Timing of Arthrocentesis in the Management of Temporomandibular Disorders: A Systematic Review and Meta-analysis

The aim of this study was to assess the best timing to perform arthrocentesis in the management of temporomandibular disorders with regards to conservative treatment. A systematic search based on PRISMA guidelines, including a computer search with specific keywords, reference list search, and manual search was done. Relevant articles were selected after 3 search rounds for final review based on 6 predefined inclusion criteria, followed by a round of critical appraisal. Eleven publications, including 5 randomized controlled trials and 6 prospective clinical studies informed this review. The studies were divided into 3 groups based on the timing of arthrocentesis: 1). Arthrocentesis as the initial treatment, 2). Early arthrocentesis, and 3). Late arthrocentesis. Meta-analyses compared the efficacy of improvement in mouth opening and pain reduction in the 3 groups. The results were statistically significant and favoured early arthrocentesis, followed by late arthrocentesis in terms of both improvements in mouth opening and pain reduction. All 3 groups showed improvement in mouth opening and pain reduction. We conclude that early arthrocentesis may be able to produce the best clinical results, while arthrocentesis before attempting conservative treatment may produce less favourable outcomes.


INTRODUCTION
Temporomandibular disorders (TMDs) are a group of common facial pain conditions affecting 3.7-12% of the general population 1, 2 , and is 5 times more prevalent in women 2 . TMDs can be from the joint itself, the muscles of mastication, or a combination of both 3 . Although up to two-thirds of these patients will seek treatment for TMDs, about 15% of them will develop into chronic disease 1 . According to the National Institute of Dental and Craniofacial research, the estimated annual cost incurred in the United States related to TMDs is about $4 billion 1 . Because of their high prevalence, tendency to develop into chronicity, and significant financial burden to the society, TMDs are no doubt a public health problem.
The conventional approach to the management of most types of TMDs begins with conservative treatment, such as non-steroidal anti-inflammatory drugs (NSAIDs), physiotherapy, soft diet, and occlusal splint etc. 4 . Positive results are seen in between 75-90% of the patients with TMDs with conservative treatment 5 . However, there remains a small proportion of patients that are refractory to conservative treatment. In general, as the duration of pain symptoms increase, patient response to intervention diminishes 6,7 , and they become difficult to treat. This phenomenon applies to TMDs as it has been concluded that patients with chronic temporomandibular joint (TMJ) pain respond less favourably than in patients where the pain is non-chronic in nature [8][9][10] .
Temporomandibular joint arthrocentesis was introduced as a minimally invasive treatment for severe, limited mouth opening in the early 1990s 11 . It involves lavage of the superior joint space using 2 needles with an irrigation fluid such as normal saline, 4 with or without additional medications injected into the joint. While arthrocentesis has been proposed to be a first-line treatment for closed-lock of the TMJ 12 , different authors have also noted the effectiveness in the treatment of non-locking joints, and mechanisms have been proposed, such as removal of pathological joint fluid, and elimination of the negative pressure termed "the suction cup" effect 8,13,14 . Being cost-effective and safe, with a success rate of over 80% 15 , arthrocentesis is now a standard procedure not only for closed lock of the TMJ, but also for TMJ arthralgia of a non-locking nature when the outcomes of conservative treatment are unsatisfactory. However, how long should conservative treatment be attempted before moving on to the next step in the treatment algorithm when results are unsatisfactory remains a question to be answered.
Unfortunately, some patients are shunted between conservative treatment options for an extended period of time without observable progress, while early minimally invasive treatment, such as arthrocentesis, may have the potential to provide timely symptomatic relief as well as functional improvement. This early clinical benefit may prevent the development of chronic pain and psychological deterioration in some of the patients with TMDs. As said earlier, the difficulty of treating pain symptoms increases with the duration of those pain symptoms experienced by the patients. Therefore, the importance of understanding when to performing TMJ arthrocentesis cannot be overstated. However, there is no consensus in the literature regarding the timing of arthrocentesis with regards to conservative treatment at present.
The aim of this systematic review and meta-analysis was to determine whether timing of arthrocentesis, with regards to conservative management, has any effects on the treatment outcomes in patients with temporomandibular disorders.

Eligibility criteria
The following PICOTS criteria were applied: Relevant articles from the first and second rounds were included for the third-round evaluation.
In the third-round evaluation, full-texts of the included studies were evaluated based on the following inclusion criteria:

Statistical analysis
A study that included all these criteria was classified as having a low risk of bias.
A study that did not include one of these criteria was classified as having a moderate risk of bias When two or more criteria were missing, the study was considered to have a high risk of bias.
When there was any discrepancy during the appraisal process between the 2 reviewers, consensus was reached with discussion.

Risk of bias across studies
Publication bias using funnel plot were performed, as appropriate given the known limitations of these methods if the number of studies is ten or larger recommended by Cochrane handbook 19 .

Ethical approval
Ethical approval was not necessary as this study was a systematic review.

Study selection
The study selection process is shown in Figure 1. From the electronic database search, a total of 1999 articles were retrieved. After removal of duplicates, 896 articles remained.
Abstract screening of the resulted articles was done, and 714 articles were excluded. A total of 182 relevant studies were included in the second-round search.
In the second-round search, manual search from 2010-2020 and reference list search from the included studies did not result in any additional articles. 182 articles were included in the third-round evaluation, where 171 articles were excluded due to failing of one or more of the inclusion criteria mentioned above. Thus, 11 studies were included in the meta-analysis.

Study characteristics
The characteristics of the 11 included studies are shown in Table 1 showed a high risk of bias. (Table 2)

Synthesis of results
We used the WMD and SD of the continuous variables for meta-analyses. In one of the studies where the range and median were given 20 , the mean and standard deviation were estimated with statistical formulas 18 . In another study, the standard deviation for the preoperative and post-operative pain measurements (VAS) were not available 27 , therefore, pain measurements from that study were excluded from the meta-analysis. In another study 22 , the VAS scores were extracted from graphs by measurement, and that the In 3 of the studies, arthrocentesis was performed as the initial treatment for TMDs 20-23 , in 2 of the studies, arthrocentesis was done within 3 months of failing conservative treatment 24,25 , and in 5 of the studies, arthrocentesis was performed after at least 3 months of conservative treatment [26][27][28][29][30] . Both random effects model and fixed effects models were used for construction of forest plots, due to the intention of generalization inference and the high heterogeneity found in the included studies, but also because of the small number of studies included in the meta-analysis.

Pain (VAS)
All the included studies showed a reduction in pain in VAS after arthrocentesis (ranged 1.23-6.29), with arthrocentesis done within 3 months of conservative treatment producing the greatest reduction in pain, followed by arthrocentesis done after 3 months of conservative treatment. When using fixed effects model, arthrocentesis done within 3 months of conservative treatment was found to produce the greatest improvement of pain score (WMD 5.86mm, 95% CI: 5.49 to 6.23, p<0.001), followed by arthrocentesis after 3 months (WMD 5.65mm, 95% CI: 5.48 to 5.83, p<0.001) ( Figure 6). Similar results were obtained using random effects model, with arthrocentesis done within 3 months of conservative treatment found to produce the greatest improvement of MMO (WMD 5.39mm, 95% CI: 3.80 to 6.97, p<0.001), followed by arthrocentesis after 3 months (WMD 4.72mm, 95% CI: 3.44 to 5.99, p<0.001) (Figure 7).

Assessment of publication bias
Publication bias was not assessed as there were inadequate numbers of included trials within each group to properly assess publication bias. Traditional approach to management of arthrogenous TMDs usually begin with conservative treatment options, such as splint, medications, physiotherapy, and soft diet 31,32 . Surgical treatment is not offered until these conservative options have been exhausted without satisfactory improvement of pain symptoms and jaw function. This is due to the fact that many TMD symptoms do improve without any surgical treatment 4,33,34 . However, about 5-10% of patients with TMDs remain unresponsive to conservative options and require surgical treatment [35][36][37][38] . Of all the surgical options available for TMDs, arthrocentesis is the most minimally invasive which does not cause any irreversible damage to tissues. Due to the fact that arthrocentesis has been shown to produce a more rapid clinical response than conservative treatment 22 , it has led to recent speculation that perhaps early arthrocentesis should be considered when initial conservative treatment do not show any obvious benefits.
A recent network meta-analysis of randomized clinical trials by Al-Moraissi et al. osteoarthritis 46,47 , and rheumatoid arthritis 48 . Although it seems that arthroscopy may be more effective in improving joint movement and reducing pain 39,49 , arthrocentesis has the advantage of being simpler, more minimally invasive, and can be done at an outpatient setting.
There were a number of limitations to this study. As mentioned earlier, there was only a limited number of studies included in this systematic review. Specifically, after exclusion of studies with high risk of bias in the sensitivity analysis, only 1 study was included in the group where arthrocentesis was performed within 3 months of failing conservative treatment. Therefore, generalization of the results of this study cannot be made with high certainty. In addition, different diagnoses of TMDs were used across studies, and it is possible that the outcomes are affected due to the inherent differences in the nature of the various diagnoses. Another shortcoming of this study was that confounding variables were present, as different intra-articular injections were used along with arthrocentesis in some of the studies, and that different modalities of conservative treatment were used.
Additionally, some studies used different techniques of arthrocentesis, such as single and double-puncture arthrocentesis. Therefore, more RCTs are needed to decipher the optimal time to perform arthrocentesis in the treatment of various TMDs.
In conclusion, this study suggested that when conservative treatment fails to produce satisfactory results, early arthrocentesis may result in the greatest improvement of mouth opening and pain relieve. However, when arthrocentesis is used as an initial treatment without first attempting conservative treatment, the outcome may be less optimal than attempting conservative treatment first. Nevertheless, arthrocentesis performed at all time points produced improvement in mouth opening and pain reduction. It is important to note that the current meta-analysis only provides preliminary evidence of low to moderate quality level, as there is a paucity of well-designed studies in the literature regarding this topic. Therefore, more well-designed RCTs are required before we can draw more definitive conclusions.