Manic symptoms in schizophrenia spectrum disorders

Introduction

Schizophrenia is a severe mental disorder characterized by significant alterations in thought, perception, emotion, and behavior. Often regarded as a single mental disorder, it appears to reflect considerable heterogeneity. Schizophrenia symptoms are typically grouped into positive, negative, and disorganized symptoms, but no single symptom cluster is pathognomonic of schizophrenia (1). While mania is generally easy to recognize (2), severe cases with psychotic features can be misdiagnosed as schizophrenia, and milder cases may be mistaken for personality disorders (3). Furthermore, it is critical to distinguish between mania and manic symptoms. Whereas a manic episode significantly impacts various domains of functioning, may include psychotic symptoms, and usually requires hospitalization (4), manic symptoms do not necessarily meet the criteria for a full-blown manic episode.

Manic symptoms can appear within a range of psychiatric diagnoses, and do not exclusively form part of manic episodes. The relationship between schizophrenia and manic symptoms remains an area of limited research. Evidence, however, indicates the presence of manic symptoms across various diagnostic categories, including schizophrenia and schizoaffective disorder (5). Moreover, van Os and Kapur (2009) proposed changing the categorical dichotomy of schizophrenia and bipolar disorder (BD) to a dimensional conceptualization (6). Diagnosis of schizoaffective disorder requires meeting criteria for a major mood episode for most of the lifetime of the illness as well as psychotic symptoms without overt mood symptoms within a 2-week period (7). Manic symptoms are often part of schizoaffective disorder, but no consensus currently exists as to whether this disorder lies within the schizophrenia spectrum disorders (SSDs), the mood disorders or both (8–11).

Manic symptoms can significantly impact the clinical course and prognosis of schizophrenia (12) and as such, their assessment has been included in the mania domain of the Clinical-Rated Dimensions of Psychosis Symptom Severity (CRDPSS) scale proposed by DSM-5 for assessing the severity of psychotic symptoms in schizophrenia (13). A Korean study examining the psychometric properties of the Young Mania Rating Scale (YMRS) in SSDs, using a receiver operating characteristic analysis, found the optimal cut-off score for distinguishing schizophrenia patients with manic symptoms from those without to be 10, with a sensitivity of 88.3% and a specificity of 75.6% (14). This is in contrast to the cut-off point of 12 taken to be the threshold for diagnosing mania in mood disorders (15). They concluded that a YMRS score of 10 indicates mild mania severity on the Clinical Global Impression (CGI) scale, making it a reasonable threshold for identifying manic symptoms in patients with SSD.

Overlooking mania could result in missed opportunities to use pharmacological treatments and may lead clinicians to make excessively pessimistic prognoses (16). Individuals with schizophrenia often experience a more severe course of illness and have worse prognoses than those with schizoaffective disorder. Further research is necessary to categorize better the various clinical phenomena that fall under the umbrella of manic syndromes and SSDs. This study aims to investigate the presence of mania in stable patients diagnosed with SSDs. We hypothesized that in patients with SSD, manic symptoms are associated with clinical psychopathology.

Results

A total of 75, 44 male (58.7%) patients with SSD [mean age 43.55 (±11.800) years] and 31 female (41.3%) patients with SSD [mean age 47.81 (±10.521) years] met the inclusion criteria. The mean YMRS score for the total sample was 6.36 (±5.753). We dichotomized our group according to the severity of YMRS scoring and used the YMRS cut-off score of 10, suggested to be appropriate for the detection of mania in SSDs (Kim et al., 2018). The total sample was dichotomized into two groups: Group 1, without or with minimal symptoms of mania (YMRS ≤ 10) and Group 2, with distinct manic symptoms (YMRS ≥ 10). Group 1 (N = 55) had a mean YMRS total score of 3.42 (±3.004) and Group 2 (N = 20) had a mean YMRS total score of 14.45 (±3.052). The two groups differed significantly in terms of total mean YMRS scoring [M = 11.032 [95% confidence interval (CI): 9.462 to 12.602], t (73) = 14.005, p = 2.252 × 10⁻²², d = 3.657], as expected. As Levene's test for equality of variances was nonsignificant (p = 0.807), we could safely assume that the data were normally distributed. A post-hoc power calculation for independent t-tests at α = 0.05 found the statistical power to be equal to 1.000.

Descriptive statistics were used to provide a comprehensive summary of the mean YMRS individual item and total score for the total sample (Table 1). Table 2 summarizes the demographic characteristics of the participants without or with minimal manic symptoms and mania. Sex distribution and family history of mental illness did not differ significantly between the two groups (Fisher's exact test = 0.147 and Fisher's exact test = 1.000, respectively). For both, a post-hoc power calculation for chi-square tests with df = 1 at α = 0.05 and for a medium effect size of 0.33, found the statistical power to be equal to 0.820.

Interestingly, we found a positive significant correlation between YMRS total score and the Positive and Negative Syndrome Scale (PANSS) total score (r² = 0.516, p = 2.147 × 10⁻⁶), the PANSS-Positive subscore (r² = 0.600, p = 1.310 × 10⁻⁸; Figure 1), and the PANSS-General Psychopathology subscore (r² = 0.444, p = 6.646 × 10⁻⁵), but the correlation between YMRS total score and PANSS-Negative subscore failed to reach significance (Table 3). Furthermore, independent-samples t-tests were performed to explore mean differences in PANSS total and subtest, Generalized Anxiety Disorder -7 scale (GAD-7), and Risk Assessment of Suicidality Scale (RASS) scores between participants without or with minimal symptoms of mania, and in those with manic symptoms. Positive symptoms scoring, as assessed by the PANSS-Positive subscale, was the only psychopathology measure that stood the stringent criterion of the Bonferroni multiple correction and showed a statistically significant difference between the two YMRS groups [PANSS-Positive subscore mean diff = 6.841 (95% CI: 3.417to 10.265), t(73) = 3.982, p = 0.00016, d = 1.040] (Table 4). A post-hoc power calculation based on data from this independent t-test comparison, for d = 1.040 and α = 0.05, found the statistical power to be over 0.999, a more than adequate value for detecting an effect. More specifically, analyzing correlations between individual YMRS items and the PANSS total and subscale scores (Table 5), PANSS-Positive subscore was positively correlated with YMRS item 7 on language and thought disorder (r² = 0.449, p = 5.239 × 10⁻⁵) and YMRS item 11 on insight (r² = 0.522, p = 1.593 × 10⁻⁶) at a post-hoc calculated power of 0.990 and 0.999, respectively, given α = 0.05.

View Figure

• Download as Powerpoint
• Download Figure

Finally, linear regression established that the PANSS-Positive subscale score could significantly predict the YMRS total score [F(1,73) = 36.851, p = 5.214 × 10⁻⁸]. The YMRS total score accounted for 32.6% of the explained variability in PANSS-Positive score.

Discussion

In this study, we found an increased number of manic symptoms denoting the presence of mania in just over one in four (26.7%) stable participants with SSDs. An earlier epidemiological study from Canada found that the prevalence of an episode of mania in patients with schizophrenia in the community was 17.7 % (17), and a more recent study showed significant subthreshold manic symptoms (YMRS score > 7) to be present in 25.1% of patients (18).

In this group of patients with SSD, we showed that positive symptoms were associated with mania. Interestingly, the severity of positive symptoms was found to predict the presence of manic symptoms, such that, the higher the PANSS-Positive score, the more likely the presence of manic symptoms. Our findings replicate results from a previous study on 175 patients with schizophrenia aiming to search for patterns in clinical symptomatology suggestive of the presence of mood disorders under the label of schizophrenia, also showing that mood symptoms correlate with positive symptoms (18).

Furthermore, to ascertain whether metabolic variability is associated with the clinical features of schizophrenia, Malaspina et al. (2021) examined the association of N-acetylaspartate (NAA) and choline (Cho) levels with clinical symptoms in patients with schizophrenia. They found a positive correlation between manic symptoms, as assessed by the YMRS, and whole-hippocampus multivoxel average choline millimolar concentration of Cho, denoting that both manic symptoms and positive symptoms reflect demyelination. On the contrary, negative symptoms were correlated with decreased NAA hippocampal levels reflecting a different pathophysiologic process, consistent with microgliosis/astrogliosis and/or lower vitality (19).

Significant progress has been made in understanding the genetics of schizophrenia over the last 15 years, shedding light on the close relationship between SSDs and other conditions, particularly BD and childhood neurodevelopmental disorders. A clearer picture is emerging, suggesting that clinical heterogeneity partly reflects etiological heterogeneity. For example, several etiological pathways are influenced by the catechol-O-methyltransferase gene (COMT), including prefrontal cognition or emotional processing in the amygdala and the prefrontal cortex, in addition to other insults to the brain such as adolescent cannabis use. This means that the individual clinical phenotype may result from a combination of distinct symptom dimensions and their associated genetic risk factors (20).

COMT is an enzyme catalyzing the breakdown of dopamine and norepinephrine, thought to be involved in the pathophysiology of BD and schizophrenia. COMT striatal activity, but not the rs4680 (COMT Val/Met) functional polymorphism, may be a biomarker for manic symptoms (21), and research has suggested that the effect of this variant may be associated with comorbid manic symptoms in schizophrenia (22). Using the OPCRIT criteria (23), an Irish study showed significant overtransmission of the Val allele for mania in patients with schizophrenia (24).

Interestingly, it has been suggested that second-generation antipsychotics, with the exception of clozapine, may induce states of agitation often resembling manic states, possibly via their antidepressant actions on serotonergic and noradrenergic neurotransmission (25).

In the era of promoting health economics through screening (26), we suggest that administering the YMRS, a relatively easy-to-use and cost-effective tool, to screen readily for mania in SSDs may prove a valuable strategy for the busy clinician. YMRS could help identify mania in SSDs, as early intervention may lower the costs of treating poorly responding revolving-door patients (27) and improve patient outcomes (28), thus decreasing costs for the patient and the mental health and welfare systems. Adding mood stabilizers (29) and engaging the patient in psychoeducation (30) may prevent frequent relapses associated with high expenditure for the patient and society.

It is crucial, however, to interpret our findings by considering the various limitations of this study. First, the subjective nature of YMRS introduces inter-rater variability, which could affect the reliability and validity of the assessments. It has recently been suggested that implementing flags and mitigation strategies during trials may enhance the value of YMRS data, direct emphasis toward rater training, and bolster the reliability and validity of trial outcomes (31). Therefore, future YMRS assessments will have to be undertaken by the same trained rater for all patients. This study could serve as a pilot study, as a more representative and larger sample size is required in future studies to enhance data reliability. In the future, genotyping either for known genetic polymorphisms or within a genome-wide association study (GWAS) protocol, without an a priori hypothesis, holds promise for disentangling the dimensional etiology of SSDs, part of which seems to stem from the presence of manic symptoms. Another strategy for furthering our understanding of manic symptoms in SSDs may be to focus on patients with drug-naïve first-episode SSD, to eliminate any drug-induced agitation.

Nevertheless, this study highlights that by addressing manic symptoms contributing to and associated with positive psychotic psychopathology in individuals with SSDs, we could improve the management of acute SSD episodes and promote remission, especially in poorly responding patients with undetected, hence suboptimally treated manic symptoms.

Conclusions

This study explored the presence of manic symptoms in patients with SSDs. We showed the severity of positive symptoms to correlate with an increased number of manic symptoms, as assessed by the YMRS. In this group of patients, positive symptoms also predicted the presence of manic symptoms. It, therefore, appears that in SSDs, YMRS could be used as a friendly and reliable screening tool promoting individualized and precision management, increasing the cost-effectiveness of interventions. Further, beyond cross-sectional studies, the high degree of phenomenological pleiotropy within SSDs points to the need for extensive transdiagnostic research to delineate biologically distinct entities incorporating carefully collected phenotypical data from diverse global communities, with the application of new and emerging technologies.

Materials and Methods

Study Design/Procedures

Assessment was performed during three sessions on separate days within 1 month. The first session was physician-led and included a thorough medical and psychiatric history taking. The second session, usually no later than a week after the second session, was psychologist-led under the supervision of a psychiatrist. It comprised the clinical interview for assessing psychopathology, including completion of the YMRS, by one of four clinical psychology research assistants. Self-report scales were completed during the third and final session (Figure 2).

View Figure

• Download as Powerpoint
• Download Figure

Data Availability

Data availability is restricted due to human subject involvement and is non-public. All data used in the analysis are available upon reasonable request to the corresponding author.

Acknowledgments

A special thanks to Anna-Maria Andreopoulou, Sophia Athanasiou, Styliani Stai, Thalia Deligianni, Mariliza Papadatou, and Evangelia Vlachou for helping with data entry.

Author Contributions

E.M.T. oversaw clinical research coordination and organized, managed and analyzed the database of results. She was the primary author of the manuscript and orchestrated and contributed to the intellectual conceptualization of the perspective paper; she also participated as a clinical supervisor in data collection. She also acted as research coordinator. D.P. and M.K. conducted semistructured interviews on participants, collected clinical data under supervision and edited the manuscript. K.C. and S.F. collected clinical data and collated relevant data and edited the manuscript. G.K. conducted all primary assessments and screening. K.N.F. conceptualized the research project, was the primary investigator of this study and contributed to the writing and editing of this manuscript.

Funding Sources

This article was not funded by external sources.

Author Disclosures

The authors have confirmed that no conflict of interest exists. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. The manuscript has been read and approved by all authors.