What we learn about bipolar disorder from large-scale neuroimaging: Findings and future directions from the ENIGMA Bipolar Disorder Working Group

Abstract

MRI-derived brain measures offer a link between genes, the environment and behavior and have been widely studied in bipolar disorder (BD). However, many neuroimaging studies of BD have been underpowered, leading to varied results and uncertainty regarding effects. The Enhancing Neuro Imaging Genetics through Meta-Analysis (ENIGMA) Bipolar Disorder Working Group was formed in 2012 to empower discoveries, generate consensus findings and inform future hypothesis-driven studies of BD. Through this effort, over 150 researchers from 20 countries and 55 institutions pool data and resources to produce the largest neuroimaging studies of BD ever conducted. The ENIGMA Bipolar Disorder Working Group applies standardized processing and analysis techniques to empower large-scale meta- and mega-analyses of multimodal brain MRI and improve the replicability of studies relating brain variation to clinical and genetic data. Initial BD Working Group studies reveal widespread patterns of lower cortical thickness, subcortical volume and disrupted white matter integrity associated with BD. Findings also include mapping brain alterations of common medications like lithium, symptom patterns and clinical risk profiles and have provided further insights into the pathophysiological mechanisms of BD. Here we discuss key findings from the BD working group, its ongoing projects and future directions for large-scale, collaborative studies of mental illness.

Keywords: ENIGMA; MRI; bipolar disorder; cortical surface area; cortical thickness; mega-analysis; meta-analysis; neuroimaging; psychiatry; volume.

FIGURE 1

Major challenges facing neuroimaging studies of BD and how the ENIGMA BD Working Group meets these challenges

FIGURE 2

(a) ENIGMA Bipolar Disorder Working Group sites across the world including over 150 researchers from 20 countries and 55 institutions. (b) Schematic of ENIGMA Bipolar Disorder Working Group as it fits into the larger ENIGMA Consortium network. rsfMRI, resting‐state functional MRI; tbfMRI, task‐based functional MRI; WM, white matter; DTI, diffusion tensor imaging; MDD, major depressive disorder; PTSD, post‐traumatic stress disorder; OCD, obsessive–compulsive disorder; CNVs, copy number variants; Familial Risk, relatives of individuals with psychiatric illness (including bipolar disorder and schizophrenia)

FIGURE 3

(a) Outline of the ENIGMA BD Working Group guiding principles. (b) Flow diagram showing working group logistics including memorandum of understanding, participation in and development of new research proposals, data sharing, etc. Ethics/IRB: The ENIGMA BD Working group is experienced with navigating international research ethics and institutional review boards, which may require additional approval depending on project specifics. More information on the ENIGMA BD Working group including the Memorandum of Understanding can be found online (http://enigma.ini.usc.edu/ongoing/enigma‐bipolar‐working‐group/)

FIGURE 4

Findings from Subcortical volumetric abnormalities in bipolar disorder (Hibar et al., 2016). (a) Cohen's d effect size estimates for subcortical differences between individuals with BD versus healthy controls (HC) using ENIGMA‐standardized FreeSurfer volumes. Statistical model accounts for age, sex, and intracranial volume. Error bars indicate mean effect size ± standard error of the mean. Results passing study‐wide significance threshold are indicated by (*) including the amygdala which showed a trending effect. (b) Forest plots displaying the effect size estimates (adjusted Cohen's d) for each of the 20 study sites in the comparison of individuals with BD versus HC at each subcortical structure along with the overall inverse variance‐weighted random‐effects meta‐analysis results (RE Model)

FIGURE 5

Findings from Cortical abnormalities in bipolar disorder: an MRI analysis of 6,503 individuals from the ENIGMA Bipolar Disorder Working Group (Hibar et al., 2018). (a) A widespread pattern of thinner cortex in adult individuals with BD versus HC. Cohen's d effect sizes plotted in regions passing correction for multiple comparisons. (b) Thicker cortex in adult individuals with BD taking lithium medication at time of scan. (c) Thinner cortex in adult individuals with BD associated with anticonvulsant treatment at time of scan

FIGURE 6

Findings from Using structural MRI to identify bipolar disorders ‐ 13 site machine learning study in 3020 individuals from the ENIGMA Bipolar Disorders Working Group (Nunes et al., 2018). (a) Support vector machine (SVM) classifier performance trained on each site independently, including mean and 95% confidence intervals for accuracy, area under the receiver operating curve (ROC‐AUC), sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). (b) Receiver operating curves from aggregate individual‐level analysis with dashed line indicating chance performance, blue line indicating mean ROC and gray lines indicating ROC curves from individual folds

FIGURE 7

Findings from Widespread white matter microstructural abnormalities in bipolar disorder: evidence from mega‐ and meta‐analyses across 3,033 individuals (Favre et al., 2019). Mega‐analysis fractional anisotropy (FA) differences between BD and HC across 43 white matter (WM) tracts and the whole‐brain skeleton with R squared effect sizes and confidence intervals ranked by increasing order of magnitude for the regions showing significant group differences. R, right; .L, left; CC, corpus callosum; BCC, body of the corpus callosum; GCC, genu of the corpus callosum; CGC, cingulum; SCC, splenium of corpus callosum; FX, fornix; PTR, posterior thalamic radiation; EC, external capsule; ACR, anterior corona radiata; SLF, superior longitudinal fasciculus; UNC, uncinate fasciculus; CR, corona radiata; SS, sagittal stratum; IFO, inferior fronto‐occipital fasciculus, SFO, superior fronto‐occipital fasciculus; Average FA, average FA across full skeleton; PCR, posterior corona radiata; ALIC, anterior limb of the internal capsule; FXST, fornix (cres) / stria terminalis

FIGURE 8

Findings from The association between familial risk and brain abnormalities is disease‐specific: an ENIGMA–Relatives study of schizophrenia and bipolar disorder (de Zwarte et al., 2019). Top: Cohen's d effect sizes comparing BD and SCZ relatives and healthy controls across global brain measures. Bottom: global effect sizes adjusted for total intracranial volume (ICV). *Nominally significant (p < .05 uncorrected); **q < .05 corrected for multiple comparisons

FIGURE 9

Cortical thickness differences across ENIGMA working groups. Cohen's d effect sizes comparing cases versus healthy controls (HC) plotted across 34 bilateral cortical ROIs from ENIGMA‐standardized FreeSurfer protocol (http://enigma.ini.usc.edu/protocols/). Warmer colors indicate lower thickness in cases/patients, whereas cooler colors indicate greater thickness in cases/patients versus HC. Results derived from published ENIGMA studies: bipolar disorder (N = 4,419, 28 sites, Hibar et al., 2018), major depressive disorder (N = 10,105, 15 sites, Schmaal et al., 2017), schizophrenia (N = 9,572, 39 sites, van Erp et al., 2018), attention deficit hyperactivity disorder (ADHD N = 4,180, 36 sites, Hoogman et al., 2019), obsessive–compulsive disorder (OCD N = 3,665, 27 sites, Boedhoe et al., 2018) and autism spectrum disorder (ASD N = 3,222, 49 sites, van Rooij et al., 2018)