Efficacy of Continuous Theta Burst Stimulation to the Pre-supplementary Motor Area in Gambling Disorder: a Randomized Double-Blind Controlled Trial

Abstract

Gambling disorder (GD) is a condition constituting a public health concern due to its impactful detrimental consequences for affected individuals and its very high social costs. Given the lack of FDA-approved treatments, novel non-pharmacological interventions are warranted. In this double-blind randomized controlled trial, we aimed to investigate the effects of continuous Theta Burst Stimulation (cTBS) in patients with GD. Twenty-nine patients with a diagnosis of GD were included and randomly assigned to active or sham condition. In the active condition, participants received 10 sessions of cTBS over pre-SMA bilaterally. Patients were evaluated at baseline and at the end of treatment with the Pathological Gambling version of the Yale-Brown Obsessive-Compulsive Scale (PG-YBOCS) as primary outcome measure, and with the following secondary measures: Hamilton Anxiety Scale (HAM-A), Hamilton Depression Scale (HAM-D), Gambling Urges Scale (GUS), Barratt Impulsiveness Scale (BIS-11), Sheehan Disability Scale (SDS), Fagerstrom Test for Nicotine Dependence (FTND). A two-way repeated measures ANOVA revealed a significant “group” × “time” interaction for PG-YBOCS and GUS measures. Within group comparisons revealed that PG-YBOCS scores (p<.001) and GUS scores (p<.001) improved significantly in the group receiving the active stimulation, and positively correlated with a reduction of gambling-related disability, as measured with the SDS. This study shows that cTBS could be an effective treatment for GD, since it reduces one of its main dimensions: craving. Trial registration: ClinicalTrials.gov Identifier: NCT03669315

Gambling disorder (GD) concerns the urge of gambling and is characterized by the inability to control/stop to gamble (APA, 2013). It is a main health concern (David et al., 2020) due to its impactful detrimental consequences for affected individuals, their families and its very high social costs (Walker and Sobel, 2016). It is a widespread disorder with a prevalence of 1.3–7.7% (Harrison et al., 2018), whose impact can be even exacerbated by the COVID-19 pandemic (Salerno & Pallanti, 2021). In light of its clinical and neurobiological similarities to substance addiction (Potenza et al., 2006; Pallanti, 2006; Potenza et al., 2009; Leeman and Potenza, 2012; Goundriaan et al., 2014), GD is considered a behavioural addiction and has been categorized in the DSM-5 (APA, 2013) under the category of “substance-related and addictive disorders”. Pivotal mechanisms of GD are craving, reward system alterations and decreased inhibition (Potenza et al., 2019; Antons et al., 2020). Specifically, craving has been described as a predictor of GD severity, number of gambling episodes, chasing persistence and has been associated with mood alterations and negative urgency (Mallorquì-Bagué et al., 2023; Quintero et al., 2020). Furthermore, high levels of response impulsivity have been correlated to the severity of GD (Mestre-Bach et al., 2020). A meta-analysis showed that GD was associated with significant impairments in motor and attentional inhibition, discounting and decision-making tasks. Indeed, neuroimaging studies have found in people with GD alterations in decision-making processes (Pettorruso et al., 2019; Winstanley & Clark, 2015), reward mechanisms (Clark et al., 2019) and faulty inhibitory control mechanisms (Verdejo-Garcia et al., 2015).

Non-Invasive Brain Stimulation (NIBS) has been shown to be particularly effective for impulsive-compulsive spectrum disorders and substance use disorders (SUDs) (Stein et al., 2019; Ekhtiari et al., 2019). Repetitive Transcranial Magnetic Stimulation (rTMS), which uses magnetic fields to stimulate nerve cells in the brain, emerged as being able to modulate both decision-making, reward functioning and inhibitory control (Juan and Muggleton, 2012; Stanford et al., 2013; Levasseur-Moreau and Fecteaus, 2012) which are all altered functions in GD, as reported before. Regarding the current evidence about the application of TMS in GD, Pettorusso et al. (2019) in an open-label study treated patients with GD with rTMS at high frequency (15 Hz) over the left DorsoLateral Prefrontal Cortex (DLPFC). At the end of the treatment, a reduction in symptoms severity was observed. However, the low sample size and the lack of control group did not allow to draw clear conclusions. A recent preliminary study with continuous Theta Burst Stimulation (cTBS) in 6 patients with GD showed the potential beneficial effect of TBS over the pre-Supplementary Motor Area (pre-SMA) (Salerno et al., 2022), an area with a role in risky decisions and inhibitory control (Obeso 2017; Tosun 2017). No study with rTMS in GD adopted a randomized double-blind design.

TBS is a brief form of rTMS, which typically lasts for 2–3 min, which mimics the theta bursts recorded from the hippocampus (Suppa et al., 2016). It can be intermittent, which increases cortical excitability, or continuous, which decreases cortical excitability (Huang et al., 2005). Different studies showed that it could be effective in the treatment of depression (Williams et al., 2018), schizophrenia (Pallanti et al., 2022), addiction (Su et al., 2020) and, as reported before, in a preliminary study, in GD (Salerno et al., 2022). Since no FDA-approved treatment exists for GD, new treatment options are needed. In this randomized, controlled and double-blind study, we aimed to investigate the effects of cTBS over the pre-SMA in patients with GD, assessing symptoms severity before and after treatment. We hypothesized that GD could be treated with cTBS targeting the altered connectivity (e.g. inhibitory control networks), with the aim to reduce craving and increase inhibitory control over impulses and behaviours. We expected to observe a reduction in gambling urge and impulsivity.

Methods

Participants

Twenty-nine patients (8 females, mean age: 44.2, range 21–66) with a diagnosis of GD, according to criteria of the DSM-5 (APA, 2013) assessed through clinical interview with a licensed clinician, were consecutively admitted at the Institute of Neuroscience (Florence, Italy). Participants were recruited between September 2018 and December 2019. All participants had a history of illness of at least 1 year. Exclusion criteria were comorbidity for bipolar disorders, psychotic disorders, history of seizures or epilepsy, in vivo diamagnetic metal implants such as cardiac pacemakers, pregnancy and neurological disorders. The study was designed as double-blind, controlled and randomized. Participants as well as expert clinicians assessing them were blind about the treatment condition. Participants were randomly assigned to active (continuous TBS over the pre-SMA bilaterally) or sham conditions. Random assignment was done by computerized random number generator. In agreement with the 1964 Helsinki declaration, the study was approved by the local ethics committee (ref. 11992) and, after complete description of the study to the subjects, written informed consent was obtained.

Procedure

cTBS was administered with the MagVenture MagPro R30 stimulator with add-on Theta Burst option (MagVenture INC.) using a Cool D-B80 figure-of-eight coil. Sham stimulation was performed with a sham-coil (MCF-B 65, MagVenture), which replicates stimulation sounds, and mimics the sensation of magnetic stimulation. cTBS consists of bursts of 3 pulses separated by 20ms (i.e. 50 Hz). Stimulus intensities were set at 80% of Resting Motor Threshold (RMT). Two trains of 600 pulses (a total of 1200 pulses) were used with an 8-s intertrain interval. The bilateral pre-SMA was targeted using individual MRI and a neuronavigation system (SofTaxic Optic 2.0) (see Fig. 1). RMT was defined as the minimum magnetic flux needed to elicit a threshold EMG response (50 mV in peak-to-peak amplitude) in a resting target muscle (abductor pollicis brevis or APB) in 5/10 trials using single-pulse TMS administered to the contralateral primary motor area.

Fig. 1.

The anatomical location of the pre-supplementary motor area (pre-SMA) and its topographical relationship with the supplementary motor area (SMA) is illustrated on a midsagittal T1-MRI view in a representative human hemisphere. The pre-SMA is located in the mesial aspect of the hemisphere and is constituted mainly by Brodmann areas 6 and 32. It is delimited by the cingulate sulcus inferiorly (solid white line), the hemispheric margin superiorly (h.m., dotted white line), a coronal plane at the genu of the corpus callosum anteriorly (represented by a vertical dotted black line) and a coronal plane at the anterior commissure posteriorly (represented by a vertical dotted black line at the anterior commissure) (see Nachev 2005, 2007, 2008; Salerno et al., 2022). Pre-SMA was the target in this study using cTBS. Image from the Center for Morphometric Analysis (Massachusetts General Hospital, Harvard Medical School)

Psychometric Measures

Participants were assessed by expert clinicians blinded to the treatment condition at the baseline and at the end of the treatment with the Pathological Gambling version of the Yale-Brown Obsessive-Compulsive Scale (PG-YBOCS; Pallanti et al., 2005), the Hamilton Anxiety Scale (HAM-A; Maier et al., 1988) and the Hamilton Depression Scale (HAM-D; Beck et al., 1981), the Gambling Urges Scale (GUS; Raylu et al., 2004), the Barratt Impulsiveness Scale (BIS-11; Fossati et al., 2001), the Sheehan Disability Scale (SDS; Leon et al., 1997) and the Fagerstrom Test for Nicotine Dependence (FTND; Healtherton et al., 1991). At the end of the treatment, patients were re-assessed by a licensed psychiatrist to verify if they still met the DSM-5 criteria for a diagnosis of GD.

Statistical Analysis

A two-way repeated measures ANOVA was run to assess the interaction between groups and time. Depending on the distribution, assessed with the Shapiro-Wilk test a parametric (t-test) or non-parametric test (Wilcoxon test) was run to determine if there were differences in PG-YBOCS and the other rating scales before treatment between the two groups. T-test or Wilcoxon test was also used for within group comparison in time (pre- vs post-treatment). Analyses were performed with R software (version 4.0.5) (R Core Team. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing, 2021). Statistical significance was set at p<.05, two-sided. Bonferroni correction for multiple comparisons was applied.

Results

For each group, the numbers of participants who were randomly assigned as well as their demographical data are reported in Table 1.

Table 1.

Demographical data of the two groups (Active and Sham)

Group	Sample	Gender	Age
Active	16	4 F, 12 M	Mean= 42.71; SD=15.51
Sham	13	4 F, 9 M	Mean= 42.77; SD= 12

The comparison of baseline measures between the two groups (active vs sham) was not significant for any measure, meaning that the two groups did not differ before the treatment. When the same analysis was repeated for post-treatment measures, PG-YBOCS (p-value<.001) as well as the HAM-A (p<.05), the GUS (p<.01) and the SDS (p<.01) scores resulted being statistically different between the two groups. ANOVA analysis revealed a significant “group” × “time” interaction for PG-YBOCS (F=6.571; p<0.05) and GUS measures (F=5.045; p<0.05). Depending on the normality test results, one sample paired t-test or Wilcoxon test was run to compare measures before and after treatment within the real stimulation group. PG-YBOCS scores resulted to be significantly different (V=136, p<0.001, r=0.671) as well as the GUS scores (t=4.76, p<0.001, d=1.19) (see Figs. 2 and 3). Importantly, 93% of patients (15 out of 16) showed a reduction in PG-YBOCS scores of at least 35%, which is the threshold for the definition of full response to treatment in impulsive disorders (Pallanti et al., 2002). Furthermore, SDS scores improved significantly after the treatment (t=3.9, p<0.001, d=0.72) and positively correlated with the reduction in GD severity (R=0.69, p<0.001) and craving (R=0.46, p<0.05). Improvement in PG-YBOCS scores positively correlated with GUS score change (R=0.49, p<0.01). Other measures did not show an improvement after the treatment. All patients were re-assessed at the end of the treatment by a licensed psychiatrist. It was shown that those who showed an improvement of at least 35% in symptoms severity did not meet anymore the criteria for a diagnosis of GD, i.e. 15 out of 16 patients, according to the DSM-5 criteria.

Fig. 2.

Pathological Gambling–Yale-Brown Compulsive Obsessive Scale (PG-YBOCS) median scores before (pre-test) and after treatment (post-test) in the two groups: active vs sham stimulation

Fig. 3.

Gambling Urges Scale (GUS) mean scores before (pre-test) and after treatment (post-test) in the two groups: active vs sham stimulation

Discussion

This is the first randomized, double-blind study assessing the effect of MRI-guided cTBS over the pre-SMA for the treatment of GD. We reported a reduction in symptom severity in the active compared to the placebo group, as shown by the significant reduction of the PG-YBOCS score after only 10 sessions of cTBS, along with that of GUS. The results of this study confirm the results of the previous proof of concept (Salerno et al., 2022) strengthening their importance due to the presence of a control arm and the larger sample size. The clinical value of these results is mainly represented by the improvement of one of the main dimensions of GD, that is craving. Indeed, both the PG-YBOCS and GUS scores showed a reduction, after treatment, in thoughts related to gambling and urges, meaning a decrease in craving. It is noteworthy to mention that these findings were obtained after only 10 sessions of cTBS. The lack of significant changes in impulsivity, as measured by the BIS, was not unpredictable, since BIS measures mainly the trait impulsiveness, which is hardly modifiable. Overall, this study suggests cTBS as a new therapeutic option for the treatment of GD.

Previously, studies adopting rTMS in GD mainly targeted the left DLPFC at high frequency (Pettorusso et al., 2019, 2020, 2021). The main result was the improvement of gambling-related decision-making. However, at our knowledge, no studies adopted a double-blind study with a control group, which represents the main limitation of previous studies (Pettorusso et al. 2020) for the generalizability of their results. Other studies, such as Gay (2017), reported a reduction of craving, but only after one single session of TMS over the right PFC, which did not result in a reduction of gambling behaviour. cTBS of the right dorsolateral PFC has been adopted by Zack et al. (2016). However, the treatment did not result in a decrease in urges and bet size, possibly due to the target of the stimulation.

About the clinical significance of craving, it should be noted that it represents a very real phenomenon, which impacts patients’ daily lives, dominating their thoughts and causing distress (Tiffany and Wray, 2012; Mallorquí-Bagué et al., 2023). In this study, the intervention leads to the amelioration of craving based on the premise that craving may increase drug usage or serve as a sign of relapse. Therefore, a treatment that directly reduces craving distress should reasonably promote long-term reductions in drug use, and, importantly, enhance quality of life. Coherently, we have shown that a reduction in craving as well as in symptoms’ severity was correlated to a reduction in the related disability measured through the SDS, which assesses the impacts of a disorder in everyday life, work and social relationships.

While the present results are strictly clinical, we may hypothesize some mechanisms for the described effects. Considering that TMS can influence the activity of brain areas distant from the site of stimulation (Siebner et al., 2009), TBS could have had not only a direct effect on the pre-SMA and inhibitory control, but also on the striatum, implicated in key gambling behaviours (Pallanti et al., 2010), such as monetary choices (Hollander et al., 2005) and on regions implicated in the reward system, such as the anterior cingulate cortex (Hartstra et al., 2010). Indeed, cTBS over the pre-SMA could act by modulation of inhibitory control circuitries (Borgomaneri et al., 2020; Obeso et al., 2017), by increasing cognitive shifting ability or improving decision-making (Georgiev et al., 2016). Furthermore, cTBS applied to the pre-SMA has shown to increase the efficiency of inhibitory control over prepotent ongoing responses in healthy subjects (Obeso et al. 2013) and in improving decision-making (Tosun et al., 2017). Future studies should address this point with neuroimaging, in order to understand at which neural level changes occurred.

This study overcame some of the main limitations of our previous report (Salerno et al., 2022) since this is a double-blind study and introduced a control arm. However, some limitations are still identifiable, concerning the lack of a follow-up, the short duration of the treatment protocol and the small sample size. Furthermore, future studies should include neuropsychological assessment and effective measures of win/losses of gamblers. Larger multicentric studies will be needed to confirm the efficacy of cTBS on the pre-SMA in GD.