Managing non-communicable diseases among people living with HIV

Non-communicable diseases (NCDs) – including cardiovascular diseases, diabetes, cancers, and other illnesses – will represent a significant challenge for HIV care in low- and middle-income countries as the population of people on HIV treatment grows and ages, the 21st International AIDS Conference (AIDS 2016) held in Durban heard in July.

Developing models of care which can manage non-communicable diseases in people living with HIV, as well as in the general population, will be a critical part of developing a health system that can provide different patterns of care to people with HIV according to their needs – so-called `differentiated` care.

Kara Wools-Kaloustian from Indiana University School of Medicine, co-primary investigator for the East African International Epidemiologic Database to Evaluate AIDS (IeDEA) cohort, believes the best response would be to develop and strengthen systems that support health in the general population, whether for HIV or for NCDs.

"An integrated chronic disease management model will likely be the most cost-effective and sustainable approach," she said.

Burden of NCDs

According to the World Health Organization, NCDs kill 38 million people each year. The majority of these deaths by far (28 million) are thought to occur in low- and middle-income countries, which have been slower to develop effective responses to detect, prevent, and treat these diseases. Deaths from NCDs occur at earlier ages in these countries, with 82% in people younger than 70 years.

The majority (56%) of NCD-related deaths are caused by cardiovascular diseases such as hypertension, cerebrovascular disease including strokes, coronary artery disease, and cardiomyopathy. Cancers (primarily non-AIDS-related) are responsible for 26% of the deaths, while respiratory diseases such as chronic obstructive pulmonary disease (COPD) and pulmonary hypertension account for 13%, and diabetes for 5%. Other conditions including chronic kidney disease, liver disease, neurodegenerative disorders, osteoporosis, and frailty also contribute to NCD-related mortality. Moreover, NCDs cause illness, disability, and greatly reduced quality of life for an even greater number of people.

However, getting a precise estimate of the burden of NCDs in some low- or middle-income countries – and comparing the burden of disease between people living with HIV and HIV-negative people, or between treatment-naive HIV-positive people and those on antiretroviral therapy (ART) – can be difficult due to a dearth of data. Furthermore, there is limited laboratory capacity to assess the risk of NCDs and limited clinical capacity to diagnose them in many of the countries.

"Our HIV treatment programmes do not routinely collect information on NCD risk factors or NCD occurrence," said Paula Munderi of the Medical Research Council/Ugandan Viral Research Institute’s Unit on AIDS. "Most of our countries do carry out national HIV surveillance and NCD surveys, but these 2 surveys are not linked."

Munderi shared data from a few East African cohorts. A survey of four districts in Uganda and Tanzania found a prevalence of hypertension in the general population that ranged from 19 to 27% (depending upon whether one lived in an urban or rural community). This was generally much higher than the prevalence of HIV, which ranged from 6 to 12%. The prevalence of diabetes ranged from 2 to 4%, heart failure from 2 to 9%, and COPD from 2 to 6%. 

Another much larger survey of hypertension that included 65,000 adults from 20 rural communities in Uganda, with an overall HIV prevalence of 5%, found a hypertension prevalence of 14%. The majority (79%) of these cases had not been previously diagnosed and only 15% of those found to have hypertension were currently on treatment. 

While the risk of NCDs appears to be higher among people living with HIV compared to the general population in most industrialised countries, in these East African studies the prevalence appeared roughly the same. But NCD cases may be under-diagnosed or under-reported at facilities where HIV care is provided.

In addition, the Ugandan/Tanzanian survey found that the diagnosis and management of NCDs was only taking place in district hospitals and larger health centres, not at the lower level health centres and dispensaries where most of the community lives and where efforts are underway to decentralise HIV treatment. Healthcare providers at these smaller sites told the investigators that they had had recent training on HIV, but no training on NCD management.

Also, the risk of NCDs may not yet be as great in some of the poorer countries where diet and lifestyles differ from those of wealthier countries. For instance, a survey looking at cardio-metabolic risk factors in Malawi showed that the overall prevalence of hypertension and diabetes did not differ between people who were HIV-positive and HIV-negative. However, in this setting the risk of being overweight was lower among HIV-positive individuals, possibly due to weight loss in people with advanced immune suppression.

In contrast, in South Africa, where obesity is much more prevalent, hypertension is the most common reason for visits to primary health clinics, but there also appears to be a substantial number of people with multiple diagnoses, according to a presentation by Tolu Oni from the University of Cape Town School of Public Health and Family Medicine.

A recently published cross-sectional study looking at hypertension, diabetes, HIV, and tuberculosis-related consultations among people attending Michael Mapongwana Clinic, a primary care health facility in Khayelitsha near Cape Town, found that the burden of hypertension and diabetes was higher among HIV-positive people younger than 46 years of age, but this was not observed among older people coming in for care. Almost a quarter of people with multiple diagnoses had HIV with either hypertension or diabetes or both.

Oni noted that in South Africa there may be differences in how patients prioritise HIV versus hypertension or diabetes care. This could lead to an under-reporting of NCDs, as well as poorer treatment outcomes that people living with HIV can ill afford.

Pathophysiology of NCDs

Data from the Danish cohort study, which compares the average life expectancy of the average 50-year-old living with HIV in the modern treatment era to an HIV-negative individual in the general population, showed that there is about a ten-year shorter life expectancy. This is true even when restricted to individuals who do not have any other co-morbidities.

The outlook is even worse among people who started HIV treatment with a CD4 count of less than 350 cells/mm³, according to data from the NA-ACCORD cohort, which suggested a 20-year shorter life expectancy among individuals who start treatment late. Most of this mortality is attributed to the early onset on NCDs.

"This is particularly important since the vast majority of the 17 million people living with HIV around the world, most of whom are from low- or middle-income countries, have started ART at low CD4 counts. And this is a problem we are going to see for a long time to come," said Peter Hunt of the University of California at San Francisco, who gave a presentation on why HIV appears to be associated with a higher risk of NCDs over and above what could be explained by lifestyle risk factors.

Hunt focused primarily on the role of persistent inflammation, which fails to normalise to the levels seen in individuals who do not have HIV despite years of viral suppression on ART. Possible causes for this inflammation include low-level HIV replication from viral reservoirs, cytomegalovirus (CMV) co-infection, and microbial translocation – where a leaky gut barrier allows bacteria to get into the bloodstream – all of which can drive chronic immune activation.

Regardless of the cause, markers of innate immune activation and inflammation remain abnormal in people with HIV and predict morbidity and mortality quite strongly. For instance, a single measurement of the inflammatory cytokine IL-6 is strongly predictive of subsequent serious non-AIDS events – mostly NCDs – and mortality over subsequent years in people on ART.

"This is not just a high-income country phenomena; this is also occurring in low- and middle-income countries as well," Hunt emphasised.

For now, approaches to reducing NCD-related illness and death are primarily limited to moderate exercise and reducing lifestyle risk factors such as smoking, drinking, and illicit drug use. However, Hunt is optimistic that the use of statins – which reduce monocyte and T-cell activation as well as lowering blood lipids – will have a role in treatment.

This hypothesis may be confirmed by the large international REPRIEVE trial, which is testing the new and better-tolerated pitavastatin in 6500 people with HIV. Unfortunately, pitavastatin is not yet available in all low- and middle-income countries.

In the future, Hunt believes it may be possible to develop other treatments that more effectively address the common immunological pathways that drive numerous NCDs.

Prevention would be better than a cure, and part of Oni’s presentation focused on opportunities to integrate HIV and NCD prevention that she believes are being missed in adolescents, as many NCD lifestyle risk factors such as smoking, an unhealthy diet, and lack of physical activity often develop in adolescence.

Models for delivery

In many ways, the rollout of ART in low- and middle-income countries has blazed a trail that could show how to respond more effectively to NCDs. HIV and NCDs both require a healthcare infrastructure that supports continuity of care, incorporation of community education, prevention, screening, treatment, and palliation.

"The most important thing is political commitment with sustained funding, because we cannot care for HIV, nor can we care for chronic diseases, without commitment and funding," Wools-Kaloustian said. This may prove challenging as NCDs do not benefit from the same degree of grassroots activism that drove the HIV response.

Wools-Kaloustian noted that the HIV field has learned to use alternate structures and mobile services, and standardised and simplified treatment approaches have allowed task-shifting to address the shortage of providers. Better point-of-care diagnostics are already available for diabetes management, though they may prove challenging for some other NCDs such as cancer. Access to cheaper generic drugs was fast-tracked and government procurement systems have been strengthened, which will be critical in making NCD treatment accessible and affordable.

Questions going forward

During the discussion at the session there were questions about whether NCD services should be stand-alone or integrated with HIV services.

"My personal sense is that the conditions are perfectly integrated in the patient – and it is us who seem to dis-integrate things in our thinking," Munderi said.

However, concerns were raised about trying to shift too much of the burden of chronic care onto peer groups and community health workers.

"Community health workers are wonderful tools but they cannot do everything – and they can’t do anything without being paid," said Wools-Kaloustian. "We cannot expect community health workers to be doctors, and the more we put on their plate, the more there is a need for formal education. We need more doctors and nurses, and need to incentivize them to remain in their countries after they have been trained."