Introduction
Adolescence is a crucial period for the development of the nervous system; cognitive, emotional, and social functions; and mental health.Reference Foulkes and Blakemore 1 A meta-analysis of 133,693 patients with schizophrenia revealed that schizophrenia risk increases from adolescence (1.05 per 10,000 person-years) and reaches its peak (4.15 per 10,000 person-years) at the age of 20 to 29 years.Reference van der Werf, Hanssen and Kohler 2 Evidence has indicated that up to 45% of first episodes of bipolar disorder occur before the age of 20 years, with a mean age of approximately 17 years.Reference Bolton, Warner, Harriss, Geddes and Saunders 3 According to the U.S. Adolescent Depression Project, 28% of adolescents experience at least one episode of major depressive disorder by the age of 19 years, with the median disease-onset age of 15.5 years.Reference Lewinsohn, Rohde and Seeley 4
Increasing evidence supports the dysregulation of pro- and anti-inflammatory cytokines during the first episodes and early disease stages of schizophrenia, bipolar disorder, and major depressive disorder.Reference Noto, Maes and Nunes 5 –Reference Lesh, Careaga and Rose 7 Noto et al found that inflammatory markers, particularly interleukin (IL)-6 and IL-8, are significantly correlated with negative, psychotic, and affective symptom dimensions in drug-naïve patients aged 25.8 ± 6.4 years old with first-episode psychosis.Reference Noto, Maes and Nunes 5 A meta-analysis of 3453 and 1095 patients with first-episode schizophrenia and major depressive disorder, respectively, revealed increased levels of IL-6, C-reactive protein (CRP), and tumor necrosis factor (TNF)-α in these patients.Reference Cakici, Sutterland, Penninx, Dalm, de Haan and van Beveren 6 A population-based study of 2472 patients with schizophrenia, 2236 with bipolar disorder, and 12,606 with depression found that CRP differed significantly between mental disorders, being highest in individuals with bipolar disorder, followed by schizophrenia and depression.Reference Horsdal, Kohler-Forsberg, Benros and Gasse 8 Lesh et al compared the cytokine profiles of 69 and 16 patients with first-episode schizophrenia and bipolar disorder, respectively, and reported that, compared with healthy controls, patients with schizophrenia had higher levels of IL-1β, IL-2, and IL-6, and those with bipolar disorder had higher levels of IL-10.Reference Lesh, Careaga and Rose 7 Goldsmith et al further suggested that increased levels of TNF-α and IL-6 were the most reliable results in patients with schizophrenia, bipolar disorder, and major depressive disorder compared with healthy controls.Reference Goldsmith, Rapaport and Miller 9 However, mixed samples of adolescents and adults, particularly middle-aged people, with first-episode severe mental disorders may confound the real effect of pro- and anti-inflammatory cytokines on the pathophysiology of first-episode schizophrenia, bipolar disorder, and major depressive disorder.
Cognitive deficit is a crucial clinical and etiological concern in patients with first-episode schizophrenia, bipolar disorder, and major depressive disorder.Reference Bombin, Mayoral and Castro-Fornieles 10 , Reference Barrett, Mulholland, Cooper and Rushe 11 Bombin et al demonstrated that adolescents aged approximately 16 years with first-episode schizophrenia and those with first-episode bipolar disorder share common impairments in cognitive function, including working memory, attention, and executive function.Reference Bombin, Mayoral and Castro-Fornieles 10 Barrett et al assessed the neurocognitive function of 32 and 46 patients with first-episode bipolar disorder and first-episode schizophrenia, respectively, through analysis of covariance and demonstrated that memory, executive functioning, and language were the most impaired.Reference Barrett, Mulholland, Cooper and Rushe 11 A cross-sectional study of 36 drug-naïve patients with first-episode major depressive disorder revealed impairment in their visual, working, and verbal memory.Reference Chen, Jiang and Wang 12 A meta-analysis of 31 studies including 994 patients with first-episode major depressive disorder indicated persistent impairment in inhibition control, verbal fluency, and processing speed in acute and remitted phases.Reference Ahern and Semkovska 13 However, the major concern regarding the aforementioned studies was that psychological symptoms were not adjusted during the analyses of cognitive function.
Studies on the association of pro- and anti-inflammatory cytokines with cognitive function in first-episode severe mental disorders are still limited.Reference Orhan, Fatouros-Bergman and Schwieler 14 , Reference Tian, Li and Xu 15 Orhan et al revealed that IL-18 is associated with cognitive impairment measured by using MATRICS Consensus Cognitive Battery in patients aged 28.5 ± 1.07 years with first-episode schizophrenia.Reference Orhan, Fatouros-Bergman and Schwieler 14 Tian et al revealed that abnormal levels of nucleotide-binding oligomerization domain NOD-like receptor 3 inflammasome, IL-18, and nuclear factor-κB are associated with poor performance in the Wechsler Memory Scale in patients aged 49.45 ± 7.32 years with first-episode major depressive disorder.Reference Tian, Li and Xu 15 However, their study enrolled adults and did not particularly focus on adolescents.Reference Orhan, Fatouros-Bergman and Schwieler 14 , Reference Tian, Li and Xu 15
In the current study, we specifically investigated the cytokine profiles and cognitive function in adolescents with first-episode schizophrenia, bipolar disorder, or major depressive disorder. In the context of any type of psychiatric condition resulting in elevated levels of risk-measures vs lower levels of protective measures, we hypothesized that adolescents with first-episode schizophrenia, bipolar disorder, or major depressive disorder exhibit higher levels of proinflammatory cytokines, such as IL-6 and TNF-α, and perform poorly in executive and working memory function tasks compared with controls.
Methods
Participants
We included adolescents aged between 12 and 19 years who met the criteria specified in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition for schizophrenia, bipolar disorder, or major depressive disorder. They had a first episode of illness, and had a Clinical Global Impression–Severity scale for schizophrenia, bipolar disorder, or major depressive disorder score of ≤3 in a relatively stable condition. Patients with intellectual disability, organic mental disorder, substance use disorder, pregnancy or breastfeeding, severe autoimmune/immune diseases, stroke, epilepsy, and unstable physical illnesses were excluded. We also enrolled age- and sex-matched healthy controls without any of the mentioned physical conditions or psychiatric disorders based on the Mini-International Neuropsychiatric Interview. Demographic characteristics, including age, sex, body mass index (BMI), were recorded, and clinical assessments, including 17-item Hamilton Depression Rating Scale (HDRS),Reference Hamilton 16 Young Mania Rating Scale (YMRS),Reference Young, Biggs, Ziegler and Meyer 17 and Positive and Negative Syndrome Scale (PANSS),Reference Andreasen and Olsen 18 were performed. The Institutional Review Board of the Taipei Veterans General Hospital approved our study, and it was performed in accordance with the Declaration of Helsinki. All patients and their parents provided written informed consent prior to their inclusion in the study.
Measurement of inflammatory cytokines
Proinflammatory cytokines, including IL-2, IL-6, TNF-α, and CRP, were assayed using enzyme-linked immunosorbent assay (ELISA) kits (R&D systems, Minneapolis, MN, USA) for all participants. Fasting serum samples were collected in serum separator tubes, clotted for 30 minutes, and stored at −80°C until use. All assays were performed according to the vendor’s instructions. The final absorbance of each sample of the mixture was measured and analyzed at 450 nm using an ELISA plate reader with Bio-Tek Power Wave Xs and Bio-Tek’s KC junior software (Winooski, VT, USA). The standard range was considered as specified in the vendor’s instructions. A linear regression R-squared value of at least 0.95 was considered a reliable standard curve.
Measurement of neuropsychological-cognitive functions
In the current study, the Wisconsin Card Sorting Test (WCST) and working memory task were examined for executive function and working memory. WCST required strategic planning, organized searching, utilizing environmental feedback to shift cognitive sets, directing behavior toward achieving a goal, and modulating impulsive responding. In the working memory task, subjects were asked to respond as quickly as possible when they saw a number that appeared again only separated by one other number (i.e., 23–45–23; participants responded to the second 23 as quickly as possible). Executive functions are a wide number/type of functions, and there is evidence in the literature that not all the psychiatric conditions present the same deficits in executive function. Apart from working memory, there is no gold-standard executive function for any psychiatric conditions. Working memory task and WCST were commonly used in our previous studies.Reference Chen, Li and Lin 19 –Reference Chen, Chang and Bai 22
Statistical analysis
For between-group comparisons, the F-test was used for continuous variables, and Pearson’s test was used for categorical variables. General linear models (GLMs) with the adjustment of demographic data (age, sex, and BMI) and clinical symptoms (HDRS, YMRS, and PANSS) were performed to examine the levels of proinflammatory (IL-6, TNF-α, and CRP) and anti-inflammatory (IL-2) cytokines between groups. GLMs with adjustment of demographic data and with/without adjustment of clinical symptoms were used to examine the cognitive function (WCST and working memory task) between groups. GLMs with demographic data, groups, and clinical symptoms were performed to assess the association of proinflammatory and anti-inflammatory cytokines with cognitive function. A two-tailed P-value of less than .05 was considered statistically significant. All data processing and statistical analyses were performed using the SPSS version 17 software (SPSS Inc., Chicago, IL, USA).
Results
In all, 26 adolescents with first-episode schizophrenia, 35 with first-episode bipolar disorder, 29 with first-episode major depressive disorder, and 22 age-/sex-matched controls were included in the current study, with a female predominance (Table 1). Adolescents with schizophrenia and bipolar disorder had a higher BMI (P = .039) than those with major depressive disorder and controls (Table 1). Adolescents with schizophrenia exhibited the highest scores in PANSS (P < .001), those with bipolar disorder with the highest scores in YMRS (P < .001), and those with major depressive disorder with the highest scores in HDRS (P < .001; Table 1). Adolescents with bipolar disorder had higher CRP levels (P = .051) than other three groups, and adolescents with schizophrenia, bipolar disorder, and major depressive disorder had lower IL-2 levels than the controls (P = .001; Table 1). Adolescents with schizophrenia exhibited the poorest performance in WCST and working memory task (all P < .05) compared with other three groups (Table 1).
Table 1. Demographic Characteristics, Proinflammatory Cytokines, and Cognitive Function between Groups
Abbreviations: BMI, body mass index; CRP, C-reactive protein; HDRS, 17-item Hamilton Depression Rating Scale; IL, interleukin; PANSS, Positive and Negative Syndrome Scale; SD, standard deviation; TNF, tumor necrosis factor; WCST, Wisconsin card sorting test; YMRS, Young Mania Rating Scale.
GLMs with the adjustment of demographic data and clinical symptoms found that patents with bipolar disorder exhibited highest levels of CRP (P = .023), IL-6 (P = .022), and TNF-α (P = .011), and had lowest IL-2 levels (P = .034; Figure 1). Levels of IL-6 (P = .054) and TNF-α (P = .021) in major depressive disorder group ranged between those in bipolar disorder group and in control group (Figure 1). Adolescents with schizophrenia exhibited similar cytokine profiles compared with the controls, except higher IL-6 levels (P = .077; Figure 1). GLMs with the adjustment of demographic data showed a dose-dependent pattern between diagnoses and cognitive function, which meant that adolescents with schizophrenia exhibited the poorest performance in cognitive tasks, those with bipolar disorder with intermediate performance, and those with major depressive disorder with relatively well performance (Figure 2). However, the significance of the differences in cognitive function between groups disappeared after adjusting for clinical symptoms (HDRS, YMRS, and PANSS; Table 2). Finally, only CRP levels (P = .027) were negatively associated with executive function measured by WCST (Table 2).
Figure 1. Estimated marginal means of cytokines between groups with adjustment of age, sex, body mass index, and symptom scores.
Figure 2. Estimated marginal means of cognitive function between groups with adjustment of age, sex, and body mass index. (a) Wisconsin Card Sorting Test scores. (b) Working memory task scores.
Table 2. GLMs with Adjustment of Age, Sex, BMI, and Symptom Scores between Executive Function (% Conceptual-Level Responses in WCST) and Cytokines
Abbreviations: B, parameters to be estimated; BMI, body mass index; CRP, C-reactive protein; GLM, general linear model; HDRS, 17-item Hamilton Depression Rating Scale; IL, interleukin; PANSS, Positive and Negative Syndrome Scale; t, t-test value; TNF, tumor necrosis factor; WCST, Wisconsin card sorting test; YMRS, Young Mania Rating Scale. Bold type indicates statistical significance (p < 0.05)
Discussion
Our study findings partially support our hypothesis. Adolescents with first-episode bipolar disorder had higher levels of proinflammatory cytokines (IL-6, TNF-α, and CRP) and lower levels of anti-inflammatory cytokines (IL-2) than controls. Adolescents with first-episode major depressive disorder had increased levels of TNF-α. Adolescents with first-episode schizophrenia exhibited only slightly increased IL-6 levels. Among the four groups, adolescents with schizophrenia performed the worst in cognitive tasks followed by those with bipolar disorder and major depressive disorder. Surprisingly, the significant difference in cognitive function between the disease and control groups disappeared after complete adjustment of psychological symptoms.
Counterintuitively, the worst cytokine profile (highest IL-6, TNF-α, and CRP levels and lowest IL-2 levels) was found in adolescents with first-episode bipolar disorder but not in those with first-episode schizophrenia, although patients with schizophrenia exhibited significantly greater psychiatric symptoms as measured using PANSS and poorer cognitive function than those with major affective disorders. Siwek et al found that the later stages of bipolar disorder were characterized by increased soluble TNF-receptor 80-kDa levels and decreased sIL-2R levels.Reference Siwek, Sowa-Kucma and Styczen 23 However, we observed increased levels of TNF-α, IL-6, and CRP and decreased levels of IL-2 in patients with first-episode bipolar disorder; these findings are compatible with the results of Lin et al and Goldstein et al that the concentrations of IL-6 and TNF-α are higher in young patients with bipolar disorder and at-risk individuals.Reference Goldstein, Lotrich and Axelson 24 , Reference Lin, Shao and Wang 25 Furthermore, IL-6 is regarded as a potential biomarker of early-stage schizophrenia.Reference Stojanovic, Martorell and Montalvo 26 Stojanovic et al demonstrated that higher IL-6 levels predicted the conversion from the at-risk mental state to schizophrenic episode.Reference Stojanovic, Martorell and Montalvo 26 Interestingly, a study suggested that increased TNF-α levels triggered an acute illness (mania and depression) in major affective disorders but returned to baseline during the remission of diseases.Reference Sowa-Kucma, Styczen and Siwek 27 Finally, based on our findings with complete adjustment of BMI and symptom scores, increased IL-6 levels are commonly related to severe mental disorders, high TNF-α levels are specifically associated with bipolar disorder and major depressive disorder, and increased CRP and reduced IL-2 levels are specifically associated with bipolar disorder. We may further hypothesize that the discrepancies between the cytokine profiles of first-episode schizophrenia, bipolar disorder, and major depressive disorder implied that in the first episode and early stages of bipolar disorder and major depressive disorder are related to an episodic cytokine storm (ie, IL-6, TNF-α, and IL-2) with a fluctuating pattern, whereas schizophrenia is associated with insidious accumulation of proinflammatory cytokines, particularly IL-6. Thus, imbalance between immune inflammatory response system and compensatory immune regulatory reflex system may be associated with the pathophysiology of schizophrenia, bipolar disorder, and major depressive disorder.Reference Misiak, Stanczykiewicz, Kotowicz, Rybakowski, Samochowiec and Frydecka 28 , Reference Roomruangwong, Noto and Kanchanatawan 29 The optimal modulation of imbalanced immune systems may be beneficial to achieve the improved therapeutic effect in the treatment of severe mental disorders.Reference Radtke, Chapman, Hall and Syed 30 Furthermore, additional studies with a longitudinal study design may be necessary to elucidate the aforementioned hypothesis and to revalidate our findings.
Studies have reported persistent deficits in cognitive function in patients with first-episode schizophrenia, bipolar disorder, and major depressive disorder before and after pharmacological treatment.Reference Hill, Reilly and Harris 31 –Reference Reyes, Cardoso and Jansen 33 Hill et al enrolled patients with first-episode psychosis (30 patients with schizophrenia, 22 patients with bipolar disorder patients with psychosis, and 21 patients with psychotic depression) and assessed their neuropsychological function at baseline and after a 6-week risperidone treatment; they noted no differential improvement in cognition (ie, working memory, attention, and memory) over time or across neuropsychological domains in the three patient groups with risperidone treatment.Reference Hill, Reilly and Harris 31 Bilder et al reported that residual psychiatric symptoms are associated with increased deficits in executive function and attention in patients with first-episode schizophrenia treated according to a standardized algorithm.Reference Bilder, Goldman and Robinson 32 However, Reyes et al compared the cognitive functions of 91 young adults with bipolar disorder, 90 with major depressive disorder, and 1077 community controls using the Montreal Cognitive Assessment (MoCA), and revealed that MoCA scores of patients with bipolar disorder and major depressive disorder were similar with those of controls after adjustment for psychiatric symptoms.Reference Reyes, Cardoso and Jansen 33 In the current study, we found that adolescents with first-episode schizophrenia exhibited the lowest performance in WCST and working memory tasks before adjustment for psychiatric symptoms. However, the three disease groups did not differ from the control group in terms of cognitive function after completely controlling for psychiatric symptom, which may imply that perhaps the clinical symptoms (clinical severity) rather than the clinical diagnosis (dichotomic diagnosis) provided a better account of the biological expression, including cognitive function. Further studies are necessary to elucidate whether cognitive impairment in the early stage (first episode and at-risk state) of these illnesses may be related to residual and prodromal symptoms of psychopathology.
In the current study, we found a negative relationship between CRP levels and executive function in adolescents with first-episode mental disorders. A systematic review reported that the most consistent result of worse cognitive performance with high CRP levels was observed in patients with schizophrenia and also indicated potential roles of CRP, IL-6, and TNF-α in the cognitive impairment of patients with bipolar disorder.Reference Misiak, Stanczykiewicz, Kotowicz, Rybakowski, Samochowiec and Frydecka 28 A recent British biobank study revealed that the highest serum CRP quintile was significantly associated with impairment in cognitive performance in major affective disorders.Reference Milton, Ward and Ward 34 Millett et al identified CRP as a significant positive predictor of proxy cognitive decline in euthymic patients with bipolar disorder.Reference Millett, Perez-Rodriguez and Shanahan 35 Krogh et al demonstrated that higher CRP levels were associated with lower psychomotor speed both at baseline and a 3-month exercise intervention in patients with major depressive disorder.Reference Krogh, Benros, Jorgensen, Vesterager, Elfving and Nordentoft 36
This study has several limitations. First, only the WCST and working memory task were used for measuring cognitive function. Additional studies may be required to comprehensively evaluate cognitive functions of adolescents with first-episode severe mental disorders. Second, the medications used by the patients were not discontinued during the cognitive function assessment and cytokine examination in the current study, and they may have affected the assessment and cytokine level results, although previous studies suggested that psychotropic medications, including atypical antipsychotics, mood stabilizers, and antidepressants, may exhibit the anti-inflammatory effect in the patients with severe mental disorders.Reference Baumeister, Ciufolini and Mondelli 37 , Reference Stapel, Sieve, Falk, Bleich, Hilfiker-Kleiner and Kahl 38 Allowing patients to continue their medications was ethically appropriate and prevented disease exacerbation and relapse during the study. Furthermore, it provided more naturalistic data. However, a drug-free study design would be required to validate our findings. Third, there was no measure or discussion of psychosocial stress in the current study, which may also influence the levels of proinflammatory cytokines and might contribute to answer why the bipolar group had abnormalities in these markers but not the schizophrenia group.Reference Wieck, Grassi-Oliveira and Prado 39 Finally, the study sample size was still relatively small in the current study. Further studies with larger sample sizes would be required to validate our results.
In conclusion, adolescents with first-episode bipolar disorder exhibited the highest levels of proinflammatory cytokines, such as CRP, IL-6, and TNF-α, and the lowest levels of anti-inflammatory cytokines, such as IL-2, but adolescents with first-episode schizophrenia performed the worst in cognitive tasks. Dysregulated cytokine profiles and impaired cognitive function were observed in the early stage (first episode) of severe mental disorders. Furthermore, altered proinflammatory cytokines, particularly CRP, were found to be related to cognitive impairment in adolescents with first-episode schizophrenia, bipolar disorder, and major depressive disorder.
Acknowledgment
We would like to thank I-Fan Hu for his friendship and support.
Funding Statement
The study was supported by grant from Taipei Veterans General Hospital (V106B-020, V107B-010, V107C-181, V108B-012, V110C-025, and V110B-002), the Yen Tjing Ling Medical Foundation (CI-109-21, CI-109-22, and CI-110-30), and the Ministry of Science and Technology, Taiwan (107-2314-B-075-063-MY3, 108-2314-B-075-037, 110-2314-B-075-026, and 110-2314-B-075-024-MY3). The funding source had no role in any process of our study.
Disclosures
Y.M.B., J.W.H., and M.H.C. designed the study. M.H.C. analyzed the data and drafted the first version of the manuscript. K.L.H., S.J.T., and P.C.T. performed literature search and reviewed the revised manuscript. All authors contributed substantially to the manuscript and approved the final manuscript for submission. All authors are responsible for the integrity, accuracy, and presentation of the data. The authors do not have any conflict of interest to disclose. The authors have no financial relationships relevant to this article to disclose.
