Molecular Psychiatry (2016) 21, 554–564

© 2016 Macmillan Publishers Limited All rights reserved 1359-4184/16

www.nature.com/mp

ORIGINAL ARTICLE
C-reactive protein is increased in schizophrenia but is not
altered by antipsychotics: meta-analysis and implications
BS Fernandes1,2, J Steiner3, H-G Bernstein3, S Dodd1,4, JA Pasco1,5, OM Dean1,4, P Nardin2, C-A Gonçalves2 and M Berk1,4

The inflammatory hypothesis of schizophrenia (SZ) posits that inflammatory processes and neural–immune interactions are
involved in its pathogenesis, and may underpin some of its neurobiological correlates. SZ is the psychiatric disorder causing the
most severe burden of illness, not just owing to its psychiatric impairment, but also owing to its significant medical comorbidity.
C-reactive protein (CRP) is a commonly used biomarker of systemic inflammation worldwide. There are some conflicting results
regarding the behaviour of CRP in SZ. The aims of this study were to verify whether peripheral CRP levels are indeed increased in
SZ, whether different classes of antipsychotics divergently modulate CRP levels and whether its levels are correlated with positive
and negative symptomatology. With that in mind, we performed a meta-analysis of all cross-sectional studies of serum and plasma
CRP levels in SZ compared to healthy subjects. In addition, we evaluated longitudinal studies on CRP levels before and after
antipsychotic use. Our meta-analyses of CRP in SZ included a total of 26 cross-sectional or longitudinal studies comprising 85 000
participants. CRP levels were moderately increased in persons with SZ regardless of the use of antipsychotics and did not change
between the first episode of psychosis and with progression of SZ (g = 0.66, 95% confidence interval (95% CI) 0.43 to 0.88,
P o 0.001, 24 between-group comparisons, n = 82 962). The extent of the increase in peripheral CRP levels paralleled the increase in
severity of positive symptoms, but was unrelated to the severity of negative symptoms. CRP levels were also aligned with an
increased body mass index. Conversely, higher age correlated with a smaller difference in CRP levels between persons with SZ and
controls. Furthermore, CRP levels did not increase after initiation of antipsychotic medication notwithstanding whether these were
typical or atypical antipsychotics (g = 0.01, 95% CI − 0.20 to 0.22, P = 0.803, 8 within-group comparisons, n = 713). In summary, our
study provides further evidence of the inflammatory hypothesis of SZ. Whether there is a causal relationship between higher CRP
levels and the development of SZ and aggravation of psychotic symptoms, or whether they are solely a marker of systemic low-
grade inflammation in SZ, remains to be clarified.

Molecular Psychiatry (2016) 21, 554–564; doi:10.1038/mp.2015.87; published online 14 July 2015

INTRODUCTION inflammatory biomarkers in the peripheral blood. Interestingly, a

The inflammatory hypothesis of major psychiatric disorders,
such as schizophrenia (SZ) and mood disorders, posits that
inflammatory processes and neural–immune interactions are
involved in the pathogenesis of psychiatric conditions and
may underpin some of their neurobiological correlates.1,2 It has
attracted great interest in recent years, with an increasing
body of evidence providing support for the existence of
central and peripheral immune and inflammatory biomarker
alterations in numerous neuropsychiatric conditions, addi-
tionally supporting the view of these pathologies as systemic
disorders.3–5
SZ is the psychiatric disorder causing the most severe burden
of illness for the individual, not just due to its significant
psychiatric, cognitive and social impairments but also due to its
significant medical comorbidity.6 It is associated with increased
odds for obesity, diabetes, atherosclerosis, osteoporosis and
cardiovascular diseases.6,7 Individuals with autoimmune diseases
and severe infections have persistent or acutely elevated levels of
large epidemiologic study identified autoimmune diseases and
infections as risk factors for developing SZ.8,9 Prenatal infection
appears to increase the risk for the development of SZ and it is
hypothesized that this is predicated on persistent inflammatory
upregulation.10 Similarly, many of the other known risk factors
for psychiatric disorders, as diverse as stress, poor diet, smoking,
physical inactivity and childhood abuse, are transduced via
inflammatory signalling.3 Moreover, several cross-sectional studies
and an early meta-analysis found increased inflammatory bio-
markers, mostly cytokines, in the plasma and the serum of
subjects with SZ.11 Some inflammatory cytokines,14 in particular
interleukin 6 (IL-6) and interleukin 1β (IL-1β), have also been
suggested to represent state markers for acute exacerbations of
psychosis.11 Taken together, these findings suggest that immune
dysregulation may indeed be involved in the pathogenesis of SZ,
and that SZ is a systemic disease with an inflammatory
component (at least in patient subgroups) that could at least in
part be responsible for common risk pathways in SZ and obesity.

1
IMPACT Strategic Research Centre, Barwon Health, Deakin University, School of Medicine, Geelong, VIC, Australia; 2Laboratory of Calcium Binding Proteins in the Central Nervous
System, Department of Biochemistry, Federal University of Rio Grande do Sul, UFRGS, Porto Alegre, Brazil; 3Department of Psychiatry, University of Magdeburg, Magdeburg,
Germany; 4Department of Psychiatry and Orygen, Florey Institute for Neuroscience and Mental Health, The National Centre of Excellence in Youth Mental Health, University of
Melbourne, Parkville, VIC, Australia and 5Department of Medicine, NorthWest Academic Centre, University of Melbourne, St Albans, VIC, Australia. Correspondence:
Dr BS Fernandes, IMPACT Strategic Research Centre, Barwon Health, Deakin University, School of Medicine, Geelong, VIC, Australia.
E-mail: brisasf@gmail.com
Received 10 March 2015; revised 1 May 2015; accepted 26 May 2015; published online 14 July 2015

Although much attention has been paid to the study of cyto-
kines, other biomarkers of inflammation, in particular C-reactive
protein (CRP), could also play a role in the context of SZ. CRP
possesses specific advantages over cytokines as biomarkers. CRP is
one of the most commonly used biomarkers of inflammation in
clinical practice. It is not just used as a measurement of systemic
inflammation mirroring inflammatory conditions in studies
devoted to unveil altered pathophysiological states; it is also the
most employed laboratory test in clinical settings worldwide, to
assess systemic, peripheral inflammatory status. High-sensitivity
CRP is the most sensitive assay for assessing CRP levels.12 Usually,
in apparently healthy subjects, CRP levels are lower than 3 mg l − 1,
but slightly increased levels can occur in normal individuals. These
indicate a state of low-grade inflammation, which is frequently
associated with obesity and metabolic syndrome, and predicts an
increased risk of depression13 and cardiovascular diseases—
common comorbidities in persons with SZ.14 CRP is mainly
produced by hepatocytes and is directly modulated by IL-6 and
IL-1β, both of which appear to be increased particularly during
exacerbations of psychotic status.11,15,16 This raises the possibility
that CRP, in addition to inflammatory cytokines, may play a role in
the inflammatory hypothesis of SZ and therefore be involved in its
pathophysiology in a similar manner.
Several studies have explored CRP in SZ; however, the results
are inconsistent, with approximately half of the studies reporting
an association between CRP levels and SZ, whereas the other
half reported no association.14,17–41 Furthermore, the results are
constrained by diverging methodological characteristics among
the different reports, such as small sample sizes and presence of
systemic clinical diseases (for instance, diabetes and atherosclero-
sis). It is also controversial whether CRP is increased in SZ per se
or whether it is merely a marker of the associated systemic
inflammation commonly found in this disorder, due to increased
prevalence of obesity and metabolic syndrome, among others.
Importantly, it is debatable whether CRP levels are correlated
with severity of positive and negative symptoms in SZ, which is
paramount to reveal a relationship between CRP and the
pathophysiology of SZ.
To make the field even more confusing, many antipsychotics,
the first and sole line of medication to treat psychosis, are
habitually responsible for obesity, consequent insulin resistance
and metabolic syndrome, possibly contributing to the increased
cardiovascular mortality in this population.22,40 Once more, the
body of evidence here is contradictory, with some longitudinal
clinical studies showing worsening of metabolic syndrome and,
ultimately, increased levels of CRP after initiation of antipsychotics,
some showing no change and some showing decreased levels of
CRP after initiation of these medications.20–23,26,27,32,37,40 On the
other hand, animal studies suggest that antipsychotics may
decrease IL-6 and, as a theoretical consequence, negatively
modulate CRP.42,43
Inconsistent findings among the studies might be caused by
heterogeneous patient populations, variations in methodology or
small sample sizes lacking statistical power. Meta-analysis is a
recognized technique used to resolve discrepancies between
studies. It is a quantitative method that combines results from
independent studies to increase statistical power in order to
derive more solid conclusions. This allows one to distinguish small
effects from no effect. It also helps to determine whether the
variation in effects between studies is merely due to the expected
random statistical fluctuation or instead to sample variations or
trait assessment. In addition, the technique of meta-regression
may be used to evaluate confounders and discrepancies among
different studies.43
The aims of this study were therefore firstly to verify if
peripheral CRP levels are indeed increased in SZ, secondly if the
increases occur since the occurrence of the first episode of
psychosis, thirdly if different classes of antipsychotics divergently
© 2016 Macmillan Publishers Limited
C-reactive protein in schizophrenia
BS Fernandes et al
555
modulate CRP levels, and fourthly whether its levels are correlated
with positive and negative symptomatology and obesity. Lastly,
we wanted to find out whether CRP levels change following
antipsychotic therapy, indicating either potential harmful or
protective metabolic effects of antipsychotic medication regard-
ing inflammation. With this in mind, we performed a meta-
analysis of all cross-sectional studies of serum and plasma CRP
levels in SZ compared to healthy subjects. In addition, we
evaluated longitudinal studies on CRP levels before and after
antipsychotic use. This will help to clarify the role of CRP as a
biomarker of low-grade inflammation in SZ, ultimately clarifying
the role of CRP as an underpinning factor involved in the
inflammatory hypothesis.
MATERIALS AND METHODS
This study is comprised of a between-group meta-analysis comparing
serum and plasma CRP levels in persons with SZ and healthy controls, and
of two within-group meta-analyses of longitudinal studies of CRP levels
before and after initiation or change of class of antipsychotics. We adhered
to the Preferred Reporting Items for Systematic Reviews and Meta-
Analyses statement and by the Cochrane Collaboration.43 The literature
search, decisions on inclusion, data extraction and quality control were all
performed independently by two of the authors (BSF and CAG).
Search strategy
We conducted a systematic search for all possibly eligible English and non-
English peer-reviewed articles to avoid language publication bias44,45 using
Medline, the Cochrane Library, Scielo, Scopus and Web of Knowledge. No
year or country restrictions were used. The Boolean search term used for
the electronic database search was (CRP OR C-reactive protein OR hsCRP
OR hs-CRP) and (schizophrenia OR psychosis OR schizophreniform OR
schizoaffective). The last search was performed in January 2015. We then
manually checked the reference sections of the publications found
through our electronic search, to identify additional studies that may
have been missed. Study selection eligibility and exclusion criteria were
pre-specified.
Study selection
Inclusion criteria were as follows: (1) adult subjects with SZ, schizophreni-
form, or schizoaffective disorder, as defined by the American Psychiatric
Association, which from now on will collectively be referred to as SZ for the
sake of simplicity; (2) pairwise comparison with a control group of healthy
volunteers for the between-group meta-analysis or pairwise comparisons
before and after antipsychotics in longitudinal studies for the within-group
meta-analyses; and (3) studies assessing circulating serum or plasma CRP
or high-sensitivity CRP in human blood samples in vivo. Exclusion criteria
were as follows: (1) duplicate reports; (2) studies that employed CRP assays
that provide dichotomous results (‘positive‘ or ‘negative‘); and (3) lack of a
control group for the between-group meta-analysis, or lack of follow-up for
the within-group meta-analyses (Figure 1). The decision of whether to
include studies in the meta-analyses was made based on the above
criteria, and a consensus was reached among the authors on those
decisions.
Data extraction
To avoid potential errors, two reviewers (BSF and CAG) inde-
pendently extracted data (n, mean and s.d.). The following data were
extracted: diagnostic status, CRP levels, sex, age, duration of illness, body
mass index (BMI), type of antipsychotics, Positive and Negative Symptoms
Scale (PANSS) total scores and its positive and negative sub-scores.46
According to information available in the studies, sub-group analysis
according to antipsychotic status (that is, drug-naive or -free, or on
antipsychotics) was performed. Subjects with SZ were considered drug-
free if they were off psychiatric medication before blood sampling for at
least 2 weeks. Discrepancies in data entry were double-checked by the
reviewers with the original published data and consensus was reached.
When the necessary data were not available from the published paper, we
contacted the authors and requested the necessary information. If the
results were graphically presented and the authors could not provide the
data, we used the method for data extraction from graphs by Sistrom
Molecular Psychiatry (2016), 554 – 564
 C-reactive protein in schizophrenia
BS Fernandes et al
556

Potentially relevant, published articles identified

(n = 124)

Studies screened on the basis of title and abstract

(n = 48)

Studies excluded: 22
Duplicate report
No control group
CRP assessed before onset of disease

Studies included in the meta-analyses

(n = 26)

Within-group meta-analysis Within-group meta-analysis

Between-group meta-analysis
Start of antipsychotics Change of antipsychotics
19 studies included
6 studies included 3 studies included
24 pairwise comparisons
8 pairwise comparisons 12 pairwise comparisons

Figure 1. Flow diagram of the systematic review showing the study inclusion and exclusion process. Some studies included subjects in more
than one mood state, being included in more than one analysis.

et al.47 Whenever multiple reports pertained to the same groups of
patients, we retained only the most comprehensive report for the meta-
analytic calculations to avoid duplication of information.43
Quality assessment
We used the Newcastle–Ottawa Scale (NOS)48 as recommended by the
Cochrane Collaboration,43 to assess the quality of the eligible observational
studies. For the between-group and within-group meta-analyses, we used
the NOS scale for case–control and for cohort studies, respectively. Overall
quality score was defined as the frequency of criteria that were met by the
particular study. The NOS scale contains eight items in the case–control
section and nine items in the cohort section, categorized into the three
domains of selection, comparability, exposure and outcome (the last only
for the cohort scale). A series of response options is provided for each item.
A star system was used to enable semi-quantitative assessment of study
quality, such that the highest quality studies are awarded a maximum of
one star for each item, with the exception of the comparability domain,
which allows the assignment of two stars. As such, the NOS scale ranges
between 0 and 9 stars.48 Item ‘non-response rate’ from Exposure in the
case–control scale was not applicable; therefore, a maximum of eight stars
was considered. For the cohort scale we considered the maximum as nine
stars. The quality score of the included studies ranged from 1 to 7. All
studies were included in the posterior analyses.
Publication bias
Studies with negative results are less likely to be published than studies
with positive results. To account for significant publication bias, we
analyzed a funnel plot graph, a scatter plot of the effect size (ES) against a
measure of study size, and the Egger test.45 In addition, the Orwin’s fail-safe
N-test (file drawer statistic)49 was used to quantify the number of possible
negative omitted studies that would be required to make our results
nonsignificant (P40.05).
Statistical analysis
Comprehensive Meta-analysis Software version 2.0 (CMA, Borenstein, USA)
was employed in all analyses. As studies used different measurement
methods, standardized mean difference estimates of the differences in CRP
levels between subjects with SZ and healthy controls were used as the ES,
using Hedges’ adjusted g. This provides an unbiased ES adjusted for small
sample sizes. The 95% confidence interval (95% CI) of the ES was
also computed. An ES of 0.2 is considered as indicating a small effect,
meaning a small difference in CRP levels between subjects with SZ and
controls, of 0.5 indicating a moderate effect, and of 0.8 indicating a
large effect.43
We assessed the heterogeneity across studies using the Cochran Q
test,50 a weighted sum of the squares of the deviations of individual study
ESs estimates from the overall estimate, and a P-value of o0.10 was
considered significant (that is, showing heterogeneity). Inconsistency
across studies was then quantified with the I2 metric.45,51 It can be
interpreted as the percentage of total variation across several studies due
to heterogeneity, and it is considered moderate when between 50% and
75% and high when > 75%. As the analyses showed that the studies were
heterogeneous, we pooled ES results from different studies according to
the inverse variance method of accounting for random effects, which
allows population level inferences and is more stringent than the fixed-
effect models.52. Random-effect modelling assumes a genuine diversity in
the results of various studies and incorporates a between-study variance
into the calculations, and is considered the best modelling in the presence
of heterogeneity. The direction of the ES was positive if subjects with SZ
showed increased CRP and negative if they showed decreased CRP levels
when compared with controls for the between-group meta-analysis, and
positive if it increased after the use of antipsychotics for the within-group
meta-analyses.
Unrestricted maximum likelihood random-effects meta-regressions52 of
ESs were performed with mean age, length of follow-up, BMI, NOS score
and severity of disease as assessed by PANSS as moderators, to determine
whether these covariates influenced the ES. Studies were weighted such
that the studies with the most precise parameters, quantified by the
sample size and 95% CI, had more influence in the regression analyses.43

Molecular Psychiatry (2016), 554 – 564 © 2016 Macmillan Publishers Limited


The meta-analyses consisted of three steps. Firstly, we performed the
overall analysis for all meta-analyses. Secondly, sensitivity analyses were
conducted to ascertain whether the results of our analyses were strongly
influenced by any single study or studies sharing some characteristic. The
overall significance was recomputed after each study or group of studies
with a common feature were deleted from the analysis. Thirdly, we
performed meta-regression analyses in order to investigate possible
moderators of CRP levels. The level of significance for the ES estimates was
set at Po0.05.
RESULTS
We identified 124 studies through electronic searches (Figure 1),
all of which, with the exception of one Russian study, had an
English abstract;27 it was translated and later included by one of
the authors (HGB) who is a Russian speaker. Of those, 76 were
excluded on the basis of the title and the abstract, leaving 48
studies for further evaluation. Twenty-six studies fulfiled our
inclusion criteria, 19 for the between-group meta-analysis of
SZ versus control,14,17–19,21,24,25,27–31,33–36,38,39,41 6 for the within-
group meta-analysis of CRP changes after initiation of
antipsychotics20,21,23,27,32,37 and 3 for the within-group meta-
analysis of CRP changes with substitution of the type of the
antipsychotic.22,26,40 Some studies provided more than one
pairwise comparison, as whenever possible we extracted data
according to medication status.
CRP levels are increased in SZ regardless of the use of
antipsychotics
Supplementary Table S1 summarizes the 19 studies,14,17–19,21,24,25,
27–31,33–36,38,39,41
with 24 pairwise comparisons included in the
between-group meta-analysis. The studies were published
between 2007 and 2014, and varied in sample size (from 60 to
78 810). The mean participant age varied from 16 to 57 years.
Eight studies assessed CRP levels in the serum24,27,31,33,34,36,39,41
and 11 in the plasma.14,17–19,21,25,28–30,35,38 Of the 19 included
studies, 5 provided data on drug-naive or drug-free
subjects.27,31,34,35,38 In 11 of the studies, the control group was
matched by age and BMI to the case group.17,18,21,24,25,29,30,33–35,41
Four studies were population studies14,18,30,34 and the other 15
were case–control studies conducted in in-patient or out-patient
units.17,19–29,31–33,35–41 Five studies provided data on subjects with
first episode psychosis19,27,31,34,38 and 14 studies provided data on
subjects with chronic SZ14,17,18,21,24,25,28–30,33,35,36,39,41 (that is, less
or more than 1 year since the occurrence of psychosis, respec-
tively). One study did not provide data on BMI27 and 13 studies
showed a lack of smoking status.14,17,21,24,25,27,29,31,33,34,38,39,41
Seventeen of the 19 studies measured CRP levels employing
a high-sensitivity assay.14,17–19,21,24,25,28–31,33,35,36,38,39,41
Overall, random-effects between-group meta-analysis showed
that serum and plasma CRP levels were moderately increased in
subjects with SZ when compared with healthy controls (g = 0.66,
95% CI 0.43 to 0.88, P o0.001, 24 between-group comparisons,
n = 82 962). When we carried out sub-group analyses according to
the use of antipsychotics, we verified that CRP levels were
increased with large ESs in subjects with SZ who were drug-naive
or -free (g = 0.87, 95% CI 0.14 to 1.60, P = 0.021, 6 between-group
comparisons, n = 708), and were moderately increased in subjects
with SZ on antipsychotics (g = 0.58, 95% CI 0.36 to 0.79, P = 0.001,
18 between-group comparisons, n = 82 254). When analysing the
subjects on psychiatric medication according to antipsychotic
type, CRP levels were increased with large ESs among persons on
typical antipsychotics (g = 0.87, 95% CI 0.67 to 1.08, P = 0.001, 2
between-group comparisons, n = 414), were moderately increased
on atypical antipsychotics (g = 0.53, 95% CI 0.14 to 0.93, P = 0.008,
7 between-group comparisons, n = 1366), and again moderately
increased on subjects on typical and/or atypical antipsychotics
(g = 0.52, 95% CI 0.27 to 0.76, P = 0.001, 9 between-group
© 2016 Macmillan Publishers Limited
C-reactive protein in schizophrenia
BS Fernandes et al
557
comparisons, n = 80 474) (Figure 2 and Table 1). In addition, CRP
levels were moderately increased in both subjects with SZ during
a first episode of psychosis, here defined as less than 1 year since
the onset of psychosis (g = 0.63, 95% CI 0.03 to 1.22, P = 0.038,
6 between-group comparisons, n = 708), and in other SZ subjects,
defined as more than 1 year since the onset of psychosis (g = 0.76,
95% CI 0.43 to 1.08, P = 0.001, 18 between-group comparisons,
n = 82 254) (Figure 2 and Table 1).
Studies conducted in clinical samples usually report a greater
effect when compared with population-based studies. Four of the
19 studies included were population based.22,28,32,38 In general,
CRP was increased with a small effect in population samples
(g = 0.18, 95% CI 0.06 to 0.30, P = 0.003, 4 between-group compa-
risons, n = 79 506) and with a large effect in clinical samples
(g = 0.77, 95% CI 0.52 to 1.02, P o 0.001, 20 between-group
comparisons, n = 5390).
As differences in BMI and weight between the patient and
control groups could confound results, we re-ran the analysis
including only the 11 studies with a control group matched
by BMI and age to the SZ group21–23,25,28,30–35 and verified that
CRP remained increased with a moderate effect (g = 0.51, 95% CI
0.43 to 0.59, P o0.001, 14 between-group comparisons,
n = 80 693). In addition, when we sub-grouped the studies
according the presence or the absence of clinical chronic
comorbidities such as diabetes mellitus and dislipidemia, the ESs
remained significant in both groups (Table 1).
CRP levels are positively related to the severity of positive
symptoms and BMI, and negatively related to age:
meta-regressions
We performed meta-regression analyses in the between-group
meta-analysis in an exploratory attempt to identify the sources of
heterogeneity between studies and the effect of moderators. In
univariable meta-regression models, we found no relationship
between CRP levels and severity of positive and negative
symptoms as assessed by the PANSS total scores. When we did
a separate analysis of positive and negative subscores of the
PANSS, we verified that the severity of positive symptoms as
assessed by PANSS-positive subscale was positively related to the
magnitude of the ES, indicating that the greater the severity of
positive symptoms, the greater the increase in CRP levels were
(slope = 0.12, 95% CI 0.03 to 0.23, P = 0.013). We found no
relationship between CRP levels and negative symptoms (slope =
0.04, 95% CI − 0.10 to 0.20, P = 0.533) (Table 1).
Furthermore, there was a negative relationship between CRP
levels and age in the drug-naive or -free subjects, meaning that
the greater the age, the smaller the difference in CRP between
drug-naive or -free subjects with SZ and healthy controls were
(Table 1).
Lastly, we also found a positive relationship between BMI and
CRP levels regardless the use of antipsychotics, meaning that the
higher the BMI, the higher CRP levels were in persons with SZ
when compared with controls (Table 1).
CRP levels do not increase after initiation or change of
antipsychotics
In order to verify if antipsychotics induced changes in CRP levels,
we conducted two within-group meta-analyses of longitudinal
studies, one of CRP changes before and after initiation of
antipsychotics, and one of CRP levels before and after change of
type of antipsychotics.
Six studies with eight pairwise comparisons20,21,23,27,32,37 were
included in the meta-analysis of alterations in CRP levels after
initiation of antipsychotics, with a total of 713 persons with SZ.
Five studies20,21,23,27,37 referred to drug-naive or -free subjects with
SZ, who were put on typical, atypical, or a combination of these
antipsychotics. One of these studies referred to persons with SZ
Molecular Psychiatry (2016), 554 – 564
 C-reactive protein in schizophrenia
BS Fernandes et al
558
Study g p
Fawzi et al., 2011a 1.45 0.000
Fawzi et al., 2011b 1.91 0.000
Fernandez-Egea et al., 2009 0.04 0.858
Sarandol et al., 2007 -0.25 0.276
Hepgul et al., 2012 0.39 0.105
Berardis et al., 2014 1.68 0.000
0.87 0.021
Akanji et al., 2009 0.87 0.000
Carrizo et al., 2008a 0.40 0.259
Carrizo et al., 2008b 0.53 0.133
Carrizo et al., 2008c 0.91 0.010
Dickerson et al., 2013 0.51 0.000
Hope et al., 2011 0.08 0.451
Antipsychotics

Kliushnik et al., 2008 1.12 0.000

Kuo et al., 2013 2.06 0.000
Lin et al., 2013a 0.55 0.205
Lin et al., 2013b 0.31 0.404
Lin et al., 2013c 0.24 0.545
Suvisaari et al, 2011 0.38 0.007
Vuksan-Cusa et al., 2013 0.32 0.082
Joshi et al, 2013 0.63 0.004
Frydecka et al, 2014 0.35 0.021
Klemettila et al., 2014 1.12 0.000
Stojanovic et al., 2014 0.12 0.607
Wium-Andersen et al., 2014 0.20 0.010
0.58 0.000
Overall 0.60 0.000
-2.00 -1.00 0.00 1.00 2.00

CRP Decreased CRP Increased

Figure 2. Forest plot for random effects between-group meta-analysis on serum and plasma C-reactive protein (CRP) levels in persons with
schizophrenia (SZ) and healthy controls. The sizes of the squares are proportional to sample size. A total of 19 studies were included,
comprising 1995 subjects with SZ and 80 967 healthy controls.

on olanzapine with aripiprazole as an add-on therapy.32 The
follow-up varied from 4 to 52 weeks and the antipsychotic
treatment initiated included haloperidol, clozapine, risperidone,
olanzapine, aripiprazole, or a combination of typical and atypical
antipsychotics. The mean BMI of the study participants was 26.65
at baseline and 27.63 at the end of follow-up (P = 0.901)
(Supplementary Table S2). Overall, initiation of antipsychotics
was not associated with an increase in CRP levels (g = 0.01, 95% CI
− 0.20 to 0.22, P = 0.803, 8 within-group comparisons, n = 713).
When we analysed the changes of CRP levels according the type
of antipsychotic initiated, we verified that our results remained
nonsignificant regardless of whether the antipsychotic was typical,
atypical, or a combination of both (Figure 3 and Table 1).
Furthermore, in the meta-regressions we verified that neither age,
weeks of follow-up nor BMI at the baseline influenced these
results (see Table 1 for details).
In addition, we analysed if the change of type of antipsychotics
altered CRP levels. Three studies22,26,40 comprising 12 pairwise
comparisons with a total of 1325 persons with SZ were included.
Again, change of antipsychotics did not produce a change in CRP
levels (g = − 0.03, 95% CI − 0.11 to 0.06, P = 0.059, 12 within-group
comparisons, n = 1325), irrespective of whether the change was to
a typical or atypical antipsychotic (Figure 4 and Table 1). Once
more, the meta-regressions showed that age, weeks of follow-up
and BMI at the baseline did not influence these results (Table 1).
Sensitivity analyses
In the between-group meta-analysis, the heterogeneity for the
whole analysis was high. When we analysed the studies according
to the source of the sample, that is, population-based or clinical
studies, heterogeneity was mostly high in the clinical studies and
low in the population-based studies. In all other subgroup
analyses, the heterogeneity remained high and could not be
explained by any single variable (Table 1). Thereafter, we
conducted a sensitivity analysis in all sub-group meta-analyses
excluding studies one at a time to determine the robustness of the
analyses and to verify whether a particular study was responsible
for the high heterogeneity. No single study thoroughly explained
the heterogeneity and the results remained significant in all cases,
even when the the larger population study conducted by Wium-
Andersen et al.14 were excluded.
We ran the analysis including only the 17 studies that employed
a highly sensitive CRP assay and the results remained significant
with a moderate ES (g = 0.67, 95% CI 0.43 to 0.90, P o0.001, 22
between-group comparisons, n = 82 744) and the heterogeneity
did not change. In addition, CRP levels were equally increased in
serum and plasma (Table 1).
In the within-group meta-analyses of CRP alterations after the
initiation of antipsychotics and after a change of antipsychotics,
the results remained nonsignificant in the sensitivity analysis
when one study at a time was excluded, again showing that these
negative results were not due to a particular study.
Publication bias
We found no evidence of publication bias in the funnel plot of the
between-group meta-analysis and the Egger’s test was nonsigni-
ficant (P = 0.11). Moreover, the Orwin’s fail-safe N-test yielded 5564
studies as the number of negative studies necessary to turn our

Molecular Psychiatry (2016), 554 – 564 © 2016 Macmillan Publishers Limited

 C-reactive protein in schizophrenia
BS Fernandes et al
559
Table 1. Statistics on –between and –within group meta-analyses regarding serum and plasma CRP levels in schizophrenia

Group-Wise No. of pairwise No. of subjects Meta-analysis Heterogeneity

Between-Group SZ vs. HC SZ HC Hedges’ g 95% CI P I2 Q P

All 24 1,995 80,967 0.60 0.43 0.88 0.001 91.14 259.68 0.001
Drug-naïve or free 6 348 360 0.87 0.14 1.60 0.021 94.99 99.78 0.001
All antipsychotics 18 1,647 80,607 0.58 0.36 0.79 0.001 87.42 135.22 0.001
Typical antipsychotics 2 237 177 0.87 0.67 1.08 0.001 0.00 0.08 0.928
Atypical antipsychotics 7 346 1020 0.53 0.14 0.93 0.008 76.99 26.08 0.001
Typical/atypical antipsychotics 9 1064 79,410 0.52 0.27 0.76 0.001 85.34 54.56 0.001
First episode SZ 6 348 360 0.63 0.03 1.22 0.038 86.99 30.74 0.001
Chronic SZ 18 1,647 80,607 0.76 0.43 1.08 0.001 94.93 256.67 0.001
Populational samples 4 473 79,033 0.18 0.06 0.30 0.003 14.96 3.53 0.317
Clinical samples 20 1,934 3,456 0.77 0.52 1.02 0.001 88.43 164.29 0.001
Age and BMI paired 14 1,096 79,599 0.51 0.43 0.59 0.001 93.15 189.98 0.001
hsCRP only 22 1,842 80,902 0.67 0.43 0.90 0.001 91.53 248.06 0.001
Without clinical conditions 17 1,227 961 0.78 0.69 0.87 0.001 88.42 138.19 0.001
With clinical conditions 7 768 79,736 0.45 0.36 0.54 0.001 93.62 94.14 0.001
Serum 18 1,344 1,053 0.65 0.38 0.91 0.001 87.97 141.40 0.001
Plasma 6 651 80,565 0.69 0.22 1.16 0.004 95.57 112.91 0.001
Within-Group SZ before and after No. of pairwise SZ sample size Hedges’ g 95% CI P I2 Q P
start of antipsychotics

All 8 713 0.01 − 0.20 0.22 0.803 74.33 27.27 0.001

Typical antipsychotics 2 36 0.01 − 0.82 0.84 0.983 84.29 6.36 0.012
Atypical antipsychotics 3 62 0.32 − 0.55 1.24 0.494 88.41 17.69 0.001
Typical/atypical antipsychotics 3 615 0.01 − 0.07 0.08 0.849 0.00 1.21 0.544
Within-Group SZ before and after No. of pairwise SZ sample size Hedges’ g 95% CI P I2 Q P
change of antipsychotics

All 12 1325 − 0.03 − 0.11 0.06 0.059 66.23 27.27 0.001

Typical antipsychotics 2 159 − 0.08 − 0.23 0.07 0.075 0.00 0.52 0.471
Atypical antipsychotics 9 1086 0.04 − 0.07 0.15 0.467 69.42 26.18 0.001

Moderator in Meta-regressions No. of pairwise No. of subjects Meta-regression Meta-regression

Between-Group SZ vs. HC SZ HC Slope 95% CI P Intercept Z P

BMI (all*) 23 1,892 80,952 0.21 0.04 0.29 0.004 − 1.16 − 0.83 0.401
BMI (drug-naïve or free*) 6 348 360 0.29 0.16 0.38 0.001 − 6.41 − 5.78 0.001
BMI (on all antipsychotics*) 17 1,544 80,592 0.11 0.03 0.20 0.009 − 2.53 − 2.14 0.032
WHR (all) 5 490 445 3.18 − 15.98 22.35 0.744 − 1.65 − 0.18 0.852
Waist circumference (all) 10 558 511 0.01 − 0.07 0.09 0.867 0.22 0.05 0.954
Age (all) 24 1,995 80,967 − 0.02 − 0.04 0.01 0.169 1.34 2.62 0.008
Age (drug-naïve or free*) 6 348 360 − 0.20 − 0.34 -0.06 0.006 6.84 3.12 0.001
Age (on all antipsychotics) 18 1,647 80,607 − 0.01 − 0.03 0.01 0.558 0.84 1.78 0.074
Age of onset (all) 10 639 592 0.05 − 0.01 0.12 0.087 − 0.76 − 0.90 0.365
Length of illness (all) 10 639 592 − 0.03 − 0.09 0.02 0.219 1.01 2.93 0.003
PANSS Total Scores (all) 12 1,047 920 − 0.01 − 0.03 0.02 0.750 0.99 1.27 0.201
PANSS Positive Scores (all*) 7 401 349 0.12 0.03 0.23 0.013 − 1.25 − 1.48 0.137
PANSS Negative Scores (all) 7 401 349 0.04 − 0.10 0.20 0.533 − 0.08 − 0.06 0.947
Sample size (all) 23 1,892 80,952 − 0.01 − 0.02 0.01 0.392 0.68 5.51 0.001
NOS (all) 23 1,892 80,952 − 0.02 − 0.04 0.16 0.189 1.39 1.89 0.392
Within-Group SZ before and after No. of pairwise SZ sample size Slope 95% CI P Intercept Z P
start of antipsychotics

Length of follow-up in weeks 8 713 − 0.01 − 0.02 0.01 0.583 0.04 0.37 0.708
Age 8 713 0.01 − 0.02 0.02 0.910 − 0.03 − 0.09 0.923
BMI in the baseline 8 602 − 0.06 − 0.14 0.02 0.144 1.58 1.42 0.154
Within-Group SZ before and No. of pairwise SZ sample size Slope 95% CI P Intercept Z P
after change of antipsychotics

Length of follow-up in weeks 12 1,325 -0.01 − 0.02 0.02 0.612 0.02 0.45 0.403
Age 12 1,325 0.02 − 0.03 0.03 0.485 − 0.03 − 0.12 0.763
BMI in the baseline 12 1,325 -0.12 − 0.19 0.12 0.319 1.93 1.95 0.261
Abbreviations: BMI, body-mass index in kg/m2; CI, confidence interval; HC, healthy control; hsCRP, high-sensitivity C-reactive protein; NOS, Newcastle-
Ottawa Scale; PANSS, Positive and Negative Symptoms Scale; SZ, schizophrenia; WHR, waist-hip ratio. *Po0.05.

© 2016 Macmillan Publishers Limited Molecular Psychiatry (2016), 554 – 564

 C-reactive protein in schizophrenia
BS Fernandes et al
560
Study g p

Typical
Diaz et al., 2010a 0.43 0.053

Henderson et al., 2009 -0.43 0.096

0.01 0.983

Loffler et al., 2010 2.72 0.000

Atypical
Diaz et al., 2010b -0.12 0.521

Diaz et al., 2010c -0.46 0.015

0.32 0.494
Typical/Atypical

Sarandol et al., 2007 0.18 0.267

Kliushnik et al., 2008 0.01 0.945

Schwarz et al., 2012 -0.01 0.901

0.01 0.849

Overall 0.01 0.803

-2.00 -1.00 0.00 1.00 2.00
CRP Decreased CRP Increased

Figure 3. Forest plot for random effects within-group meta-analysis on serum and plasma C-reactive protein (CRP) level alterations after
initiation of antipsychotics in persons with schizophrenia (SZ). The sizes of the squares are proportional to sample size. A total of 6 studies
were included, comprising 713 subjects with SZ.

Study g p
Typical

Meyer et al., 2009b -0.06 0.471

Kramer et al., 2011e -0.24 0.312

-0.08 0.312

Baptista et al., 2007 -0.18 0.162

Meyer et al., 2009a 0.35 0.000

Meyer et al., 2009c 0.07 0.350

Atypical

Meyer et al., 2009d -0.10 0.185

Meyer et al., 2009e -0.00 0.981

Kramer et al., 2011a 0.00 1.000

Kramer et al., 2011b 0.14 0.257

Kramer et al., 2011c 0.00 1.000

Typical/Atypical

Kramer et al., 2011d 0.02 0.860

0.06 0.058

Kramer et al., 2011f -0.17 0.134

-0.17 0.134

Overall 0.03 0.322

-1.00 -0.50 0.00 0.50 1.00

CRP Decreased CRP Increased

Figure 4. Forest plot for random effects within-group meta-analysis on serum and plasma C-reactive protein (CRP) levels alterations after
change of type antipsychotics in persons with schizophrenia (SZ). The sizes of the squares are proportional to sample size. A total of 3 studies
were included, comprising 1325 subjects with SZ.

positive results into negative ones. In addition, the heterogeneity DISCUSSION

was not explained by the covariates of sample size or NOS study
quality, which reinforces the fact that our results are likely not due
to publication bias. There were also no evidence of publication
bias for both within-group meta-analyses of initiation or change of
antipsychotics in the funnel plots, and again the Egger’s test were
nonsignificant (P = 0.57 and 0.48, respectively). Again, there were
no relationship between the covariates of sample size or NOS
study quality (see Table 1 for details).
Our three meta-analyses of CRP in SZ included a total of 26 cross-
sectional or longitudinal studies comprising 85 000 participants.
CRP levels were moderately increased in persons with SZ
regardless the use of antipsychotics and did not change between
the first episode of psychosis and with progression of SZ. The
extent of the increase in peripheral CRP levels paralleled the
increase in severity of positive symptoms, but was unrelated to

Molecular Psychiatry (2016), 554 – 564 © 2016 Macmillan Publishers Limited


the severity of negative symptoms. It is also aligned with the
increase in BMI. Conversely, higher age correlated with a smaller
difference in CRP levels between persons with SZ and controls.
Furthermore, CRP levels did not change after the initiation of
antipsychotics notwithstanding if these were typical or atypical
antipsychotics.
One of the most common explanations of the inflammatory
hypothesis of SZ presumes that the inflammatory changes found
in this disorder are attributable to and are merely a marker of the
metabolic syndrome and the unhealthy lifestyle that commonly
accompanies severe psychiatric conditions.22,40 Our analyses
uncover that CRP levels are moderately increased in subjects with
SZ within 1 year of the occurrence of the first episode of psychosis
and remains moderately increased with progression of SZ. In order
to further clarify this point, we conducted a series of analyses of
sub-groups and verified that CRP levels are increased in SZ in
comparison with controls even when only studies paired by BMI
and age were considered, and also when only studies that
included subjects free of any clinical condition besides SZ were
examined. When we analysed CRP levels according to the setting
of the study, that is, population-based studies versus studies
conducted in clinical settings, we once more verified that CRP
levels remained increased in SZ in both scenarios, although the
magnitude of the effect dropped from large, in clinical settings, to
small, in population-based samples. Taken altogether, these
findings indicate that although lifestyle factors and clinical
comorbidities are of cardinal importance in the machinery of SZ,
SZ per se is also associated with an inflammatory burden.
We found that the greater the severity of positive symptoms—
the hallmark and most prominent feature of SZ—the greater
the CRP levels were. In addition, subjects with SZ presented a
proportionally higher increase in CRP levels with increases in BMI
than controls, again suggesting that SZ is associated with a
particularly high inflammatory burden. Interestingly, we found an
inverse correlation of higher age with smaller differences in CRP
levels between persons with SZ and healthy controls. This has two
possible explanations. Firstly, the decreased difference in CRP
between persons with SZ and controls with age may be due to the
expected increase in CRP in controls with age, which could
hamper the association. Secondly, it may suggest that the early
stages of SZ present a particularly prominent inflammatory
component, possibly related to its development, which burns
out over time.53
Possible CRP effects in the brain and its modulation by the
periphery
The biological mechanism linking inflammation and SZ is not
entirely understood. There is evidence from animal studies that
CRP per se, which is produced in the periphery, does not cross the
blood–brain barrier under ordinary circumstances; however, high
levels of peripheral CRP can increase the blood–brain barrier
paracellular permeability by affecting the function of tight
junctions.16 This allows pro-inflammatory cytokines and CRP itself
to enter the central nervous system and ultimately turn the brain
susceptible to CRP. Thus, increased CRP levels may explain at least
in part the increased permeability of the blood–brain barrier that
is assumed to have a role in psychiatric disorders, including
SZ.54,55 Once within the central nervous system, CRP may directly
induce neuroinflammation. In primary cell culture, CRP induces a
pro-inflammatory state in microglia.56
Hyper-reactivity of the microglia can induce changes in
astrocytes, which in turn release IL-6 and transforming growth
factor-β.56,57 This induces the metabolism of tryptophan to
kynurenic acid by the astrocytes. Kynurenic acid, in turn, acts as
an antagonist of NMDA receptors, creating a hypoglutamatergic
state with decreased brain-derived neurotrophic factor (BDNF)
and ultimately potentiating the dopaminergic hyperfunction in
© 2016 Macmillan Publishers Limited
C-reactive protein in schizophrenia
BS Fernandes et al
561
the limbic system that is responsible for the development of
positive symptoms characteristic of SZ.15,58–60 In our study, we
found that positive symptoms appear to be most robustly linked
to inflammation, with more severe symptomatology correlated
to higher levels of CRP. Thus, a pro-inflammatory state with
activation of the kynurenine pathway may be one possible
explanation for the association of CRP with SZ. This is in line with
the results of a meta-analysis of cytokines in SZ, which found IL-6
and transforming growth factor-β to be state markers, exacer-
bated during acute relapse and decreasing after antipsychotic
treatment.11 Elevated kynurenic acid seems to be characteristic of
SZ, being elevated in the cerebrospinal fluid and in the prefrontal
cortex of persons with SZ, and post-mortem studies demonstrated
increased density and activation of the microglia in SZ.15,61,62.
Moreover, the protein S100B, which is increased in SZ,63 denotes
astrocytic activity and can activate microglia,56 has been found to
be increased in SZ, and two recent meta-analyses found the
neurotrophin BDNF,64,65 which is decreased by pro-inflammatory
cytokines, to be decreased in SZ as well.
Another feasible explanation for increased CRP in SZ would be
that CRP may be acting as a proxy for IL-6, as it is directly and
positively modulated by it. There is some evidence that IL-6
is correlated with positive symptoms,57 akin to CRP as observed
in this analysis. In this case, increased CRP levels would be
an epiphenomenon and a peripheral consequence of a
pro-inflammatory state during acute psychosis. Either way, in
both scenarios the mechanisms whereby the inflammatory
hypothesis has a role in SZ would complement the glutamatergic
and dopaminergic hypotheses rather than demand an entirely
new framework.
It remains true that CRP is elevated by many lifestyle and
environment factors that are disproportionately present across
psychiatric disorders, including poor diet, smoking and obesity.3,13
In addition, it is known that these environmental factors are
transduced using cytokine and related signalling pathways, and
are all capable of increasing inflammatory markers such as CRP.3
As such, CRP may be a proxy of an allostatic environmental load as
well as a marker of non-communicable disorders, many of which
share CRP as a biomarker.66
It should be emphasized that the peripheral and central
inflammatory systems probably operate in parallel in a synchro-
nized way. In animal models, peripheral inflammatory stimuli can
induce expression of inflammatory mediators in the brain, and it
has been shown that overexpression of IL-1 in the central nervous
system can increase peripheral production of inflammatory
mediators, pointing out this bidirectional communication between
central and peripheral inflammatory systems. Treatment with
interferon-α can induce psychiatric symptoms, and increases IL-6
levels in the cerebrospinal fluid, again suggesting that peripherally
activating the immune system activates inflammatory pathways in
the brain.57 In addition, in SZ, treatment with antipsychotics can
peripherally decrease the levels of some pro-inflammatory
cytokines, such as IL-6 and transforming growth factor-β.11,42 This
bidirectional interaction between the brain and the periphery
regarding inflammatory pathways contributes to the increased
peripheral inflammation found in the periphery in SZ, possibly
playing a role in the occurrence of pathologies commonly
associated with SZ.
Effects of antipsychotics on central and peripheral inflammation
Another common explanation for the increased inflammation in
SZ is the potentially harmful metabolic profile of antipsychotics.22
However, we found largely increased CRP levels in drug-naive or
-free subjects with SZ and only moderately increased ones in
those on antipsychotics. Of note, the magnitude of the ES was
smaller in persons on atypical antipsychotics, which are usually
Molecular Psychiatry (2016), 554 – 564
 C-reactive protein in schizophrenia
BS Fernandes et al
562
associated with a particularly unfavourable metabolic profile, than
on those on typical antipsychotics.
If one considers that SZ is a pathology associated with increased
inflammation per se as substantiated by elevated levels of CRP,
and that its severity is positively correlated with CRP levels, it is
tempting to speculate that medications with the ability to
ameliorate psychotic symptoms might, theoretically, have the
potential to decrease CRP levels. It was shown in pre-clinical
studies that antipsychotics can decrease cytokines, including IL-
6,42 which directly modulates CRP. A meta-analysis of peripheral
cytokine levels in subjects with SZ found IL-6 and transforming
growth factor-β to be increased during acute relapses and
decreased after treatment with antipsychotics.11 Considering
CRP in particular, treatment with risperidone decreased CRP levels
in one rodent study.42 Antipsychotics can decrease microglial
activation, one of the sources of pro-inflammatory cytokines in SZ,
such as tumour necrosis factor-α, nitric oxide, IL-1β and IL-2.56
Regarding the kynurenine pathway, chronic treatment with
antipsychotics can decrease kynurenic acid in the brain.15,16 In
our longitudinal meta-analyses, there were no increases in CRP
levels after the initiation of antipsychotic treatment. This null
result may be due to the possibility that antipsychotics ameliorate
positive symptoms by decreasing inflammation and the acute
phase response, regardless of their hazardous peripheral meta-
bolic effects driving metabolic syndrome. This aligns with results
of a longitudinal clinical study in SZ, which found that
antipsychotics increased CRP levels only in subjects with normal
CRP levels at baseline, and that it had no effect in those with
already increased CRP, and probably higher general inflammation,
at the beginning.26 Either way, our null result is reassuring to the
psychiatrist in clinical settings, suggesting that antipsychotics do
not increase CRP in SZ.
CRP levels as a possible biomarker in SZ
CRP, as opposed to cytokines and other biomarkers, is a standard
laboratory exam and can be measured in the peripheral blood of
persons with SZ and analysed in any clinical laboratory around the
globe. Since its measurement is already well defined worldwide,
CRP is a very attractive potential clinical biomarker in psychiatric
disorders.67 We found in our meta-regressions that the greater
the severity of positive symptoms, the greater the CRP levels
were. This makes CRP an attractive possible biomarker of disease
activity in SZ, mirroring severity. If future studies reveal an
increase in subjects with SZ right before exacerbations of positive
symptoms, and that it increases predicts the occurrence of
exacerbations of the disorder, this would possibly make CRP
a very useful biomarker to guide strategies in the clinical
setting.67–71
Whether an increase in CRP or an improvement of positive
symptoms occurs first is unknown. This is a pivotal point and
needs further studies in order to clarify the possible role of CRP in
the mechanism of action of antipsychotics. To provide a definitive
answer, longitudinal studies with frequent blood draws are
warranted. The fact that there were no changes in CRP levels
with treatment with antipsychotics limits the use of peripheral CRP
as a surrogate biomarker of response to antipsychotics.
Strengths and limitations
Our study relied on a large sample size (26 studies with 85 000
subjects), which permitted us to draw conclusions through meta-
analyses and meta-regression techniques. Our results are unlikely
to be caused by publication bias, as the funnel plots were
symmetrical, and a very large amount of negative unpublished
studies would be necessary to turn our positive results into
nonsignificant ones. In addition, through a series of sensitivity
and sub-group analyses, we were able to rule out the possibility
that the results were biased due to a unique outlier. The above-
Molecular Psychiatry (2016), 554 – 564
mentioned approach also allowed us to investigate and rule out
any single study as the sole source of the high heterogeneity
found in most analyses.
Notwithstanding its strengths, our study has some inherent
limitations due to its design and statistical methods employed.
Firstly, meta-analysis is retrospective research in nature, affected
by the methodological rigour of the studies included, comprehen-
siveness of search strategies and possible publication bias. We
tried to keep the probability of bias to a minimum by doing a
thorough search for data and by using explicit criteria for study
selection, data collection and data analysis. In addition, we
evaluated the quality of the studies using the NOS scale and
verified that most of the studies included were at least of
moderate quality. We believe that using this approach, the results
and conclusions can provide reliable information. Secondly, some
of the meta-regressions may have failed to achieve statistical
significance due to a lack of power in these specific analyses,
giving us a false-negative result. This may particularly have been
the reason for the lack of association between severity of negative
symptoms and CRP levels. Thirdly, the meta-analysis on CRP levels
in persons with SZ compared with controls provides us with a
pooled result originating from cross-sectional studies. Therefore,
we cannot draw any causal associations. Thus, we do not know if
an increase in CRP levels is a cause and pre-requisite for SZ
development or an allostatic counterbalancing mechanism as a
consequence of SZ development, related to environmental risk
factors or systemic metabolic changes associated with SZ such as
obesity. However, it is worth noting that CRP levels were increased
since the first episode of psychosis and in drug-naive subjects. CRP
remained increased when we considered just the studies with a
control group paired by BMI and studies which excluded systemic
clinical conditions. Fourthly, we considered as first-episode
psychosis subjects within the first year after the onset of
psychosis; therefore, our findings related to this population does
not refer solely to the early onset phase of an acute first episode.
Fifthly, as most studies did not present data separately according
the diverse types of antipsychotics, only an analysis considering
antipsychotics as typical, atypical, or a combination, was possible.
Lastly, as in other meta-analyses, our results should be interpreted
with caution, because individual studies varied greatly with
respect to the demographic characteristics of participants, type
of antipsychotics on use and duration of follow-up, which ranged
from 4 to 54 weeks. It is not possible therefore to extrapolate our
negative results regarding the absence of increased CRP levels
over a longer period of time. It is also worth mentioning that the
mean BMI presented in the longitudinal studies showed just a
slight albeit nonsignificant increase during the follow-up; there-
fore, we cannot extrapolate our results regarding the absence of
increase in CRP levels after initiation of antipsychotics to subjects
with substantial weight increase.
Final conclusions
Our three meta-analyses of 26 cross-sectional or longitudinal
studies comprising 85 000 persons with SZ and healthy controls
provide evidence that CRP levels are moderately increased in SZ
since at least the first episode of psychosis, and regardless of
whether the subjects were drug-naive or on typical or atypical
antipsychotics. The extent of the increase in peripheral CRP levels
paralleled the severity of positive symptoms, providing further
evidence of a systemic inflammatory component in this disorder.
Interestingly, CRP levels did not increase after initiation of
antipsychotics; we speculate that the potential harm caused by
antipsychotics on CRP may be counterbalanced by their benefits
ameliorating positive symptoms and decreasing inflammation. In
summary, our study provides further evidence of the inflammatory
hypothesis of SZ. If there is a causal relationship between higher
CRP levels and the development of SZ and aggravation of
© 2016 Macmillan Publishers Limited

psychotic symptoms, or if they are solely a marker of systemic low-
grade inflammation in SZ, remains to be clarified.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGMENTS
We thank all the authors of the included papers, in particular Drs Domenico De
Berardis, Nilay Hepgul, Susanne Kraemer, Vanessa Mondelli and Jaana Suvisaari, who
very kindly provided us with unpublished data for our paper. BSF is supported by a
scholarship and by a research grant MCTI/CNPQ/Universal 14/2014461833/2014-0,
both from CNPq, Brazil. MB is supported by a NHMRC Senior Principal Research
Fellowship 1059660.
REFERENCES
1 Maes M. Evidence for an immune response in major depression: a review and
hypothesis. Prog Neuropsychopharmacol Biol Psychiatry 1995; 19: 11–38.
2 Gibney SM, Drexhage HA. Evidence for a dysregulated immune system in the
etiology of psychiatric disorders. J Neuroimmune Pharmacol 2013; 8: 900–920.
3 Berk M, Williams LJ, Jacka FN, O'Neil A, Pasco JA, Moylan S et al. So depression is
an inflammatory disease, but where does the inflammation come from? BMC Med
2013; 11: 200.
4 Brietzke E, Stertz L, Fernandes B, Kauer-Sant'anna M, Mascarenhas M, Vargas A
et al. Comparison of cytokine levels in depressed, manic and euthymic patients
with bipolar disorder. J Affect Disord 2009; 116: 214–217.
5 Kunz M, Cereser KM, Goi PD, Fries GR, Teixeira AL, Fernandes BS et al. Serum levels
of IL-6, IL-10 and TNF-alpha in patients with bipolar disorder and schizophrenia:
differences in pro- and anti-inflammatory balance. Rev Brasil Psiquiatr 2011; 33:
268–274.
6 Deh M, Correll CU, Bobes J, Cetkovich-Bakmas M, Cohen D, Asai I et al. Physical
illness in patients with severe mental disorders. I. Prevalence, impact of medi-
cations and disparities in health care. World Psychiatry 2011; 10: 52–77.
7 De Hert M, Cohen D, Bobes J, Cetkovich-Bakmas M, Leucht S, Ndetei DM et al.
Physical illness in patients with severe mental disorders. II. Barriers to care,
monitoring and treatment guidelines, plus recommendations at the system and
individual level. World Psychiatry 2011; 10: 138–151.
8 Canetta S, Sourander A, Surcel HM, Hinkka-Yli-Salomaki S, Leiviska J, Kellendonk C
et al. Elevated maternal C-reactive protein and increased risk of schizophrenia in a
national birth cohort. Am J Psychiatry 2014; 171: 960–968.
9 Canetta SE, Bao Y, Co MD, Ennis FA, Cruz J, Terajima M et al. Serological doc-
umentation of maternal influenza exposure and bipolar disorder in adult off-
spring. Am J Psychiatry 2014; 171: 557–563.
10 Debnath M, Venkatasubramanian G, Berk M. Fetal programming of schizophrenia:
Select mechanisms. Neurosci Biobehav Rev 2015; 49C: 90–104.
11 Miller BJ, Buckley P, Seabolt W, Mellor A, Kirkpatrick B. Meta-analysis of
cytokine alterations in schizophrenia: clinical status and antipsychotic effects.
Biol Psychiatry 2011; 70: 663–671.
12 Ruth MR, Port AM, Shah M, Bourland AC, Istfan NW, Nelson KP et al. Consuming a
hypocaloric high fat low carbohydrate diet for 12 weeks lowers C-reactive protein,
and raises serum adiponectin and high density lipoprotein-cholesterol in obese
subjects. Metabolism 2013; 62: 1779–1787.
13 Pasco JA, Nicholson GC, Williams LJ, Jacka FN, Henry MJ, Kotowicz MA et al.
Association of high-sensitivity C-reactive protein with de novo major depression.
Br J Psychiatry 2010; 197: 372–377.
14 Wium-Andersen MK, Orsted DD, Nordestgaard BG. Elevated C-reactive protein
associated with late- and very-late-onset schizophrenia in the general population:
a prospective study. Schizophr Bull 2014; 40: 1117–1127.
15 Kegel ME, Bhat M, Skogh E, Samuelsson M, Lundberg K, Dahl ML et al. Imbalanced
kynurenine pathway in schizophrenia. Int J Tryptophan Res 2014; 7: 15–22.
16 Campbell BM, Charych E, Lee AW, Moller T. Kynurenines in CNS disease: regula-
tion by inflammatory cytokines. Front Neurosci 2014; 8: 12.
17 Vuksan-Cusa B, Sagud M, Jakovljevic M, Peles AM, Jaksic N, Mihaljevic S
et al. Association between C-reactive protein and homocysteine with the sub-
components of metabolic syndrome in stable patients with bipolar disorder and
schizophrenia. Nord J Psychiatry 2013; 67: 320–325.
18 Suvisaari J, Loo BM, Saarni SE, Haukka J, Perala J, Saarni SI et al. Inflammation
in psychotic disorders: a population-based study. Psychiatry Res 2011; 189:
305–311.
19 Stojanovic A, Martorell L, Montalvo I, Ortega L, Monseny R, Vilella E et al. Increased
serum interleukin-6 levels in early stages of psychosis: associations with at-risk
© 2016 Macmillan Publishers Limited
C-reactive protein in schizophrenia
BS Fernandes et al
563
mental states and the severity of psychotic symptoms. Psychoneuroendocrinology
2014; 41: 23–32.
20 Schwarz E, Guest PC, Steiner J, Bogerts B, Bahn S. Identification of blood-based
molecular signatures for prediction of response and relapse in schizophrenia
patients. Transl Psychiatry 2012; 2: e82.
21 Sarandol A, Kirli S, Akkaya C, Ocak N, Eroz E, Sarandol E. Coronary artery disease
risk factors in patients with schizophrenia: effects of short term antipsychotic
treatment. J Psychopharmacol 2007; 21: 857–863.
22 Meyer JM, McEvoy JP, Davis VG, Goff DC, Nasrallah HA, Davis SM et al. Inflam-
matory markers in schizophrenia: comparing antipsychotic effects in phase 1 of
the clinical antipsychotic trials of intervention effectiveness study. Biol Psychiatry
2009; 66: 1013–1022.
23 Loffler S, Loffler-Ensgraber M, Fehsel K, Klimke A. Clozapine therapy raises serum
concentrations of high sensitive C-reactive protein in schizophrenic patients. Int
Clin Psychopharmacol 2010; 25: 101–106.
24 Lin CC, Chang CM, Liu CY, Huang TL. Increased high-sensitivity C-reactive
protein levels in Taiwanese schizophrenic patients. Asia Pac Psychiatry 2013; 5:
E58–E63.
25 Kuo FC, Lee CH, Hsieh CH, Kuo P, Chen YC, Hung YJ. Lifestyle modification and
behavior therapy effectively reduce body weight and increase serum level of
brain-derived neurotrophic factor in obese non-diabetic patients with schizo-
phrenia. Psychiatry Res 2013; 209: 150–154.
26 Kraemer S, Minarzyk A, Forst T, Kopf D, Hundemer HP. Prevalence of metabolic
syndrome in patients with schizophrenia, and metabolic changes after 3 months
of treatment with antipsychotics--results from a German observational study. BMC
Psychiatry 2011; 11: 173.
27 Kliushnik TP, Siriachenko TM, Sarmanova ZV, Otman IN, Dupin AM, Sokolov RE.
[Changes of the level of serum antibodies to neuroantigens in patients with
schizophrenia during the treatment]. Zh Nevrol Psikhiatr Im S S Korsakova 2008;
108: 61–64.
28 Klemettila JP, Kampman O, Seppala N, Viikki M, Hamalainen M, Moilanen E et al.
Cytokine and adipokine alterations in patients with schizophrenia treated with
clozapine. Psychiatry Res 2014; 218: 277–283.
29 Joshi KB, Nillawar A, Thorat AP. Cardiovascular disease risk in schizophrenia
patients: a case control study. J Clin Diagn Res 2013; 7: 2694–2696.
30 Hope S, Dieset I, Agartz I, Steen NE, Ueland T, Melle I et al. Affective symptoms are
associated with markers of inflammation and immune activation in bipolar
disorders but not in schizophrenia. J Psychiatr Res 2011; 45: 1608–1616.
31 Hepgul N, Pariante CM, Dipasquale S, DiForti M, Taylor H, Marques TR et al.
Childhood maltreatment is associated with increased body mass index and
increased C-reactive protein levels in first-episode psychosis patients. PsycholMed
2012; 42: 1893–1901.
32 Henderson DC, Fan X, Copeland PM, Sharma B, Borba CP, Boxill R et al. Aripi-
prazole added to overweight and obese olanzapine-treated schizophrenia
patients. J Clin Psychopharmacol 2009; 29: 165–169.
33 Frydecka D, Misiak B, Pawlak-Adamska E, Karabon L, Tomkiewicz A, Sedlaczek P
et al. Interleukin-6: the missing element of the neurocognitive deterioration in
schizophrenia? The focus on genetic underpinnings, cognitive impairment and
clinical manifestation. Eur Arch Psychiatry Neurosci 2014.
34 Fernandez-Egea E, Bernardo M, Donner T, Conget I, Parellada E, Justicia A et al.
Metabolic profile of antipsychotic-naive individuals with non-affective psychosis.
Br J Psychiatry 2009; 194: 434–438.
35 Fawzi MH, Fawzi MM, Fawzi MM, Said NS. C-reactive protein serum level in
drug-free male Egyptian patients with schizophrenia. Psychiatry Res 2011; 190:
91–97.
36 Dickerson F, Stallings C, Origoni A, Vaughan C, Khushalani S, Yang S et al.
C-reactive protein is elevated in schizophrenia. Schizophr Res 2013; 143: 198–202.
37 Diaz FJ, Perez-Iglesias R, Mata I, Martinez-Garcia O, Vazquez-Barquero JL, de Leon
J et al. Possible effects of some antipsychotic drugs on C-reactive protein in a
drug-naive psychotic sample. Schizophr Res 2010; 121: 207–212.
38 De Berardis D, Conti CM, Marini S, Serroni N, Moschetta FS, Carano A. C-reactive
protein level and its relationship with suicide risk and alexithymia among newly
diagnosed, drug-naive patients with non-affective psychosis. Eur J Inflammation
2013; 11: 0–0.
39 Carrizo E, Fernandez V, Quintero J, Connell L, Rodriguez Z, Mosquera M et al.
Coagulation and inflammation markers during atypical or typical antipsychotic
treatment in schizophrenia patients and drug-free first-degree relatives. Schizophr
Res 2008; 103: 83–93.
40 Baptista T, Davila A, El Fakih Y, Uzcategui E, Rangel NN, Olivares Y et al. Similar
frequency of abnormal correlation between serum leptin levels and BMI before
and after olanzapine treatment in schizophrenia. Int Clin Psychopharmacol 2007;
22: 205–211.
41 Akanji AO, Ohaeri JU, Al-Shammri S, Fatania HR. Association of blood levels of
C-reactive protein with clinical phenotypes in Arab schizophrenic patients. Psy-
chiatry Res 2009; 169: 56–61.
Molecular Psychiatry (2016), 554 – 564
 C-reactive protein in schizophrenia
BS Fernandes et al
564
42 McNamara RK, Jandacek R, Rider T, Tso P. Chronic risperidone normalizes elevated
pro-inflammatory cytokine and C-reactive protein production in omega-3 fatty
acid deficient rats. Eur J Pharmacol 2011; 652: 152–156.
43 JP H. Handbook for systematic reviews of Interventions Version 5.1.0. The
Cochrane Collaboration 2011.
44 Egger M, Zellweger-Zahner T, Schneider M, Junker C, Lengeler C, Antes G. Lan-
guage bias in randomised controlled trials published in English and German.
Lancet 1997; 350: 326–329.
45 Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected
by a simple, graphical test. BMJ 1997; 315: 629–634.
46 Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS)
for schizophrenia. Schizophr Bull 1987; 13: 261–276.
47 Sistrom CL, Mergo PJ. A simple method for obtaining original data from published
graphs and plots. Am J Roentgenol 2000; 174: 1241–1244.
48 Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of
the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010; 25:
603–605.
49 Soeken KL, Sripusanapan A. Assessing publication bias in meta-analysis. Nurs Res
2003; 52: 57–60.
50 Bowden J, Tierney JF, Copas AJ, Burdett S. Quantifying, displaying and accounting
for heterogeneity in the meta-analysis of RCTs using standard and generalised Q
statistics. BMC Med Res Methodol 2011; 11: 41.
51 Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in
meta-analyses. BMJ 2003; 327: 557–560.
52 Lau J, Ioannidis JP, Schmid CH. Quantitative synthesis in systematic reviews. Ann
Int Med 1997; 127: 820–826.
53 Goto M, Sugimoto K, Hayashi S, Ogino T, Sugimoto M, Furuichi Y et al. Aging-
associated inflammation in healthy Japanese individuals and patients with
Werner syndrome. Exp Gerontol 2012; 47: 936–939.
54 Najjar S, Pearlman DM, Devinsky O, Najjar A, Zagzag D. Neurovascular unit
dysfunction with blood-brain barrier hyperpermeability contributes to major
depressive disorder: a review of clinical and experimental evidence. J Neuroin-
flammation 2013; 10: 142.
55 Uranova NA, Zimina IS, Vikhreva OV, Krukov NO, Rachmanova VI, Orlovskaya DD.
Ultrastructural damage of capillaries in the neocortex in schizophrenia. World J
Biol Psychiatry 2010; 11: 567–578.
56 Adami C, Sorci G, Blasi E, Agneletti AL, Bistoni F, Donato R. S100B expression in
and effects on microglia. Glia 2001; 33: 131–142.
57 Schwieler L, Larsson MK, Skogh E, Kegel ME, Orhan F, Abdelmoaty S et al.
Increased levels of IL-6 in the cerebrospinal fluid of patients with chronic schi-
zophrenia - significance for activation of the kynurenine pathway. J Psychiatry
Neurosci 2015; 40: 126–133.
58 Koshy Cherian A, Gritton H, Johnson DE, Young D, Kozak R, Sarter M. A
systemically-available kynurenine aminotransferase II (KAT II) inhibitor restores
Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)
Molecular Psychiatry (2016), 554 – 564
nicotine-evoked glutamatergic activity in the cortex of rats. Neuropharmacology
2014; 82: 41–48.
59 Schwarcz R, Bruno JP, Muchowski PJ, Wu HQ. Kynurenines in the mammalian
brain: when physiology meets pathology. Nat Rev Neurosci 2012; 13: 465–477.
60 Wu HQ, Okuyama M, Kajii Y, Pocivavsek A, Bruno JP, Schwarcz R. Targeting
kynurenine aminotransferase II in psychiatric diseases: promising effects of an
orally active enzyme inhibitor. Schizophr Bull 2014; 40: S152–S158.
61 Catts VS, Wong J, Fillman SG, Fung SJ, Shannon Weickert C. Increased expression
of astrocyte markers in schizophrenia: association with neuroinflammation. Aust N
Z J Psychiatry 2014; 48: 722–734.
62 Fillman SG, Cloonan N, Miller LC, Weickert CS. Markers of inflammation in the
prefrontal cortex of individuals with schizophrenia. Mol Psychiatry 2013; 18: 133.
63 Schroeter ML, Steiner J. Elevated serum levels of the glial marker protein S100B
are not specific for schizophrenia or mood disorders. Mol Psychiatry 2009; 14:
235–237.
64 Fernandes BS, Berk M, Turck CW, Steiner J, Goncalves CA. Decreased peripheral
brain-derived neurotrophic factor levels are a biomarker of disease activity in
major psychiatric disorders: a comparative meta-analysis. Mol Psychiatry 2014; 19:
749–751.
65 Fernandes BS, Steiner J, Berk M, Molendijk ML, Gonzalez-Pinto A, Turck CW et al.
Peripheral brain-derived neurotrophic factor in schizophrenia and the role of
antipsychotics: meta-analysis and implications. Mol Psychiatry 2014; doi:10.1038/
mp.2014.117 (e-pub ahead of print).
66 ON A, Jacka FN, Quirk SE, Cocker F, Taylor C, Oldenburg B et al. A shared fra-
mework for the common mental disorders and non-communicable disease:
key considerations for disease prevention and control. BMC Psychiatry 2015;
15: 15.
67 Frey BN, Andreazza AC, Houenou J, Jamain S, Goldstein BI, Frye MA et al. Bio-
markers in bipolar disorder: a positional paper from the International Society for
Bipolar Disorders Biomarkers Task Force. Aust N Z J Psychiatry 2013; 47: 321–332.
68 Fernandes BS, Gama C, Kauer-Sant'anna M, Lobato M, Belmonte-De-Abreu P,
Kapczinski F. Serum brain-derived neurotrophic factor in bipolar and unipolar
depression: a potential adjunctive tool for differential diagnosis. J Psychiatric Res
2009; 43: 1200–1204.
69 Fernandes BS, Gama CS, Cereser KM, Yatham LN, Fries GR, Colpo G et al. Brain-
derived neurotrophic factor as a state-marker of mood episodes in bipolar dis-
orders: a systematic review and meta-regression analysis. J Psychiatric Res 2011;
45: 995–1004.
70 Kapczinski F, Fernandes B, Kauer-Sant'anna M, Gama C, Yatham L, Berk M. The
concept of staging in bipolar disorder: the role of BDNF and TNF-alpha as bio-
markers. ACTA Neuropsychiatr 2009; 21: 272–274.
71 Davis J, Maes M, Andreazza A, McGrath JJ, Tye SJ, Berk M. Towards a classification
of biomarkers of neuropsychiatric disease: from encompass to compass. Mol
Psychiatry 2015; 20: 152–153.
© 2016 Macmillan Publishers Limited