Auditory Thalamocortical Resting-State Hyperconnectivity in Schizophrenia
Poster No:
696
Submission Type:
Abstract Submission
Authors:
John Williams1, Zu Zheng2, Philip Tubiolo3, Roberto Gil1, Greg Perlman1, Jodi Weinstein1, Natalka Haubold1, Eilon Silver-Frankel1, Mark Slifstein1, Guillermo Horga4, Anissa Abi-Dargham1, Jared Van Snellenberg1
Institutions:
1Renaissance School of Medicine at Stony Brook University, Stony Brook, NY, 2SUNY Downstate Health Sciences University, Brooklyn, NY, 3Stony Brook University, Stony Brook, NY, 4New York State Psychiatric Institute, New York, NY
First Author:
Co-Author(s):
Introduction:
Auditory hallucinations, and positive symptoms broadly, are a core component of schizophrenia (SCZ) and psychosis generally, yet their etiology remains unknown. Findings of aberrant thalamocortical resting-state (RS) functional connectivity (RSFC) have been mixed, though convergently showing hyperconnectivity between thalamus and somatosensory cortex. However, limitations in demarcating medial geniculate nucleus (MGN) from other thalamic nuclei suggests that the degree to which these results represent abnormalities within the auditory processing pathway is unclear.
Here, we aim to specifically measure RSFC between auditory thalamus (MGN) and auditory cortex (AC) in unmedicated patients with SCZ (PSZ ) and healthy control participants (HC), using a sensory thalamic localizer (TL) fMRI task in tandem with resting-state fMRI data acquired from the same subjects.
Here, we aim to specifically measure RSFC between auditory thalamus (MGN) and auditory cortex (AC) in unmedicated patients with SCZ (PSZ ) and healthy control participants (HC), using a sensory thalamic localizer (TL) fMRI task in tandem with resting-state fMRI data acquired from the same subjects.
Methods:
53 PSZ and 45 HC completed study procedures. PSZ were medication-free for at least 3 weeks for reasons unrelated to this study . Symptom severity was assessed using the Positive and Negative Syndrome Scale (PANSS). Participants completed 4 runs each of resting-state fMRI and the TL fMRI task. Data were preprocessed using HCP minimal preprocessing pipeline 4.2.0. During the TL task, alternating auditory and visual stimuli were displayed, with task details shown in Figure 1.
TL task data was analyzed by first identifying AC and visual cortex (VC) in each hemisphere. Participant-level Auditory − Visual contrast images were generated using SPM12. In each hemisphere, a single, contiguous cluster was generated within an AC mask and VC mask from among the top and bottom 10% of contrast values, respectively, and bilateral mean AC and VC time series were extracted. Hemispheric auditory and visual thalamic search region (TSR) masks were generated for each subject. Connectivity between each voxel's time series and mean AC and VC time series were calculated, to produce TSR-AC and TSR-VC connectivity maps, which were then thresholded based on the size of the TSR. The largest contiguous cluster remaining within each hemi-thalamus TSR-AC was identified as MGN.
For each RS run of each participant, average time series were extracted from hemispheric MGN and LGN ROIs. RSFC correlations with both ipsilateral and contralateral AC were estimated for each hemispheric MGN, and then averaged across hemispheres and runs. Group differences in MGN-AC connectivity were assessed using an unequal variances t-test. Associations between MGN-AC connectivity and positive symptoms were primarily assessed across groups using a GLM with an intercept, and predictors for diagnosis (PSZ), lifetime medication exposure (dichotomous), and PANSS Positive scores. Follow-up analyses were conducted to assess the specificity of this associations by using this GLM with additional predictors for scores from the PANSS Negative and General Psychopathology Scales.
TL task data was analyzed by first identifying AC and visual cortex (VC) in each hemisphere. Participant-level Auditory − Visual contrast images were generated using SPM12. In each hemisphere, a single, contiguous cluster was generated within an AC mask and VC mask from among the top and bottom 10% of contrast values, respectively, and bilateral mean AC and VC time series were extracted. Hemispheric auditory and visual thalamic search region (TSR) masks were generated for each subject. Connectivity between each voxel's time series and mean AC and VC time series were calculated, to produce TSR-AC and TSR-VC connectivity maps, which were then thresholded based on the size of the TSR. The largest contiguous cluster remaining within each hemi-thalamus TSR-AC was identified as MGN.
For each RS run of each participant, average time series were extracted from hemispheric MGN and LGN ROIs. RSFC correlations with both ipsilateral and contralateral AC were estimated for each hemispheric MGN, and then averaged across hemispheres and runs. Group differences in MGN-AC connectivity were assessed using an unequal variances t-test. Associations between MGN-AC connectivity and positive symptoms were primarily assessed across groups using a GLM with an intercept, and predictors for diagnosis (PSZ), lifetime medication exposure (dichotomous), and PANSS Positive scores. Follow-up analyses were conducted to assess the specificity of this associations by using this GLM with additional predictors for scores from the PANSS Negative and General Psychopathology Scales.
Results:
Relative to HC, MGN-AC connectivity was significantly greater in PSZ (x̄HC=0.0539, x̄PSZ=0.0898, p=0.0332). MGN-AC connectivity was positively associated with PANSS Positive scores (β=0.0310, p=0.0229), but not diagnosis or medication exposure, across groups. When evaluated within PSZ separately, MGN-AC connectivity was associated with PANSS Positive scores (β=0.0276, p=0.0365).
In a follow-up analysis including regressors for other symptom categories using the PANSS Negative and General Psychopathology Scales in addition to diagnosis and medication exposure, MGN-AC connectivity was again associated with PANSS Positive scores (β=0.0395, p=0.0310), but no other predictors.
In a follow-up analysis including regressors for other symptom categories using the PANSS Negative and General Psychopathology Scales in addition to diagnosis and medication exposure, MGN-AC connectivity was again associated with PANSS Positive scores (β=0.0395, p=0.0310), but no other predictors.
Conclusions:
We used a dual fMRI paradigm in order to: 1) functionally localize AC and MGN, and 2) measure RSFC between these regions, in unmedicated PSZ and HC. We found that PSZ display significant hyperconnectivity between MGN and AC, and that this hyperconnectivity additionally predicts the severity of positive symptoms of psychosis dimensionally across groups.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Connectivity (eg. functional, effective, structural) 2
fMRI Connectivity and Network Modeling
Perception, Attention and Motor Behavior:
Perception: Auditory/ Vestibular
Keywords:
ADULTS
FUNCTIONAL MRI
Psychiatric
Psychiatric Disorders
Schizophrenia
Thalamus
Other - Thalamocortical
1|2Indicates the priority used for review
Provide references using author date format
Anticevic, A., M. W. Cole, G. Repovs, J. D. Murray, M. S. Brumbaugh, A. M. Winkler, A. Savic, J. H. Krystal, G. D. Pearlson and D. C. Glahn (2014). "Characterizing thalamo-cortical disturbances in schizophrenia and bipolar illness." Cereb Cortex 24(12): 3116-3130.
Giraldo-Chica, M. and N. D. Woodward (2017). "Review of thalamocortical resting-state fMRI studies in schizophrenia." Schizophr Res 180: 58-63.
Fischl, B. (2012). "FreeSurfer." Neuroimage 62(2): 774-781.
Friston, K. J., J. Ashburner, S. Kiebel, T. Nichols and W. D. Penny (2007). Statistical parametric mapping : the analysis of funtional brain images. Amsterdam ; Boston, Elsevier/Academic Press,: 1 online resource (vii, 647 pages).
Glasser, M. F., S. N. Sotiropoulos, J. A. Wilson, T. S. Coalson, B. Fischl, J. L. Andersson, J. Xu, S. Jbabdi, M. Webster, J. R. Polimeni, D. C. Van Essen, M. Jenkinson and W. U.-M. H. Consortium (2013). "The minimal preprocessing pipelines for the Human Connectome Project." Neuroimage 80: 105-124.
Iglesias, J. E., R. Insausti, G. Lerma-Usabiaga, M. Bocchetta, K. Van Leemput, D. N. Greve, A. van der Kouwe, I. Alzheimer's Disease Neuroimaging, B. Fischl, C. Caballero-Gaudes and P. M. Paz-Alonso (2018). "A probabilistic atlas of the human thalamic nuclei combining ex vivo MRI and histology." Neuroimage 183: 314-326.
Jiang, F., G. C. Stecker and I. Fine (2013). "Functional localization of the auditory thalamus in individual human subjects." Neuroimage 78: 295-304.
Kastner, S., D. H. O'Connor, M. M. Fukui, H. M. Fehd, U. Herwig and M. A. Pinsk (2004). "Functional imaging of the human lateral geniculate nucleus and pulvinar." J Neurophysiol 91(1): 438-448.
Kay, S. R., L. A. Opler and J. P. Lindenmayer (1989). "The Positive and Negative Syndrome Scale (PANSS): rationale and standardisation." Br J Psychiatry Suppl(7): 59-67.