Are CD4 counts still useful in the 'treat all' era?

Keith Alcorn
Published: 08 March 2019

CD4 cell testing before starting treatment is still essential even in the era of 'treat all' guidelines, two studies from southern Africa presented at the Conference on Retroviruses and Opportunistic Infections (CROI 2019) in Seattle this week have concluded.

CD4 cell counts measure the health of the immune system and the degree of damage caused by HIV. A CD4 count below 200 cells/mm3 indicates advanced disease and a high risk of opportunistic infections. Treatment guidelines in most countries now recommend that everyone diagnosed with HIV should start treatment as soon as possible, regardless of CD4 cell count. Up until 2016, many countries restricted treatment to people with CD4 counts below 500 cells/mm3 or below 350 cells/mm3, so everyone underwent CD4 cell testing to check eligibility for treatment.

Data from six countries in southern Africa shows that while viral load testing is on the increase in the region, the number of CD4 cell tests being carried out was beginning to fall by 2017 (Egger). This decline worries some researchers, who say that a CD4 cell count before treatment is still needed in the 'treat all' era to identify people with CD4 cell counts below 200 cells/mm3 who need a rapid treatment start, closer monitoring and prophylaxis against infections.

But others think that resources are being wasted on CD4 cell testing and that expansion of viral load testing is the priority, to ensure that people achieve and maintain undetectable viral load on treatment, and to prevent the emergence of drug resistance.

Zambia: CD4 counts before treatment still needed

Zambian researchers reported that the absence of a pre-treatment CD4 cell count was associated with an increased risk of death.

They looked at the relationship between CD4 cell testing and mortality between 2013 and 2015 in people receiving HIV treatment in four provinces. Zambia switched from starting treatment at a CD4 count below 350 cells/mm3 to a CD4 count below 500 cells/mm3 in April 2014.

The researchers looked at mortality in 33,911 new antiretroviral therapy (ART) users, of whom 20,911 had a CD4 cell measurement less than six months before starting treatment. They incorporated data from an existing tracing study of people lost to follow-up to calculate mortality in people lost to follow-up, so are unlikely to have under-counted deaths in this group.

They found that people who lacked a CD4 count measurement were 45% more likely to die after starting treatment. Whereas 6.6% of those tested less than six months prior to starting treatment died during the follow-up period, 1.5% of those without a measurement died (p = 0.0009).

The majority of deaths occurred within the first 90 days after starting treatment, suggesting that opportunistic infections are not being diagnosed at the time of treatment initiation.

They conclude that CD4 counts are still useful for identifying people at higher risk of illness who may need specialised management.

Botswana: CD4 counts on treatment not needed for over-200s

The Botswana-Harvard AIDS Institute investigated whether CD4 counts still provided useful information in the 'treat all' era by looking at baseline CD4 counts and changes in CD4 count after starting treatment, especially the proportion of people with a CD4 count above 200 who experienced a decline in CD4 count below 200 after starting treatment. Data come from the greater Gaborone region between 2015 and 2017.

During the study period Botswana switched from guidelines recommending treatment initiation at any CD4 cell count below 350 to a 'treat all' guideline in June 2016. Guidelines on monitoring CD4 counts after starting treatment also changed. Prior to June 2016, CD4 counts were carried out three months after starting treatment and then six-monthly if the CD4 count was below 300 cells/mm3 or 12-monthly if it was above 300. After adopting the 'treat all' policy, CD4 counts were recommended three months after starting ART, at month 6 if the previous CD4 count was below 200, at month 12 and every year thereafter.

A total of 14,425 new patients who underwent CD4 cell testing during the study period were identified. Half had only one test, 18.6% had two, 12.5% three tests, 9.3% four tests and 9.2% had five tests or more. Viral load testing was carried out in 79.4% of the patients who underwent CD4 cell testing (viral load testing was recommended for all patients three to six months after starting treatment during this period).

Twenty-five per cent of people tested had CD4 counts below 200 cells/mm3 and half of this group had CD4 cell counts below 100 cells/mm3. Men were more than twice as likely to have CD4 counts below 200 cells/mm3 as women and people with CD4 counts below 200 cells/mm3 were twice as likely to be over 35. The researchers found no clear trend towards a reduction in the proportion of people starting treatment with a CD4 count below 200 cells/mm3 over the three years of the study.

A total of 10,854 people had a baseline CD4 count above 200 cells/mm3 and 5062 of these people had at least two CD4 counts during this period. Only 3.6% experienced a drop in CD4 count below 200 after their baseline CD4 count. Men were twice as likely to experience a CD4 decline below 200 cells/mm3 and drops below 200 cells/mm3 occurred predominantly in people with baseline counts between 350 and 200 cells/mm3.

Of those who experienced a CD4 cell drop below 200, 82% had a viral load test within six months of the CD4 cell drop and only 21% had a detectable viral load (> 400 copies/ml). However, in the vast majority of cases the subsequent CD4 cell measurement was above 200 cells/mm3, indicating that the CD4 cell drop was transient.

The researchers concluded that while there is very limited benefit to ongoing CD4 monitoring if people have CD4 counts above 200 at the time they start treatment, baseline CD4 cell tests remain essential to identify people with advanced disease.


Egger M et al. Trends in CD4 and viral load testing in southern Africa: analysis of 6 countries. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 150, 2019.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.

Lemme TB et al. Utility of CD4 cell count monitoring in Botswana: analysis of routine laboratory data. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 149, 2019.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.

Sikombe K et al. Early mortality in HIV-infected patients initiating ART without a pre-therapy CD4. Conference on Retroviruses and Opportunistic Infections, Seattle, abstract 148, 2019.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.

Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.