Fast forward thirty years and developments in the drug treatment of HIV have changed it from a death sentence to a manageable, chronic condition, frequently compared to diabetes. Research has shown that when patients are diagnosed and treated early, before their immune system has been severely compromised, they can be advised that they are likely to have a near normal life expectancy and that their deaths may well not be related to HIV.1, 2
This shift from an imminently life threatening condition to one that can be controlled with regular medication is of course welcome. It does however bring with it a whole new set of challenges. Certainly, HIV physicians can no longer afford to ignore the 50 years plus group as the prevalence of HIV in this age group continues to rise.
The focus of this article will be on the challenges specific to the management of older people with HIV. These include the challenge of diagnosing the older person in the first place; the challenge of managing the effects of HIV and of highly active anti-retroviral therapy (HAART) on age-related pathologies; and the need to consider the potential interactions between HAART and the drugs commonly used to treat these pathologies (Box 1).
Making the diagnosis
By the end of 2011, an estimated 96,000 people were living with HIV in the UK; approximately one quarter of whom were undiagnosed and unaware of their infection.3 Given that the full benefits of HAART rely largely on early diagnosis, there has been a national drive to expand HIV testing and diagnose people earlier. The logic behind the HIV testing guidelines published by the British HIV Association in 2008 was that earlier diagnosis would translate into less morbidity and mortality for people with HIV and also reduce transmission rates. The recommendation was that in populations where the prevalence of HIV exceeds two per 1000 population, HIV testing should be offered to all adults registering with a GP and all general medical admissions.4
Progress has been made in this respect and in 2011, the need for earlier diagnosis via expanded testing was acknowledged by NICE with the publication of guidelines designed to facilitate and normalise HIV testing in primary and secondary care.5,6 Despite this progress, the statistics on HIV prevalence in the over 50s give cause for concern.
Increasing prevalence and late diagnosis in the over 50s
The proportion of adults accessing HIV services who were above the age of 50 rose from 12% in 2002 to 22% in 2011.7 Worryingly, the number of older adults being diagnosed late, after HAART should have been commenced, is disproportionately high; 61% in over 50s versus 45% in those younger than 50 years.7 It is well established that late diagnosis is associated with increased morbidity and mortality8 and a poorer response to HAART.9
Challenging misconceptions, perceptions and attitudes
The reasons for the trend towards late diagnosis in the over 50s are not clear but are likely to do with the fact that HIV prevention, counselling, testing, and education efforts are largely directed at the younger generation. This may be because widely-held societal beliefs are that after the age of 50 years, people become either sexually inactive or are in monogamous heterosexual relationships.
The fact that the majority of newly infected older people have acquired HIV through sexual transmission indicates that these beliefs are simply not true and there is evidence that older men with or at risk for HIV infection are sexually active, participate in risky sexual behaviour, and need safer sex interventions.11
This may in part due to the fact that there are an increasing number of older people finding themselves single after a divorce, coupled with the increasing acceptability of internet dating. This generation may not be as au fait with condom use as their younger counterparts, thereby putting themselves at increased risk of sexually transmitted infections, including HIV.
Another factor contributing to late diagnosis in the over 50s is that physicians are missing opportunities to test for HIV in older patients. Amongst the reasons for HIV not being considered as a differential in the elderly patient, is the fact that many of the "indicator conditions" that should prompt an HIV test, such as bacterial pneumonia and recurrent herpes zoster, are much more common amongst older people and can be easily explained away by "ageing".
Physicians without an awareness of the growing problem of HIV in older people are therefore less likely to look for an underlying cause than they would be if seeing these conditions in a younger patient (Box 2).
Unfortunately, even when HIV is on the list of differentials, barriers to testing remain. These may include physician-held assumptions about the necessary amount of counselling required prior to HIV testing and uncertainty about onward referral, as well as patient-held misconceptions about implications for mortgage and insurance applications.
How to overcome late diagnosis in the over 50s
Ultimately, more HIV testing in older people is what is needed to overcome the issue of late diagnosis and to improve morbidity and mortality rates in this group. This necessitates a greater awareness of the scale of the problem amongst physicians and the general public, misconceptions about sexual activity in the over 50s to be challenged and perhaps most importantly, the normalisation of HIV testing especially in areas of high HIV prevalence.
Managing comorbidities and drug toxicities
HIV and HAART are associated with metabolic problems and cardiovascular disease as well as with neurocognitive impairment and renal disease. These associations are compounded by advancing age and mean that there is a need for both physicians caring for HIV patients and those caring for older patients to be aware of the complex interaction between age-related pathologies and HIV.
Metabolic and bone complications of HIV and HAART
HIV and HAART have been associated with a range of metabolic problems including dyslipidaemia, diabetes and osteoporosis. Prior to the widespread use of HAART, dyslipidaemia, characterised by reduced HDL levels and raised triglyceride levels, was common in HIV. The mechanisms by which this occurs have not been conclusively established but chemokine-mediated inhibition of lipoprotein lipase activity may play a role.12
The use of HAART, particularly the PI (protease inhibitor) class of drugs, has been shown to compound the situation and cause a rise in LDL and total cholesterol levels in addition to low HDL and high triglyceride levels. The situation is exacerbated by increasing age with one large Italian study demonstrating significantly higher cholesterol and triglyceride levels in HIV patients on HAART over the age of 50 as compared to those under the age of 35.13
An increased prevalence of insulin resistance, glucose intolerance and diabetes has been reported in HIV in the HAART era, with one large study showing rates of diabetes four times higher in HIV positive men than in those without HIV.14 The risk increases with length of time on HAART. As with dyslipidaemia, the link between HIV treatment and diabetes is stronger in older patients.15
There is a high prevalence of osteopenia and osteoporosis among HIV patients, with more rapid progression of bone mineral loss than in the HIV negative population.16 Several factors contribute to bone loss including HIV infection itself, associated chronic inflammation and exposure to HAART. Specifically, tenofovir, currently recommended as first line treatment for HIV, has been associated with loss of bone mineral density and with increased fracture rates in several studies, although the mechanism by which this occurs is not well established.16a Furthermore, chronic HIV infection is associated with vitamin D deficiency and this too may play a role in poor bone health. One meta-analysis estimated that 67% of HIV-infected people exhibited reduced BMD, and 15% had osteoporosis.17 This is clinically significant, with research data from 2011 showing that HIV infected patients are more likely to suffer bony fractures than the general population, especially older patients with low nadir CD4 counts.18 Information about bone disease and HIV has now been incorporated into the British HIV Association monitoring guidelines, which recommend routine DEXA scanning in men over the age of 70 years and women over the age of 65 years.19
Cardiovascular complications of HIV and HAART
High rates of dyslipidaemia and diabetes, in addition to the fact that HIV infection itself seems to be a driver for atherosclerosis and MI, mean that it is no surprise that CVD has emerged as a major problem in the ageing HIV population. Numerous population-based studies confirm higher rates of CVD in HIV-infected patients than in people without HIV.20, 21 One study demonstrated that the risk of preclinical atherosclerosis conferred by HIV infection was similar to that conferred by smoking22 while another demonstrated that HIV infection had a similar impact on MI risk to diabetes.23
The World Health Organization has predicted that by 2030, ischaemic heart disease and HIV infection will rank among the top 10 causes of global mortality.24 The likely association and possible synergy between the two diseases suggest they will have an even greater effect on global mortality and pose major clinical and public health problems.
The potential mechanisms by which HIV increases CVD risk are the subject of ongoing study. They are likely to include exposure to HAART; a well documented excess of traditional risk factors such as smoking, diabetes and hypertension in HIV patients, unrelated to HAART; as well as HIV-specific factors including inflammation, immune dysregulation and endothelial dysfunction. All of these factors may contribute to the higher risk of CVD and may act synergistically.
Managing CVD in the HIV population comes with a number of challenges. One of these challenges is determining the degree to which the findings of general population studies can be extrapolated to the HIV population. For instance, when screening HIV patients for CVD risk, one needs to consider that there is debate around the degree to which the Framingham risk equations are applicable in HIV infected persons.20,25 Designing a cardiac friendly HAART regimen has become easier with the development of newer agents that are not associated with as much dyslipidaemia and insulin resistance as older drugs. Nevertheless, with other side effects, drug interactions and
pill burden still to consider, this can remain a challenge with some patients.
Despite this, extensive research suggests that long term, uninterrupted virological suppression, the goal of HAART, does reduce overall CVD risk despite the role of HAART in increasing certain risk factors. Current UK management of CVD risk in HIV includes a baseline and annual CVD risk assessment, most commonly using the Framingham score, addressing modifiable risk factors such as smoking and hypertension and paying attention to differences in cardiovascular risk profiles of various anti-retrovirals when planning treatment.
Neurocognitive impairment in HIV
The detrimental effect of HIV on the brain is well established, with studies demonstrating that HIV associated neuro-cognitive disturbances are very common even in patients on HAART26 and that there are similarities between HIV associated dementia and Alzheimer's disease.27 A more recent study28 has shown that HIV infection is associated with the premature loss of brain functional abilities by approximately 15 to 20 years. In this study, functional magnetic resonance imaging revealed reduced cerebral blood flow in HIV patients as compared to uninfected controls. The reduction was to levels usually seen in people 15 to 20 years older. HIV patients also had greater increases in blood flow when required to perform visual tasks, suggesting more neurological effort was required by the HIV patients to achieve the same goal. Cerebral damage is thought to be mediated by neurotoxic immune factor release in the brain. These research findings may influence the choice of HAART in the older HIV patient, suggesting a requirement for regimens tailored to neuro-protection if they begin to show signs of cognitive impairment.
Renal complications of HIV and HAART
The risk of renal disease increases with age as renal function slowly declines. Data from various HIV studies have demonstrated a faster decline in estimated glomerular filtration rate in patients exposed to various HAART regimens,29, 30 in particular tenofovir and atazanavir, which are both preferred first line treatments for HIV. These data suggest that clinicians need to be vigilant in monitoring renal function in ageing patients.
Managing drug interactions and polypharmacy
Many HIV patients on HAART take multiple medications concomitantly, with an analysis of a large HIV cohort study demonstrating that 40% of patients had at least one potential drug-drug interaction.31 Countless studies have demonstrated the fact that polypharmacy is the norm in elderly patients and this just increases the potential for drug-drug interactions. The majority of clinically significant drug interactions arise from the fact that many HIV drugs are substrates, inducers or inhibitors of the cytochrome P450 enzyme pathway. The second largest contributor to HIV drug metabolism and therefore, clinically significant drug interactions, is the UGT pathway. Both the cytochrome P450 and the UGT pathways are utilised by many of the drugs commonly used to treat age-related pathologies.
Depending on the type of interaction, there is the potential for sub-therapeutic levels of HAART, leading to detectable HIV replication and resistance problems; sub-therapeutic levels of concomitant drugs and the subsequent mismanagement of comorbidities; increased levels of HAART or concomitant drugs with an increased likelihood of side effects. The potential for multiple drug interactions is the reason that a two-way communication between HIV physicians prescribing HAART and doctors in primary and secondary care prescribing other drugs, is crucial.
Highlighting potential interactions and educating HIV patients about their existence have also become a part of the role of the HIV pharmacist. Furthermore, there are a number of online resources available to help prescribers, the most widely used being the database developed at Liverpool University, at www.hivdruginterations.org, a constantly updated, comprehensive and invaluable resource.
In conclusion, it is clear that the challenges facing HIV physicians managing older patients nowadays are different to those of thirty years ago. The first hurdle is to overcome barriers to making an HIV diagnosis in this population in the first place. This will require a normalisation of HIV testing and a move towards universal screening as well as ongoing education efforts targeting both physicians in contact with older patients and patients themselves. Secondly, it is clear that HIV infection seems to act synergistically with a number of age-related comorbidities including renal, bone and neuro-cognitive disease as well as a range of metabolic problems. To ensure optimal management of these patients, communication and two-way education between HIV physicians and physicians caring for older people, are key.
Conflict of interest: none declared
References available at www.gmjournal.co.uk