Efficacy of adjunctive theta burst transcranial magnetic stimulation in acute mania: A randomized and placebo-controlled study

Subjects

To participate in the study, the right-handed patients between the ages of 18 and 59 were recruited. Individuals who met the diagnostic requirements for mania according to the International Classification of Diseases-tenth edition (ICD-10) (World Health Organization, 1992) were contacted.^[20] The effect size calculated of prior randomized controlled trial (RCTs) by Kaptsan et al. (Cohen’s d = 0.29) and Praharaj et al. ([Cohen’s d = 0.74) is 0.51 (avg. Cohen’s d).^[16,17] With the expectation of a similar modest effect size, number of groups, and repetition of evaluation being 2 and 3, using G-power and a critical F of 3.175, the total sample size was calculated to be 40 participants with an alpha error of 0.05, an anticipated 25% dropout rate and 1-beta error of 0.95. We enrolled 20 subjects in the “active” group and another 20 in the “sham” group [Figure 1]. Subjects with at least moderate-to-severe illness: Baseline YMRS Score >26 were included in the study.^[21]

Close

Figure 1:

Flowchart displaying the participant requirement process ‘ITT: Intension to treat analysis’

Export to PPT

Exclusion criteria included any comorbid psychosis or substance use disorders excluding caffeine and nicotine (according to ICD-10 DCR), a history that suggests a brain injury or surgery, any substantial brain malformation or tumor, neurodegenerative disease or history of seizures, subjects who have received ECT in the preceding 6 months, and subjects with metallic parts in the body, such as pacemakers.

The research has been listed in the Clinical Trials Registry India (CTRI), Number: CTRI/2022/04/041957. The research was conducted at the psychiatry department, Sri Guru Ram Rai Institute of Medical and Health Sciences, Dehradun. The ethical committee of the institution approved the study protocol. (SGRR/IEC/2/19; IEC Registration No. ECR/710/ Inst/UK/2015/RR-18).

Seventy-two patients were screened for the study and among them, 40 patients in all were signed up. The patients were all admitted to the inpatients in the department of psychiatry and underwent regular general physical and systemic examination, basic metabolic profile, and electrocardiography. Block randomization was used to assign patients at random to the active and sham groups. The patients were hospitalized for 2 weeks during which iTBS and cTBS sessions were given. The follow-up evaluation was conducted post-discharge, on outpatient visits. A detailed written informed consent was signed by the patient and their caregivers before enrolment, after explaining the full procedure in detail along with the expected benefits and risks involved. The patients continued the psychotropics for the complete study duration. No psychotherapy was provided for any of the patients enlisted in the study.

Two patients in the active group and where as one in the sham group discontinued the study (1 – worsening of symptoms and 2 – uncooperative). Nobody among the patients discontinued treatment as a result of rTMS side effects.

rTMS stimulation parameters

TBS was done using a “MagVenture - MagPro- R30.” Theta Burst’s booster and the figure-of-8-shaped B65 coil are both found in the MagPro- R30 device. The dorsol lateral prefrontal cortex was the favored stimulation location. The worldwide 10–20 Electroencephalography (EEG) system was utilized for the positioning of the TMS coil [Figure 2]. The system was chosen since it often appears in other clinical TMS applications and takes participant skull size diversity into account.^[22]

Close

Figure 2:

Site of theta burst stimulation stimulation as per international 10–20 system

Export to PPT

There were 20 TBS sessions in total for both groups (two sessions/day, first left TBS and after 30 min right TBS, a rigorous protocol for 5 days a week, intensive protocol in accordance with prior studies in other disorders [Figure 3].^[23]

Close

Figure 3:

Days of stimulation.

Export to PPT

An active cTBS protocol burst of three pulses was used, delivered at 50 Hz, which was repeated every 200 MS (at a frequency of 5 Hz) lasting 40 s and containing a total of 600 pulses.

The iTBS, which delivered a total of 600 pulses in 20 trains of 10 bursts at intervals of 8 s and 3 pulses at 50 Hz, repeated at a rate of 5 Hz. A total of 1200 pulses were delivered/day. Instead of the standard 600 pulses/day, as suggested by Huang et al. and colleagues, we provided 1200 pulses/day.^[24]

We delivered a stronger version of TBS of rTMS with a large size and better study design stimulating both right and left DLPFC.

We used TBS at 80% of the resting motor threshold (RMT) since the plasticity of a stimulus is affected by its intensity. Before each session of stimulation, this was established. with the lowest intensity according to the Rossini-Rothwell method. In each of the ten trials, RMT generated five MEP responses that were at least 50 uV.^[25,26] Sham rTM was delivered using the opposite side of the coil of 8, which resembles active stimulation in terms of sound and scalp contact. The previous research has utilized a sham stimulation technique similar to this one.^[27,28]

Clinical measures

The intensity of manic symptoms was evaluated by YMRS. The YMRS gives each item a severity rating. Irritability, thought content, speech, and disruptive/aggressive behavior are the four categories that are assessed on a 0–8 scale, and the remaining seven items are scored on a 0–4 scale.^[29]

Clinical global impression (CGI)-Bipolar illness: For evaluating the severity of manic episodes the degree of change from the illness’s worst phase and the phase that came right before it.^[30] A tailored checklist based on rTMS-related unfavorable events from prior studies was used to assess rTMS-related adverse events after each session.

Blinding procedure

An impartial evaluator who was unaware of the randomization of the stimulation groups into active or sham conditions evaluated the subjects at baseline, 7^th day, and on day 21. Blocks with random-number patterns were used to divide the patients into active and sham groups. There were unopened envelopes with these numbers. The clinician who gave the TBS immediately before the first session for each patient opened the envelope. In addition, the patients were unaware of the treatment arm. The rater was required to retrospectively infer the treatment arm for each trial participant based on the pre- and post-psychopathological ratings. A Cohen’s Kappa coefficient of −0.016 for the guess matrix^[31] indicated “no agreement,” which is a sign of the blind’s good integrity. Independent verification of blinding of participants to treatment arms was not possible.^[27,28]

Statistical analysis

SPSS was used to analyze the study’s data (Version 28). YMRS score served as the main outcome variable. Scores on the CGIBP-S, CGIBP-P, and CGIBP-W were secondary outcome factors. The Kolmogorov–Smirnov test and normal probability plots were used to confirm the assumption of normality. Independent t-tests and Chi-square tests were used to look at group differences in sample characteristics (wherever applicable). Using the Mann–Whitney U test, variables such as the number of episodes and motor threshold that violated the assumption of normality were compared. Regardless of whether a patient completed the trial or not, all of the patients who were enlisted were added to the analysis using the intention-to-treat (ITT) method. The latest observation carried forward approach was used to replace the missing data. In the double-blind phase, the primary analysis focused on the interaction effect of treatment over time (baseline, 7^th, and 21^st days) between the active and sham group. Using the restricted maximum likelihood mixed (growth curve) model analysis, the overall treatment effect over time for the two groups was examined, with time (0, 7^th, and 21^st) as the within-subject component, treatment (active/sham) serving as the between-subject factor, with allocation order serving as the “subjects.”

Sample characteristics

Comparable characteristics between the two groups included gender, marital status, religion, habitat, employment, socioeconomic status, and education. Medical morbidity (χ² (1, 40) = 2.125, P = 0.764) and number of episodes (U = 226.500, P = 0.478) present in comparable proportions in both the “active” and “sham” group individuals. The subjects randomly assigned to the “active” and “sham” groups were comparable in age (t (40) = 0.041, P = 0.778), Motor Threshold (t (40) = 0.229, P = 0.400), and antipsychotic dose per trial (As determined by Woods, the chlorpromazine equivalent dose) [Table 1].^[32]

Safety and side effects

TBS was not associated with any severe adverse effects. Nine patients (six from the “active” group and three from the “sham”) complained of headaches in the early sessions, which were relieved by analgesics. There are no reports of any people quitting because of side effects.

Outcome measures

At the baseline psychopathology scores such as YMRS (t = 0.081; P = 3.280), CGIBP-S (t = 1.295; P = 0.136), CBIBP-P (t = 0.309; P = 0.192), and CGIBP-W (t = 0.312; P = 0.354) were comparable for both the “active” and “sham” groups.

[Table 2] compares the pre-post effects from the pre-treatment period, 7^th and 21^st across the two groups, taking into account the effects of group and time on the ITT analysis.

Repeated measures-ANOVA did not find a significant group*time effect for each of the variables, that is, YMRS (F = 0.300; P ≤ 0.612), CGIBP-S (F = 0.432; P = 0.562), CGIBP-P (F = 0.202; P = 0.701), and CGIBP-W (F = 200; P =0.705).

According to YMRS scores, 85% of the active group and 80% of the sham group responded. The active group had a 1.417 odds ratio for response when compared to the sham group.

A drop in the overall YMRS scores of at least 50% from baseline till day 21 was taken into account as a response.^[33]