Craniofacial growth disturbance is related to temporomandibular joint abnormality in patients with juvenile idiopathic arthritis, but normal facial profile was also found at the 27-year follow-up

Abstract

Objectives: To assess the long-term outcome of craniofacial morphology related to disease variables and temporomandibular joint (TMJ) involvement as demonstrated with computed tomography (CT) and magnetic resonance imaging (MRI) in adult patients with juvenile idiopathic arthritis (JIA).

Methods: Sixty of 103 patients participated in a re-examination on average 27 years after baseline. Craniofacial morphology, with emphasis on size and position of the mandible, was assessed in lateral cephalographic images and related to disease variables and TMJ involvement by CT and MRI. Definitions of craniofacial growth disturbances were based on measurements outside 2 SD from the mean of healthy adult controls.

Results: Sagittal craniofacial growth disturbances were found in 57% and micrognathia in 27% of the 60 patients. Of those with JIA TMJ involvement, 70% had some form of growth disturbance. Micrognathia occurred only in patients with bilateral TMJ involvement. The bilateral TMJ group had significantly different craniofacial morphology than healthy controls and patients without TMJ involvement. Growth disturbances and TMJ involvement were present in all subtypes of JIA, except for one subtype comprising one patient. Patients with growth disturbances had more severe disease than patients with normal craniofacial growth, regarding both present and previous disease activity. Unexpectedly, half of the patients without craniofacial growth disturbances also had TMJ involvement, many from before the age of 12.

Conclusions: Craniofacial growth disturbances were found to be frequent in adult JIA patients, especially in those with bilateral TMJ involvement. However, growth disturbances did not always follow TMJ involvement, not even when affected early.

Acknowledgements

We thank Dr Odd Vinje and Dr Jørn Thoen for valuable advice and help with data collection, Dr Leiv Sandvik for help with the statistical analysis, and Håkon Størmer for help with the illustrations. This work was supported by grants from the University of Oslo, Norway. LZA and MGF are financed by PhD stipends from the University of Oslo, Norway.