Simplified MRI scoring system for TMJ assessment in juvenile idiopathic arthritis - 21/04/26

Doi : 10.1016/j.jormas.2026.102804

Steven Gernandt ^a,^⁎ , Céline Habre ^b, Sylviane Hanquinet ^b, Gregory S. Antonarakis ^c, Raffaella Carlomagno ^d, Michaël Hofer ^e, Delphine S. Courvoisier ^f, Paolo Scolozzi ^a, Seema Toso ^b

^a Division of Oral and Maxillofacial Surgery, Department of Surgery, Geneva University Hospitals, Geneva, Switzerland

^b Pediatric Radiology Unit, Radiology Division, Diagnostic Department, Children’s Hospital, Geneva University Hospitals, Geneva, Switzerland

^c Division of Orthodontics, University Clinics of Dental Medicine, University of Geneva, Geneva, Switzerland

^d Rheumatology, Immunology and Allergology Unit, Department of Pediatrics, Lausanne University Hospital, Lausanne, Switzerland

^e Department of Paediatrics, Hôpital Riviera-Chablais, Rennaz, Switzerland

^f Division of Rheumatology, University Hospitals of Geneva, Geneva, Switzerland

^⁎ Corresponding author at: Division of Oral and Maxillofacial Surgery, Department of Surgery, University Hospitals of Geneva, Switzerland, Rue Gabrielle Perret-Gentil 4, 1211 Geneva, Switzerland. Division of Oral and Maxillofacial Surgery Department of Surgery University Hospitals of Geneva, Switzerland Rue Gabrielle Perret-Gentil 4 Geneva 1211 Switzerland

Abstract

Objectives

The temporomandibular joint (TMJ) is a small, yet complex joint. MRI interpretation of TMJ involvement in children with juvenile idiopathic arthritis (JIA) is challenging but essential for the diagnosis and treatment monitoring. We propose a simplified and reproducible MRI scoring system to facilitate both initial diagnosis as well as longitudinal follow-up.

Methods

Two pediatric radiologists independently scored 139 TMJ MRIs from 66 children with JIA using a detailed literature-based score (Score A) and a newly developed simplified score (Score B). For all 278 TMJs, reading time was recorded, and interrater reliability was assessed using intraclass correlation coefficients (ICCs). The primary objective was to assess interrater reliability, with construct validity evaluated in an exploratory manner.

Results

The mean patient age was 12.0 years (SD: 3.7). For Score A, interrater agreement showed moderate agreement for the acute score with an ICC of 0.67–0.77 and good agreement for the chronic score (ICC: 0.75–0.86). In contrast, Score B demonstrated excellent agreement for the acute score (ICC: 0.94–0.99) and good agreement for the chronic score (ICC: 0.85–0.90). Both scores showed similar associations with relevant clinical criteria, including the decision to treat. Image interpretation using Score B required 83% less time than with Score A.

Conclusion

These findings support the reproducibility and feasibility of the simplified score (Score B) for evaluating TMJ involvement in children with JIA, although further studies are required to evaluate its diagnostic performance, external validity, and impact on clinical decision-making.

Le texte complet de cet article est disponible en PDF.

Keywords : Temporomandibular joint, Children, Magnetic resonance imaging, Juvenile idiopathic arthritis