Neuroimaging biomarkers of the brain-heart axis in Takotsubo syndrome: regional convergence without directional consistency: a systematic review and biomarker qualification appraisal

BACKGROUND: Takotsubo syndrome (TTS) is an acute stress-induced cardiomyopathy increasingly recognised as a disorder of the brain-heart axis. Neuroimaging studies have identified structural, functional and metabolic abnormalities within the central autonomic network, particularly the amygdala. Still, the role of amygdalar activity as a biomarker in TTS has not been systematically synthesised. METHODS: A systematic review was conducted according to PRISMA guidelines and registered in PROSPERO (CRD420251177487). PubMed, Web of Science, Cochrane Library, Google Scholar and ClinicalTrials.gov were searched through February 2025 for studies reporting neuroimaging findings in confirmed TTS. Data were extracted, narratively synthesised, and appraised against biomarker-qualification criteria. Directional consistency was additionally quantified by vote counting with exact binomial (sign) tests at the level of study-region-imaging domain, with Holm correction for multiplicity and sensitivity analyses excluding studies at high risk of bias. RESULTS: The search identified 489 records, of which 29 underwent full-text assessment; 10 studies met the inclusion criteria. Multimodal neuroimaging consistently implicated the amygdala, insula, hippocampus, thalamus, anterior cingulate and prefrontal cortex, with reduced amygdalar and insular grey matter, altered limbic structural connectivity, and disrupted central autonomic and default mode network connectivity. Two ¹⁸F-FDG PET/CT cohorts, the only included studies in which imaging preceded the outcome, found raised amygdalar metabolic activity before TTS onset that independently predicted subsequent syndrome development; the remaining eight were cross-sectional and cannot separate predisposition from consequence. Convergent, though directionally inconsistent, evidence supports a working model of chronic limbic sensitisation and exaggerated sympathoadrenal activation contributing to myocardial stunning. Vote counting confirmed that an abnormality was reported by almost every study assessing the amygdala (8/9), insula (6/6), prefrontal cortex (6/6) and hippocampus (5/5), but no region survived correction for multiplicity, and only the prefrontal cortex approached directional consistency (5/5 reduced; p = 0.062); amygdalar findings were directionally split (4 increased versus 3 decreased; p = 1.00). Excluding the two studies at high risk of bias altered no conclusion. CONCLUSIONS: Neuroimaging findings support TTS as a neurocardiac disorder characterised by dysfunction of limbic and autonomic brain networks. Amygdalar hyperactivity is the most frequently reported candidate marker of vulnerability, but none of these findings is clinically actionable, and the reasons are specific rather than generic: effect sizes are modest (Cohen's d = 0.37-0.53), the direction of abnormality is inconsistent for every region except the prefrontal cortex, no included study reported test-retest reliability for its own primary measure, no study reported discrimination or calibration, and none tested incremental value over inexpensive psychometric or autonomic measures. These findings are therefore hypothesis-generating; adequately powered prospective cohorts and formal biomarker qualification are required before neuroimaging-guided risk stratification of the brain-heart axis can be considered for clinical use.