Structural MRI findings in MS and comorbid major depressive disorder
| Study group | Patient group |
MRI Scanner, slices (mm), method | Psychometric instruments | Disability status | Findings | ||
| MS | MS+comorbid psychiatric disorder | Healthy controls | |||||
| Reischies et al (1988)126 | 46 | Neurologist ratings (depression=11%, euphoria=11%, irritability=22%, impaired judgement=9%, reduced drive=2%, lability of affect=11%) | – | 0.35 T Magnetom, 10 mm slices Radiological evaluation of lesions | No standard instruments, 7-point scale for each of five symptoms (depression, euphoria, irritability, impaired judgement, reduced drive, with lability of mood scored for extent of occurrence of depressed mood and euphoria simultaneously) | – | The presences of lesions in the frontal lobes were related to higher total scores of psychopathology. Periventricular lesions associated with lability of affect, disorders of judgement, depression and euphoria |
| Möller et al (1994)15 | 19 | 6 | – | 0.5 T Scanner (make not given), 5 mm slices, enhanced with gadolinium-pentetic acid. Radiological evaluation of lesions, video display ball and track cursor method used for VBR and CC measurements | Fatigue Severity Scale HDRS MADRS | Mean EDSS depressed group=3.2 and in non-depressed group=3.1 | No association between scores of depression and lesion load, gadolinium-enhanced lesions or VBR |
| Tsolaki et al (1994)127 | 22 | Mean HDRS=11.8 (SD=6) and not significantly greater than controls | 15 | 0.5 T GE Scanner, 5 mm slices Radiological evaluation of lesions and measurement of CC | HDRS (several cognitive scales) | Mean KDSS score=3.3 | No correlation between HDRS and lesion load or volume of the CC |
| Sabatini et al (1996)128 | 10 (All RR) | 10 (All RR) | – | 1.5 T Siemens Magnetom Scanner, 6 mm slices with 0.3 mm gap. Radiological evaluation of lesions. SPECT | BDI HDRS SAS | – | No difference in number, size, location or area of demyelinating lesions SPECT demonstrated higher CBF in left limbic cortex and lower CBF in right limbic cortex in depressed group |
| Fassbender et al (1998)26 | 23 (All RR) | 4/23 individuals fulfilled DSM-IIIR criteria for major depression | 17 | 1.5 T superconducting unit, Siemens Scanner, enhanced with gadolinium-pentetic acid. Radiological evaluation of lesions | HAS HDRS | Mean EDSS=3.5 | Individuals with gadolinium-enhancing lesions had higher scores of depression and anxiety |
| Diaz-Olavarrieta et al (1999)124 | 44 | 42/45 had symptoms on NPI (dysphoria=79%, agitation=40%, anxiety=37%, apathy=20%, euphoria=13%, irritability=13%, hallucinations=10%, delusions=7%) | 25 (84% had no symptoms based on the NPI) | Gadolinium-enhanced MR images Radiological evaluation of lesions | MMSE NPI | Mean EDSS=3.3 | Only euphoria and hallucinations correlated with the severity of demyelination |
| Bakshi et al (2000)36 | 29 (RR=23, SP=6) | 19 (RR=10, SP=9) | – | 1.5 T Philips Gyroscan Scanner, 5 mm slices, radiological evaluation of lesions and atrophy | BDI HDRS | Mean EDSS=4.3 for depressed group and 2.9 for non-depressed group | Higher lesion load in superior frontal and superior parietal lobes in depressed group. Lateral and third ventricular enlargement and frontal atrophy noted in depressed group |
| Berg et al (2000)37 | 47 (RR=20, SP=23, PP=4) | 31 (RR=17, SP=10, PP=4) | – | 1.5 T Siemens Magnetom Vision or 1.5 T Gyroscan 15 Philips Scanner, 5 mm slices, radiological evaluation of lesions | BDI Fatigue Severity Scale GHQ HDRS MADRS | Median EDSS=6 for depressed group and 5 for non-depressed group | Higher lesion load in right temporal lobe in depressed patients with a trend towards an increased lesion load in the right parietal lobe, right frontal lobe and cerebellum. Lesion load correlated with BDI, HDRS and MADRS scores in the right temporal lobe |
| Pujol et al (2000)129 | 45* | 9 | – | 1.5 T Sigma Systems, GE Scanner, 5 mm slices with a 2.5 mm gap, evaluation of myelin loss through coronal inversion recovery sequence Tissue segmentation | BDI | Mean EDSS=3.3 | Lesions in the left arcuate fasciculus were significantly related to affective symptoms, somatic complaints and BDI scores |
| Zorzon et al (2001)34 | 77 (predominantly RR, but exact figures not given)† | 18 | 110=HC (4=depression) 97=chronic diseases (16=depression) | 1.5 T Philips, Gyroscan Scanner, 5 mm slices, semiautomated measurements of lesion load and regional brain volumes | FIM HAI HDRS MMSE | Median EDSS=2.0 | Greater depression scores in patients with mainly cerebral involvement than those with spinal cord lesion localization The severity of depressive symptoms was weakly correlated with right frontal lesion load and right temporal lobe volume |
| Benesova et al (2003)33 | 10 | 10 | – | 1.5 T Scanner Radiological evaluation of lesions | HDRS MADRS SDS | Mean EDSS=2.5 | Increased lesion load area in right frontal lobe in MS+MDD group. Highest number of lesions was observed in the right frontal lobe in MS+MDD group |
| Feinstein et al (2004)35 | 19 (RR=7, SP=10; PP=2) | 21 (RR=8, SP=12, PP=1) | – | 1.5 T Sigma Systems, GE Scanner, 3 mm slices, Semiautomated measurements of lesion load and regional brain volumes | BDI HDRS | Mean EDSS 4.7 in depressed group and 4.6 in non-depressed group) | Less GM volume and more CSF volume in left anterior temporal region Greater T2 weighted lesion volume and extensive T1 weighted lesion volume in left medial inferior prefrontal cortex |
| Feinstein et al (2010)43 | 32 (BDI<19) (13 taking antidepressants) (RR=26; SP=4; PP=2) | 30 (BDI>19) (20 taking antidepressants) (RR=23, SP=7) | – | 1.5 T Sigma Systems Scanner, 1.2–1.4 mm slices, manual tracing of lesions and semiautomated brain region extraction DTI analysis, FA and MD maps attained | BDI | Mean EDSS in depressed group=3.5 and 2.5 in non-depressed group | Individuals with depression had reduced white matter volume in the left superior frontal lobe and greater lesion volume in the right medial inferior frontal lobe Higher MD in the grey matter of the left anterior temporal lobe and in lesions in the right inferior frontal region. Reduced FA in the white matter of the left anterior temporal lobe |
| Gold et al (2010)39 | 21 (BDI<13) (All RR) | 8 (BDI>13) (All RR) | 20 | 3 T Siemens Scanner, 1 mm slices. Automated measurement of brain volume and manual tracing of lesions. Segmentation of hippocampus | BAI BDI | Mean EDSS=2.5 | No difference in brain atrophy or lesion volume between depressed and non-depressed groups. Both MS groups demonstrated smaller hippocampal volume compared with controls. The depressed group had smaller dentate gyrus volume compared with non-depressed MS individuals and dentate gyrus was inversely correlated with depression as measured by the BDI (r=−0.38, p=0.04) |
| Papadopolou et al (2013)41 | 91 | Of 89 tested, mean CES-D=13.6, SD=10.5) | – | 1.5 TAvanto, Siemens Scanner, 1.5 mm slices. Radiological evaluation of lesions and semiautomated mechanism for lesion volume used | CES-D (several cognitive scales) | Median EDSS=3.6 | No association with lesion load or volume (WM or cortical lesions) and depression scores |
| Gobbi et al (2014)42 | 54 | 69 | 90 | 3 T scanner Automatic segmentation technique for evaluation of lesions VBM | MADRS | Median EDSS in depressed group=2.5, and 2 in non-depressed group | No difference in lesion frequency or location between the MS groups No difference in WM atrophy between the MS groups. Atrophy of the left middle frontal gyrus and right inferior frontal gyrus was selectively related to depression |
| Gold et al (2014)40 | 83 (CES-D<20) (women) | 26 (CES-D>20) (women) | – | 3 T Scanner, 1 mm slices, a number of scanners used SIENA X programme generated hippocampal volumes. Shape analysis with surface mesh modelling | CES-D | Mean EDSS 3.36 in depressed group and 3.08 in non-depressed group | No difference in lesion load between the two groups. Smaller right hippocampal volume in depressed group (p=0.04), but no difference in left hippocampal or total hippocampal volume. Areas of hippocampal volume reduction included the CA2-3 region and posterior subiculum Right hippocampal shape (reduced thickness) associated with depressive symptoms (affective but not vegetative symptoms) |
| Kern et al (2015)130 | 27 RR | 4/27 RR had severe depressive symptoms | 20 | 3 T Scanner Trackvis Deterministic fibre tracking | EDSS BDI-II | Mean EDSS 2.5 | Depressive symptoms associated with smaller volumes in hippocampal subregions CA2, CA3, and dentate gyrus Depressive symptoms associated with lower scores in spatial memory and attention/executive function Hippocampal and thalamic atrophy in RRMS group. Reduced white matter FA in cingulum, fornix and uncinate fasciculus |
| Shen et al (2014)47 | 15 RR | 9/15 depressive symptoms | 15 | 3 T Scanner TBSS | EDSS HAMD | Mean EDSS 1.73 | HAMD scores positively correlated with FA values in left hypothalamus, right posterior middle cingulate gyrus and hippocampus, right precentral gyrus and posterior cingulate WM Depressive symptoms negatively associated with demyelinating lesions in limbic system and frontal lobe |
| Hanken et al (2015)49 | 49 RR | 37/49 cognitive fatigue patients with depressive symptoms | 14 | 3 T Scanner NeuroQLab3.531 Fibre tracking | EDSS BDI | Mean EDSS (3.2 CF/4.1 CNF) | Interaction effects between BDI scores and WM between prefrontal cortex and pons |
| Nigro et al (2015)50 | 22 | 20 | 16 | 3 T Scanner LST lesion segmentation toolbox Brain connectivity toolbox | EDSS BDI-II | Mean EDSS (2.3 non-MDD, 2.8 MDD) | Increased nodal path length in right hippocampus and amygdala in patients with MD in MS Increased shortest distance between right hippocampus and right amygdala and dorsolateral and ventrolateral prefrontal cortex, orbitofrontal cortex, and sensorimotor cortices and supplementary motor area |
| Nygaard et al (2015)46 | 61 RR | 16/61 | 61 | 1.5 T Scanner Cortical thickness, surface area and volume | EDSS BDI | Mean 1.9 | Depressive symptoms related to reduced surface area in frontal pole, pars orbitalis, orbital frontal, rostral and caudal middle frontal, and pre-central and post-central regions bilaterally, and left hemisphere included middle temporal, fusiform and parahippocampal regions Volume reductions were associated with similar regions associated to surface area findings and extended to right supramarginal and superior temporal regions and inferior temporal of left hemisphere. |
*Study includes the individuals investigated in Pujol et al (1997).129.
†A 2-year follow-up study on same cohort of patients demonstrated no difference in lesion load between the groups, but noted increased brain atrophy in left frontal lobe in the depressed cohort (Zorzon et al,).44
BDI, Beck Depression Inventory; CBF, cerebral blood flow; CC, corpus callosum; CES-D, Center for Epidemiological Studies Depression Scale; CIL, clinically isolated lesions; CP, chronic progressive; CIS, Clinical Interview Schedule; DSM, Diagnostic Statistics Manual; DTI, diffusion tensor imaging; EDSS, Expanded Disability Scale; FA, fractional anisotropy; FIM, Functional Independence Measure; HADS, Hospital Anxiety and Depression Scale; HAS, Hamilton Anxiety Scale; HC, healthy controls; HDRS, Hamilton Depression Rating Scale; KDSS, Kurtzke Disability Status Scale; MADRS, Montgomery and Asberg Depression Scale; MMSE, Mini-Mental State FPujilExamination; MD, mean diffusivity; MS, multiple sclerosis; NPI, Neuropsychiatric Inventory; PD, physically disabled; PSE, Present State Examination; RR, relapse remitting; SAS, Zung Self-rated Anxiety Scale; SCL, symptom check-list; SDS, Zung Self-rated Depression Scale; SPECT, single-photon emission CT; SSSI, Social Stress and Support Interview; VBM, voxel-based morphomety; VBR, ventricular brain ratio.