Italian study shows growth in heart and kidney problems in people living with HIV

Reduced kidney function, high blood pressure and cardiovascular disease have each become more common in a large cohort of Italians living with HIV followed since 2004, with a majority of people now having high cholesterol and high blood pressure, researchers from the Italian ICONA cohort report in HIV Medicine. By 2014, half of all people followed since 2004 were at high risk of a major cardiovascular event such as a stroke or heart attack.

The researchers say that the dramatic increase in the prevalence of non-communicable diseases usually attributed to ageing requires special attention.

More than half of all deaths in people taking antiretroviral therapy (ART) are due to non-communicable diseases such as non-AIDS-defining cancers, cardiovascular disease, or end-stage liver or kidney disease. Many of these conditions are a consequence of growing older but can also be exacerbated by HIV infection, so they may occur at a younger age, or at greater frequency, in people living with HIV.

These co-morbidities also complicate the prescribing of antiretroviral drugs and any medications needed to treat other health conditions, due to drug-drug interactions. People with more co-morbid conditions are more likely to be at risk of unexpected drug-drug interactions if their doctors are unaware of all potential drug interactions.

To examine whether co-morbid conditions are becoming more common in people living with HIV as they age, Italian investigators looked at the prevalence of these conditions in two cross-sectional analyses, in 2004 and 2014, in a closed cohort. They also looked at the prevalence of these conditions according to HIV treatment history in an open cohort, to see whether a lack of treatment – or more prolonged treatment – had any impact on the prevalence of co-morbidities.

Closed cohort

The closed cohort consisted of individuals who were seen for care both in 2004 and 2014. By definition, each participant was ten years older in 2014, therefore the prevalence of non-communicable diseases would be expected to increase, but not by as much as was observed.

It consisted of 1517 people (33% women) with a median age of 41 years in 2004 and 51 years in 2014. By 2014, 13% of the cohort were aged 61 or over. Forty-three per cent had acquired HIV through injecting drug use, 23% by heterosexual contact and 27% were men who have sex with men.

In 2004, 21% were not taking ART but by 2014 only 2% were not on treatment. The median CD4 cell count in the cohort rose from 507 to 706 cells/mm³ and by 2014, 92% had a CD4 cell count above 350 cells/ mm³.

The prevalence of obesity, measured by body mass index, did not increase significantly. The prevalence of smoking declined from 55% to 46%, a significant reduction (p < 0.001). However, alcohol consumption increased: in 2004, 34% of the cohort reported no alcohol use, falling to 19% in 2014 (p < 0.001).

The prevalence of hepatitis C in the cohort was high: 33% had the virus in 2014 and 5% were hepatitis B surface antigen positive.

Several co-morbidities increased in prevalence between 2004 and 2014. Whereas 5% had impaired kidney function in 2004 (defined as eGFR < 60ml/min), almost a third (30%) had impaired kidney function (p < 0.001) in 2014.

In 2004, 18% of the cohort had cardiovascular disease, defined as a heart attack, stroke or invasive coronary procedure occurring at least one year prior to 2004. By 2014 almost a third of the cohort had a history of cardiovascular disease (32%), a very high burden of cardiovascular disease in a cohort with a median age of 51 years.

The proportion of people at high risk of a cardiovascular event within the next five or ten years also increased sharply. In 2004, 20% were at high risk of a cardiovascular event within five years (> 10% risk) using the D:A:D scoring system. By 2014, 50% were at high risk (p < 0.001).

The prevalence of dyslipidaemia (at least one of total cholesterol > 6.2mmol/l, HDL cholesterol > 0.9mmol/l or triglycerides > 2.3mmol/l in the previous year) rose from 75% in 2004 to 91% in 2014. The prevalence of hypertension rose from 67% to 84% between 2004 and 2014 (p < 0.001).

Open cohort

The investigators also looked at an open cohort of 3688 patients receiving treatment in 2004 and 6679 receiving treatment in 2014. This analysis covered everyone in the closed cohort, everyone in care in 2004 who subsequently died or became lost to follow-up, as well as individuals who joined the ICONA cohort after 2004 and were still in care in 2014.

With this continuous enrolment, the demographic profile of the cohort changed over time. The purpose of the open cohort analysis was to investigate whether these demographic changes affected the prevalence of co-morbidities.

Over the ten-year study period, the proportion of people who acquired HIV through injecting drug use fell from 32% to 11%, the proportion who were female fell from 32% to 22% and proportion with HIV and hepatitis C co-infection fell from 37% to 13% (all p < 0.001). These changes indicate a shift in the nature of people receiving care in Italian HIV clinics, from being predominantly people who inject drugs and people with hepatitis C co-infection, including many women, to a population of mostly men who have sex with men who were less likely to have hepatitis co-infection.The proportion who smoked also fell, from 54% to 37%.

The median age of the open cohort increased less markedly (from 41 to 44), reflecting the entry of newly diagnosed people into the cohort.

In contrast to the closed cohort, the prevalence of most co-morbidities fell in the open cohort. There was no increase in the prevalence of impaired kidney function and only modest increases in the proportion of the cohort at high risk of a cardiovascular event.

However, when investigators looked at the prevalence of co-morbidities according to treatment history, they found that the greatest increases in impaired kidney function occurred in people already on treatment at least one year prior to their first evaluation (either 2004 or 2014). Similarly, the greatest risk of cardiovascular events was seen in the same group. The analysis does not look at the relationship between specific drugs and co-morbidities, or other variables that might affect the development of co-morbidities.

Conclusions

“The burden of NCDs [non-communicable diseases] in PLWHIV [people living with HIV] in Italy appears to have markedly worsened over a 10-year span, which is likely to be the result of both ageing and HIV infection as well as their interaction,” the investigators conclude. “Particular attention needs to be given to drug interactions between ART and co-medications for NCDs,” they note, as well as early detection of co-morbidities and optimisation of ART regimens to reduce the risk of any co-morbidities, especially reduced kidney function and cardiovascular disease.