disease including HIV infection. The comparison is therefore of little interest. The 11.8% implantation rate obtained in our series is rather encouraging. In our non-HIV-infected population concerning 1151 patients, the current overall implantation rate is 14.5% during the same period which is not significantly different (unpublished data).

Due to thelong wait for scientific and legal progress, some women in our series have begun assisted reproduction treatment quite belatedly, with impaired results just because of advanced age. Half of our population is aged > 36 years. The mean age was higher in our series (35.0 years) than the age of the non HIV-infected population of our centre (33.7 years) (unpublished data). Other authors report the same phenomenon(Guibert, 2004Go).As access to assisted reproduction treatment becomes easier for HIV-infected women and the number of centres increases, this age difference should progressively tend to disappear.

Assisted reproductive techniques require a controlled ovarian stimulation to allow an acceptable efficiency but, in our population, several patients were poor responders. Ten stimulations were cancelled: six before IVF/ICSI leading to a 9.7% [95% confidence interval (CI): 3.6–19.9] cancellation rate, two before IUI and two before frozen–thawed embryo transfer.In poor responders, spontaneous ovulation remains possible but in this population a spontaneous pregnancy cannot occur as unprotected repeated intercourse is not recommended and repeated self-inseminations are tedious.

Only for two patients did oocyte retrieval fail. The 8.4 mean number of oocytes per aspiration in our series was similar to the 8.5 mean number observed in our general population during the same period (unpublished data). This comparable result has been achieved thanks to a higher dose of gonadotrophins: 2793 IU in our study versus 1878 IU in the general non HIV-infected population (unpublished data).

Our earlier results (Ohl et al., 2003Go), together with those of other groups (Harlow et al., 2000Go; Clark et al., 2001Go) had suggested a negative effect of HIV infection on ovarian function. Whether the virus or the antiretroviral therapy have a specific impact, distinct from the effect of age, on the ovarian function is still unknown and further studies are necessary.

In our series the 63.8% fertilization rate is encouraging. When transfer occurs,the 23.9% pregnancy rate per transfer is acceptable in this series despite older age. Results obtained here are better than those we reported in an earlier series (i.e. 9.1%) (Ohl et al., 2003Go). The small size of the initial sample, i.e. 10 patients, can probably account for some of the difference.

Nowadays, IVF centres are more and more concerned about frequent multiple pregnancies leading to increased rates of prematurity and perinatal mortality. Recently, elective single embryo transfer has been proposed in selected favourable populations to avoid multiple pregnancies after IVF, without reported decrease of the pregnancy rate (Thurin et al., 2004Go). For HIV+ patients, twin pregnancies are not desirable. Prematurity together with obstetric complications increase the risk of maternal-to-fetal HIV transmission (Weigel et al., 2001Go). Moreover, HAART and particulary those containing protease inhibitors may increase the risk of prematurity (European Collaborative Study, 2000Go; Thorne et al., 2004Go). Nonetheless, these inhibitorsare recommended to prevent maternal-to-fetal transmission according to current guidelines (Coll et al., 2002Go; Public Health Service Task Force, 2003Go; Delfraissy et al., 2004Go). Anaemia induced by the treatment could also enhance the risk of premature uterine contractions.

In our series, the double embryo transfer remained the standard procedure only for women aged > 40 years of age. For patients aged < 40 years, a single embryo transfer was systematically performed even if the embryo quality was not optimal or even if the couple had already encountered one or more previous IVF failures. The transfer of three embryos was not an option for us in this context whatever the age of the patient and even with frozen–thawed embryos. A triple pregnancy with the need of an invasive selective reduction would have been unacceptable considering the increased risk of maternal–fetal viral transmission (Guibert, 2004Go). For frozen–thawed embryo transfers, we performed a single embryo transfer during the first attempt and a double embryo transfer was discussed in the case of failure.

In our series, the rather encouraging implantation rate of 11.8% has led us to consider an authentic multiple pregnancy probability and has supported the idea of a reduction of the number of transferred embryos. Of course the pregnancy rate observed may have been impaired by the small number of embryos replaced in the uterus (1.8 ± 0.6) but our strategy was deliberate and the absence of multiple pregnancy should itself be considered a success.

HIV-infected women have been described as being at increased risk of miscarriage (Brocklehurst and French, 1998Go). Introduction of antiretroviral therapy and especially HAART seems to have decreased the miscarriage rate (Massad et al., 2004Go). In our series, the 30.8% miscarriage rate is high but the small size of the population must be taken into account and the role of other factors remains to be determined.

Safety of assisted reproduction treatment
Whether assisted reproduction treatment in itself could be responsible for an increased risk of vertical transmission is unknown since little work has been published about assisted reproduction in HIV+ women. Nevertheless it is well known thatmaternal-to-fetal HIV transmission occurs late in pregnancy, due to fetal–maternal exchanges at the end of gestation, especially during labour (Rouzioux et al., 1995Go). Assisted reproductive techniques are concerned only by the beginning of pregnancy and therefore they cannot be responsible for HIV transmission.

Choice of technique
The choice of the technique used depended on the overall medical presentation of each patient. Because of the viral specificity of our population, infertility factors were frequently identified and IVF often had to be proposed in the first place.

Even when intrauterine insemination was initially considered possible, in the absence of severe infertility, other factors were taken into consideration. It soon appeared to us that stimulations for IUI in HIV+ women would be difficult to perform. High doses of gonadotrophins could not be used because of the risk of multiple pregnancy and results with inseminations, in our series, were very disappointing. So, IVF was rapidly proposed, providing both better results and the opportunity of a single embryo transfer.

ICSI has often been the technique of choice. The age of the patients has particularly been taken into account, higher than usual for a general IVF population. We also considered impaired ovarian response as an indication per se. Our three couples with a previous IVF failure obtained embryos after a subsequent ICSI. In addition, for three patients interruption of the antiretroviral treatment was recommended because of the planned pregnancy, so we used ICSI to avoid unexpected fertilization failure and thus to reduce the delay of conception. These different reasons led us rapidly to perform ICSI rather than classical IVF as an elective procedure.

In conclusion, treatment progress and radical changes in social attitudes now make it possible for assisted reproduction to be offered to HIV+ women. Assisted reproduction treatments are performed with the double benefit of controlling the risk of vertical viral transmission by multidisciplinary care and of treating sterility.

In our series, difficulties included the age of the patients, the existence of an authentic infertility and our deliberate policy of preferential single embryo transfer. Nevertheless the absence of multiple pregnancies associated with an encouraging pregnancy rate are rewarding for the efforts of the multidisciplinary team.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Brocklehurst P and French R (1998) The association between maternal HIV infection and perinatal outcome: a systematic review of the literature and meta-analysis. Br J Obstet Gynaecol 105,836–848.[ISI][Medline]

Clark RA, Mulligan K, Stamenovic E, Chang B, Watts H, Andersen J, Squires K and Benson C (2001) Frequency of anovulation and early menopause among women enrolled in selected adult AIDS clinical trials group studies. J Infect Dis 184,1325–1327.[CrossRef][ISI][Medline]

Coll O, Fiore S, Floridia M, Giaquinto C, Grosch-Wörner I and Guiliano M (2002) Pregnancy and HIV infection: a European Consensus on management. AIDS 16(Suppl 2), S1–S18.[Medline]

Cooper E, Charurat M, Mofenson L, Hanson C, Pitt J, Diaz C, Hayani K, Handelsman E, Smeriglio V, Hoff R et al (2002) Combination antiretroviral strategies for the treatment of pregnant HIV-1-infected women and prevention of perinatal HIV-1 transmission. J AIDS 29,484–494.[ISI][Medline]

Delfraissy JF, Rouzioux C, Bujan L, Faucher JM, Heard I, Jouannet P, Le Mercier Y, Leruez M, Meier A and Prestel T (2002) Désir d’enfant et assistance médicale à la procréation. In Prise en charge thérapeutique des personnes infectées par le VIH. Paris. Rapport 2002. Ministère de l’emploi et de la solidarité, Secrétariat d’Etat à la santé et à l’action sociale. Médecine-Sciences, Flammarion (eds), p 282.

Delfraissy JF, Matheron S, Blanche S, Faye A, Jeantils V, Mandelbrot L, Rouzioux C and Tubiana R (2004) Grossesse. In Prise en charge thérapeutique des personnes infectées par le VIH. Paris. Rapport 2004. Ministère de l’emploi et de la solidarité, Secrétariat d’Etat à la santé et à l’action sociale. Médecine-Sciences, Flammarion (eds), Paris pp 193–196.

European Collaborative Study (2005) Mother-to-child transmission of HIV infection in the era of highly active antiretroviral therapy. CID, 40,458–465.[CrossRef]

European Collaborative Study and the Swiss Mother + Child HIV Cohort Study (2000) Combination antiretroviral therapy and duration of pregnancy. AIDS 14,2913–2920[CrossRef][ISI][Medline]

Gilles C, Manigart Y, Konopnicki D, Barlow P and Rozenberg S (2005) Management and outcome of cervical intraepithelial neoplasia lesions: a study of matched cases according to HIV status. Gynecol Oncol 96,112–118.[CrossRef][ISI][Medline]

Guibert J, Merlet F, Le Dû A, Mandelbrot L, Leruez-Ville M, Costagliola D, Kunstmann JM, De Almeida M, Heard I, Dulioust E et al (2001) Prise en charge des couples séro-différents pour le VIH. Résultats du protocole NECO (ANRS 092) à Paris. Reprod Humaine Horm 14,363–364.

Guibert J (2004) La femme VIH séropositive: problème de VIH ou/et de fertilité. Désir d’enfant et VIH, Troisièmes Journées Nationales, Toulouse, 6–7 mai.

Harlow S, Schuman P, Cohen M, Ohmit S, Cu-Uvin S, Lin X, Anastos K, Burns D, Greenblatt R, Minkoff H et al (2000) Effect of HIV infection on menstrual cycle length. J AIDS 24,68–75.[ISI][Medline]

Kind C, Rudin C, Siegrist CA, Wyler CA, Biedermann K, Lauper U, Irion O, Schupbach J and Nadal D (1998) Prevention of vertical HIV transmission: additive protective effect of elective cesarean section and zidovudine prophylaxis. Swiss Neonatal HIV Study Group. AIDS 12,205–210.[CrossRef][ISI][Medline]

Mandelbrot L, Le Chenadec J, Berrebi A, Bongain A, Benifla JL, Delfraissy JF, Blanche S and Mayaux MJ (1998) Perinatal HIV-1 transmission: interaction between zidovudine prophylaxis and mode of delivery in the French Perinatal Cohort. J Am Med Assoc 280,55–60.[Abstract/Free Full Text]

Mandelbrot L, Landreau-Mascaro A, Rekacewicz C, Berrebi A, Benifla JL, Burgard M, Lachassine E, Barret B, Chaix ML and Bongain A (2001) Lamivudine–zidovudine combination for prevention of maternal-infant transmission of HIV-1. J Am Med Assoc 285,2083–2093.[Abstract/Free Full Text]

Massad S, Springer G, Jacobson L, Watts H, Anastos K, Korn A, Cejtin H, Stek A, Young M, Schmidt J and Minkoff H (2004) Pregnancy rate and predictors of conception miscarriage and abortion in US women with HIV. AIDS 18,281–286.[CrossRef][ISI][Medline]

Ohl J, Partisani M, Wittemer C, Schmitt MP, Cranz C, Stoll-Keller F, Rongières C, Bettahar-Lebugle K, Lang JM and Nisand I (2003) Assisted reproduction techniques for HIV serodiscordant couples: 18 months of experience. Hum Reprod 18,1244–1249.[Abstract/Free Full Text]

Palermo G, Joris H, Devroey P and Van Steirteghem AC (1992) Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 340,17–18.[CrossRef][ISI][Medline]

Partisani M, Ohl J, Schmitt MP, Wittemer C, Cranz C, Rongières C and Lang JM (2005) Assisted reproductive techniques for HIV-seropositive men. 12th Conference on Retroviruses and Opportunistic Infections, Boston, February 22–25, 2005 (abstract 540).

Public Health Service Task Force (2003) Recommendations for use of antiretroviral drugs in pregnant HIV-1-infected women for maternal health and interventions to reduce perinatal HIV-1 transmission in the United States. AIDSinfo Web site (http://AIDSinfo.nih.gov).

Rouzioux C, Costagliola D, Burgard M, Blanche S, Mayaux MJ, Griscelli C and Valleron AJ (1995) Estimated timing of mother-to-child human immunodeficiency virus type I transmission by use of a Markov model. Am J Epidemiol 142,1330–1337.[Abstract]

Saada M, Le Chenadec J, Berrebi A, Bongain A, Delfraissy JF, Mayaux MJ and Meyer L (2000) Pregnancy and progression to AIDS: results of the French prospective cohorts. AIDS 14,2355–2360.[CrossRef][ISI][Medline]

Semprini AE, Vucetich A, Oneta M, Rezek D, Rubino P, Scarselli F and Hollander LH (2002) Spermatozoa washing and ICSI in HIV discordant couples: >50% pregnancy rate. Hum. Reprod 17(Abstract Book 1), P-340.

Sobel JD (2000) Gynecologic infections in human immunodeficiency virus-infected women. CID, 31,1225–33.[CrossRef]

Soncini E and Condemi V (2003) Intraepithelial cervical carcinoma and HIV. Prevalence, risk factors and prevention strategies. Minerva Ginecol 55,51–55.[Medline]

Testart J, Lassalle B, Belaisch-Allart J, Hazout A, Forman R, Rainhorn JD and Frydman R (1986) High pregnancy rate after early human embryo freezing. Fertil Steril 46,268–272.[ISI][Medline]

Thorne C, Patel D and Newell ML (2004) Increased risk of adverse pregnancy outcomes in HIV-infected women treated with highly active antiretroviral therapy in Europe. AIDS 18,2337–2339.[CrossRef][ISI][Medline]

Thurin A, Hausken J, Hillensjo T, Jablonowska B, Pinborg A, Strandell A and Bergh C (2004) Elective single-embryo transfer versus double-embryo transfer in in vitro fertilization. New Engl J Med 351,2392–2402.[Abstract/Free Full Text]

Weigel MM, Kremer H, Sonnenberg-Schwan U, Gölz J, Gürtler L and Doerr HW (2001) Diagnostics and treatment of HIV-discordant couples who wish to have children. Eur J Med Res 6,317–321.[ISI][Medline]

Weisser M, Rudin C, Battegay M, Pfluger D and Kully C (1998) Does pregnancy influence the course of HIV infection? Evidence from two large swiss cohort studies. J Acquir Immun Defic Syndr Hum Retroviral 17,404–410.[ISI][Medline]

Submitted on March 18, 2005; resubmitted on May 6, 2005; accepted on May 31, 2005.