The American Journal of Drug and Alcohol Abuse, 37:131–136, 2011 Copyright © Informa Healthcare USA, Inc. ISSN: 0095-2990 print / 1097-9891 online DOI: 10.3109/00952990.2010.543998 Characterizing methamphetamine withdrawal in recently abstinent methamphetamine users: a pilot field study Michael J. Mancino, M.D.1, Brooks W. Gentry, M.D.2, Zachary Feldman, M.S.1, John Mendelson, M.D., Ph.D.3 and Alison Oliveto, Ph.D.1 1Department of Psychiatry, Psychiatric Research Institute, University of Arkansas for Medical Sciences, Little Rock, AR, USA, 2Department of Anesthesia, University of Arkansas for Medical Sciences, Little Rock, AR, USA, 3Department of Internal Medicine, Drug Dependence Research Laboratory, Langley Porter Institute, University of California at San Francisco, San Francisco, CA, USA INTRODUCTION Background: Methamphetamine dependence has become a significant problem, but methamphetamine Over the last two decades methamphetamine use has withdrawal symptoms have not been well escalated in the USA as well as around the world (1,2). studied. Methods: This prospective observational Methamphetamine abuse carries with it profound medical pilot study was designed to examine withdrawal (3,4), psychiatric (1,3), legal (5), and social (1) problems. symptoms, mood, anxiety, cognitive function, and The increasing prevalence and adverse consequences of subjective measures of sleep over a 4-week period methamphetamine use highlight the importance of char- in six patients entering residential treatment for acterizing better the pharmacology of methamphetamine methamphetamine dependence. Results: to develop promising treatment strategies for metham- Methamphetamine withdrawal symptoms, phetamine dependence. mood, and anxiety symptoms all resolve fairly The Diagnostic and Statistical Manual defines crite- quickly within 2 weeks of cessation of ria for cocaine and amphetamine withdrawal syndromes methamphetamine. Sleep was disrupted (6), and 87–97% of recently abstinent amphetamine users over the course of the 4-week study. No clinically are thought to experience withdrawal (7,8). Nevertheless, significant alterations in blood pressure or heart rate amphetamine withdrawal, much less methamphetamine were identified. This study did not demonstrate any withdrawal, has been studied on a very limited basis alterations in cognitive function over the 4 weeks of in humans (7–9). The duration of amphetamine with- the residential stay. Conclusions: This pilot study drawal is generally considered to be much longer than points toward the need for a double-blind, cocaine withdrawal (10), reportedly lasting from 5 days placebo-controlled amphetamine withdrawal to more than 2 weeks (7,11). Acute withdrawal from paradigm in humans where changes in sleep, chronic amphetamine use in humans has been associ- cognitive function, and withdrawal measures can be ated with depression, fatigue, sleepiness (12), aggression, explored more fully. Scientific Significance: This study agitation, anxiety, hyperphagia, hypersomnia, and psy- extends the literature by pointing toward a chomotor retardation (13,14) and has also been linked methamphetamine withdrawal syndrome that with sleep abnormalities such as early onset of the first includes alterations in measures of sleep quality and REM period, greatly increased total REM, and total sleep refreshed sleep, early improvement in depression and length that increased for several weeks (15). anxiety symptoms, most striking during the first Given that the period of drug or substance withdrawal week, but persisting into the second week. is often cited as time during which risk for relapse is very high (16,17), better characterizing the withdrawal Keywords: methamphetamine withdrawal, mood, sleep, cognitive syndrome associated with cessation of methamphetamine function use would be of great importance for developing more Address correspondence to: Michael J. Mancino, M.D., Department of Psychiatry, Center for Addiction Research, University of Arkansas for Medical Sciences, 4301 West Markham, Slot 848, Little Rock, Arkansas 72205 USA. Tel: +501-526-8442. Fax: +501-526-7816. E-mail: [email protected] 131 132 M. J. MANCINO ET AL. efficacious treatment strategies. Therefore, we conducted or eating too much; (x) decreased appetite; (xi) vivid this prospective observational pilot study to exam- or unpleasant dreams; (xii) sleeping too much; and ine mood, anxiety, cognitive function, and subjective (xiii) insomnia. Each item was scored on a 0–4 scale measures of sleep over a 4-week period in patients (choices: “not at all,” “a little,” “moderately,” “quite a entering residential treatment for methamphetamine bit,” and “extremely”). Hamilton Depression (HAM-D; dependence. 18) and Anxiety (HAM-A; 19) scales were adminis- tered at baseline and thrice weekly. Sleep quality and duration were assessed at baseline and thrice weekly METHODS utilizing items from the Sleep Inventory Questionnaire Subjects (20). Domains assessed by this sleep questionnaire Consecutive male and female treatment-seeking volun- include overall quality of sleep rated on a 0–4 scale teers were recruited from a local residential treatment (choices: “very poor,” “poor,” “average,” “good,” and facility between August 2006 and January 2008. Subjects “very good”) and how refreshed the subject felt after between the ages of 18 and 65 were recruited by flyers awakening in the morning rated on a 0–4 scale (choices: and staff referrals at the treatment facility. Participants “not at all,” “a little,” “moderately,” “quite a bit,” and had to report methamphetamine as their primary drug “extremely”). Cognitive assessments were obtained at problem with recent use verified by urine toxicology baseline, week 2, and week 4 and included the trail screen positive for methamphetamine. Subjects were not making parts A and B (21), the digit symbol sub- included if they had an unstable medical condition requir- test portion of the Wechsler Adult Intelligence Scale ing acute inpatient hospitalization; current diagnosis of – Revised (22), and the Stroop color–word interfer- other drug or alcohol physical dependence (other than ence test (23). Heart rate and blood pressure (sitting tobacco); history of psychosis, schizophrenia, or bipo- and standing) were measured at baseline and thrice lar disorder; current suicidality; or current use of over- weekly. Body weight was measured at baseline and the-counter or prescription psychoactive drugs that may weekly. affect mood ratings. Women were not eligible if they Urine samples were obtained thrice weekly and ana- were pregnant or lactating. Assessments were reviewed lyzed for the presence of methamphetamine (includ- by a study physician prior to study entry to ensure ing metabolites) and other drugs using Hitachi 717 these criteria were met. All participants gave written Automated Analyzers (Boehringer Mannheim Corp., informed consent to participate in the study and were Indianapolis, IN, USA). A quantitative test was used to compensated monetarily for participation. This proto- examine changes in urinary levels of methamphetamine col was approved by the UAMS Institutional Review over the course of participation to ensure that metham- Board. phetamine levels decreased over time as would be expected with abstinence. Setting Participants were residing at the Recovery Centers of Arkansas (RCA), a licensed and accredited community- Data Analysis based provider of residential services for substance abuse Means and ranges were computed for continuous baseline and related disorders. characteristics. For longitudinal data, Proc GLIMMIX in SAS was used to fit a Mixed Model with Random Research Design and Procedures Intercept. The GLIMMIX procedure in SAS allows for After informed consent, demographic data, drug use the generalization of the standard linear model (GLM) history, medical and psychiatric histories, and a urine by allowing for the data to exhibit correlation and non- sample to test for recent methamphetamine use were constant variability. Additionally, the random intercept obtained. Participants were then observed over a 4- allows the model to be fit separately for each individ- week period, during which time drug use, physiolog- ual, and then combined for the total model. This model ical, mood, sleep, cognitive, and withdrawal assess- was used initially to determine if there was a significant ments were obtained. The Methamphetamine Selective overall effect of time. Significant time point differences Severity Assessment (MSSA) was administered at base- were identified utilizing post hoc Holm–Bonferroni cor- line to determine severity of methamphetamine depen- rections for multiple comparisons. For simplicity, a mean dence (11). Methamphetamine withdrawal symptoms weekly score was calculated for measures that were were measured at baseline and thrice weekly using obtained thrice weekly and these mean weekly scores the Methamphetamine Withdrawal Assessment (MAWA) are reported. For data with five time points (baseline based on DSM-IV criteria for amphetamine withdrawal and weeks 1 through 4), hypothesis tests were per- (6). Thirteen items of amphetamine withdrawal symp- formed to test for differences between baseline and each toms were included in the MAWA. These items were subsequent time point as well as differences between (i) drug craving; (ii) dysphoric mood; (iii) loss of plea- each time point (e.g., baseline and week 1, week 1 sure; (iv) loss of interest in activities previously enjoyed; and week 2, etc.). All tests performed were two-tailed (v)