Skip to main content
NCBI home page
Schizophrenia Bulletin logoLink to Schizophrenia Bulletin
. 2019 Oct 11;46(3):722–731. doi: 10.1093/schbul/sbz100

Choroid Plexus Enlargement and Allostatic Load in Schizophrenia

Yan-Fang Zhou 1, Jun-Chao Huang 1, Ping Zhang 1, Feng-Mei Fan 1, Song Chen 1, Hong-Zhen Fan 1, Yi-Min Cui 2, Xing-Guang Luo 3, Shu-Ping Tan 1, Zhi-Ren Wang 1, Wei Feng 1, Ying Yuan 4, Fu-De Yang 1, Anya Savransky 5, Meghann Ryan 5, Eric Goldwaser 5, Joshua Chiappelli 5, Laura M Rowland 5, Peter Kochunov 5, Yun-Long Tan 1,✉, L Elliot Hong 5
PMCID: PMC7147577  PMID: 31603232

Abstract

Although schizophrenia is a brain disorder, increasing evidence suggests that there may be body-wide involvement in this illness. However, direct evidence of brain structures involved in the presumed peripheral-central interaction in schizophrenia is still unclear. Seventy-nine previously treatment-naïve first-episode schizophrenia patients who were within 2-week antipsychotics initial stabilization, and 41 age- and sex-matched healthy controls were enrolled in the study. Group differences in subcortical brain regional structures measured by MRI and the subclinical cardiovascular, metabolic, immune, and neuroendocrine biomarkers as indexed by allostatic load, and their associations were explored. Compared with controls, patients with schizophrenia had significantly higher allostatic load (P = .001). Lateral ventricle (P < .001), choroid plexus (P < .001), and thalamus volumes (P < .001) were significantly larger, whereas amygdala volume (P = .001) was significantly smaller in patients. The choroid plexus alone was significantly correlated with higher allostatic load after age, sex, education level, and the total intracranial volume were taken into account (t = 3.60, P < .001). Allostatic load was also significantly correlated with PANSS positive (r = 0.28, P = .016) and negative (r = −0.31, P = .008) symptoms, but in opposite directions. The peripheral multisystemic and central nervous system abnormalities in schizophrenia may interact through the choroid plexus during the early stage of the illness. The choroid plexus might provide a sensitive structural biomarker to study the treatment and prevention of brain-periphery interaction abnormalities in schizophrenia.

Keywords: schizophrenia, chronic stress, choroid plexus, allostatic load

Introduction

Schizophrenia is a devastating brain disorder with a decreased life expectancy of approximately 15–20 years, largely due to increased common medical illnesses such as cardiovascular and metabolic diseases1,2 that are often attributed to the side effects of antipsychotic medications.3,4 Although there is no dispute that antipsychotic medications contribute to these risks, increased comorbid medical illnesses were reported even before the use of antipsychotics became widespread,5,6 and have also been observed even in first-episode, antipsychotic-naïve patients.6,7 The underlying mechanism connecting schizophrenia as primarily a brain illness to its peripheral medical vulnerability remains obscure.

The stress response system demands close central-peripheral coordination, and as such its abnormality may play an important role. The “stress cascade” system is known to contribute to the vulnerability of schizophrenia and its pathophysiology.8–10 When faced with repeated stressors, organisms adjust their homeostatic set points to achieve “allostasis.” 11,12 Allostatic mechanisms are adaptive in the short term, but chronic elevation of allostatic activity due to prolonged stress or abnormal responses may lead to “wear and tear” physiological changes, referred to as allostatic load (AL).13 AL has been clinically estimated through cardiovascular, metabolic, neuroendocrine, and immune-inflammatory markers.14 Elevated AL has been associated with increased risk of cardiovascular disease and mortality15 and psychiatric illness14,16 including schizophrenia.10 Three recent independent studies showed that elevated AL may already be present in first-episode schizophrenia.17–19

Assessing AL early during psychosis onset is critical to determine whether increased AL is a consequence of schizophrenia itself vs antipsychotic medications. A study with 28 apparently later-onset (~32 years of age) first-episode patients, mostly antipsychotic-naïve, already showed elevated AL compared with that of controls.18 Another study of 36 first-episode patients aged 27.5 years on average also showed significantly higher AL within 30 days of antipsychotic medication,19 and the higher AL was associated with impaired working memory.20 A third study in 21 schizophrenia patients (averaged age = 23.4 years) who had up to 5 years of antipsychotics treatment showed elevated AL compared with age-matched controls.17 Overall, there is evidence suggesting that elevated AL in schizophrenia occurs independent of antipsychotics exposure, but the sample sizes in these studies were modest. Our first goal was to determine whether elevated AL can be replicated in first-episode psychosis in a large sample that had minimal antipsychotic medication exposure.

If the peripherally measured AL is already elevated at treatment onset, the next critical question is how high AL is related to the brain at this early stage. We have found that high AL is associated with reduced cortical thickness21 and compromised integrity of the fornix,22 the white matter connecting the hippocampus and hypothalamus, although both findings were based on chronic schizophrenia where measures are likely affected by aging and chronicity of disease. Besides the hippocampus and hypothalamus that are known to regulate stress responses,23,24 another brain structure most likely sensitive to AL is the choroid plexus (CP), as it is a major gateway for peripheral-central exchanges, maintaining cerebrospinal fluid (CSF) ion homeostasis, and critically supporting neuronal and glial development including adult neurogenesis.25,26 Importantly, CP is highly sensitive to immunological,27–31 cardiovascular,32 and metabolic changes,33 measures that are part of the AL construct. Therefore, among neuroanatomical regions, CP represents a structure highly interactive with the periphery. Interestingly, CP calcifications have been associated with more serious psychotic symptoms.34–36 Others have reported significant relationships between chronic stress and subcortical brain regions particularly the hippocampus and the CP,1,37,38 although none have assessed AL associations with these structures in schizophrenia. Therefore, our second goal was to test the hypothesis that elevated AL in schizophrenia may be associated with abnormal subcortical brain regional structures, particularly with CP and hippocampus.

Methods

Participants and Clinical Protocol

First-episode patients of schizophrenia (FES) within 2 weeks of antipsychotic medication initiation (n = 79) and age- and sex-matched healthy controls (HC, n = 41) were recruited. All participants were Han Chinese and the recruitment occurred between the years 2017 and 2018. Patients were enrolled during their first hospitalization of Beijing Huilongguan Hospital, and inclusion criteria were: (a) met the criteria for schizophrenia of the Structured Clinical Interview of the DSM-IV; (b) aged between 16 and 45 years; (c) total illness duration less than 3 years and otherwise no previous exposure to antipsychotic medications; (d) within 2 weeks of antipsychotic medication initiation. Individuals with mental retardation or organic brain disorders, neurological and unstable medical illnesses, and alcohol or substance use disorder were excluded. Participants were also screened and excluded if they were on regular administration of neurotrophic agents, immune modulators, or antioxidants within the past 8 weeks. Healthy volunteers were recruited from nearby communities and were excluded if they had a history of psychiatric disorders or psychosis among their first-degree relatives. The research was approved by the ethics committee and institutional review board of the Beijing Huilongguan Hospital, and all participants provided written informed consent.

Most patients at admission had severe psychosis and were treated without delay. Initial treatment also helped avoiding AL assessments to be conducted under an acutely psychotic state. Imaging data were collected within 2 weeks of treatment initiation, when four patients were medication-free, 16 patients were on a first-generation of antipsychotics (haloperidol) combined with second-generation antipsychotics (either risperidone or olanzapine), and the remaining patients were on the following second-generation antipsychotics: risperidone (33), olanzapine (13), aripiprazole (7), paliperidone (2), quetiapine (1), and iloperidone (3).

Symptom Assessments

The severity of clinical symptoms was assessed using the Positive and Negative Syndrome Scale (PANSS).39 Two trained psychiatrists conducted the PANSS assessments, and they achieved an intraclass correlation coefficient (ICC) above 0.80 before the study. The PANSS positive, negative, and general psychopathology scale scores were calculated.

Assessment of Biological Indicators

Thirteen biomarkers representative of cardiovascular, metabolic, inflammation, and stress hormone were used to calculate AL.21,22 Cardiovascular indicators included systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate, measured in the sitting position after 10 min of rest. Metabolic indicators included body-mass index (BMI), waist-hip ratio, high-density lipoprotein (HDL), total cholesterol, and glycated hemoglobin (HbA1c). Immune indicators included high-sensitivity C-reactive protein (h-CRP). Neuroendocrine hormones included overnight urine epinephrine, norepinephrine, cortisol, and fasting blood dehydroepiandrosterone (DHEA). Blood samples were collected between 8:00 and 9:00 am after overnight fasting and serum was stored in aliquots at −80°C until assay. For overnight urine, participants emptied their bladder at 7:00 pm, following which all urine was collected in a container that was kept on ice until 7:00 am the next morning. Among female participants, blood and urine samples were collected outside of the menstrual period to reduce potential confounds from sex hormone fluctuations.

Consistent with previous calculations of the AL index,14,21 we identified the 75th percentile (or the 25th percentile for HDL and DHEA) of the 13 biomarkers based upon the levels observed in the healthy controls. Participants who had a biomarker value greater than the 75th percentile of the biomarkers (or less than the 25th percentile for HDL and DHEA) received a score of 1. The actual thresholds used for the thirteen indicators are shown in Table 1. Participants who were on medications for hyperlipidemia, diabetes, and hypertension were automatically given a 1 for the respective component.

Table 1.

Biomarkers Included in the Allostatic Load Index

Mean(SD)
Biomarker FES HC Threshold F Value P Value
Allostatic Load 4.68(1.79) 3.59(2.33) NA 11.87 .001
Cardiovascular
 SBP, mmHg 111.33(10.79) 114.90(13.23) ≥121.5 2.87 .09
 DBP, mmHg 71.77(8.05) 72.24(8.46) ≥80.0 0.01 0.91
 Heart rate, bpm 80.90(5.89) 73.71(9.35) ≥81.50 28.76 <.001*
Metabolic
 BMI, kg/m2 21.44(3.15) 22.88(3.50) ≥24.57 2.39 .13
 Waist–hip ratio 0.86(0.07) 0.86(0.10) ≥0.90 0.17 .68
 HDL, mmol/L 1.34(0.38) 1.35(0.26) ≤1.1 0.003 .96
 Total cholesterol, mmol/L 4.08(0.88) 4.43(0.78) ≥4.91 1.62 .21
 HbA1c, % 5.51(0.88) 5.26(0.28) ≥5.5 4.11 .045
Inflammation
 H-CRP, mg/L 2.34(3.09) 1.31(1.86) ≥1.55 7.21 .008
Stress/Neuroendocrine
 Urine E, μg/g creatine 5.27(6.63) 6.88(7.98) ≥8.99 3.20 .08
 Urine NE, μg/g creatine 35.58(30.10) 37.78(33.98) ≥40.7 0.68 .41
 Urine cortisol, μg/g creatine 23.92(18.11) 15.59(13.06) ≥16.64 7.39 .008
 Blood-DHEA, μg/dl 175.52(141.67) 197.62(153.31) ≤102.59 1.26 .26
Open in a new tab

Note: NA, not applicable; SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; BMI, Body-mass Index; HDL, High-density Lipoprotein; HbA1c, Glycated Hemoglobin; H-CRP, Hypersensitive C-reactive Protein; Urine E, Overnight Urine Epinephrine; Urine NE, Overnight Urine Norepinephrine; Blood-DHEA, Dehydroepiandrosterone. Statistics for each measure was based on raw data.

Threshold: the cut-off value above (or below for HDL and DHEA) which the item was scored as 1 for the calculation of AL.

*Significant after Bonferroni correction, P < .004.

Image Acquisition and Processing

Imaging data were collected at Beijing Huilongguan Hospital magnetic resonance imaging research center with a 3.0-T Prisma MRI scanner (Siemens, Germany) and a 64-channel head coil. Structural T1-weighted images were acquired using a sagittal 3D magnetization-prepared rapid acquisition gradient echo (MP-RAGE) sequence. The scan parameters were as follows: echo time (TE) = 2.98 ms, repetition time (TR) = 2,530 ms, flip angle (FA) = 7°, matrix size = 256 × 224, field-of-view (FOV) = 256 × 224 mm, inversion time (TI) = 1,100 ms, and thickness/gap = 1/0 mm. Volumetric processing was performed using FreeSurfer, version 5.3 (http://surfer.nmr.mgh.harvard.edu/). The volumetric measurements were extracted using a standard procedure that included motion correction, removal of nonbrain tissue using a hybrid watershed/surface deformation procedure, automated Talairach transformation, and segmentation of the subcortical volumetric structures.40,41 Intensity normalization and automated topology corrections were performed. Images were manually checked for quality and inspected for motion correction, and the left and right of the same structures were summed to generate a volume score for each subcortical structure. Subcortical structures including the lateral ventricle, lateral-ventricle choroid plexus, thalamus, amygdala, hippocampus, caudate, putamen, pallidum, accumbens, and the total intracranial volume were extracted for statistical analysis.

All participants completed AL assessment. MRI data in 19 of the 79 patients and 1 of the 41 controls were missing due to refusal or incompletion of MRI. One patient did not complete clinical assessments.

Statistical Methods

Demographic data were compared between the two groups by χ 2 test or t-tests. ANCOVA was used to explore the group differences in AL and volumes of brain regions adjusting for age, sex, and education level. For brain volume data, the covariates also included total intracranial volume. Bonferroni correction was used for multiple comparisons, eg, the significance level in comparing the 13 AL components was P < .004 (.05/13). To account for multiple potential confounds affecting the AL effects on brain regions, for each brain region, multiple regression analyses of the entire sample were performed to include age, sex, education level, intracranial volume, diagnosis (coded as 0 for controls and 1 for patients), AL, and AL × diagnosis interaction as predictors. Bonferroni-corrected P < .0056 (.05/9 brain regions) was used as the significance threshold for all model and predictor statistics. Follow-up explorations between significant AL and brain regions used partial correlation analysis, also co-varying for age, sex, education level, and total intracranial volume. The relationships with clinical symptoms and AL were similarly examined. All tests were two-tailed, with a significance level set at corrected P < .05.

Results

Clinical Characteristics

Demographic and clinical information is presented in Table 2. The FES and HC groups were frequency-matched in age and sex. Chlorpromazine equivalent antipsychotic dosage (CPZ)42,43 in patients was 319.8 ± 209.4 mg/d (mean ± SD), and they were treated for an average of 6.3 ± 2.6 days (range: 0–13 days) at the time of MRI. The interval between AL assessment and MRI was 3.3 ± 1.9 days.

Table 2.

Participant Demographics and Clinical Characteristics

Mean(SD)
FES HC Statistics
(t or χ 2 Value)		 P Value
Sample Size 79 41
Age, year 27.2(7.6) 29.8(6.4) −1.92 .06
Sex, M/F 38/41 21/20 0.11 .75
Education, year 12.9(3.3) 14.0(2.5) −1.93 .037*
Illness Duration, month 11.4(12.4) NA NA NA
PANSS
 Total Score 78.1(12.4) NA NA NA
 Positive Subscale 22.5(5.2) NA NA NA
 Negative Subscale 17.6(6.3) NA NA NA
 General Psychopathology 37.8(6.7) NA NA NA
Open in a new tab

Note: FES, First-Episode Schizophrenia; HC, Healthy Controls; NA, not applicable; PANSS, Positive and Negative Syndrome Scale.

*Significant at P < .05.

AL Index and Individual Components

The AL index was significantly higher in the FES group than in the HC group (F1,115 = 11.87, P = .001). Within the AL composite measure, heart rate (F1,115 = 28.76, P < .001), h-CRP (F1,115 = 7.21, P=.008), HbA1c (F1,115 = 4.11, P = .045), and overnight urine cortisol level (F1,115 = 7.39, P = .008) were nominally and significantly higher in FES. As the four nominally different measures were from the four categories of the AL construct (Table 1), the group differences in AL may not be driven primarily by any one subcomponent.

Males and females did not significantly differ in AL in either patients (P = .13) or controls (P = .20). We also recalculated the AL in males and females separately and found that male-specific AL (P=.031) and female-specific AL (P=.027) were also significantly higher in FES than HC. The relationship between disease duration and AL was not significant (P = .07). As expected, age significantly influenced AL in the HC group (β = 0.45, P = .003), although this relationship was not significant in FES group (P = .61).

Subcortical Brain Regions and Associations With AL

The total intracranial volume in FES and HC groups showed no significant difference (1505.2 ± 127.7 vs 1507.6 ± 146.5 cm3, F1,95 = 0.02, P = .90). The volumes of the lateral ventricle (F1,94 = 14.01, P < .001), CP (F1,94 = 31.26, P < .001), and thalamus (F1,94 = 24.88, P < .001) were significantly larger, whereas amygdala (F1,94 = 10.81, P = .001) was significantly smaller in FES compared with HC (Table 3).

Table 3.

Subcortical Brain Structures Comparisons

Mean(SD) in cm3
Region FES HC F Value P Value
Lateral ventricle 15.72 (6.12) 11.51 (4.30) 14.01 <.001*
Choroid plexus 1.04 (0.31) 0.73 (0.20) 31.26 <.001*
Thalamus 18.65 (2.86) 16.07 (1.91) 24.88 <.001*
Amygdala 3.26 (0.37) 3.48 (0.41) 10.81 .001*
Hippocampus 8.22 (0.68) 8.19 (0.77) 0.13 .72
Caudate 7.60 (0.97) 7.09 (1.06) 5.38 .023
Putamen 10.45 (1.48) 10.58 (1.42) 0.26 .61
Pallidum 4.23 (0.57) 4.12 (0.50) 1.26 .26
Accumbens 0.99 (0.18) 0.99 (0.18) <.001 .99
Open in a new tab

*Significant after Bonferroni correction, P < .0056.

Multiple regression analyses were performed to account for multiple potential confounds affecting the AL effects on each of the nine brain regions (Table 4). There were no significant AL × diagnosis interactions in any model (all P > .0056). Only the higher AL was independently and significantly associated with larger CP (t = 3.60, P < .001), in addition to a significant diagnosis effect on CP (t = 4.58, P < .001). The regression analyses of AL with other brain regions were not significant after these other covariates were accounted for (Table 4). Collinearity was examined using variance inflation factor (VIF), and all VIF values were less than 1.5 (range: 1.0–1.4), suggesting acceptable collinearity in all of the models.

Table 4.

Multiple Regression Analysis Results of AL on Brain Regions

Model Age Sex Education ICV Dx AL AL×Dx
F Value t Value t Value t Value t Value t Value t Value t Value
Region P Value P Value P Value P Value P Value P Value P Value P Value
Lateral ventricle 10.92
<.001*		 1.05
.30		 0.32
.75		 −1.84
.07		 3.82
<.001*		 3.62
<.001*		 1.80
.08		 1.35
0.18
Choroid plexus 16.82
<.001*		 0.31
.76		 −0.68
.50		 −2.45
.02		 3.19
.002*		 4.58
<.001*		 3.60
<.001**		 1.11
.27
Thalamus 17.81
<.001*		 −0.91
.36		 −0.31
.76		 −0.02
.99		 4.55
<.001*		 5.15
<.001*		 0.45
.65		 −0.19
.85
Amygdala 28.76
<.001*		 −1.58
.12		 −0.97
.34		 2.50
.01		 7.78
<.001*		 −3.00
.003*		 0.20
.84		 0.55
.58
Hippocampus 15.08
<.001*		 −1.37
.17		 −2.67
.009		 2.18
.03		 3.32
.001*		 0.71
.48		 1.78
.08		 1.60
.11
Caudate 15.20
<.001*		 −2.56
.01		 1.45
.15		 1.00
.32		 5.16
<.001*		 2.22
.03		 0.27
.79		 1.63
.11
Putamen 47.95
<.001*		 −0.83
.41		 −0.39
.70		 0.89
.37		 6.92
<.001*		 −0.47
.64		 −0.40
.69		 −0.39
.69
Pallidum 26.05
<.001*		 −1.11
.27		 −0.52
.60		 0.93
.36		 5.10
<.001*		 1.14
.26		 0.71
.48		 1.36
.18
Accumbens 20.07
<.001*		 −2.14
.04		 −1.37
.17		 2.49
.01		 6.32
<.001*		 −0.07
.95		 −0.24
.81		 0.16
.87
Open in a new tab

Note: ICV, Intracranial Volume; Dx, Diagnosis; AL×Dx, AL×Diagnosis Interaction.

Each row represented a multiple linear regression analysis model of different brain regions, data on the final models were presented.

*Significant at P < .0056 to correct for 9 modeling analyses.

**Significantly associated with AL after other potential confounds accounted for and P < .0056 to correct for nine modeling analyses.

Partial correlations were used to further explore the relationships between subcortical brain regions and AL in separate groups (Figures 1A and 1B). Only the CP volume was significantly correlated with AL in FES group (r = 0.34, P = .01); this effect was also present in the controls but to a weaker degree (r = 0.27, P = .11). These r values were not significantly different (z = 0.37, P = .71). AL was significantly correlated with CP volume in the combined sample (r = 0.42, P < .001; Figure 2C). As expected, the CP volume was positively correlated with lateral ventricle size in both FES and HC groups (r = 0.37, P = .005; r = 0.66, P < .001, respectively), but the lateral ventricle size was not significantly correlated with AL (r = 0.19, P = .17 in patients and r = 0.10, P = .56 in controls).

Fig. 1.

Fig. 1.

Open in a new tab

Partial correlations between allostatic load (AL) and the subcortical regional structures. (A) Partial correlations in patients. (B) Partial correlations in healthy controls. Y-axis: Partial correlation coefficients. *Nominally significant at P < .05 after adjusting for age, sex, education, and total intracranial volume.

Fig. 2.

Fig. 2.

Open in a new tab

Correlations between choroid plexus (CP) volume and the components of allostatic load (AL). (A) Partial correlations between CP volume and the components of AL in patients. (B) Partial correlations between CP volume and the components of AL in healthy controls. Y-axis: Partial correlation coefficients. *Nominally significant at P < .05 after adjusting for age, sex, education level, and total intracranial volume; none of these correlations was significant after Bonferroni correction of 13 comparisons at P < .004. (C) Scatter plot of correlations between AL and CP volume in FES, HC, and the combined sample. (D) Location and shape of a typical lateral-ventrical CP.

There is also the possibility that even brief exposure to antipsychotics may have increased the CP volume. We correlated days of exposure (0–13 days) and lifetime total cumulative CPZ exposure (0–7731.1 mg) with CP volume and found none to be significant (r = 0.0 and 0.14, P = 1.0 and .36, respectively).

On specific component(s) within the AL construct, higher overnight urine epinephrine (r = 0.30, P = .008), and norepinephrine (r = 0.35, P = .008) were nominally associated with larger CP volume in FES, whereas diastolic blood pressure (r = 0.42, P = .012) and BMI (r = 0.34, P = .044) were nominally correlated with CP volume in HC (Figures 2A and 2B), none was significant after Bonferroni correction.

Associations Between AL and Symptom Severity and Antipsychotics

Higher AL was significantly correlated with more positive symptoms (r = 0.28, P = .016) but less negative symptoms (r = −0.31, P = .008). There was no significant association with the PANSS general psychopathology (r = 0.23, P = .052) or total score (r = 0.09, P = .46). CPZ was not correlated with AL (P = .47), heart rate (P = .22), HbA1c (P = .75), h-CRP (P = .34), or overnight urine cortisol level (P = .35). As anticholingergic drugs may increase heart rate, we also compared seven patients on anticholinergic drugs to the rest of the patients, and the difference in heart rate was not significant (81.0 ± 5.7 vs 80.3 ± 8.7, t = 0.30, P = .77).

Discussion

First-episode patients within 2 weeks of treatment initiation already had significantly larger choroid plexus volume compared with controls, and higher allostatic load was further significantly related to larger choroid plexus. This is the largest study of AL in schizophrenia, using a balanced strategy to minimize antipsychotic medication exposure effects while reducing the possible effects of acute distress on AL measurement due to untreated psychosis. The larger CP at this early stage of the illness appeared not primarily driven by any one subcomponent within the AL index. The results identify new empirical evidence to support that a relationship between brain structural abnormalities and multisystemic pathophysiology is already present at the initiation of treatment in schizophrenia.

CP is one of the main structures involved in CNS-periphery exchanges. CP epithelial cells and the connecting tight junctions form the blood-cerebrospinal fluid barrier (BCSFB), and together with the endothelial system of the blood-brain barrier (BBB), are the primary CNS-circulation interfaces responsible for maintaining the homeostasis of the brain microenvironment.44 The epithelium expresses corticotrophin-releasing hormone receptors,45 receives abundant sympathetic innervations, and releases noradrenaline in response to stimulations.46 Peripheral proinflammatory cytokines and infection affect the tight junction proteins and the epithelial cells of the CP47,48; and CP also regulate stress/inflammatory responses31 in both normal29 and traumatic conditions.31 As CP is closely involved in the regulation of these stress, inflammatory, neurotrophic, and neuroendocrinological processes, which are important mediators of the allostatic system,12,49 it may not be surprising that higher AL in schizophrenia is associated with larger CP.

The multiple regression analysis showed that larger volume of CP is associated with schizophrenia independent of and in addition to the elevated AL (Table 4). The reason is unknown, although we speculate that the current AL measure may not capture all the variance contributing to abnormal CP in schizophrenia. Besides interaction with the peripheral circulation through its basal side, CP epithelium is exposed to neural signaling through the CSF at its apical side. CP is critically regulated by neuronal activities and involved in neuronal repair and restoration.50,51 Active psychosis itself has been proposed as potentially biologically “toxic” to the brain52,53 and may increase neuroinflammation,54 which may require hyperactivity of the CP for their clearance55 and thus the larger CP seen in patients. Furthermore, as lateral ventricular CP is necessary for supporting adult neurogenesis at the subventricular neural stem cells,26 it is unclear whether CP abnormality in the early stage of the illness may be related to abnormal adult neurogenesis proposed to be associated with psychosis.56 CP is also shown to be the main site for harboring CNS T-cells, which can be neuroprotective and control excessive neuroinflammation at diseased CNS conditions.57 If this neuroprotective mechanism was also present at the early stage of psychosis, it could potentially activate CP leading to the enlarged CP. However, neuroimaging does not directly measure CP cellular morphology, for which postmortem studies become valuable. In the only postmortem study of CP of schizophrenia that we are aware of, proinflammatory gene expressions in CP were up-regulated in schizophrenia compared with that of controls, which were positively correlated with postmortem serum immune and neuroendocrine markers,58 in part supporting our in vivo observations. Based on our data alone, it is unclear whether the enlarged CP volume is a cause or an effect in patients with schizophrenia. It is possible that this may occur prior to the first episode of schizophrenia. If that in fact occurs, CP assessment may provide a useful early marker of risk for schizophrenia in clinical high-risk individuals for guiding prevention and early interventions efforts.

Besides CP, the volumes of lateral ventricle and thalamus were larger, and the amygdala was smaller in FES compared with HC (Table 3). Larger ventricular volumes in FES patients have been previously reported,35 with likely progressive ventricular enlargement in schizophrenia.59,60 Importantly, AL was significantly associated only with lateral ventricular CP but not lateral ventricular volume itself. We did not find significantly smaller hippocampal volume in FES. Meta-analysis of studies in (mostly chronic) schizophrenia had reported smaller of hippocampus, amygdala, and thalamus.61 However, whether smaller hippocampal volumes had already been present in FES is still controversial.62,63 Overall, at this early stage, the associations of AL with other subcortical structures appear not as strong as with CP.

Resting heart rate, HbA1c, h-CRP, and overnight urine cortisol levels were also found to be nominally higher in FES compared with HC. High cortisol levels have been associated with psychotic symptoms64; h-CRP is a predictor of cardiovascular events and is elevated in schizophrenia65–67; higher HbA1c in FES has been observed6; and increased heart rate in patients may be due to dysregulation of the autonomic nervous system68 or use of medications,69 although our analysis did not support the medication contributions. Collectively, these findings may support the notion that patients with schizophrenia experience multisystem subclinical dysfunction even around the time of disease onset.1

At present, there is no consensus regarding the specific biomarkers the AL index should include.18 The measures included in the current study may not capture all the effects of cumulative stress the AL concept attempts to encompass, which is a limitation. There was also a considerable difference in the number of markers representing each subsystem in our current approach that followed the original AL study,70 which is another limitation. For instance, h-CRP was the only marker of inflammation but metabolic dysregulations were represented by several biomarkers. Additional inflammation measures can be readily added, but as different studies may add different potential combinations of measures, comparisons of findings across studies may become challenging. What would be the optimal combination of measures to index AL may in part depend on research progress.

AL was associated with severity of positive symptoms, in agreement with previous research showing that elevated stress sensitivity is related to psychotic experiences.71 The inverse correlation between AL and negative symptoms is more puzzling. Negative symptoms measured at the first episode of psychosis may not be as stable compared with the chronic phase of illness.72

The cross-sectional nature of the study limits our ability to interpret potential causal pathways underlying the observed associations. Longitudinal analyses are needed to ascertain whether CP enlargement and higher AL at the early stage of the disorder predict the alarmingly high rates of medical comorbidity and mortality in schizophrenia. Antipsychotic medication exposure can be a prominent confounding factor; however, our data do not support this at this stage of the illness. This is consistent with another study in which high AL in unmedicated patients was normalized after 6–12 weeks of antipsychotic treatment, suggesting that antipsychotic medications may not increase AL in this treatment duration.18 Finally, as CSF is the primary product of CP, the lack of CSF assessment remains an important limitation and our finding suggests a need to include CSF investigations in future CP-related studies in schizophrenia.

In summary, this study showed that choroid plexus may be a bellwether connection between heightened peripheral multisystem subclinical abnormalities and the clinical brain dysfunction at the early stage of schizophrenia. The finding is novel in suggesting that non-CNS subclinical comorbidity of psychosis73 can be linked to specific brain structural assessments. The CP is infrequently investigated in schizophrenia etiopathophysiological research, but may have an important role in understanding schizophrenia as a multisystem illness.

Acknowledgments

We are grateful to all patients and healthy volunteers who participated in this research. Supports were received from the National Key R & D Program of China (grant number 2016YFC1307000), National Natural Science Foundation of China (grant numbers 81761128021 and 81771452), and the National Institutes of Health (grant numbers R01MH112180 and R01MH116948).

Conflict of Interest Statement

Dr. Hong has received or is planning to receive research funding or consulting fees from Mitsubishi, Your Energy Systems LLC, Neuralstem, Taisho, Heptares, Pfizer, Sound Pharma, Luye Pharma, Takeda, and Regeneron. Other authors declare that the research was conducted without any relationship that could be interpreted as a potential conflict of interest or financial conflict.

References

  • 1. Pillinger T, D’Ambrosio E, McCutcheon R, Howes OD. Is psychosis a multisystem disorder? A meta-review of central nervous system, immune, cardiometabolic, and endocrine alterations in first-episode psychosis and perspective on potential models. Mol Psychiatry. 2019;24(6):776–794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Mitchell AJ, Vancampfort D, Sweers K, et al. Prevalence of metabolic syndrome and metabolic abnormalities in schizophrenia and related disorders–a systematic review and meta-analysis. Schizophr Bull. 2013;39(2):306–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Park YM, Lee SH, Lee BH, et al.. Prolactin and macroprolactin levels in psychiatric patients receiving atypical antipsychotics: a preliminary study. Psychiatry Res. 2016;239:184–189. [DOI] [PubMed] [Google Scholar]
  • 4. De Hert M, Detraux J, van Winkel R, Yu W, Correll CU. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nat Rev Endocrinol. 2011;8(2):114–126. [DOI] [PubMed] [Google Scholar]
  • 5. Pillinger T, Beck K, Gobjila C, et al. Impaired glucose homeostasis in first-episode schizophrenia: a systematic review and meta-analysis. JAMA Psychiatry. 2017;74(3):261–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Zhang XY, Chen DC, Tan YL, et al.. Glucose disturbances in first-episode drug-naïve schizophrenia: relationship to psychopathology. Psychoneuroendocrinology. 2015;62:376–380. [DOI] [PubMed] [Google Scholar]
  • 7. Guest PC, Wang L, Harris LW, et al.. Increased levels of circulating insulin-related peptides in first-onset, antipsychotic naïve schizophrenia patients. Mol Psychiatry. 2010;15(2):118–119. [DOI] [PubMed] [Google Scholar]
  • 8. Corcoran C, Walker E, Huot R, et al.. The stress cascade and schizophrenia: etiology and onset. Schizophr Bull. 2003;29(4):671–692. [DOI] [PubMed] [Google Scholar]
  • 9. Seeman TE, McEwen BS, Rowe JW, Singer BH. Allostatic load as a marker of cumulative biological risk: macArthur studies of successful aging. Proc Natl Acad Sci U S A. 2001;98(8):4770–4775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Misiak B, Frydecka D, Zawadzki M, Krefft M, Kiejna A. Refining and integrating schizophrenia pathophysiology – relevance of the allostatic load concept. Neurosci Biobehav Rev. 2014;45:183–201. [DOI] [PubMed] [Google Scholar]
  • 11. Sterling P, Eyer J. Allostasis: a new paradigm to explain arousal pathology. In: Fisher S, Reason J, eds. Handbook of Life Stress, Cognition and Health. New York, NY: John Wiley, Sons; 1988: 629–649. [Google Scholar]
  • 12. McEwen BS. Protection and damage from acute and chronic stress: allostasis and allostatic overload and relevance to the pathophysiology of psychiatric disorders. Ann N Y Acad Sci. 2004;1032:1–7. [DOI] [PubMed] [Google Scholar]
  • 13. Nugent KL, Chiappelli J, Rowland LM, Hong LE. Cumulative stress pathophysiology in schizophrenia as indexed by allostatic load. Psychoneuroendocrinology. 2015;60:120–129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Bizik G, Picard M, Nijjar R, et al.. Allostatic load as a tool for monitoring physiological dysregulations and comorbidities in patients with severe mental illnesses. Harv Rev Psychiatry. 2013;21(6):296–313. [DOI] [PubMed] [Google Scholar]
  • 15. Gruenewald TL, Mroczek DK, Ryff CD, Singer BH. Diverse pathways to positive and negative affect in adulthood and later life: an integrative approach using recursive partitioning. Dev Psychol. 2008;44(2):330–343. [DOI] [PubMed] [Google Scholar]
  • 16. Pettorruso M, De Risio L, Di Nicola M, et al. Allostasis as a conceptual framework linking bipolar disorder and addiction. Front Psychiatry. 2014;5:173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Savransky A, Chiappelli J, Fisseha F, et al.. Elevated allostatic load early in the course of schizophrenia. Transl Psychiatry. 2018;8(1):246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Berger M, Juster RP, Westphal S, et al.. Allostatic load is associated with psychotic symptoms and decreases with antipsychotic treatment in patients with schizophrenia and first-episode psychosis. Psychoneuroendocrinology. 2018;90:35–42. [DOI] [PubMed] [Google Scholar]
  • 19. Misiak B, Kotowicz K, Loska O, et al.. Decreased use of active coping styles contributes to elevated allostatic load index in first-episode psychosis. Psychoneuroendocrinology. 2018;96:166–172. [DOI] [PubMed] [Google Scholar]
  • 20. Misiak B, Kotowicz K, Loska O, et al.. Elevated allostatic load index is associated with working memory deficits in first-episode psychosis. Schizophr Res. 2019;204:439–441. [DOI] [PubMed] [Google Scholar]
  • 21. Chiappelli J, Kochunov P, Savransky A, et al.. Allostatic load and reduced cortical thickness in schizophrenia. Psychoneuroendocrinology. 2017;77:105–111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Savransky A, Chiappelli J, Rowland LM, et al.. Fornix structural connectivity and allostatic load: empirical evidence from schizophrenia patients and healthy controls. Psychosom Med. 2017;79(7):770–776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. McEwen BS, Nasca C, Gray JD. Stress effects on neuronal structure: hippocampus, amygdala, and prefrontal cortex. Neuropsychopharmacology. 2016;41(1):3–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. McCarty R. Learning about stress: neural, endocrine and behavioral adaptations. Stress. 2016;19(5):449–475. [DOI] [PubMed] [Google Scholar]
  • 25. Spector R, Keep RF, Robert Snodgrass S, Smith QR, Johanson CE. A balanced view of choroid plexus structure and function: focus on adult humans. Exp Neurol. 2015;267:78–86. [DOI] [PubMed] [Google Scholar]
  • 26. Silva-Vargas V, Maldonado-Soto AR, Mizrak D, Codega P, Doetsch F. Age-dependent niche signals from the choroid plexus regulate adult neural stem cells. Cell Stem Cell. 2016;19(5):643–652. [DOI] [PubMed] [Google Scholar]
  • 27. Marques F, Sousa JC, Coppola G, et al.. The choroid plexus response to a repeated peripheral inflammatory stimulus. BMC Neurosci. 2009;10:135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Johanson C, Johanson N. Merging transport data for choroid plexus with blood-brain barrier to model CNS homeostasis and disease more effectively. CNS Neurol Disord Drug Targets. 2016;15(9):1151–1180. [DOI] [PubMed] [Google Scholar]
  • 29. Szmydynger-Chodobska J, Strazielle N, Zink BJ, Ghersi-Egea JF, Chodobski A. The role of the choroid plexus in neutrophil invasion after traumatic brain injury. J Cereb Blood Flow Metab. 2009;29(9):1503–1516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Xiang J, Routhe LJ, Wilkinson DA, et al.. The choroid plexus as a site of damage in hemorrhagic and ischemic stroke and its role in responding to injury. Fluids Barriers CNS. 2017;14(1):8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Szmydynger-Chodobska J, Strazielle N, Gandy JR, et al.. Posttraumatic invasion of monocytes across the blood-cerebrospinal fluid barrier. J Cereb Blood Flow Metab. 2012;32(1):93–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Murphy VA, Johanson CE. Adrenergic-induced enhancement of brain barrier system permeability to small nonelectrolytes: choroid plexus versus cerebral capillaries. J Cereb Blood Flow Metab. 1985;5(3):401–412. [DOI] [PubMed] [Google Scholar]
  • 33. Redzic ZB, Segal MB. The structure of the choroid plexus and the physiology of the choroid plexus epithelium. Adv Drug Deliv Rev. 2004;56(12):1695–1716. [DOI] [PubMed] [Google Scholar]
  • 34. Sandyk R. Choroid plexus calcification as a possible marker of hallucinations in schizophrenia. Int J Neurosci. 1993;71(1-4):87–92. [DOI] [PubMed] [Google Scholar]
  • 35. Steen RG, Mull C, McClure R, Hamer RM, Lieberman JA. Brain volume in first-episode schizophrenia: systematic review and meta-analysis of magnetic resonance imaging studies. Br J Psychiatry. 2006;188:510–518. [DOI] [PubMed] [Google Scholar]
  • 36. Palha JA, Santos NC, Marques F, et al.. Do genes and environment meet to regulate cerebrospinal fluid dynamics? Relevance for schizophrenia. Front Cell Neurosci. 2012;6:31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Zaletel I, Filipović D, Puškaš N. Chronic stress, hippocampus and parvalbumin-positive interneurons: what do we know so far? Rev Neurosci. 2016;27(4):397–409. [DOI] [PubMed] [Google Scholar]
  • 38. Demeestere D, Libert C, Vandenbroucke RE. Therapeutic implications of the choroid plexus-cerebrospinal fluid interface in neuropsychiatric disorders. Brain Behav Immun. 2015;50:1–13. [DOI] [PubMed] [Google Scholar]
  • 39. Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull. 1987;13(2):261–276. [DOI] [PubMed] [Google Scholar]
  • 40. Ségonne F, Dale AM, Busa E, et al.. A hybrid approach to the skull stripping problem in MRI. Neuroimage. 2004;22(3):1060–1075. [DOI] [PubMed] [Google Scholar]
  • 41. Fischl B, Salat DH, Busa E, et al.. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron. 2002;33(3):341–355. [DOI] [PubMed] [Google Scholar]
  • 42. Andreasen NC, Pressler M, Nopoulos P, Miller D, Ho BC. Antipsychotic dose equivalents and dose-years: a standardized method for comparing exposure to different drugs. Biol Psychiatry. 2010;67(3):255–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Gardner DM, Murphy AL, O’Donnell H, Centorrino F, Baldessarini RJ. International consensus study of antipsychotic dosing. Am J Psychiatry. 2010;167(6):686–693. [DOI] [PubMed] [Google Scholar]
  • 44. Kaur C, Rathnasamy G, Ling EA. The choroid plexus in healthy and diseased brain. J Neuropathol Exp Neurol. 2016;75(3):198–213. [DOI] [PubMed] [Google Scholar]
  • 45. Wong ML, Licinio J, Pasternak KI, Gold PW. Localization of corticotropin-releasing hormone (CRH) receptor mRNA in adult rat brain by in situ hybridization histochemistry. Endocrinology. 1994;135(5):2275–2278. [DOI] [PubMed] [Google Scholar]
  • 46. Nilsson C, Kannisto P, Lindvall-Axelsson M, Owman C, Rosengren E. The neuropeptides vasoactive intestinal polypeptide, peptide histidine isoleucine and neuropeptide Y modulate [3H]noradrenaline release from sympathetic nerves in the choroid plexus. Eur J Pharmacol. 1990;181(3):247–252. [DOI] [PubMed] [Google Scholar]
  • 47. Coisne C, Engelhardt B. Tight junctions in brain barriers during central nervous system inflammation. Antioxid Redox Signal. 2011;15(5):1285–1303. [DOI] [PubMed] [Google Scholar]
  • 48. Cardia E, Molina D, Abbate F, et al.. Morphological modifications of the choroid plexus in a rodent model of acute ventriculitis induced by gram-negative liquoral sepsis. possible implications in the pathophysiology of hypersecretory hydrocephalus. Childs Nerv Syst. 1995;11(9):511–516. [DOI] [PubMed] [Google Scholar]
  • 49. McEwen BS, Wingfield JC. The concept of allostasis in biology and biomedicine. Horm Behav. 2003;43(1):2–15. [DOI] [PubMed] [Google Scholar]
  • 50. Marques F, Sousa JC, Brito MA, et al.. The choroid plexus in health and in disease: dialogues into and out of the brain. Neurobiol Dis. 2017;107:32–40. [DOI] [PubMed] [Google Scholar]
  • 51. Baruch K, Deczkowska A, David E, et al.. Aging. aging-induced type I interferon response at the choroid plexus negatively affects brain function. Science. 2014;346(6205):89–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Rund BR. Does active psychosis cause neurobiological pathology? A critical review of the neurotoxicity hypothesis. Psychol Med. 2014;44(8):1577–1590. [DOI] [PubMed] [Google Scholar]
  • 53. Andreasen NC, Liu D, Ziebell S, Vora A, Ho BC. Relapse duration, treatment intensity, and brain tissue loss in schizophrenia: a prospective longitudinal MRI study. Am J Psychiatry. 2013;170(6):609–615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Orlovska-Waast S, Köhler-Forsberg O, Brix SW, et al.. Cerebrospinal fluid markers of inflammation and infections in schizophrenia and affective disorders: a systematic review and meta-analysis. Mol Psychiatry. 2019;24(6):869–887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Strazielle N, Khuth ST, Ghersi-Egea JF. Detoxification systems, passive and specific transport for drugs at the blood-CSF barrier in normal and pathological situations. Adv Drug Deliv Rev. 2004;56(12):1717–1740. [DOI] [PubMed] [Google Scholar]
  • 56. Iannitelli A, Quartini A, Tirassa P, Bersani G. Schizophrenia and neurogenesis: a stem cell approach. Neurosci Biobehav Rev. 2017;80:414–442. [DOI] [PubMed] [Google Scholar]
  • 57. Schwartz M, Baruch K. The resolution of neuroinflammation in neurodegeneration: leukocyte recruitment via the choroid plexus. EMBO J. 2014;33(1):7–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58. Kim S, Hwang Y, Lee D, Webster MJ. Transcriptome sequencing of the choroid plexus in schizophrenia. Transl Psychiatry. 2016;6(11):e964. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59. Saijo T, Abe T, Someya Y, et al.. Ten year progressive ventricular enlargement in schizophrenia: an MRI morphometrical study. Psychiatry Clin Neurosci. 2001;55(1):41–47. [DOI] [PubMed] [Google Scholar]
  • 60. DeLisi LE, Sakuma M, Maurizio AM, Relja M, Hoff AL. Cerebral ventricular change over the first 10 years after the onset of schizophrenia. Psychiatry Res. 2004;130(1):57–70. [DOI] [PubMed] [Google Scholar]
  • 61. van Erp TG, Hibar DP, Rasmussen JM, et al. Subcortical brain volume abnormalities in 2028 individuals with schizophrenia and 2540 healthy controls via the ENIGMA consortium. Mol Psychiatry. 2016;21(4):547–553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62. Bois C, Levita L, Ripp I, et al.. Hippocampal, amygdala and nucleus accumbens volume in first-episode schizophrenia patients and individuals at high familial risk: a cross-sectional comparison. Schizophr Res. 2015;165(1):45–51. [DOI] [PubMed] [Google Scholar]
  • 63. Scanlon C, Anderson-Schmidt H, Kilmartin L, et al.. Cortical thinning and caudate abnormalities in first episode psychosis and their association with clinical outcome. Schizophr Res. 2014;159(1):36–42. [DOI] [PubMed] [Google Scholar]
  • 64. Dinan TG. Stress and the genesis of diabetes mellitus in schizophrenia. Br J Psychiatry Suppl. 2004;47:S72–S75. [DOI] [PubMed] [Google Scholar]
  • 65. Balõtšev R, Koido K, Vasar V, et al.. Inflammatory, cardio-metabolic and diabetic profiling of chronic schizophrenia. Eur Psychiatry. 2017;39:1–10. [DOI] [PubMed] [Google Scholar]
  • 66. Danesh J, Whincup P, Walker M, et al.. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. BMJ. 2000;321(7255):199–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67. Fernandes BS, Steiner J, Bernstein HG, et al.. C-reactive protein is increased in schizophrenia but is not altered by antipsychotics: meta-analysis and implications. Mol Psychiatry. 2016;21(4):554–564. [DOI] [PubMed] [Google Scholar]
  • 68. Reinertsen E, Shashikumar SP, Shah AJ, Nemati S, Clifford GD. Multiscale network dynamics between heart rate and locomotor activity are altered in schizophrenia. Physiol Meas. 2018;39(11):115001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69. Tajiri M, Suzuki Y, Sugai T, Tsuneyama N, Someya T. Effects of olanzapine on resting heart rate in Japanese patients with schizophrenia. PLoS One. 2018;13(7):e0199922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70. Seeman TE, Singer BH, Rowe JW, Horwitz RI, McEwen BS. Price of adaptation–allostatic load and its health consequences. MacArthur studies of successful aging. Arch Intern Med. 1997;157(19):2259–2268. [PubMed] [Google Scholar]
  • 71. Reininghaus U, Kempton MJ, Valmaggia L, et al.. Stress sensitivity, aberrant salience, and threat anticipation in early psychosis: an Experience Sampling Study. Schizophr Bull. 2016;42(3):712–722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72. Mezquida G, Cabrera B, Bioque M, et al. ; PEPs Group. The course of negative symptoms in first-episode schizophrenia and its predictors: a prospective two-year follow-up study. Schizophr Res. 2017;189:84–90. [DOI] [PubMed] [Google Scholar]
  • 73. Dantzer R. Cytokine-induced sickness behaviour: a neuroimmune response to activation of innate immunity. Eur J Pharmacol. 2004;500(1-3):399–411. [DOI] [PubMed] [Google Scholar]

Articles from Schizophrenia Bulletin are provided here courtesy of Oxford University Press

RESOURCES