How do morphological alterations caused by chronic pain distribute across the brain? A meta-analytic co-alteration study

Abstract

HighlightsIn chronic pain, gray matter (GM) alterations are not distributed randomly across the brain.The pattern of co-alterations resembles that of brain connectivity.The alterations' distribution partly rely on the pathways of functional connectivity.This method allows us to identify tendencies in the distribution of GM co-alteration related to chronic pain.

Addressing reverse inference in structural brain alterations

“…We run a simulation to understand the capacity of BF to highlight the earliest areas to be altered. As already showed (18,24,(62)(63)(64)(65)(66), neuropathological alterations are supposed to be distributed across the brain following structural and functional connectivity pathways. In order to simulate the alteration spread related to a certain pathology we used the anatomical connectivity matrix derived from Hagmann, Cammoun (67).…”

“…Finally, our simulations of alteration spreads related to different pathologies are based on the premise that alterations move diffusively along brain connectivity pathways. Although this underlying mechanism has been confirmed by recent research (18,24,62,63,65,66,71,109,110), it is not the only one that might play a role in the alteration spread. Moreover, the contributions of different mechanisms can vary with regard to the type of pathology affecting the brain, so that our simulations, even though they offer in our view the best approximation to real pathological spreads with the available data, do not pretend to grasp all the complexities of the actual phenomenon.…”

Coordinate based meta-analysis: new clustering algorithm, and inclusion of region of interest studies

“…Cluster-wise analysis also makes it possible to use cooccurrence analysis, with a hypothesis that the multiple reported peak coordinates form a consistent pattern of activation or grey matter alteration by testing a null hypothesis of no cooccurrence between different brain regions; there is also a voxelwise cooccurrence algorithm but it tests a different hypothesis of spatially random coordinates (Chu et al, 2015). Some algorithms achieve this using just coordinates (no Z scores) by forming clusters from reported coordinates within a fixed spherical region placed at locations identified using the ALE algorithm Tatu et al, 2018), or within ALE clusters themselves Neumann et al, 2005). There are problems with this approach in that: cooccurrence and CBMA analyses can involve different regions since they test different hypotheses (see table 1) so some regions might be missed, and selecting nodes based on the highest ALE preselects the nodes of highest occurrence, which is not independent of cooccurrence and might be considered 'double-dipping' (Kriegeskorte et al, 2009).…”

“…For CBMAN a map of network like connections, and the standardised Z score effect size correlation inspected (figure (6)). Network analysis has also been used on such results Lancaster et al, 2005;Neumann et al, 2005;Tatu et al, 2018), offering alternatives to CBMA.…”

“…For CBMAN a map of network like connections, and the standardised Z score effect size correlation inspected (figure (6)). Network analysis has also been used on such results Lancaster et al, 2005;Neumann et al, 2005;Tatu et al, 2018), offering alternatives to CBMA. (Csardi and Nepusz, 2006), while bottom shows clusters placed such that the x and y locations are the Talairach x and y locations scaled such that the radius depends on the z Talairach coordinate (z=0 is indicated by the circle), which is generated by CBMAN.…”