Connectivity-guided theta-burst targeting gave a faster early response than the 5-cm rule

Design

Single-centre, randomised, double-blind, three-arm trial

Population

123 adults with major or bipolar II depression and at least one antidepressant failure

Primary outcome

Percentage reduction in HAM-D-17 at week 2

Effect

Structural-connectivity vs 5-cm: +8.79 points (95% CI 2.19-15.40); no difference at week 12

In a single-centre, double-blind trial in China (American Journal of Psychiatry, 12 August 2026), 123 adults with major depression or bipolar II depression who had failed at least one antidepressant received 20 sessions of accelerated intermittent theta-burst stimulation (iTBS) over two weeks with robotic neuronavigation. They were randomised to one of three left prefrontal targets: the conventional 5-cm rule, a site chosen by functional connectivity with the subgenual anterior cingulate, or a site chosen by structural connectivity (tractography) to that region.

At two weeks, structural-connectivity targeting produced a greater percentage reduction in Hamilton depression score than the 5-cm rule (difference 8.79 percentage points, 95% CI 2.19-15.40). At six weeks both connectivity-guided arms outperformed the 5-cm rule (structural 12.87 points, 6.21-19.54; functional 9.41 points, 2.84-15.98). By 12 weeks the differences were no longer significant. Adverse events were similar, with no seizures or mania.

This supports the idea that targeting matters, but it is one centre with about 40 people per arm and the advantage faded. It is not yet a reason to require MRI-guided targeting before offering TMS.

• Conventional 5-cm targeting remains acceptable practice; this trial does not make it obsolete.
• Connectivity-guided targeting may speed early response but needs MRI and neuronavigation few services have.
• Expect the early advantage to narrow; differences were not significant at 12 weeks.
• No seizures or mania were reported in any arm in this sample.

Why it matters

It challenges the assumption that where the coil sits on the scalp makes little difference to outcome.

Don't overread it

A single-centre trial of 119 analysed patients with an effect that faded by 12 weeks does not establish superior long-term outcomes.

The statistics, in plain English

The confidence intervals are wide: at two weeks the true advantage could be as small as 2 points or as large as 15. Cohen's d of 0.70 sounds large, but small trials tend to overestimate effects.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.