Effects of High-frequency Left Dorsolateral Prefrontal rTMS on Heart-brain Coupling in Women With Recurrent Pregnancy Loss and Elevated BMI

Recurrent pregnancy loss is increasingly recognized not only as a reproductive disorder but also as a condition associated with broader systemic vulnerability. In women with elevated BMI, recurrent pregnancy loss may reflect a composite phenotype characterized by reproductive burden, metabolic stress, emotional distress, and altered autonomic regulation. These converging factors may contribute to exaggerated cardiac responses to stress and early cardiovascular vulnerability. The present study is based on the hypothesis that the left dorsolateral prefrontal cortex, DLPFC, is an important cortical target for top-down modulation of autonomic and heart-brain processes in this population. This study is designed as a randomized, sham-controlled, single-session mechanistic clinical trial. Sixty women with recurrent pregnancy loss and BMI ≥24 kg/m² will be enrolled and randomly assigned in a 1:1 ratio to real-rTMS or sham-rTMS. Both groups will be stimulated over the same left DLPFC target identified using the adjusted BeamF3 method. In the sham condition, the coil will be positioned over the same target but angled at approximately 45° relative to the scalp to markedly reduce effective cortical stimulation while preserving, as far as possible, the acoustic cue and part of the scalp sensation. Participants and outcome assessors will remain blinded to group assignment; only the operator delivering stimulation will know the assignment on the study day. The stimulation protocol will use high-frequency rTMS at 10 Hz and 100% motor threshold. Stimulation will be delivered in 30 cycles, each consisting of a 5 s train and an 11 s inter-train interval, for a total of 1,500 pulses over approximately 8 min. During the rTMS-ECG protocol, continuous 3-lead electrocardiography will be recorded at a sampling rate of 1,000 Hz. After preprocessing, heart-brain coupling, HBC, will be quantified from cycle-locked cardiac oscillatory power at the target frequency. HBC is defined in this study as the time-averaged normalized power within the 0.062-0.063 Hz band during each 16 s stimulation cycle. The primary endpoint will be the between-group difference in mean HBC across 30 stimulation cycles, calculated with the Morlet wavelet using the default wavenumber of 6. Secondary endpoints will include mean HBC across 30 cycles under higher-frequency-resolution and higher-temporal-resolution Morlet conditions. Exploratory analyses will include repeated HBC extraction using DoG and Paul wavelets, as well as a group × stimulation progression analysis based on three sequential stages: cycles 1-10, 11-20, and 21-30. Baseline assessments will include demographic characteristics, reproductive and medical history, comorbidities, medication use, psychiatric ratings, resting heart rate, blood pressure, and 12-lead ECG. Safety monitoring will cover common rTMS-related adverse events, including pain at the stimulation site, facial muscle contraction, mild headache, and dizziness, as well as serious adverse events such as syncope and seizure. After completion of the protocol, participants will remain under observation for 1 h, with reassessment of blood pressure and resting heart rate when clinically indicated. This trial is intended to establish a mechanistic framework for left DLPFC rTMS modulation of heart-brain coupling in women with recurrent pregnancy loss and elevated BMI and to inform the design of future multi-session randomized controlled studies.