Lavage Volume of Arthrocentesis in the Management of Temporomandibular Disorders: A Systematic Review and Meta-Analysis

The aim of this study was to investigate the most effective lavage volume of arthrocentesis in the management of temporomandibular disorders. A comprehensive electronic search, based on the PRISMA guidelines, was performed, which included a computer search with specific keywords, a reference list search and a manual search. The inclusion criteria were the following: a randomized controlled trial, at least 20 subjects who underwent arthrocentesis, mention of the irrigation materials used for the arthrocentesis, mention of the irrigation volumes used for the arthrocentesis, MMO and pain measured as VAS or NRS, were reported as outcome figures, mention of a specific diagnosis or signs and symptoms, and inclusion of the data on the MMO or VAS/NRS at 6-month follow-up. Sixteen publications were enrolled in the meta-analysis, comparing arthrocentesis with a lavage volume <150 mL and arthrocentesis with a lavage volume ≥150 mL, in the efficacy of the improvement in the mouth opening and pain reduction. The results revealed the group with a lavage volume <150 mL had a greater improvement in the mouth opening and pain reduction. However, results are to be interpreted with caution, due to the paucity of the randomized controlled literature and other confounding factors. Further high-quality studies are required to provide a better conclusion to the treatment outcomes of the different lavage volumes.


Introduction
Temporomandibular disorders (TMDs) are a series of clinical problems which affect the temporomandibular joint (TMJ), masticatory muscles and associated structures [1]. Based on the diagnostic criteria for a TMD (DC/TMD), established in 2014, a TMD was diagnosed, based on the physical examination (Axis I) and assessment of the psychosocial status and pain-related disability (Axis II) [1]. In more detail, the DC/TMD Axis I includes: (a) muscle disorders, including myalgia, myofascial pain and myofascial pain with referral, etc., (b) intra-articular joint disorders, including disc displacement with or without the reduction or mouth opening limitation; (c) other articular conditions, including arthralgia.
TMDs are the second most common musculoskeletal problem following chronic lower back pain. Around 6-12% of the general population is thought to be affected by TMDs [2,3]. TMDs usually affect people between 20 to 40 years of age, and are more prevalent in females [4,5]. Some of the common signs and symptoms are facial pain, limited mouth opening and joint sounds. In the United States, the estimated management cost was about USD 4 billion per year [6]. Moreover, patients with TMDs used a broader range of services and hence, consumed more resources [7].

Data Synthesis and Analysis
The extracted WMDs of the continuous data (MMO and VAS) were used in the meta-analysis. Forest plots were constructed using the random effects model with a 95% confidence interval (CI). Heterogeneity between the studies was evaluated using the Chi 2 with p < 0.10 or I 2 statistic of >50% [28]. In order to access whether the lavage volume had any effect on the clinical outcomes, we divided the included studies into two groups: (1) arthrocentesis with a lavage volume <150 mL; and (2) arthrocentesis with a lavage volume ≥150 mL. The MMO and VAS of the two groups were compared. The STATA (StataCorp. 2019. Stata Statistical Software: Release 16.0. College Station, TX, USA) was used for all statistical analyses.

Risk of Bias across Studies
Funnel plots were used to measure the publication bias, defined as the tendency to publish the results that are statistically or clinically significant. This method is deemed suitable, with more than 10 studies in the meta-analysis [28].

Study Selection
PRISMA 2020 flow diagram of the study is presented in Figure 1. An electronic database search resulted in a total of 2648 articles. Then, 645 articles remained after the removal of the duplicates. Following an initial screening of the titles and abstracts, 420 articles were excluded, due to the irrelevant topics. A total of 225 articles were included in the second round search. In the second round search, a manual search from 2010 to 2020 and a reference list search from the included studies resulted in five additional articles. Two hundred and thirty articles were included in the third-round evaluation. Of those, 214 articles were excluded due to the failure of one or more of the inclusion criteria, mentioned above. Thus, a total of 16 studies were included in the meta-analysis. and a reference list search from the included studies resulted in five additional articles. Two hundred and thirty articles were included in the third-round evaluation. Of those, 214 articles were excluded due to the failure of one or more of the inclusion criteria, mentioned above. Thus, a total of 16 studies were included in the meta-analysis.

Study Characteristics
Details of the included studies are shown in Table 1. All studies are randomized control trials in the critical appraisal and meta-analysis. A total of 677 patients received arthrocentesis treatment for TMDs in the 16 included studies. In the included studies, the diagnoses included disc displacement with or without reduction (DDWR/DDWOR), Wilkes stages 3 and 4, internal derangement, osteoarthritis and arthralgia. In five of the included studies, less than 150 mL of lavage volume was used while eleven studies used more than or equal to 150 mL. Regarding the operative technique, 12 studies performed the traditional single-needle puncture while four other studies employed the double-needle technique. The intra-articular irrigants were saline and lactated Ringer's solution. Some studies adopted the adjunctive injection of hyaluronic acid (HA), dexamethasone, platelet-rich plasma (PRP) and bone marrow nucleated cells. The follow-up period in the included studies, ranged from immediately after the procedure to 24 months. For this review, the outcome data for 6 months, 24 or 26 weeks, or 180 days of follow-up were extracted.

Study Characteristics
Details of the included studies are shown in Table 1. All studies are randomized control trials in the critical appraisal and meta-analysis. A total of 677 patients received arthrocentesis treatment for TMDs in the 16 included studies. In the included studies, the diagnoses included disc displacement with or without reduction (DDWR/DDWOR), Wilkes stages 3 and 4, internal derangement, osteoarthritis and arthralgia. In five of the included studies, less than 150 mL of lavage volume was used while eleven studies used more than or equal to 150 mL. Regarding the operative technique, 12 studies performed the traditional single-needle puncture while four other studies employed the double-needle technique. The intra-articular irrigants were saline and lactated Ringer's solution. Some studies adopted the adjunctive injection of hyaluronic acid (HA), dexamethasone, plateletrich plasma (PRP) and bone marrow nucleated cells. The follow-up period in the included studies, ranged from immediately after the procedure to 24 months. For this review, the outcome data for 6 months, 24 or 26 weeks, or 180 days of follow-up were extracted.  Vos et al., (2014) [19] 2014 RCT Arthrocentesis only 40 38.3 300 mL B NaCl Guarda-Nardini et al.,

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies showed a low risk of bias, three studies showed some concerns and one study showed a high risk of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Twelve studies a low risk of bias, three studies showed some concerns and one study showed a h of bias (Table 2).

Risk of Bias within the Studies
The assessment of the quality of the studies was carried out. Tw a low risk of bias, three studies showed some concerns and one stud of bias (Table 2).

Synthesis of the Results
The WMD and SD of the continuous variables were used for the meta-analyses. In one of the studies where the range and median were given [32], the mean and SD were estimated with statistical formulae [27]. In one study [34], the SD was calculated from the confidence intervals [28]. In two studies, the SD for the preoperative and postoperative pain measurements (VAS) were not available [29,30], therefore, the pain measurements from those studies were excluded from the meta-analysis. In three studies, the SD values of the postoperative MMO and VAS were not available. The SD of the postoperative MMO and VAS were assumed to be the same as the pre-operative values.
Random effects model was used for the construction of the forest plots, due to the intention of the generalization inference and the substantial heterogeneity found in the included studies (I 2 ranged from 70.02% to 98.5%; all p < 0.1).

MMO
The forest plots of the pooled WMD, between the <150 mL and ≥150 mL groups, in the improvement of the MMO, are summarized in Figure 2. All included studies demonstrated an improvement in the MMO after arthrocentesis (range 1.85-16.14 mm), with a greater improvement in the MMO in the group with a lavage volume <150 mL. There is no statistically significant difference in the MMO after the 6 months or equivalent (WMD: 9.62, 95% CI: 6.17 to 13.07, I 2 = 99.1%, p = 0.392) follow-up between the two groups.
A sensitivity analysis is not performed as there is only one study with high risk of bias.

Pain (VAS)
The forest plots of the pooled WMD, between the <150 mL and ≥150 mL groups, in the reduction in VAS, are summarized in Figure 3. All included studies demonstrated a reduction in the VAS after arthrocentesis (range −1.23-−8.20), with a greater improvement in the VAS in the group with a lavage volume <150 mL. There is no statistically significant difference in the VAS after the 6 months or equivalent (WMD: −4.91, 95% CI: −3.89 to −5.93, I 2 = 97.9%, p = 0.696) follow-up between the two groups.  A sensitivity analysis is not performed as there is only one study with a high risk of bias.

Assessment of the Publication Bias
The publication bias was assessed using the funnel plot techniques and Begg's rank test. The funnel plots of the MMO and VAS mean differences were both not in a severe asymmetry, which are shown in Figures 4 and 5. These implied the small study effect and thus a publication bias was not significant. Begg's rank test also suggests no significant publication bias in the MMO (p = 0.163 in overall and 0.640 and 0.086 in the subgroups) and the VAS differences (p = 0.155 in overall and 0.210 and 0.734 in the subgroups).

Discussion
In the management of arthrogenous TMDs, arthrocentesis has become a standard treatment option, due to its high efficacy and safety [14,43]. However, different techniques of arthrocentesis exist in the literature and in clinical practice, such as additional injection materials into the superior joint space [41,42,[44][45][46][47][48], the use of ultra-sound guidance [49][50][51][52] single versus double puncture techniques [30,32,53,54] and the timing of the procedure [19,55]. Specifically, the ideal irrigation volume for arthrocentesis of the TMJ remains a controversy. For example, in the studies included in this review, the irrigation volume ranged from 60 mL to 301 mL. Studies regarding the ideal irrigation volume for the TMJ arthrocentesis are few and far between, let alone the prospective clinical trials. If arthrocentesis performed with a smaller irrigation volume results in similar clinical outcomes, then arthrocentesis with a larger irrigation volume would be unnecessary, and the proce-

Discussion
In the management of arthrogenous TMDs, arthrocentesis has become a standard treatment option, due to its high efficacy and safety [14,43]. However, different techniques of arthrocentesis exist in the literature and in clinical practice, such as additional injection materials into the superior joint space [41,42,[44][45][46][47][48], the use of ultra-sound guidance [49][50][51][52] single versus double puncture techniques [30,32,53,54] and the timing of the procedure [19,55]. Specifically, the ideal irrigation volume for arthrocentesis of the TMJ remains a controversy. For example, in the studies included in this review, the irrigation volume ranged from 60 mL to 301 mL. Studies regarding the ideal irrigation volume for the TMJ arthrocentesis are few and far between, let alone the prospective clinical trials. If arthrocentesis performed with a smaller irrigation volume results in similar clinical outcomes, then arthrocentesis with a larger irrigation volume would be unnecessary, and the procedure could be completed in a timely fashion with increased patient comfort. Unfortunately, such a recommendation is difficult to make, due to the obvious knowledge gap at present. To the authors' knowledge, the present study is the first systematic review and meta-analysis to investigate the ideal irrigation volume for the TMJ arthrocentesis.
The results of our study suggested that arthrocentesis with a smaller irrigation volume (<150 mL) may be superior to that performed with a larger irrigation volume (≥150 mL), in terms of pain reduction and jaw function. Our results showed that there was a greater improvement in the MMO and VAS in the group with a small irrigation volume (<150 mL). Nevertheless, the evidence may be inconclusive as there was no statistically significant difference between the smaller (<150 mL) and larger (≥150 mL) irrigation volumes in the MMO (WMD: 9.62, 95% CI: 6.17 to 218 13.07, I 2 = 99.1%, p = 0.392) and VAS (WMD: −4.91, 95% CI: −3.89 to −5.93, I 2 = 236 97.9%, p = 0.696). The pooled analyses represented by the forest plots are, however, from a limited number of studies. Although whether such difference seen in this study represents any actual clinical significance, is unknown, it may be safe to propose that arthrocentesis with a smaller volume (<150 mL), is at least as effective as that performed with a larger volume. However, since only the data from the 6-month follow-up was extracted from the included studies for the meta-analysis, it is not possible to say whether this is also true in the longer term.
The view that a smaller volume used for arthrocentesis of the TMJ may be just as effective as a larger volume from the current study, is shared with the few studies in the current literature on the topic. In a clinical study by Grossman et al., arthrocentesis of the TMJ was used to treat patients presented with disc displacement without reduction [22]. The patients were divided into two groups, with 50 mL or 200 mL of irrigation volume used in the procedure. While it was found that the favorable clinical outcomes were seen in both groups, in terms of pain reduction and improvement of the jaw function, no significant difference was found between the two groups. In another randomized controlled trial in 2017, no statistical significant difference was found between the groups with an irrigation volume of either 100 mL or 250 mL, while the clinical improvement of the TMD symptoms were seen in both groups [24]. Moreover, in a recent cadaveric study, it was found that 25 mL of the irrigation solution was sufficient to remove methylene blue from the TMJ space of fresh human cadavers [56]. In contrast, in a clinical trial by Kaneyama et al., it was suggested that the ideal lavage volume for the removal of inflammatory mediators from the TMJ joint space, was between 300-400 mL, although the statistical analysis of the clinical variables of the TMDs, such as pain reduction and mouth opening, were not reported in that study [57]. Therefore, more clinical studies are required in order to further understand the ideal irrigation volume for TMJ arthrocentesis.
There were a number of limitations to the current study. Due to the paucity of randomized controlled studies in the literature, it was not possible to compare the results of those studies performed with control groups that investigated the effect of different irrigation volumes. Rather, the results of the randomized controlled trials, which have reported the irrigation volume and outcomes of arthrocentesis and not focusing on the irrigation volumes, were pooled for our meta-analysis. This methodology is not ideal and therefore the results from this study cannot be interpreted with high certainty. Moreover, confounding factors were present across the studies, such as different diagnoses of TMDs, different materials used for the lavage and intra-articular injection, different techniques of arthrocentesis across centers and the difference in the mean age and sex distribution in the included studies.
The association of ethnic background and prevalence of TMD were discussed in a number of studies [58][59][60]. It has been a less discussed factor and deserved more discussion in the context of the arthrocentesis outcome. Nevertheless, due to a limited number of studies specifying the patients' ethnicity in the arthrocentesis treatment, a comprehensive systematic review becomes difficult. Other factors, such as age, duration of the symptoms and oral habits are believed to affect the prognosis of arthrocentesis [61]. Therefore, ethnicity can be investigated as an effect modifier in the arthrocentesis when a different lavage volume is applied. A more in depth understanding of the relationship between ethnicity, age, gender and different irrigation volumes of arthrocentesis is feasible when a diverse population is included in further studies.
Moreover, only 6-months of data were selected for the meta-analysis, the effects of the different irrigation volume at different time points are unknown. Further review on the impact of the irrigation volume can be analyzed according to the duration of the follow-up. In fact, one study had identified the impact of the follow-up time of the different treatment for TMDs, by performing sub-group analyses, according to the duration of follow-up [14]. Therefore, more randomized controlled trials of arthrocentesis involving different follow-up periods are required before a final conclusion can be drawn.

Conclusions
In conclusion, the current systematic review and meta-analysis suggests that arthrocentesis of the TMJ is at least as effective, if not more, when a smaller lavage volume is used (<150 mL). However, due to the limitations with the methodology and confounding factors, the evidence is weak at this time. Future randomized clinical trials are needed to better understand the clinical outcomes related to the different irrigation volumes used for the arthrocentesis of the TMJ.