Abstract
HIV-1-infected patients are surviving longer and by 2015 half will be older than 50 years of age. Frailty is a syndrome associated with advanced age but occurs in HIV-1-infected patients at younger ages. One hundred outpatient HIV-1-infected persons were prospectively tested for clinical markers of frailty: shrinking weight, slowness in walking, decrease in grip strength, low activity, and exhaustion. Age, length of infection with HIV, CD4 count, HIV-1 RNA, and comorbidities were compared. CD4 counts <200 cells/mm3 were associated with 9-fold increased odds of frailty relative to patients with a CD4 count >350 cells/mm3 (odds ratio [OR] 9.0, 95% confidence interval [CI] 2.1-44). Seven frail patients were measured 6 months later: 2 died refusing therapy, 4 were no longer frail, and 1 patient remained frail. We conclude that frailty is common in HIV outpatients and is associated with low CD4 counts. However, our data suggest that frailty is transient, especially in younger patients who may revert to their prefrail state unlike uninfected elderly individuals in whom a stepwise decline in function occurs.
Introduction
The increasing age and comorbidities of HIV-1-infected patients may result in significant changes in their daily living activities. The mean age of HIV-1-infected individuals is increasing in the United States, and the Centers for Disease Control projects that more than half of HIV-1-infected individuals in the United States will be over 50 years by the year 2015. As this cohort ages, the consequences of the aging process coupled with chronic HIV-1 infection potentially place this group at increased risk of morbidity and mortality as a result of the aging process. Aging as well as HIV-1 infection have both been associated with similar immunologic and metabolic changes. 1 Some of the immunologic changes seen in aging and HIV are B-cell dysfunction, a decrease in the number of peripheral T-cells and gut-associated lymphoid tissue, and T-cell senescence, 1 while the metabolic changes seen in HIV are mostly related to antiretroviral therapy (ART) and the associated comorbidities. One aspect of aging, the newly described syndrome of frailty, may play an important role in aging of HIV-1-infected adults. 1,2 –5
Frailty has been defined in the elderly individuals as a biologic syndrome of decreased reserve and resistance to stressors, resulting from the cumulative decline of physiologic systems. In the elderly individuals, frailty pursues a stepwise decline in function over time. The clinical importance of frailty is that the syndrome is considered a high-risk state, predictive of adverse health outcomes such as decreased mobility, hospitalization, and death. 6 The relationship between frailty and chronic disease was poorly understood when frailty was first investigated. 7 Numerous studies in the past 10 years have tried to assess frailty in different populations and measure immunologic, hormonal, and clotting changes that occur with aging. Fried et al studied frailty in men and women older than 65 years who were enrolled in a cardiovascular study. 8 Their definition of frailty was validated in a study of aging women. 9 Modifications of their definition have been used in other studies including HIV-1-infected individuals. 2 –5 Fried et al described a frailty phenotype that even in the absence of disabilities or comorbidities showed that 7% of the population older than 65 years are frail, whereas 20% to 26% older than 80 years of age were frail. 8 Frailty can be a primary finding but also a secondary diagnosis as a result of an acute event or comorbidity like malignancy, atherosclerosis, infection, or depression. 10
Frailty has been found in HIV-1 infected patients at younger ages than non-HIV-infected patients. 2 Older HIV-1-infected individuals frequently present with more severe HIV disease and have a shorter survival time than younger individuals, often because they are not diagnosed until very late in the disease process. 11 Another reason may be that older patients have more comorbid diseases interacting with HIV-1. Older HIV-1-infected individuals have been described as frailer than age-matched control individuals without HIV-1 infection. 2 The introduction of a combination antiretroviral treatment (ART) has improved not only the immunity of patients but has increased the life expectancy of treated patients. However, the toxicity of ART drugs and/or a history of opportunistic infection/infections can add to the number of comorbidities of HIV-1-infected individuals.
The purpose of our study was to measure frailty and determine its relationship to aging in HIV-1-infected individuals. The clinical observation that led to this study was that patients (young and old alike) appeared frail when the CD4 counts were low, but as CD4 counts rose, the patients reverted to their prefrail state. This study assessed the extent of frailty in a group of HIV-1-infected clinic patients and compared the presence of frailty by age and CD4 counts. We also investigated the significance of comorbid diseases and the duration of ART for HIV-1 and their relationship to the syndrome of frailty, if any.
Methods
Study Population and Recruitment
A total of 100 patients with appointments with an HIV provider were assessed in an outpatient clinic from May to December 2010. Study participants were evaluated during 1 outpatient visit. HIV-1-infected patients from the University of Arizona Petersen Clinics aged 18 years or older including men and women were eligible for this study. Random patients were evaluated after obtaining their informed consent. The protocol for the study was approved by the Institutional Review Board of the University of Arizona. The clinic is in an academic setting and enrolls patients from all socioeconomic and education backgrounds. The patients belong to a mixed population characteristic of Southern Arizona, the majority being of white, non-Hispanic and white, Hispanic population groups. Approximately 70% of the patients are men who have sex with men, 80% of all patients are on ART, and ~90% of patients on ART have undetectable viral loads.
During a routine clinical appointment, clinical markers of frailty were measured (see below). Patients’ charts were reviewed for the following information: the most recent CD4 count and ultraquantitative HIV-1 RNA viral load, comorbid diseases, length of HIV treatment, and years of disease with HIV-1. Patients on ART as well as patients on no ART were included. All patients on ART were on a minimum of 3 antiretroviral drugs. The duration of ART was calculated from the start of ART, including mono and dual therapy. This information was retrieved from clinic records and patient recall.
We reassessed a subgroup of patients with CD4 counts <200 cells/mm3 at 6 months for frailty. Six months for a follow-up visit was chosen based on prior observations that reversion to the prefrail state usually occurred within several months of beginning ART.
Definition of Frailty
We used the definition of frailty as validated by Fried et al.
8
The definition consists of 5 characteristics of which at least 3 criteria need to be present in order for a patient to be considered frail. These are: Shrinking: measured unintentional weight loss of ≥10 pounds in the prior year; Slowness: decreased walking time as defined by a timed 15-foot walk test; time adjusted for gender and standing height. Men with a height of <173 cm and women with a height <159 cm who walked 15 feet in >7 seconds are considered frail; men >173 cm and women >159 cm who walked 15 feet in >6 seconds are considered frail; Weakness: decreased grip strength measured by a dynamometer; value adjusted for sex and body mass index (BMI). Men with a BMI Low physical activity level: a weighted score of kilocalories expended per week measured by the Minnesota Leisure Time Activity Questionnaire that asks about activities like daily living, sports, and hobbies. There is frailty when males use <383 kcal/week and females <270 kcal/week
12
; Exhaustion: self-reported by answering 2 questions from the Center for Epidemiologic Studies Depression scale.
13
The questions asked were how often in the last week did you feel: (a) that everything I did was an effort; or (b) I could not get going. Answers were 0 = less than 1 day, 1 = 1 to 2 days, 2 = 3 to 4 days, and 3 = most of the time. Answering 2 or 3 to either of these questions is a positive criterion for frailty.
Statistical Methods
We evaluated associations between frailty and clinical factors, stratified by age, 50 years and older versus younger than 50 years. Patients older than 50 years are considered older HIV-1-infected patients. 11 CD4 counts were categorized as less than 200 cells/mm3 (corresponding to the definition of AIDS), between 200 and 350 cells/mm3 and more than 350 cells/mm3. HIV-1 plasma RNA (viral load) was denoted as either detectable (≥48 copies/mL) or undetectable <48 copies/mL). We assessed associations between frailty and age, CD4 count, viral load, comorbidities, and length of ART utilizing the t test for continuous variables, and chi-square test for categorical variables. We assessed the association between frailty and CD4 counts using a 1 degree of freedom chi-square for trend. Comorbidities that were assessed included dyslipidemia, psychiatric diagnosis (depression, anxiety disorders, bipolar, and personality disorders as diagnosed by psychiatrist or psychologist), neuropathy, hypertension, hepatitis C (ongoing viremia without therapy), diabetes mellitus, coronary artery disease, malignancy, lung disease, chronic hepatitis B (as defined by prior measurement of HBV DNA and/or HBsAg in the serum), and coccidioidomycosis. Statistical significance was achieved when the P value was <.05. Mantel-Haenszel stratified analysis (producing odds ratios) was used to test the association between age and frailty, stratified by CD4 count. Multiple logistic regression analyses were used to analyze the effects of age, CD4 count, plasma viral load, and length of ART on the risk of frailty.
Results
The general characteristics of the study patients are shown in Table 1. There were 74 men and 26 women; 58 patients were younger than 50 years of age and 42 patients were older than 50 years of age. Ages ranged from 21 to 78 years. Most patients were white, non-Hispanic (89 patients), a minority were African-Americans (8), and there was 1 Native American, 1 Pacific Islander, and 1 Asian. Nineteen patients were judged to be frail on the basis of the 5 parameters used to measure frailty. The most common comorbidities were dyslipidemia (33 patients, 33%), psychiatric illness (31 patients, 31%), neuropathy (23 patients, 23%), hypertension (19 patients, 19%), chronic hepatitis C (18 patients, 18%), diabetes mellitus (9 patients, 9%), and coronary artery disease (6 patients, 6%). Other comorbidities were found in less than 5% of patients and included lung disease, malignancy, HIV wasting syndrome, chronic liver disease due to alcohol, chronic hepatitis B infection, avascular necrosis of the hip, and coccidioidomycosis. Among psychiatric diagnoses the most common were depression and anxiety disorders. However, frailty was significantly associated by age group in patients with 2 comorbidities, neuropathy and hepatitis C. Neuropathy was significantly associated with frail patients younger than 50 years of age (P < .044) but not in the older population. Hepatitis C was significantly associated with frailty in the older population (P < .023), but not with patients less than 50 years. The other comorbidities listed in Table 1 were not found to be significantly associated with frailty in either age group. It was found that frail patients younger than 50 years had more comorbidities related to opportunistic infections, while older patients had more comorbidities related to chronic diseases like dyslipidemia and hypertension.
Characteristics of the Study Group.
Table 2 shows the relationship between frailty and the CD4 count by age group. A low CD4 count was significantly associated with frailty in both age groups. Among patients with CD4 counts of <200 cells/mm3, we found a higher percentage (60%) of frail patients older than 50 years in comparison to patients younger than 50 years of age (39%). Similar findings occurred in the patients with CD4 counts >350 cells/mm3, 10% of patients older than 50 years in comparison to 4% of patients younger than 50 years. The prevalence of frailty was very high (43.5%) in patients of all ages with CD4 counts <200 cells/mm3. Frailty prevalence was 19.2% in patients with CD4 counts of 200 to 350 cells/mm3 and only 7.8% in patients of all ages with CD4 counts > 350 cells/mm3.
Frail Patients and the Relationship of Age and CD4 Count.
Stratified analyses showed no significant association between age and frailty, adjusted for CD4 count category, but age and CD4 count are strongly associated. CD4 counts <200 cells/mm3 were associated with a 9-fold increased odds of frailty relative to CD4 count >350 cells/mm3 (odds ratio [OR] 9.0, 95% confidence interval [CI] 2.1-44). CD4 counts of 200 to 350 cells/mm3 were not associated with frailty. No significant interaction was found between age, frailty, and the presence of a detectable plasma viral load.
Eighty-nine (89) patients were on ART, some for only several weeks, others as long as 24 years. Eleven (11) patients were antiretroviral drug naive. Figure 1 shows the relationship of the length of ART and frailty. Logistic regression models for the entire cohort showed that a longer duration of ART is protective for frailty even after controlling for CD4 count. The model included all patients (including those on no ART) and suggests that for every additional year on treatment, the OR for frailty decreases about 20%. The same logistic regression models were applied to patients on ART only (89 patients). The model suggests similar findings. For each additional year on ART, there was an 18% reduction in the OR for frailty.
Relationship of length of antiretroviral therapy and the presence or absence of frailty.
A small subgroup of 7 frail patients with CD4 counts lower than 200 cells/mm3 (Table 3) was assessed again for frailty within 6 months. Two patients eventually died, one refusing ART and the other stopping ART. Three patients had an improvement of the CD4 count to a level >200 cells/mm3. Only 1 of these 3 remained frail on retesting 6 months later. This patient was >50 years and had hepatitis C which is associated with frailty in the older age group and remained frail despite marked improvement in CD4 count. Two patients improved their CD4 count but not to >200 cells/mm3 and their viral loads were undetectable. They were not frail upon repeat measurements 6 months later. Thus, of the 5 surviving patients with <200 CD4 cells/mm3 and frailty, only 1 patient remained frail after adhering to an ART regimen.
Characteristics and Outcomes of 7 Frail Patients with Low CD4 Counts Who Began Antiretroviral Therapy and the Results of Retesting 6 Months Later.
Abbreviations: DVT, deep venous thrombosis; COPD, chronic obstructive lung disease.
Discussion
Two retrospective studies by Desquilbet et al assessed frailty in a cohort of men who have sex with men from the Multicenter AIDS Cohort Studies (MACS). Both studies used a shortened definition of frailty containing fewer criteria than did our study. The first study compared frailty in HIV-1-infected men in the pretreatment era to a control group of HIV uninfected men. 2 There were similar rates of frailty in HIV-positive men older than 55 years and HIV-negative men older than 65 years; frailty was found to occur earlier in HIV-1-infected men. Our study had similar findings of an earlier occurrence of frailty phenotype, but we obtained higher rates of frailty compared with MACS, possibly because of a more complex definition of frailty that we used but also because of the different population of patients in our study.
The second study by Desquilbet et al evaluated CD4 count and HIV viral load as predictors of frailty in HIV-positive men and found that lower CD4 counts and viral loads of more than 50 000 copies/mL of RNA were significantly associated with frailty. 3 Also the prevalence of frailty declined in the era of ART. Despite differences in measuring frailty and in population characteristics our study concluded like Desquilbet et al that a low CD4 count is significantly associated with frailty and that patients on long-term ART have less likelihood of developing frailty.
A study by Onen et al found that HIV was associated with premature frailty in an urban outpatient care setting. The investigators looked at characteristics that affected frailty such as comorbidities, CD4 count, and socioeconomic factors. 4 Our results had significant differences as well as similarities. For example, our patients were assessed for comorbidities, CD4 count, and length of HIV treatment. A premature prevalence of frailty, more comorbidities, and lower CD4 count were found in both studies, but we did not find a strong association between psychiatric diagnosis and frailty. We found a positive relationship between length of ART and not being frail.
The fourth published study on HIV and frailty was performed with women only by Terzian et al. 5 Frailty was assessed in HIV-1-infected women compared to non-HIV-infected women, and it was found that a low CD4 count (<100 cells/mm3) was associated with a higher prevalence of frailty.
All the aforementioned studies had slight modifications in the definition of frailty, and there were differences in the population characteristics as well. The common themes of all the studies were the premature occurrence of frailty in HIV-1-infected patients and the association of a low CD4 count with a higher frailty prevalence.
Our patient cohort consisted of a mixed population with more men than women of all ages above 18 years, but the number of patients (100) could be a limitation of the study. We looked specifically to determine whether there are any differences in frailty in HIV-1-infected populations younger than 50 years versus greater than 50 years of age and at factors that affect these 2 groups. Given the clinical significance of frailty, this additional study contributes to our knowledge of the relationship between HIV-1 infection and frailty.
Our initial hypothesis that age was not significantly important when measuring frailty of patients with low CD4 counts was confirmed. The frailty is likely more causally related to the profound immunosuppression found in the patients with low CD4 counts. Many of these patients had a history of recently treated opportunistic infections. Because of these observations we propose that an active diagnosis of AIDS (CD4 count <200 cells/mm3) is a significant comorbidity itself and highly predisposes patients to being frail.
All of our frail patients had at least 1 comorbidity besides HIV itself. Our frail patients <50 years had significantly less comorbidities than the frail population >50 years, but in comparison, the younger people had a lower CD4 count. We had a higher number of frail patients in the younger population than expected, but likely this was because of a greater number of younger patients with CD4 counts of <200 cells/mm3.
Our other hypothesis was that frailty may be temporary in younger patients with low CD4 counts and reverts when CD4 counts improve. We were limited by the low number of patients and this hypothesis could not be proven, but it was a likely explanation for the small number of patients in which this was demonstrated (Table 3). Longer ART was found to be protective for frailty (Figure 1). This fact on its own would support the recommendations of starting ART at higher CD4 counts and continuing ART without any treatment breaks. We believe that the main reason by which length of ART was shown to be protective for frailty is that patients on long-term treatment are more likely to have better control of comorbidities as well as HIV and less likely to be frail.
Conclusions
We have observed an association between low CD4 counts and frailty, which is not affected by age, viral load, or the presence of comorbidities. Effective treatment with ART plays a protective role against frailty, reinforcing the importance of effective ART. Early implementation of ART in the care of HIV patients may protect against frailty.
Footnotes
Acknowledgment
The authors thank Michael Lebowitz, MD, MPH, for his suggestions and review of the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.