Elsevier

European Journal of Radiology

Volume 85, Issue 12, December 2016, Pages 2225-2230
European Journal of Radiology

Real-time MRI of the temporomandibular joint at 15 frames per second—A feasibility study

https://doi.org/10.1016/j.ejrad.2016.10.020Get rights and content

Abstract

The purpose of this study was to develop and evaluate a novel method for real-time MRI of TMJ function at high temporal resolution and with two different contrasts. Real-time MRI was based on undersampled radial fast low angle shot (FLASH) acquisitions with iterative image reconstruction by regularized nonlinear inversion. Real-time MRI movies with T1 contrast were obtained with use of a radiofrequency-spoiled FLASH sequence, while movies with T2/T1 contrast employed a gradient-refocused FLASH version. TMJ function was characterized in 40 randomly selected volunteers by sequential 20 s acquisitions of both the right and left joint during voluntary opening and closing of the mouth (in a medial, central and lateral oblique sagittal section perpendicular to the long axis of the condylar head). All studies were performed on a commercial MRI system at 3 T using the standard head coil, while online reconstruction was achieved with a bypass computer fully integrated into the MRI system. As a first result, real-time MRI studies of the right and left TMJ were successfully performed in all 40 subjects (80 joints) within a total examination time per subject of only 15 min. Secondly, at an in-plane resolution of 0.75 mm and 5 mm section thickness, the achieved temporal resolution was 66.7 ms per image or 15 frames per second. Thirdly, both T1-weighted and T2/T1-weighted real-time MRI movies provided information about TMJ function such as disc position, condyle mobility and disc-condyle relationship. While T1 contrast offers a better delineation of structures during rapid jaw movements, T2/T1 contrast was rated superior for characterizing the articular disc. In conclusion, the proposed real-time MRI method may become a robust and efficient tool for the clinical assessment of TMJ function.

Introduction

Temporomandibular disorders (TMD) are a frequent cause of pain in the orofacial system. Approximately 5%–12% of the total population is affected by this clinical condition, which involves pain as well as alterations of the masticatory muscles and the temporomandibular joint (TMJ) [1]. Because a successful treatment of TMD depends on an adequate diagnosis, the visualization of TMJ alterations such as abnormal disc-condyle relationship and degenerative changes usually involves magnetic resonance imaging (MRI). However, static images as acquired by conventional MRI do not cover the proper physiologic process, i.e. the active, muscularly guided TMJ function [2], [3]. This is because conventional scans suffer from measuring times of minutes, so that sequential scans during several passively opened mouth positions are typically recorded with use of a bite block [3]. Accordingly, static views of TMJ internals can only be derived before or after disc reduction [4]. Dynamic imaging of the TMJ was previously attempted on a single patient [5] but hampered by insufficient spatial and temporal resolution as required for the diagnosis of TMD.

Recently, a novel real-time MRI technique based on highly undersampled radial fast low angle shot (FLASH) acquisitions with iterative image reconstruction by regularized nonlinear inversion (NLINV) was developed [6], which allows for dynamic visualization of arbitrary physiologic processes up to very high frame rates. An earlier MRI study of TMJ function using 1.67 s acquisitions in combination with a sliding-window reconstruction only achieved a nominal resolution of 3 frames per second (fps), but promised valuable dynamic information about the disc and condyle mobility [7].

The aim of the present study is to report and evaluate further advances of this technology. First, the temporal resolution for TMJ studies could be improved to 66.7 ms per image or 15 fps, which more reliably captures the often rapid and discontinuous TMJ movements seen in patients, and second, the approach now offers two different contrasts, namely T1 contrast for a radiofrequency-spoiled FLASH sequence [8] and T2/T1 contrast for a gradient-refocused FLASH version [9]. T1-weighted images exhibit excellent anatomic detail because of the use of very short echo times with negligible sensitivity to susceptibility differences. These problems are unavoidable at air-tissue interfaces as, for example, in the vicinity of the ear canal and therefore may affect the visibility of neighboring structures. On the other hand, MRI T2 contrast, which also dominates T2/T1-weighted images as obtained here, is generally known for its sensitivity to pathologic tissue alterations which are often characterized by prolonged T2 relaxation times. For example, in TMD T2-weighted images have been recommended for the detection of synovial fluid and joint effusion [10], [11]. Here we studied a group of unselected volunteers in order to evaluate feasible experimental conditions and potential clinical settings for real-time MRI of TMJ function.

Section snippets

Subjects

This study was approved by the ethics committee of the Göttingen University. The subject group consisted of randomly selected volunteers as the central aim was to explore the feasibility of a novel real-time MRI method at high spatiotemporal resolution and the utility of two different contrasts regardless of TMJ condition. A total of 43 participants (18 females, 22 males) between 23 and 51 years of age (mean 28.6 ± 5.6 years) were recruited (students and members of the local University). Three

Results

Real-time MRI of the right and left TMJ at 15 fps and with T1 or T2/T1 contrast was successfully performed in all 40 subjects (80 joints). Because the entire examination could be completed within 15 min, real-time MRI appears as a fast and robust method in terms of practical performance.

Discussion

Medical imaging procedures such as invasive arthroscopy and arthrography using computed tomography [21] have successfully been applied to characterize TMJ alterations and joint related complaints, in particular when offering reproducible and accurate visualization of soft and hard tissues [20]. Arthrographic imaging allows for the evaluation of the bony structures of the TMJ in many planes, but suffers from difficulties in positioning of the patient and increased exposure to radiation [21]. To

Conclusions

In conclusion, the present study advances real-time MRI of physiologic TMJ movements to a temporal resolution of 15 fps, while retaining access to high spatial resolution. The use of two sequence versions offering either T1 or T2/T1 contrast in a group of unselected volunteers demonstrates access to valuable information about TMJ function such as disc position, condyle mobility and disc-condyle relationship. While T2/T1 contrast revealed statistically significant better ratings regarding image

Conflicts of interest

JF is co-inventor of a patent covering the real-time MRI technique used in this study.

Acknowledgments

AAJ received financial support by the DZHK (German Center for Cardiovascular Research) and by the BMBF (German Ministry of Education and Research).

References (30)

  • S. Zhang et al.

    Real-time magnetic resonance imaging of the temporomandibular joint dynamics

    Open Med. Imaging J.

    (2011)
  • C. Kober et al.

    Do we need real-time MRI for diagnosis of temporomandibular joint disorders?

    Int. J. Comput. Dent.

    (2011)
  • M. Uecker et al.

    Real-time MRI at a resolution of 20 ms

    NMR Biomed.

    (2010)
  • O. Kling et al.

    Real-time MRI as a new technique for the functional assessment of the temporomandibular joint

    J. Craniomandib. Funct.

    (2013)
  • V.B. Roeloffs et al.

    Spoiling without additional gradients −Radial FLASH MRI with randomized RF phases

    Magn. Reson. Med.

    (2016)
  • Cited by (33)

    • Diagnostic value of dynamic magnetic resonance imaging of temporomandibular joint dysfunction

      2021, European Journal of Radiology Open
      Citation Excerpt :

      Further, some patients had difficulty to open and close the mouth in 14 steps of approximately the same distance. Developments towards faster data acquisition such as real-time sequences [18] with simultaneously high quality images could eliminate these weaknesses. There are several limitations of our study which need to be mentioned.

    View all citing articles on Scopus
    View full text