Epidemiological studies and AIDS surveillance data showed that HIV transmission depends on contact with infected body fluids, primarily blood and semen . These compartments are established as reservoirs of HIV-1 infection [2,3], and monitoring the HIV-1 viral load in compartments is essential for assessing transmission risk, disease progression, and response to antiretroviral regimens.
Recent evidence suggests that the oral cavity may actually be an HIV-1 reservoir. Indeed, Freel et al.  documented one case of discordant HIV-1 subpopulations in peripheral blood and in the oral cavity, suggesting that selected individuals may harbour different viral populations in the blood and saliva, as has been widely documented for blood and other body fluids .
Therapeutic regimens including HIV protease inhibitors greatly reduce the viral load in plasma . To gain insights into the efficacy of antiretroviral therapy in reducing the seminal viral load, we evaluated in a longitudinal study the effects of three antiretroviral regimens on changes of viral load in plasma and semen of HIV-1-infected individuals, and demonstrated that highly active antiretroviral therapy (HAART) can dramatically reduce the viral load in the semen of HIV-1-infected subjects .
In the present study, we aimed at evaluating the effects of indinavir-containing regimens on HIV-1-RNA levels in saliva.
Seventeen anti-HIV-1-positive, antiretroviral-naive subjects provided paired (i.e. collected on the same day) saliva and plasma specimens at the beginning of the study and between 4 and 12 weeks later. All patients were men, and their ages ranged between 24 and 45 years. None had received previous lamivudine or protease inhibitor treatment. The median plasma HIV-1-RNA level at enrolment was 59 260 copies/ml, and the CD4 cell count was 234/mm3. All patients received 300 mg zidovudine or 40 mg stavudine with 150 mg lamivudine, given twice a day in combination with 800 mg indinavir three times a day.
The presence in the plasma and saliva of specific HIV-RNA sequences was first assessed by qualitative reverse transcriptase–polymerase chain reaction (RT–PCR). A competitive RT–PCR-based assay was used to quantify cell-free HIV-1-RNA molecules in plasma and saliva samples; the limit of detection of competitive RT–PCR is one copy per 100 μl plasma .
At baseline, HIV-1 RNA was detectable in 14 out of the 17 samples of saliva (82.4%), but only five out of 17 patients showed HIV-1-RNA levels greater than 500 copies/ml. By contrast, HIV-1 RNA in plasma was greater than 500 copies/ml in all patients. The mean concentration was 247 copies/ml (range 0–86 240 copies/ml) in saliva compared with 59 260 copies/ml (1034–886 780) in plasma. After a period of treatment ranging between 8 and 12 weeks, 13 subjects (76.5%) were free of HIV-1 RNA in saliva when tested by RT–PCR, and the HIV-RNA level in saliva was less than 100 copies/ml in the remaining four patients. By contrast, HIV-1 RNA in plasma was less than 100 copies/ml in only eight out of 17 patients (47.1%), although all experienced at least a 1.5 log reduction (Table 1).
The transmission of HIV-1 through the oral cavity may occur during breastfeeding [8,9], oral–genital sex , and direct deposition in macaques , despite the observed risk being low.
Therefore, oral sexual activities that entail a massive exchange of saliva cannot be considered ‘safe sex’ even if the risk of infection is exceedingly small. In some patients, the HIV-1-RNA level in saliva was found to be higher than that in semen . This is very telling, because within the framework of the sexual transmission of HIV, semen is considered to be highly infective. Moreover, the medium level of viral load in saliva, measured using competitive RT–PCR, in subjects with CD4 cell counts of less than 200/mm3, was higher than that found in maternal milk . This finding is also relevant because breastfeeding is thought to play an important role in the increased transmission rate of HIV-1 from mother to infant [8,9].
Antiretroviral therapy has a powerful effect on reducing HIV-1-RNA levels in blood  and semen . Our data demonstrated that HAART can also constantly reduce the HIV-RNA level in saliva, achieving very low levels in all patients analysed and undetectable levels in the majority of patients, also considering that the more sensitive qualitative method was used. The proportion of patients reaching an undetectable RNA level in saliva was considerably higher than that in plasma. This result was probably related to the baseline RNA levels in saliva, which were generally 1–2 logs lower than in plasma.
Our findings, observed in 8–12 weeks of HAART in naive patients, imply that saliva is virtually risk free, with regard to HIV transmission.
1. Sande MA. Transmission of AIDS. The case against casual contagion.
N Engl J Med 1986, 314: 380–382.
2. 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: F51–F56.
3. 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.
4. Freel SA, Williams JM, Nelson JAE. et al
. Characterization of human immunodeficiency virus type 1 in saliva and blood plasma by V3-specific heteroduplex tracking assay and genotype analyses. J Virol 2001, 75: 4936–4940.
5. Schrager LK, D'Souza MP. Cellular and anatomical reservoirs of HIV-1 in patients receiving potent antiretroviral combination therapy. JAMA 1998, 280: 67–71.
6. Hammer SM, Squires KE, Hughes MD. et al
. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and Cd4 counts of 200 per cubic millimeter or less. N Engl J Med 1997, 337: 725–733.
7. Liuzzi G, Chirianni A, Bagnarelli P, Clementi M, Piazza M. A combination of nucleoside analogues and a protease inhibitor reduces HIV-1 RNA levels in semen: implications for sexual transmission of HIV infection. Antivir Ther 1999, 4: 95–99.
8. Lewis P, Nduati R, Kreiss JK. et al
. Cell-free human immunodeficiency virus type 1 in breast milk. J Infect Dis 1998, 177: 34–39.
9. Dunn DT, Newell ML, Ades AE, Peckham CS. Risk of human immunodeficiency virus type 1 transmission through breast feeding. Lancet 1992, 7: 177–184.
10. Rothenberg RB, Scarlett M, del Rio C, Reznik D, O'Daniels C. Oral transmission of HIV. AIDS 1998, 12: 2095–2105.
11. Baba TW, Trichel AM, Li A. et al
. Infection and AIDS in adult macaques after nontraumatic oral exposure to cell-free SIV. Science 1996, 272: 1486–1489.