Elevated levels of C-reactive protein (CRP) have repeatedly been observed in schizophrenia (SZ) and related disorder but without clear description of the associated clinical variables. The objectives of this study were:

– to determine the prevalence of abnormal CRP levels in an observational sample of patients with SZ or schizoaffective disorders;

– to identify the clinical variables associated with elevated CRP levels as well as the effects of treatments.

Two hundred and nineteen stable patients with schizophrenia or schizoaffective disorder (mean age = 31.6 years, 75.3% male gender) were systematically included in the network of FondaMental Expert Center for schizophrenia and assessed with a dedicated electronic medical records including the Structured Clinical Interview for DSM-IV Axis I Disorders and validated scales for depressive and psychotic symptomatology. High sensitivity CRP (hs-CRP) was measured with an assay using nephelometry (Dade Behring). Abnormal CRP level was defined by levels > 3 mg/L. Metabolic syndrome was defined according to the International Diabetes Federation.

Overall, 63 patients (28.8%) were found to have abnormal CRP levels. In univariate analysis, abnormal CRP levels were found to be significantly associated with the DSM-IV-TR schizophrenia diagnosis compared to schizoaffective disorder (32.6% vs. 10.5%, P = 0.006). This association remained significant after adjustment for BMI (P = 0.007) and antidepressants (P = 0.043). Abnormal CRP levels were also found to be significantly associated with BMI (P < 0.0001), hypertriglyceridemia (P = 0.0015), high waist circumference (P < 0.0001), metabolic syndrome (P = 0.0011) and abdominal obesity (P < 0.0001), while current tobacco status, hypertension or high fasting glucose were not (all P > 0.05). All patients were treated by antipsychotics. Patients treated by antidepressant were found to have less abnormal CRP levels than others (P = 0.01), contrary to those treated by mood stabilizing agents (P > 0.05).

CRP may be considered as a biomarker of interest to differentiate schizophrenia from schizoaffective disorder, and as a marker of inflammation induced by perivisceral fat. Treatment with antidepressant appears as a protective anti-inflammatory agent.