Fabienne Cyprien1, Philippe Courtet2, Vanessa Poulain3, Jerome Maller4, Chantal Meslin5, Alain Bonafé6, Emmanuelle Le Bars7, Marie-Laure Ancelin8, Karen Ritchie9, Sylvaine Artero10. 1. Inserm, U1061, La Colombière Hospital, Montpellier F-34093, France; University of Montpellier 1, Montpellier F-34000, France; CHRU Carémeau, Nîmes, France. 2. Inserm, U1061, La Colombière Hospital, Montpellier F-34093, France; University of Montpellier 1, Montpellier F-34000, France; CHRU Montpellier, Montpellier, France. 3. Inserm, U1061, La Colombière Hospital, Montpellier F-34093, France. 4. Monash Alfred Psychiatry Research Centre, The Alfred & Monash University School of Psychology and Psychiatry, Melbourne, Australia. 5. Centre for Mental Health Research, Australian National University, Canberra, Australia. 6. University of Montpellier 1, Montpellier F-34000, France; CHRU Montpellier, Montpellier, France. 7. CHRU Montpellier, Montpellier, France. 8. Inserm, U1061, La Colombière Hospital, Montpellier F-34093, France; University of Montpellier 1, Montpellier F-34000, France. 9. Inserm, U1061, La Colombière Hospital, Montpellier F-34093, France; University of Montpellier 1, Montpellier F-34000, France; Faculty of Medicine, Imperial College, St Mary׳s Hospital, London, United Kingdom. 10. Inserm, U1061, La Colombière Hospital, Montpellier F-34093, France; University of Montpellier 1, Montpellier F-34000, France. Electronic address: sylvaine.artero@inserm.fr.
Abstract
BACKGROUND: Recent research on late-life depression (LLD) pathophysiology suggests the implication of abnormalities in cerebral white matter and particularly in interhemispheric transfer. Corpus callosum (CC) is the main brain interhemispheric commissure. Hence, we investigated the association between baseline CC measures and risk of LDD. METHODS: We studied 467 non-demented individuals without LLD at baseline from a cohort of elderly community-dwelling people (the ESPRIT study). LLD was assessed at year 2, 4, 7 and 10 of the study follow-up. At baseline, T1-weighted magnetic resonance images were manually traced to measure the mid-sagittal areas of the anterior, mid and posterior CC. Multivariate Cox proportional hazards models stratified by sex were used to predict LLD incidence over 10 years. RESULTS: A significant interaction between gender and CC size was found (p=0.02). LLD incidence in elderly women, but not in men, was significantly associated with smaller anterior (HR 1.37 [1.05-1.79] p=0.017), mid (HR 1.43 [1.09-1.86] p=0.008), posterior (HR 1.39 [1.12-1.74] p=0.002) and total (HR 1.53 [1.16-2.00] p=0.002) CC areas at baseline in Cox models adjusted for age, education, global cognitive impairment, ischemic pathologies, left-handedness, white matter lesion, intracranial volume and past depression. LIMITATIONS: The main limitation was the retrospective assessment of major depression. CONCLUSION: Smaller CC size is a predictive factor of incident LLD over 10 years in elderly women independently of cognitive deterioration. Our finding suggests a possible role of CC and reduced interhemispheric connectivity in LLD pathophysiology. Extensive explorations are needed to clarify the mechanisms leading to CC morphometric changes in mood disorders.
BACKGROUND: Recent research on late-life depression (LLD) pathophysiology suggests the implication of abnormalities in cerebral white matter and particularly in interhemispheric transfer. Corpus callosum (CC) is the main brain interhemispheric commissure. Hence, we investigated the association between baseline CC measures and risk of LDD. METHODS: We studied 467 non-demented individuals without LLD at baseline from a cohort of elderly community-dwelling people (the ESPRIT study). LLD was assessed at year 2, 4, 7 and 10 of the study follow-up. At baseline, T1-weighted magnetic resonance images were manually traced to measure the mid-sagittal areas of the anterior, mid and posterior CC. Multivariate Cox proportional hazards models stratified by sex were used to predict LLD incidence over 10 years. RESULTS: A significant interaction between gender and CC size was found (p=0.02). LLD incidence in elderly women, but not in men, was significantly associated with smaller anterior (HR 1.37 [1.05-1.79] p=0.017), mid (HR 1.43 [1.09-1.86] p=0.008), posterior (HR 1.39 [1.12-1.74] p=0.002) and total (HR 1.53 [1.16-2.00] p=0.002) CC areas at baseline in Cox models adjusted for age, education, global cognitive impairment, ischemic pathologies, left-handedness, white matter lesion, intracranial volume and past depression. LIMITATIONS: The main limitation was the retrospective assessment of major depression. CONCLUSION: Smaller CC size is a predictive factor of incident LLD over 10 years in elderly women independently of cognitive deterioration. Our finding suggests a possible role of CC and reduced interhemispheric connectivity in LLD pathophysiology. Extensive explorations are needed to clarify the mechanisms leading to CC morphometric changes in mood disorders.
Authors: Marnix J M van Agtmaal; Alfons J H M Houben; Frans Pouwer; Coen D A Stehouwer; Miranda T Schram Journal: JAMA Psychiatry Date: 2017-07-01 Impact factor: 21.596