MRI Study On Brain Functional And Structural Changes In First-episode Patients With Major Depressive Disorder

Object The aim of this study was to explore the brain functional and structuralpathological mechanism in first-episode major depressive disorder (MDD), by bloodoxygenation level dependent functional magnetic resonance imaging (BOLD-fMRI),three dimension MRI (3D), and diffusion tensor imaging (DTI) techniques.Methods Twenty seven first-episode treatment-naive young adult with MDD(according to the diagnosis critietia of CCMD-Ⅲand DSM-Ⅳ) and twenty eight ageand gender-matched healthy controls were assessed using 17-item HamiltonDepression Rating Scale (HAMD), Hamilton Anxiety Rating Scale (HAMA), LifeEvent Scale (LES), and Understanding and recognition test of Wechsler MemoryScale (WMS). HAMD sorces were requested≥24 in patients and≤7 in controls. Then,MRI scan were obtained ordering to BOLD-fMRI (including four task, emotionalcount Stroop, emotional face, emotional picture and the recall of short-term memory),3D, DTI. We assessed patients like above again after six months antidepressant(Venlafaxine) treatment.Results Twenty two patients and 24 controls finished MRI scan. 16 patientswere followed. Some data were excluded because of head movement. Data could beanalyzed finally including 18 pre-treatment patients (4 males) and 16 post-treatmentpatients (4 males) in fMRI, 20 pre- patients (4 males) and post- patients (4 males) in3D; 19 pre- patients (4 males) and 13 post- patients (4 males) in DTI. Data of sixteenpatients and sixteen age and gender-matched healthy controls were analyzed in fMRIpart, and 13 patients and 13 age and gender-matched healthy controls were analyzedin 3D and DTI part. 1. HAMD and HAMA scores in post-treatment patients weresignificantly lower than those in pre-treatment (P=0.000), however still higher thanthose in healthy controls (P=0.01). Understanding and recognition tests scores werelower significantly in pre-treatment patients compared with healthy controls (P=0.001, P=0.04), and were similar between post-treatment patients and controls(P＞0.05). LES total scores and negative scores were different between pre-treatmentpatients and controls (P=0.000). 2. fMRI(1) ecStroop task The bilateral cingulategyfi, several regions of prefrontal lobe, temporal and parietal lobe had increasedactivation in negative words compared with neutral words in patients, which isconverse in controls. And bilateral cerebellum had decreased activation in negativewords than neutral words both in patients and controls. Pre-treatment patients showed decreased activation in right cingulate and left middle temporal gyrus, increasedactivation in left cerebellum compared with controls. No areas with significantlydifferent activation were found between pre- and post-treatment patients and betweenpost-treatment patients and controls. (2) Emotional face task There were increasedvisual cortex and parahippocampal gyrus activation and decreased right prefrontalcortex activation in sad face compared with happy face in both patients and controls.The caudate nucleus activation increased in sad face than happy face in patients,which is converse in controls. Compared with healthy controls, MDD patients showedincreased activation in right posterior cingulated and left insula, decreased activationin bilateral prefrontal, left inferior parietal lobule, left fusiform gyrus, right uncus andright parahippocampal gyrus. Post-treatment patients showed increased activation inright prefrontal cortex, left temporal lobe, bilateral visual cortex and left basal gangliathan pre-treatment, and increased activation in bilateral prefrontal lobe, right cingulategyrus and right temporal lobe than controls. (3) Emotional picture Both patientsand controls showed increased activation in bilateral visual cortex in negative picturecompared with positive picture. Compared with positive picture, negative pictureincreased bilateral amygdala and left uncus activation, and decreased activation in leftmiddle frontal gyrus, right anterior and posterior cingulate, right superior temporalgyrus, right parietal lobe and bilateral insula in MDD patients. Both negative andpositive picture activated prefrontal lobe in controls. Patients showed decreasedactivation in several areas of bilateral prefrontal lobe and left uncus compared withcontrols. Right middle frontal gyrus, right posterior cingulate, bilateral temporal lobeand left thalamus activation increased in post-treatment patient than pre-treatment.However, the right middle frontal gyrus, right inferior parietal lobule and leftprecuneus showed lower activation than controls yet. (4) Recall of short-termmemory task Compared with positive picture, negative picture increased theactivation in bilateral prefrontal and left parietal lobe, left anterior cingulate, lefthippocampus, left basal ganglia and left middle occipital lobe, and decreasedactivation in left caudate head and bilateral insula in MDD patients. Controlsshowed smaller decreased activation of left anterior cingulate and right medial frontalgyrus in negative picture than positive picture. Compared with controls, MDDpatients showed increased activation in several regions of bilateral prefrontal andparietal lobe, right middle temporal gyrus, bilateral middle occipital gyrus. Aftertreatment, the bilateral prefrontal lobe, temporal lobe and visual cortex activation decreased than pre-treatment patients. However, the bilateral prefrontal lobe, rightangular gyrus and right visual cortex still showed increased activation than controls.(5) Stressful life event (SLE) Patients with and without SLE also showed increasedactivation in prefrontal lobe in ecStroop task, and patients with SLE showeddecreased activation in right insula and left inferior temporal lobe than those withoutSLE. Compared with those without SLE, patients with SLE showed increasedactivation in prefrontal lobe, right parahippocampal gyrus, right caudate body,bilateral thalamus in face task, right anterior cingulate in picture task, whereas,decreased activation in left anterior cingulate in face task, and bilateral parietal lobeand middle occipital gyrus and left canduate body in picture task, and left precentralgyrus, left parietal lobe and bilateral temporal lobe in recall task. Furthermore, inpicture task, both patients with and without SLE showed increased activation inseveral areas of prefrontal lobe, right temporal lobe and left putamen. (6) SeverityCompared with patients with lower HAMD scores, those patients with high HAMDscores showed decreased activation in many regions, including bilateral middlefrontal gyrus, left superior temporal lobe and left insula in ecStroop task, bilateralprefrontal lobe, right cingulate, right thalamus and right caudate body in face task, leftuncus, right pamhippocampal gyrus, right putamen and bilateral insula in picture task,and bilateral prefrontal lobe and anterior cingulate, left insula, right cuneus and rightparahippocampal gyrus in recall task. Moreover, patients with higher HAMD scoresshowed decreased activation in several areas in frontal lobe and left hippocampus inpicture task than those with lower HAMD scores. (7) Course of disease Comparedwith patients with short course, the patients with long course showed increasedactivation in right middle frontal gyrus, right inferior parietal lobule and right superioroccipital gyrus in ecStroop task, in left uncus, left caudate tail and right thalamus inpicture task, bilateral prefrontal lobe (BA10), right middle occipital and rightparahippocampal gyrus in recall task. However, patients with long duration showeddecreased activation in some other regions than those patients with long duration,including bilateral temporal lobe, left superior parietal lobule and left insula inecStroop task, right superior frontal gyrus, right thalamus, right claustrum, right visualcortex and left precuneus in face task, several areas of prefrontal lobe, bilateralparietal lobe and visual cortex in picture task. (8) Short-time memory (WMSUnderstanding score) Compared with patients with lower Understanding scores, thepatients with higher Understanding scores showed increased activation in right prefrontal lobe, left anterior cingulated and middle temporal gyrus, parietal cortex,bilateral visual cortex, left thalamus and caudate body in ecStroop task, in bilateralprefrontal, temporal and visual cortex, left cingulate, right parietal cortex and insula,and bilateral cerebellum in face task, in bilateral prefrontal cortex, superior temporalgyri and parietal cortex, left insula in picture task. However, patients with highUnderstanding scores showed decreased activation in some other regions than thosepatients with lower Understanding scores, including left amygdala andparahippocampal gyrus, right uncus, bilateral caudate in face task, and bilateralparahippocampal gyri, left subcallosal gyrus, right medial frontal gyrus, right caudateand cerebellum in picture task, and left anterior cingulate, bilateral superior temporalgyri and right middle occipital gyrus in recall task. 3. 3D MDD patients showedsignificantly lower gray matter density than healthy controls in right medial frontalgyrus, right temporal pole, bilateral putamen and globus pallidus, bilateral cerebellum.After treatment, the gray matter density in right superior frontal gyrus, bilateralprecentral gyrus, temporal pole, putamen and globus pallidus, and cerebellum inpatients were still significantly lower than controls. 4. DTI Compared with controls,MDD patients exhibited significant decreased fractional anisotropy (FA) values in leftsuperior frontal gyrus, bilateral middle frontal gyrus and bilateral anterior cingnlate.FA values in bilateral superior and middle gyrus, right inferior frontal gyrus increasedafter treatment. However, FA values in left middle gyrus in post-treatment patientswere still lower than controls. Furthermore, FA values in right middle and superiortemporal gyrus were significantly higher than controls. 5. There were no regionswith significant different gray matter density and white matter FA values were foundbetween patient with long and short duration, between patients with high and lowHAMD scores, and between patients with or without SLE.Conclusions 1. Decreased gray matter density in right medial frontal gyrus, righttemporal pole, bilateral putamen and globus pallidus, bilateral cerebellum may be theneuropathological substrate of MDD rather than a result. 2. This is the first study toreport that white matter microstructure abnormalities of prefrontal lobe and anteriorcingulate in first-episode, treatment-naive young adult MDD, suggesting thatabnormalities of brain white matter may be present early in the course of MDD. 3.Dysfunction in bilateral prefrontal lobe, cingulate, parietal and visual cortex maycontribute to the neuropathology of cognitive impairments in MDD. 4. Abnormalhyperfunction in limbic system and hypofunction in prefrontal and parietal lobe may contribute to neuropathology of mood processing bias in MDD. 5. Functional andwhite matter microstructure abnormalities could be improved by effectiveantidepressant treatment. 6. Occurring of SLE may induce greater bias to perceive sadstimuli in MDD, but no impairment to attention and execution function. And patientswith SLE may have little memory impairment than those without SLE. 7. Memoryand the ability to response to emotional stimuli may be negative correlation withsymptoms severity and illness course of MDD. Execution function and attention maybe negative correlation with symptoms severity of depression. 8. Lower gray matterdensity and FA values may increase the risk of MDD episode, however, no correlationwith the course, severity and the occurring of SLE.