Methadone is a schedule I drug in Russia, and the Ministry of Health has not accepted Western data on the benefits of agonist maintenance therapy. This approach is similar in many ways to the United States from the mid- 1920s to late 1960s, when physicians could lose their licenses or be arrested and jailed if they used opioids to treat opioid addicts. However, unlike the United States during those years, Russia has committed significant resources to detoxification and residential treatment. For example, state-supported alcohol and drug treatment is provided in 138 dispensaries (115 of which have inpatient units) and 12 addiction hospitals with more than 25 000 beds in total and from 50 to 2000 beds per hospital depending on the region. In addition, several hundred commercial and nongovernmental organizations and more than 5600 psychiatrist-narcologists work in the addiction field (Evgenia Koshkina, MD, PhD, personal communication, September 2011). This treatment is readily available, as seen in the , where study patients averaged 4 to 5 prior treatment episodes.
Baseline Demographics and Clinical Characteristicsa
Starting naltrexone therapy for these patients and under these conditions is easy because the patients undergo routine detoxification. Study findings show that an extended-release implant can alter the course of the addiction, at least for 6 months in about half the patients; however, the degree to which patients will accept longer courses of treatment is a topic for future studies. Unfortunately naltrexone, in the oral or extendedrelease form, is not widely available in Russia owing to costs, but this situation could change. Whatever the future may bring, patients in this study likely received better treatment than they otherwise would have, including those in the placebo group who received counseling from experienced therapists that was integrated into the study procedures and available immediately after completing detoxification and residential treatment.
Although results clearly favored the implant, patients who received oral naltrexone had fewer urine tests yielding results positive for opiates compared with the placebo group. In addition, the primary outcome showed a nonsignificant trend (P
=.07) favoring oral naltrexone compared with placebo that might be significant with a larger sample size. This difference was significant (P
=.02) when survival was measured for verified relapse; thus, oral naltrexone appeared to improve on the results of usual treatment with a few patients. These findings differ from earlier Russian studies where patients receiving oral naltrexone treatment had better outcomes than those of the placebo control group starting in the first month and continuing through month 6.11,12
A possible reason for these differences is that the older patients in the implant study (average age, 28.2 years) may have been less influenced by and dependent on close relatives for support than the younger patients (aged 21-23 years) in the earlier oral naltrexone studies. Genetic differences in μ-receptors may also play a role, and we are exploring this possibility in collaboration with other investigators.
These findings are similar to those from the recent trial of sustained-release injected naltrexone, where about half of the patients in the medication group remained in treatment for 6 months and had fewer urine tests with results positive for opioids than the placebo control group.9
From follow-ups on the limited sample of patients who remained in treatment without relapse and who returned for 9- and 12-month follow-ups, we can determine that approximately half relapsed after treatment ended. However, by counting missed appointments as relapses, almost all patients had a relapse, suggesting that for most patients, naltrexone therapy probably needs to be continued for an extended period.
Fourteen patients who received the naltrexone implant (13.7%) experienced a relapse between implantations, and 12 relapses occurred in weeks 6 through 8. The following 5 possibilities might account for this finding: fibrosis around the implant reduced dissemination of naltrexone; the patients metabolized naltrexone rapidly; patients had access to large amounts of high-grade heroin that they used to overcome the blockade as blood levels dropped toward the end of the dosing cycle; the implant released naltrexone more quickly than intended, resulting in low blood levels toward the end of the dosing cycle; or the subcutaneous tissue where the implant was placed did not have enough blood supply to absorb the naltrexone and maintain opioid blockade.
The possibility of patients unmasking the study by using heroin is not as likely as it may appear. In Russia, a sort of placebo effect is associated with getting an injection: patients often think injections are stronger regardless what is injected. In addition, the quality of heroin is sometimes poor, which might reduce the effect of a single heroin injection, and the effect also depends to some extent on expectation and setting. Thus lack of an effect from a single injectionmaynot necessarily be attributable to opioid blockade. In addition, the placebos were not active and had only a visual similarity to the active medication.
Similar to earlier studies, we saw no evidence of increased depression, anxiety, or anhedonia associated with naltrexone.34
In fact these symptoms, along with craving, appeared to drop for patients who continued treatment without relapse, as seen in other naltrexone studies with opioid-dependent patients35,36
and in studies of alcohol-dependent patients treated with extendedrelease injected naltrexone who did not experience dysphoria or lack of pleasurable stimuli.37
Tolerability of the implant was generally good, and no serious AEs attributable to the study medications were reported; however, AEs at the implant site were more common among patients who received the naltrexone implant. This finding could reflect contamination in some of the implants, local irritation caused by naltrexone or other components of the implant, or patients’ attempts to remove the implant, although none were reported. The proportion of other AEs was comparable across groups and also to those in a study using the Australian implant16
; however, in that study, 3 of the 56 patients had implants removed at their request.
Previous studies have shown that any effective treatment for opioid dependence reduces risk of HIV due to injections.38
This finding is very relevant to countries such as Russia, where HIV is being spread largely by injected drug use as reflected by these and other data from St Petersburg showing that more than 40% of opiateaddicted patients are seropositive for HIV.39,40
Given the potential for reduction in HIV risk among patients who remained in naltrexone treatment and did not relapse, combined with the apparent unshakable resistance to using agonist therapies in Russia and the widespread availability of inpatient detoxification, naltrexone and in particular extended-release formulations could play a meaningful role in reducing the spread of HIV if the treatment was more readily available throughout the network of state and private treatment facilities.
The limitations of naltrexone implants include the surgical procedure, possibility of wound infection or local irritation, cosmetic defects (scars), need for high opioid doses if the patient develops amedical condition that requires opioid therapy, and possible removal of the implant by the patient within 7 to 14 days after receiving it. Limitations of the study include the limited amount of data on patients who did not remain in treatment, thus making it difficult to obtain more accurate information on the proportions with relapse at 9- and 12-month followups and other secondary outcomes. Strengths include the randomized, prospective, double-dummy design; the large number of participants; involvement of close relatives to provide additional information; and determination of the primary outcome by objective data.