Comparative effectiveness of Korean Medicine treatments for temporomandibular joint disorders : A systematic review and network meta-analysis

다국어 초록 (Multilingual Abstract)

Introduction: Temporomandibular joint disorders (TMDs) is characterized by pain and functional restrictions in the temporomandibular joint that interfere with daily activities such as speaking, chewing, and breathing. A range of Korean Medicine modalities offer potential solutions for TMD management. A systematic review and network meta-analysis was conducted to evaluate and compare the effectiveness of various Korean Medicine interventions in addressing TMDs.

Methods: Electronic searches encompassed various Korean Medicine interventions, including acupuncture, moxibustion, cupping, herbal medicine, and chuna, specifically targeting TMDs. The search was conducted on November 12, 2023 across multiple databases, including MEDLINE, CENTRAL (Cochrane Central Register of Controlled Trials), EMBASE, AMED (Allied and Complementary Medicine Database), CINAHL (Cumulative Index to Nursing and Allied Health Literature), OASIS (Oriental Medicine Advanced Searching Integrated System), KISS (Korean studies Information Service System), RISS (Research Information Service System), KMbase (Korean Medical Database), KCI (Korea Citation Index), CNKI (China National Knowledge Infrastructure), and CiNii. Network meta-analysis was performed using frequentist methods, and a random effects model was applied based on the clinical diversity of the disease. Primary outcome focused on pain intensity assessed through validated self-reported tools such as visual analogue scale and numerical rating scale, with secondary outcomes including maximum mouth opening and adverse event incidence. Sensitivity analyses were conducted to enhance robustness, and the Confidence In Network Meta-Analysis methodology were employed to assess evidence quality.

Results: A total of 4,172 records were identified from the searched databases. After screening the records, 45 studies were finally included in qualitative synthesis, and 36 of 45 studies were included in quantitative synthesis. The following treatments were compared in terms of effectiveness as assessed by visual analogue scale: (1) acupuncture, (2) acupotomy, (3) electroacupuncture, (4) exercise, (5) laser therapy, (6) manipulation, (7) no treatment (8) sham treatment, (9) scalp acupuncture, and (10) splint. Compared to sham treatment, there were significant differences in acupotomy (Mean difference (MD) -5.071, 95% Confidence interval (CI) -7.366 to -2.777) and acupuncture (MD -1.181, 95% CI -2.276 to -0.087). According to the SUCRA, acupotomy was considered the most optimal treatment, followed by electroacupuncture. However, the effect size may have been exaggerated because no studies directly compared acupotomy and electroacupuncture with sham treatment, and each was directly linked to one study comparing acupuncture in the network plot. A total of 12 studies reported incidence of adverse events, and none were serious adverse events.

Conclusion: This study aimed to evaluate the comparative effectiveness and acceptability of Korean Medicine treatments for TMDs through a systematic review and network meta-analysis. Acupotomy showed significant superiority in pain relief compared to sham treatment, but the level of evidence was low as it was based on a small number of studies. Therefore, future rigorous, well-designed, large-scale randomized controlled trials are needed to draw clear conclusions.

Keywords: Temporomandibular joint disorders, Korean Medicine, Systematic review, Network Meta-analysis

목차 (Table of Contents)

• Abstract
• Ⅰ. Introduction 1
• Ⅱ. Methods 4
• 2.1. Protocol and registration 4
• 2.2. Information sources and search strategy 4
• 2.3. Eligibility criteria 4
• 2.3.1. Type of participants 5
• 2.3.2. Type of interventions 5
• 2.3.3. Type of outcome measurements 6
• 2.3.4. Type of study design 6
• 2.3.5. Other criteria 6
• 2.4. Study selection 7
• 2.5. Data management 7
• 2.6. Data extraction 7
• 2.7. Risk of bias assessment 8
• 2.8. Data synthesis 8
• 2.8.1. Pairwise meta-analysis 8
• 2.8.2. Network meta-analysis 9
• 2.8.2.1. Data coding 10
• 2.8.2.2. Network model 10
• 2.8.2.3. Connectivity test 10
• 2.8.2.4. Transitivity test 10
• 2.8.2.5. Consistency test 11
• 2.8.2.6. Effect sizes 11
• 2.8.3. Additional analysis 12
• 2.9. Missing data management 12
• 2.10. Reporting bias assessment 12
• 2.11. Certainty of evidence 13
• Ⅲ. Results 14
• 3.1. Study selection 14
• 3.2. Characteristics of the included studies 15
• 3.2.1. Study design 16
• 3.2.2. Participant characteristics 16
• 3.2.3. Intervention treatment 17
• 3.2.4. Outcome measures 17
• 3.3. Risk of bias (RoB 2.0) included studies 28
• 3.3.1. Bias arising from the randomization process 28
• 3.3.2. Bias due to deviations from intended interventions 28
• 3.3.3. Bias due to missing outcome data 29
• 3.3.4. Bias in measurement of the outcome 29
• 3.3.5. Bias in selection of the reported result 30
• 3.3.6. Overall bias 30
• 3.4. Comparative effectiveness 35
• 3.4.1. Primary outcome (pain intensity, VAS) 35
• 3.4.1.1. Network model establishment and connectivity test 35
• 3.4.1.2. Transitivity test 37
• 3.4.1.3. Consistency test 37
• 3.4.1.4. Effect size 37
• 3.4.1.5. Publication bias 43
• 3.4.2. Secondary outcome (range of motion, MMO) 44
• 3.4.2.1. Network model establishment and connectivity test 44
• 3.4.2.2. Transitivity test 46
• 3.4.2.3. Consistency test 46
• 3.4.2.4. Effect size 46
• 3.4.2.5. Publication bias 52
• 3.4.3. Secondary outcome (incidence of adverse events) 53
• 3.5. Sensitivity analysis 53
• 3.5.1. Excluding studies using estimated values and assessed as high risk of overall bias 53
• 3.5.1.1. Primary outcome (pain intensity, VAS) 53
• 3.5.1.2. Secondary outcome (range of motion, MMO) 59
• 3.5.2. Short-term effect 64
• 3.5.2.1. Primary outcome (pain intensity, VAS) 64
• 3.5.2.2. Secondary outcome (range of motion, MMO) 69
• 3.6. Level of evidence 74
• 3.6.1. Primary outcome (pain intensity, VAS) 74
• 3.6.2. Secondary outcome (range of motion, MMO) 75
• Ⅳ. Discussion 85
• 4.1. Summary of findings 85
• 4.2. Comparison with previous study – Clinical perspectives 87
• 4.3. Comparison with previous study – Methodological perspectives 88
• 4.4. Implication for further study and clinical practice 88
• 4.5. Strength and Limitations 89
• Ⅴ. Conclusion 92
• Ⅵ. References 93
• 국문초록 101
• Appendix 1. PRISMA NMA checklist 104
• Appendix 2. Search strategy 110
• Appendix 3. Excluded studies and reason after full-text review 115
• Appendix 4. Forest plot showing node-split analysis for pain intensity 120
• Appendix 5. Forest plot showing node-split analysis for range of motion 121