Semen characteristics in human immunodeficiency virus (HIV)- and hepatitis C (HCV)-seropositive males: predictors of the success of viral removal after sperm washing

Nicolás Garrido1,2,5, Marcos Meseguer1,2, José Remohí1,2,3, Carlos Simón1,2,3 and Antonio Pellicer1,2,4

1 IVI Valencia, 2 FIVI (Fundación IVI), 3 Departments of Paediatrics, Obstetrics and Gynaecology, Valencia University School of Medicine and 4 Department of Obstetrics and Gynaecology Hospital Dr Peset, Valencia, Spain

5 To whom correspondence should be addressed at: Instituto Valenciano de Infertilidad, Andrology Laboratory and Semen Bank, Plaza de la Policía Local 3, Valencia 46015, Spain. Email: nicolas.garrido{at}ivi.es


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Human immunodeficiency virus (HIV)/hepatitis C virus (HCV)-seropositive males can now father children safely, avoiding transmission risks to the mother and the children using sperm washing and nested PCR (nPCR) techniques. Nevertheless, we still lack enough data to determine the reasons why ~10% of the performed sperm washes remain positive, thus forcing the repetition of the treatment. Semen quality in infected males is also essential to these procedures. We aimed to determine the predictive value of the semen parameters, sperm washing procedure and the infection status for the post-wash viral positivity, as well as the correlation between the semen and the disease features. METHODS: Semen characteristics were evaluated in 136 samples provided from 125 males. We also included a control group of 125 males matched by age and length of sexual abstinence. At the time of semen retrieval, 70 of them were infected with HIV (45 also with HCV, 64.3%), and 55 of them with HCV alone. nPCR for viral detection was performed in each sample. RESULTS: Thirteen out of 136 (9.5%) of the samples were positive for one or more viral detections (HIV RNA, HIV DNA and HCV RNA, when needed). From a total of 240 nPCR viral analyses, 16 were positive (6.6%). None of the seminal parameters were adequate to predict post-wash results, nor was a positive result dependent on the volumes used in the semen wash. A positive correlation was found between post-wash progressive motility and CD4 blood levels, as well as a negative correlation between progressive motility and time of evolution of the disease in HIV-infected males. CONCLUSIONS: Semen analysis, according to the World Health Organization criteria, of HIV- and HCV-affected patients showed no differences from that of non-infected males. Moreover, low CD4 blood levels, and a long evolution of the disease do not negatively affect sperm motility.

Key words: HIV and HCV/ICSI/nested PCR/semen wash/serodiscordant


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Couples with a seropositive human immunodeficiency virus (HIV) male, or couples undergoing assisted reproduction techniques (ARTs) with the male infected with hepatitis C virus (HCV), are increasingly requesting techniques to avoid the risk of transmission to the female partner and offspring (Semprini et al., 1992Go; Levy et al., 2000Go).

Previous work on HIV epidemics has clearly demonstrated the presence of the virus in the ejaculates obtained from infected men (Ho et al., 1984Go; Zagury et al., 1984Go). Moreover, the presence of HCV in the whole ejaculate has been demonstrated by several investigations, although some controversial results have been described (Levy et al., 2000Go; Garrido et al., 2001Go; Meseguer et al., 2002Go).

For viral detection in sperm, different methodologies have been employed, although nested PCR (nPCR) has been demonstrated to be the most sensitive, since it permits the detection of a single viral RNA or DNA particle (Meseguer et al., 2002Go). Other works employed different technologies with the extraction step performed by using either silica beads, a specific viral kit or guanidinium thiocyanate to bypass the difficulties encountered when seminal plasma is present. For the cDNA amplification step, several other techniques have been employed such as Amplicor HCV Cobas assay, one-round PCR, NASBA and branched DNA (Bourlet et al., 1992Go; Peter and Sevall, 2004Go).

Actually, we can eliminate the presence of both viruses with sperm washing procedures,which can be employed safely in ART treatments after nPCR determination of the absence of the virus (Garrido et al., 2004Go). Interestingly, recently published articles point to a CD4-independent receptor for HIV in sperm, thus indicating that sperm can host viral particles (Bandivdekar et al., 2003Go). Its repercussions on male fertility are still unknown, but the direct evidence of healthy newborns and the absence of horizontal transmission to the partner in >2000 reported treatments in serodiscordant couples in the world argue against this possibility.

Nevertheless, a variable percentage of sperm washes remain positive, and these washes need to be repeated. Interestingly, repeat sperm washes prove successful in viral depletion.

So far, there are not enough data addressing the influence of the semen parameters and physical condition of the males on the success of sperm wash in the removal of viral particles. Also, information about the relationship between the status of the infection, male characteristics, antiretroviral treatments, etc. is lacking.

Therefore, the aim of this work was to analyse for the first time the relevance of the sperm washing protocol, the ability of the analysis of seminal parameters to predict sperm wash success, as well as the influence of the male's health status on the semen characteristics.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Patients
Semen samples were obtained from males with HIV attending Instituto Valenciano de Infertilidad for ART procedures to avoid viral transmission, or from males belonging to an infertile couple who requested sperm washing due to the presence of HCV antibody.

We also included a control population consisting of 73 healthy seronegative males, partners of women with tubal infertility, within the same period of time as the HIV-infected population. Controls and HIV males were matched by age (±2 years) and the number of days (±0.5) of sexual abstinence. This was performed using a similar design to that previously described (Dulioust et al., 2002Go).

All paticipants signed an informed consent. The study was approved by the Ethics Committee of the Instituto Valenciano de Infertilidad.

A total of 125 males provided 136 samples to be washed between August 2001 and November 2003. One of them was not analysed by nPCR, and one patient rejected sperm wash repetition after a positive result in one sample. Their ages ranged between 21 and 54 years and the median was 37.1 in the males.

The mode of viral acquisition was, for the 70 HIV-infected males (45 of them also co-infected with HCV, 64.3%), 32 ex-addictions to parenteral drugs (45.7%), two plasma donations (3.0%), 15 heterosexual transmissions (21.4%) and 21 of unknown origin (30%). The duration of the HIV disease was 9.6 years (range 1–20). In the Centre of Diseases Control (CDC) classification of the disease, 12 of them (17.1%) were A1, 23 Aa (32.8%), five A3 (7.1%), three B1 (4.3%), seven B2 (10%), 11 B3 (15.7%), two C2 (2.8%) and six C3 (8.5%).

Regarding antiretroviral treatments, 21 (30%) of them were treatment free, while the remainder were receiving any combination: two monotherapies (2.8%), four bi-therapies (5.7%), 36 (51.4%) tri-therapies, six tetra-therapies (8.5%) and one hexa-therapy (1.4%).

HIV blood load was 19 408 IU/ml (ranging from undetectable to 525 000 IU/ml), while 39 patients (55.7%) showed a negative viral blood load. Median peripheral blood CD4 levels were 497.5, ranging from 26 to 1064.

There were 55 HCV-infected males. Among them, only eight (14%) had viral load determinations in peripheral blood, median 577 000 IU/ml (range 12 699–2 450 000), and the mode of viral acquisition was unknown in all except for one blood transfusion at birth. Only one received a combined treatment of rivabirine and interferon a few years earlier.

Semen analysis
All the samples were processed in a dedicated laboratory, situated in a different room from where we work with non-infectious samples. All the safety guidelines followed can be found in previous works (Bellver et al., 2002Go). Briefly, the viral risk area was dedicated purely to these biological samples, including a biosafety cabinet workstation, an exclusive centrifuge with anti-aerosol lids, and a nitrogen tank solely for storage of these potentially infectious samples.

Semen parameters were analysed according to the World Health Organization (1999)Go criteria. Total count and motility before and after the wash were recorded. Sperm morphology was not analysed for safety reasons: it is not recommended to work with cutting elements when treating HCV/HIV-positive samples and, as these samples are always for ICSI treatments, fresh sperm morphology is not needed. Sperm wash was only performed in those samples with >2 x 106 total progressive motile sperm in the whole semen sample. When we attempted sperm washes on samples below these thresholds, no spermatozoa were found after the wash, to be frozen or analysed. This was repeated at least four times, and it is obviously caused by the exigent protocols needed to eliminate viral load.

Sperm wash
Ejaculates obtained after a sexual abstinence of 3–5 days were allowed to liquefy for 10 min at 37°C, and then were diluted 1:1 (v:v) with sperm medium (MediCult, Jyllinge, Denmark). Then, they were pelleted at 400 g 10 min, and the supernatants were discarded.

A volume of sperm medium equal to the initial volume was added, and then layered onto a triple density gradient (90, 70 and 45%, PureSperm, Nidacon, Goteborg, Sweden) of varying volumes depending on the sperm sample quality (increasing volumes in samples with greater numbers of progressive sperm cells), usually 1–1.2 ml of each layer, and centrifuged for 20 min at 300 g.

Each pellet was obtained and washed with 5 ml of sperm medium, and re-pelleted again. Supernatants were discarded and a swim-up of 0.5–1.5 ml (depending on the pellet size) was done. After 45 min, the upper 0.35 ml of each tube was obtained and pooled.

One half was immediately submerged in liquid nitrogen for PCR determinations, and the other half was frozen with sperm freezing medium (MediCult, Jyllinge, Denmark), as previously described (Meseguer et al., 2002Go), and stored until their use after a negative result regarding viral presence.

PCR techniques for HIV/HCV applied to sperm
Nucleic acids were extracted from the washed spermatozoa using the Nuclisense® method (Organon Teknika, Barcelona, Spain) following the instructions of the manufacturer. Two extractions were done in parallel, one from the spermatozoa sample and the other with the spermatozoa samples after the addition of HIV RNA obtained from HIV-infected plasma to detect the presence of transcription or amplification inhibitors after the nucleic acid extraction procedure. Both extracted samples were used for two HIV RNA transcriptions to detect genes from the gag and pol region, followed by a nested DNA amplification. The same samples were used to amplify HIV proviral DNA by a nested amplification to detect both genes (gag and pol). The other extraction, run in parallel with added HIV RNA before nucleic acid extraction, was used as a positive control to detect the presence of inhibitors of the transcription or amplification. Negative controls to detect the presence of amplicon contamination were also performed.

For HIV RNA transcription, we used the antisense external primers to anneal with nucleotides 1696–1676 and 3286–3265 for gag and pol genes, respectively. Standardized conditions for transcription were followed using 100 µmol/l dithiothreitol (DTT), 1 mmol/l each dNTP, 0.2 µmol/l antisense primer, 20 U of RNasin (Promega, Barcelona, Spain) and 5 U of AMV transcriptase (Promega) in a final volume of 20 µl. Nested DNA amplification used the external primers to anneal with nucleotides (from the ARV2/SF2 sequence) 1224–1243 and 1696–1676 and internal primers to anneal with nucleotides 1316–1335 and 1524–1504 for the gag region. External primers annealing with nucleotides 2623–2642 and 3286–3265 and internal primers annealing with nucleotides 2716–2741 and 3250–3227 for the pol region were used. PCR standardized conditions were followed including 6 µl of reverse transcriptase or previous PCR product, 2.5 and 2 mmol/l MgCl2 for the gag and pol region, respectively, 0.2 mmol/l each dNTP (Amersham Pharmacia, Madrid, Spain), 0.5 mol/l each primer and 2 U of Taq polymerase (Promega) in a 50 µl final volume. {beta}-Actin gene amplification was performed to confirm the presence of DNA in the extraction from the spermatozoa suspension. Results were read after a 2% agarose gel electrophoresis subsequent to ethidium bromide staining. In all the samples, a consistent result (either positive or negative) was obtained. The assay failure rate was zero.

The HCV nPCR protocol was described in our previous work. It amplifies the viral NCR region in a protocol comparable with reverse transcription and nPCR for HIV but with the adequate specific primers (Meseguer et al., 2002Go).

Statistical analysis
All data were abnormally distributed. Subsequently, in these cases, matched Wilcoxon tests were carried out. Significance was defined as P<0.05.

Correlations between CD4, evolution of the disease and semen parameters of HIV males were performed. Since the distribution of variables was not normal, the degree of correlation between the variables can be calculated using Spearman's rank test.

The diagnostic performance of a test or the ability of a test to discriminate between situations is evaluated using receiver operating characteristic (ROC) curve analysis (Zweig and Campbell, 1993Go). ROC analyses between having or not having a positive result were compared for all the parameters.

We also determined the positive predictive value and negative predictive value of time elapsed between sperm wash results and different parameters, which are the probability that samples are positive when the test of these parameters is positive and the probability that samples are negative when the same test is negative, respectively.

Fisher's exact test was performed (Table IV) to compare positive and negative samples in relation to blood viral load.


View this table:
[in this window]
[in a new window]
 
Table IV. Comparison between nPCR result and semen parameters (mean ± SEM) in HIV-infected males

 
The statistical analysis was performed using the MedCalc Software, Ghent, Belgium.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Semen characteristics depending on the infection
As reflected in Table I, seminal characteristics are comparable between groups independently of the infection. Seminal volume and concentration means are above the normal results considered by the World Health Organization, while sperm motility was slightly below the normal parameters of 50% (A + B) forms.


View this table:
[in this window]
[in a new window]
 
Table I. Seminal characteristics depending on the infection.

 
Although HIV/HCV-co-infected males and HCV-infected ejaculates presented differences in the number of total cells with progressive motility (115.9x106 versus 145.2x106/ejaculate), this was not statistically different.

Predictive values of the semen parameters for the sperm wash results
With the available data, we calculated the effectiveness of the semen parameters in the prediction of the effectiveness of the wash procedure. To this end, ROC curve analysis was employed.

In Table II the results of the ROC curve analysis can be found, demonstrating the lack of a good predictive factor either within the disease parameters or with semen parameters.


View this table:
[in this window]
[in a new window]
 
Table II. ROC curve analysis for the prediction of the post-wash PCR result for infectious particles by semen parameters and HIV patients' features

 
ROC curve analysis is performed to analyse the diagnostic features of a test; the area under the curve (AUC) represents the value of a test as a diagnostic tool. The maximum value is 1. This will correspond to a 100% sensitive and accurate test. When we look at the results, all values are low and, among them, only semen concentration (AUC=0.65) in raw semen and total sperm with progressive motility (TMP; AUC=0.68) displayed a mild predictive value.

Nevertheless, this is insufficient to employ these parameters to counsel patients about the success of the sperm wash. With the threshold values for both the semen and medical features of our HIV population, we were unable to find a single factor able to predict if a sperm wash will prove positive (thus forcing us to discard the semen sample), or not.

Regarding CD4 count or blood viral load in HIV patients (two of the most relevant parameters in the description of the disease status of HIV-infected males), they only have clinical relevance for male health, and they are not helpful for prediction of the wash result. No cut-off value was found to be informative regarding the chances of having a positive or negative result. We lack information regarding the CD4 blood levels of the uninfected males in this study.

Influence of the density gradient and swim-up volumes employed on the sperm wash results
From the evaluation of the volumes of the media employed in the preparation of samples from HCV/HIV-seropositive males, we can conclude that the amount of culture medium employed to wash the semen samples is not relevant for the viral detection results.

Regarding the triple density gradient layers, comparable volumes were prepared in the samples which yielded positive results and those that proved negative (1.0±0.2 versus 1.2 ± 0.2 ml).

Also, the mean number of tubes employed for the gradients was 2.8 ± 1.3 in the positive samples and 2.5 ± 1.2 in the negative samples; this difference was not statistically different. The mean number of tubes employed in the swim-up procedures of the positive samples was 2.6 ± 1.5, while the negatives samples were processed in 2.4 ± 1.2 tubes.

Also, the mean volume of the positive samples was 0.9 ± 0.3 ml, and that employed for the swim-up when the samples were negative was equivalent (0.9 ± 0.3 ml).

Correlation between HIV infection and semen parameters
When semen parameters were correlated with CD4 blood levels, evolution time of the disease and blood viral load, we only found a significant negative correlation between B and (A+B) forms in the ejaculate and the time of the evolution of the disease, and a positive correlation between B and (A+B) forms after the sperm wash with the blood CD4 numbers (Table III).


View this table:
[in this window]
[in a new window]
 
Table III. Correlation between CD4, evolution of the disease and semen parameters of HIV males

 
Moreover, as reflected in Table IV, there are no differences in the semen or disease parameters between those samples that were positive in comparison with those that were negative.

Semen characteristics depending on antiretroviral treatment in HIV males
Antiretroviral drug intake did not significantly affect seminal quality, as reflected by the results of Table V. All the parameters analysed were comparable between groups, irrespective of whether or not the patient was taking any antiretroviral therapy.


View this table:
[in this window]
[in a new window]
 
Table V. Results between HIV patients receiving or not receiving antiretroviral drugs

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
In this work, we have provided relevant data regarding the semen features of HIV and HCV males. Moreover, we have demonstrated that no parameter can predict the success of a sperm wash protocol. Whether a positive result is a consequence of an incorrect laboratory manipulation still needs to be determined. We believe that an ‘incorrect manipulation’ could occur when a positive result after wash is found, because PCR techniques are extremely sensitive, and there is a possibility that some viral particles still remained within the washed samples, and this could be provoked by faulty manipulation when eliminating seminal plasma or density gradients. The working area is disinfected extensively after each protocol, thus eliminating any possible cross-contamination.

Moreover, we do not believe in the possibility of the presence of intracellular viral load, although an alternative route to CD4 is found by the sperm to get viral particles inside the cell (Bandivdekar et al., 2003Go), since we are able to detect even one single copy, meaning that no sperm has virus inside.

Also, a positive result is not characteristic of an individual, since every second wash performed on patients whose first sperm wash resulted positive had the second result of the wash free of viral load.

The modus operandi in the laboratory is focused on two aspects: first, we need to eliminate any viral particle, since nPCR is an extremely sensitive method, and any viral residue will be detected, thus yielding a positive result and forcing us to discard the sample. Secondly, we must optimize the sperm washing procedure to ensure that we maintain as many motile sperm as possible, to permit as many ICSI procedures as possible. Then, if the manipulation in the laboratory is relevant for the results, this is a delicate equilibrium that sometimes can be broken, therefore producing a positive result.

To this end, we have demonstrated that there is no relevance of the density gradients or swim-up volume in the results, since no differences can be found between the treatments practised on the samples that yielded positive results after the sperm wash and those that were negative after the nPCR determinations.

Regarding the correlation between semen parameters, blood CD4 levels and viral load levels, there is a clear confirmation that, as the disease evolves in time, sperm motility is not impaired. This is in contradiction to the results found by other groups, in which the total volume is diminished as long as the infection progresses in time, thus increasing sperm concentration since total progressive sperm numbers were maintained (Dulioust et al., 2002Go).

Concerning CD4 cell counting, there is a positive correlation with the ability of spermatozoa to undergo capacitation. This unequivocally indicates that an adequate immune status favours the production of spermatozoa which are able to trespass two consecutive semen preparations. We were unable to observe any change in sperm concentration, as other findings show. (Crittenden et al., 1992Go; Politch et al., 1994Go; Dondero et al., 1996Go; Muller et al., 1998Go).

Both findings are important just in males where semen characteristics are severely impaired, and these small changes in motility related to CD4 cell blood counts and time of the disease's evolution can make the sperm wash procedure impractical.

Receiving antiretroviral treatment seems not to impair spermatogenesis in HIV-seropositive males, since the results of the semen analysis were comparable. Studies favouring (Robbins et al., 2001Go) or against this hypothesis are available in the literature. Moreover, this had no influence either on the results of the sperm wash or on the semen characteristics after the sperm wash.

The frequency of sperm wash-positive samples found in our work is slightly higher than previous works have demonstrated. For instance, Marina et al. (1998)Go found that 5% of the samples are positive. Other works analysing the effectiveness of sperm wash were unable to detect positive samples after the procedure in the laboratory (Kim et al, 2001Go). Moreover, this group was unable to detect the presence of HIV receptors on the prepared sperm cells, thus adding strength to the theory that spermatozoa cannot be infected by HIV.

Also, we are unable to find a factor able to predict the result accurately after sperm wash. All the parameters analysed had very low predictive values, and the only information that we can provide to the patients undergoing a sperm wash is what our experience reflects: ~10% of the samples resulted positive and these probabilities are higher as the number of determinations increases. For instance, a HIV/HCV-infected male needs three PCR determinations (HIV DNA, HIV RNA and HCV RNA), while an HCV-infected patient needs only one PCR (HCV RNA), and a patient infected with HIV needs two (viral and proviral HIV detection).

In conclusion, HIV- and HCV-infected patients now have the possibility to become fathers, avoiding viral transmission to the mother and future child. Sperm parameters are not significantly different from those that the World Health Organization considers to be normal. We were unable to detect a parameter that could help us in the prediction of the sperm wash results, which remained positive in 10% of the samples. Low CD4 counts together with a long time of evolution of the disease are relevant for the sample, but insignificant for the whole treatment. Given these results, patients must be counselled accordingly before the initiation of a treatment.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors wish to acknowledge Drs Gimeno and Garcia de Lomas, from the Instituto Valenciano de Microbiología, for their kind help in developing nested PCR for HIV and HCV determinations in washed sperm, and M.J.Morata, Y.Márquez, A.Zurilla, P.Rodriguez and C.Rico for their assistance with processing of infected semen.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Bandivdekar AH, Velhal SM and Raghavan VP (2003) Identification of CD4-independent HIV receptors on spermatozoa. Am J Reprod Immunol 50, 322–327.[CrossRef][ISI][Medline]

Bellver J, Meseguer M, Garrido N and Ruiz A (2002) Virus de la inmunodeficiencia humana y reproducción. In Remohí J, Simón C, and Pellicer A (eds) Reproducción Humana. pp. 373–384, Mc Graw-Hill, Madrid, Spain.

Bourlet T, Levy R, Laporte S, Blachier S, Bocket L, Cassuto G, Chollet L, Crittenden JA, Handelsman DJ and Stewart GJ (1992) Semen analysis in human immunodeficiency virus infection. Fertil Steril 57, 1294–1299.[ISI][Medline]

Crittenden JA, Handelsman DJ, Stewart GJ (1992) Semen analysis in human immunodeficiency virus infection. Fertil Steril 57(6), 1294–9.[ISI][Medline]

Dondero F, Rossi T, D'Offizi G, Mazzilli F, Rosso R, Sarandrea N, Pinter E and Aiuti F (1996) Semen analysis in HIV seropositive men and in subjects at high risk for HIV infection. Hum Reprod 11, 765–768.[Abstract]

Dulioust E, Du AL, Costagliola D, Guibert J, Kunstmann JM, Heard I, Juillard JC, Salmon D and Leruez-Ville M (2002) Semen alterations in HIV-1 infected men. Hum Reprod 17, 2112–2118.[Abstract/Free Full Text]

Garrido N, Meseguer M, Simón C, Pellicer A and Remohí J (2001) Detección del virus de la Hepatitis C en semen lavado de varones seropositivos para su posterior uso en reproducción asistida: resultados preliminares. Actualidad Andrológica 9(2–3), 56–60.

Garrido N, Meseguer M, Simon C, Pellicer A and Remohí J (2004) Assisted reproduction in HIV and HCV infected men of serodiscordant couples. Arch Androl 50, 105–111.[ISI][Medline]

Ho DD, Schooley RT, Rota TR, Kaplan JC, Flynn T, Salahuddin SZ, Gonda MA and Hirsch MS (1984) HTLV III in the semen and blood of the semen and blood of healthy homosexual man. Science 226, 451–453.[ISI][Medline]

Kim LU, Johnson MR, Barton S, Nelson MR, Sontag G, Smith JR, Gotch FM and Gilmour JW (2001) Evaluation of sperm washing as a potential method of reducing HIV transmission in HIV-discordant couples wishing to have children. AIDS 16, 429–430.[CrossRef]

Levy R, Tardy JC, Bourlet T, Cordonier H, Mion F, Lornage J and Guerin JF (2000) Transmission risk of hepatitis C virus in assisted reproductive techniques. Hum Reprod 15, 810–816.[Abstract/Free Full Text]

Marina S, Marina F, Alcolea R, Exposito R, Huguet J, Nadal J and Verges A (1998) Human immunodeficiency virus type 1-serodiscordant couples can bear healthy children after undergoing intrauterine insemination. Fertil Steril 70, 35–39.[CrossRef][ISI][Medline]

Meseguer M, Garrido N, Gimeno C, Remohi J, Simon C and Pellicer A (2002) Comparison of polymerase chain reaction-dependent methods for determining the presence of human immunodeficiency virus and hepatitis C virus in washed sperm. Fertil Steril 78, 1199–1202.[CrossRef][ISI][Medline]

Muller CH, Coombs RW and Krieger JN (1998) Effects of clinical stage and immunological status on semen analysis results in human immunodeficiency virus type 1-seropositive men. Andrologia 30 (Suppl 1), 15–22.

Peter JB and Sevall JS (2004) Molecular-based methods for quantifying HIV viral load. AIDS Patient Care, 75–79. STDS18.

Politch JA, Mayer KH, Abbott AF and Anderson DJ (1994) The effects of disease progression and zidovudine therapy on semen quality in human immunodeficiency virus type 1 seropositive men. Fertil Steril 61, 922–928.[ISI][Medline]

Robbins WA, Witt KL, Haseman JK, Dunson DB, Troiani L, Cohen MS, Hamilton CD, Perreault SD, Libbus B, Beyler SA et al. (2001) Antiretroviral therapy effects on genetic and morphologic end points in lymphocytes and sperm of men with human immunodeficiency virus infection. J Infect Dis 184, 127–135.[CrossRef][ISI][Medline]

Semprini AE, Levi-Setti P, Bozzo M, Ravizza M, Taglioretti A, Sulpizio P, Albani E, Oneta M and Pardi G (1992) Insemination of HIV-negative women with processed semen of HIV-positive partners. Lancet 340, 1317–1319.[CrossRef][ISI][Medline]

World Health Organization (1999) WHO Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interactions, 4th edn. Cambridge University Press,.

Zagury D, Bernard J, Leibowitch J, Safai B, Groopman JE, Feldman M, Sarngadharan MG and Gallo RC (1984) HTLV III in cells cultured from two patients with AIDS. Science 226, 449–451.[ISI][Medline]

Zweig MH and Campbell G (1993) Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in medicine. Clin Chem 39, 561–577.[Abstract/Free Full Text]

Submitted on October 26, 2004; resubmitted on November 22, 2004; accepted on November 26, 2004.