Add-On Deep Transcranial Magnetic Stimulation (Dtms) in Patients with Dysthymic Disorder Comorbid with Alcohol Use Disorder: a Comparison with Standard Treatment

The World Journal of Biological Psychiatry

ISSN: 1562-2975 (Print) 1814-1412 (Online) Journal homepage: http://www.tandfonline.com/loi/iwbp20

Add-on deep transcranial magnetic stimulation (dTMS) in patients with dysthymic disorder comorbid with alcohol use disorder: A comparison with standard treatment

Paolo Girardi, Chiara Rapinesi, Flavia Chiarotti, Georgios D. Kotzalidis, Daria Piacentino, Daniele Serata, Antonio Del Casale, Paola Scatena, Flavia Mascioli, Ruggero N. Raccah, Roberto Brugnoli, Vittorio Digiacomantonio, Vittoria Rachele Ferri, Stefano Ferracuti, Abraham Zangen & Gloria Angeletti

To cite this article: Paolo Girardi, Chiara Rapinesi, Flavia Chiarotti, Georgios D. Kotzalidis, Daria Piacentino, Daniele Serata, Antonio Del Casale, Paola Scatena, Flavia Mascioli, Ruggero N. Raccah, Roberto Brugnoli, Vittorio Digiacomantonio, Vittoria Rachele Ferri, Stefano Ferracuti, Abraham Zangen & Gloria Angeletti (2015) Add-on deep transcranial magnetic stimulation (dTMS) in patients with dysthymic disorder comorbid with alcohol use disorder: A comparison with standard treatment, The World Journal of Biological Psychiatry, 16:1, 66-73, DOI: 10.3109/15622975.2014.925583 To link to this article: https://doi.org/10.3109/15622975.2014.925583

Published online: 20 Aug 2014. Submit your article to this journal

Article views: 182 View related articles

View Crossmark data Citing articles: 9 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=iwbp20 The World Journal of Biological Psychiatry, 2015; 16: 66–73

BRIEF REPORT

Add-on deep transcranial magnetic stimulation (dTMS) in patients with dysthymic disorder comorbid with alcohol use disorder: A comparison with standard treatment

PAOLO GIRARDI1,2 , CHIARA RAPINESI1,2 , FLAVIA CHIAROTTI3 , GEORGIOS D. KOTZALIDIS 1 , DARIA PIACENTINO1 , DANIELE SERATA1,2 , ANTONIO DEL CASALE 1,4 , PAOLA SCATENA2 , FLAVIA MASCIOLI2 , RUGGERO N. RACCAH 5 , ROBERTO BRUGNOLI1 , VITTORIO DIGIACOMANTONIO2 , VITTORIA RACHELE FERRI 1,2 , STEFANO FERRACUTI1 , ABRAHAM ZANGEN6 & GLORIA ANGELETTI1

1 NESMOS Department (Neurosciences, Mental Health, and Sensory Organs), Sapienza University of Rome, School of Medicine and Psychology, Sant’ Andrea Hospital, Rome, Italy, 2 Alcohology Service, Villa Rosa, Suore Hospitaliere of the Sacred Heart of Jesus, Viterbo, Italy, 3 Department of Cell Biology and Neurosciences, Istituto Superiore di Sanità , Rome, Italy, 4 Department of Psychiatric Rehabilitation, Fondazione P. Alberto Mileno Onlus, Vasto, CH, Italy, 5 ATID Ltd Advanced Technology Innovation Distribution, Rome, Italy, and 6 Department of Life Sciences, Ben Gurion University of the Negev, Be ’ er Sheva, Israel

Abstract Objectives. Dorsolateral prefrontal cortex (DLPFC) is dysfunctional in mood and substance use disorders. We predicted higher efﬁ cacy for add-on bilateral prefrontal high-frequency deep transcranial magnetic stimulation (dTMS), compared with standard drug treatment (SDT) in patients with dysthymic disorder (DD)/alcohol use disorder (AUD) comorbidity. Methods. We carried-out a 6-month open-label study involving 20 abstinent patients with DSM-IV-TR AUD comorbid with previously developed DD. Ten patients received SDT for AUD with add-on bilateral dTMS (dTMS-AO) over the DLPFC, while another 10 received SDT alone. We rated alcohol craving with the Obsessive Compulsive Drinking Scale (OCDS), depression with the Hamilton Depression Rating Scale (HDRS), clinical status with the Clinical Global Impressions scale (CGI), and global functioning with the Global Assessment of Functioning (GAF). Results. At the end of the 20-session dTMS period (or an equivalent period in the SDT group), craving scores and depressive symptoms in the dTMS-AO group dropped signiﬁ cantly more than in the SDT group ( P Ͻ 0.001 and P Ͻ 0.02, respectively). Conclusions. High frequency bilateral DLPFC dTMS with left preference was well tolerated and found to be effective as add-on in AUD. The potential of dTMS for reducing craving in substance use disorder patients deserves to be further investigated.

Key words: alcohol , dysthymia, depression, dorsolateral pre-frontal cortex (DLPFC), deep Transcranial magnetic stimulation (dTMS)

Introduction to drink or intense thoughts about alcohol (World Health Organization 1992), and its development Alcohol use disorder (AUD), currently the most common psychiatric disorder, represents a wide- plays a major role in maintaining alcohol intake and spread and serious personal and public health prob- dependence and may promote relapse (Drummond lem in the United States (Kessler et al. 1994; Hasin 2001). The development of craving is correlated et al. 2007). Alcohol craving is an irresistible urge with changes in the brain reward circuitry, which

Correspondence: Georgios D. Kotzalidis, MD, NESMOS Department (Neurosciences, Mental Health, and Sensory Organs), Sapienza University, School of Medicine and Psychology, Psychiatry Unit, Sant’ Andrea Hospital, Rome, Italy. Tel: ϩ 39 0633775951. Fax: ϩ 39 0633775342. E-mail: [email protected]

(Received 30 July 2013 ; accepted 14 May 2014 ) ISSN 1562-2975 print/ISSN 1814-1412 online © 2015 Informa Healthcare DOI: 10.3109/15622975.2014.925583 dTMS in dysthymia plus alcoholism 67 includes the medial forebrain bundle and the meso- Materials and methods cortical and meso-limbic dopamine pathways (Park Patients et al. 2007). Prefrontal abnormalities have been documented The study was conducted at the Alcoholism in many substance abuse disorders (Lang et al. Service, Villa Rosa, Viterbo, Italy. We admitted to 2008; Moreno-Ló pez et al. 2012; Bosch et al. the Alcoholism Service’ s day hospital 10 consecu- 2013) and the dorsolateral prefrontal cortex tive patients with comorbid long-term DSM-IV-TR (DLPFC) is a major component of the neural sub- DD and AUD (American Psychiatric Association strate of craving for several psychoactive substances, 2000) in their detoxifi cation phase (i.e., abstaining including alcohol (Olbrich et al. 2006). DLPFC for at least 1 month), from December 2011 to June abnormalities have been demonstrated in alcohol 2012. Patients agreed to undergo dTMS as an use disorder (AUD) using different methodologies, add-on to their current treatment, which was left including functional near-infrared spectroscopy unmodifi ed, and did not use their abuse substance (Ernst et al. 2014), functional magnetic resonance throughout the study period. imaging (Park et al. 2010), structural magnetic Inclusion criteria, besides the above-mentioned resonance imaging (Makris et al. 2008), and post- DSM-IV-TR diagnoses, were age between 16 and mortem proteomic analysis (Alexander-Kaufman 65 years; at least 5-year duration of illness; avail- et al. 2007). Functional and structural DLPFC ability of reliable informants; wish to participate in alterations have also been reported in mood disor- the study; and providing consent for undergoing ders, namely bipolar (Townsend et al. 2010), major dTMS. Exclusion criteria were other concurrent depressive (Chang et al. 2011; Oh et al. 2012; Ye substance use except nicotine; specifi c contraindica- et al. 2012), and dysthymic disorders (DD; Ravin- tions to dTMS (history of seizures and carrying a dran et al. 2009). Indeed, alcohol dependence and pacemaker); and having received dTMS in the past mood disorders often co-occur and complicate the 12 months. course and outcome of one another (Grant et al. DSM-IV-TR diagnoses were established after 2004). structured interviews with all patients (Structured The effectiveness of available anticraving drugs for Clinical Interviews for DSM-IV Axis I and II Dis- alcohol dependence is limited (O’ Brien 2005). Inter- orders; SCID-1 and SCID-2, respectively) (First estingly, transcranial magnetic stimulation (TMS) et al. 1997, 2002). All patients met also criteria for techniques are increasingly employed and reported DSM-5 Persistent Depressive Disorder (Dysthy- to benefi t several psychiatric conditions, including mia) (American Psychiatric Association 2013). cocaine (Politi et al. 2008) and alcohol abuse in The rating scales used to assess patient status abstinent patients (Mishra et al. 2010; Rapinesi et al. were the Obsessive Compulsive Drinking Scale 2013), nicotine abuse (Amiaz et al. 2009), and mood (OCDS; Anton et al. 1995), the Hamilton Depres- disorders (Wassermann and Zimmermann 2012). sion Rating Scale (HDRS; Hamilton 1960), the Several studies have shown the potential anti- Clinical Global Impressions scale, severity (CGIs; craving effects of repetitive transcranial magnetic Guy 1976), and the Global Assessment of Func- stimulation (rTMS; Politi et al. 2008; Amiaz et al. tioning Scale (GAF; Endicott et al. 1976). The 2009; Mishra et al. 2010) and deep TMS (dTMS) OCDS is a 14-item self-rated scale; individual or in substance dependence (Rapinesi et al. 2013), combined items are Likert-like scales ranging from hence TMS may be effective for treating addictive 0 to 4, with higher scores refl ecting worse alcohol behaviours (Feil and Zangen 2010). dTMS, in consumption attitude. Its score ranges from 0 to 40, contrast to standard rTMS, by using a specially with 0– 20 ranges on each of the Obsessive and constructed H-coil, is able to stimulate neural tis- Compulsive subscales; there are no explicit cut-off sue beyond 1.5 cm from the scalp and allows for scores, but the tool is able to assess severity, clinical better precision. course, and treatment outcome (Anton et al. 1996). The other scales are clinician-rated; these were completed by trained clinicians at baseline, i.e., Objective before the fi rst session, at the end of the fi rst, sec- e aimed to investigate the effect of add-on dTMS ond, and third weeks of treatment, and at the end (dTMS-AO) on craving and mood in patients with of the 20-session treatment for the dTMS-AO comorbid DD and AUD treated with standard group (4 weeks) and at similar time-points for the detoxifi cation treatment (SDT), in the short-term SDT group (OCDS, HDRS, and CGIs, but not and in the long run. These patients were compared GAF), and at the end of the 6-month follow-up with a control group consisting of age-matched period (all scales). The HDRS is a 21-item scale patients receiving only SDT. whose fi rst 17 items are added to obtain the total 68 P. Girardi et al. score; single items are Likert, ranging 0 – 4 (8 items) consisted of 55 trains with a 2-s duration each and or 0 – 2 (9 items); 0– 7 is normal, 8– 13 mild depres- an inter-train interval of 20 s. The complete cycle of sion, 14 – 18 moderate depression, 19– 22 severe the dTMS treatment consisted of fi ve consecutive depression and Ն 23 very severe depression. A drop session days in a week for four consecutive weeks, of at least 50% from baseline scores is considered for a total of 20 sessions for each patient. The treat- as treatment response, while a score lower than 8 is ment was well tolerated by all patients. a remission. The HDRS interviews were carried-out by certifi ed clinicians whose interrater reliability Statistical analysis was 0.875 (Fleiss’ kappa). The CGIs ranges 1 (not at all ill) to 7 (extremely ill). Dropping to 1 or 2 Baseline socio-demographic and clinical character- (borderline ill) is considered an additional measure istics of the two groups of patients (dTMS-AO and of clinical remission or response. The GAF is a SDT) were compared with t -test for independent pseudocontinuous 1– 100 scale subdivided in ten groups for continuous variables (i.e., age, years of 10-point content layers with higher scores indicat- alcohol abuse) and Chi-squared test for categorical ing better psycho-socio-occupational functioning variables (i.e., gender, years of education, type of (American Psychiatric Association 2000, p. 34). treatment). For each psychopathological measure All patients were undergoing SDT for alcohol (HDRS, OCDS, CGI), score changes at the 1-, 2-, detoxifi cation which remained unmodifi ed through- 3-week (1w, 2w, 3w), post-treatment (p-t) and out the study (5– 20 mg/day diazepam, 75– 150 mg/ 6-months (6m) time-points with respect to baseline day trazodone, or a combination of the two). The 10 were computed. Changes were then analysed by dTMS-AO patients underwent add-on dTMS, fi ve mixed-model analysis of covariance (ANCOVA), sessions per week, for a total of 20 sessions for each with treatment (dTMS-AO vs. STD) and gender patient in a 4-week span. Another 10 patients with (male vs. female) as between-subject factors, time the same diagnoses, matched for age, educational (1w, 2w, 3w, p-t, 6m) as repeated-measures factor, level, drug treatment and score on the OCDS and and age, educational level and baseline value of the the HDRS, constituted the SDT group; all patients psychopathological measure as covariates. Green- were users of the same Service and signed a free house-Geisser (G-G) correction for repeated mea- informed consent for participation in the study. The sures was used to take into account possible study received approval from the Ethical Committee violation of the sphericity assumption. In case of of the Villa Rosa Hospital. signifi cant interaction treatment-by-time, multiple comparisons were performed by the Tukey test, to assess differences between treatments at each time- dTMS treatment point (fi ve comparisons), and differences between Add-on dTMS patients were abstaining from time-points within each treatment group (10 ϩ 10 alcohol for at least 1 month before the fi rst dTMS comparisons). As for the GAF, the change between session. The dTMS sessions were conducted using the 6m follow-up and baseline value was computed, Brainsway’ s H1 coil deep TMS System (Brainsway, and analysed by analysis of covariance including Har Hotzvim, Jerusalem, Israel). The H1 coil is treatment and gender as between-subject factors, designed to elicit neuronal activation in medial and and age, educational level, and baseline value of the lateral prefrontal regions, including the orbitofrontal psychopathological measure as covariates. Since cortex, with a preference for the left hemisphere only one time-point was considered, no multiple (Roth et al. 2007). The H1 coils were positioned over comparisons could be performed. Correlations the patient’ s scalp. The optimal spot on the scalp for between HDRS or OCDS and pharmacological stimulation of the right abductor pollicis brevis muscle treatments were sought through Pearson’ s r . Cut- was located, and the motor threshold was established off for statistical signifi cance was set at P Ͻ 0.05. All by delivering single stimulations to the motor cortex. P values were two-tailed. We used the IBM SPSS The motor threshold, defi ned as the lowest stimula- Statistics 21.0 (August 2012) and STATA 8.0 for tion intensity which produced fi ve motor evoked all analyses, as appropriate. potentials (MEPs) of at least 50 μ V in fi ve of 10 stimulations, was measured by gradually increasing Results stimulation intensity. The site of stimulation was located 5.5 cm anterior to the point at which maxi- Baseline socio-demographic and clinical character- mum stimulation of the abductor pollicis brevis muscle istics of the sample are summarised in Table I. was achieved. dTMS treatment was delivered by an There were no differences between the two groups expert physician in trains of 20 Hz at 120% of the (dTMS-AO and SDT) regarding the above-con- measured motor threshold. Each dTMS session sidered characteristics (Table I) and there was no dTMS in dysthymia plus alcoholism 69

Table I. Comparison of baseline socio-demographic and clinical characteristics of the two groups of patients.

dTMS-AO group SDT group Variables ( N ϭ 10) ( N ϭ 10) Test (value) P *

Age in years (mean Ϯ SD) 52.6 Ϯ 7.7 54.1 Ϯ 11.4 t -test (t ϭ 1.21) 0.244 Gender (% male) 50% 70% Chi-squared test (χ 2 ϭ 0.833) 0.361 Years of education (%) Յ 8 years 30% 20% Chi-squared test (χ 2 ϭ 0.573) 0.751 9 – 12 years 50% 60% Յ 13 years 20% 20% Years of alcohol abuse (mean Ϯ SD) 9.6 Ϯ 4.7 12.6 Ϯ 8.2 t -test (t ϭ 0.36) 0.727 Standard treatment (%) Diazepam 5– 25 mg/day 10% 30% Chi-squared test (χ 2 ϭ 1.291) 0.524 Trazodone 75– 150 mg/day 30% 20% Diazepam 5– 25 mg/day ϩ trazodone 75– 150 mg/day 60% 50% HDRS (mean Ϯ SD) 19.1 Ϯ 5.4 19.7 Ϯ 5.1 t -test (t ϭ – 0.26) 0.801 OCDS (mean Ϯ SD) 24.8 Ϯ 5.7 23.7 Ϯ 5.6 t -test (t ϭ 0.44) 0.668 CGIs [median (Q1– Q3)] 6.0 (5.0– 6.2) 6.0 (5.0– 6.2) Mann – Whitney U test (U ϭ 109.0) 0.791 GAF (mean Ϯ SD) 45.6 Ϯ 4.2 46.6 Ϯ 4.6 t -test (t ϭ – 0.51) 0.620

AO, add-on; SD, standard deviation; SDT, standard drug treatment; * signifi cance (chosen cut-off: P Ͻ 0.05). correlation between the pharmacological treat- 6.41, G-G P ϭ 0.0086). The Tukey test showed that ments and OCDS (r ϭ 0.03) or HDRS (r ϭ 0.12) the reduction of HDRS from baseline was signifi - total scores at the time of evaluation. cantly larger in the experimental than in the control group at the 1w, 2w, 3w and p-t time-points; no difference was observed between treatments as for the HDRS change at the 6m follow-up. When considering the Mean HDRS score differences from baseline in the comparisons within treatment group among time- two treatment groups are shown in Figure 1. We points, the change at 6m follow-up was signifi cantly found (i) a signifi cant main effect of treatment larger than that at 1w alone in the experimental (F (1,13) ϭ 11.74, P ϭ 0.0045), with the experimental group, while in the control group the change at 6m group showing a larger reduction in HDRS score follow-up was signifi cantly larger than those at all than the control group, (ii) a signifi cant main effect other time-points. of time ( F (4,64) ϭ 19.20, G-G p Ͻ 0.0001), with According to the HDRS, three dTMS-AO patients HDRS reduction increasing over time, and (iii) a (30%) were remitters (HDRS less than 8) at the end signifi cant interaction treatment-by-time (F (4,64) ϭ of treatment and one (10%) was responder (at least 50% drop from baseline), compared to none (0%) of the SDT patients at the 4-week time-point. How- ever, quite surprisingly, at the 6-month follow-up, fi ve dTMS-AO patients were remitters (50%) and one (10%) was responder, while in the SDT group there were six remitters (60%) and one responder (10%). There was a trend towards better response/ remission rate in the dTMS-AO vs . SDT group (Chi-squared test, χ 2 ϭ 3.662, P ϭ 0.056).

OCDS Mean OCDS score differences in the two treatment group are presented in Figure 2. Again, we observed (i) a signifi cant main effect of treatment (F (1,13) ϭ Figure 1. Mean drops on the HDRS from baseline, to 1st, 2nd, 85.21, P Ͻ 0.0001), with the experimental group 3rd, and 4th weeks of treatment (end of the 20-session cycle), and showing a larger improvement than the control to the 6-month follow-up (with no maintenance dTMS) in the group, (ii) a signifi cant main effect of time dTMS-AO and SDT groups. Error bars are SD; solid lines refer ϭ Ͻ to dTMS-AO, shaded lines to SDT. P -values refer to Tukey test ( F (4,64) 56.46, G-G P 0.0001), with the for multiple comparisons at each time-point (cut-off, P Ͻ 0.05; improvement increasing over time, and (iii) a * P Ͻ 0.05; * * P Ͻ 0.01). signifi cant interaction treatment-by-time (F (4,64) ϭ 70 P. Girardi et al.

Figure 2. Mean drops on the OCDS from baseline, to 1st, 2nd, Figure 3. Mean drops on the CGIs from baseline, to 1st, 2nd, 3rd, 3rd, and 4th weeks of treatment (end of the 20-session cycle), and and 4th weeks of treatment (end of the 20-session cycle), and to to the 6-month follow-up (with no maintenance dTMS) in the the 6-month follow-up (with no maintenance dTMS) in the dTMS-AO and SDT groups. Error bars are SD; solid lines refer dTMS-AO and SDT groups. Error bars are SD; solid lines refer to dTMS-AO, shaded lines to SDT. P -values refer to Tukey test to dTMS-AO, shaded lines to SDT. P -values refer to Tukey test for multiple comparisons at each time-point (cut-off, P Ͻ 0.05; for multiple comparisons at each time-point (cut-off, P Ͻ 0.05; * * P Ͻ 0.01). * P Ͻ 0.05; * * p Ͻ 0.01).

13.39, G-G P Ͻ 0.0001). The Tukey test showed that While post-treatment all dTMS-AO patients were the reduction of OCDS from baseline was signifi - remitters or responders according to the CGIs cantly larger in the experimental than in the control (100%), as compared with eight out of 10 (80%) in group at all time-points. When considering the with- the SDT group, at the 6-month follow-up, all patients in-treatment-group comparisons among time-points, (100%) regardless of whether they had dTMS or in the experimental group the change at 1w was sig- SDT alone, were in the responder/remitter range nifi cantly lower than those at 3w, p-t and 6m fol- (1– 2, normal to borderline ill). There was no differ- low-up; in addition, the change at 2w was signifi cantly ence in response/remission rate between the dTMS- lower than that at 6m follow-up alone. In the control AO and the SDT groups (χ 2 ϭ 0.117, P ϭ 0.732). group, all paired comparisons between time-points were signifi cant, except for 1w vs. 2w. GAF CGIs A main effect of treatment ( F (1,13) ϭ 29.64, P ϭ 0.0001) was observed, with the experimental CGIs score differences in the two treatment groups group showing a larger increase than the control are shown in Figure 3. We found (i) a signifi cant group in GAF score. main effect of treatment (F (1,13) ϭ 15.36, P ϭ 0.0018), with the experimental group showing a larger reduction in CGIs score than the control Discussion group, (ii) a signifi cant main effect of time ( F (4,64) ϭ 39.86, G-G P Ͻ 0.0001), with CGIs In this study we observed a rapid improvement of reduction increasing over time, and (iii) a signifi cant depressive and craving symptoms in patients with interaction treatment-by-time (F (4,64) ϭ 4.21, G-G comorbid DD and AUD with add-on dTMS, as P ϭ 0.0203). The Tukey test showed that the reduc- shown by HDRS and OCDS scores. This improve- tion of CGIs from baseline was signifi cantly larger ment, which started early on in the dTMS-AO group, in the experimental than in the control group at 1w, reached a maximum at the end of the 20-session 2w, and 3w; no difference was observed between cycle and was present at the 6-month follow-up, treatments as for the change at p-t and 6m follow-up. although no maintenance sessions were provided. When considering the comparisons within treatment The improvement at the end of the treatment was group among time-points, the change at 1w was sig- signifi cantly greater (faster) than the one obtained in nifi cantly lower than that at all other time-points in a standard drug detoxifi cation treatment comparison the experimental group; in the control group the group in the same time period. However, the advan- changes at 1w and 2w were signifi cantly lower than tage of add-on dTMS over standard detoxifi cation all others, and the change at 3w was signifi cantly treatment over all measures considered, disappeared higher than those at 1w and 2w, and signifi cantly at the 6-month follow-up for mood and general psy- lower than those at p-t and 6m follow-up. chiatric conditions, although it was maintained for dTMS in dysthymia plus alcoholism 71 craving and functioning. Although it may be argued bilaterally. One study, which employed left DLPFC that other effects could have played a role in the rTMS with 20 Hz stimulation frequency for 10 ses- maintenance of the response, such as both dTMS-AO sions, did not fi nd differences between real and sham and SDT groups improving because they were taken TMS in clinical measures in alcohol-dependent care of, or due to other Hawthorne-type effects, i.e., detoxifi ed patients (OCDS, HDRS, and Beck better performing while knowing to be observed or Depression Inventory), but the real rTMS group dif- cared for, or regression to the mean, or other effects fered from sham for an increased attentional blink that have been ascribed to placebo, or still that the response to alcohol-related cues after completion of real response to add-on dTMS was short-lived, the the TMS cycle (Hö ppner et al. 2011). Another study fact that a sham dTMS control group or a double- used right DLPFC rTMS with 10 Hz stimulation blind design were not employed may have obscured frequency vs . single-blind sham TMS in ICD-10 the factors involved in the response we obtained. A alcohol dependence syndrome patients and found a Hawthorne-like effect may be hypothesised to under- signifi cant reduction of craving, as assessed by the lie the timeϫ treatment interaction we observed with Alcohol Craving Questionnaire (ACQ-NOW), in the all assessment scales. Furthermore, the lack of main- real TMS group of add-on dTMS compared with tenance sessions for the rest of the follow-up period the sham-treated group (Mishra et al. 2010). may have played a role in the decreased differences Both 10-Hz rTMS for 3 weeks in drug-free patients between the experimental and the control group. As (George et al. 2010) and for 2 weeks in add-on for global functioning, measured with the GAF and (Huang et al. 2012) vs . sham TMS, as well as open- resulting from the combination of a psychological, label, H-coil 20-Hz dTMS for 4 weeks (Rosenberg social and occupational evaluation, a signifi cant ten- et al. 2010) and 18-week continuation (Harel et al. dency toward improvement from baseline to the end 2012), over the left DLPFC, proved to be useful in of treatment, which persisted at follow-up (unlike improving major depressive disorder. Yet, no specifi c the other measurements), was observed. GAF scores studies focused on DD; dysthymia symptoms were showed a greater improvement in the dTMS-AO reported to improve with 0.5 Hz slow rTMS in par- group than in the SDT group. The stimulation of the kinsonian patients (Dragaš evic et al. 2002). We here DLPFC may have improved the decisional abilities report the superior effectiveness of H-coil dTMS-AO (cfr. Hecht et al. 2010) of the patients who under- in patients with both DD and AUD in the short went dTMS, thus allowing them to cope better with term, compared to standard treatment alone; our everyday tasks; in turn, this may have resulted in the preliminary results need to be replicated in a larger improved functioning observed here. CGIs scores sample, using sham dTMS and continuation treat- did not allow differentiating the two groups as clearly ment to assess the value of this novel treatment in as the two psychometric scales, i.e., HDRS and the long run. OCDS; however, the scores were skewed towards the fl oor in the post-treatment and 6-month evaluations and this may have originated the lack of differences. Limitations of the study Interestingly, the dTMS-AO group had a better This was a one-centre, open-label study, and its lim- functioning outcome at these time-points, and this ited sample size refl ects this fact. Despite our long- may relate to improved social functioning induced term follow-up, we did not control for the possible by this technique, similarly to what has been shown effect of maintenance dTMS treatment, which could in people with autism (Enticott et al. 2011). possibly reveal an end-point difference between However, the main fi nding of this study is the dTMS-AO and SDT. Furthermore, we did not use accelerated response we obtained in the dTMS-AO a sham dTMS control group, and the two groups group; it is possible that early response may predict differed for number of encounters with the person- better outcome in this patient population, as it was nel, which could have boosted Hawthorne-like pla- shown to predict better response and higher remis- cebo effects of the procedure. sion rates (Henkel et al. 2009). The delayed antide- pressant response model has been recently questioned, with improvement within 2 weeks to predict later Conclusions stable responses/remissions (Szegedi et al. 2003). dTMS has not been tested in comorbid AUD and The add-on of dTMS to SDT and psychotherapy is DD, except for one small case series, reported by our associated with a signifi cant and rapid improvement group (Rapinesi et al. 2013). There are other types of the clinical picture, which is maintained at the of repetitive TMS (rTMS) treatments that have been 6-month follow-up in patients with comorbid diag- tested in alcoholism and alcohol craving, which do nosis of DD and AUD, although at this end-point it not use the H-coil and do not stimulate the DLPFC did not differentiate from SDT in all measures. 72 P. Girardi et al.

With respect to the SDT group, dTMS reduced References signifi cantly both craving and depression during the Alexander-Kaufman K , Cordwell S , Harper C , Matsumoto I . period it was administered. The dTMS effect was an 2007. A proteome analysis of the dorsolateral prefrontal cortex early one, signifi cantly better than SDT only, becom- in human alcoholic patients. Proteomics Clin Appl 1 : 62 – 72 . ing so from the fi rst week to the end of treatment American Psychiatric Association. 2000 . Diagnostic and Statisti- both for depression and craving (whereas SDT cal Manual of Mental Disorders, Fourth Edition-Text Revision (DSM-IV-TR). Washington, DC: American Psychiatric Asso- determined a reduction in depression from the third ciation . week to the end of treatment and of craving from the American Psychiatric Association. 2013 . Diagnostic and Statisti- second week to the end of treatment). cal Manual of Mental Disorders, Fifth Edition (DSM-5). Our results possibly suggest that in the absence Arlington, VA: American Psychiatric Association. of maintenance dTMS, the advantage over standard Amiaz R , Levy D, Vainiger D, Grunhaus L , Zangen A . 2009 . treatment may diminish after 6 months. Our fi nd- Repeated high-frequency transcranial magnetic stimulation over the dorsolateral prefrontal cortex reduces cigarette craving ings need replication in a wider sample and with a and consumption. Addiction 104 : 653 – 660 . protocol providing for maintenance treatment. Such Anton RF , Moak DH , Latham P . 1995 . The Obsessive Compul- maintenance could resolve the doubt whether the sive Drinking Scale: a self-rated instrument for the quantifi ca- effect of dTMS is linked to its constant administration of thoughts about alcohol and drinking behavior. Alcohol tion or whether it is subjected to tolerance with Clin Exp Res 19 : 92 – 99 . Anton RF , Moak DH , Latham PK . 1996 . The obsessive compul- time. Our data, despite being obtained in a small sive drinking scale: a new method of assessing outcome in alco- sample, suggest the possible utility of add-on dTMS holism treatment studies . Arch Gen Psychiatry 53 : 225 – 231, in the treatment of patients with comorbid DD and erratum 576. AUD. Bosch OG , Wagner M , Jessen F , K ü hn KU , Joe A , Seifritz E , et al . 2013. Verbal memory defi cits are correlated with prefrontal hypometabolism in 18 FDG PET of recreational MDMA users. Acknowledgements PLoS One 8 : e61234. Chang CC , Yu SC , McQuoid DR , Messer DF , Taylor WD , Singh The authors wish to thank Prof. Patrizio Pezzotti, K , et al . 2011 . Reduction of dorsolateral prefrontal cortex gray Istituto Superiore di Sanit à , Rome, Italy, for criti- matter in late-life depression. Psychiatry Res 193 : 1 – 6 . cally reviewing the manuscript and for helping with Draga š evic N , Potrebi A , Damjanovi A , Stefanova E , Kosti VS . 2002. Therapeutic effi cacy of bilateral prefrontal slow repetitive the statistics, Engineer David Hazani for continuous transcranial magnetic stimulation in depressed patients with support and care of the dTMS apparatus, Ms Mimma Parkinson ’ s disease: an open study. Mov Disord 17 : 528 – 532 . Ariano, Ms Ales Casciaro, Ms Teresa Prioreschi, and Drummond DC . 2001 . Theories of drug craving, ancient and Ms Susanna Rospo, Librarians of the Sant’ Andrea modern . Addiction 96 : 33 – 46 . Hospital, School of Medicine and Psychology, Sapi- Endicott J , Spitzer RL , Fleiss JL . 1976 . The global assessment scale . A procedure for measuring overall severity of psychiatric enza University, Rome, for rendering precious bib- disturbance. Arch Gen Psychiatry 33 : 766 – 771 . liographical material accessible, as well as their Enticott PG , Kennedy HA , Zangen A , Fitzgerald PB . 2011 . Deep Secretary Lucilla Martinelli for her assistance during repetitive transcranial magnetic stimulation associated with the writing of the manuscript. improved social functioning in a young woman with an autism spectrum disorder. J ECT 27 : 41 – 43 . Ernst LH , Plichta MM , Dresler T , Zesewitz AK , Tupak SV , Hae- Statement of interest ussinger FB , et al . 2014 . Prefrontal correlates of approach pref- erences for alcohol stimuli in alcohol dependence. Addict Biol. This work has not been supported by any funding. 19 : 497 – 508 . In the past 2 years, Paolo Girardi has received Feil J , Zangen A . 2010 . Brain stimulation in the study and treat- research support from Lilly, Janssen, and Springer ment of addiction. Neurosci Biobehav Rev 34 : 559 – 574 . Healthcare, and has participated in Advisory Boards First MB, Gibbon M, Spitzer RL, Williams JBW, Benjamin LS. 1997. Structured Clinical Interview for DSM-IV Axis II for Lilly, Otsuka, Pfi zer, Schering, and Springer Personality Disorders, (SCID-II). Washington, DC: American Healthcare and received honoraria from Lilly and Psychiatric Press. Springer Healthcare. Ruggero N. Raccah is scien- First MB , Spitzer RL , Gibbon M , Williams JBW . 2002 . Structured tifi c consultant to ATID Ltd, distributor of deep Clinical Interview for DSM-IV Axis I Disorders, Research Ver- r-TMS (Brainsway) technology in Italy. All other sion, Patient Edition . New York, NY: Biometric Research, New York State Psychiatric Institute, A 1 – 46 . authors of this paper have no relevant affi liations or George MS , Lisanby SH , Avery D , McDonald WM , Durkalski fi nancial involvement with any organization or entity V , Pavlicova M , et al . 2010. Daily left prefrontal transcranial with a fi nancial interest in, or fi nancial confl ict with magnetic stimulation therapy for major depressive disorder: a the subject matter or materials discussed in the sham-controlled randomized trial. Arch Gen Psychiatry 67 : manuscript. This includes employment, consultan- 507 – 516 . Grant BF , Stinson FS , Dawson DA , Chou SP , Dufour MC , cies, honoraria, stock ownership or options, expert Compton W, et al . 2004 . Prevalence and co-occurrence of testimony, grants or patents received or pending, or substance use disorders and independent mood and anxiety royalties. disorders: results from the National Epidemiologic Survey on dTMS in dysthymia plus alcoholism 73

Alcohol and Related Conditions. Arch Gen Psychiatry Oh DH , Son H , Hwang S , Kim SH . 2012 . Neuropathological 61 : 807 – 816 . abnormalities of astrocytes, GABAergic neurons, and pyrami- Guy W . 1976. ECDEU Assessment Manual of Psychopharmacol- dal neurons in the dorsolateral prefrontal cortices of patients ogy – Revised (DHEW Publ No ADM 76– 338) . Rockville, MD, with major depressive disorder. Eur Neuropsychopharmacol USA: Department of Health, Education, and Welfare, Public 22 : 330 – 338 . Health Service, Alcohol, Drug Abuse, and Mental Health Olbrich HM , Valerius G , Paris C , Hagenbuch F , Ebert D , Administration, NIMH Psychopharmacology Research Branch, Juengling FD . 2006 . Brain activation during craving for alcohol Division of Extramural Research Programs; p. 218 –222 . measured by positron emission tomography. Aust NZ J Hamilton M . 1960 . A rating scale for depression. J Neurol Psychiatry 40 : 171 – 178 . Neurosurg Psychiatry 23 : 56 – 62 . Park MS , Sohn JH , Suk JA , Kim SH , Sohn S , Sparacio R . 2007 . Harel EV , Rabany L , Deutsch L , Bloch Y, Zangen A , Levkovitz Y . Brain substrates of craving to alcohol cues in subjects with 2012. H-coil repetitive transcranial magnetic stimulation for alcohol use disorder. Alcohol Alcohol 42 : 417 – 422 . treatment resistant major depressive disorder: an 18-week con- Park SQ , Kahnt T , Beck A , Cohen MX , Dolan RJ , Wrase J , et al . tinuation safety and feasibility study. World J Biol Psychiatry. 2010. Prefrontal cortex fails to learn from reward prediction 15(4): 298 – 306 . errors in alcohol dependence. J Neurosci 30 : 7749– 7753. Hasin DS , Stinson FS , Ogburn E , Grant BF . 2007. Prevalence, Politi E , Fauci E , Santoro A , Smeraldi E . 2008 . Daily sessions of correlates, disability, and comorbidity of DSM-IV alcohol transcranial magnetic stimulation to the left prefrontal cortex abuse and dependence in the United States: results from the gradually reduce cocaine craving. Am J Addict 17 : 345 – 346 . National Epidemiologic Survey on Alcohol and Related Condi- Rapinesi C , Kotzalidis GD , Serata D , del Casale A , Bersani FS , tions . Arch Gen Psychiatry 64 : 830 – 842 . Solfanelli A , et al . 2013. Effi cacy of add-on deep transcranial Hecht D, Walsh V, Lavidor M . 2010 . Transcranial direct current magnetic stimulation in comorbid alcohol dependence and stimulation facilitates decision making in a probabilistic guess- dysthymic disorder: three case reports. Prim Care Compan- ing task . J Neurosci 30 : 4241– 4245. ion CNS Disord 15(1) pii : PCC.12m01438; doi:10.4088/ H ö ppner J , Broese T , Wendler L , Berger C , Thome J . 2011 . Repet- PCC.12m01438. itive transcranial magnetic stimulation (rTMS) for treatment Ravindran AV , Smith A , Cameron C , Bhatla R , Cameron I , Geor- of alcohol dependence. World J Biol Psychiatry 12(Suppl 1): gescu TM , et al . 2009. Toward a functional neuroanatomy of 57 – 62 . dysthymia: a functional magnetic resonance imaging study. J Huang ML , Luo BY , Hu JB , Wang SS , Zhou WH , Wei N , et al . Affect Disord 119 : 9 – 15 . 2012. Repetitive transcranial magnetic stimulation in combina- Rosenberg O, Shoenfeld N , Zangen A , Kotler M , Dannon PN . tion with citalopram in young patients with fi rst-episode major 2010. Deep TMS in a resistant major depressive disorder: a depressive disorder: a double-blind, randomized, sham- brief report. Depress Anxiety 27 : 465 – 469 . controlled trial. Aust NZ J Psychiatry 46 : 257– 264 . Roth Y, Amir A , Levkovitz Y, Zangen A . 2007 . Three-dimensional Kessler RC , McGonagle KA , Zhao S , Nelson CB , Hughes M , distribution of the electric fi eld induced in the brain by tran- Eshleman S , et al . 1994 . Lifetime and 12-month prevalence of scranial magnetic stimulation using fi gure-8 and deep H-coils . DSM-III-R psychiatric disorders in the United States. Results J Clin Neurophysiol 24 : 31 – 38 . from the National Comorbidity Survey. Arch Gen Psychiatry Szegedi A , M ü ller MJ , Anghelescu I , Klawe C , Kohnen R , Benk- 51 : 8 – 19 . ert O . 2003 . Early improvement under mirtazapine and parox- Lang N , Hasan A, Sueske E , Paulus W , Nitsche MA . 2008. Corti- etine predicts later stable response and remission with high cal hypoexcitability in chronic smokers? A transcranial magnetic sensitivity in patients with major depression . J Clin Psychiatry stimulation study. Neuropsychopharmacology 33: 2517 – 2523 . 64 : 413 – 420 . Makris N , Oscar-Berman M , Jaffi n SK , Hodge SM , Kennedy Townsend J , Bookheimer SY , Foland-Ross LC , Sugar CA , DN , Caviness VS , et al . 2008 . Decreased volume of the brain Altshuler LL . 2010 . fMRI abnormalities in dorsolateral pre- reward system in alcoholism. Biol Psychiatry 64 : 192 – 202 . frontal cortex during a working memory task in manic, euthymic Mishra BR , Nizamie SH , Das B , Praharaj SK . 2010. Effi cacy of and depressed bipolar subjects. Psychiatry Res 182 : 22 – 29 . repetitive transcranial magnetic stimulation in alcohol depend- Wassermann EM , Zimmermann T . 2012. Transcranial magnetic ence: a sham-controlled study. Addiction 105 : 49 – 55 . brain stimulation: therapeutic promises and scientifi c gaps. Moreno-Ló pez L , Stamatakis EA , Ferná ndez-Serrano MJ , Pharmacol Ther 133 : 98 – 107 . G ó mez-R í o M , Rodr í guez-Fern á ndez A , P é rez-Garcí a M , et al . World Health Organization. 1992 . The ICD-10 Classifi cation of 2012. Neural correlates of the severity of cocaine, heroin, alco- Mental and Behavioral Disorders: Clinical Descriptions and hol, MDMA and cannabis use in polysubstance abusers: a Diagnostic Guidelines. Geneva: World Health Organization. resting-PET brain metabolism study. PLoS One 7 : e39830. Ye T, Peng J, Nie B, Gao J, Liu J, Li Y, et al. 2012. Altered func- O ’ Brien CP . 2005 . Anticraving medications for relapse preven- tional connectivity of the dorsolateral prefrontal cortex in fi rst- tion: a possible new class of psychoactive medications. Am J episode patients with major depressive disorder. Eur J Radiol Psychiatry 162 : 1423– 1431. 81:4035– 4040.