In bipolar depression, how do theta burst and standard repetitive transcranial magnetic stimulation compare on effectiveness and manic switch risk?

How do theta burst and standard rTMS compare on antidepressant effect?

The most direct evidence comes from a 2025 systematic review and meta-analysis that pooled 56 articles covering 1,709 patients with bipolar depression [1]. When the researchers compared active TMS against sham (inactive) stimulation in randomized controlled trials, active treatment produced a moderate antidepressant effect (Cohen's d = 0.40) [1]. When they pooled active treatment arms from randomized trials together with uncontrolled studies, the effect size was large (Cohen's d = 1.4), with response and remission rates of about 47% and 28% respectively—similar to what is seen in unipolar major depression [1].

Crucially for the comparison question, the meta-analysis examined whether different protocols performed differently. It found that high-frequency protocols, including intermittent theta burst stimulation (iTBS) delivered to the left dorsolateral prefrontal cortex, and low-frequency protocols delivered to the right dorsolateral prefrontal cortex were both robustly associated with positive outcomes [1]. The abstract states that left DLPFC iTBS showed noninferiority to more widely used high-frequency rTMS protocols [1]. In plain terms, theta burst—which delivers pulses in rapid bursts and takes far less time per session—did not appear worse than standard rTMS at relieving bipolar depression.

One randomized trial complicates the picture. A double-blind trial of 37 patients with acute bipolar depression found no significant difference between active iTBS and sham iTBS on depression scores; the trial was terminated for futility, and response rates were low in both groups (about 16–17%) [2]. That result suggests iTBS may not work for every patient, particularly those already taking mood stabilizers or antipsychotics [2]. The trial authors themselves noted that iTBS has not been adequately studied in bipolar disorder and called for more research on how TMS protocols differ between unipolar and bipolar depression [2]. So the meta-analysis supports noninferiority across pooled studies, while at least one rigorous trial found iTBS no better than sham in its specific population—a discrepancy that likely reflects differences in patient selection, sample size, and treatment parameters rather than a clear verdict against theta burst.

Does either protocol raise the risk of a manic switch?

The same 2025 meta-analysis directly addressed manic switch risk. Rates of treatment-emergent mania or hypomania were low and statistically equivalent to those seen with sham stimulation, with an odds ratio of 1.3 and a 95% confidence interval spanning 0.7 to 2.4 [1]. Because that interval includes 1.0, the data do not show a meaningful increase in manic switch risk from active TMS compared with placebo—and this finding applied across the pooled protocols, including iTBS [1].

The randomized iTBS trial provides a closer look at theta burst specifically. Among 18 patients receiving active iTBS, one developed treatment-emergent hypomania during the double-blind phase, and a second episode occurred during open-label treatment [2]. With numbers that small, the trial cannot establish a precise risk rate, but it is consistent with the meta-analysis finding that manic switches are uncommon [1][2].

A small pilot study of an accelerated, high-dose theta-burst protocol called Stanford Neuromodulation Therapy (SNT) reported no treatment-emergent hypomania or mania in seven participants with bipolar depression, even though the protocol delivered 90,000 pulses over five days—far more than the roughly 18,000 pulses in a conventional six-week iTBS course [4]. The authors cautioned that most participants were on mood stabilizers and that only two had bipolar I disorder, so the risk in that group remains particularly uncertain [4]. This is an early safety signal from an open-label study, not proof that high-dose theta burst is safe for everyone.

What else shapes the comparison between these protocols?

The meta-analysis found that higher baseline illness severity and more treatment sessions predicted greater antidepressant effect [1]. That means the practical difference between theta burst and standard rTMS may depend less on the pulse pattern itself and more on how sick the patient is and how many sessions they receive. The finding that iTBS was noninferior to standard high-frequency rTMS is notable because iTBS sessions are much shorter—a practical advantage for patients and clinics—though the meta-analysis did not report a head-to-head randomized comparison isolating delivery time as a variable [1].

For context on manic switch risk in bipolar depression generally, a large naturalistic cohort study of antidepressant medications found no significant increase in manic switch after adjusting for confounding (adjusted hazard ratio 0.77, 95% CI 0.51–1.08) [3]. That study examined drugs, not TMS, so it cannot be used to estimate TMS risk directly. But it does suggest that the concern about manic switches in bipolar depression treatment is not unique to brain stimulation, and that careful adjustment for patient characteristics can change how risk looks. The TMS meta-analysis similarly found manic switch rates equivalent to sham, reinforcing that the risk with TMS is low in the populations studied [1].

Sources used in this answer

Related research questions