Depression Improvement Correlates With Lower TMS Intensity in a Randomized Trial With Real-Time E-Field Modeling

Current transcranial magnetic stimulation (TMS) practice uses fixed-percentage motor threshold dosing, conventionally 120% rMT, for depression treatment. Individual anatomical variability may result in substantially different cortical electric-Field (E-Field) strengths across patients. We used prospective real-time E-Field modeling to characterize individual TMS intensity requirements and examine associations with clinical outcomes. Twenty-eight subjects with major depressive disorder received single-day accelerated intermittent theta-burst stimulation (10 sessions, 1800 pulses/session) with real-time E-Field-guided dosing targeting M1-equivalent stimulation at left dorsolateral prefrontal cortex (DLPFC). Required %rMT for motor-equivalent DLPFC E-Field ranged from 49.7%-150.4% rMT (mean = 99.7% ± 18.9%), with 53.6% of subjects requiring less than 100% rMT. Real-time E-Field-guided dosing achieved 48.1% better precision than conventional 120% rMT in approximating motor-equivalent E-Field delivery (t(27) = 2.45, p = 0.021). Three dosing frameworks were tested against change in QIDS-SR16 scores: delivered %rMT was not significantly associated with outcomes (ρ = -0.15, p = 0.436), while both the ratio of DLPFC to M1 E-Field strength (ρ = -0.49, p = 0.008) and absolute DLPFC E-Field strength (ρ = -0.49, p = 0.008) were significantly negatively correlated with greater symptom reduction. These findings demonstrate substantial interindividual variability in cortical E-Field delivery under fixed-percentage dosing, and that real-time E-Field guidance more precisely approximates motor-equivalent stimulation than conventional 120% rMT. The association of lower absolute DLPFC E-Field strength with greater symptom reduction challenges the assumption that higher stimulation intensity produces better clinical outcomes, and warrants systematic investigation in larger trials.