US is successfully used to identify inflammatory changes in large joints [23, 24]. It is a noninvasive, less expensive technique than MRI. The use of US for TMJ exploration is currently controversial in the literature [8, 26]. This is due to the fact that US is a method that strictly depends on the operator and on the frequency of the transducer that is being utilized. High frequency transducers have improved the diagnostic quality of US [27]. US is an easy technique to diagnose TMJ disc displacements. In identifying the position of the disc, there are authors which argue that US is a useful diagnostic method, while others have shown that the diagnostic accuracy of US is relatively low [25, 26, 28].
Detection of inflammatory changes in the TMJ (synovitis, effusion) can be carried out directly with US by identifying the fluid collection as a hypoechoic area or indirectly by quantifying the degree of capsular distension. Very small differences (0.2–0.3 mm) in measuring capsular distension significantly influence US sensitivity and specificity as a diagnostic test for TMJ effusion [29].
MRI investigation is the reference standard in the diagnosis of inflammatory diseases of the TMJ [20]. TMJ changes that occur in inflammatory diseases have similar manifestations as in other joints: subchondral bone modifications and synovial collection with synovitis. Bone pathological changes are represented by subchondral edema, subchondral erosions and bone resorption with shape modifications of the condyle and of the temporal fossa. MRI examination is the only imaging method that can highlight subchondral edema, which is one of the most important sign in the early stages of the disease. Although not pathognomonic (it can also occur in mechanical dysfunctions), subchondral edema is the result of inflammatory osteitis that occurs in the acute phase and precedes the onset of subchondral erosions. The MRI examination shows a diffuse T2 hypersignal in the subchondral bone marrow. Early signs of TMJ inflammation, such as subchondral oedema or minimal synovial thickening, can be found in T2-weighted sequences [30].
The AUC obtained in this study was 0.78 (95% CI 0.68–0.87), indicating a good diagnostic accuracy when using US to identify capsular width as an indirect marker of fluid collection in the TMJ.
The best cut-off value of the capsular distension was 2.05 mm, with a sensitivity of 55.9% and a specificity of 94.7%. The second optimal threshold value was 1.75 mm, with a sensitivity of 82.4% and a specificity of 67.6%. Sensitivity increases at values below the optimum threshold of 2.05 mm, while specificity increases at values above 2.05 mm.
The optimal threshold value obtained in our study was consistent with that obtained by Manfredini et al. [31]. Bas et al. [32] identified a threshold value of capsular distension of 1.65 mm. The differences between these values are due to the operator-dependent character of US.
The diagnostic accuracy of US in the detection of TMJ effusion has been evaluated by several authors. Manfredini et al. [31] obtained a Se of 80%, while Jank et al. [14] obtained a Se of 81%. Tognini et al. [33] reported a good diagnostic accuracy of US, studying the ROC curve. Melchiore et al. [7] reported a Se of 70.6%. All these studies used MRI investigation as a gold standard.
Elias et al. [34] attempted to establish normal values of lateral capsular thickness. This value was between 1.4–1.6 mm. Capsular thickness may increase in the case of lateral displacements of the articular disc (Fig. 5). Strictly lateral discal displacements are less common [3]. In our study, we had four joints with lateral disc displacement. In these situations, it is difficult to make a differential ultrasonographic diagnosis between effusion and lateral disc displacement. However, joint pain may be a sign of differential diagnosis in such cases. Synovitis and fluid accumulation inside the TMJ are often accompanied by pain. The correlation between joint pain, capsular distension and the presence of joint fluid confirmed by MRI was made by the study of Bas et al. [32]. A positive correlation was found between pain scores and the degree of fluid accumulation confirmed by MRI. The optimal threshold value of the ultrasound-measured capsular distension obtained by this group of researchers was 1.65 mm. The link between joint pain and the presence of fluid collection in TMJ, diagnosed with MRI imaging has been investigated by several authors, most obtaining a close link between pain and the presence of intra-articular fluid [5, 35–37]. Other authors [38, 39] concluded that there is no link between joint pain and intra-articular fluid accumulation.
The strength of this study lies in the large number of joints examined (204) and the short time in which the clinical examination, US and MRI were performed. In other studies [31–33], the interval between MRI and US was 2–3 weeks. A large chronological difference between the two investigations may influence the results of the study. Inflammatory changes may remit or, on the contrary, may intensify.
Scarce research on this topic suggests the need for conducting new studies to assess intra- and inter-observational variability. It was found that for minor differences in threshold values for capsular distension, sensitivity and specificity can change substantially. These minor differences of 0.2, 0.3 mm may be due to inter- and intra-observational variability. The evaluation of this variability is difficult to achieve only by retrospective analysis of static images. For this reason, storage and analysis of videos, that can provide more information for the examiner, is recommended.