HIV may be present in many body fluids, but it does not follow that its presence will constitute a risk for infection. This is because it may be present in many body fluids only in very low concentrations. Research shows that levels of HIV vary hugely across time in the same individual, with the highest levels detectable just after infection and at the development of symptomatic HIV disease. It is estimated (Galvin) that the risk of HIV infection to a woman from a single instance of unprotected vaginal intercourse varies from one in 25 to one in 1,000 if the man is in the first three weeks of HIV infection, but that this falls to one in 1,000 to one in 10,000 if he is chronically infected.

It is important to note that it is almost impossible to isolate HIV from some body fluids, such as urine.

In what form is HIV infectious?

HIV's genetic material may be passed from one individual to another in several different forms.

It may be passed as free virus particles (HIV RNA), which are able to bind onto cells with CD4 markers.

It may be passed within HIV–infected lymphocytes as proviral DNA. In this case the virus is transcribed into the genetic material of the host cell and will begin to replicate when that immune system cell is activated.

In semen, for instance, it exists in both forms; as virus particles in the nutrient carrier fluid; and as proviral DNA within lymphocytes. It is not present in sperm cells themselves.

The difference between viral load and infectious particles

Throughout this chapter you will see the terms `viral load' and `infectious particles' used. These terms refer to different measurements:

• `Infectious particles' refers to the number of viruses present in millilitre of a body fluid. This is assessed by a test which detects whole viruses. It is slow and laborious to conduct, and if there are very low levels of HIV present in a body fluid the test will not detect the HIV particles.
• `Viral load' refers to the results of a test designed to amplify any genetic material of the virus in a sample of a body fluid. This test uses a template which corresponds to a genetic sequence of HIV which does not tend to vary between individuals. The test detects the genetic sequence and amplifies it by attaching a marker chemical to make the genetic sequence visible. The test will go through several rounds of amplification to increase the chance that HIV's genetic sequence can be detected. The final result is described as the number of HIV RNA copies per millilitre. The number is usually at least tenfold greater than the number of infectious HIV particles.

Repeat viral load tests of the same blood sample can give results that vary by a factor of 3. This means that a meaningful change would be a drop to less than 1/3 or an increase to more than 3 times the previous test result. For example, a change from 200,000 to 600,000 is within the normal variability of the test. A drop from 50,000 to 10,000 would be significant.

It is important to note than no more than 2% of the HIV in a person’s body is in the blood at any one time; the viral load test does not measure how much HIV is in body tissues like the lymph nodes, spleen, or brain. HIV levels in lymph tissue and semen go down when blood levels go down, but not at the same time or the same rate.

The viral load test, also known as the polymerase chain reaction test, is a technology devised to detect very small genetic fragments. It cannot judge whether these fragments are infectious.

The viral load tests currently available all have lower limits of detection which vary between 500 and 5 copies per millilitre.

Log changes in viral load

Throughout this chapter reference will be made to `log' changes in viral load. The log scale is the measurement used by researchers to describe the changes in viral load in an infected individual, and in this case refers to changes to the power of 10. The scale is used because the magnitude of changes in viral load during antiretroviral therapy is immense, and can only be captured on graphs or diagrams by using a logarithmic scale.

A 1 log change refers to a tenfold change, a 2 log change to a hundredfold change and a 3 log change to a thousand-fold change.

Viral load below 10,000 copies is considered to be low viral load, whilst viral load in the range of 10,000 to 50,000 copies is considered to be moderately high, and above 50,000 copies is considered to be high.

Blood

HIV has been isolated from blood at levels of up to 5,000 infectious particles per ml, but some researchers have found it much more difficult to isolate HIV from the blood of HIV–positive individuals.

Although levels of free infectious virus may be very low in the blood of asymptomatic individuals (below 200 HIV RNA copies per ml), asymptomatic individuals may have viral load above 1,000,000 HIV RNA copies per ml.

Antiretroviral therapy using a combination of drugs usually reduces viral load by at least 90% within a few weeks of initiating treatment. Blood from individuals receiving antiretroviral therapy may be less infectious, although many cases have been reported of viral burden rebounding to extremely high levels when resistance to antiretroviral drugs emerges, or when treatment is discontinued.

Semen

Between 10 and 30% of seminalfluid samples studied have been shown to contain either free infectious virus or cell–associated virus (infected lymphocytes). However, levels of cell–associated virus are much higher; there are up to one million lymphocytes in each millilitre of semen; even if only one per cent of these lymphocytes are infected with HIV, this means that each millilitre of semen may contain up to 1,000 infected lymphocytes. The average male ejaculation is 5ml of semen, although this varies considerably.

Lymphocytes and cell-associated virus are most likely to be present in individuals suffering from infections which cause inflammation in the urethra and other regions of the genital tract. Untreated sexually transmitted infections, even those which cause no detectable symptoms, are likely to increase the viral load in semen.

The relationship between viral load in blood, immune status and viral load in semen is full of uncertainties. Recent studies have produced very contradictory results.

Viral load in semen is usually greater in those with lower CD4 counts. Several studies have shown that HIV is more frequently isolated from the semen of men with CD4 counts below 200, especially if they are symptomatic. A 1994 study found HIV in the semen of 87% of men with CD4 count below 100, but in only 40% of men with CD4 counts between 100 and 200. Only in men with CD4 counts below 100 was it possible to find more virulent forms of HIV (the SI phenotype), which are believed to be more easily transmitted during vaginal intercourse (Vernazza). A 1996 study found that HIV RNA levels were at least tenfold (1 log) higher in men with CD4 counts below 200 compared to men with CD4 counts above 500 (Dyer).

A 1991 study found that whilst HIV was most commonly isolated from the semen of men with CD4 counts below 200, it could also be isolated from the semen of 43% of a group of asymptomatic men on at least one occasion out of six monthly tests, suggesting fluctuating levels of HIV in the semen of infected men.

Although most studies show that the majority of men treated with antiretroviral drugs demonstrated parallel declines in plasma and seminal viral load, all studies have shown considerable individual variation in responses.

The following patterns have been observed:

• Viral load becomes undetectable in plasma weeks or months before doing so in semen.
• Viral load becomes undetectable in semen but not in plasma.
• Plasma viral load rebounds after a period of undetectability but seminal viral load remains undetectable.

Studies of gay men (Taylor) suggest that at any one time about 12% of HIV-positive men on ARV treatment may have detectable HIV in their semen even if they do not have it in their blood; this has been linked to the presence of (often asymptomatic) urethritis.

Seminal and plasma viral loads were obtained from 72 HIV-positive men who were not taking HIV therapy in a research project involving investigators from London and Birmingham. Earlier research from these investigators established that in men taking HAART, a detectable seminal viral load was associated with untreated gonorrhoea.

Although none of the men in the study had a blood plasma viral load below the limit of detection (400 copies/ml), 22 men (30% of the sample) did have undetectable seminal viral loads. These men were classified as “non-shedders. ” Low levels of HIV, on average 4,000 copies/ml, were detected in the semen of 58% of men, who were classified as “seminal shedders. ” Their blood plasma viral load was significantly higher than non-shedders at 100,000 copies/ml versus 10,000 copies/ml.

12% (n=9) of men had extremely high seminal levels of HIV, on average 398,000 copies/ml, and these nine men were classified as “seminal super- shedders” and were considered particularly likely to pass on HIV during unprotected sex.

Blood plasma viral loads, CD4 cell counts and stage of HIV disease did not differ significantly between super-shedders and seminal shedders. However, the investigators did find that super-shedders were generally older, with an average age of 48 years compared to 35 for the sample as a whole, and were more likely to have an STI (33% of super-shedders compared to 3% of total sample).

The investigators conclude that men shedding viral load in their semen were likely to have higher blood plasma viral loads than non-shedders. In addition, a small number of men are super-shedders, and although they have comparable blood levels of HIV to normal shedders, super-shedders tend to be older and have an STI which may assist local reproduction of HIV in the genital tract.

A comprehensive discussion of research findings in this area can be found in the “HIV & AIDS Treatments Directory”, or at aidsmap.com. See Anti-HIV Therapy: Choosing your treatment strategy.

Vaginal and cervical secretions

Levels of HIV in the vagina increase at the time of menstruation because of the presence of blood and other cell–associated virus. Levels of macrophages which could harbour HIV also increase in the vaginal fluid at the time of menstruation. It is believed that sexually transmitted infections and inflammation in the vagina will boost the level of lymphocytes. The hormone progesterone, a component of some oral contraceptives, has also been associated with increased HIV RNA levels (Mostad).

Levels of HIV RNA in plasma and vaginal fluid have been shown to correlate very closely, but nearly 35% of women tested in one study had no detectable HIV RNA in their vaginal fluids despite detectable HIV in their plasma once the single occasion sampling took place (Hart).

HIV RNA levels in vaginal fluid have been shown to decline after the initiation of antiretroviral therapy. Antiretroviral therapy was associated with a median 1.4 log decline (approx 95%) in viral burden in vaginal fluid after three weeks of therapy. Antiretroviral therapy consisted of one or two nucleoside analogues (Lennox).

Rectal secretions

A 2004 study of gay men found that levels of HIV RNA in rectal mucosal secretions were higher than those in blood or semen -- by about 500 percent in the case of blood and 2,500 percent in the case of semen.

The study found that viral loads in swabs taken from the rectum were 25 times higher than semen viral loads in the 58 percent of the men not taking HIV treatment, and four times higher in the 37 percent of men on treatment who had detectable viral loads.

The average rectal viral loads were 316,000 and 4,000 respectively.

There was also a much less strong relationship between the use of treatment and reduction of viral load in the rectal swabs than there was with blood or semen.

During anal intercourse, the rectum secretes a protective, lubricating mucus to accommodate penetration. Although the mucus eases penetration and repeated thrusting, it by no means protects either partner from cuts and abrasions that can lead to the passage of HIV infection, which is why proper lubrication and condom use is necessary. In fact, the mucus may be far more dangerous than any other fluids either partner might be exposed to.

The study suggests that gay or bisexual men who have insertive unprotected anal sex and encounter this mucus are at far higher risk of HIV infection than previously thought. Formerly, most literature on HIV protection named blood or semen as the main culprit in HIV transmission.

Brain tissue and cerebrospinal fluid

HIV was cultured from the cerebrospinal fluid of 13 out of 14 people suffering from AIDS–associated neurological disorders, in much higher quantities than any other body fluid apart from blood (Levy). Another study was able to isolate HIV from brain tissue and cerebrospinal fluid in 24 out of 33 people suffering from AIDS–associated neurological disorders (Ho).

Bronchoalveolar fluid

Faeces

Saliva

Tears

Researchers have found it difficult to isolate HIV from the tears of HIV–positive people. In one experiment it was possible to find HIV in the tears of only one person with AIDS out of seven tested, and impossible in five HIV–positive people with no symptoms (Fujikawa). Another experiment using highly sensitive techniques looked at 50 HIV–positive people and failed to find HIV in the tears of any of them (Mueller).

Urine