| |
|
Greater Effect of Highly Active Antiretroviral Therapy on Survival in People Aged 50 Years Compared with Younger People in an Urban Observational Cohort
|
| |
| |
This study was presented previously at an AIDS conference and reported on the NATAP website and is now published in Clinical Infectious Diseases.
When older patients (>50) were untreated they tended to die more quickly. But, when treated with HAART older patients did not die sooner than younger patients at 3 years of follow-up: there was no significant difference in the cumulative survival rates between the older and younger treated groups at 3 years (83% vs. 89%, respectively; P = .29).... there was a 72% decrease in the hazard for older treated people compared with older untreated people.. In the present study, we have shown that HIV-infected individuals aged 50 years and unexposed to HAART had double the hazard rate for death than did younger, untreated, HIV-infected people... we have shown that HIV-infected individuals aged 50 years and unexposed to HAART had double the hazard rate for death than did younger, untreated, HIV-infected people. Older patients receiving HAART had >2-fold reduction in the hazard for death after controlling for enrollment CD4 cell count and demographic variables...among patients who had been treated with HAART for >90 days, there were no statistically or clinically significant differences in the survival rate between the younger and older groups...study limitation: compliance was not followed in this study (editorial note: older folks may be more compliant)...but, the effect of comorbidities (diabetes, elevated lipids, etc) would bias the survival against the treated elderly; instead, this group benefited the most from treatment.
Summary: Although human immunodeficiency virusinfected people aged 50 years have a blunted CD4 cell recovery when receiving highly active antiretroviral therapy (HAART), there are few data on mortality. Mortality rates were studied for 253 individuals aged 50 years and a younger group of 535 people in a retrospective cohort; for untreated persons in each age group, the proportions surviving at 3 years were 83% and 70% (P < .01), respectively. No significant difference in the survival rate was found between the older (83%) and younger (89%) patients who received HAART (P = .29). The hazard ratio for death in the older untreated group was 2.4 (95% confidence interval [CI], 1.43.9) when exposed to HAART. However, compared with older untreated patients, the hazard ratio for death decreased to 0.28 (95% CI, 0.150.52) for treated older adults. The effect of HAART substantially improves the survival rate for older individuals and supports the importance of treatment in this group.
Clinical Infectious Diseases 2003;36:212-218
John L. Perez and Richard D. Moore; Division of Infectious Diseases, Johns Hopkins University School of Medicine, Baltimore, Maryland
As of December 2000, >774,000 cases of HIV infection in the United States had been reported to the US Centers for Disease Control and Prevention, and 58% of the affected patients had died. During this time, the number of cases reported among those aged 50 years was >84,000 (10.9%). Several cohort studies have shown that older people have progression to AIDS faster than younger people. For example, a French cohort found that older age was associated with a more rapid progression to symptomatic disease. In persons with hemophilia, the relative risk of developing AIDS after seroconversion is 1.45 for each 10-year increase in age. The duration of survival is also significantly shorter for elderly persons, and older patients have poorer outcomes of opportunistic infections than do younger individuals. These data suggest that older patients might benefit from treatment more than do younger people.
HAART is effective in correcting CD4 lymphocytopenia and decreasing the virus load; this has been associated with a 50% decrease in morbidity and mortality associated with AIDS. The degree of immune recovery is dependent on the regenerative capacity of the thymus, which is lost with advanced age. However, in one prospective cohort, the extent of CD4 recovery could be demonstrated in the oldest subjects in the cohort, although CD4 recovery was negatively correlated with age. Although the median age was only 37 years in this cohort, it is reasonable to suspect that CD4 recovery may be slow, but present, as age increases. Recently, one case-control study examined 52 HIV-infected individuals aged 50 years and matched them to a random sample of 52 younger control subjects. They found no differences in CD4 counts, virus loads, frequency of opportunistic infections, hospitalizations, drug-related side effects, and death. However, they did not include people who were untreated; thus, no estimate of the relative risk for death for treated versus untreated patients could be determined.
Given the question of meaningful immune restoration in older people and the significant toxicities of HAART, it is important to determine in a community-based setting whether HAART is effective in changing the level of HIV-related mortality in older people with HIV infection. We have conducted such an analysis by comparing the rate of survival for patients aged 50 years with the rate for younger patients in a large urban HIV clinical cohort.
Subjects for this analysis were enrolled at the Johns Hopkins HIV Clinic (Baltimore, Maryland), which provides primary and specialty care for HIV infection to 2500 patients. In 1990, an observational clinical database was established for patients receiving longitudinal primary care for HIV infection at the clinic. All patients were eligible for inclusion in the study if they enrolled during the period of 1 July 1990 through 28 February 2001.
We used a retrospective cohort design. The older group consisted of patients aged 50 years at the time of enrollment in the clinic (n = 259), whereas the younger group consisted of patients aged 1849 years (n = 538). The age cutoff of 50 years was chosen for 2 reasons: the World Health Organization does not report the prevalence of HIV infection among persons aged 50 years, and previous studies have used this cutoff. The younger group was chosen at random from the total cohort at a ratio of younger to older patients of approximately 2 : 1. The primary outcome of interest was all-cause mortality. Patients who were observed in the clinic for <90 days were excluded from the survival analysis (n = 9) to ensure an adequate follow-up time necessary to evaluate the effect of receipt of HAART.
RESULTS
A total of 253 patients aged 50 years were identified from the clinic; for comparison, 535 younger patients were chosen randomly among the 2500 people enrolled in the clinic. The mean age at enrollment was 36.5 years in the younger group (range, 1849 years) and 55.2 years in the older group (range, 5083 years; P < .05). Patients in the older group were more likely than patients in the younger group to be male (83% vs. 72%; P < .05) and to not be men who have sex with men (26% vs. 34%; P < .05).
At enrollment, there was no difference between the groups with regard to the stage of HIV disease. The median HIV RNA level at enrollment was lower in the older group (4.3 log10 copies/mL) than the younger group (4.5 log10 copies/mL; P < .05), but the median CD4 cell count for the older group (289 cells/mm3) was significantly higher than that for the younger group (244 cells/mm3; P < .05).
There was no statistically significant difference between the older and younger groups with regard to the percentage of patients receiving HAART (P = .30), the year of enrollment (P = .25), or the year that HAART was initiated (P = .23). Among those who never received HAART, there was no significant difference between the older and younger groups in the median CD4 cell count (300 vs. 290 cells/mm3, respectively; P = .46) or HIV RNA load at enrollment (4.4 vs. 4.4 log10 copies/mL, respectively; P = .49). However, among patients who received HAART, older patients had a significantly higher median enrollment CD4 cell count (273 vs. 224 cells/mm3; P < .05) and a lower virus load (4.2 vs. 5.2 log10 copies/mL; P < .05) than did the younger patients.
There were more deaths in the older group (20%) than in the younger group (12%; P < .01). The median times to death in the older and younger groups were 663.5 and 828 days, respectively (P = .29). The conditional Kaplan-Meier survival analysis was performed by dividing the patients into 4 groups: young untreated patients (n = 193), young patients who were receiving HAART (n = 342), older untreated patients (n = 101), and older patients who were receiving HAART (n = 152). The median length of follow-up time for the cohort was 1095 days (3 years). The cumulative proportions of patients who survived at 3 years were 83% in the younger group and 70% in the older group (P < .01). However, there was no significant difference in the cumulative survival rates between the older and younger treated groups at 3 years (83% vs. 89%, respectively; P = .29).
These results are consistent with those of earlier studies, which showed that older patients have a higher mortality rate than do younger patients and that age is inversely correlated with the time of AIDS diagnosis. AIDS-defining illnesses have also been shown to be present at higher CD4 cell counts in people aged >30 years. The magnitude and speed of CD4 cell count recovery after the initiation of HAART is also negatively correlated with age in people with AIDS and in elderly persons with malignancies that occur after they have undergone marrow ablative chemotherapy. The observation that the ability of the thymus to repopulate CD4 cells is inversely correlated with age has been proposed as a mechanism for the blunted T cell response to HAART in elderly individuals. These findings imply that treatment of elderly persons may not lead to clinically meaningful responses. However, our study used a direct measure of treatment response (mortality) rather than surrogate markers (CD4 cell count and virus load). By means of this approach, we have now demonstrated that patients aged 50 years have a great mortality benefit associated with HAART. The reason for the discrepancy between the CD4 cell count and the survival rate may be the lag between diagnosis of HIV/AIDS and death and the subsequent follow-up required for cohort studies.
|
|
| |
| |
|
|
|
