Glutamate modulates temporomandibular joint bone tissue resorption in patients with early rheumatoid arthritis

Abstract: The general aim was to investigate the occurrence and development of TMJ bone tissue resorption and to identify predictors for progression during the two first years with rheumatoid arthritis. The thesis comprises one cross-sectional part and one longitudinal part. The aim of the cross-sectional part was to investigate whether there was an association between the radiographic sign (TMJ erosions) or the clinical sign (crepitus) of TMJ bone tissue resorption and glutamate and if that relationship was influenced by serotonin, tumor necrosis factor, interleukin-1β, interleukin-6, vascular endothelial growth factor, C-reactive protein, erythrocyte sedimentation rate, rheumatoid factor or sex hormones. The first aim of the longitudinal part was to investigate whether TMJ erosions or crepitus, glutamate, inflammatory mediators and markers changed over time and if the change in TMJ erosions was related to these factors. The second aim was to investigate the predictive value of TMJ erosions, glutamate and crepitus as well as the inflammatory mediators and markers for change in TMJ erosions over two years after diagnosis. Both parts of the thesis comprised the same group of patients with recent RA diagnosis. They were 47 at first visit and 41 attended the two year follow-up. Similar examination protocols were used for the cross-sectional and the longitudinal part. The subjects were examined for TMJ erosions on cone-beam computerized tomographs and the clinical examination included the articular sound of crepitus. Blood samples were obtained and analyzed for glutamate, serotonin, tumor necrosis factor, interleukin-1β, interleukin-6 and vascular endothelial growth factor, C-reactive protein, erythrocyte sedimentation rate and rheumatoid factor, as well as for estradiol and testosterone. Significance of univariate correlations was calculated with Spearman rank correlation coefficient. To test the significance of the differences between visits the Friedman ANOVA on ranks was used. Multivariate regression was used to investigate the relative influence by independent variables on the presence, development or prediction of the TMJ erosions. Interaction analysis was performed by including an interaction term in addition to the main effects in the multiple linear regression model. The significance of the differences between groups was calculated with Fisher´s exact test for variables on nominal scale and Mann-Whitney U-test for variables on ordinal scale. A probability level of less than 0.05 was considered significant in all analyses. Seventy‐two percent of the patients showed TMJ erosions at the beginning of the study. TMJ erosion score was positively correlated to both glutamate and crepitus. There was also a positive correlation between erosions and C-reactive protein in the patients with estradiol levels ≥ 65 pmol/L. Multiple stepwise regression showed that TMJ crepitus (p < 0.001), glutamate (p = 0.005) and erythrocyte sedimentation rate (p = 0.010) explained 41% of the variation in the dependent variable TMJ bone erosion score (p < 0.001). Glutamate was positively correlated to TMJ erosion score in the patients without TMJ crepitus, with C-reactive protein < 3mg/L, C-reactive protein < 3 mg/L and estradiol < 50 pmol/L as well as C-reactive protein < 3 mg/L and testosterone ≤ 1.2 pmol/L. Progression of the erosion score was found in 29% of the patients and 43% showed regression during the first two years after RA diagnosis. The progression of erosion score was explained to 34% by testosterone (p = 0.012) and interleukin‐1β (p = 0.024). Patients with higher than median level of glutamate at the first visit showed more reduction in erosion score than those with lower level (p = 0.035). TMJ erosion score at the first visit was negatively correlated to progression of the score (rs = -0.64, p < 0.001). The multivariate analysis showed that erosion score was the only significant predictive factor that explained 53% of the regression (p < 0.001). TMJ bone tissue resorption was found to be present in a majority of patients with early RA. Glutamate modulates TMJ bone tissue resorption in the early phase of disease with absence of crepitus and under the influence by systemic inflammation and sex hormones. Regression in TMJ bone tissue resorption was the most frequent finding over a two-year period directly after diagnosis. Progression however seems to be associated with both inflammatory and non-inflammatory mechanisms while progression of crepitus is associated with inflammation when glutamate is increased. Regression of TMJ bone tissue resorption could be predicted by early TMJ erosions and glutamate. The results indicate that TMJ bone tissues resorption can be expected early in RA