Marseille 2007
| Abstract #302 - Withdrawal symptoms as a predictor of mortality in patients HIV-infected through drug use receiving antiretroviral treatment (ART) | ||||
|
Session: 41.4: Drugs Alcohol and Potions (Parallel) on Tuesday @ 16.30-18.30 in 3 Chaired by Jeffrey Weiss, Biljana Ristic Authors: Presenting Author: Mrs Maria Patrizia Carrieri - INSERM, France |
||||
| Additional Authors: Dr Roch Giorgi, Ms Virginie Villes, Dr Isabelle Poizot-Martin, Prof Pierre Dellamonica, Dr Bruno Spire, | ||||
| Aim: Even in the ART era, individuals HIV-infected through drug use (IDUs) are known to have higher mortality rates than those in other HIV-transmission categories. This increased risk of death is mainly attributable to other competing causes, such as suicide, overdose, HCV progression, that may affect survival of IDUs. However, to date, few studies have investigated to what extent life event experience may be predictive of an increased risk of death once that HIV-disease status is taken into account. In this analysis, the MANIF 2000 cohort allowed us to investigate the role of such life events experience on survival of IDUs living with HIV and receiving ART. | ||||
| Method / Issue: We used 9-year longitudinal data from MANIF 2000 cohort of patients receiving ART (N=294) who accounted for 1430 PY of follow-up. Clinical and biological data were derived by medical records. Ninety-three percent of patients were HCV positive. At each given visit, face to face interviews and self-administered questionnaires collected, among other information, data about life events, social conditions, history of prison, substitution treatment, injection practices, addictive behaviours, depression and self-reported side effects. Negative life events were grouped into several categories: financial problems, negative relational events, violence, hospitalisation, illnesses or accidents in family members, drug-related problems. This latter variable included negative experience with detoxification, death for overdose of a family member or a friend and withdrawal symptoms. Survival analyses (Kaplan Meier and Cox model) were used to study the effect of possible baseline predictors and time dependent covariates on the risk of death. To account for missing information in the explanatory variables, we used multiple imputation to estimate HR in the final Cox model. | ||||
| Results / Comments: During the study period, 26 deaths occurred, 2 were attributable to HIV, 8 to liver disease, 2 to suicides and 1 road accident, 4 to cardiovascular disease, 4 unspecified at home or in a car, and the remaining for other causes. Sixty patients reported negative experience with withdrawal symptoms during follow up. Interestingly self-report of drug use as time dependent variable was not associated with a reduced risk of survival while negative experience of withdrawal symptoms was significantly associated with reduced survival (HR[95%CI]=4.7[1.1-19.9] after adjustment for viral load >100000 cp/ml (HR[95%CI]=4.7[2.0-10.9]) and age (HR[95%CI]=1.11[1.02-1.21]). Adjustments for initial CD4<200 and initial adherence to ART did not significantly change the strength of this association. This association remained significant even after exclusion of the two deaths for suicide. and also after multiple imputation of missing data of the other possible predictors. | ||||
| Discussion: Experience of withdrawal symptoms in IDUs living with HIV not only reflects difficulties in the management of their opiate dependence, but also may be a sign of decreased pharmacological tolerance to opiates. These results suggest that a proportion of these patients die from drug use as they have less tolerance. However, this association needs to be better investigated in larger populations. Withdrawal-like symptoms should be better detected and managed in the aim of properly delivering comprehensive care to opiate dependent individuals on ART and preventing drug-related deaths. | ||||
| Go Back | ||||
