Paradox: people with moderate viral load more likely to pass on HIV

This article is more than 16 years old. Click here for more recent articles on this topic

HIV may have evolved so that the average viral load set point – around 33,000 copies/ml – seen in most untreated people during chronic infection is finely balanced between being the optimal for HIV transmission and the optimal for host survival according to a study published online this week in the journal, Proceedings of the National Academy of Sciences.

The findings may have important policy implications for ‘imperfect’ HIV prevention technologies, such as vaccines that may reduce an individual's viral load so they can live longer without the need for treatment, but which then increases that individual's transmission potential over a lifetime.

When are people with HIV most likely to sexually transmit the virus?

There is currently some disagreement regarding the impact of untreated primary (also called acute) HIV infection versus untreated chronic HIV infection on infectiousness and sexual transmission, and their relative contributions to onward transmission.

Although studies have found that viral load during untreated primary HIV infection is much higher – and therefore individuals are more infectious – than during untreated chronic infection, it is also understood that since untreated primary HIV infection only lasts a matter of weeks or months, the opportunity for transmission is lower compared with untreated chronic HIV infection, which can last for years or even decades.


primary infection

In HIV, usually defined as the first six months of infection.

mathematical models

A range of complex mathematical techniques which aim to simulate a sequence of likely future events, in order to estimate the impact of a health intervention or the spread of an infection.

set point

The viral load that the body settles at within a few weeks to months after infection with HIV. Immediately after infection, a person’s viral load is typically very high. After a few weeks to months, this rapid replication of HIV declines and the person's viral load drops to its set point. A higher viral set point suggests that, in the absence of treatment, disease will progress faster than in a person with a lower set point. 


In HIV, different strains which can be grouped according to their genes. HIV-1 is classified into three ‘groups,’ M, N, and O. Most HIV-1 is in group M which is further divided into subtypes, A, B, C and D etc. Subtype B is most common in Europe and North America, whilst A, C and D are most important worldwide.


The power of bacteria or viruses to cause a disease. Different strains of the same micro-organism can vary in virulence.


In fact, a recent US study utilised mathematical modelling to estimate that fewer than 9% of all new sexually transmitted HIV infections originated in people with untreated primary HIV infection, compared with 48% of new infections resulting from sexual contact with people with untreated chronic HIV infection.

However, another recent study from Canada used very different (and arguably more robust) methods – phylogenetic analysis to track the impact of actual primary infections on sexual transmission. Here, almost half of all sexually transmitted HIV infections were found to be due to primary infection.

Adding to the debate is this study by researchers from Imperial College, London, who again utilised mathematical modelling to estimate the impact of untreated HIV infection on transmission.

Quantifying the impact of the relationship between viral load and infectiousness

The main aim of the study was to quantify the relationship between viral load and infectiousness and to estimate its epidemiological impact. The investigators did this by re-examining existing data on viral load during the natural history of HIV infection (from recently-infected gay men in Amsterdam) and data on viral load and transmission (from heterosexual men and women in Zambia).

The first part of this mathematical model involved calculating treatment-free survival based on an individual’s viral load set point (i.e. the average viral load over time that is established following primary infection). To do this they used data from the Amsterdam Seroconverters Cohort, which followed the natural history of HIV infection in 123 men (and 504 person-years) between 1982 and 1993.

They estimated that an untreated individual with an average viral load of 1,000 copies/ml would have 15.6 years of symptom-free life. As the viral load increased, the length of asymptomatic HIV infection reduced: 9.7 years for 10,000 copies/ml; 4.9 years for 100,000 copies/ml, and 2.1 years for a million copies/ml.

They then used data from a Zambian study that took place between 1994 and 1998 and which included 983 serodiscordant couples. Even with condoms and safer sex counselling, 107 HIV infections occurred during the course of the study.

From these data, the researchers were able to estimate the annualised transmission rate (within a stable long-term discordant partnership). This ranged from a rate of 0.02 per year for someone with a set-point viral load of 1,000 copies/ml; to 0.132 per year for 10,000 copies/ml; 0.279 per year for 100,000 copies/ml; and 0.313 per year for one million copies/ml.

They compared these estimates with data on HIV transmission from Rakai, Uganda, which they found to be consistent with their estimates. However, they note, “both these studies involved a degree of counselling to reduce unprotected sex, so transmission rates within uncounselled partnerships could be somewhat higher, although the dependence on viral load would be similar.”

Estimating the transmission potential

What, then, is the impact of HIV viral load and longevity on the epidemiology of sexual HIV transmission? The investigators calculated this by estimating the expected number of people someone is likely to infect over their entire lifespan. Of course, as they note, “many factors can affect this, including host behaviour, coinfections, and the state of epidemic itself, because opportunities for transmission are reduced when prevalence is already high.”

Nevertheless, they define the ‘transmission potential’, as follows: “The average number of people potentially infected over the duration of the whole infectious period, in circumstances where most people are uninfected, for an infected individual with a particular viral load; where the rate of partner change is sufficiently high that it does not limit transmission, and where the transmission rate within partnerships is similar to that reported by the two cohorts studied here.”

They found that the periods of highest viral load (during primary infection and during late-stage HIV disease) did not actually have the highest transmission potential: 0.67 (infections per person per lifespan) for primary infection and 0.50 for late-stage HIV disease.

In fact, the viral load with the highest transmission potential (of close to 1.5 infections per person per lifespan) was found to be during chronic infection: 33,113 (4.52 log10) copies/ml. This, the investigators point out , is very close to the average viral load set point seen in both the Dutch (22,908 or 4.36 log10 copies/ml) and Zambian (54,954 or 4.74 log10 copies/ml) cohorts.

Policy implications for public health

Given this analysis – that medium levels of HIV provide the greatest opportunity for onward transmission – they warn against utilising “imperfect” prevention technologies – such as vaccines, immunotherapy, or microbicides – which may result in ‘perverse outcomes’. “If the intervention reduces patients’ viral load in such a way as to increase their transmission potential on average,” they write, “then incidence will increase, not decrease. An intervention that reduces viral loads from high to intermediate levels and is therefore beneficial to the individual may nevertheless increase overall incidence and thus cause more overall harm than benefit.”

Their analysis also supports the findings of a 2006 Canadian modelling study which suggests that treating all diagnosed people with antiretrovirals in order to reduce infectiousness would have a major impact on the HIV epidemic .

However, they point to a “significant limitation” of their analysis: they combined viral load survival data from a cohort infected with subtype B virus with transmission data from a cohort infected with non-B subtypes. Their calculations assumed that the relationship between the duration of asymptomatic infection and set-point viral load is similar in different settings, but they are aware of studies which found rapid disease progression in individuals infected with subtype D that was not explained by higher viral loads.

Has HIV evolved for optimal transmission potential?

The investigators also present a hypothesis regarding the evolution of HIV’s virulence. “Seen from the perspective of the virus,” they argue, “a negative correlation between infectiousness and duration of infection could be interpreted as a trade-off between two viral life-history traits, with natural selection leading to an optimal balance in this trade-off."

They wonder if this is “not coincidence but, rather, an outcome of natural selection acting on HIV-1 to maximize opportunities for onwards transmission?”

Lead author Dr Christophe Fraser says that further studies are necessary to prove their hypothesis that it is evolution, and not coincidence, that the average viral load seen in untreated people is finely balanced between optimal HIV transmission and optimal host survival.

“We now want to see whether the virus has adapted in order to allow it to infect the most people, which seems plausible given the results of our study. This would have serious implications for public health policy, because if it is true then some strategies to prevent transmission could end up making the virus more virulent by accident. While it is too early to sound the alarm, more research to prove or disprove this theory is urgently needed. That is what we are focusing on now.”