http://www.medscape.com/viewarticle/550458
Posted 02/12/2007
G Orlando; P Meraviglia; L Cordier; L Meroni; S Landonio; R Giorgi; M Fasolo; I Faggion; A Riva; A Zambelli; R Beretta; G Gubertini; G Dedivitiis; G Jacchetti; A Cargnel
Abstract
Background: The availability of several therapeutic regimens has transformed HIV infection from a life-threatening disease into a chronic condition. Older patients (>50 years old) with HIV infection constitute a new treatment challenge in terms of the cumulative effects of ageing and antiretroviral therapy (ART).
Methods: The immunovirological effects and metabolic interactions of 48 weeks of ART in older patients followed up in three Infectious Diseases Units in Milan, Italy since 1994 were compared with those in younger controls aged 25-35 years.
Results: The 159 older patients and 118 controls enrolled in the study were comparable for HIV stage, baseline CD4 cell count and viral load but differed for mode of HIV transmission, comorbid conditions and related chronic treatments. Mean viral load decreased after 48 weeks of treatment by 2.6 log10 HIV RNA copies/mL and CD4 count increased by 137.5 cells/μL in older patients, and similar values for immunovirological effects were obtained in the young controls. The relative risk (RR) of an abnormal test in older patients was 7.33 [95% confidence interval (CI) 4.36-12.36] for glucose, 1.73 (95% CI 1.45-2.07) for total cholesterol, 1.56 (95% CI 1.22-2.0) for high-density lipoprotein cholesterol, 1.26 (95% CI 1.02-1.56) for triglycerides, 6.48 (95% CI 4.36-9.66) for serum creatinine, and 0.45 (95% CI 0.35-0.58) for ALT. Moderate/severe liver and renal toxicities were recorded in the older patients but not in the controls. The tolerability of ART did not differ between the older patients and the controls. Thirty-nine new cardiovascular, endocrine-metabolic and neuralgic disorders (24.52 per 100 person-years) were diagnosed in the older patients and four (3.39 per 100 person-years) in the controls (P<0.0001).>Conclusions: Diseases induced by, or related to, the toxic effects of antiretrovirals interact with age-specific health profiles, raising new questions and challenges. Comparative epidemiological studies, research studies addressing specific questions and surveillance are needed to answer the questions that arise in clinical monitoring.
Introduction
In recent years, the clinical history of HIV infection has been modified by three main factors: (i) the widespread use of highly active antiretroviral therapy (HAART), which has reduced mortality; (ii) new therapeutic strategies, which have transformed HIV infection from a life-threatening disease into a chronic condition, and (iii) the identification of risk factors for side effects and untoward effects of each antiretroviral drug, which has allowed treatment to be tailored to each patient.
As a consequence, a new variable must be considered by healthcare professionals: the ageing of HIV-infected people. HIV-infected adults older than 50 years represent more than 10% of the HIV-infected population and 15% of all people living with AIDS. These percentages are expected to increase with time. Individuals with HIV/AIDS are living longer following the introduction of HAART, and the mean age of HIV-infected people has increased since the beginning of the epidemic in industrialized countries.[1]
Among patients with AIDS in Italy, the median age has increased from 29 years for male patients and 24 years for female patients in 1985 to 41 and 38 years, respectively, in 2004. The overall rate of AIDS diagnosis among patients older than 50 years has increased from 5.3% to 10.2% in 10 years (1990-2001).[2]
More than 50% of patients on HAART are expected to live beyond their 60s. Reliable epidemiological estimates are lacking for elderly patients, and little is known about the interactions between ageing and HIV infection, both in the field of disease progression and in the field of antiretroviral treatment effectiveness, tolerability and short- and long-term toxicity, the pharmacokinetics of antiretroviral drugs, and interactions between HAART and underlying diseases and their treatments.
Older age has been associated with faster progression of HIV infection and shorter survival time after the diagnosis of AIDS.[3-8] Randomized, controlled clinical trials for the evaluation of antiretroviral drugs or therapeutic strategies generally exclude older patients and/or those with concurrent disorders. No recommendations about the most appropriate timing and type of antiretroviral treatment are given in the international guidelines for the use of antiretroviral agents in older HIV-1-infected patients.[9] Scientific data, based on case reports or limited studies, do not help healthcare professionals to properly treat older HIV-infected individuals.
Physiological changes observed with ageing, including increased risk of infection, reduced immunocompetence, the appearance of several comorbid conditions which can affect the disease process and complicate its management, and interactions among antiretrovirals and drugs used for the treatment of other diseases, underline the need for age-related evaluations of treatment and management strategies.
In this study, we evaluated immunovirological outcomes and the incidence rate of adverse metabolic events or new comorbidities in the first year of antiretroviral treatment in an HIV-infected cohort of patients aged 50 years or older, followed up in three Infectious Diseases Units of the L Sacco Hospital, Milan, Italy.
Patients and Methods
This was a longitudinal, historical, cohort study in which all HIV-infected patients followed up in three Infectious Diseases Units of L Sacco Hospital (Milan, Italy) who started their first antiretroviral treatment after they had reached 50 years of age and who had completed at least 48 weeks of antiretroviral treatment were included in the analysis. A control group was identified among randomly selected HIV-infected out-patients who started their first antiretroviral treatment when they were aged between 25 and 35 years. Demographic, anamnestic, clinical, immunovirological and biochemical data, including information on all non-HIV-related ongoing treatments at the time of inclusion in the study, were collected. Effects of antiretroviral treatment on immunovirological parameters and on liver and kidney function tests, lipid and glucose metabolism, and body weight were recorded at 4, 8, 12, 24, 36 and 48 weeks from the beginning of antiretroviral treatment; any changes in antiretroviral treatment and the causes of such changes were also recorded.
Comorbid conditions to be included in our analysis were identified on the basis of the most common interactions with HIV infection or antiretrovirals described in the literature; drugs for comorbid conditions were selected using the same criterion. We included: cardiovascular disorders (arterial hypertension requiring pharmacological treatment; ischemic cardiovascular diseases), endocrine-metabolic disorders (diabetes treated with insulin or oral hypoglycaemic drugs; hyperlipidaemia; thyroid dysfunction), neuropsychiatric disorders (peripheral neuropathy; epilepsy; any psychiatric disorder needing pharmacological continuative treatment), neoplastic diseases (any neoplastic disease unrelated to HIV infection), hepatic diseases (severe chronic viral or dismetabolic liver disease) and renal dysfunction (chronic renal insufficiency with creatinine clearance≤50 mL/min). Non-AIDS-related drugs included in the analysis were: cardiovascular (calcium channel blockers, antihypertensive drugs, antiarrhythmics, ergotamine derivatives and erectile dysfunction agents), endocrine-metabolic (lipid-lowering agents, hormone substitutive therapy, insulin, oral hypoglycaemic drugs and corticosteroids), gastrointestinal (proton pump inhibitors and H2 blockers) and neuropsychiatric (neuroleptic, psychotropic and antidepressant) drugs.
Statistical analyses were performed using GraphPad Prism version 4.00 for Windows (GraphPad Software, San Diego, CA, USA). Categorical data were analysed by the χ2 test or Fisher's exact test, and continuous data were analysed by t-tests or the nonparametric Mann-Whitney test.
Results
A total of 159 patients who started any antiretroviral regimen after the age of 50 years and 118 younger controls met the inclusion criteria and were included in the analysis. At baseline evaluation, the older patients and control patients were comparable for Centers for Disease Control and Prevention (CDC) stage of HIV infection, CD4 cell count, viral load and percentage of HAART vs non-HAART antiretroviral regimens ( Table 1 ). Older patients and controls differed in sociodemographic and epidemiological variables; there was a greater preponderance of male patients in the older group (male to female ratio 4.5:1 vs 1.4:1 in older patients and controls, respectively; P<0.0001);>
AIDS presenters (defined as patients who seek medical attention when an AIDS defining condition is already present) were similarly distributed between the older patients and the controls (20.12% and 20.33%, respectively). The late presenters (defined as patients who needed antiretroviral treatment within 6 months of their first HIV-positive test) constituted 55.35% of older patients and 36.44% of the controls (P=0.0023). The median time between the first HIV-positive test and treatment was 4.5 months [interquartile range (IQR) 1.5-27.5] for older patients and 12 months (IQR 2.5-64.3) for controls (Mann-Whitney test >P=0.0049).
At baseline evaluation, no differences in the rate of comorbid conditions were identified between older patients and controls: 59 comorbidities were found in 52 older patients (32.7%; rate 1.13/patient) and 32 in 30 controls (25.42%; rate 1.06/patient). However, cardiovascular and endocrine-metabolic disorders were significantly more frequent in the older group, while liver diseases were more common in the younger controls ( Table 2 ).
Long-term treatments for these conditions, including several drugs that could potentially interfere with antiretroviral treatments, as defined in the Methods section, were prescribed for 42 older patients and six controls [P<0.0001; onclick="resizeWin('Tables',500,650)" target="Tables" href="http://www.medscape.com/viewarticle/550458_Tables#T2">Table 2 ).
A total of 53 opportunistic infections or AIDS-defining conditions (rate 1.15/patient) in older patients and 32 (rate 1.14/patient) in controls were reported in clinical records. Pneumocystis carinii pneumonia (PCP) was the most frequent opportunistic infection in both groups; no differences in opportunistic infections were observed between the two groups, although a higher incidence of AIDS-related neoplastic diseases (Kaposi's sarcoma and non-Hodgkin's lymphoma) was reported in the older patients ( Table 3 ).
Immunovirological Effects of the Treatments
Ninety-five older patients (59.7%) and 78 controls (66.1%) had been treated with several HAART regimens, while the remaining older patients and controls, most of whom started their first antiretroviral treatment before 1996, had received one or two antiretroviral drugs (non-HAART). Mean HIV RNA values decreased in 48 weeks of treatment by 2.6 log10 HIV-1 RNA copies/mL both in older patients and in controls (from 4.91±0.74 to 2.31±0.8 log1010 copies/μL, respectively) (Fig. 1). Viral suppression, defined as HIV RNA <2.7>10 copies/mL to include patients treated in the early 1990s when the test cut-off was 500 copies/mL, was achieved in 88.3% of older patients and in 74.04% of controls after 8 weeks of treatment, and in 78.2% and in 78.8% of patients at 48 weeks, respectively (Fig. 1). copies/μL and from 4.83±2.22 to 2.22±0.83 log
 | Figure 1. (click image to zoom) Immunovirological parameters in 159 older HIV-infected patients and 118 younger HIV-infected controls during 48 weeks of antiretroviral treatment. (a) Mean viral loads (log10 copies/mL); (b) mean CD4 counts (cells/µL) and (c) percentages of older patients and controls with suppressed viraemia. |
The median baseline absolute CD4 count at the time of enrolment was 202 cells/μL (IQR 76.5-338 cells/μL) for older patients and 188 cells/μL (IQR 69-294 cells/μL) in controls. The mean CD4 count increased from 230.3±176.6 to 367.8±198.5 cells/μL in 48 weeks of treatment. In controls, the mean CD4 cell count increased from 199.4±145.9 to 391.2±213.1 cells/μL. The CD4 cell count was significantly higher in older patients than in controls at week 12 (Mann-Whitney test P=0.049) but thereafter no significant differences in mean CD4 values were observed.
Biochemical Changes
The percentage of patients with abnormal biochemical tests [plasma glucose >6.1 mmol/L, total cholesterol >4.9 mmol/L, high-density lipoprotein (HDL) cholesterol <1.04>2.1 mmol/L, serum creatinine >105.6 mmol/L] was higher in older patients than in controls during the period considered, while the percentage of patients with alamine aminotransferase (ALT) levels >50 IU/L was higher in controls (Fig. 2).
 | Figure 2. (click image to zoom) Percentage of older patients (P) and controls (C) with abnormal biochemistry values during 48 weeks of antiretroviral treatment. ALT, alamine aminotransferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein. |
The relative risk of abnormal tests during treatment was high for fasting glucose (RR 7.33, 95% CI 4.36-12.36), serum creatinine (RR 6.48, 95% CI 4.36-9.66), total cholesterol (RR 1.73, 95% CI 1.45-2.07) HDL cholesterol (RR 1.56, 95% CI 1.22-2.0) and triglycerides (RR 1.26, 95% CI 1.02-1.56), and low for ALT (RR 0.45, 95% CI 0.35-0.58).
During the 48 weeks of follow-up, episodes of increased ALT levels at liver toxicity grades 1-4 were recorded more frequently among controls than among older patients in whom, however, more severe toxicity episodes with ALT>500 IU/L were recorded. Grade 1 toxicity was recorded in 74 controls and 25 older patients, grade 2 in 22 controls and seven older patients, grade 3 in two controls and five older patients, and grade 4 in no controls and four older patients.
Moderate to severe renal toxicity episodes (creatinine >132.6 and >176.8 mmol/L) were recorded in 21 and two older patients, respectively. No abnormal creatinine values were found in controls.
Cholesterol levels >6.47 mmol/L were observed in 18.23% and 5.84% of tests in older patients and in controls, respectively (P<0.0001).>P=0.023); triglyceride levels >3.39 mmol/L were found in 43 older patients and in 28 controls (P=0.047). Fasting plasma glucose levels >8.325 mmol/L were found in 35 tests for older patients and in one test for controls (P<0.0001).
Tolerability of Antiretroviral Treatment and New Comorbidities
In 48 weeks of treatment, 141 adverse or untoward effects were recorded: 80 in older patients and 61 in controls. Gastrointestinal intolerance was the most frequent complaint in the first 12 weeks of treatment, and metabolic disorders and peripheral neuropathy had a peak incidence after 24 weeks of treatment in both groups.
The treatment regimen was modified in 75 older patients (47.16%) and 51 controls (43.22%), with no significant differences in the causes of change in both groups ( Table 4 ).
Thirty-nine new comorbidities requiringtreatment among those included in our study were diagnosed, with a rate of 24.52 per 100 person-years follow up among older patients. In controls, newly diagnosed comorbid conditions were significantly lower in frequency: four new diagnoses (rate 3.39 per 100 person-years follow up; P<0.0001).
Discussion
Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents[9] focus particular attention on the treatment of acute HIVinfection in women with child-bearing potential, pregnant women, patients with severe liver disease, and patients with tuberculosis coinfections. No data have been reported for older HIV-infected persons who constitute, in our opinion, a subgroup of patients for whom specific considerations are critical for the appropriate selection and monitoring of safe and effective antiretroviral treatment.
In this work, we found that older HIV-infected patients differ from younger patients in epidemiology, timing of HIV diagnosis, age-specific health challenges and organ dysfunctions.
A higher rate of HIV transmission through heterosexual contacts (52.8%) was found in older patients compared with younger controls and with an Italian AIDS cohort,[2] in whom the rate of infection via the heterosexual transmission route from the beginning of the epidemic until December 2004 was 20.0%. In older patients, a very high rate of transmission through blood transfusions (4.4% vs 0% in controls and 0.8% in the Italian cohort) and unknown or unreported HIV exposure (18.9% vs 2.5% in controls and 3.3% in the Italian cohort) was found.
Older AIDS patients did not differ from controls for opportunistic or AIDS-defining conditions but they came later to medical attention, as outlined by the higher rate of late presenters, namely those who needed antiretroviral treatment within a very short time after their first medical visit. Such late presentation is probably a result of a lack of awareness of risk factors, leading to a late diagnosis. These epidemiological data suggest that older patients are more likely to have difficulties in assessing their exposure risk or to show psychological denial concerning their disease, and thus that there is a need for specific studies on the psychological profile of older HIV-infected people and for targeted information campaigns.
The most frequent comorbid conditions in older patients were cardiovascular and endocrine-metabolic diseases, while in the younger control population chronic liver diseases were the main comorbid conditions, probably as a result of the higher rate of injecting drug use in these patients.
We did not find differences in immunovirological recovery between these two groups of patients. Independent of the therapeutic regimen, we observed a sharp decrease in viral load within 12 weeks of treatment both in older patients and in controls (2.6 log10 copies/mL in both groups). The overall increase in CD4 cell count did not differ between the groups, although CD4 cell reconstitution was slightly reduced in the older group (137.3 and 191.8 cells/μL in older patients and controls, respectively) as previously described by Grabar et al..[10] The durability of the immune recovery obtained and the effects on overall survival were not evaluated in this study. Perez[8] observed a 2-fold increased hazard rate for death in untreated patients more than 50 years old than in a younger group, but did not find differences in survival between HAART-treated older and younger HIV-infected patients.
HIV infection is changing from a life-threatening to a chronic disease, and the proportion of deaths unrelated to HIV infection is growing. In this context, underlying health problems, comorbidities and senescence itself could become very important variables that must be taken into account in determining treatment efficacy or health policy.
In a probabilistic simulation study, Scott Braithwaite et al.[11] calculated that the proportion of patients infected with HIV on HAART who would die of comorbid conditions would be higher in a group of 50-year-old patients than in a group of 30-year-old patients (72% vs 45%) sharing the same favourable prognostic indicators (CD4 counts of 800 cells/μL and viral loads of 10 000 copies/mL). They estimated that the main causes of death unrelated to HIV infection would be cardiovascular diseases (35%), cancer (26%) and liver failure (12%).
More than one-third of our older patients had a non-HIV-related disease before antiretroviral treatment, with a total of 59 diagnoses for the conditions included in our analysis, and one-quarter of them were taking HIV-unrelated medication; cardiovascular problems and endocrine-metabolic disorders were the most significant baseline conditions. In 48 weeks of treatment, a significantly higher number of new disorders were diagnosed in older patients than in controls. The 39 newly diagnosed conditions were related to neuropsychiatric, cardiovascular and metabolic disorders. This may have been a result of naturally occurring age-related events and/or untoward and toxic effects of antiretrovirals acting synergistically with senescence.
HIV-associated cardiovascular manifestations, described since the beginning of the HIV pandemic, have been associated with several pathogenic mechanisms unrelated to HIV infection per se:[12] coronary heart disease has been correlated with protease inhibitor-induced metabolic and coagulative disorders, and systemic arterial hypertension has been related to HIV endothelial dysfunction, to atherosclerosis induced by HAART or to protease inhibitor-induced insulin resistance with increased sympathetic activity and sodium retention. In general, HIV-infected people are at higher risk of developing hypertension at a young age than the general population.[13] Seaberg et al.[14] observed that HIV-positive men taking HAART had a higher risk of systolic hypertension after 5 or more years of treatment compared with HIV-negative men, and that HAART treatment did not affect diastolic pressure, thus producing a syndrome similar to the isolated systolic hypertension commonly seen in elderly people.
Prolonged HAART has been correlated with an increased cardiovascular risk and also with increases in total cholesterol levels, triglyceride-rich very low density lipoprotein (VLDL) cholesterol and low-density lipoprotein (LDL) cholesterol, abnormalities in glucose metabolism with overt diabetes, impaired glucose tolerance and insulin resistance.[15]
The mechanisms of such metabolic alterations have not yet been completely elucidated, but may be related to altered nutrient metabolism and changes in body composition induced by the HIV infection itself, and/or by the direct effects of antiviral agents.[12] Age is probably an adjunctive risk factor for the development both of altered lipid and glucose metabolism and, ultimately, of increased cardiovascular risk. In the different definitions proposed for the metabolic syndrome,[16] high blood pressure and high levels of triglycerides, high-density lipoprotein (HDL) cholesterol and fasting glucose constitute the main risk factors for the development of cardiovascular disease.[14] Lifestyle modifications, including changes in diet, increased physical activity and smoking cessation, and changes in HIV treatment to regimens with lower risk drugs have to be considered in the management of this population at high risk of cardiovascular disease.
In our cohort, the percentage of older patients with abnormal values for glucose, total and HDL cholesterol, and triglycerides was higher at baseline evaluation than the percentage for young controls, and after 48 weeks of antiretroviral treatment the gap between the two groups had increased, with the risk of abnormal biochemical values increasing in the older population for all the parameters evaluated. The only exception was ALT levels, which were more frequently abnormal in young controls, probably as a consequence of the higher rate of chronic liver disease in this group. It is noteworthy, however, that the most severe liver toxicities (grades 3 and 4) were recorded in the older population and that several episodes of renal insufficiency with moderate to severe renal toxicity were recorded in the older patients but not in the younger patients, suggesting that adequate drug dosages should be identified for older HIV-infected patients.
Fick et al.,[17] in the most recent updating of Beer's criteria for potentially inadequate medication use in older adults, do not mention these drugs.
No differences were found in tolerability of antiretroviral treatments, although the rate of peripheral neuropathy was significantly higher in older patients than in controls, but it was not possible to determine the role of drugs and of comorbidities (diabetes, vascular disorders, etc.) in the pathogenesis of this symptom.
As the HIV-infected population has changed from the beginning of the epidemic, with a significant increase in mean age, the overall profile of HIV infection, opportunistic infections and comorbidities associated with immune suppression or toxic effects of antiretroviral drugs has increasingly been overlaid on the age-specific health profile, giving rise to new questions and challenges. In addition to the problems indicated in this study, it has been reported that older HIV-infected patients seem to be at a higher risk of developing depression and memory problems[18,19] which could affect antiretroviral treatment compliance, adherence and effectiveness, and eventually promote the development of resistance to antiretroviral treatments.
The limitations of this study are the relatively low numbers of older patients and controls included compared with the high numbers of variables, and the very long period of patient enrolment, with a wide variation in antiretroviral regimens, which may have affect the results obtained. Targeted comparative epidemiological, research and surveillance studies are needed to answer the questions identified in this work and those arising in clinical practice.
Reprint Address
Dr Giovanna Orlando, II Divisione Malattie Infettive, AN - Polo Universitario L Sacco, Via GB Grassi, 74, 20157 Milano, Italy. Tel:+0239042572; fax:+0238200909; e-mail: g.orlando@hsacco.it
HIV Med. 2006;7(8):549-557. ©2006 Blackwell Publishing
AIDS quilt is being unfurled by activists
