Experts publish recommendations on managing low-level but detectable HIV

An international panel of scientists reviewed the evidence to construct a framework of recommendations on the management of low-level but detectable HIV. The guidance, published in The Lancet HIV by Dr Tommaso Clemente and colleagues, brings together evidence from a scoping review and expert judgement in order to create a shared point of reference for clinicians.

While some other guidelines map out some of the scenarios and how to act on them, a thorough approach to low-level but detectable viral load has not been available so far. Clinicians have therefore lacked a clear roadmap for distinguishing between its different possible causes and for managing the more challenging cases. This persistent low but detectable viral load is often called low-level viraemia.

The recommendations include repeat viral load testing as well as reviewing adherence and drug interactions. Drug resistance testing is recommended especially at viral loads above 200 copies, as well as CD4-to-CD8 ratio assessment if low-level viraemia has lasted at least three to six months. They also agreed on considering viral load in the central nervous system if neurological symptoms are present. Closer viral load follow-up every three to four months is also advised if treatment is continued with no change, and HIV prevention measures should be considered if the viral load is between 200 and 1000 copies.

Background on low-level viraemia

The team defined low-level viraemia as multiple viral load measurements between 50 and 1000 copies, but also made recommendations for a viral load between 50 and 200 copies separately. Generally, a viral load above 1000 copies is used as the cut-off for treatment failure, although in most resource-rich settings readings of above 200 may already be seen as treatment failure.

The authors note that low-level viraemia may be seen in between 3% and 26% of people with HIV, depending on the definitions and the settings. The research so far has been inconsistent on the effects and risk of low-level viraemia on health outcomes as well as on the risk of treatment failure. However, it is often concluded that a viral load between 200 and 1000 copies increases the risk of resistance and potential treatment failure, while the conclusions about other health outcomes are less clear.

The causes of low-level viraemia are even less clear, but there are certain factors that are associated with it such as low treatment adherence, being on a therapy with a lower genetic barrier to resistance, and starting treatment at very high viral load or very low CD4 counts.

Integrase inhibitors such as dolutegravir or bictegravir may be protective against low-level viraemia showing the value of a higher barrier to resistance. To explain, the barrier to resistance is a characteristic of drugs and reflects how hard and biologically costly it is for the disease-causing microorganism to become resistant to the drug. A drug with a high genetic barrier to resistance means the microorganism will have a hard time developing resistance and when it does, it may lead to the microorganism losing its fitness or efficiency at infecting cells.

The review and guidance

The authors gathered existing evidence and sought consensus with an international panel of experts to strengthen the overall reliability of their recommendations. Their review focused on adults taking HIV treatment who had persistent low-level viraemia. Studies published between 2008 and 2025 were included if they provided clinical data on aspects such as the factors that predict low-level viraemia, the emergence of drug resistance, and compartmentalisation (ongoing viral replication in a reservoir site such as the brain).

There were 84 studies that were included for the analysis. The researchers then used the evidence from the review for the international consensus process by surveying experts on statements tied to the review findings. Sixty-three international experts in the field of HIV reached consensus on 18 statements (over 70% of the experts agreed).

The consensus part was crucial, because low-level viraemia is still a ‘grey area’ in HIV management and research findings have been uneven. Among some of the reasons the research has not been conclusive so far is that different studies use different criteria such as varying cut-off values for low-level viraemia, different definitions of virological failure or different follow-up periods and interventions.

The expert panel considered low-level viraemia in two different ranges – 50 to 200 copies and 200 to 1000 copies. They also had to consider whether this viraemia is happening while on drugs with a high barrier to resistance or not. The high resistance barrier drugs were usually considered to be the boosted protease inhibitors such as darunavir (in Symtuza) and the two integrase inhibitors – bictegravir and dolutegravir (found in Biktarvy and Dovato or Juluca respectively).

Definitions of low-level viraemia

Most studies included in the review used 50 copies as lower cut-off for low-level viraemia, while the upper boundary varied. Some studies used 200 copies as the upper boundary, while others used 400, 500 or 1000 copies. The reason this matters is because a viral load that would be considered a virological failure in one healthcare setting may still be treated as low-level viraemia in another. 

Differing causes of low-level viraemia

In some people with HIV, low-level viraemia may reflect actual viral replication, meaning that new cells are being infected despite treatment. This is the type of low-level viraemia that might be the most concerning, because it can allow the virus to develop drug resistance.

However, in other cases, the virus may come from long-lived cells that release viruses that are often defective and unable to successfully complete their infectious lifecycle. These cells are often referred to as the reservoir. In this situation the viraemia may be less concerning, but it may still trigger low-grade inflammation. Also, it doesn’t seem to respond to treatment: because these are defective viral particles that cannot infect new cells and are coming from already infected cells, the drugs cannot block them.

This is because antiretroviral drugs are designed to block steps in the lifecycle of the virus, but with viruses that are pumped out by already infected cells, there aren’t many steps left to block. However, in current clinical practice, there is no way of differentiating productive viraemia, where new rounds of cell infections happen, from reservoir-driven viraemia.

A related issue is the so-called compartmentalisation of the virus. Low-level viraemia in the blood may be caused by active viral replication in a different body compartment (site) – particularly the central nervous system (CNS). There was limited evidence available for the review, due to the anatomical difficulty of measuring viral loads in the CNS, which can only be done via spinal tap (drawing fluid from the spine). However, the researchers note that people experiencing low-level viraemia with unexplained neurological, neuropsychiatric or cognitive symptoms may benefit from a viral load measurement via spinal tap as the CNS viral load may be higher.

Risk of resistance

The other main concern in the presence of low-level viraemia is the emergence of resistance. Resistance mutations (changes in the gene sequences of the virus that help it evade the drug) have been detected during low-level viraemia, including at levels below 200 copies. However, the risk for resistance seems to become more pronounced at viral loads of above 200 copies. Below 200 copies, testing for resistance also becomes difficult as there may not be enough copies to run the test or there may not be enough viral particles carrying the resistance, leading to false results or the inability to sequence the viral genes altogether.

Treatment switch

To see whether drug-related factors were linked to the condition, such as low barrier to resistance of the treatment at the time of low-level viraemia, outcomes from treatment switch studies were also included. There were nine studies that assessed virological suppression after switching treatment. The clearest evidence came from a study where participants switching from an NNRTI-based regimen to a protease inhibitor-based one had more virological suppression. NNRTIs are a class of HIV drugs that are considered to have a lower barrier to resistance in general, while protease inhibitors are considered to have a higher barrier to resistance.

Several studies reported better rates of viral re-suppression (getting back to undetectable) after treatment switch or intensification than without changing treatment. Some also suggested that switching treatment reduced the risk of later virological failure. However, the evidence was not strong enough to support switching treatment in all cases, especially when there was no resistance detected and the person was already on a high-resistance-barrier regimen.

Clinical impact

When it comes to broader clinical outcomes, the evidence was inconsistent. Some studies linked low-level viraemia with increased mortality, or AIDS-related and non-AIDS-related conditions, but other studies came to different conclusions. The evidence on CD4 counts was more reassuring: it does not appear that low-level viraemia lowers CD4 counts. Still, a good CD4 count alone does not rule out the broader effects of the virus on the immune system and some smaller studies suggested there was increased immune activation (meaning more inflammation) when low-level viraemia was present.

Low-level viraemia and risk of transmission

The evidence on transmission risk was nuanced, but one fact is clear – there are no documented cases of transmission below 200 copies. A review by the World Health Organisation (WHO) concluded that there is almost zero risk of sexual transmission of HIV with viral loads of less than 1000, while a completely undetectable viral load carries zero risk. In the context of low-level viraemia, there is a risk that the viral load may fluctuate or even start to rise before the next measurement potentially putting sexual partners at risk, hence safer practices may be advisable. The review did not find any evidence on transmission through non-sexual routes.

What predicted low-level viraemia

While there is no single explanation for what causes low-level viraemia, the review identified several contributing factors.

Some studies found that low-level viraemia was more likely in people who had higher viral loads before treatment, periods of detectable viral load, or viral load blips earlier on. (A blip is a one-off detectable reading that returns to undetectable at the next test.)

Apart from these, a lower CD4 count before treatment or before the onset of low-level viraemia was also associated with persistent low-level viraemia. This suggests low-level viraemia doesn’t just reflect what is happening at the moment, but also the longer virological and immunological history of the person.

The findings on treatment adherence were more complicated and partly counter-intuitive. Although missed doses remain one of the most important and actionable causes for low-level viraemia, only some studies found a significant link between adherence and detectable viral load. More controversially, the studies that looked directly at drug concentrations in blood failed to establish a link. These findings highlight the complexity and multifaceted causes of low-level viraemia – they show that not every case of low-level viraemia can be reduced to low adherence and the person’s regularity taking their treatment.  

The guidance

All these factors together make low-level viraemia challenging for clinicians, but can also have serious negative effects on the mental health of people with HIV. They may have no idea why low-level viraemia is happening to them and what to expect.

While this guidance does not answer every question on the topic, it tries to offer a more universal and informed way of approaching low-level viraemia. The final list of recommendations, where evidence from the review and experts’ rating is combined, can be seen below.

The recommendations in full

The recommendations have been adapted in a way to preserve their clinical and scientific accuracy but with small wording changes or added explanations to improve readability.

After each recommendation, letters A to D represent the quality of the existing evidence, A being the highest and D the lowest quality.

The researchers also used the numbers 1 and 2 which describe the strength of the recommendation, where 1 stands for strong and 2 for weak recommendation.

In brackets there is also the agreement percentage among the surveyed experts in the field, where a higher value means more of them agree on a given recommendation. For example, a statement that has received a C2 grade with 92.1% agreement means that the quality of the evidence is low, the recommendation is weak and 92.1% of the experts agree on it.

Definitions and outcomes

• In people with HIV on ART, a viral load between 50 and 200 copies should raise initial concern (grade C2), but its clinical significance depends on the wider clinical context. This includes the stage of HIV disease, the genetic barrier of the antiretroviral regimen, previous virological failure or drug resistance, and whether the viraemia could be coming from clonal expansion (multiplication of infected cells) rather than ongoing viral replication (grade D2). Viral loads between 200 and 1000 copies were considered more concerning (grade B1) and, when confirmed, are generally used to define virological failure (grade C2; 92.1% expert agreement).
• Any single viral load result between 50 and 1000 copies should be repeated in a timely way, generally within one month (grade D1). However, for people with a viral load between 50 and 200 copies who are taking a high-genetic-barrier regimen and have no history of resistance to their current antiretrovirals, repeat testing could be done within around three months (grade D2; 84.1% expert agreement).
• If the low-level viraemia is persistent, the panel recommends a thorough review of the person’s treatment history which should include previous viral load and CD4 count patterns, past instances of virological failure or drug resistance, previous regimens, current adherence, and possible interactions with other drugs, food, herbs or supplements (grade C1; 93.7% expert agreement).

Factors that may predict low-level viraemia

• Because persistent low-level viraemia may be caused by lower adherence, the panel says that adherence counselling should come before any other intervention (grade C1). In some cases, practical measures such as directly observed therapy may also be considered (grade D2; 93.7% expert agreement).
• Where available, drug exposure testing – checking the concentrations or presence of the antiretroviral drugs in blood or urine – may be considered, especially during pregnancy, in suspected malabsorption (inability to absorb nutrients in the intestines due to conditions affecting the digestive system), or when viral load is at least 200 copies (grade D2; 79.4% expert agreement).

Testing for drug resistance during low-level viraemia

• Resistance testing is advised when confirmed low-level viraemia is above 200 copies, if a validated method is available (grade B2). It may also be considered when viral load is between 50 and 200 copies, particularly if the person is taking a low-genetic-barrier regimen (grade C2; 81.0% expert agreement).
• Although resistance testing can be technically possible below 200 copies, sequencing is more likely to succeed and detect resistance when viral load is above 200 copies (grade B2). The panel also recommends that clinicians tell the laboratory the viral load value when requesting resistance testing below 1000 copies (grade D2; 84.1% expert agreement).
• To improve the chance of successful sequencing, the laboratory may increase the volume of plasma (the liquid part of blood) tested, if the assay is validated for this approach. This allows the sample to be concentrated (grade D2; 88.9% expert agreement).

Monitoring the immune system during low-level viraemia

• For people whose low-level viraemia persists for three to six months, CD4 count testing is advised at least once, and CD4-to-CD8 ratio may also be checked (grade D2; 88.9% expert agreement). While CD4 counts may fluctuate during the day or for other reasons, the CD4-to-CD8 ratio is a more stable marker of overall immune health and recovery in people with HIV.

Low-level viraemia that is in or originating from the reservoir or tissues

HIV may specifically prefer to hide in immune organs or the brain, which is usually difficult to access (these hiding places are called viral reservoirs).

• The panel says clinicians should consider the possibility of viral escape, especially in people with a low CD4 nadir (lowest CD4 count someone ever had, usually right before treatment). CNS viral escape means that HIV may be replicating or evolving in the central nervous system in a way that is partly separated from the virus measured in blood. (grade D2; 74.6% expert agreement).
• If persistent low-level viraemia is accompanied by neurological symptoms without another clear cause, cerebrospinal fluid (the fluid that surrounds the brain and the spinal cord) viral load should be measured after brain imaging to assess possible CNS compartmentalisation or viral escape (grade D2; 84.1% expert agreement).

Changing treatment during low-level viraemia

• Treatment decisions depend on the range of the viral load, resistance results and the genetic barrier of the current regimen. If no new resistance mutations are found, or if RNA-based resistance testing is not available, people on a high-genetic-barrier regimen may be able to continue their current ART, especially when viral load is between 50 and 200 copies (grade D2). However, switching from a low-genetic-barrier regimen to a high-genetic-barrier regimen should be considered even if resistance is not detected (grade C2; 87.3% expert agreement).
• If resistance mutations are detected during persistent low-level viraemia, ART should be changed. The choice of new regimen should consider the mutations found, predicted drug susceptibility (how will the virus respond to other drugs based on the present mutations), treatment history and remaining treatment options (if some drugs can no longer be used due to resistance) (grade D1; 98.4% expert agreement).
• If the only new mutation detected is M184I/V in someone taking emtricitabine or lamivudine, changing ART should be considered, especially for people on low-genetic-barrier or dual therapies containing these drugs (grade D1). However, if nucleoside reverse transcriptase inhibitors (a class of HIV drugs such as lamivudine) are included in the new regimen, keeping emtricitabine or lamivudine as part of triple therapy may still be considered, because M184I/V can reduce the likelihood of other NRTI mutations emerging (grade D2; 88.9% expert agreement).
• For people who remain on their current regimen during low-level viraemia, viral load should be monitored more often than in routine follow-up; for example every three to four months during at least the first year (grade D1; 90.5% expert agreement).
• If someone switches regimen during low-level viraemia but does not achieve viral load below 50 copies within six months, repeated resistance testing may be considered (grade D2; 82.5% expert agreement).
• If adherence is good and no new resistance mutations are detected, persistent low-level viraemia may actually be non-suppressible viraemia. This could result from defective viruses or from replication-competent viruses that are unable to complete new replication cycles (viruses that can complete the first part of their replication cycle, but not finish it successfully) (grade D2). Although uncommon, APOBEC3-related mutations detected on RNA-based genotypic testing may support defective viral production as the explanation (grade D2; 88.9% agreement agreement).

Risk of transmission during low-level viraemia

• For transmission risk, the panel notes that no sexual transmission of HIV has been documented with low-level viraemia between 50 and 600 copies. However, when viral load is between 200 and 1000 copies, additional prevention strategies such as condoms or PrEP for partners should be considered, taking into account individual risk factors, possible active viral replication, unpredictable fluctuations in viral load and any upward trend over time (grade B1; 79.4% expert agreement).