Neuroimaging Modalities

Brain Structure

Main measures of brain structure include white matter, gray matter, and cerebrospinal fluid volumes, as well as cortical thickness (see Figure 7–1A, B, and C). These measures are usually computed by analyzing T1-weighted images obtained with the magnetic resonance imaging (MRI) technique. MRI uses magnetic fields to polarize hydrogen ions in water throughout the tissues of the brain, which in turn can be used to quantify and visualize a variety of brain structural characteristics. Depending on a study’s aims and hypotheses, researchers perform analyses in the whole brain or in specific regions of interest. Changes in brain structure occur during normal aging. Aberrant changes in brain structure may indicate brain atrophy or other pathological conditions. For example, cortical thinning is related to the progression of the brain atrophy (Fischl and Dale 2000) and significantly correlated with declines in cognitive abilities and general intelligence (e.g., Menary et al. 2013). Before volume loss becomes detectable, myelin and axonal abnormalities may be detected using magnetization transfer imaging (Filippi et al. 1995; Silver et al. 1997). The decreases in the magnetization transfer ratio may reflect a reduction in size and number of neurons and a reduction in dendritic density (Bruno et al. 2006). The presence of lesions in the brain also indicates abnormality in brain structure. White matter hyperintensities are white matter lesions that appear on a T2-weighted MRI image in a bright white color. The presence of such lesions may be an important indicator of decline in cognitive function (Debette and Markus 2010).

Key Brain Structures Affected in Bipolar Disorder

Below we describe proposed functions of several brain regions and white matter tracts relevant to bipolar disorder. Note that the list is not exhaustive. The references we provide are just examples of corresponding neuroimaging work that can serve as a good source of information for an interested reader.

• Dorsolateral and ventrolateral prefrontal cortices. These regions play roles in working memory, episodic memory, attention, regulation of behavior, and control of responses (Wood and Grafman 2003).
  
• Ventromedial prefrontal cortex. A key structure for emotion regulation, decision making, and evaluation of future consequences of behavior (Bechara et al. 2000).
  
• Ventral striatum (Figure 7–2A1). A region that supports reward anticipation, reward prediction, and prediction error (Haber and Knutson 2010).
  
• Amygdala (Figure 7–2A2). A region in the medial temporal lobe that supports automatic processing of emotional and social stimuli (LeDoux 1996), and emotional learning and memory (Cahill et al. 1995).
  
• Hippocampus (Figure 7–2A3). A region in the medial temporal lobe that is critical for episodic memory that integrates information into relational associations (Moscovitch et al. 2016).
  
• Superior longitudinal fasciculus (Figure 7–2B1). A tract that connects frontal cortex to parietal, temporal, and occipital cortices (Jellison et al. 2004).
  
• Uncinate fasciculus (Figure 7–2B2). A tract that connects orbital and inferior frontal cortices to the anterior temporal cortex (Jellison et al. 2004).
  
• Inferior longitudinal fasciculus (Figure 7–2B3). A tract that connects temporal and occipital cortices (Jellison et al. 2004).

Correlation Between Brain Measures and Clinical Characteristics in Individuals With Bipolar II Disorder

Understanding brain structural and functional changes in individuals with BD II relative to HC subjects and individuals with other mood disorders deepens our knowledge about the neurobiological underpinnings of BD II. Understanding how these changes are related to current or lifetime symptoms of depression and mania as well as other clinical and cognitive characteristics may provide us with potential therapeutic targets to address the individual needs of patients with BD II.

We reviewed all available neuroimaging literature related to BD II and found that the number of published studies was extremely small even when all neuroimaging modalities were taken into account (Figure 7–3), the individual samples were small and often included patients across mood states, and the statistical methods of data analyses were not always state-of-the-art. Taken together, this did not allow us to draw firm conclusions about the neural underpinnings of BD II. However, some general themes emerge around possible neural processes that may explain some of the clinical characteristics of BD II.

Impulsivity, sensation seeking, and risk taking characterizing BD II could be related to dysfunctional ventromedial cortex and ventral striatum. Ventromedial cortex plays a role in emotion regulation and decision making. People with damage to this region guide their behavior based on immediate but not future positive or negative prospects. Bechara et al. (2000) coined such behavior as “myopia for the future.” The finding of reduced volumes of ventromedial prefrontal cortex in individuals with BD II who experienced more mood episodes during their lifetime and rapid cycling (Narita et al. 2011) can be one explanation for impulsive risk-taking behaviors in BD II. The ventral striatum supports reward anticipation, reward prediction, and prediction error (see, e.g., Haber and Knutson 2010; Knutson et al. 2001; Schultz et al. 2000). Increased activation in ventral striatum in individuals with BD II relative to HC subjects and relative to individuals with BD I during reward anticipation, but decreased activation in the same region during loss anticipation, suggests that individuals with BD II may disregard potential negative consequences from risky behaviors.

Efficient inhibitory control and executive function require adequate involvement of prefrontal cortices and communication of prefrontal regulatory signal to other cortical and subcortical brain structures. Reduced white matter integrity in long-range, interhemispheric, and limbic tracts, and altered brain activation during task difficulty increase and response inhibition, make functioning of this pathway suboptimal, and decrease the ability to downregulate emotions and impulsive behaviors in BD II.

The studies that included BD II, BD I, and HC subjects often reported that the same brain abnormalities characterized both bipolar disorder subtypes. Thus, individuals with either BD I or BD II, compared with HC subjects, had significant reductions in gray matter volume in medial and lateral prefrontal cortex, amygdala, hippocampus, and thalamus (Ha et al. 2009; Hibar et al. 2016; Janiri et al. 2017), and aberrant white matter integrity in the internal capsule that carries information among striatal regions and inferior longitudinal fasciculus that connects occipital and temporal cortices (Ambrosi et al. 2016). Surprisingly, GABA levels (Atagün et al. 2017) and serotonin transporter availability (Chou et al. 2010, 2013) did not differ for individuals with BD II and HC subjects.

The differences between brain structural, functional, and metabolic abnormalities for BD II individuals relative to BD I individuals may be more subtle than those between BD II and HC subjects because of the overall similarities in mood symptoms. Studies comparing various neuroimaging measures in individuals with BD II and BD I revealed that the latter had reduced gray matter volumes in temporal, posterior cingulate, and frontal regions (Ha et al. 2009; Maller et al. 2014), reduced white matter integrity in the right inferior longitudinal fasciculus (Ha et al. 2011) and uncinate fasciculus (Caseras et al. 2015), increased number of white matter hyperintensities that may reflect a decrease in the white matter density (Altshuler et al. 1995; Kieseppä et al. 2014); increased dorsolateral activation during working memory tasks (Dell’Osso et al. 2015), including those with happy distractors (Caseras et al. 2013); and decreased ventral striatal activation during reward anticipation, but increased during reward outcome processing (Caseras et al. 2013). Taken together with altered patterns of glucose metabolism in prefrontal cortical, temporal, parietal, and striatal regions (Ketter et al. 2001; Li et al. 2012), these findings reflect more severe neural pathology in BD I than in BD II across multiple brain regions.

The inferior longitudinal fasciculus connects temporal and occipital lobes, so damage to this structure may affect visual perception and recognition in both bipolar disorder types. The exact mechanism, as well as the way it may mediate potential problems with processing of visual stimuli, has yet to be identified. Yip and colleagues (2013) proposed that white matter microstructures—but not gray matter macrostructures—can be affected in BD II at an early age and stage, and before clinically significant symptoms manifest. They also proposed that white matter development may be a candidate target for understanding genetic and environmental antecedents to bipolar disorder, and mood disorders more broadly. In addition to the disrupted integrity of the commissural and projection fibers, the pathophysiology of BD II may be related to an involvement of the limbic association fibers (Ha et al. 2011; Kurumaji et al. 2017). Relative sparing of the dorsal system and long association fibers may differentiate BD II from BD I.

Trait markers of BD II that probably developed over the course of the disorder and were not related to mood state are brain structural abnormalities, aberrant functional connectivity in medial superior frontal gyrus during task performance (Marchand et al. 2014), the default mode network, and limbic system at rest (Wang et al. 2016, 2017a), as well as disrupted interhemispheric coordination (Wang et al. 2015a, 2015b). In young adults with BD II, increased connectivity in temporo-insular regions at rest may be a candidate vulnerability marker for bipolar disorder (Yip et al. 2014).

Only a small number of studies examined the effect of mood state in BD II. These studies found that BD II depression may be linked to widespread functional abnormalities in cortical midline, frontal, limbic, and parietal circuitry supporting emotional face processing (Vizueta et al. 2012) and posterior midline abnormalities (Caseras et al. 2015). Correlation analyses indicated that although current depression and mania symptoms were not related to brain volume measures in individuals with BD II, greater lifetime number of mood episodes was related to reduced gray matter volume in ventromedial prefrontal cortex (Narita et al. 2011) involved in decision making (Bechara et al. 2000). Individuals with higher depression scores had lower absolute prefrontal and paralimbic cortical but higher anterior paralimbic and thalamic glucose metabolism (Ketter et al. 2001; Marchand et al. 2011b), as well as greater brain activation in cingulate regions, precuneus, and paracentral lobule in response to fearful and happy faces (Marchand et al. 2011a).

Cognitive difficulties such as decreased ability to memorize information or concentrate on task are common complaints of individuals with all bipolar disorder types. Some of these difficulties in BD II may be related to aberrant structure and white matter integrity in temporal and anterior cingulate cortices, which have been associated with executive functioning and IQ changes (Bruno et al. 2006; Liu et al. 2010). Interestingly, contrary to other researchers, Bruno and colleagues (2006) suggested that BD II has more extensive structural abnormalities than BD I, which could be related to persistent depression, rather than mania, and may be a key pathophysiological factor associated with increased risk of developing cognitive abnormalities.

Given that BD II is often misdiagnosed as MDD, there is a need for more studies that elucidate the differences in brain structure, function, and metabolism in BD II relative to MDD. On the basis of previous longitudinal clinical research, it has been shown that between 7.5% (Coryell et al. 1995) and 10.5% (Alloy et al. 2012) of individuals with BD II convert to BD I within 4.5–10 years. It is important therefore to understand underlying changes in brain structure, function, and metabolism that occur during conversion to identify ways to prevent these changes. This further emphasizes the need for longitudinal studies that can relate longitudinal changes in brain structure, function, and metabolism measures to changes in clinical symptoms. So far, longitudinal studies of BD II progression to BD I have not included any neuroimaging components. The relationship between mood changes over time and corresponding brain response, as well as the developmental trajectories in individuals with BD II, are also unknown because of a lack of longitudinal studies.

Here we attempt to make several predictions about brain changes that occur during conversion of BD II to BD I based on the cross-sectional results described above. Although it is tempting to propose that BD II is qualitatively different from BD I in abnormal patterns of brain structure, function, and metabolism, current research does not provide sufficient evidence for this assertion. Most studies show that BD II is quantitatively different from BD I, suggesting further deterioration of already aberrant brain measures. Specifically, the conversion from BD II to BD I may be related to further reduction in gray matter volume and cortical thickness in frontal, temporal, and parietal lobules; increase in the number and volume of white matter lesions; and decrease in white matter integrity in the inferior longitudinal, superior longitudinal, and uncinate fasciculi. During BD II to BD I conversion, significant deterioration in normal functioning and connectivity, as well as altered glucose metabolism, may be observed in prefrontal cortex, striatal regions, amygdala, and hippocampus. Given that illness duration per se did not correlate with the measures of brain structure and function in BD II, it is possible that other factors (e.g., childhood trauma, life stress, frequency of acute mood episodes) affect brain deterioration, leading to the increased probability of BD II to BD I conversion. There were a very limited number of studies that examined the effect of trauma on brain structure and function in BD II. Meanwhile, trauma and life stress may be important factors in developing symptoms such as instability, irritability, and depression in mood disorders generally and BD II specifically. It remains unclear whether conversion from BD II to BD I can be prevented by targeting the key brain regions and white matter tracts described above by using emotion and cognitive training techniques, psychotherapy, or other methods.

Another important question concerns the changes in brain structure and function related to the use of psychotropic medications. Although limited evidence suggests that the use of medication may normalize the size of some brain structures (Elvsåshagen et al. 2013a; Hibar et al. 2016), more studies are needed to understand the relationship among specific psychotropic drugs, their dosage, the changes in the brain structure, and the timeframe for such changes.

KEY POINTS

• There are relatively few neuroimaging studies of bipolar II disorder (BD II).
  
• Firm conclusions cannot be drawn about neural markers of BD II because of the small and diverse samples in available studies.
  
• Relative to healthy adults, individuals with bipolar I disorder (BD I) and individuals with BD II show reduced gray matter volume along with aberrant brain activation and glucose metabolism in regions responsible for reward anticipation, reward processing, executive function, and cognitive control.
  
• Less severe presentation of manic symptomatology in BD II versus BD I may be explained by less prominent abnormalities in white and gray matter, and functioning of frontal cortical and striatal brain regions.
  
• Ventral striatal activation during reward anticipation and processing distinguishes BD II from BD I.
  
• Longitudinal neuroimaging studies of BD II are needed to identify state and trait neural markers of the disorder.

Bechara A, Tranel D, Damasio H: Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 123(Pt 11):2189–2202, 2000 11050020

Brooks JO 3rd, Vizueta N, Penfold C, et al: Prefrontal hypoactivation during working memory in bipolar II depression. Psychol Med 45(8):1731–1740, 2015 25752642

Bruno SD, Papadopoulou K, Cercignani M, et al: Structural brain correlates of IQ changes in bipolar disorder. Psychol Med 36(5):609–618, 2006 16469198

Cahill L, Babinsky R, Markowitsch HJ, et al: The amygdala and emotional memory. Nature 377(6547):295–296, 1995 7566084

Caseras X, Lawrence NS, Murphy K, et al: Ventral striatum activity in response to reward: differences between bipolar I and II disorders. Am J Psychiatry 170(5):533–541, 2013 23558337

Caseras X, Murphy K, Lawrence NS, et al: Emotion regulation deficits in euthymic bipolar I versus bipolar II disorder: a functional and diffusion-tensor imaging study. Bipolar Disord 17(5):461–470, 2015 25771686

Chou Y-H, Wang S-J, Lin C-L, et al: Decreased brain serotonin transporter binding in the euthymic state of bipolar I but not bipolar II disorder: a SPECT study. Bipolar Disord 12(3):312–318, 2010 20565438

Chou Y-H, Lin C-L, Wang S-J, et al: Aggression in bipolar II disorder and its relation to the serotonin transporter. J Affect Disord 147(1–3):59–63, 2013 23123132

Coryell W, Endicott J, Maser JD, et al: The likelihood of recurrence in bipolar affective disorder: the importance of episode recency. J Affect Disord 33(3):201–206, 1995 7790673

Cui X, Bray S, Bryant DM, et al: A quantitative comparison of NIRS and fMRI across multiple cognitive tasks. Neuroimage 54(4):2808–2821, 2011 21047559

Debette S, Markus HS: The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ 341:c3666, 2010 20660506

Dell’Osso B, Cinnante C, Di Giorgio A, et al: Altered prefrontal cortex activity during working memory task in bipolar disorder: a functional magnetic resonance imaging study in euthymic bipolar I and II patients. J Affect Disord 184:116–122, 2015 26074021

Elvsåshagen T, Westlye LT, Bøen E, et al: Evidence for reduced dentate gyrus and fimbria volume in bipolar II disorder. Bipolar Disord 15(2):167–176, 2013a 23317454

Elvsåshagen T, Westlye LT, Bøen E, et al: Bipolar II disorder is associated with thinning of prefrontal and temporal cortices involved in affect regulation. Bipolar Disord 15(8):855–864, 2013b 23980618

Filippi M, Campi A, Dousset V, et al: A magnetization transfer imaging study of normal-appearing white matter in multiple sclerosis. Neurology 45(3 Pt 1):478–482, 1995 7898700

Fischl B, Dale AM: Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc Natl Acad Sci USA 97(20):11050–11055, 2000 10984517

George MS, Ketter TA, Post RM: SPECT and PET imaging in mood disorders. J Clin Psychiatry 54(Suppl):6–13, 1993 8270597

Gujar SK, Maheshwari S, Björkman-Burtscher I, et al: Magnetic resonance spectroscopy. J Neuroophthalmol 25(3):217–226, 2005 16148633