LAUSR

C-reactive protein (CRP), a humoral component of the innate immune system with important functions in host-defense, is extensively used as a sensitive biomarker of systemic inflammation. During inflammation, hepatocyte-derived CRP rises dramatically in the blood due to increased interleukin-6 (IL-6) levels. Reliable detection of CRP in saliva, instead of blood, would offer advantages regarding sampling procedure and availability but using saliva as a diagnostic body fluid comes with challenges. The aims of this study were to evaluate associations between salivary CRP, total protein levels in saliva and serum CRP. Furthermore, we examined associations with plasma IL-6, body mass index (BMI), tobacco smoking and age. Salivary CRP was investigated by ELISA in 107 middle-aged participants from the general population. We employed spectrophotometric determination of total protein levels. Correlation analyses were used for associations of salivary CRP with serum CRP (turbidimetry), plasma IL-6 (Luminex®), BMI and smoking habits. Salivary median CRP was 68% higher (p=0.009), and total protein levels were 167% higher (p<0.0001), in morning compared to evening saliva. The correlation coefficients between serum and salivary CRP were low to moderate, but stronger for evening than morning saliva. Plasma IL-6 correlated significantly with serum CRP (rs=0.41, p<0.01), but not with morning or evening salivary CRP. Non-smokers showed 103% higher salivary CRP levels (p=0.015), whereas serum CRP was independent of smoking status. As opposed to CRP in serum, salivary CRP was not associated with BMI. Salivary CRP was 90% higher among the age interval 60–69 years compared to subjects aged 45–59 (p=0.02) while serum CRP levels did not differ between the age groups. In conclusion, CRP in saliva did not straightforwardly reflect serum concentrations. This raises questions regarding adequate reflection of biological events. The pronounced diurnal salivary CRP pattern accentuates the importance of standardizing the time-point of sampling.