Litt MD, Kreutzer D, Shafer D.

Alterations of inflammatory cytokines in temporomandibular dysfunction as a function of cognitive-behavioral treatment.

University of Connecticut Health Center, Department of Behavioral Sciences—MC3910, 263 Farmington Ave, Farmington, CT 06030. Litt@nso.uchc.edu

PURPOSE: To determine whether Cognitive-Behavioral treatment for temporomandibular dysfunction (TMD) pain acts in part by altering inflammatory function, as measured by levels of circulating cytokines.

METHODS: Thirty men and women suffering from jaw pain for at least 3 months and never treated were randomly assigned to either a standard conservative care group (STD; n=15) or to standard care plus cognitive-behavioral treatment (STD+CBT; n=15). Treatment took place one hour per week for 6 weeks. STD consisted of placement of a flat plane splint plus prescription of nonsteroidal anti-inflammatory medication. The STD+CBT condition added relaxation training, stress management, cognitive restructuring intended to promote self-efficacy, and habit modification. Cytokine measures examined included IL-10, IL-6, TNFα, IL-3, IL-2, and IL-10. Patients were followed at post-treatment (6 weeks), 12 weeks, 18 weeks, and 24 weeks. Cytokine assays of patient plasma were conducted using high sensitivity ELISAs. Detectable levels were found for IL-10, IL-6, and TNFα at baseline, and at all follow-ups for 27 subjects.

RESULTS: Repeated measures ANOVA indicated a significant Treatment X Time effect on reported pain over the follow-up period showing ad advantage for STD+CBT over STD alone. Results of repeated measures ANCOVA yielded significant Treatment X Time effects on IL-6 (p < .05) and on TNFα (p < .05), such that these cytokine levels tended to decrease over time in the STD+CBT patients, but did not decrease in the STD patients. IL-10 levels tended to increase over time in both groups. Separate linear mixed model analyses were conducted in which pain was the repeated dependent variable, and each of the three cytokine levels were independent variables. A significant Treatment X Time X Cytokine level effect emerged for TNFα (p < .05), and significant Time X Cytokine effects emerged for both IL-6 and IL-10 (p < .05 for each cytokine), indicating that reported pain was in part a function of inflammatory and ant-inflammatory cytokine levels. These results are consistent with a pattern of gradually decreasing pro-inflammatory action in IL-6 and TNFα, and gradually increasing anti-inflammatory action from IL-10, coinciding with pain levels over time, and varying by Treatment type.

CONCLUSIONS: Results are seen as supporting the idea that Cognitive-Behavioral treatments may act in part by effecting changes in cytokine networks and inflammation in chronic pain states.

 

Back