Testing the expanded continuum hypothesis of schizophrenia and bipolar disorder. Neural and psychological evidence for shared and distinct mechanisms

Abstract

Despite the traditional view of Schizophrenia (SZ) and Bipolar disorder (BD) as separate diagnostic categories, the validity of such a categorical approach is challenging. In recent years, the hypothesis of a continuum between Schizophrenia (SZ) and Bipolar disorder (BD), postulating a common pathophysiologic mechanism, has been proposed. Although appealing, this unifying hypothesis may be too simplistic when looking at cognitive and affective differences these patients display. In this paper, we aim to test an expanded version of the continuum hypothesis according to which the continuum extends over three clusters: the psychotic, the cognitive, and the affective. We applied an innovative approach known as Source-based Morphometry (SBM) to the structural images of 46 individuals diagnosed with SZ, 46 with BD and 66 healthy controls (HC). We also analyzed the psychological profiles of the three groups using cognitive, affective, and clinical tests. At a neural level, we found evidence for a shared psychotic core in a distributed network involving portions of the medial parietal and temporo-occipital areas, as well as parts of the cerebellum and the middle frontal gyrus. We also found evidence of a cognitive core more compromised in SZ, including alterations in a fronto-parietal circuit, and mild evidence of an affective core more compromised in BD, including portions of the temporal and occipital lobes, cerebellum, and frontal gyrus. Such differences were confirmed by the psychological profiles, with SZ patients more impaired in cognitive tests, while BD in affective ones. On the bases of these results we put forward an expanded view of the continuum hypothesis, according to which a common psychotic core exists between SZ and BD patients complemented by two separate cognitive and affective cores that are both impaired in the two patients' groups, although to different degrees.

Keywords: Bipolar disorder; Continuum hypothesis; Psychosis; Schizophrenia; Source-based morphometry.

Fig. 1

Neuroanatomical profile. A) Results of the first analysis (HC-SZ-BD). IC 18 resulted significantly different between HC, SZ and BD (p = .0145). In particular, it showed a difference between HC and SZ (p = .019), between HC and BD (p = .011), but no difference between SZ and BD (p = .832), indicating a common altered network characterizing both SZ and BD patients. On the right, graphical representation of the loading coefficients' means for healthy controls (HC), schizophrenia (SZ) and bipolar (BD) patients, meaning how IC18 is expressed in each group. B) Results of the second analysis (SZ-BD). IC6 (p < .001) resulted significantly different between SZ and BD; On the right, the histogram shows how IC6 is expressed in each group through the loading coefficients, showing that this network is reduced in SZ.

Fig. 2

Psychological profile. A) Manova results. Each point represents the values of the first two canonical variables (c1 and c2) of each participant (HC in blue, SZ in red and BD in yellow). It is evident that c1 separates the HC group from SZ and BD, while c2 separates the two clinical groups. B) Cognitive and Affective Scores. Each point represents the mean values for each participant (HC in blue, SZ in red and BD in yellow) of the standardized scores of some cognitive and affective scores. It is evident that both dimensions rely on a continuum: HC < BD < SZ when considering the cognitive alteration, and HC < SZ < BD when considering the affective alteration.

Fig. 3

The expanded continuum model. A) Neuronatomical evidence. The “P” (Psychotic Core) represented by IC18, the common altered network between SZ and BD when compared to HC. We suggest that this network could be responsible of a common shared psychotic functioning, probably involving an altered information processing and elaboration (Fig. 1A). The “C” (Cognitive Core) could be represented by IC6, the reduced network characterizing SZ when compared to BD (Fig. 1B). The “A” (Affective Core) could be represented by IC3, IC10, IC20 (Supplementary Materials). B) Psychological evidence. The “P” represents a common latent variable that emerged in the MANOVA analysis (c1), which seems to separate HC on the one side and SZ and BD on the other side (see Fig. 2A). The “C” is more impaired in SZ (Fig. 2B), and includes working memory, executive, reasoning, and vocabulary tasks. The “A” Core is more impaired in BD (Fig. 2B), and includes different impulsivity, temperament and character measures and bipolar symptoms.