HUMAN IMMUNODEFICIENCY VIRUS (HIV) IS transmitted primarily through vaginal and anal intercourse. The presence of HIV in both blood and semen has been evident since the earliest days of the HIV epidemic.1 Efficient transmission of infectious pathogens requires a critical level of exposure to the infectious agent. The concentration of HIV in semen and its access to susceptible cells ultimately determines HIV transmission from infected men to their sex partners. The genital fluid system and blood are separate immunologic compartments within which HIV harbors and mutates to varying degrees.2–5 Concentrations of HIV in semen, for example, are associated with but not equal to the quantity of HIV in blood plasma. The ability to estimate risks for sexually transmitting HIV based on knowing blood plasma viral load (BPVL), therefore, depends on the magnitude of the association between HIV in blood plasma and HIV in genital fluids.
The degree to which semen viral load (SVL) is associated with BPVL may be influenced by multiple factors. Understanding the relationship between BPVL and SVL is necessary for estimating the potential benefits of antiretroviral (ARV) therapies for reducing HIV transmission risks. In addition, patients who are told that their BPVL is below the level of detection may assume that their viral load test results equally apply to their genital secretions, affirming a sense of less infectiousness.6,7 However, BPVL only reflects the infectiousness of genital secretions to the degree that BPVL is correlated with SVL.
Predicting HIV transmission from BPVL has been most strongly suggested by Quinn et al.8 In a prospective cohort study of HIV serodiscordant couples in Rakai Uganda, Quinn et al. found that the majority of HIV seroconversions occurred in the first weeks of HIV infection. In fact, Quinn et al. observed a dose effect for index patients’ BPVL and the risk for HIV seroconversion in their partners, with seroconversions rarely occurring when partners had viral loads under 1500 copies. These findings suggest that knowing one’s BPVL informs HIV sexual transmission risks and implies a strong association between BPVL and SVL. Examining the correspondence between BPVL and SVL in studies that have directly tested their association may prove useful in extrapolating SVL from BPVL in predicting sexual HIV transmission.
In this article, we review studies that have contemporaneously measured HIV BPVL and SVL. We examined the correlations between BPVL and SVL as well as the factors that influence their concordance and discordance. The strength and variability of BPVL-SVL associations are discussed in terms of their implications for HIV prevention and communicating BPVL results to patients.
We conducted automated searches using PubMed in January and February 2007 by using the following key terms: semen, HIV, viral load. We inspected articles identified in the search for whether the results reported the correlation between BPVL and SVL. We then manually examined the references within each of the identified articles and repeated the search. We also searched abstracts from the Conference on Retroviruses and Opportunistic Infections and the International AIDS Society. We identified 18 articles that met the inclusion criterion of reporting a correlation between BPVL and SVL. For one additional study, we calculated the BPVL-SVL correlation using subject-level data reported in the article.9
The greatest limitation of the studies that have assessed the correlation between BPVL and SVL is their generally small samples. Correlation coefficients are sensitive to sample sizes, particularly when there are extreme values in the distributions. The sample sizes of studies should therefore be taken into account when interpreting these findings. In terms of sample characteristics, studies have varied in their recruitment and selection procedures. Some studies, for example, have recruited men who were participating in medication trials where all men were treated with the same drug regimen, whereas other studies have recruited from community social services that included men who may not have been treated at all. Heterogeneous HIV treatment regimens influence the strength of the association between BPVL and SVL because of differential suppressive effects of ARV medications in the blood and genital compartments. Changes in HIV treatment regimens available over the years may also influence the association of BPVL and SVL, although the extent of the influence is unclear. As treatments change over time there will likely be more changes in the association between BPVL and SVL.
Studies have also varied in their detection and treatment of sexually transmitted infections (STI). Confidence in detecting recent STI is of particular importance when interpreting the association between BPVL and SVL because urethritis increases viral shedding in the genital tract but not in the blood. Another important limitation of the literature is its reliance on single time cross-sectional assessments. Repeated correlations allow for estimates of the stability of the association between BPVL and SVL over time as well as the predictors of changes in the correlation. Nearly every study we reviewed reported a single correlation between BPVL and SVL from cross-sectional data, with only 2 studies reporting prospective data.
Finally, a noteworthy strength of this literature is that nearly all of the studies we reviewed used the same assay for detecting HIV RNA in semen, the NASBA or nucleic acid sequence based amplification (NucliSens) assay; an assay that is relatively unaffected by factors in semen that inhibit HIV RNA detection.10 The few studies did not use a NucliSens assay did use a silicon process for removing inhibitors that are known to affect polymerase chain reaction when applied to SVL. These characteristics of the studies reviewed should be considered when interpreting their results.
Associations Between BPVL and SVL
Table 1 presents the summary of 19 studies that report the correlation between BPVL and SVL. As shown in the table, correlations ranged between 0.07 and 0.64. One study found nearly perfect (94%) concordance between SVL and BPVL,21 precluding calculating a correlation coefficient, whereas other studies did not find the correlation significant.23,29 The total number of men included across studies was 1226 and the mean correlation was 0.45 (SD = 0.20, median = 0.45).
Examining the studies in relation to their year of publication shows that the magnitude of correlations did not show a consistent pattern of change over time. The 8 studies conducted before 2000, midway between the first and last studies reviewed, had a mean correlation of 0.52, whereas the 9 studies conducted 2000 and later had a mean correlation of 0.41. It is noteworthy that the outlying study that found the highest degree of concordance between BPVL and SVL had the most stringently selected sample of men in terms of treatment status and not having STIs,21 whereas the study with the lowest correlation only required being HIV-positive as the sole selection criterion.23
Studies have consistently found that HIV is less concentrated in semen than in blood plasma. However, there is considerable variability in the relative concentrations of HIV in the 2 components both between and within studies. In most cases, men whose viral load was undetectable in their semen had corresponding undetectable viral loads in their blood plasma. Studies, however, identify men with SVL that is equal to or greater than their BPVL,11,28 and studies also indicate detectable SVL among men whose BPVL was below levels of detection, with the difference often exceeding 1 log. For example, Vernazza et al.29 compared the viral loads of 28 men who had detectable BPVL and SVL and found that only 12 of the men had RNA levels in blood plasma and semen within 1 log of each other. The investigators found that 9 men demonstrated SVL 10-fold lower than their BPVL, whereas 7 men had SVL 10-fold higher than their BPVL. Also noteworthy is research conducted in Malawi, which found a consistent pattern of higher SVL among Malawian men compared with a US/Swiss sample,30–32 a difference that may be attributable to co-occurring STI.
Factors Influencing the Association Between BPVL and SVL
Our review identified 4 major factors that potentially influence the relationship between BPVL and SVL: co-occurring STI, ARV treatments and adherence, HIV treatment resistance, and stage of HIV disease.
There is overwhelming evidence that co-occurring STI increase viral shedding in the genital tract and that STI significantly alter the correspondence between BPVL and SVL.33 For example, Cohen et al.30 reported that HIV shedding in semen is 8 times greater in Malawian men with urethritis when compared with men matched on CD4 cell counts and BPVL; men who had urethritis had a median SVL of 12.4 log compared with 1.5 log in the matched control group. In a case-control study, Sadiq et al.27 found a 5-fold increase in SVL among men with gonorrhea and chlamydia compared with men who were not diagnosed with an STI. It appears that asymptomatic urethritis similarly influences viral shedding. Winter et al.16 found that asymptomatic urethritis was strongly associated with increases in SVL. The increased number of HIV-infected cells in semen that accompanies genital tract inflammation is also a potential factor in sexual transmission of HIV.34 Elevated HIV shedding and increases in infected cells in semen of men who have asymptomatic urethritis is of particular concern because it is these men who are unlikely treated for urethritis and are most likely engaging in unprotected sex. Greater numbers of sex partners and higher rates of sexual intercourse may be associated with more persistent genital shedding of HIV even in the absence of STI.23 Because STI increases genital HIV shedding but not HIV shedding in blood plasma, the correlation between BPVL and SVL is suppressed by STI. In fact, studies with the lowest correlations between BPVL and SVL are those that are most likely to have included men with co-occurring STI.
HIV Treatment and Adherence
ARV therapies suppress HIV replication in the genital tract to similar degrees as observed in peripheral blood.35 Although most studies have reported cross-sectional findings, Neto et al.9 reported serial correlations between BPVL and SVL taken over the course of initiating ARV therapy. In this study, the strength of the association between BPVL and SVL increased over the time men were receiving treatments. ARV therapies can markedly reduce HIV concentrations in blood plasma and semen,13 supporting the hypothesis that suppressive ARV therapies reduce the risk for HIV transmission.36 Most studies investigating the effects of ARVs on SVL show suppression of HIV replication in the genital compartment.37 HIV treatment regimens do not equally penetrate the blood plasma and genital tract compartments.22,37 Therefore, various ARV regimens differentially influence the correlation between BPVL and SVL. For example, in a study of men who had been treated with a combination ARV regimen for at least 6 months, penetration of protease inhibitors in the genital tract was variable.38 Chakraborty et al.39 showed that BPVL and SVL are not correlated when men are not treated for HIV regardless of their CD4 count and that, in contrast, the correlation between BPVL and SVL is strongest (approximately 0.40) when men are treated with ARVs and their CD4 cell counts are low (<200). Under optimal conditions, when men are treated with a highly suppressive ARV regimen, are adherent to taking their medications, and do not have co-occurring STI, one study estimates with 95% certainty that <4% of men who have an undetectable BPVL will have detectable SVL.21 Unfortunately, these optimal conditions are rarely met outside of research settings.
Failure for ARVs to achieve undetectable SVL is attributable to several factors, including drug potencies, absorption, and drug penetration.22 Nonadherence to ARV regimens is another important source of suboptimal viral suppression. Discontinuation of ARVs results in viral rebound in SVL similar to that seen in BPVL.40 In studies of treatment-naive men who initiated ARVs, significant numbers of men still have detectable SVL after months of treatment.19,41,42 This study showed that the single best predictor of persistent genital shedding of HIV was nonadherence to ARVs. Barroso and Schechter41 found a dose-response association between ARV adherence and SVL, where men who missed the fewest doses of their medications had the greatest degree of viral suppression in their semen over time.
HIV Treatment Resistance
Men who receive combination ARV regimens have demonstrated drug-resistant mutations in the genital tract.4,42–44 The varieties of HIV mutations observed in blood plasma and semen often differ in the 2 compartments, including among treatment-naive men.45 Multiple variants of HIV occur within an individual’s blood and semen. Even within the blood and semen compartments there are heterogeneous cell-free and cell-associated variants of HIV.46 In addition, multiple drug-resistant HIV can occur in semen without a blood plasma complement.47 Thus, differential treatment resistance in the blood and genital compartments, similar to differential drug penetration, suppresses the association between BPVL and SVL. There is considerable alarm about the potential spread of multiple drug-resistant HIV from men with resistant HIV in their semen who contract a co-occurring STI.43 Although suppressive ARV therapy can reduce viral shedding in men who have urethritis, when HIV is poorly controlled, the risk for transmitting treatment-resistant variants is particularly high.48
Stage of HIV Disease
HIV shedding in the genital compartment occurs over the entire course of HIV disease.37 Studies that correlate SVL with CD4 cell counts show a wide range of associations. For example, Ball et al.17 found a close association between CD4 counts, the most common marker of HIV disease progression and HIV proviral DNA in both blood and semen, whereas Winter et al.16 found no association between CD4 counts and SVL. In addition, studies that report HIV symptoms as well as other markers of HIV disease progression do not find stages of HIV infection to influence the association between BPVL and SVL. For example, Vernazza et al.12 found that the association between SVL and HIV symptoms was not significant. Other studies have reported that HIV concentrations in semen are independent of HIV disease stage.49 In addition, one study compared serial samples from 5 untreated men and found very stable levels of SVL over 8 weeks,50 suggesting that SVL is relatively stable at least over a short period of time.
Conclusions and Implications
HIV concentrations in blood plasma and semen are correlated, with the magnitude of the associations ranging from 0.07 to over 0.60. The strength of the association between BPVL and SVL is influenced by multiple factors; co-occurring STI/urethritis weakens the association whereas ARV treatments strengthen the association. In contrast, there is limited evidence that the association between BPVL and SVL is influenced by HIV disease stage. Thus, for ARV treatments to significantly impact seminal infectivity, men must be free of co-occurring STI/urethritis and remain adherent to an HIV-suppressive treatment regimen that optimally penetrates the genital tract and to which they are not resistant in their genital compartment.
Our review was limited to studies of HIV in blood plasma and semen. However, research has investigated the correlation between HIV in female genital secretions and blood plasma with very similar results. For example, Hart et al.51 reported that BPVL correlated 0.64 with HIV RNA in vaginal secretions. The degree to which factors that influence HIV in the female genital compartment parallel the male genital tract, as well as factors that are unique to women (e.g., the menstrual cycle, changes in vaginal environment, etc.) should be the subject of review. Our literature review consisted of research from North and South America, Western Europe, and one study from Africa. Further research is therefore needed to investigate the association between BPVL and SVL in countries where HIV prevalence is greatest and ARV treatments are just becoming available.
An implication of the association between BPVL and SVL is how individual viral load test results are communicated to patients. Research shows that both HIV-infected and uninfected men who believe that having an undetectable viral load reduces the risks for HIV transmission are more likely to engage in higher-risk sexual behaviors.6 The disinhibiting effects of having an undetectable viral load on sexual risk behavior will likely offset any potential protective benefits of reductions in semen infectivity. However, semen that has an undetectable viral load remains infectious12 and knowing one’s BPVL at best only accounts for one third of the variance (the squared correlation) in knowing one’s SVL.
Individuals who engage in unprotected sex are often less adherent to their medications than those who practice safer sex,23,52,53 and they are therefore more likely to contract STI. In addition, treatment nonadherence has been associated with increased viral shedding in semen.23,41 Men who engage in unprotected sexual intercourse are also likely to use their viral load status in their sexual decision making, where having an undetectable viral load relaxes concerns about transmitting HIV.7 Therefore, those who are at greatest risk for transmitting HIV may believe that they are actually at less risk based on false assumptions of their SVL in relation to their BPVL. Also of concern is the role of cell-associated virus in HIV transmission. Seminal cells that harbor HIV proviral DNA can serve as vehicles for sexual transmission of HIV,54 particularly HIV-infected lymphocytes and macrophages.55 SVLs can be greater than BPVL,18 and men who have undetectable SVL must be considered at least somewhat infectious because the minimum quantity of HIV required for sexual transmission of the virus is unknown12 and because cell-associated virus may play a role in HIV transmission. HIV prevention messages targeted to both infected and uninfected persons should communicate the importance of condoms and other risk reduction strategies regardless of HIV treatment status and throughout all stages of HIV disease. Perhaps most critically, HIV prevention for people living with HIV/acquired immunodeficiency syndrome must include regular monitoring and aggressive treatment of co-occurring STI.
1. Ho DD, Schooley R, Rota T, et al. HTLV-III in the semen and blood of a healthy homosexual man. Science 1984; 226:451–453.
2. Byrn RA, Zhang D. HIV-1 in semen: An isolated virus reservoir. Lancet 1997; 350:1141.
3. Craigo J, Patterson B, Paranjpe S, et al. Persistent HIV type 1 in semen and blood compartments in patients after long-term and potent antiretroviral therapy. AIDS Res Hum Retroviruses 2004; 20:1196–1209.
4. Eron JJ, Vernazza PL, Johnston DM, et al. Resistance of HIV-1 to antiretroviral agents in blood and seminal plasma: Implications for transmission. AIDS 1998; 12:F181–F189.
5. Ping LH, Cohen MS, Hoffman I, et al. Effects of genital tract inflammation on human immunodeficiency virus type 1 V3 populations in blood and semen. J Virol 2000; 74:8946–8952.
6. Crepaz N, Hart T, Marks G. Highly active antiretroviral therapy and high risk sexual behavior: A meta-analytic review. JAMA 2004; 292:224–236.
7. Kalichman SC, Eaton L, Cain D, et al. Changes in HIV treatment beliefs and sexual risk behaviors among gay and bisexual men, 1997–2005. Health Psychol. In press.
8. Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med 2000; 342:921–929.
9. Neto VNP, Soprani M, Cunha CB, et al. Longitudinal comparison between plasma and seminal HIV-1 viral loads during antiretroviral treatment. Rev Soc Bras Med Trop 2003; 36:689–694.
10. Dunne AL, Mitchell F, Allen KM, et al. Analysis of HIV-1 viral load in seminal plasma samples. J Clin Virol 2003; 26:239–245.
11. Liuzzi G, Chirianni A, Clementi M, et al. Analysis of HIV-1 load in blood, semen and saliva: Evidence for different viral compartments in a cross-sectional and longitudinal study. AIDS 1996; 10:1058–1059.
12. Vernazza PL, Gilliam BL, Flepp M, et al. Effect of antiviral treatment on the shedding of HIV-1 in semen. AIDS 1997; 11:1249–1254.
13. Gupta P, Mellors J, Kingsley L, et al. High viral load in semen of human immunodeficiency virus type 1-infected men at all stages of disease and its reduction by therapy with protease and non-nucleoside reverse transcriptase inhibitors. J Virol 1997; 71:6271–6275.
14. Coombs R. Association between culturable human immunodeficiency virus type 1 (HIV-1) in semen and HIV-1 RNA levels in semen and blood: Evidence for compartmentalization of HIV-1 between semen and blood. J Infect Dis 1998; 177:320–330.
15. Dulioust E, Tachet A, De Almeida M, et al. Detection of HIV-1 in seminal plasma and seminal cells of HIV-1 seropositive men. J Reprod Immunol 1998; 41:27–40.
16. Winter AJ, Taylor S, Workman J, et al. Asymptomatic urethritis and detection of HIV-1 RNA in seminal plasma. Sex Transm Infect 1999; 75:261–263.
17. Ball JK, Curran R, Irving WL, et al. HIV-1 in semen: Determination of proviral and viral titres compared to blood, and quantification of semen leukocyte populations. J Med Virol 1999; 59:356–363.
18. Tachet A, Dulioust E, Salmon D, et al. Detection and quantification of HIV-1 in semen: Identification of a subpopulation of men at high potential risk of viral sexual transmission. AIDS 1999; 13:823–831.
19. Barroso PF, Schechter M, Gupta P, et al. Effect of antiretroviral therapy on HIV shedding in semen. Ann Intern Med 2000; 133:280–284.
20. Shepard RN, Schock J, Robertson K, et al. Quantitation of human immunodeficiency virus type 1 RNA in different biological compartments. J Clin Microbiol 2000; 38:1414–1418.
21. Vernazza P, Troiani L, Flepp MJ, et al. Potent antiretroviral treatment of HIV-infection results in suppression of the seminal shedding of HIV. AIDS 2000; 14:117–121.
22. Eron JJ, Smeaton LM, Fiscus SA, et al. The effects of protease inhibitor therapy on human immunodeficiency virus type 1 levels in semen (AIDS Clinical Trials Group Protocol 850). J Infect Dis 2000; 181:1622.
23. Kalichman SC, Cage M, Rompa D, et al. Human immunodeficiency virus in semen and plasma: Investigation of sexual transmission risk and behavioral correlates. AIDS Res Hum Retroviruses 2001; 17:1695–1703.
24. Celum C, Buchbinder S, Donnell D, et al. Early human immunodeficiency virus (HIV) infection in the HIV Network for prevention trials vaccine preparedness cohort: Risk behaviors, symptoms, and early plasma and genital tract virus load. J Infect Dis 2001; 183:23–35.
25. Stekler J, Sycks B, Holte S, et al. Semen HIV dynamics and effect of ART following primary infection. Presented at: the 10th Conference on Retroviruses and Opportunistic Infections, 2006. Abstract available at: www.retroconference.org/2006/cd/Abstract/26208.htm
26. Medeiros RP, Munerato P, Diaz RS. HIV-1 viral load in blood and semen plasma of Brazilian patients under antiretroviral therapy. J Clin Virol 2004; 30:46–7.
27. Sadiq S, Taylor S, Copas AJ, et al. The effects of urethritis on seminal plasma HIV-1 RNA loads in homosexual men not receiving antiretroviral therapy. Sex Transm Infect 2005; 81:120–123.
28. Taylor S, Sadiq T, Sabin C, et al. Seminal super shedding of HIV: Implications for sexual transmission. Presented at: the 10th Conference on Retroviruses and Opportunistic Infections, 2003. Abstract available at: www.retroconference.org/2003/cd/Abstract/454.htm
29. Vernazza P, Gilliam B, Dyer J, et al. Quantification of HIV in semen: Correlation with antiviral treatment and immune status. AIDS 1997; 11:987–993.
30. Cohen M, Hoffman I, Royce R, et al. Reduction of concentration of HIV-1 in semen after treatment of urethritis: Implications for prevention of sexual transmission of HIV-1. Lancet 1997; 349:1868–1873.
31. Dyer JR, Hoffman IF, Kazembe P, et al. Association of CD4 cell depletion and elevated blood and seminal plasma human immunodeficiency virus type 1 (HIV-1) RNA concentrations with genital ulcer disease in HIV-1-infected men in Malawi. J Infect Dis 1998; 177:224–227.
32. Dyer JR, Vernazza PL, Gilliam BL, et al. High levels of human immunodeficiency virus type 1 in blood and semen of seropositive men in sub-Saharan Africa. J Infect Dis 1998; 177:1742–1746.
33. Atkins MC, Carlin EM, Emery VC, et al. Fluctuations of HIV load in semen of HIV positive patients with newly acquired sexually transmitted diseases. BMJ 1996; 313:341–342.
34. Xu C, Politch J, Tucker L, et al. Factors associated with increased levels of human immunodeficiency virus type 1 DNA in semen. J Infect Dis 1997; 176:941–947.
35. Gilliam DJ, Fiscus SA, Marcus C. Effects of reverse transcriptase inhibitor therapy on the HIV-1 viral burden in semen. J AIDS 1997; 151:54–60.
36. Taylor S, Boffito M, Vemazza PL. Antiretroviral therapy to reduce the sexual transmission of HIV. J HIV Ther 2003; 8:55–66.
37. Kashuba A, Dyer J, Kramer L, et al. Antiretroviral drug concentration in semen: Implications for sexual transmission of human immunodeficiency virus type 1. Antimicrob Agents Chemother 1999; 43:1817–1826.
38. Solas C, Lafeuillade A, Halfon P, et al. Discrepancies between protease inhibitor concentrations and viral load in reservoirs and sanctuary sites in human immunodeficiency virus-infected patients. Antimicrob Agents Chemother 2003; 47:238–243.
39. Chakraborty H, Helms RW, Sen PK, et al. Estimating correlation by using a general linear mixed model: Evaluation of the relationship between the concentration of HIV-1 RNA in blood and semen. Stat Med 2003; 22:1457–1464.
40. Liuzzi G, D’Offizi G, Topino S, et al. Dynamics of viral load rebound in plasma and semen after stopping effective antiretroviral therapy. AIDS 2003; 17:1089–1092.
41. Barroso PF, Schechter M. Adherence to antiretroviral therapy and persistence of HIV RNA in semen. J AIDS 2003; 32:435.
42. Mayer KH, Boswell S, Goldstein R, et al. Persistence of human immunodeficiency virus in semen after adding indinavir to combination therapy. Clin Infect Dis 1999; 28:1252–1259.
43. Taylor S. Drug-resistant HIV-1 in the semen of men receiving antiretroviral therapy with acute sexually transmitted infections. Antivir Ther 2003; 8:479–483.
44. Taylor S, Weller I, Kaye S, et al. Drug-resistant HIV-1 in the semen of men receiving antiretroviral therapy with acute sexually transmitted infections. Antivir Ther 2003; 8:479–483.
45. Liuzzi G, Chiarianni A, Zaccarelli M, et al. Differences between semen and plasma of nucleoside reverse transcriptase resistance mutations in HIV-infected patients, using a rapid assay. In Vivo 2004; 18:509–512.
46. Zhu T, Wang N, Carr A, et al. Genetic characterization of human immunodeficiency virus type 1 in blood and genital secretions: evidence for viral compartmentalization and selection during sexual transmission. J Virol 1996; 70:3098–3107.
47. Eyre RC, Zheng G, Kiessling AA. Multiple drug resistance mutations in human immunodeficiency virus in semen but not blood of a man on antiretroviral therapy. Urology 2000; 55:591.
48. Sadiq ST, Taylor S, Kaye S, et al. The effects of antiretroviral therapy on HIV-1 RNA loads in seminal plasma in HIV-positive patients with and without urethritis. AIDS 2002; 16:219–225.
49. Krieger JN, Collier AC, Ross SO, et al. Recovery of human immunodeficiency virus type 1 from semen: Minimal impact of stage of infection and current antiviral chemotherapy. J Infect Dis 1991; 163:386–388.
50. Hamed WM, Holodniy M, Katzenstein DA, et al. Detection of human immunodeficiency virus type 1 in semen: Effects of disease stage and nucleoside therapy. J Infect Dis 1993; 167:798–802.
51. Hart CE, Lennox JL, Pratt-Palmore M, et al. Correlation of human immunodeficiency virus type 1 RNA levels in blood and the female genital tract. J Infect Dis 1999; 179:871–882.
52. Diamond C, Richardson J, Milan J, et al. Use of and adherence to antiretroviral therapy is associated with decreased sexual risk behavior in HIV clinic patients. J AIDS 2005; 39:211–218.
53. Wilson TE, Barron Y, Cohen M, et al. Adherence to antiretroviral therapy and its association with sexual behavior in a national sample of women with human immunodeficiency virus. Clin Infect Dis 2002; 34:529–534.
54. Zhang H, Dornadula G, Beumont M, et al. Human immunodeficiency virus type 1 in the semen of men receiving highly active antiretroviral therapy. N Engl J Med 17 1998; 339:1803–1809.
55. Quayle A, Mayer K, Anderson D. T-Lymphocytes and macrophages, but not motile spermatozoa, are a significant source of human immunodeficiency virus in semen. J Infect Dis 1997; 176:960–968.