Mononeuritis multiplex as clinical presentation of systemic lupus erythematosus.

Attention‐deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder with a worldwide prevalence of approximately 7.2% in children and 2.5% in adults []. Despite its prevalence, the neuroanatomical basis of ADHD remains less clear [] than that of other neuropsychiatric conditions such as depression [, ] or schizophrenia []. Many studies have identified brain regions with decreased brain volume or cortical thickness in ADHD compared with typically developing cohorts, including the anterior cingulate cortex [], supplementary motor cortex [], prefrontal cortex [, ], regions in the basal ganglia [], amygdala [], hippocampus [], corpus callosum [], thalamus [, ], and cerebellum []. However, these regions are inconsistently reported across studies and have not been demonstrated to be specific to ADHD. Several meta‐analyses have sought to identify consistent results from this disparate literature with minimal success. Most employ a neuroimaging meta‐analytic technique called activation or anatomic likelihood estimate (ALE), which examines the spatial convergence of coordinates identified from individual clusters/regions []. The largest meta‐analysis to date used ALE to analyze data from 96 structural ( = 22) and functional ( = 74) magnetic resonance imaging (MRI) studies on ADHD []. Using rigid guidelines, this study found no structural differences and a single task‐based activation cluster in the left pallidum/putamen in participants with ADHD ( = 68) versus controls. While specific alteration of the putamen may be a factor in ADHD, the heterogeneity of ADHD presentation and the literature to date implies likely involvement of a more complex set of regions. Interestingly, in prior studies of ADHD, the putamen has also been found to be anti‐correlated with the default mode network (DMN) [], which has exhibited atypical connectivity in this population []. These findings raise the possibility that connectivity patterns involving the putamen and/or the DMN are a convergent finding that may provide a more complete understanding of ADHD‐related brain alterations. Many of these prior studies and meta‐analyses have primarily focused on identifying consistent localized differences in brain activity or volume between ADHD and controls, based on the hypothesis that symptoms of ADHD originate from specific regions. However, an alternative hypothesis is that similar symptoms may arise from disruptions across distributed networks rather than isolated locations, for which a traditional mass univariate approach is not ideal. Therefore, in light of the inconsistent localization of specific regions in the ADHD literature, adopting a network‐based approach is a reasonable and rational “next step” approach to identify convergence across the literature. One such method is a meta‐analytic technique called coordinate network mapping (CNM) []. CNM is derived from lesion network mapping, which uses resting‐state functional connectivity to identify networks associated with symptoms linked to specific brain lesions [, , , , , , , , , ]. Although lesions may occur in separate brain regions across patients with similar symptoms, they often localize to a specific network. CNM generalizes this method to identify common networks that are functionally connected to a reported set of representing differences in volume or blood oxygenation–level dependent (BOLD) activity across studies. Unlike ALE analyses, which focus on spatial convergence at individual locations, CNM evaluates spatial convergence at the network level, that is, how functionally connected networks might underpin a disorder, even when specific regions are inconsistently implicated across specific studies. For disorders like ADHD, differences in methodology and task paradigms can make comparing task‐based fMRI studies difficult. However, voxel‐based morphometry studies, which measure brain volume and cortical thickness, are well suited for meta‐analyses like CNM. Importantly, control analyses, such as in non‐ADHD cohorts, are critical to ensuring that the observed connectivity patterns are specific to ADHD and do not represent general features of brain connectivity or artifacts of the methodology. Also of note, we use the term “decreased brain volume” as we cannot distinguish regions that are underdeveloped versus those that were of normal size at some point and then atrophied over time, either prior or subsequent to symptom onset.