3.2. Methodological Quality
The quality of the RCTs was measured according to the PEDro scale. This scale consists of 11 items that are answered with “yes” or “no” responses based on whether they are fulfilled or not. The score is established over 10 points, as the first item is not counted towards the final score. Documents with a score ≥6 were classified as high quality, and those with a score <6, as low quality.
Following the instructions of the PEDro scale, the articles included in this analysis scored between 4 [18
] and 8 [19
], with two considered low-quality studies [18
] and four considered high quality (Table 1
None of the studies fulfilled the blinded therapist item (Item 5), although all of them fulfilled randomized allocation (Item 1), similar baseline measurements (Item 3), between-group comparisons (Item 9), and point measures and measures of variability (Item 10). Only one of the studies [19
] used blinded subjects (Item 4).
In total, the analysis included 304 people who suffered from TMD, 233 of whom were women (76.64%). The average subject age was 41.5 years. The studied pathologies included mouth opening pain [18
], mouth opening limitation [19
], myofascial symptoms [14
], non-reducing disc displacement [19
] and chronic migraine [14
The treatments employed in the studies were diverse. The most common treatment was TE, used in every study except one [19
]. Other techniques that were employed in more than one study were caudal mobilization of the TMJ [19
], health education and good habits communicated during the session by the clinical therapist, and manual treatment of the temporal and masseter muscles, which were all used in three out of the six articles. The following additional treatments were used in only one of the studies: cervical region treatment and TMJ neurodynamics [14
], TMJ manipulation [19
], botulinum toxin injections and manual pressure on craniocervical coordination centers [18
], lateral and medial pterygoid muscle and sphenopalatine ganglion treatments [21
], and Michigan splints.
Regarding the frequency of sessions, two sessions per week was most common, performed in three out of the six studies. Among the remaining studies, one study included one session every two weeks [19
], and the other two [14
] did not clarify the frequency. The treatment duration showed a greater variation, ranging from 2–4 weeks [20
] up to 18 weeks [19
]. The total number of sessions also presented great variability, from 3 (±1) sessions [18
] up to 24 [21
], giving a median of 9.5 sessions (Table 2
The most frequent outcomes were pain, assessed using the VAS, and active MMO, measured in millimeters, both of which were included in all of the articles. Two studies measured the pain pressure threshold (PPT) in the temporal and masseter muscles [14
], and two recorded the passive MMO [20
]. The remaining outcomes appeared in only one of the studies.
One study [19
] conducted a 4-month follow-up, two studies [20
] had a 12-month follow-up, and the rest conducted a 3-month follow-up.
In every article, a significant improvement in pain (measured with the VAS) and MMO compared to baseline was observed after MT for TMJ. Comparing the effects of MT, TE and education, the differences in pain and MMO seemed to be non-significant in the medium term (3–6 months) [19
When comparing the effect of MT for TMJ to other therapies, some differences were observed. Compared to MT in the cervical region, a significant reduction in pain was found from 3 months onwards, as well as a significant increase in MMO until 3-month follow-up [14
]. Comparing treatment with manual pressure on cranio-cervical coordination centers to botulinum toxin injection, no significant differences were found during the follow-up, except for laterotrusion movements, which improved more after MT treatment [18
]. The use of Michigan splints showed similar results, whether it was complemented with MT and TE or not, except when measuring active MMO, in which case a greater improvement was found when MT and TE were also used (Table 3
shows a forest plot derived from the meta-analysis performed on the change experienced in the pain variable, immediately after the intervention as well as at three months from the baseline. It is observed how immediately after the intervention, the mean improvement in the pain variable was around 4/10 points, although the range of improvement was very wide depending on the study consulted (Figure 2
). However, when the results were analyzed at 3 months from the start, the average improvement remained at around 4/10, but the results were more homogeneous when comparing the different studies.
On the other hand, Figure 3
shows the changes experienced in the MMO both at the end of the intervention and at three months from the baseline. At the first time of measurement (after the intervention), it is observed how the average improvement of almost all the groups was close to 15 mm, with the exception of the Garrigós-Pedrón study [14
], where an improvement of 4.35 mm is observed, which makes the trend line be to the left of the vast majority of the studies analyzed. When the results are analyzed at three months, it is observed that five of the seven results analyzed showed a failure to maintain the improvement experienced after the intervention, while the studies by Craane B. [20
] and Kalamir A. [21
] not only show maintenance of the improvement but an almost doubling of it. The reason why these differences in results are observed between the different studies at 3 months may be the types of therapeutic exercise used.