Abstract

Introduction

Cerebellar alterations, including both volumetric changes in the cerebellar vermis and dysfunctions of the corticocerebellar connections, have been documented in psychotic disorders. Starting from the clinical observation of a bipolar patient with cerebellar hypoplasia, the purpose of this review is to summarize the data in the literature about the association between hypoplasia of the cerebellar vermis and psychotic disorders [schizophrenia (SCZ) and bipolar disorder (BD)].

Methods

A bibliographic search on PubMed has been conducted, and 18 articles were finally included in the review: five used patients with BD, 12 patients with SCZ and one subject at psychotic risk.

Results

For SCZ patients and subjects at psychotic risk, the results of most of the reviewed studies seem to suggest a gray matter volume reduction coupled with an increase in white matter volumes in the cerebellar vermis, compared to healthy controls. Instead, the results of the studies on BD patients are more heterogeneous with evidence showing a reduction, no difference or even an increase in cerebellar vermis volume compared to healthy controls.

Conclusions

From the results of the reviewed studies, a possible correlation emerged between cerebellar vermis hypoplasia and psychotic disorders, especially SCZ, ultimately supporting the hypothesis of psychotic disorders as neurodevelopmental disorders.

Introduction

Neuroimaging applications offer an important tool of investigation in the research of psychiatric disorders, and, despite the limitations related to the reproducibility of techniques and samples, they are providing more and more useful insight for clinical practice (Anderson et al., 2019). So far, various alterations of the central nervous system have been described, both at a morphological and functional level, especially in the field of major psychosis (Voineskos et al., 2020). In this context, several biomarkers have been hypothesized to improve the diagnostic and prognostic definition of these patients, with promising results (Aydin et al., 2019; Keshavan et al., 2020).

At the brain level, morphological and connectivity alterations were frequently found in psychotic subjects (O’Donoghue et al., 2015). In particular, many studies suggested the presence of gray matter and white matter reductions in patients with schizophrenia (SCZ), especially in prefrontal regions, thalamus, cingulate cortex, temporal lobe, basal ganglia, amygdala and hippocampus, also coupled with total ventricular volume increase (Brambilla et al., 2013; Delvecchio et al., 2017; Aydin et al., 2019). Moreover, the ability to discriminate patients with SCZ from healthy controls (HC) was also highlighted by a meta-analysis of multivariate pattern recognition studies of neuroimaging-based diagnostic biomarkers, which found an overall sensitivity and specificity of approximately 80% (Kambeitz et al., 2015; Kraguljac et al., 2021). It is important to underline that brain alterations seem more consistent in psychotic patients with a chronic course than in those with a first psychotic episode, indicating an increase in the burden of the disease over time (Hulshoff Pol and Kahn, 2008; Kambeitz et al., 2015). Interestingly, the same alterations have been found in ultra-high-risk (UHR) subjects, therefore suggesting the ability of specific biomarkers to predict the risk of evolution to psychosis in UHR subjects (Koutsouleris et al., 2015; Del Fabro et al., 2021; Kraguljac et al., 2021). As for bipolar disorder (BD), it has been shown that patients with BD have structural alterations in several cortical regions, including prefrontal, anterior temporal and insula cortices (Ganzola and Duchesne, 2017; Dusi et al., (2019); Prunas et al., (2018); Brambilla et al., (2009); Ching et al., 2022), and subcortical areas, such as hippocampus, thalamus and amygdala (Hajek et al., 2009; Rimol et al., 2010), as well as functional connectivity deficits, especially within and between cortico-limbic structures (Cattarinussi et al., 2022), compared to HC.

Moreover, when SCZ patients and BD patients undergo a direct comparison, SCZ patients present more extensive gray matter alterations than BD patients, especially in the frontotemporal cortex, thalamus, hippocampus and amygdala (Maggioni et al., 2016). In contrast, it has been suggested that BD patients showed a greater reduction in gray matter volumes of the cerebellum compared to SCZ patients (Farrow et al., 2005; Molina et al., 2011; Maggioni et al., 2016). This evidence is not surprising, especially because the cerebellum, and in particular the cerebellar vermis, is considered an ‘emotional pacemaker’ specifically engaged in the modulation of emotional processing and therefore particularly involved in affective disorders (Heath, 1977; Minichino et al., 2014). In general, the cerebellar alterations have been extensively described to be related to the pathophysiology of other psychiatric disorders, including attention deficit hyperactivity disorder, autism spectrum disorders, SCZ, major depressive disorder and anxiety disorders (Phillips et al., 2015; Laidi et al., 2019). The cerebellum has multiple control functions, especially those related to motor coordination, cognitive abilities and emotion regulation (Koziol et al., 2014; Mariën and Borgatti, 2018). Although it is not yet certain which function is particularly involved in the pathogenesis of psychiatric disorders, it seems reasonable to support Minichino’s hypothesis, which underlines the centrality of this structure in cognitive and emotional functions, abilities that are often impaired in these conditions (Minichino et al., 2014).

The idea of this article arises from the observation of a clinical case, reported below, of a female patient with BD in which hypoplasia of the cerebellar vermis was found to be associated with functional glucose hypo-uptake. The purpose of this article is therefore to review the literature to highlight the correlations described between hypoplasia of the cerebellar vermis and psychotic disorders, in particular UHR, SCZ and BD, to allow an evaluation in diagnostic and prognostic terms.

Methods

The review was designed according to the PRISMA model for systematic reviews (Page et al., 2021). To identify potentially relevant records, a bibliographic search was conducted in the PubMed database without any prespecified restriction. To review all scientific articles exploring the association between alterations in the cerebellar vermis and psychosis, we carried out a comprehensive search, where the keywords ‘vermis’, ‘vermal’, ‘medial’, ‘paravermis’, ‘paravermal’, ‘paramedial’ were matched with the terms ‘cerebellar’, ‘cerebellum’, and with the terms ‘psychotic disorder’, ‘psychosis’, ‘schizophrenia’, and ‘bipolar disorder’.

Eligibility criteria consisted of (1) articles written in English (or with an English translation available), (2) original studies on human samples: review articles and case reports were excluded, (3) patients with a diagnosis of SCZ or BD and (4) neuroimaging studies using structural MRI. Studies using other neuroimaging methods or samples with comorbid substance abuse and patients with specific neurological or genetic syndromes were excluded to reduce the heterogeneity across the reviewed studies. Records were also considered for inclusion if quoted in the reference lists of articles screened for inclusion. All the resulting records were screened by title and abstract by CME and then full-text articles were evaluated by CME and ND. Briefly, starting from 149 articles screened, 18 articles were finally included in the review. The selection of the articles has been reported in Fig. 1.

Characteristics of the reviewed articles have been summarized in Table 1. For each article, we extracted the following variables: general information (author, year of publication), study design, sample characteristics (demographics and numerosity), diagnosis of the sample (BD, SCZ or UHR), methods (structural MRI with 1.5 Tesla or 3 Tesla), and main results.

Results

Eighteen articles were included in the review. Of these, five articles were on patients with BD, 12 on patients with SCZ and one on subjects at psychotic risk. The study designs were cross-sectional for all studies, covering a time span from 2001 to 2021. As regards those studies that assessed patients with a diagnosis of SCZ, the comparison between psychotic patients and HC was carried out on heterogeneous populations, which varied for sample size, length of illness, onset period or clinical status. Either the total brain volume or the white matter and gray matter volumes of the cerebellar vermis were investigated, with the majority of the reviewed studies showing a reduction of total and gray matter volumes as well as an increase in white matter volumes.

In particular, the study by Ichimiya et al. (2001) highlighted significantly smaller volumes of the cerebellar vermis, but not of other cerebellar structures, in 20 drug-naive SCZ patients compared to 20 HC. Moreover, a correlation between the reduction in vermal volumes and specific subscales (depression and paranoia) of the Brief Psychiatric Rating Scale was highlighted (Ichimiya et al., 2001). Similarly, Loeber et al. (2001) confirmed the reduction of total vermis volumes in 38 SCZ patients compared to 26 HC. Indeed, the authors showed abnormalities in the posterior and anterior vermis areas but not in the inferior vermian level (Loeber et al., 2001). These results were further confirmed by three subsequent studies, which reported a reduction in total volumes of the cerebellar vermis in patients with chronic SCZ (Okugawa et al., 2003) and in patients with childhood-onset SCZ (Greenstein et al., 2011) and in adolescent patients diagnosed with SCZ (Henze et al., 2011) compared to HC. Also, a study by Wang et al. (2017) showed a reduction in gray matter density in cerebellar vermis in SCZ patients with or without cognitive impairments compared to HC.

Interestingly, Joyal et al. (2004) also reported selective reductions in the inferior vermis in 19 SCZ patients compared to 19 HC whereas Segarra et al. (2008) found a reduction in the total gray matter volume of the cerebellar vermis in adult patients compared to age- and gender-matched HC, respectively. Notably, cerebellar deficits have also been observed in subjects at psychotic risk in a study carried out by (Roman-Urrestarazu et al., 2014). Indeed, the authors highlighted a cluster of lower gray matter at the cerebellar vermis level, in patients with both clinical and familiar risk of psychosis (Roman-Urrestarazu et al., 2014) . Moreover, areas of specific reduction in gray matter volumes were identified in the whole superior vermis in 13 first-episode SCZ patients compared to 13 HC (Rasser et al., 2010) and only in the posterior superior vermis in 71 SCZ patients compared to 65 HC (Laywer et al., 2006). Finally, Lee et al. (2007) highlighted an increase in cerebellar vermis white matter volume while the study by Lawyer et al. (2009) demonstrated an increased mean of gray matter/white matter ratios in cerebellar vermis in chronic SCZ patients with respect to HC.

With regard to BD patients, the results from the reviewed studies were highly heterogeneous. In particular, two studies seemed to highlight no differences between BD patients compared to HC (Brambilla et al., 2001; Monkul et al., 2008), with the former study showing, though, a reduction in total vermis volumes in family members of BD patients compared to HC whereas the latter highlighting a trend to smaller vermis V2 areas in BD patients compared to HC. Instead, two studies (Baldaçara et al., 2011; Lupo et al., 2021) found a significant reduction of cerebellar volumes in BD patients compared to HC without highlighting correlations with suicide attempts (Baldaçara et al., 2011). In contrast, the study by Womer et al. (2009) showed an increase in total vermis volumes in males but not in females affected by BD.

Case presentation

When it comes to our observation, D.S. was a 35-year-old woman with a long history of psychiatric disorders with several past pharmacologic and psychotherapeutic treatments. Substantially normal psychomotor development was described, although her parents reported that she often stumbled as a child and was clumsy in her movements. She had a regular school background until she enrolled in university. Currently, she has been stuck in her university career for over 10 years, alternating between moments of great commitment and periods of intolerance and difficulty. The psychopathologic onset seemed to be around the age of 18 when, following family grief, the patient would have experienced an episode of depressive mood, with anhedonia and an increase in anxiety and brooding. Over the years, both the patient and family members have reported frequent mood swings in a depressive and hypomanic sense. In general, depressive experiences were accompanied by abandonment of studies, an intense sense of uselessness and defeat, as well as the need for continuous reassurance from family members. In particular, toward the mother, she undertook a relationship style of important emotional dependence, which in the depressive phases seemed particularly evident. In the hypomanic phases, D.S. had the feeling of being able to study, and appeared talkative, very casual, seductive, at times incongruous in the interaction.

When she was first admitted to our psychiatry ward, D.S. was going through a depressive phase, characterized by the fear of losing her reference figures. The anguish at the thought of separation from her mother, in the hypothesis of illness and death, which was not based on a concrete risk, triggered a solid suicidal ideation in her. The depressive thought, regarding the inevitable end of everything after the death of the mother, was not subjected to criticism and took on delusional characteristics. The thought of suicide seemed to be ruminative and stereotypical, not authentically connoted of a real planning. There were also anxious symptoms, such as the constant need for reassurance from the outside, as well as a certain infantilism in the behaviors. A therapy based on valproic acid 1000 mg, olanzapine 20 mg and clomipramine 25 mg per day was settled, to help the patient managing the suicidal rumination, the depressive symptoms and the delusional ideation. Thus, we assisted in the remission of the clinical picture (brief psychiatric rating scale from 44 to 27; Hamilton Rating Scale for Depression from 15 to 4, Young Mania Rating Scale from 0 to 9) (Hamilton, 1960; Overall and Gorham, 1962; Young et al., 1978).

In the following months, we continued to witness D.S.’s mood swings, which profoundly influenced her planning skills in her study and life commitments. During the outpatient visits and in the two subsequent hospitalizations, psychometric assessment scales were carried out and the diagnosis of BD (via Structured Clinical Interview for Diagnostical and Statistical Manual of Mental Disorders—5° edition—Clinical Version) and of Dependent Personality Disorder (via Structured Clinical Interview for Diagnostical and Statistical Manual of Mental Disorders—5° edition—Personality Disorders) was formulated (APA, 2013; First et al., 2015a, 2015b). While depressive symptoms configured real depressive episodes, the counter-polar episodes seemed hypomanic. Frank delusional ideation did not manifest itself outside the acute phase of the disorder. In the context of the diagnostic assessment, neuroimaging evaluations were carried out. Brain MRI (Fig. 2) showed a large cisterna magna with hypoplasia of the inferior vermis and an accentuated cerebellar cortical pattern; ventricular size at upper limits of norm, minimal asymmetry of the lateral ventricles, midline structures in axis. On brain PET/computerized tomography, a picture of modest hypocaptation in the cerebellar vermis site was highlighted. At the specialist evaluation, there were no objective neurological alterations. A genetic evaluation was carried out in the suspicion of a syndromic picture, but no significant alterations were evidenced on the molecular cytogenetic analysis through an array of comparative genomic hybridization. On cognitive evaluations, the patient showed a normal IQ (92) (29.7% percentile), measured with the Raven’s Progressive Matrices (Bingham et al., 1966; SPM, 2008). On a brief neuropsychologic examination, the patient was found to be deficient (77/100) (5% percentile) in the global score (ENB-2) (Mondini et al., 2011), and in some cognitive domains, including abstraction, cognitive estimates, tangled figure test and clock test.

In light of the continuous fluctuations in D.S.’s mood, the poor therapeutic adherence and the ambivalent relationship with her mother, it was proposed to enter a rehabilitation residential program, which the patient accepted. After several therapeutic changes, D.S. found a partial thymic balance with the following therapy: valproic acid 1000 mg, aripiprazole 15 mg and fluvoxamine 50 mg daily. The choice to continue antipsychotic therapy is due on the one hand to the stabilization capacity of aripiprazole (Nestsiarovich et al., 2022), and on the other to the individual tendency towards clinical worsening in the face of attempts at pharmacological descalation. In recent months, she has been supported from the rehabilitation point of view with respect to the management of her daily life, seeking a mediation between excessive study commitments and surrender. Her social skills have also been strengthened to allow her a more functional social insertion. The patient was informed of the drafting of this case report and gave her consent to its publication.

Discussion

The finding of an MRI picture of hypoplasia of the cerebellar vermis in the case of D.S. provided the opportunity for a review of the updated literature on cerebellar vermis alterations in subjects with psychotic disorders.

The cerebellum is a central brain structure with multiple functions, mainly motor coordination, cognitive and emotional control (Koziol et al., 2014; Adamaszek et al., 2017; Mariën and Borgatti, 2018; Schmahmann, 2019). It is physiologically connected to the spinal cord and the cerebrum; it has a complex regional organization, where the cerebellar vermis represents the medial cerebellar cortex (D’Angelo, 2018). Recent evidence is leading us to consider the cerebellum not only from the point of view of the motor regulation function but also from the behavioral and emotional perspective. Indeed, according to the hypothesis expressed by Schmahmann (2019), the cerebellum would implement the same regulatory function on both movement and behavior, thus maintaining an implicit homeostatic balance around the baseline (Schmahmann, 2019). In fact, according to Schmahmann (2019), the cerebellum maintains behavior around a homeostatic baseline, automatically, without conscious awareness. While the anterior regions are mainly involved in sensorimotor and vestibular control, the posterior regions are particularly involved in cognitive and emotional control (D’Angelo, 2018). In fact, a lesion in the posterior cerebellar lobe can give rise to a cerebellar cognitive affective syndrome, which is characterized by disorders in language, executive functions, visual processing and emotion regulation (Argyropoulos et al., 2020). The hypothesis of the emotion regulation function of the cerebellum has also been investigated from a psychopathological point of view: Minichino et al. (2014), in a review on cerebellar alterations associated with BD, express the theory of the cerebellum as an ‘emotional pacemaker’, therefore suggesting a role in emotional processing that would make the cerebellum an essential research target in affective disorders.

Notably, the majority of the reviewed studies reported the presence of structural abnormalities in cerebellar vermis in both SCZ and BD patients, selective with respect to the entire brain, with a mixed picture of either increased or decreased gray matter volumes in these patients compared to HC, which seemed to be maintained even in different types of patients (childhood-onset SCZ, adolescent or adult SCZ patients) and at a different stage of the course of illness (chronic psychosis, first-episode psychosis). Overall, these results would suggest an involvement of cerebellar alterations in psychotic disorders, as it is already documented for neurodevelopmental disorders. In fact, early cerebellar damage seems to impact the long-term outcome more than the late one, suggesting a function of the cerebellum in brain development (Stoodley, 2016; Sathyanesan et al., 2019). In line with this hypothesis is the evidence reporting the involvement of the cerebellum in the pathogenesis of autism spectrum disorders, attention-deficit hyperactivity disorder and psychotic disorders, which indeed further suggests the putative link between brain development and the emergence of these disorders (Phillips et al., 2015).

Many studies have demonstrated that patients affected by SCZ or BD show neurodevelopmental abnormalities, which seem to accumulate from the prenatal period to adolescence, constituting a biological vulnerability to developing the disorders (Rund, (2018)). Therefore, since cerebellar abnormalities are often related to neurodevelopmental problems, their presence in psychotic patients seems to confirm the hypothesis of SCZ as a neurodevelopmental disorder.

Furthermore, in recent years, many theories have proposed the key role of the cerebellum in emotion regulation (i.e. emotional pacemaker theory) (Heath, 1977; Minichino et al., 2014), it is not surprising that many studies on affective disorders found selective deficits in this region (Adamaszek et al., 2017). Also, it might be possible that structural abnormalities in this structure have a role in the pathogenesis and in the progress of psychotic disorders, whose psychopathologic features are characterized by the difficulty of recognizing internal and external stimuli and creating a valid construct of reality.

Nevertheless, it is important to highlight that the reviewed studies suffer from some limitations. First, the heterogeneity, in terms of illness severity, course of illness (first-episode psychosis vs. chronic SCZ) and medication status (drug-naïve patients vs. patients in treatment with antipsychotic drugs), of the sample employed might have negatively impacted the generalizability of the findings. Second, the reviewed studies used MRI scanners with different magnetic fields (1.5T vs. 3T), which could impact the uniformity of the results. Finally, all studies included are cross-sectional, which limits the possibility of longitudinal evaluations of the results.

In conclusion, while it is possible to hypothesize an association between cerebellar vermis hypoplasia and SCZ, further studies are needed to define this association in the context of BD. Interestingly, this case report highlights how, in a case of BD with psychotic symptoms, there was an incidental finding of hypoplasia of the cerebellar vermis, not associated with neurological symptoms. It would be interesting to understand if cerebellar vermis abnormalities could be more frequently related not only to SCZ but also to other psychotic disorders, such as BD. Although further studies are needed to fully understand the relationship between cerebellar vermis hypoplasia and psychotic disorders, the interesting finding of this case report could provide a link between neuroimaging research and clinical practice. Indeed, even if the clinical use of neuroimaging tools has still a not decisive impact as an aid in the formulation of the diagnosis, the study of individual cases in which brain abnormalities are highlighted can provide a solid foundation for the promising developments of treatment personalization (Aydin et al., 2019; Keshavan et al., 2020).

Acknowledgement

This study was (partially) supported by the Italian Ministry of Health (Ricerca Corrente 2023) and by the Italian Ministry of Health (Grant No. GR-2019-12369100).

Conflicts of interest

There are no conflicts of interest.