Skip to Content
Discovering the causes of cancer and the means of prevention

Publications Search - Abstract View

Title: Cause-specific mortality associated with HIV and HTLV-II infections among injecting drug users in the USA.
Authors: Goedert JJ,  Fung MW,  Felton S,  Battjes RJ,  Engels EA
Journal: AIDS
Date: 2001 Jul 6
Branches: IIB
PubMed ID: 11426075
PMC ID: not available
Abstract: BACKGROUND: Human T-lymphotropic virus type II (HTLV-II) is widespread among injecting drug users (IDU) and may contribute to the risk of leukemia/lymphoma, neurodegenerative disease, and perhaps pneumonia, especially with HIV co-infection. METHODS: In 1987--1991, 6570 IDU were tested for HIV and HTLV-II antibodies. In 1998, they were matched to the National Death Index. Numbers of observed deaths of each cause were compared by standardized mortality ratios with the numbers expected, using sex-, race-, age-, and year-specific rates in the general population. Relative risk (RR) associated with each virus, compared to uninfected drug users, was estimated by Poisson modeling. RESULTS: There were 1351 deaths, including 683 (15%) of 4604 participants who enrolled seronegative for both viruses; 328 (47%) of 701 who had HIV but not HTLV-II infection; 220 (21%) of 1033 who had HTLV-II but not HIV infection; and 120 (52%) of 232 who were infected by both viruses. Compared to the general population, mortality for participants with neither virus was increased 4.3-fold [95% confidence interval (CI), 4.0--4.7] and was significantly elevated for virtually every cause of death. With HIV, mortality from medical causes, but not external causes, was increased 3.7-fold (95% CI, 3.3--4.2), particularly with AIDS and related conditions. With HTLV-II, all-cause mortality was reduced (RR, 0.8; 95% CI, 0.7--0.9), with no statistically significant reduction or elevation for any specific cause. A non-significant excess of tuberculosis deaths (RR, 4.6; 95% CI, 0.8--25.2) was noted with HTLV-II, but there was no excess mortality from leukemia/lymphoma, other malignancies, or neurodegenerative disease. CONCLUSIONS: Without HIV or HTLV-II, IDU had profoundly increased mortality from medical and external causes. HIV was specifically associated with death due to AIDS and related conditions. HTLV-II infection was not significantly associated with mortality from any cause, suggesting that it is not a significant human pathogen, even when present with HIV infection.