The amber light: the World Health Organization’s position when an HIV viral load is “suppressed but not undetectable”

Dr Lara Vojnov at IAS 2023.
Dr Lara Vojnov at IAS 2023. Photo by Roger Pebody.

The risk of transmission by someone who has a viral load too low to be quantified by some viral load tests, but where the test still detects HIV’s presence, is “almost zero or negligible” in the words of the World Health Organization (WHO), the 12th International AIDS Society Conference on HIV Science (IAS 2023) in Brisbane, Australia heard yesterday.

Dr Lara Vojnov, Diagnostics Advisor in WHO’s Global HIV, Hepatitis and STI Programme, was launching the WHO’s new policy brief, The role of HIV viral suppression in improving individual health and reducing transmission. In simple language, it explains viral load measurements and their relevance to transmission for an audience of “people living with HIV, providers, laboratory staff, programme managers, global and national policymakers, and clinical and diagnostic partners.”

At IAS 2023, Dr Lara Vojnov talks to NAM aidsmap's Roger Pebody about WHO's statement on undetectable viral load.

The brief distinguishes between three viral load test results: “Unsuppressed” means a viral load of at least 1000 copies/ml. “Undetectable”, in the conventional lab tests used in clinics, mainly in higher-income settings, means a viral load below 200 copies/ml  – or less, depending on the sensitivity of the particular test. Anything over 200, 20 or whatever the test's limit of detection is has traditionally been called 'detectable'.

'Suppressed’ is the hinterland between those two definitions, whose significance WHO now feel they need to clarify. This is because  some tests now being used in the field can detect the presence of HIV at viral loads below 1000, but sometimes at such low levels that they cannot come up with a specific figure for the viral load.

If someone’s viral load test places them in this “suppressed” category, it means their risk of transmission is “almost zero or negligible”. This phrase was chosen carefully to align with the “zero risk” already assigned to transmission from people with viral load results below 200, which was the end result of the zero infections observed in the PARTNER 1 and 2 and Opposites Attract studies and formed the scientific underpinning of the Undetectable = Untransmittable (U=U) campaign.   

So why now add this ‘amber’ category of ‘suppressed but detectable’ between the green light of ‘undetectable’ and the red light of ‘unsuppressed’, as WHO does in its new policy brief?

This is because of the urgent need to introduce viral load testing as a standard part of HIV treatment and care in all income settings. Years ago, viral load tests were seen as a technology too complex and expensive for lower income settings. Clinical monitoring – basing decisions about changing treatment on symptoms – was seen as being as effective as CD4 or viral load monitoring and more cost effective.

However, it soon became clear that delaying regimen change until people became ill imperilled future treatment options, as it caused widespread drug resistance, which has severely limited second-line treatment options and has cost lives.

Now, long-awaited viral load tests have been developed that do not have to be processed in a laboratory (they can be used at the ‘point of care’) and give results within minutes. Many countries test a drop of dried blood on a card, with obvious benefits in terms of where a test can be done – the card can be stored at room temperature and posted. Both point-of-care and dried blood spot tests have been evaluated and found to be appropriately sensitive (detecting real cases of detectable viral load) and specific (not detecting false cases).


Undetectable = Untransmittable (U=U)

U=U stands for Undetectable = Untransmittable. It means that when a person living with HIV is on regular treatment that lowers the amount of virus in their body to undetectable levels, there is zero risk of passing on HIV to their partners. The low level of virus is described as an undetectable viral load. 

detectable viral load

When viral load is detectable, this indicates that HIV is replicating in the body. If the person is taking HIV treatment but their viral load is detectable, the treatment is not working properly. There may still be a risk of HIV transmission to sexual partners.

second-line treatment

The second preferred therapy for a particular condition, used after first-line treatment fails or if a person cannot tolerate first-line drugs.

treatment failure

Inability of a medical therapy to achieve the desired results. 

case-control study

An observational study in which a group of people with an infection or condition (called ‘cases’) are compared with a group of people without the infection or condition (called ‘controls’). The past events and behaviour of the two groups are compared. Case-control studies can help us understand the risk factors for having an infection or a condition. However, it is difficult both to accurately collect information about past events and to eliminate bias from case-control studies.

However, Lara Vojnov emphasised, the clinical threshold in these tests for a definitely unsuppressed viral load is in the region of 1000. This is not due to any technological limit for detecting lower viral loads, but because the specimens used are smaller, and this leads to results, in the cases of lower viral loads, that are positive but not quantifiable – in other words, the test does not give a precise numerical value.

A dried blood spot test can give an 'undetectable' result too. This means it's failed to find any virus and there is zero risk of transmission. But we can't attach a definite figure such as 'below 50' to an undetectable result in a dried blood spot test - just that it didn't find any.

As regards the implications for treatment and for the individual, having a viral load in this range is quite uncommon: one study presented at the 2020 Conference on Retroviruses and Opportunistic Infections found that less than 5% of people with a viral load under 1000 had one over 200. A viral load in the high hundreds is often transient, being 'on the way down’ (as in people new to therapy) or 'on the way up’ (as in treatment failure, or in adherence problems that could lead to failure).

So the new category can be a warning not of a present hazard, but of a future one, just like an amber traffic light. It allows for a specific procedure for people who present as 'suppressed but detectable'. After the first such test result, WHO recommends that people should be given enhanced adherence counselling, and that a viral load test should be repeated in three months. If the result is still suppressed but detectable, then the regimen should be changed as this may indicate low-level resistance or imminent treatment failure.

(This, by the way, is no different from what the guidelines recommend for people who have an unsuppressed viral load of over 1000, unless they are on a regimen based on non-nucleoside drugs like efavirenz or rilpivirine, where resistance can develop fast and immediate switching is recommended.)

What of the implications for transmission and prevention? The problem here is that the limit set by the PARTNER studies to define ‘zero risk’ was a viral load below 200. So WHO needed to quantify the risk of transmission from someone with a viral load that was known to be under 1000 but might be above 200.

Lara Vojnov and colleagues undertook a review of all studies they could find that measured viral load and related it to transmission events. This review was published in The Lancet on Saturday.

They found that, from as long ago as the year 2000, there have been eight such studies, including treatment-as-prevention studies like PARTNER 1 and 2 and HPTN 052, PrEP studies like Partners PrEP, and smaller epidemiological and treatment studies. In total these studies enrolled 7762 serodifferent (one positive, one negative) couples.

Some studies like the PARTNER and Opposites Attract studies did not shed much light, because to be enrolled in them the positive partner already had to have a viral load below 200.

Discounting these, and one case-control study where timing of infection could not be measured, left 4773 couples, with 323 HIV transmissions recorded. There were just two transmissions where the last viral load in the HIV-positive partner before transmission was below 1000: one from someone in HPTN 052 with a viral load of 617 and one from someone in Partners PrEP with a viral load of 872. 

However, in both cases the viral load test was taken more than 50 days before transmission occurred. In the first case the viral load could easily have been ‘on its way up’ to a higher point at transmission. The second case is not so easily dismissed, though: this person had a viral load in the 700-800 region even before they started ART and they may have had a degree of immune control over their virus – which their partner might not have had (this was the case with the ‘Esperanza patient’, for instance).

So we can say that on the basis of these studies (which are unlikely to be repeated, as viral load is better controlled and transmission between couples rarer these days) that transmission from a person with a viral load between 600 and 1000 was very rare, forming a maximum of 0.6% of observed transmissions; none were seen when the viral load was below 600. Transmission is still uncommon even in viral loads of a few thousand: in two studies conducted in the 1990s in Uganda and Zambia, respectively 81% and 92% of transmissions occurred where the HIV-positive partner had a viral load over 10,000.

The Lancet report did note that these data do not apply in two areas. One is in vertical transmission, where we cannot with the same assurance say that “suppressed equals near-zero risk” because there are multiple modes of transmission from mother to baby – through blood and through breast milk. Here more conventional sensitive viral load testing in mothers recently starting ART and prophylaxis for the baby are still recommended.

And WHO acknowledges that there is a complete lack of evidence when it comes to the relationship between viral load and transmission through needles or injections. This is a big research gap as it leaves people who inject drugs uncertain about transmission or infection risks. It also fails to finally settle the question of risk to medical staff and others performing exposure-prone procedures.

That’s the science on which the WHO’s new categorisation is based. Lara Vojnov emphasised that the original message of U=U remains unchanged, which is that, “People living with HIV who have an undetectable viral load have zero risk of transmitting HIV to their sexual partners.” (This earned a round of applause.)

The ‘suppressed’ category meant, she said, that, “there is some virus replicating and present, but too little to be quantified. It could be due to missing doses, recent treatment initiation or drug resistance,” and she repeated that “people living with HIV who have a suppressed viral load have almost zero or negligible risk of transmission to their sexual partners”.

“We should be expanding viral load access to all people living with HIV using all of the tools that we have, so they can be aware of their status,” she added.


World Health Organization The role of HIV viral suppression in improving individual health and reducing transmission. WHO policy brief, 22 July 2023.

Broyles LN et al. The risk of sexual transmission of HIV in individuals with low-level HIV viraemia: a systematic review. The Lancet, 23 July 2023. 

What's new in WHO guidelines: innovations, treatment, integration and monitoring. 12th IAS Conference on HIV Science, Brisbane, satellite SAT026, 2023.

View the details of the session on the conference website.

Update: this article was amended on 26 and 29 July 2023 to clarify some details of the definition of undetectable.