Inflammatory processes and neural–immune interactions are implicated in the pathogenesis of psychiatric disorders.1, 2, 3 Bipolar disorder is one of the psychiatric disorders with the most severe burden of illness,4 not only because of its effects on mood and cognition but also because of its association with clinical comorbidities such as obesity, diabetes, and cardiovascular disease. Such an association led to the suggestion that bipolar disorder is a systemic disease with an inflammatory component that could at least partly account for common risk pathways in bipolar disorder and comorbid conditions.5 In preclinical and clinical studies,6, 7, 8 concentrations of pro-inflammatory cytokines were found to be increased in bipolar disorder in both the brain and the peripheral blood. Additionally, results from several longitudinal studies9, 10, 11, 12, 13 showed that people with inflammatory clinical pathologies are at an increased risk of developing a psychiatric event in the future.
C-reactive protein (CRP) is an acute-phase protein produced in response to an inflammatory stimulus and is mostly induced by the pro-inflammatory cytokines interleukin-6 and interleukin-1β.3 Increased concen-trations of CRP have been reported in meta-analyses of cross-sectional studies in schizophrenia2 and in major depressive disorder.14 Raised concentrations of CRP increase the risk of a first episode of depression,9 schizophrenia,12 and bipolar disorder.13 Findings from two mendelian randomisation studies13, 15 suggested a causal relation between increased CRP concentrations and the occurrence of bipolar disorder; however, a longitudinal study16 published in 2016 reported no association between CRP concentrations in childhood and risk of manic symptoms in young adulthood.
Several clinical studies17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36 have investigated the association between CRP concentration in the blood and different mood states in bipolar disorder. Although most studies found an increase in CRP concentration during manic states, the data regarding depressive and euthymic states are less certain, with both positive and null findings. Whether CRP concentrations are associated with the severity of mania or depression, or with age and length of illness, is also uncertain, and such knowledge could shed light on the importance of CRP in the context of neuroprogression.5 Regarding changes in CRP concentrations after amelioration of symptoms of index manic or depressive episodes, the data are varied, with studies showing increase, decrease, or no changes in CPR concentrations.20, 32, 33, 34, 37, 38, 39, 40, 41, 42 A meta-analysis of 11 cross-sectional studies of CRP concentrations in bipolar disorder43 found that CRP concentrations increased in mania and euthymia but not in depression. However, owing to the small number of studies included, it did not analyse effects of any moderators such as severity of symptoms or the effects of treatment.
Research in context
Evidence before this study
We searched MEDLINE, the Cochrane Library, Scopus, and Web of Knowledge for peer-reviewed articles on C-reactive protein (CRP) concentrations in bipolar disorder from inception of the database to Aug 14, 2016, with no language restrictions. The search term was “(CRP OR C-reactive protein OR hsCRP OR hs-CRP) AND (bipolar OR psychosis OR mania OR manic)”. We identified a meta-analysis on CRP concentrations in bipolar disorder that included 11 cross-sectional studies. Although the authors of this meta-analysis found some evidence of increased CRP concentrations in mania and euthymia, the number of studies included in each mood state (ie, mania, depression, and euthymia) was small and therefore no subgroup analysis was done. The authors reported no changes in CRP concentrations in depression, probably because only four studies involving this subgroup of patients were included. Longitudinal changes in CRP concentrations were not included. We identified 16 additional studies of CRP concentrations in bipolar disorder. However, most of these studies and those included in the meta-analysis had small sample sizes with conflicting results.
Added value of this study
To our knowledge, this study is the first to investigate the association between CRP concentrations and bipolar disorder that takes into consideration BMI, severity of mood episodes, and use of psychiatric medications. We were also the first to analyse dynamic changes in CRP concentrations with psychiatric medications after an index manic or depressive episode. We did five meta-analyses on CRP concentrations across the mood spectrum in bipolar disorder and included 27 studies with 84 093 participants. We found that CRP concentrations are moderately increased in people with bipolar disorder during depression and euthymia and more substantially increased during mania. In manic and depressive states, CRP concentrations are higher in people who have not taken psychiatric medication for at least 1 week. The extent of the increases in peripheral CRP concentrations in mania and depression showed no association with the severity of manic or depressive symptoms. We noted a moderate decrease in CRP concentrations with achievement of euthymia after an index manic episode, and a small decrease after an index depressive episode.
Implications of all the available evidence
Our findings provide further evidence for an increased inflammatory state in bipolar disorder across the mood states, with inflammation being particularly high during mania. Treatment of manic and depressive episodes with psychiatric medications is associated with a decrease in systemic inflammation, as shown by the decrease in CRP concentrations. Longitudinal studies are needed to clarify the association between CRP concentrations and the development of bipolar disorder.
In this study, we aimed to assess peripheral CRP concentrations in bipolar disorder across the mood states, whether concentrations are correlated with the severity of mood symptoms, and whether CRP concentrations change after resolution of an index mood episode.