Substance use, particularly the compulsive behaviors associated with addiction, lead to unhealthy behaviors including non-adherence to antiretroviral therapy (ART) and treatment failure. High on the list of disorders leading to non-adherence is heroin addiction as a wide range of impulsive, high-risk behaviors accompanies it. The science of adherence would be improved by developing new methods to prevent relapse to heroin addiction, especially methods that can be used in settings that are not limited by the aims to test such a method using an implantable naltrexone formulation (IN) that is approved in Russia and blocks opioid effects for 3 months. The efficacy of the IN should be better than oral naltrexone (ON) because it does not depend on daily behavior to take a tablet and maintains a constant plasma level for months, which should result in sustained blockade, less relapse, and better ART adherence and treatment response.

This is a double blind, double-dummy, placebo controlled, randomized trial of a 48-week course of implant naltrexone vs. oral naltrexone, each arm with drug counseling every two weeks, for 200 HIV+ patients who are in early remission from opioid dependence, and who are interested in relapse prevention treatment medication, and starting their first episode of antiretroviral therapy at the Botkin Infectious Disease Hospital in St. Petersburg or the Leningrad Regional AIDS Center. Early remission was chosen because relapse risk is the highest at this point, thus maximizing the chances for detecting a naltrexone effect. The first antiretroviral therapy treatment episode was chosen because it is feasible (relatively few opioid addicted Russians have been treated with antiretroviral therapy), and because the virus is less likely to have developed secondary resistance.

Participants will be recruited from the AIDS and addiction programs and who meet study admission criteria will be stratified within each site according to baseline viral load (>100,000 copies/<100,000 copies) and cluster of differentiation 4 (CD4) count (>50/<50 copies). Participants will be randomized to a treatment condition, receive a naloxone challenge, and if pass be prescribed oral naltrexone or oral placebo and implant/implant placebo), and given a schedule for addiction counseling and HIV treatment appointments. A 2-week supply of oral medication will be provided at each bi-weekly counseling session, and will be re-implanted at weeks 12, 24, and 36. Only the research pharmacist will know the group assignments, however the blind can be broken in case of emergency.

The primary outcome measure will be to compare implanted naltrexone versus oral naltrexone on ability to achieve a viral load of <400 copies at weeks 24 and 48.

Secondary outcomes are to compare the efficacy of the two addiction treatments; to study the adherence to antiretroviral therapy; to evaluate time to relapse and the number of days to relapsed; to evaluate decline in CD4 counts; to evaluate HIV risk behavior; to evaluate opioid positive urine tests; and to evaluate the number of days that patients will keep their scheduled appointments. The Investigator will also monitor psychiatric symptoms, other drug use, and overall adjustment.

Inclusion Criteria:

• must be HIV+ men/women starting their first episode of ART or starting a new ART treatment episode and was prescribed medications that suppressed the virus to <400 copies during their last treatment
• understand that they will be prescribed ART medications that they have never received and to which their virus is likely to be susceptible
• viral loads of 1,000 copies or more
• meet the Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition (DSM-IV) criteria for opioid dependence in early remission
• have a negative opiate urine toxicology and alcohol breath test
• show no evidence of physiologic dependence on physical exam and following a naltrexone challenge
• have a stable address in the St. Petersburg or Leningrad Region of Russia area
• have a valid telephone number where subject can be reached
• have a negative pregnancy test and use adequate contraception
• have the ability to give informed consent as judged by ability to read the consent and correctly answer 9 of 10 questions about the study on a quiz that will be administered after discussing the study and reading the consent

Exclusion Criteria:

• not be currently psychotic as determined by a psychiatric examination (i.e.; schizophrenia, paranoid disorder, mania)
• not have current suicidal or homicidal ideation requiring immediate attention as determined at baseline assessment
• not have an uncontrolled seizure disorder
• not have cognitive impairment with an inability to read and understand the consent
• not have significant laboratory abnormality such as >2 grade anemia
• not have hepatic transaminase levels >5 times the upper limit of normal
• not have serum creatinine >1.5 times the upper limit of normal
• not have pending legal charges with impending incarceration
• not be concurrently participating in another treatment study
• not currently taking naltrexone