Effectiveness of accelerated intermittent theta burst stimulation for social cognition and negative symptoms among individuals with schizophrenia: A randomized controlled trial

.


Background
Schizophrenia is a leading cause of mental disability and is characterized by impairments in multiple functional domains (Abplanalp et al., 2022). The 2019 Global Burden of Disease Study reported that the number of individuals with schizophrenia cases has increased to 23.6 million, contributing 12.2% of the global disability-adjusted life years (DALYs) (2022). The psychopathological dimensions of schizophrenia include positive, negative, and cognitive symptoms (Sommer et al., 2022). Most studies have demonstrated the effectiveness of antipsychotics in alleviating the positive symptoms of schizophrenia, but treatment of the negative and cognitive symptoms is limited (Huhn et al., 2019). Compared with positive symptoms, negative and cognitive symptoms have a greater impact on the quality of life and social functioning of individuals with schizophrenia, and are also the main reasons for the chronic duration of illness (He et al., 2022).
Evidence indicates that in cognitive symptoms of schizophrenia, social cognition plays an important key role in patients' social functioning and interpersonal interactions, which is the direction of the latest research (Green et al., 2019;van Hooren et al., 2008). Social cognition refers to psychological activities during social interactions, which involve perceiving, interpreting and generating reactions to the intentions, tendencies and behaviours of others (Green et al., 2008), including emotional processing, theory of mind, social perception, and attribution bias (Pinkham et al., 2014). Although social cognition and non-social cognition share some overlapping cognitive processes, indeed, social cognitive impairment has a more-negative effect on daily functioning than does non-social cognitive impairment . Furthermore, emotion recognition and theory of mind are the two most important domains of social cognition in schizophrenia. (Tan et al., 2018). The former involves identifying, managing, understanding, and interpreting emotions, while the latter involves understanding and predicting others' beliefs, intentions, and behaviours (Lim et al., 2021b). These dysfunctions are particularly associated with difficulties in conflict resolution ability, interpersonal relationships, social interactions (Green et al., 2012).
However, there are limited studies on intervention for social cognition and negative symptoms in schizophrenia. Psychopharmacological studies suggest that there are different opinions on the use of antipsychotics for the improvement of negative symptoms, and the effectiveness for social cognition may be even less promising (Kucharska-Pietura and Mortimer, 2013;Na et al., 2015). Simultaneously, the interaction of antipsychotics with various neurotransmitters in the brain may lead to sedation, agitation, compulsion, tremor and other adverse drug reactions (Skokou et al., 2022). Although psychosocial interventions may be a feasible way to improve social cognition and negative symptoms, they require substantial time and presents objective difficulties in their promotion (Solmi et al., 2022a;Wang et al., 2013). Therefore, it is of great clinical value to explore simple and effective interventions to improve social cognition and negative symptoms in individuals with schizophrenia (Solmi et al., 2022b). Noninvasive brain stimulation (NIBS) can induce neuronal action potentials by physically regulating the brain, or modulates spontaneous neuronal activity by a tonic depolarization or hyperpolarization, which is helpful to improve negative symptoms, depressive symptoms, or cognitive impairment. Huhn et al., 2019). Due to its noninvasive characteristics, NIBS has gradually attracted attention in the clinical treatment and research of mental diseases (Lefaucheur et al., 2014). At present, repetitive transcranial magnetic stimulation (rTMS) in NIBS has gotten the most traction in psychiatry as a technique, while intermittent theta burst stimulation (iTBS) is a more novel form of rTMS (Chu et al., 2021). ITBS is a new form of NIBS, that is characterized by the delivery of 600 pulses within 3 minutes (Di Lazzaro et al., 2011). It produces an increase in motor cortex excitability in a short time, and its excitatory effect may be superior to that of rTMS (Blumberger et al., 2018). Some studies show that prefrontal rTMS may help to improve negative symptoms in individuals with schizophrenia (Freitas et al., 2009;Shi et al., 2014). Wobrock et al. (2015) believed that active 10-Hz rTMS was not superior compared with sham rTMS in improving negative symptoms. Additionally, iTBS can increase the executive function of patients with medication-resistant depression (Cheng et al., 2016), alleviate the decline on cognitive and memory abilities, and improve working memory performance (Debarnot et al., 2015;Hoy et al., 2016). Most studies of major depressive disorder (MDD) or autism spectrum disorder (ASD) have proven that iTBS is a safe and effective neuroregulatory technique (Chung et al., 2015;Demirtas-Tatlidede et al., 2010).
We attempted to explore the efficacy of accelerated iTBS for social cognition and negative symptoms in a randomized controlled trial based on a sample of individuals with chronic schizophrenia. If this hypothesis can be confirmed, it will provide insights for the use of iTBS in the field of clinical neuroregulation for schizophrenia.

Study design
The study was a randomized, double-blind, controlled clinical trial. It was based on the Neuroscience Laboratory of Shanghai Pudong Mental Health Center, School of Medicine, Tongji University. The sample size of the study was calculated by sample size power analysis software (PASS version 21.0.3 [NCSS LLC,Utah,USA]). Based on our preliminary experiment, we set the significance level as 0.05, the power value as 0.8, the standard deviation as 1.36, and a two-sided test was needed. The samples were divided into two groups: the active iTBS and sham iTBS groups, with at least 23 participants assigned to each arm. Considering that participants may be unable to drop out or complete the test for special reasons, we set a 20% loss rate and ultimately determined a sample size of 30 people in each group. The study protocol was registered with the Chinese Clinical Trial Registry (ChiCTR2100051984).

Participants
All participants were recruited from the Shanghai Pudong Mental Health Center, School of Medicine, Tongji University, from November 1, 2020, to June 30, 2021. The inclusion criteria were as follows: (1) individuals who met the DSM-5 (Roehr, 2013) diagnostic criteria for schizophrenia; (2) individuals aged between 18 years and 65 years; (3) individuals with a course of disease ≥ 5 years; (4) individuals who were right-handed; (5) individuals who were treated with a stable dose of olanzapine for at least 6 months before the iTBS intervention and until the follow-up was completed; (6) individuals who were able to cooperate with treatment and complete various assessments independently; (7) individuals who were able to express themselves through spoken or written language; and (8) participants with signed informed consent forms for participation in the study. The exclusion criteria were as follows: (1) patients who met the DSM-5 (Roehr, 2013) diagnostic criteria for brain organic mental disorder, dementia, or intellectual disability; (2) patients with severe brain trauma, neurological disorders, or physical illness; (3) history of metal implantation in the body; (4) patients with severe visual, hearing or language impairments; (5) patients who received treatment with iTBS, rTMS, or MECT within 3 months; (6) patients with changed or combined use of other antipsychotics or drug dosage adjustments during the observation period; (7) patients with a history of vertigo, syncope, epilepsy, or alcohol or drug dependence; and (8) participants or guardians did not sign the study informed consent form and those who dropped out.

Randomization
The subjects were sorted and numbered according to the time of enrolment. A researcher who did not participate in the assessments or stimulation randomly sampled the numbers through SPSS 26.0 software (SPSS, Inc., Chicago, IL, USA) and divided them into the active iTBS group and the sham iTBS group at a ratio of 1:1. After the subjects were enrolled in the group, they were assigned a ranking number according to enrolment time. Finally, they were assigned to the corresponding groups according to the results of prior random sampling. Neither the participants nor the investigators were aware of the groupings except for randomization and stimulation.

Accelerated intermittent theta burst stimulation (Accelerated iTBS)
Accelerated iTBS was performed using a Brain Ultimate M-100 transcranial magnetic stimulator (Brain Ultimate, Atlanta, USA) with a 70-mm air-cooled butterfly coil. The accelerated iTBS parameters were as follows: 3 pulses, 50 Hz bursts, 2 s on and 8 s off; 600 pulses per session; total duration of 3 min 9 s, and a resting motor threshold (RMT) intensity of 120% (Blumberger et al., 2018). Considering that high-dose TBS may cause patient discomfort (Williams et al., 2018), we applied three iTBS rounds per treatment day and separated by 15 minutes (total: 1800 pulses/day). Simultaneously, using the TMS Navigation System (Brain Ultimate, Atlanta, USA) to accurately locate the coil, the coil was placed with the Visor 2.0 and tangential to the skull with the handle pointing in the posterolateral direction. The stimulation target was located in the left dorsolateral prefrontal cortex (DLPFC), which may be an effective target for individuals with schizophrenia (Kumar et al., 2020). Treatment was performed in a sitting or semirecumbent position. Treatment was performed once a day, 5 times per week, for 4 weeks, yielding a total of 20 sessions. All instruments and parameters were standardized and adjusted by the manufacturer. During treatment, subjects were required to wear noise-cancelling earplugs and remove all objects sensitive to magnetic fields, such as watches, cell phones, jewellery, and metal dentures.

Blinding strategy
To maintain study blinding, this study used a Brain Ultimate sham butterfly coil (P/N 3950-00 S/NO105). The system provides the sensation of TBS without enough magnetic energy to reach the cortex. The system has been used in numerous TBS trials and includes specific coil windings to optimize the blinding experience. It is also visually indistinguishable from active coils to minimize accidental unblinding. To ensure consistency and blindness, all subjects and psychological assessors are blind. Only one professionally licenced TBS therapist knew about the grouping and provided the intervention throughout the study, and only one transcranial magnetic stimulator and navigator were used.

Facial emotion recognition test (FERT)
The FERT is an assessment of a person's ability to recognize and discern the facial emotions of others, and is an important research area in social cognition (Yalcin-Siedentopf et al., 2014). Huang et al. created a Chinese localized facial expression picture test system based on the Facial Action Coding System (FACS) by Ekman (Polli et al., 2012). The Cronbach's α coefficient is 0.77, and the remeasuring reliability is 0.81 (Y et al., 2020). The test consists of 42 facial emotion photographs, which show seven basic emotions (anger, disgust, fear, sadness, surprise, happiness, and neutral), and subjects are asked to indicate the emotion and intensity in each photograph. The outcome of the FERT is the accuracy of emotion category recognition, with 1 point for correct answer and 0 points for incorrect answers. Higher scores indicate better emotion judgment.

Hinting task (HT)
The HT is a linguistic measure of theory of mind and an important indicator of social cognition (Corcoran et al., 1995). It evaluates a subject's ability to infer the real intentions behind indirect speech utterances or hints. The task consists of 10 essays presenting an interaction between two characters, and the participant is asked to say the intention of the character dropping the hint. If the participant's first answer is correct, they receive a score of 2 points. If the participant gives the correct response when prompted with paraphrasing of the hint, a score of one is given. Failure to infer true intent receives a score of zero. The Cronbach's α coefficient is 0.62 and the remeasuring reliability is 0.64 (Lindgren et al., 2018). Total scores range from 0 to 20, with higher scores indicating better theory of mind performance (Mallawaarachchi et al., 2019).

Positive and negative syndrome scale (PANSS)
The PANSS scale was proposed by Kay et al. (1987). It was conceived as an operationalized, standardized measure of schizophrenia that provides a balanced representation of positive and negative symptoms and gauges their relationship to one another and to psychopathology (Lim et al., 2021a). The 30-item PANSS constitutes four scales measuring positive and negative symptoms, general psychopathology, and their differences. The Chinese version of the PANSS provides better reliability and validity; and Cronbach's α coefficient is 0.80, and the remeasuring reliability is 0.89 (Si et al., 2004). The scale is scored on a 7-point Likert scale, and the higher the score, the more severe the psychopathology in the dimension.

Procedure
All the enrolled participants underwent social cognition and psychopathology assessments at baseline, 2 weeks, and 4 weeks. Under the guidance of the experimenter, the participants completed the FERT and HT according to the instructions on the computer, and completed the PANSS assessment in a quiet room, and checked the completion of each item. The evaluator has a master's degree in psychiatry and is a registered clinical psychological scale surveyor in China. To ensure consistency and blinding, only one consistently-trained evaluator was involved, and the assignment of active and sham TBS was unknown. After the last follow-up, the subjects and the psychological evaluator filled in the blinding questionnaire, including three choices: active iTBS, sham iTBS or "Don`t know".

Statistical analysis
Data analysis was conducted using SAS software (version 9.4 SAS Institute, Inc). We performed normality tests on all data by Kolmogorov-Smirnov normality analysis, using the mean (SD) to statistically describe normal continuous data. The t test was applied for evaluation between two groups. A nonparametric Mann-Whitney U test was used for data that did not meet the median (IQR). For classified data, the frequency (percentage) was used for statistical descriptions, and the χ 2 test was used for intergroup comparisons. The Fisher exact probability method was used for data with al theoretical frequency that was too small. The analysis was performed in the intention-to-treat population, defined as all patients randomly assigned into a treatment group who started at least one treatment session. A linear mixed model regression test was used to evaluate the effect size of the FERT, HT, and PANSS. Effect sizes are described as Cohen's d. Differences were statistically significant at P < .05.

Ethics statement
The protocol for this research was approved by the Research Ethics Committee of the Shanghai Pudong New Area Mental Health Center and Tongji University School of Medicine (No: PDJWLL2020027). All procedures were performed in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and the Helsinki Declaration of 1975, as revised in 2008. Both participants and guardians provided written informed consent for participation in this study.

Recruitment
Of the 125 participants who met the diagnostic criteria, 59 were ineligible for participation based on the inclusion and exclusion criteria (Fig. 1). Nine participants were excluded because of an age over 65 years, 7 were excluded because of a disease duration less than 5 years, 13 were excluded because of the use of other antipsychotic medications, and 8 were excluded because of medication adjustments within 6 months. Three participants who received rTMS within one month and 1 who received MECT were excluded. Of the 66 enrolled participants, 2 were uncomfortable with stimulation, 1 withdrew due to their guardian, 1 was excluded due to the use of other antipsychotics, 1 was unable to complete the psychobehavioural assessment, and 1 had other serious diseases during the follow-up period. Ultimately, 30 participants were included in each of the active and sham iTBS groups and completed the entire research process.

Participant characteristics
The demographic characteristics of the participants from the active iTBS, sham iTBS, and total sample groups in terms of age, sex, education level, marital status, employment status, course of disease, family history of mental illness, and olanzapine use are shown in Table 1. Among the 66 participants, the average age was 47.53±10.17 years, and the proportion of females was 41.67%. The average education level of the participants was 7.27±2.33 years, and most of them were married (60.00%). Fifty-six percent of the participants were unemployed, and the average course of disease was 8.62±4.09 years. It is noteworthy that 36.67% of the patients had a family history of mental illness, and the average maintenance dose of olanzapine was 11.98±2.69 mg/d. There were no significant differences in the demographics between the active and sham iTBS groups (p> .05). At the end of follow-up, only 32% of subjects and 41% of evaluators could guess the intervention assignment correctly. Jame's blinding index (BI) is 0.683 (95% CI: 0.593 to 0.772), which was greater than 0.5 to determine the overall success of the implementation of blindness. Table 2 and Fig. 2 show the comparison of the main outcomes of social cognition between groups at the baseline, 2-week, and 4-week follow-ups. There was no significant difference in FERT and HT between the active-iTBS group and the sham-iTBS group at baseline (P> .05). The main outcomes of social cognition showed that the FERT scores (week 2; 0.27 [95% CI, 0.09 to 0.45]; P< .01; ES 0.14; Fig. 2A of the two groups were significantly different at the 2-week and 4-week follow-ups. On the FERT subscale, there was a significant difference between the two groups at the 2-week follow-up only for the identification of fear (P< .05; Fig. 2B) and surprise (P< .01). At the 4-week follow-up, the active iTBS group had higher scores than the sham iTBS group for the identification of anger, fear, sadness, surprise and happiness, and the difference was statistically significant (P< .001).  Table 2 and Fig. 2 show the comparison of psychopathology symptoms between groups at the baseline, 2-week, and 4-week follow-ups. There was no significant difference in PANSS, between the active-iTBS group and the sham-iTBS group at baseline (P> .05). The PANSS score (-2.47 [95% CI, -5.83 to -1.10]; P< .01; ES 0.13; Fig. 2D) and negative symptom score (-3.43 [95% CI, -4.85 to -2.01]; P< .001; ES 0.29; Fig. 2F) of the active iTBS group were significantly different from those in the control group at the 4th week of follow-up. However, there was no significant difference between the two groups in the positive symptoms and general psychopathology subscale scores at the 2-week and 4-week follow-ups (P> .05). Table 3 shows the differences of social cognition and psychopathology symptoms within the groups at the 2-week and 4-week followups compared with baseline. At the 4-week follow-up, the active iTBS group had a mean increase of points in the FERT scores (-0.76  were significantly different from baseline after 4-week follow-up in the active iTBS group, and only disgust factor in FERT subgroup showed no statistical difference (P> .05). In the sham iTBS group, only FERT scores were significantly different from baseline at the 2-week and 4-week  follow-ups (P< .05).

Adverse events
During the study, 15 (50%) participants in the active iTBS group and 3 (10%) participants in the sham iTBS group reported one or more adverse events. Numb-headedness had the highest incidence of the adverse events, reported by 10 (33%) participants in the active iTBS group and 3 (37.5%) participants in the sham iTBS group. However, no serious adverse events, such as seizure or syncope, occured. The incidence of adverse reaction records is shown in Fig. 2H.

Discussion
We used the FERT, HT, and PANSS in a randomized, double-blind, and sham-controlled trial to explore the effectiveness of accelerated iTBS on social cognition and negative symptoms in individuals with schizophrenia. To our knowledge, this is the first randomized controlled study to simultaneously focus on the use of accelerated iTBS for social cognition and negative symptoms in individuals with schizophrenia, which is an original research direction. We demonstrated that accelerated iTBS administration in the left DLPFC significantly improved negative symptoms in individuals with schizophrenia, but there were no significant changes in positive symptoms or general psychopathology.

Fig. 2.
Differences in social cognition, psychiatric symptoms, and adverse events between active and sham iTBS groups. A, Advancement in FERT score, B, Radar chart of FERT basic emotion difference between active and sham iTBS groups in different periods; C, Advancement in HT score; D, Improvement in PANSS score; E No significant difference in positive symptom score; F, Improvement in negative symptom score; G, No significant difference in general psychopathology score; H, Dot plot of the percentage of adverse events. Abbreviations: FERT, Facial emotion recognition test; HT, Hinting task; PANSS, Positive and negative syndrome scale. * p< .05, ** p< .01, *** p< .001.
Our results are consistent with the study by Zhao et al., which evaluated the therapeutic effects of four different intensities of rTMS and iTBS in individuals with schizophrenia, effective only for negative symptoms and not for positive symptoms (Zhao et al., 2014). Some recent studies supported that left DLPFC significantly decreased negative symptoms severity among individuals with schizophrenia compared with sham iTBS (Bation et al., 2021;Wang et al., 2020). Neurochemical and neuroimaging studies have reported that negative symptoms are associated with decreased frontal or prefrontal metabolism and dopamine function (Galderisi et al., 2015). ITBS ameliorates negative symptoms by stimulating the left DLPFC, increasing metabolism in this area, promoting dopamine release, and increasing regional blood flow to the resting brain (Bulteau et al., 2022). This hypothesis supports our finding that the longitudinal difference in negative symptoms at the 4-week follow-up is more significant than that at the 2-week follow-up, which means that the duration of brain stimulation may be proportional to the amount of dopamine released.
Furthermore, our results show that the improvement in emotional processing and theory of mind in the positive iTBS group was more significant than that in the control group, which indicates that iTBS can enhance the social cognitive function of individuals with schizophrenia. To date, iTBS research on facial emotion recognition and theory of mind in individuals with schizophrenia is relatively limited. Wolwer et al. found that high-frequency (10 Hz) rTMS on the left DLPFC improved impaired facial emotion recognition in individuals with schizophrenia (Wölwer et al., 2014). Dumitru et al. administered iTBS on the left DLPFC to 28 healthy subjects and found a positive effect on facial emotion recognition (Dumitru et al., 2020). Tong et al. (2021) suggested that high-frequency rTMS could significantly improve the performance on a theory of mind and executive function, improving self-other discrimination and the ability to regulate self-other representation. These results are generally consistent with our findings and confirm the effectiveness of noninvasive neuromodulation techniques for social cognition.
The mechanism of the positive effects of iTBS on social cognition remains unclear. The neuropsychological hypothesis suggests that the frontal lobe may play an important role in promoting social cognition (Lee et al., 2004), and the limited activation of the prefrontal cortex may be at the core of facial emotion recognition disorders (Li et al., 2010), leading to reduced theory of mind-related responses in the medial superior prefrontal temporal neural network (Park et al., 2011). Lantrip et al. (2017 used rTMS to intervene in the left DLPFC of patients with depression, which could accelerate the emotional cognitive process. Rassovsky et al. (2015) found that patients with schizophrenia undergoing tDCS on the DLPFC showed significant improvement in facial emotion recognition. Nitsche et al. (2012) conducted tDCS on the left DLPFC of healthy subjects, revealing that the perception of positive emotions was improved. Therefore, we hypothesized that the reason for the improvement in emotional processing and theory of mind might be related to the iTBS on the left DLPFC.
However, the neurobiological hypothesis suggests that iTBS can influence the magnitude of long-term potentiation (LTP), which has much in common with long-term memory and is one of the main molecular mechanisms of learning and memory (Feng et al., 2023). Opie et al. (Sergi et al., 2007) found that iTBS can enhance the neural regeneration response, which is also the basis of learning and memory. Interestingly, our study found that iTBS seems to have very limited improvement in disgust emotion recognition. There are few studies on emotion recognition and intensity, which cannot truely explain the neurobiological basis and can only confirm the positive role of emotion recognition and intensity in interpersonal communication (Wang et al., 2013). Moreover, no serious adverse events such as syncope or seizure occurred during the entire treatment period, and only complaints of headache, needle-tingling, numb-headedness, and dizziness were reported. These results are consistent with most studies, which confirm that iTBS is a safe and well-tolerated neuromodulation technique with good clinical application prospects (Demirtas-Tatlidede et al., 2010).

Limitations
We also note several limitations. First, this study was been greatly limited with respect to expanding the number of participants and the follow-up period due to the COVID-19 pandemic, and the follow-up was limited to 4 weeks. Second, considering the possible inconsistent attributes of social perception and attribution bias, our study only used emotional processing and theory of mind as social cognitive assessment factors. Third, we did not extend serum pro-brain derived neurotrophic factor concentration (BDNF) or brain imaging tests such as functional magnetic resonance imaging (fMRI) to explore the neuropathological mechanisms by which accelerated iTBS improves social cognition and negative symptoms. Future studies should investigate the long-term effects of accelerated iTBS intervention duration and stimulation intensity Table 3 Comparison of social cognitive and psychiatric symptoms effects in active and sham iTBS groups from baseline.

Variable
Difference from week 2 and baseline (95% CI) on social cognition and negative symptoms in individuals with chronic schizophrenia. These results provide an evidence-based foundation for the clinical application of accelerated iTBS in physical therapy for patients with schizophrenia.

Conclusions
This randomized double-blind clinical trial found that accelerated iTBS can effectively ameliorate the social cognition and negative symptoms of individuals with schizophrenia. These findings suggest that accelerated iTBS may improve the overall functional recovery of individuals with schizophrenia, and that it is a safe and effective neuromodulation technique with good clinical application prospects.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Declaration of Competing Interest
The authors declare no conflict of interest.