Antiretroviral treatment can reduce the viral load in the blood and other bodily fluids to undetectable levels and research shows this can decrease the risk of HIV transmission. This knowledge has generated interest among people living with HIV and their partners in the use of treatment as an HIV prevention strategy. But how much can treatment reduce the risk of HIV transmission and what is the transmission risk when the viral load is undetectable?
This article explores what we know and don't know about these questions and provides key messages for both people living with HIV, and their partners.
Viral Load, Treatment and HIV Transmission
Viral load refers to the amount of HIV in a bodily fluid. Blood viral load is monitored as part of routine clinical care for a person living with HIV to track HIV disease progression, inform decisions on treatment initiation, and determine whether antiretroviral treatment is working once started. Treatment is generally considered successful when the viral load is reduced and, ideally, lowered to undetectable levels.
While we have known for years that a lower blood viral load can significantly improve the health of people living with HIV, it is only more recently that we have learned about its importance for HIV transmission and prevention. As it turns out, the amount of HIV in a fluid is an important predictor of whether infection occurs after an exposure. Research shows that a lower blood viral load is generally associated with a reduced risk of sexual HIV transmission. For each 10-fold decrease in viral load (for example, from 100,000 to 10,000 copies/ml), the risk of sexual HIV transmission generally reduces by 2-3 fold.1 This is because when the viral load in the blood decreases, it generally also decreases in the sexual fluids (semen, vaginal fluid and rectal fluid),2 which are those commonly involved in the sexual transmission of HIV.
In 2011, a landmark study known as HPTN 052 demonstrated -- among heterosexual serodiscordant couples -- that early treatment can significantly reduce the risk of sexual HIV transmission under certain conditions.3 While uncertainties and information gaps remain, interest in the use of treatment as an HIV prevention strategy is growing, and research shows that some people living with HIV consider their viral load when making decisions around safer sex and condom use.4,5 As a result, frontline service providers need to be prepared to answer potentially difficult questions to help empower clients to make informed decisions.
No Simple Answers
Clients may have many questions when it comes to viral load, treatment and HIV transmission. Two commonly asked questions are: 1) how much can treatment reduce the risk of HIV transmission and 2) what is the risk of HIV transmission when the viral load is undetectable and no condom is used?
While these questions may seem similar and straightforward, they are actually two very different questions and, unfortunately, there are no simple or conclusive answers to either of them. While this uncertainty can be frustrating and makes it difficult to provide meaningful answers, it is important for frontline service providers to communicate to clients what the research does and does not tell us. Below we explore what we know and don't know for each question.
1. How Much Can Treatment Reduce the Risk of HIV Transmission?
The HPTN 052 study helps answer this question. In this randomized controlled trial, treatment reduced the risk of HIV transmission by 96% (equivalent to a 25-fold decrease) among heterosexual serodiscordant couples who had mostly vaginal sex and received ongoing services and supports including adherence and prevention counselling, free condoms, viral load tests, and testing and treatment of sexually transmitted infections (STIs).3 Although this study improved our understanding, knowledge gaps remain that require further investigation and need to be explored with clients when answering this question.
How does this transfer to the "real world?"
The HPTN 052 study took place in a very controlled setting where a motivated group of participants were provided with ongoing supports and services to help them reduce their risk of HIV transmission and maximize the prevention benefits of treatment. As a result, it is unclear if the significant reduction in risk observed in this study will also apply to heterosexual couples in the "real world" -- outside of a clinical trial setting -- who may not receive, or have access to, the same supports and services as in HPTN 052.
We know that adherence counselling is important to support daily pill-taking in order to reduce the viral load to undetectable levels; regular viral load testing is important to make sure that treatment is working (viral load is undetectable) and HIV has not developed resistance to treatment medications (if the viral load is no longer undetectable, this may indicate that resistance has developed); and STI testing and treatment are important because untreated STIs (in either the HIV-positive or HIV-negative partner) may increase the risk of HIV transmission even when on treatment.6
If medications are missed, drug resistance develops, or untreated STIs are present, treatment may be much less effective than 96% at reducing the risk of HIV transmission. For example, in a "real world" study of heterosexual serodiscordant couples who were not part of a tightly controlled randomized trial like HPTN 052, antiretroviral treatment only reduced the risk of HIV transmission by 26% (equivalent to a 1.35-fold decrease).7
What about anal sex?
Since 97% of the couples in the HPTN 052 study were heterosexual and reported mostly having vaginal sex, we don't know how much these results apply to couples who mostly have anal sex, such as some gay men and other men who have sex with men (MSM). However, the World Health Organization (WHO) recently held a meeting on the subject and concluded that there is no reason to think that treatment won't also reduce the risk of HIV transmission through anal sex, although the reduction in risk may or may not be as high as for vaginal sex.8 There are ongoing studies, such as the Opposites Attract study in Australia, which aim to determine the extent of risk-reduction among gay men and other MSM.
2. What Is the Risk of HIV Transmission When the Viral Load Is Undetectable and No Condom Is Used?
Unfortunately the HPTN 052 study did not help answer this question. While HPTN 052 showed that being on treatment and having a lower viral load can dramatically reduce the risk of HIV transmission under certain conditions (see Question 1), it remains unclear what the actual risk is reduced to when the viral load is undetectable.
A major reason for this uncertainty is the lack of research among couples who mostly have sex without a condom.9 A review of studies in the literature, performed in November 2012, did not find any reported HIV transmissions between heterosexual serodiscordant couples where the HIV-positive partner had an undetectable viral load (no studies of same sex serodiscordant couples were identified by the review).10 However, the lack of HIV transmissions does not mean the risk is zero, as most of the couples in these studies reported using condoms often. For example, 96% of couples in the HPTN 052 study reported using condoms every time they had sex. While participants of studies often say they are using condoms more regularly than they actually are, condom use may partly explain the lack of HIV transmissions in the HPTN 052 study (when the HIV-positive partner had an undetectable viral load) and other studies included in the literature review.
Also, the risk of HIV transmission when undetectable may not be the same for all types of sex. For example, the risk may be higher for anal sex than for vaginal sex, particularly if the HIV-negative partner is the receptive partner (bottom) during anal sex (also known as receptive anal sex). This is because receptive anal sex has a higher baseline risk than other types of sex. Research shows that, on average, the risk of HIV transmission to an HIV-negative partner can be 10-20-fold higher through receptive anal sex than it is through vaginal sex.11,12 However, this research did not measure the viral load of the HIV-positive partner. Therefore, we don't know if the risk is higher through anal sex when the viral load is known to be undetectable, as this has yet to be studied.
Fortunately, there are ongoing studies, such as the Partner study in Europe, which will provide a better understanding of the risk of HIV transmission when the viral load is undetectable and no condom is used, for both vaginal and anal sex. These studies are enrolling serodiscordant heterosexual and same-sex couples who are taking HIV treatment, have an undetectable viral load, and do not always use condoms.
Coming to a consensus
While we wait for research gaps to be filled, experts and community organizations are using the information currently available to develop an answer to this question.
For example, experts at the British HIV Association (BHIVA) recently released a position statement concluding the risk of HIV transmission through vaginal sex when no condom is used is "extremely low" when the blood viral load is undetectable and the following conditions are met: 1) There are no sexually transmitted infections (STIs) in either partner; 2) the HIV-positive partner has maintained an undetectable viral load for at least 6 months; and 3) the blood viral load is monitored on a regular basis. The statement went on to say that -- despite the lack of research -- they anticipate a similarly "extremely low" risk for anal sex. However, other experts think the risk may be higher for receptive anal sex than for vaginal sex.13
Some community organizations have developed tools to try to translate the research and expert statements into messages for community members. One example is ACON (the AIDS Council of New South Wales), the largest community-based gay, lesbian, bisexual and transgender HIV/AIDS organization in Australia, which recently developed their "Know the risk" website. This site states there is a "medium risk" through anal sex when the viral load is undetectable and it is known (or there is uncertainty) that an STI is present. However, the risk can become "low" when it is confirmed there is no STI and the viral load has been undetectable for six months.
It is important to keep in mind that while expressions such as "extremely low" and "medium" risk are relatively easy to communicate to clients and may reflect risk in a more meaningful way than the use of numbers, they are also open to interpretation. For example, "extremely low" risk may mean different things to different people.
Not "no risk"
There is a general consensus that the risk is not completely eliminated when the viral load is undetectable and no condom is used. In fact, there has been one published case report of HIV transmission between a same sex male serodiscordant couple where the HIV-positive partner was believed to have an undetectable viral load.14
How is HIV transmission still possible when the viral load is undetectable?
Many people who have an undetectable viral load in the blood also have an undetectable viral load in other bodily fluids. However, undetectable does not mean that there is no virus, only that the amount of virus is below the limits that tests can detect (viral load tests used in Canada cannot detect HIV in the blood if there are less than 40-50 copies/ml). Therefore, HIV transmission may still be possible because virus is present.
Also, it is possible for people who have an undetectable viral load in the blood to sometimes have detectable (although lowered) levels of virus in their other bodily fluids.16-19 A higher level of HIV in semen, vaginal fluid and rectal fluid may increase the risk of transmission when the blood viral load is undetectable. However, it is unclear how often this happens and how significant it is in terms of HIV transmission.
Some biological factors known to increase viral load in genital and rectal fluids, and potentially the risk of HIV transmission, include inflammation at the penis, vagina or rectum caused by tearing, STIs and some vaginal conditions (such as bacterial vaginosis), and hormonal changes caused by the menstrual cycle, pregnancy or use of injectable hormonal contraceptives.20-24