• DOI:10.4088/JCP.14com09617
• Corpus ID: 39039105

Buprenorphine-naloxone treatment of prescription opioid abuse: does past performance predict future results?

@article{Langleben2015BuprenorphinenaloxoneTO,
  title={Buprenorphine-naloxone treatment of prescription opioid abuse: does past performance predict future results?},
  author={Daniel D. Langleben},
  journal={The Journal of clinical psychiatry},
  year={2015},
  volume={76 2},
  pages={
          195-7
        }
}

• D. Langleben
• Published 25 February 2015
• Medicine, Psychology
• The Journal of clinical psychiatry

J Clin Psychiatry 2015;76(2):195–197 (doi:10.4088/JCP.14com09617). © Copyright 2015 Physicians Postgraduate Press, Inc. M isuse of prescription opioids is a national public health problem: In the United States, according to the 2012 National Survey on Drug Use and Health, 4.5 million, or 1.7%, of persons aged 12 or older reported current nonmedical use of pain relievers, and 335,000 reported currently using heroin.1,2 This is an upward trend, as heroin use has more than doubled since 2002.2 Transition from ingesting or insufflating (ie, snorting) prescription opioids to snorting or smoking (ie, inhaling the fumes) heroin has become more common and is driven by the increasing purity and lower cost.3 Significant numbers of heroin users progress to intravenous use,4,5 making prescription opioid abuse a risk factor for HIV and hepatitis C.6 To further complicate the matter, the prevalence of injecting prescription opioids may be increasing.6 However, compared to heroin, prescription opioid addiction has a limited literature base. In this month’s Focus on Addiction section, McDermott et al7 examine the relationship between abstinence from illicit opioids in the first and third (last) months of treatment with buprenorphinenaloxone combination tablets.8 They found that participants not abstinent from illicit opioids in the first 2 weeks were very unlikely to achieve abstinence in weeks 9 through 12 of the trial. In addition to the potential implications for practice, the McDermott et al article highlights 2 general questions: (1) To what extent do protocols developed for heroin dependence apply to prescription opioids? and (2) What is the state of the art in the prediction of treatment response in opioid use disorders? Data reported by McDermott and colleagues are extracted from weeks 12 to 24 of the 36-week Prescription Opioid Addiction Treatment Study (POATS).9 As a secondary analysis, McDermott is best understood in the context of POATS as it has been reported in several prior papers.9–16 Briefly, POATS consisted of 2 phases: a “brief ” phase 1 that entailed a 4-week buprenorphine-naloxone “stabilization” and taper and an 8-week follow-up. Of the 653 participants who entered phase 1, only 43 (7%) were either mostly or completely abstinent at week 12. Of the 610 patients who failed phase 1, 360 enrolled in the “extended” phase 2, which lasted 12 weeks and was followed by a 4-week taper of buprenorphine-naloxone and an 8-week follow-up. Therefore, one could describe 2 “initial” responses to treatment in POATS, 1 for each phase. McDermott et al found that 56% of patients who were abstinent at the 2-week timepoint were abstinent in the last 3 weeks of buprenorphine-naloxone treatment. Conversely, only 6% of those who continued using during the first 2 weeks were abstinent at the 3-month time point. Together, these results translate to a predictive value of 94% for the lack of abstinence at weeks 9 through 12, while the predictive value of early abstinence was below 60%. McDermott et al defined “lack of abstinence” as a positive test in their assigning of positive and negative predictive values. This is somewhat counterintuitive, since positive and negative predictive values are commonly used to describe a diagnostic test, in this case, urine toxicology. To rephrase McDermott and colleagues’ findings from a clinician’s perspective, opioid-positive urine toxicology in the first 2 weeks of treatment had a high positive and low negative predictive value for opioid-positive urine toxicology in the last 3 weeks. The predictive value of the early positive urine toxicology is in line with findings in other substance use disorders. However, the low predictive value of early abstinence contrasts similar studies in cocaine17 and suggests that additional variables need to be considered. Another clue that suggests unaccounted-for variables is the poor outcome of phase 1 of the POATS. A direct comparison with other studies is complicated by differences in design and outcome measures; however, Sigmon et al18 reported between 29% and 63% improvement after a 2to 4-week buprenorphine-naloxone taper in prescription opioid abusers, while only 7% of POATS participants were abstinent at the end of phase 1. Consistent with prior literature,19 nearly half of POATS participants were abstinent or reduced their use at the end of the period covered by McDermott et al, but most (> 90%) relapsed within 8 weeks after buprenorphinenaloxone was tapered off.9 Therefore, the McDermott et al data can be thought of as predicting an intermediate outcome. It would be interesting to know what bearing these data have on the patient status 8 weeks later. A more immediate conclusion from McDermott and colleagues’ article is that 12 weeks of buprenorphine-naloxone treatment was insufficient to achieve a sustained remission in the POATS cohort. However, it is unclear to what extent the McDermott et al sample was representative of the prescription opioid abuse demographic. More than 40% of POATS patients reported chronic pain, and approximately 30% reported lifetime prevalence of heroin or alcohol abuse and depression. All of these factors are likely to influence the prognosis of prescription opioid dependence,20–24 with pain20,21 playing a particularly large and complex role.24 It would be interesting to see how these variables might have changed the McDermott et al results. See article by McDermott et al p189