Abstract
We appreciate DeSouza et al.’s interest in and comments (1) on our study (2) and find the results from their recent study of great interest (3). In their study, in addition to conventional volume and thickness-based measures, a graph-theoretical approach was used to delineate differences in structural covariance networks between patients with chronic migraine (CM) and healthy controls (HC). While DeSouza et al. did not find any between-group differences using the conventional volume or thickness-based measures, we showed that cortical thickness was decreased in several pain-related cortical regions in CM patients (2). Their finding of significant differences in structural network measures leads DeSouza et al. to suggest that such measures may provide a more sensitive means to detect structural differences in CM (3). While we agree with DeSouza et al. on the importance of using structural covariance network analysis, as it provides complementary information, we want to emphasize with our study (2) that homogeneity of the patient population is of utmost importance in order to reveal subtle differences, even if those differences appear to be shared with other chronic pain conditions. Hence, if medication overuse, psychiatric comorbidities, and ongoing preventive medication are not controlled for then, irrespective of the structural measure used, it is difficult to ascribe the presence or absence of findings to CM pathophysiology alone. For example, it has previously been demonstrated by our group that medication overuse is associated with specific structural and neurochemical differences (4,5). Additionally, depression itself is known to affect brain morphology and connectivity, as revealed by graph theory-based brain network analysis (6).
As for the discrepancy between DeSouza et al.’s study (3) and ours (2), differences in the above-mentioned factors possibly account for this. Furthermore, DeSouza et al. measured the averaged cortical thickness in 68 cortical regions of interest (ROIs) (34 in each hemisphere) parcellated according to the Desikan-Killiany atlas, rather than using a vertex-based approach. The subtle differences within a sub-ROI area between groups could theoretically be averaged out by such an ROI analysis. Further studies applying all of the above-mentioned structural measures in a more homogenous patient population are warranted and may provide additional novel insights into CM-specific pathophysiology.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
