| |
Frailty in people aging with human immunodeficiency virus (HIV) infection
|
| |
| |
Download the PDF here
Journal of Infectious Diseases Advance Access published June 5, 2014
"The increasing life spans of people with HIV reflect enormous treatment successes, and present new challenges related to aging. While some people with HIV live to older ages with relatively few health problems, others accumulate multiple problems earlier in life. Risk for HANA conditions and other adverse outcomes vary significantly between individuals, and are not fully explained by age, HIV disease severity, or duration of antiretroviral treatment and toxicity. With the accumulation of multiple health problems, it is likely that many people aging with HIV may be identified as frail"
"As frailty represents an integrative marker of health and vulnerability, and the severity of frailty can worsen or improve over time [47], more longitudinal research is needed. In particular, risk factors for frailty among HIV-positive individuals aging with high CD4 counts and undetectable viral loads have not been identified. This will be critical as this profile represents many HIV-positive persons currently ageing successfully with treated HIV infection [1-3,9]."
"Two longitudinal analyses of frailty in people with HIV have been published, both from the MACS cohort. One report included data from before 2007 [17], and the second data from 2007-2011 [13]. Each report used a different modification of the frailty phenotype scale (Table 1), which complicates comparisons between the two time periods. In both studies, likelihood of presenting as frail at a later study visit was associated with lower CD4 count and no greater than high school education [13,17]. Some risk factors for frailty identified in pre-2007 data were not replicated the second analysis (e.g. white, non-Hispanic ethnicity [17]), and other risk factors were assessed in only one study. In pre-2007 data, the association between frailty and low CD4 count was identified independently of low viral load (<400 copies/mL) and hepatitis B and C coinfection.
Participants with high viral load (>50,000 copies/mL) were also significantly more likely to become frail [17]. In the analysis of data from 2007-2011, participants with detectable viral loads were not more likely to become frail than those with undetectable viral loads, but participants with depressive symptoms, diabetes mellitus, and kidney disease were more likely to become frail [13]. Also in the 2007-2011 data, HIV-positive participants with a history of AIDS had higher odds of becoming frail than HIV-negative participants, while HIV-positive participants without history of AIDS did not have higher odds [13]."
Relationship Between a Frailty-Related Phenotype and Progressive Deterioration of the Immune System in HIV-Infected Men: CD4 predicted frailty in MACS, viral load associated.......http://www.natap.org/2012/HIV/101012_05.htm
"Characteristics of frailty and opportunities for intervention should be investigated among the increasing proportion of treated HIV-positive individuals who demonstrate high CD4 counts and undetectable viral loads"
'The clinical importance of frailty is often noted as its ability to describe individuals more vulnerable to adverse health outcomes [7]. To date, knowledge is limited regarding the prognostic characteristics of frailty in people with HIV. In one sample of people who inject drugs, having HIV or being frail was associated with three-fold higher likelihood of death, while both having HIV and being frail increased the risk seven-fold compared to those with neither [32]. In the MACS study, the presence of frailty prior to HAART initiation decreased time to AIDS or death [30]. The prevalence of frailty at baseline was 8%; 36% of people who frail at baseline developed AIDs or died, while 16% of people who were not frail developed AIDS or died [30]."
"screening for frailty among patients with complex needs has been found to be both feasible and useful in primary care settings [7]. Future studies should investigate comprehensive assessments and frailty screening in the delivery of care to people aging with HIV."
"People aging while receiving HAART are also at high risk of polypharmacy and related adverse outcomes, and people who are frail are likely most vulnerable [50]. Better understanding of optimal prescribing for frail patients on HAART is needed."
--------------------
Frailty in people aging with human immunodeficiency virus (HIV) infection
Journal of Infectious Diseases Advance Access published June 5, 2014
Thomas D. Brothers1, Susan Kirkland2, Giovanni Guaraldi3, Julian Falutz4, Olga Theou1,
B. Lynn Johnston5, Kenneth Rockwood6
1Geriatric Medicine Research, Dalhousie University, Halifax, Nova Scotia, Canada
2Departments of Community Health & Epidemiology and Medicine, Dalhousie University,
Halifax, Nova Scotia, Canada
3Department of Medical and Surgical Sciences for Adults and Children, Clinic of Infectious
Diseases, University of Modena and Reggio Emilia, Modena, Italy
4Chronic Viral Illness Service, Division of Infectious Diseases and Division of Geriatrics,
McGill University Hospital Center, Montreal, Quebec, Canada
5Division of Infectious Diseases, Dalhousie University, Halifax, Nova Scotia, Canada
6Division of Geriatric Medicine, Dalhousie University, and Centre for Health Care of the Elderly,
Capital District Health Authority, Halifax, Nova Scotia, Canada
Abstract
The increasing life spans of people with HIV reflect enormous treatment successes and present new challenges related to aging. Even with suppression of viral loads and immune reconstitution, HIV-positive individuals exhibit excess vulnerability to multiple health problems that are not AIDS-defining. With the accumulation of multiple health problems, it is likely that many people aging with treated HIV infection may be identified as frail. Studies of frailty in people with HIV are currently limited, but suggest that frailty might be feasible and useful as an integrative marker of multisystem vulnerability, for organizing care and for comprehensively measuring the impact of illness and treatment on overall health status. This review explains how frailty has been conceptualized and measured in the general population, critically reviews emerging data on frailty in people with HIV infection, and explores how the concept of frailty might inform HIV research and care.
Aging with HIV infection
Effective therapies have transformed human immunodeficiency virus (HIV) infection into a chronic illness [1]. As people with HIV live longer, aging-related challenges are arising. Despite complete suppression of viral load and immune recovery, HIV-positive individuals are more vulnerable to poor health than HIV-negative individuals [2]. This vulnerability is characterized by higher risk of several common, age-related health problems, even after adjustment for established risk factors. These conditions, termed HIV-associated non-AIDS (HANA), include cardiovascular disease, osteoporosis, metabolic disorders, hepatic and renal diseases, and some cancers, as well as age-associated immunologic changes and chronic inflammation [3-5]. Each involve different physiological systems and etiologies, yet are all strongly age-associated in the general population. While HANA conditions are more common among HIV-positive individuals who are older, have more severe HIV disease, and who have longer duration of antiretroviral treatment and toxicity, these factors do not completely explain differences in risk and survival [1,4,6].
Among people without HIV, aging and the accumulation of age-related health problems are also highly heterogeneous processes. While people generally accumulate health problems with age, individuals of the same age can experience very different levels of health. Geriatricians introduced the term "frailty" to describe this variability. Frailty represents the cumulative effects of age-related deterioration in multiple physiological systems and homeostatic mechanisms, resulting in greater vulnerability to stressors [7,8]. Frail individuals often present with nonspecific health complaints, fluctuating disability, falls, and delirium, and are at higher risk for multiple adverse outcomes, including longer hospital stays, postoperative complications, poor responses to vaccination, functional decline, and death [7].
With the accumulation of multiple health problems, it is likely that many people aging with HIV may be identified as frail [9]. The concept of frailty could provide a useful tool to measure and communicate the complexity of aging and vulnerability in people living with HIV, inform the development of therapies, and guide the delivery of care. This review explains how frailty has been conceptualized and measured in the general population, critically reviews emerging data on frailty in people living with HIV, and explores how applying the concept of frailty to research and care might benefit people living with HIV.
Frailty in the context of HIV
Investigators have begun investigating frailty among people with HIV to identify individuals more vulnerable to disease progression and death, and to measure the effects of illness and treatment on health status [1,10,11]. However frailty is not yet well understood in the context of the HAART era, where most HIV-positive individuals now experience significant immune recovery, undetectable HIV viral load, and primarily HANA clinical manifestations [1,2]. Neither CD4 count nor viral load appear to be useful surrogate markers of vulnerability in this immune-reconstituted population, while frailty is strongly associated with HANA conditions and disability [12,13], and might be a more sensitive indicator of health changes [6,9,14]. Age-related and HANA conditions have been associated with both immune activation (e.g. soluble CD14 and CD163, CD16+ monocytes, HLA-DR+/CD38+ CD8+ T cells) and immune senescence markers (e.g. terminally differentiated CD45RA+CCR7- CD4+ T cells), as well as inflammatory circulating cytokines (e.g. interleukin-6, TNF-a) [1,9,15,16]. Frailty is associated with both CD4 count and viral load [17], yet relationships between frailty and markers of immune senescence and activation among HIV-positive individuals have not been established. While the clinical spectrum of HIV disease differs whether individuals experience immune deficiency or immune activation, frailty might emerge in the context of both profiles. A hypothetical representation of the association between frailty, HANA and immune system dysregulation is depicted in Figure 1. Causal pathways between these factors are not yet understood, in part because most studies investigating HANA or frailty in HIV have been cross-sectional.
Defining frailty
While "frail" is commonly used to describe vulnerable older adults, there is no consensus on the best way to define and identify frailty systematically [7]. Two conceptual models inform most approaches to frailty: the phenotype model and the cumulative deficit model [7]. The phenotype views frailty as a clinical syndrome arising from a "cycle of frailty" comprised of chronic undernutriton, sarcopenia, and weakened strength and exercise tolerance. It suggests that frailty pathophysiology is distinct from aging or other disease processes [19]. Other factors, such as cognitive impairment, have been suggested as further phenotypic characteristics of frailty [7].
The cumulative deficit model (first proposed by members of our group) views frailty as a state of vulnerability, rather than a syndrome. It suggests that frailty arises from the cumulative effects of non-specific age-related health deficits and does not have a unique pathophysiology, but rather is related to the aging process [20]. As people accumulate health deficits and homeostatic mechanisms begin to fail, those who are frail exhibit excessive changes in health in response to even minor further insults [7]. Under this model, frailty has been proposed to describe the overall health state of an individual, and therefore serve as an integrative marker of biologic aging, as opposed to chronological age [8,9]. Studies applying both frailty models have identified associations between increasing severity of frailty and age-related deterioration in multiple systems, including immunosenescence and chronic inflammation [21,22], which may be particularly relevant in people with treated HIV [4,5,15,16].
Measuring frailty
Multiple measures exist to identify and measure frailty. Some are based on clinical judgment or a single item (e.g. walking speed), but most scales assess multiple domains of age related health and grade frailty by counting the number of deficits individuals have acquired [23]. One commonly used scale, based on the frailty phenotype [24], identifies frailty by the presence of three deficits out of five specific measures originating from the Cardiovascular Health Study (an existing prospective cohort study): self-reported unintentional weight loss >10lbs or recorded weight loss ≥5% in a year, measured slow walking speed, measured weak grip strength, self-reported exhaustion (3-4 days per week or most of the time), and low activity/energy expenditure (assessed by Minnesota Leisure Time Questionnaire) [25]. The frailty phenotype scale has been widely applied and extensively validated in its ability to identify people at increased risk for a range of adverse outcomes [7,19].
Another commonly used scale, the "frailty index", counts the number of deficits individuals have accumulated out of various health measures and presents them as a proportion [20, 26]. In contrast to the phenotypic approach, any measure can be included in a frailty index if it is generally related to age and poor health, and if the group of items covers multiple physiological systems. When at least 30 items are included, the proportion of deficits accumulated appears more informative than the specific nature of those deficits. Though the effect of each individual deficit may be small, their cumulative effects can be large. This reinforces the notion that health problems in the same individual rarely arise independently from one another [7,8,26]. Each frailty index can make use of different available measures, including functional limitations, co-morbidities, cognition and affect [26]. This approach has been operationalized clinically using data from comprehensive geriatric assessments and routine medical records [7].
Many other frailty scales exist, often including more items than the five specified by the frailty phenotype but fewer than the 30 suggested by the frailty index [23]. By counting health deficits across multiple physiologic systems, frailty scales are each able to identify individuals vulnerable to adverse outcomes, and to do so better than chronological age alone [7,23]. While scales differ in the number and nature of deficits they count, people who have accumulated more deficits are more likely to be vulnerable, and therefore more likely to be frail [27]. Different scales also demonstrate remarkable consistency in characteristics, including the nonlinear relationship between frailty severity and age, greater frailty in women than same-aged men, and higher risk of death in men than women of equal frailty [27].
However, as they include different criteria, frailty scales vary in ability to predict outcomes and in operational feasibility in different settings [23]. Frailty scales that include more measures can more sensitively grade vulnerability and track improvement and decline, and are less likely to overlook individuals who have accumulated diverse deficits in health; they might, however, be relatively cumbersome to construct [23,28]. Parsimonious scales can be quicker to apply, but often require specific measures (e.g. grip strength measured by dynamometer) and might overlook people with different health problems. Modifications to such scales are common, especially replacing performance-based measures (e.g. walking speed) with self-reported measures (e.g. reported difficulty walking), or using different criteria for performance-based measures (e.g. loss of >10lbs in past year vs. loss of >5% of body weight in past six months), yet the validity of such modifications are unknown [23].
Measuring frailty in HIV-positive individuals
All published studies of frailty in HIV infection use frailty scales comprised of a limited number of specific health measures, following the phenotype approach (Table 1). For instance, analyses of the Multicenter AIDS Cohort Study (MACS) used a frailty scale based on four self reported deficits: weight loss, exhaustion, impaired physical activity, and difficulty walking [29]. One study used a single measure of unexpected weight loss to define frailty [40]. No published studies of frailty in people with HIV have used the cumulative deficit/frailty index approach, or scales based on clinical judgment.
A recently introduced measure of health status in people aging with treated HIV, the Veterans Aging Cohort Study (VACS) index, has also been proposed to measure frailty [41]. The VACS index is a prognostic tool made up of both traditional HIV-related factors, including CD4 count and viral load, as well as hepatitis C co-infection, liver fibrosis (FIB-4), hemoglobin, estimated glomerular filtration rate (eGFR), race, and age. Investigators have considered adding measures to the index, including inflammatory markers D-dimer and soluble CD14 [42]. As the VACS index is a measure of multisystem deterioration and vulnerability, we included it as a frailty scale. However, the VACS index differs from other frailty measures as it was designed to predict mortality, and includes chronological age and race [43]. Most frailty scales do not include age, as they intend to describe biological age-related changes independent from chronological age, and most do not include race, because they instead incorporate markers of individual physical and mental health.
Further work is needed to determine the best approach to measure frailty in people aging with HIV. It is important to consider the intended use and setting for a frailty scale, whether as a brief screening tool or as a comprehensive evaluation, for use in the community, hospital, or long-term care. Some scales that have been used to identify frailty in people with HIV might not be appropriate for those who are very frail or immobile, as they include measures of physical performance (e.g. walking speed [12,31-35]) or apply exclusion criteria based on disability or comorbidities. One study using a modified version of the frailty phenotype scale in an HIV clinic excluded 19% of participants because time constraints prevented assessment of grip strength and walking speed [12], and another excluded participants requiring an assistive device to walk [38].
Epidemiology of frailty in HIV infection
Before the introduction of HAART in 1996, men in the MACS study who seroconverted were nine times more likely to be identified as frail (via a modified frailty phenotype) during at least one study visit than men who remained uninfected (13.9% vs. 1.5% prevalence) [29]. Risk for frailty increased nonlinearly with age and with duration of HIV infection [29]. Frailty was also associated with CD4 count <350 cells/mm3, viral load ≥50,000 copies/mL, and AIDS [29].
With the introduction of HAART, the prevalence of frailty appeared to decrease. Among MACS participants, frailty decreased from 8% in 1994-1995, when <0.1% of participants received HAART, to 5% in 2000-2005, when almost 70% were on HAART [17]. Among participants presenting with AIDS, frailty prevalence decreased from 24% to 10% [17]. However, from 2007-2011, when grip strength was added to the MACS frailty scale, 25% of all participants were identified as frail during at least one study visit [13]. Here the use of different scales complicates comparison of estimates between studies [7,23].
Among individuals on HAART, multiple factors have been associated with frailty in cross-sectional studies, using different frailty scales (Table 2). Some are traditional HIV measures, including lower current CD4 cell count (measured continuously [12,13] and categorically, as <500 cells/mm3 [34], <350 cells/mm3 [32], <200 cells/mm3 [33], and <100 cells/mm3 [31]), lower nadir CD4 count [12], CD4/CD8 ratio ²0.29 [31], detectable viral load [13, 32], history of AIDS [13], and longer time since diagnosis [12], as well as hepatitis C coinfection [33], low body mass index (BMI) [12, 34], high BMI [38], lipodystrophy [38], depressive symptoms [12, 32], one-year history of multiple falls [36], and lower cognitive performance [12]. HIV-positive individuals who are frail are also more likely to have lower socioeconomic status, no more than high school education [12,32], current unemployment [12,35], and income <$10,000 in the prior year [12]. Among people who inject drugs, those with advanced HIV disease (defined as CD4 <350 cells/mm3 and detectable viral load) are more likely to be frail than uninfected individuals, whereas those without advanced HIV disease are not more likely to be frail [32]. Frail HIV-positive individuals are also more likely to have been on HAART for longer duration [13] and on a protease inhibitor-containing HAART regimen, and less likely to be on a non-nucleoside reverse transcriptase inhibitor-containing regimen; this disparity is not explained by differences in adherence or successful viral suppression [12]. Frail HIV-positive individuals are also more likely than the non-frail to have been hospitalized in the past year and to have longer hospital stays [12].
Also in cross-sectional studies, markers of inflammation (interleukin-6, D-dimer, and soluble CD14) are more strongly correlated with VACS index scores than an index comprised only of age, CD4 cell count and viral load [6]. VACS index scores are also associated with upper and lower extremity strength [46] and cognitive impairment [45]. While VACS index scores were suggestive of an association with one-year history of multiple falls in one study, this was not statistically significant [36]. As falls are a common outcome identified among frail HIV-negative older adults [7], further research is needed to assess whether the VACS index is measuring frailty or a different but related construct, including some common components.
Two longitudinal analyses of frailty in people with HIV have been published, both from the MACS cohort. One report included data from before 2007 [17], and the second data from 2007-2011 [13]. Each report used a different modification of the frailty phenotype scale (Table 1), which complicates comparisons between the two time periods. In both studies, likelihood of presenting as frail at a later study visit was associated with lower CD4 count and no greater than high school education [13,17]. Some risk factors for frailty identified in pre-2007 data were not replicated the second analysis (e.g. white, non-Hispanic ethnicity [17]), and other risk factors were assessed in only one study. In pre-2007 data, the association between frailty and low CD4 count was identified independently of low viral load (<400 copies/mL) and hepatitis B and C coinfection. Participants with high viral load (>50,000 copies/mL) were also significantly more likely to become frail [17]. In the analysis of data from 2007-2011, participants with detectable viral loads were not more likely to become frail than those with undetectable viral loads, but participants with depressive symptoms, diabetes mellitus, and kidney disease were more likely to become frail [13]. Also in the 2007-2011 data, HIV-positive participants with a history of AIDS had higher odds of becoming frail than HIV-negative participants, while HIV-positive participants without history of AIDS did not have higher odds [13].
As frailty represents an integrative marker of health and vulnerability, and the severity of frailty can worsen or improve over time [47], more longitudinal research is needed. In particular, risk factors for frailty among HIV-positive individuals aging with high CD4 counts and undetectable viral loads have not been identified. This will be critical as this profile represents many HIV-positive persons currently ageing successfully with treated HIV infection [1-3,9].
Frailty and health outcomes in HIV infection
The clinical importance of frailty is often noted as its ability to describe individuals more vulnerable to adverse health outcomes [7]. To date, knowledge is limited regarding the prognostic characteristics of frailty in people with HIV. In one sample of people who inject drugs, having HIV or being frail was associated with three-fold higher likelihood of death, while both having HIV and being frail increased the risk seven-fold compared to those with neither [32]. In the MACS study, the presence of frailty prior to HAART initiation decreased time to AIDS or death [30]. The prevalence of frailty at baseline was 8%; 36% of people who frail at baseline developed AIDs or died, while 16% of people who were not frail developed AIDS or died [30].
While assessments of outcomes related to frailty in people with HIV are limited, multiple prospective studies have evaluated outcomes in relation to the VACS index. Higher VACS index scores are associated with all-cause mortality [6,43], coronary heart disease-related mortality [48], and fragility fractures, suggesting that the index might indeed measure frailty as well [41]. Compared to CD4 count and viral load, VACS index scores had better predictive ability for mortality among HIV-positive individuals with viral load <500 copies/mL and those age ≥50 years [42,43].
Future directions: frailty and HIV care
While early data have identified the feasibility and usefulness of measuring frailty in people aging with HIV, the implications of incorporating frailty concepts into HIV care are unknown. The ultimate question will be whether recognizing frailty assists in the clinical management of patients with HIV who are frail. Even when immunologically stable, people with HIV accumulate a variety of health problems, and each individual problem likely cannot characterize overall vulnerability. As people with HIV live longer, many will survive to such an age that they might be frail in spite of - not because of - the disease. Models of care need to adapt to this changing paradigm, and principles of frailty management may be useful [49]. A challenge in the management of any patient with complex needs is that many clinical interventions are intended to help people with only one problem, and such interventions can do harm in people who have many problems [7]. Interdisciplinary assessment and care can improve clinical outcomes for people who are frail, and screening for frailty among patients with complex needs has been found to be both feasible and useful in primary care settings [7]. Future studies should investigate comprehensive assessments and frailty screening in the delivery of care to people aging with HIV.
Healthy aging with HIV may be promoted by early interventions among those who are at risk for becoming frail. As frailty is associated with lower CD4 count, and risk appears to decline once individuals begin HAART [17,33], early antiretroviral treatment might delay or reduce the severity of frailty. In longitudinal studies, some older HIV-negative adults show improvement in frailty status over time, and not simply progressive decline [47]. Frailty might be an especially dynamic process in people with HIV, particularly in younger people with greater physiologic reserve and greater opportunity to improve [32-34]. However contributions of long-term antiretroviral treatment and toxicity to frailty are unknown. Characteristics of frailty and opportunities for intervention should be investigated among the increasing proportion of treated HIV-positive individuals who demonstrate high CD4 counts and undetectable viral loads.
Evidence is also unavailable regarding effective interventions for HIV-positive people who are already frail. Much of the evidence for the care of frail HIV-positive people is necessarily based on trials performed on younger and fitter people. While some medical interventions developed in fit populations are less effective, or even dangerous, in people who are frail, others can continue to have important benefits. While some treatments provide smaller risk reductions in people who are frail, the high absolute risk for poor outcomes with frailty might make this smaller benefit worthwhile [7]. People aging while receiving HAART are also at high risk of polypharmacy and related adverse outcomes, and people who are frail are likely most vulnerable [50]. Better understanding of optimal prescribing for frail patients on HAART is needed.
Conclusion
The increasing life spans of people with HIV reflect enormous treatment successes, and present new challenges related to aging. While some people with HIV live to older ages with relatively few health problems, others accumulate multiple problems earlier in life. Risk for HANA conditions and other adverse outcomes vary significantly between individuals, and are not fully explained by age, HIV disease severity, or duration of antiretroviral treatment and toxicity. With the accumulation of multiple health problems, it is likely that many people aging with HIV may be identified as frail. Emerging data suggest frailty might be a feasible and useful integrative marker of multisystem vulnerability in people aging with HIV. As people with HIV live longer and with more complex health and social care needs, the concept of frailty could be useful for identifying vulnerable individuals, for organizing care and for comprehensively measuring the impact of illness and treatment on overall health status.
|
|
| |
| |
|
|
|
