ࡱ > {3 bjbjss p {+ T T $ L h, L $ + + + + + + + $ J. 0 T
,
, ",
% + a S6% : y % 8, 0 h, , @1 P @1 J @1
,
, h, @1 T ] : Supplementary material
A novel resting-state functional MRI signature of resilience to recurrent depression
Clifford I. Workman, Karen E. Lythe, Shane McKie, Jorge Moll, Jennifer A. Gethin, John F. W. Deakin, Rebecca Elliott, Roland Zahn
Supplementary Results
In an earlier cross-sectional resting-state fMRI study which included the participants studied here, our group found subgenual cingulate-amygdala resting-state functional disconnection to be distinctive of remitted depressed patients with a history of melancholic major depressive episodes (MDE) compared to non-melancholic and healthy control groups ADDIN EN.CITE Workmanin press344(Workman in press)34434417Workman, C. I.Lythe, K. E.McKie, S.Moll, J.Gethin, J. A.Deakin, J. F. W.Elliott, R.Zahn, R.Subgenual cingulate-amygdala functional disconnection and vulnerability to melancholic depressionNeuropsychopharmacologyNeuropsychopharmacologyNeuropsychopharmacologyNeuropsychopharmacologyin press(HYPERLINK \l "_ENREF_3" \o "Workman, in press #344"Workman et al. 2016). We argued that subgenual cingulate-amygdala functional disconnection is a signature of primary vulnerability for melancholic major depressive disorder (MDD). In view of the present findings suggesting lower interhemispheric connectivity between the subgenual cingulate cortices promotes resilience to recurring MDEs, we wanted to determine whether the network of lower connectivity we previously observed in the melancholic remitted MDD (rMDD) patients is better understood as promoting resilience. To this end, we first extracted the mean Fisher Z-transformed correlation coefficients from the amygdala cluster as described previously ADDIN EN.CITE Workmanin press344(Workman in press)34434417Workman, C. I.Lythe, K. E.McKie, S.Moll, J.Gethin, J. A.Deakin, J. F. W.Elliott, R.Zahn, R.Subgenual cingulate-amygdala functional disconnection and vulnerability to melancholic depressionNeuropsychopharmacologyNeuropsychopharmacologyNeuropsychopharmacologyNeuropsychopharmacologyin press(HYPERLINK \l "_ENREF_3" \o "Workman, in press #344"Workman et al. 2016) for each participant in the current study. These data were then entered into a two-way ANOVA in SPSS 20 with between-subjects factors for group (resilient or recurring episode MDD) and for subtype (melancholic or non-melancholic). Results were considered significant at p<0.05 two-tailed.
For subgenual cingulate-amygdala resting-state connectivity, we observed a main effect of subtype (F(1,43)=7.1, p=0.01) but no main effect of group (F(1,43)=0.001, p=0.97) and no subtype by group interaction (F(1,43)=0.33, p=0.57). Subsequent post-hoc pairwise comparisons revealed lower subgenual cingulate-amygdala connectivity in the melancholic rMDD group (M=0.12, SD=0.17) compared to the non-melancholic rMDD group (M=0.25, SD=0.11, p=0.01, mean difference=-0.12, 95% CI [-0.21,-0.03]). These results suggest the network of lower functional connectivity we previously reported in the melancholic rMDD patients is independent of vulnerability or resilience to recurring MDEs, which is in keeping with our original interpretation of subgenual cingulate-amygdala functional disconnection as a primary vulnerability factor for melancholia.
Supplementary Table S1. Inter-rater reliability for the SCID-I, MADRS, and PSR scales
SCID-I subtypeCurrent MADRSMADRS
previous MDECurrent PSRHighest PSR during follow-upRatersKappa ValueICC ValueICC ValueICC ValueICC ValueRZ & KL0.600.630.450.960.98RZ & JG0.910.80KL & JG1.000.860.80KL & CW0.960.99Mean0.800.800.680.960.98
Reliability for the SCID-I mood disorders module subtype diagnosis is given as a kappa value. Reliability for the MADRS and PSR are given as intra-class correlation (ICC) values (two-way mixed with absolute agreement). RZ, KL, and JG completed the recommended training for the SCID-I for DSM-IV-TR, and RZ, KL, and CW completed the recommended training for the PSR. The SCID-I was modified to allow lifetime diagnoses of MDD subtypes, including melancholic and atypical specifiers. The MADRS was used to assess depression severity at the time of the clinical interview, and was modified to allow for retrospective assessment of the last and most severe MDE. The PSR was used to assess the severity of and impairment caused by depressive symptoms present at each follow up interview and retrospectively throughout the follow up period. The Kappa values for the SCID-I subtype diagnoses reflect moderate to perfect agreement ADDIN EN.CITE Landis1977304(Landis 1977)30430417Landis, J. RichardKoch, Gary G.The measurement of observer agreement for categorical dataBiometricsBiometricsBiometricsBiometrics159-743311977(HYPERLINK \l "_ENREF_2" \o "Landis, 1977 #304"Landis et al. 1977), and ICC values for the MADRS (both current and previous MDE) and PSR reflect moderate to excellent agreement ADDIN EN.CITE Fleiss1986303(Fleiss 1986)3033036Fleiss, J.L.The Design and Analysis of Clinical Experiments1986New YorkWiley(HYPERLINK \l "_ENREF_1" \o "Fleiss, 1986 #303"Fleiss 1986). ICC, intra-class correlation; MADRS, Montgomery-sberg Depression Rating Scale; MDD, major depressive disorder; MDE, major depressive episode; PSR, Psychiatric Status Rating; SCID-I, Structured Clinical Interview-I.
Supplementary Table S2. Reasons for exclusion of volunteers from the current study
Reasons for ExclusionNTelephone ScreeningMRI contraindications77Psychiatric disorders other than MDD54Current antidepressants or other centrally active medications52Withdrawal after telephone screening33Not meeting full screening criteria for MDD30Family history of MDD/bipolar/schizophrenia (HC group)26Substance or alcohol abuse23Current antihypertensive or statin medications20Left-handed20Non-native English speaker19Thyroid function problems19Fulfilling criteria for current MDD13History of cancer7Not remitted for long enough (>6 months)7Epilepsy5No reason recorded5Other general medical conditions5Diabetes4Out of age range (18 65 years)4Excluded because of age-matching (HC group)3Multiple sclerosis3History of stroke1Vitamin D deficiency1Total excluded after the telephone screening431 / 707Clinical Interview (remitted MDD patients)Unable to schedule for additional visits10Fulfilling criteria for a bipolar disorder6Fulfilling criteria for current social anxiety disorder6Not meeting full criteria for MDD5Fulfilling criteria for past substance abuse4Not remitted for long enough (>6 months)3Residual symptoms of post-traumatic stress disorder3Probable personality disorders2Fulfilling criteria for current generalized anxiety disorder1MRI contraindications1Withdrawal after the clinical interview1Total number of remitted MDD patients excluded after the clinical interview42 / 138Clinical Interview (HC group)Unable to schedule for additional visits6Probable or definite positive first degree family history of MDD4Fulfilling criteria for a past MDE lasting less than two months1Fulfilling criteria for current adjustment disorder1Fulfilling criteria for current MDD1Fulfilling criteria for current social anxiety disorder1Non-native English speaker1Past depressive episode not fulfilling criteria for a past MDE1Total number of HC participants excluded after the clinical interview16 / 64
Of the 707 volunteers who completed the telephone screening, 276 were eligible (184 remitted MDD patients, 92 HC participants). Of these, 202 participants agreed to complete the clinical interview after having reviewed the studys participant information sheet (138 remitted MDD patients, 64 HC participants). Following the clinical interview, 144 participants were eligible to complete the remaining study visits (96 remitted MDD patients, 48 HC participants). Of these, 102 participants underwent resting-state fMRI scanning (63 remitted MDD patients, 39 HC participants). fMRI, functional magnetic resonance imaging; HC, healthy control; MDD, major depressive disorder; MDE, major depressive episode.
REFERENCES
ADDIN EN.REFLIST Fleiss, JL (1986). The Design and Analysis of Clinical Experiments. Wiley: New York.
Landis, JR & Koch, GG (1977). The measurement of observer agreement for categorical data. Biometrics 33, 159174.
Workman, CI, Lythe, KE, McKie, S, Moll, J, Gethin, JA, Deakin, JFW, Elliott, R & Zahn, R (2016). Subgenual cingulate-amygdala functional disconnection and vulnerability to melancholic depression. Neuropsychopharmacology 41, 20822090.
j l m n $ % = j m u v { ӽ}ung`gYgRKgK hD hXe$ hD hN hD h&hE