Introduction Bipolar disorder (BD) is highly heritable, yet the brain mechanisms that distinguish risk from illness remain poorly understood. Studying unaffected relatives makes it possible to examine vulnerability without the confounding effects of medication, mood episodes and chronicity. This study tested whether local resting-state brain activity follows a familial risk gradient, from healthy controls to relatives to patients, or whether relatives show a distinct profile. Materials and methods A total of 104 participants were included: 34 patients with BD, 34 unaffected first-degree relatives and 36 healthy controls, belonging to 77 families. The amplitude of low-frequency fluctuations (ALFF), fractional ALFF (fALFF) and regional homogeneity (ReHo) were computed from resting-state MRI (3 T, multiband sequence). Voxel-wise analyses used a full factorial model with pairwise comparisons and a quadratic contrast, with threshold-free cluster enhancement (TFCE) and family-wise error (FWE) correction (p < 0.05). The ordered hypothesis was tested with Bayes factors (BF) in 28 a priori regions of interest, using family-clustered robust standard errors. Results At an uncorrected threshold of p < 0.001, patients showed, relative to controls, reduced local activity (fALFF and ReHo) in the angular gyrus and superior lateral occipital cortex bilaterally and in the right middle temporal gyrus, and increased ReHo in the cerebellar vermis, right cerebellum and inferior occipital cortex. In several of these clusters, relatives occupied an intermediate position, differing from both patients and controls. Among the 28 a priori regions, the right inferior frontal gyrus (ReHo, BF = 15.2) and the right precuneus (ALFF, BF = 15.5) showed strong evidence for the order HC > Fam > BD. In most regions where the groups differed, relatives also showed a distinct profile, with values higher or lower than those of both other groups; three of these clusters survived FWE correction. After voxel-wise FWE correction, increased ReHo in the vermis was confirmed as a difference between patients and controls, whereas no voxel showed a complete gradient across the three groups. Conclusions In patients with BD, local resting-state activity is altered in the parietal component of the default mode network, in the middle temporal gyrus and in the cerebellum, with a vermal alteration that survives correction for multiple comparisons. The right inferior frontal gyrus and the right precuneus show a gradient consistent with familial risk, whereas unaffected relatives display a distinct profile in many regions, compatible with resilience mechanisms or with treatment effects in patients. As these findings are largely exploratory, they require confirmation in larger samples that also account for pharmacological treatment. Abbreviations: BD, patients with bipolar disorder; Fam, unaffected first-degree relatives; HC, healthy controls.
Introduzione Il disturbo bipolare (DB) è altamente ereditabile, ma i meccanismi cerebrali che separano il rischio dalla malattia restano poco chiari. Lo studio dei familiari non affetti permette di osservare la vulnerabilità senza l'effetto di farmaci, episodi e cronicità. Questo lavoro ha verificato se l'attività cerebrale locale a riposo segua un gradiente di rischio familiare, dai controlli ai familiari ai pazienti, oppure se i familiari mostrino un profilo proprio. Materiali e metodi Sono stati inclusi 104 partecipanti: 34 pazienti con DB, 34 familiari di primo grado non affetti e 36 controlli sani, appartenenti a 77 nuclei familiari. Dalla RM a riposo (3 Tesla, sequenza multiband) sono state calcolate ALFF, fALFF e ReHo. Le analisi voxel-wise hanno usato un modello full factorial con confronti a coppie e contrasto quadratico, con correzione TFCE (FWE p < 0,05). L'ipotesi ordinata è stata testata con fattori di Bayes in 28 regioni di interesse definite a priori, con errori standard robusti per famiglia. Risultati A p < 0,001 non corretto, i pazienti mostravano rispetto ai controlli una ridotta attività locale (fALFF e ReHo) nel giro angolare e nella corteccia occipitale laterale superiore di entrambi gli emisferi e nel giro temporale medio destro, e una ReHo aumentata nel verme cerebellare, nel cervelletto destro e nella corteccia occipitale inferiore. In diversi di questi cluster i familiari si collocavano in posizione intermedia, differendo sia dai pazienti sia dai controlli. Nelle 28 regioni definite a priori, il giro frontale inferiore destro (ReHo, BF = 15,2) e il precuneo destro (ALFF, BF = 15,5) mostravano un'evidenza forte per l'ordine HC > Fam > BD. Nella maggior parte delle regioni in cui i gruppi differivano, i familiari presentavano inoltre un profilo proprio, con valori più alti o più bassi di entrambi gli altri gruppi; tre di questi cluster superavano la correzione FWE. Dopo correzione FWE voxel- wise, l'aumento della ReHo nel verme si confermava come differenza tra pazienti e controlli, mentre nessun voxel mostrava un gradiente completo tra i tre gruppi. Conclusioni Nei pazienti con DB l'attività locale a riposo risulta alterata nella porzione parietale della rete di default, nel giro temporale medio e nel cervelletto, con un'alterazione del verme che supera la correzione per i confronti multipli. Il giro frontale inferiore destro e il precuneo destro mostrano un gradiente coerente con il rischio familiare, mentre i familiari non affetti presentano in molte regioni un profilo distinto, compatibile con meccanismi di resilienza o con effetti della terapia nei pazienti. Trattandosi di risultati in gran parte esplorativi, sono necessarie conferme su campioni più ampi, con l'inclusione della terapia farmacologica e permutazioni vincolate alla struttura familiare.
Caratterizzazione del gradiente di alterazioni neurofunzionali nel disturbo bipolare e nei familiari non affetti
ATZENI, MICHELA
2024/2025
Abstract
Introduction Bipolar disorder (BD) is highly heritable, yet the brain mechanisms that distinguish risk from illness remain poorly understood. Studying unaffected relatives makes it possible to examine vulnerability without the confounding effects of medication, mood episodes and chronicity. This study tested whether local resting-state brain activity follows a familial risk gradient, from healthy controls to relatives to patients, or whether relatives show a distinct profile. Materials and methods A total of 104 participants were included: 34 patients with BD, 34 unaffected first-degree relatives and 36 healthy controls, belonging to 77 families. The amplitude of low-frequency fluctuations (ALFF), fractional ALFF (fALFF) and regional homogeneity (ReHo) were computed from resting-state MRI (3 T, multiband sequence). Voxel-wise analyses used a full factorial model with pairwise comparisons and a quadratic contrast, with threshold-free cluster enhancement (TFCE) and family-wise error (FWE) correction (p < 0.05). The ordered hypothesis was tested with Bayes factors (BF) in 28 a priori regions of interest, using family-clustered robust standard errors. Results At an uncorrected threshold of p < 0.001, patients showed, relative to controls, reduced local activity (fALFF and ReHo) in the angular gyrus and superior lateral occipital cortex bilaterally and in the right middle temporal gyrus, and increased ReHo in the cerebellar vermis, right cerebellum and inferior occipital cortex. In several of these clusters, relatives occupied an intermediate position, differing from both patients and controls. Among the 28 a priori regions, the right inferior frontal gyrus (ReHo, BF = 15.2) and the right precuneus (ALFF, BF = 15.5) showed strong evidence for the order HC > Fam > BD. In most regions where the groups differed, relatives also showed a distinct profile, with values higher or lower than those of both other groups; three of these clusters survived FWE correction. After voxel-wise FWE correction, increased ReHo in the vermis was confirmed as a difference between patients and controls, whereas no voxel showed a complete gradient across the three groups. Conclusions In patients with BD, local resting-state activity is altered in the parietal component of the default mode network, in the middle temporal gyrus and in the cerebellum, with a vermal alteration that survives correction for multiple comparisons. The right inferior frontal gyrus and the right precuneus show a gradient consistent with familial risk, whereas unaffected relatives display a distinct profile in many regions, compatible with resilience mechanisms or with treatment effects in patients. As these findings are largely exploratory, they require confirmation in larger samples that also account for pharmacological treatment. Abbreviations: BD, patients with bipolar disorder; Fam, unaffected first-degree relatives; HC, healthy controls.| File | Dimensione | Formato | |
|---|---|---|---|
|
Atzeni_Michela.pdf
Accesso riservato
Dimensione
1.22 MB
Formato
Adobe PDF
|
1.22 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/119757