Age and CD4 count have the greatest influence on life expectancy in the modern HIV treatment era

A study published in this month’s Lancet HIV claims to have produced one of the most detailed analyses of life expectancy among people living with HIV in high-income countries in the modern treatment era. It found that for people on antiretroviral therapy (ART) and with high CD4 cell counts, life expectancy was only a few years lower than the general population, regardless of when they had started treatment.

People who started treatment after 2015 have a slightly higher life expectancy than those who started ART before 2015. However, CD4 cell count and age have the strongest influence on life expectancy, with those with a very low CD4 count (below 50) estimated to have around twenty years fewer remaining than those with a high CD4 count (500 or over). The authors argue that “our results suggest the continuing importance of early and sustained ART”.

The life expectancy of people living with HIV has improved dramatically since the introduction of effective antiretroviral treatment, with a number of studies reporting near-normal or normal life expectancy for people living with HIV compared to the general population. However, all these studies were based on data limited to the first few years after people have started treatment. This does not help people who have been on ART for many years understand how their life expectancy compares to the general population, and whether long-term HIV infection can shorten their life expectancy despite successful treatment.

To answer this question, Dr Adam Trickey from the University of Bristol formed an international team of researchers to examine data from twenty cohort studies of people living with HIV in North America and Europe. The study was limited to people who were aged 16 or over when starting treatment, therefore mostly excluding those who acquired HIV as children. One group of participants had started ART between 1996 and 2014 and were still alive and on treatment in 2015, when follow-up data began to be collected. Another group of participants started ART between 2015 and 2019 and subsequently survived for at least a year, when their follow-up data began to be collected. The 2015 cut-off was chosen as this was when treatment guidelines changed to recommend treatment for all people diagnosed with HIV, no matter what their CD4 count was.

In all, 206,891 people with HIV were included in the analyses, with 5780 deaths recorded from 2015 onwards. People who started ART after 2015 tended to be younger and have lower CD4 counts and higher viral loads at the start of follow-up compared to those who started ART before 2015. This is most likely because those in the first group had already been on treatment for some time before follow-up began in 2015 (median 7.8 years). In contrast, among those who started ART before 2015 there was a higher proportion who had received an AIDS diagnosis and been exposed to older treatment regimens with more side-effects.

Risk factors

There were various factors which the authors anticipated would influence life expectancy, including:

• Age
• Sex
• Year of starting treatment
• How someone acquired HIV (especially if they acquired HIV through injecting drug use)
• CD4 and CD8 count one year after starting treatment and/or at start of follow-up
• Their lowest CD4 and highest CD8 count before starting treatment (and between starting treatment and 2015, if they started treatment before then)
• Viral load at start of follow-up and one year after starting treatment
• Having AIDS or hepatitis C at start of follow-up
• Previous exposure to ART drugs with increased side effects (zidovudine (AZT), didanosine (ddI), zalcitabine (ddC) or stavudine (d4T)), or monotherapy and dual therapy.

After controlling for differences in these factors among the participants, the authors found that the greatest risk factor for death was CD4 count at the start of follow up (either 2015 or a year after starting treatment for those starting after 2015). The higher someone’s CD4 count, the lower the risk of death – people with a CD4 count below 50 had nearly five times the risk (372% greater chance) of death compared to those with a CD4 count above 500, while those with a CD4 count of 200-349 had almost twice the risk (92% greater chance) of death.

Unsurprisingly, age also had a considerable influence on life expectancy. People aged 60-69 had approximately three times the risk (219% greater chance) of death compared to those aged 30-39, while those aged 70 or older had nearly eight times the risk (666% greater chance) of death. Women had a slightly lower risk (23% lower chance) of death compared to men.

Compared to men who acquired HIV through sex with another man, those who acquired HIV through injecting drug use had nearly 2.5 times the risk of death (148% greater chance). This is likely due to socio-demographic factors and the health risks associated with injecting drugs. In contrast, those who acquired HIV through heterosexual sex had only a slightly higher risk of death (24% greater chance) compared to men who have sex with men.

When someone started treatment did influence their life expectancy, but less so than other factors. Those who started treatment in 1996-1999 had a 30% greater chance of death compared to those who started in 2015-2019. This risk reduced slightly for those who started treatment in later years, with those starting ART in 2000-04 having a 21% greater chance of death, and those starting in 2005-09 and 2010-14 both having an 18% greater chance of death.

Other high-risk factors included having, at the start of follow-up, a viral load above 50 copies/ml (30% greater chance of death compared to those with viral load below 50 copies/ml), hepatitis C (38% greater chance compared to those without hepatitis C), and AIDS (60% greater chance compared to those without AIDS).

Interestingly, factors measured prior to follow-up, such as a participant’s lowest CD4 count or previous exposure to ART drugs with increased side effects, did influence life expectancy but not as substantially as might be expected. Compared to those whose lowest CD4 count was above 500 before follow-up started, those who had a lowest CD4 count between 200-349 before follow-up had a 5% greater chance of death, while those who had a lowest CD4 count below 50 had a 17% greater chance of death. Similarly, those exposed to ART with increased side-effects had only an 18% greater risk of death and those exposed to monotherapy or dual therapy had only a 3% greater risk of death, compared to those who were not exposed to these treatments.

Life expectancies calculated

The authors then calculated life expectancies for the participants based on whether they started ART before or after 2015 and how they acquired HIV, as well as viral load, AIDS status, and CD4 count at the start of follow-up.

For people starting treatment before 2015, the average life expectancy for those currently aged 40 was 76 years for women and 75 for men, compared to 86 and 81 in the general population respectively.

"The findings of this study provide insight for those who started treatment before 2015 and have been on treatment for many years."

However, there was a considerable range in estimates depending on the factors outlined above. For example, a 40-year-old woman who started treatment before 2015 with a CD4 count below 50 at the start of follow-up could expect to live until 59, and a man in the same situation could expect to live until 58. Life expectancy improved with increased CD4 count, so a 40-year-old woman who started treatment before 2015 with a CD4 count between 200-349 at the start of follow up could expect to live to 74, and a man in the same situation could expect to live until 72. A woman who started treatment before 2015 with a CD4 count above 500 at the start of follow up could expect to live to 80 years on average, and a man in the same situation could expect to live to 78. If she had a suppressed viral load, no AIDS diagnosis at the start of follow-up, and didn’t acquire HIV through injecting drug use, her life expectancy increased to 82 years. A man in the same situation could expect to live to 79 years.

For those starting treatment after 2015, the average life expectancy for those currently aged 40 was 79 years for women and 77 for men. If they had a CD4 count below 50 at the start of follow-up this reduced to 65 and 64 years respectively. If they had a CD4 count between 200-349, their life expectancy was similar to the average, calculated as 78 and 77 respectively. Whereas if their CD4 count was above 500, their life expectancy increased to 82 and 79 years respectively. If, in addition to a CD4 count above 500, they had a suppressed viral load, no AIDS diagnosis at the start of follow-up, and didn’t acquire HIV through injecting drug use, a 40-year-old woman could expect to live to 83 and a 40-year old man could expect to live to 80.

Life expectancy also depended on a person’s age. A 20-year-old woman could expect to live to 72 if she started treatment before 2015, or 77 if she started treatment after 2015. A 20-year-old man could expect to live to 71 or 75, respectively.

For both men and women, those who acquired HIV through injecting drug use and those who had AIDS at the start of follow up had the lowest remaining life expectancies.

Conclusion

The findings of this study provide insight for those who started treatment before 2015 and have been on treatment for many years. While those who started treatment after 2015 had slightly higher life expectancies, the difference was reduced when limited to those with high CD4 counts at start of follow-up. Factors related to someone’s history of living with HIV, such as previous use of ART with more side effects or a low CD4 count before starting treatment, also reduced life expectancy estimates very slightly, but ultimately the most influential factors were age and CD4 count at the start of follow-up.

Find out more: Life expectancy for people living with HIV

The authors point out that their study gives no indication of the quality of health in remaining years. A previous study found that on average, people living with HIV are likely to develop major illnesses 16 years earlier than those who are not living with HIV. They were also unable to include other factors which influence everyone’s life expectancy, whether or not they are living with HIV, including social and economic circumstances and lifestyle factors such as whether someone smoked or not.