Advertisement

Arthrocentesis versus non-surgical intervention as initial treatment for temporomandibular joint arthralgia: a randomized controlled trial with long-term follow-up

Open AccessPublished:September 15, 2022DOI:https://doi.org/10.1016/j.ijom.2022.08.018

      Abstract

      Arthrocentesis for arthralgia of the temporomandibular joint (TMJ) is often only indicated when conservative, non-surgical interventions have failed. However, performing arthrocentesis as initial therapy may facilitate earlier and better recuperation of the joint. The aim of this study was to assess the efficacy of this therapy with a long-term follow-up. Eighty-four patients were randomly allocated to receive either arthrocentesis as initial treatment (n = 41) or non-surgical intervention (n = 43). Pain (100-mm visual analogue scale, VAS) and mandibular function impairment questionnaire scores (MFIQ, 0–100) were recorded at 3, 12, and 26 weeks, and ≥ 5 years (median 6.2, interquartile range 5.6–7.4 years). Univariable analyses were performed and linear mixed-effect models were constructed. Patients in the arthrocentesis group experienced significantly lower TMJ arthralgia compared to those treated non-surgically (pain during movement: −10.23 mm (95% confidence interval −17.86; −2.60); pain at rest: − 8.39 mm (95% confidence interval −13.70; −3.08)), while mandibular function remained similar in the two groups (MFIQ −2.41 (95% confidence interval −8.61; 3.78)). Of the final sample, 10 patients (10/39, 26%) in the non-surgical intervention group and two patients (2/34, 6%) in the arthrocentesis group received additional treatment during follow-up. Thus, initial treatment with arthrocentesis reduced TMJ arthralgia more efficaciously than non-surgical intervention in the long term, while maintaining similar mandibular function.

      Keywords

      Arthralgia of the temporomandibular joint (TMJ) is a frequently occurring and debilitating problem, affecting an estimated 8% of the population.
      • Marklund S.
      • Wanman A.
      Incidence and prevalence of temporomandibular joint pain and dysfunction. A one-year prospective study of university students.
      Although a multitude of causes may give rise to TMJ arthralgia, degenerative joint disease (DJD; i.e., osteoarthritis) and internal derangement (ID) are amongst the leading causes.
      • Tanaka E.
      • Detamore M.S.
      • Mercuri L.G.
      Degenerative disorders of the temporomandibular joint: etiology, diagnosis, and treatment.
      About 75% of patients with symptomatic DJD have some type of ID, while approximately 50% of all patients with symptomatic ID experience DJD to some degree.
      • Bertram S.
      • Rudisch A.
      • Innerhofer K.
      • Pümpel E.
      • Grubwieser G.
      • Emshoff R.
      Diagnosing TMJ internal derangement and osteoarthritis with magnetic resonance imaging.
      Despite often being observed simultaneously, the two conditions are different entities and do not necessarily precede each other.
      • Stegenga B.
      Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
      In both DJD and ID, the manifestation of symptoms often indicates a disturbance in the homeostasis of the articular tissue, where degenerative processes exceed the joint’s adaptive capability for tissue synthesis and rehabilitation.
      • Tanaka E.
      • Detamore M.S.
      • Mercuri L.G.
      Degenerative disorders of the temporomandibular joint: etiology, diagnosis, and treatment.
      • Stegenga B.
      Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
      Here, the synovial fluid of the TMJ contains elevated levels of proinflammatory cytokines, matrix degradation enzymes, and breakdown products that cannot easily be cleared from the synovial cavity.
      • Stegenga B.
      Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
      • Kim Y.K.
      • Kim S.G.
      • Kim B.S.
      • Lee J.Y.
      • Yun P.Y.
      • Bae J.H.
      • Oh J.S.
      • Ahn J.M.
      • Kim J.S.
      • Lee S.Y.
      Analysis of the cytokine profiles of the synovial fluid in a normal temporomandibular joint: preliminary study.
      • Wang X.D.
      • Zhang J.N.
      • Gan Y.H.
      • Zhou Y.H.
      Current understanding of pathogenesis and treatment of TMJ osteoarthritis.
      A secondary inflammatory response may arise subsequently, which is often accountable for most symptoms such as arthralgia, restricted mouth opening, blockages, and joint noise.
      • Stegenga B.
      Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
      • Hamada Y.
      • Kondoh T.
      • Holmlund A.B.
      • Yamamoto M.
      • Horie A.
      • Saito T.
      • Ito K.
      • Seto K.
      • Sekiya H.
      Inflammatory cytokines correlated with clinical outcome of temporomandibular joint irrigation in patients with chronic closed lock.
      Arthrocentesis may be performed to remove the inflammatory mediators and degradation products that are thought to be responsible for the clinical symptoms; this involves lavage of the upper TMJ compartment.
      • Vos L.M.
      • Huddleston Slater J.J.R.
      • Stegenga B.
      Arthrocentesis as initial treatment for temporomandibular joint arthropathy: a randomized controlled trial.
      • Gulen H.
      • Ataoglu H.
      • Haliloglu S.
      • Isik K.
      Proinflammatory cytokines in temporomandibular joint synovial fluid before and after arthrocentesis.
      • Nitzan D.
      • Svidovsky J.
      • Zini A.
      • Zadik Y.
      Effect of arthrocentesis on symptomatic osteoarthritis of the temporomandibular joint and analysis of the effect of preoperative clinical and radiological features.
      However, current management regimens often only indicate arthrocentesis if conservative, non-surgical interventions, based primarily on load reduction and the prescription of non-steroidal anti-inflammatory drugs (NSAIDs), prove to be insufficient in reducing symptoms.
      • Tanaka E.
      • Detamore M.S.
      • Mercuri L.G.
      Degenerative disorders of the temporomandibular joint: etiology, diagnosis, and treatment.
      • Murphy M.K.
      • MacBarb R.F.
      • Wong M.E.
      • Athanasiou K.A.
      Temporomandibular disorders: a review of etiology, clinical management, and tissue engineering strategies.
      Previous studies performed at the authors’ institution (University Medical Center Groningen) showed that arthrocentesis as the initial treatment for TMJ arthralgia was more efficacious in reducing clinical symptoms
      • Vos L.M.
      • Huddleston Slater J.J.R.
      • Stegenga B.
      Arthrocentesis as initial treatment for temporomandibular joint arthropathy: a randomized controlled trial.
      and was more cost-effective
      • Vos L.
      • Stegenga B.
      • Stant A.
      • Quik E.
      • Huddleston Slater J.
      Cost effectiveness of arthrocentesis compared to conservative therapy for arthralgia of the temporomandibular joint: a randomized controlled trial.
      than conservative, non-surgical intervention after 6 months of follow-up. However, limited evidence exists regarding the long-term efficacy of the immediate performance of arthrocentesis, particularly when compared to other therapies.
      Therefore, the aim of this study was to determine the efficacy of arthrocentesis as the initial treatment modality in reducing TMJ arthralgia, compared to non-surgical intervention, during long-term follow-up of ≥ 5 years.

      Materials and methods

      The 2010 CONSORT guidelines were followed in the reporting of this study. The study was conducted in accordance with the principles of the Declaration of Helsinki (adapted version 2013, Fortaleza, Brazil) and the Dutch ‘Medical Research Involving Human Subjects Act’ (WMO).

      Clinical trial design

      This single-centre randomized controlled trial was performed in the Department of Oral and Maxillofacial Surgery of the University Medical Center Groningen (UMCG), the Netherlands. Study subjects were recruited between January 2009 and July 2012. Approval was given by the Institutional Review Board of UMCG (METc 2008/197) and signed informed consent was obtained from all of the study participants prior to the commencement of any study-related procedures.

      Study population

      The study patients were recruited from the outpatient clinic of the Department of Oral and Maxillofacial Surgery of UMCG. A sample size calculation was performed for the primary outcome measure for the initial follow-up period of the study, as described previously.
      • Vos L.M.
      • Huddleston Slater J.J.R.
      • Stegenga B.
      Arthrocentesis as initial treatment for temporomandibular joint arthropathy: a randomized controlled trial.
      The inclusion criteria were age ≥ 16 years, pain in the TMJ region aggravated by mandibular movement (i.e., protrusion, maximal mouth opening, and/or lateral excursions) and/or during function, the presence of TMJ arthralgia after 2 weeks of 600 mg ibuprofen three times daily (to exclude acute inflammatory pain), and the disappearance of TMJ arthralgia after a diagnostic intra-articular injection with a local anaesthetic in the TMJ (Ultracain D-S Forte; Sanofi, Amsterdam, the Netherlands; to exclude myogenous pain).
      • Tjakkes G.H.E.
      • TenVergert E.M.
      • De Bont L.G.M.
      • Stegenga B.
      The effect of intra-articular injection of Ultracain in the temporomandibular joint in patients with preauricular pain: a randomized prospective double-blind placebo-controlled crossover study.
      The exclusion criteria were systemic rheumatic disease, bony ankylosis of the TMJ, prior open TMJ surgery, pregnancy, concurrent use of anti-inflammatory drugs, muscle relaxants, anti-depressants, and/or steroids, other medical contraindications, unwillingness to receive either study treatment, and the inability to speak Dutch or English.

      Study procedures

      The study patients were randomly allocated (1:1 ratio) to either the treatment group (arthrocentesis) or the control group (non-surgical intervention, NS) for initial treatment, by an independent colleague, using randomization software (StatsDirect version 2.7.7; StatsDirect Ltd, Birkenhead, United Kingdom). Patient allocation to the initial treatment modality was concealed from the physicians and researchers using opaque, sequentially numbered, sealed envelopes. An independent nurse revealed the group allocation once the patient had agreed to participate in the study.
      The patients allocated to arthrocentesis underwent lavage of the upper TMJ space with at least 300 ml isotonic saline, while the patients allocated to the NS group had to follow a strict soft diet protocol for at least 6 weeks. Patients in the NS group who received additional physical therapy and/or splint therapy were carefully selected during a critical interim evaluation, 2 weeks after the start of the soft diet protocol. Physical therapy was provided if the maximum mouth opening did not improve, while splint therapy was provided if the pain did not subside.
      All of the arthrocentesis procedures were performed under local anaesthesia by a single oral and maxillofacial surgeon (B. Stegenga). Further details of the procedure have been reported previously.
      • Vos L.M.
      • Huddleston Slater J.J.R.
      • Stegenga B.
      Arthrocentesis as initial treatment for temporomandibular joint arthropathy: a randomized controlled trial.
      Prior to treatment, panoramic radiographs, transcranial radiographs (Schüller projection), and transpharyngeal radiographs (Parma projection) of the TMJ were obtained and examined to evaluate the degree of osseous degeneration. The evaluation of the radiographic images and DJD diagnosis were done according to the criteria described by Ahmad et al.
      • Ahmad M.
      • Hollender L.
      • Anderson Q.
      • Kartha K.
      • Ohrbach R.K.
      • Truelove E.L.
      • John M.T.
      • Schiffman E.L.
      Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD): development of image analysis criteria and examiner reliability for image analysis.
      (Supplementary Material Table S1). The ID diagnostic categories were determined according to the Diagnostic Criteria for Temporomandibular Disorders (DC/TMD),
      • Schiffman E.L.
      • Ohrbach R.R.
      • Truelove E.L.
      • Look J.
      • Anderson G.
      • Goulet J.
      • List T.
      • Svensson P.
      • Gonzalez Y.
      • Lobbezoo F.
      • Michelotti A.
      • Brooks S.L.
      • Ceusters W.
      • Drangsholt M.
      • Ettlin D.
      • Gaul C.
      • Goldberg L.J.
      • Haythornthwaite J.A.
      • Hollender L.
      • Jensen R.
      • John M.T.
      • de Laat A.
      • de Leeuw R.
      • Maixner W.
      • van der Meulen M.
      • Murray G.M.
      • Nixdorf D.R.
      • Palla S.
      • Petersson A.
      • Pionchon P.
      • Smith B.
      • Visscher C.M.
      • Zakrzewska J.
      • Dworking S.F.
      Diagnostic Criteria for Temporomandibular Disorders (DC/TMD) for clinical and research applications: recommendations of the International RDC/TMD Consortium Network* and Orofacial Pain Special Interest Group.
      based on patient history and clinical examination (Supplementary Material Table S2).

      Outcome measures

      The main outcome parameter of the study was the extent of TMJ pain, according to a visual analogue scale (VAS), during mandibular movement (and/or function) (VASm; 0–100 mm, with 0 indicating no pain). The secondary outcome measures were TMJ pain at rest (VASr; 0–100 mm, with 0 indicating no pain), mandibular function impairment as determined using the validated Mandibular Function Impairment Questionnaire (MFIQ; 0–100),
      • Stegenga B.
      • de Bont L.G.
      • de Leeuw R.
      • Boering G.
      Assessment of mandibular function impairment associated with temporomandibular joint osteoarthrosis and internal derangement.
      and the number of patients who needed additional treatment during follow-up. A higher MFIQ score indicates worse mandibular function. The outcome measures were recorded at baseline and at four follow-up time points: 3, 12, and 26 weeks (in the outpatient clinic) and ≥ 5 years (via telephone interview). The patients who received additional treatment during follow-up due to insufficient symptom reduction after the primary treatment were registered.

      Statistical analysis

      Normally distributed variables were presented as the mean and standard deviation (SD) values. Non-normally distributed continuous variables were presented as the median with the first and third quartiles (interquartile range, IQR) and compared using the Mann–Whitney U-test. The assumption of a normal distribution of continuous data was assessed by visually examining the Q–Q plot and the Shapiro–Wilk test. All categorical variables were described as frequencies with percentages and compared using Fisher’s exact test. The analyses were performed using IBM SPSS Statistics for Windows, version 23 (IBM Corp., Armonk, NY, USA). Graphs were drawn using Prism, version 9 (GraphPad Software Inc., San Diego, CA, USA). Linear mixed-effect models (LMM) were fitted to assess the effect of the interventions on repeated measurements of VASm, VASr, and MFIQ. Simple multivariable models included the fixed effects of the type of intervention and follow-up in weeks. The full multivariable models included the fixed effects of the baseline score, type of intervention, follow-up in weeks, sex, age, presence and type of ID, degree of DJD seen on radiographic imaging, and any additional treatments. These variables were selected because of their prognostic nature and based on model improvement.
      • Kahan B.
      • Morris T.
      Adjusting for multiple prognostic factors in the analysis of randomised trials.
      Additional treatments were included as time-varying variables by taking into account at which point during follow-up the additional treatments were performed. The included random effects were the participants. Additionally, the fixed interaction between the intervention and time and/or the random effect of time were only included if the term significantly improved the multivariable model. Between models, model improvement was tested using likelihood ratio tests. Participants were only included in the analysis if the baseline and at least one follow-up measurement of one or more outcome variables were recorded. All models yielded an estimated regression coefficient (β) with corresponding 95% confidence interval (95% CI). P ≤ 0.05 (two-tailed) was considered statistically significant. All of the LMM analyses were performed in R, version 4.0.5 (R Core team; R Foundation for Statistical Computing, Vienna, Austria), using the lme4 package.
      • Bates D.
      • Mächler M.
      • Bolker B.
      • Walker S.
      Fitting linear mixed-effects models using lme4.

      Results

      A total of 84 patients were initially enrolled in the study (Fig. 1). Of these, 41 were allocated to receive arthrocentesis and 43 to receive NS (Table 1). After randomization, 11 patients were excluded from the statistical analysis: seven (17%) from the arthrocentesis group and four (9%) from the NS group. The reasons for exclusion are presented in Fig. 1.
      Fig. 1
      Fig. 1CONSORT flow diagram (version 2010) of subject enrolment, allocation, and follow-up. In the arthrocentesis group, two patients did not receive treatment due to the absence of arthralgia on the day of the intervention. These two patients did not meet the inclusion criteria in the first place and were therefore mistakenly randomized. The outcome measurements of the patients in both groups who were excluded after treatment were not registered at any of the follow-up moments, rendering linear mixed model analyses of these subjects impossible.
      Table 1Baseline characteristics of the study population after randomization, and the final study sample.
      PredictorsAfter randomizationFinal study sample
      ArthrocentesisNon-surgical interventionArthrocentesisNon-surgical intervention
      Sample size, n41433439
      Female/male, n (%)30 (73)/11 (27)34 (79)/9 (21)25 (74)/9 (26)32 (82)/7 (18)
      Age in years, median (IQR)28 (23–55)31 (21–49)27 (23–50)29 (21–41)
      VASm (mm), mean (SD)51.60 (18.88)54.80 (24.46)53.16 (17.82)54.45 (23.96)
      VASr (mm), median (IQR)15.00 (00.0–34.00)17.00 (0.00–44.75)17.50 (0.00–34.75)17.00 (0.00–44.25)
      MFIQ, mean (SD)49.43 (14.51)47.79 (19.10)49.16 (14.79)48.98 (18.10)
      ID, n (%)
      No ID6 (16)6 (14)6 (18)6 (15)
      ADDwR9 (24)12 (29)8 (24)11 (28)
      ADDwR with intermittent locking11 (29)10 (24)10 (29)9 (23)
      ADDwoR with limited mouth opening12 (32)13 (31)10 (29)12 (31)
      ADDwoR without limited mouth opening0 (0)1 (2)0 (0)1 (3)
      ID, missing
      Missing indicates unregistered diagnoses of the drop-outs. The missing subjects are not included in the percentages.
      3100
      DJD, n (%)
      No DJD19 (48)24 (57)16 (47)22 (56)
      Indeterminate for DJD12 (30)10 (24)9 (26)9 (23)
      Evidence of DJD9 (23)8 (19)9 (26)8 (21)
      DJD, missing
      Missing indicates unregistered diagnoses of the drop-outs. The missing subjects are not included in the percentages.
      1100
      ADDwR, anterior disc displacement with reduction; ADDwoR, anterior disc displacement without reduction; DJD, degenerative joint disease; ID, internal derangement; IQR, interquartile range (Q1–Q3); MFIQ, mandibular function impairment questionnaire; SD, standard deviation; VASm, visual analogue scale during mandibular movement (and/or function); VASr, visual analogue scale at rest.
      a Missing indicates unregistered diagnoses of the drop-outs. The missing subjects are not included in the percentages.
      The final arthrocentesis group included 25 female patients (74%) and nine male patients (26%); median age was 27 years (IQR 23–50 years). The final NS group included 32 female patients (82%) and seven male patients (18%); median age was 29 years (21–41 years). The median time interval between the baseline and the latest follow-up (≥5 years) was 6.2 years (IQR 5.6–7.4 years).
      All of the NS patients adhered to the strict soft diet protocol. Ten patients (26%) in the NS group received additional physical therapy, 15 (38%) received additional splint therapy, and two (5%) received both additional physical and splint therapy. Also, 10 patients in the NS group (26%) underwent additional arthrocentesis during follow-up (at a median of 41 weeks (IQR 30–80 weeks)); the study outcomes of six of these patients (15%) were recorded after the additional procedure (at a median 41 weeks (IQR 25–104 weeks)).
      In the arthrocentesis group, two patients (6%) received additional treatment: one patient underwent another arthrocentesis (after 75 weeks) and subsequently diagnostic arthroscopy (after 136 weeks); one patient received a methylprednisolone injection against the joint capsule (after 8 weeks). No serious adverse events were recorded; four patients experienced mild temporary swelling around the TMJ directly after arthrocentesis.
      The median scores of the outcome variables VASm, VASr, and MFIQ over time are presented in Fig. 2; the scores at the latest follow-up are reported in Table 2. The simple multivariable LMM analysis indicated a significantly lower VASm in the arthrocentesis group compared to the NS group (−11.13 mm (95% CI −20.04; −2.21)), as shown in Table 3. Of the secondary outcome measures, VASr was significantly lower in the arthrocentesis group (−8.72 mm (95% CI −16.23; −1.21)), whereas the MFIQ score did not differ significantly between the treatment groups (−3.89 (95% CI −11.16; 3.37)) (Table 3).
      Fig. 2
      Fig. 2Progression of (a) pain during movement (VASm), (b) pain at rest (VASr), and (c) perceived mandibular function (MFIQ score) over time for non-surgical intervention and arthrocentesis. The graphs are based on the raw data (not the linear mixed models), presented as the median and interquartile range error bars.VASm, visual analogue scale during mandibular movement (and/or function); VASr, visual analogue scale at rest; MFIQ, mandibular function impairment questionnaire; T0, baseline; T1, 3-week follow-up; T2, 12-week follow-up; T3, 26-week follow-up; T4, ≥ 5-year follow-up.
      Table 2Results of the study outcomes recorded at the latest follow-up (T4, ≥5 years).
      Study outcomeTreatment groupP-value
      ArthrocentesisNon-surgical intervention
      VASm (mm), median (IQR)0.00 (0.00–15.00)10.00 (0.00–57.5)0.097
      VASr (mm), median (IQR)0.00 (0.00–10.00)0.00 (0.00–37.5)0.154
      MFIQ, median (IQR)4.41 (1.47–18.38)13.97 (0.00–33.46)0.360
      IQR, interquartile range (Q1–Q3); MFIQ, mandibular function impairment questionnaire; VASm, visual analogue scale during mandibular movement (and/or function); VASr, visual analogue scale at rest.
      Table 3Simple and full multivariable linear mixed model analyses results.
      PredictorsVASmVASrMFIQ
      β (95% CI)P-valueβ (95% CI)P-valueβ (95% CI)P-value
      Simple multivariable
      Treatment (Ref. = non-surgical)− 11.13

      (−20.04; −2.21)
      0.014− 8.72

      (−16.23; −1.21)
      0.023− 3.89 (−11.16; 3.37)0.293
      Follow-up in weeks− 0.07

      (−0.09; −0.04)
      <0.001− 0.02

      (−0.04; −0.00)
      0.023− 0.08 (−0.10; −0.06)<0.001
      Full multivariable
      Treatment (Ref. = non-surgical)− 10.23

      (−17.86; −2.60)
      0.009− 8.39

      (−13.70; −3.08)
      0.002− 2.41 (−8.61; 3.78)0.445
      Follow-up in weeks− 0.07

      (−0.09; −0.05)
      <0.001− 0.02

      (−0.03; 0.00)
      0.063− 0.09 (−0.11; −0.07)<0.001
      Baseline score (in millimetres for VASm and VASr)0.58

      (0.39; 0.77)
      <0.0010.56

      (0.45; 0.67)
      <0.0010.57 (0.37; 0.76)<0.001
      Sex (Ref. = female)− 0.68

      (−10.42; 9.06)
      0.891− 3.30

      (−9.93; 3.33)
      0.330− 2.05 (−9.83; 5.72)0.605
      Age in years0.02

      (−0.26; 0.30)
      0.884− 0.00

      (−0.19; 0.18)
      0.975− 0.05 (−0.26; 0.17)0.678
      Additional treatment in non-surgical group
      Additional treatment in the conservative, non-surgical intervention group with arthrocentesis during the follow-up (n = 6).
      6.77

      (−10.71; 24.26)
      0.448− 10.88

      (−22.99; 1.23)
      0.07811.02 (−2.81; 24.86)0.118
      Additional treatment in arthrocentesis group
      Additional treatment in the arthrocentesis group during the follow-up (n = 2).
      − 1.45

      (−28.95; 26.06)
      0.918− 2.15

      (−21.12; 16.82)
      0.824− 11.48 (−47.73; 24.76)0.535
      No ID (Ref.)00.13300.17000.261
      ADDwR− 14.28

      (−26.28; −2.27)
      − 9.76

      (−18.20; −1.33)
      − 6.91 (−16.39; 2.57)
      ADDwR with intermittent locking− 10.93

      (−23.29; 1.43)
      − 4.76

      (−13.43; 3.90)
      − 6.17 (−15.55; 3.20)
      ADDwoR with limited mouth opening− 6.90

      (−18.44; 4.65)
      − 4.53

      (−12.43; 3.37)
      − 1.74 (−10.72; 7.25)
      ADDwoR without limited mouth opening− 19.49

      (−61.23; 22.25)
      − 13.07

      (−41.60; 15.47)
      NA
      No DJD (Ref.)00.54700.76900.785
      Indeterminate for DJD− 4.62

      (−13.68; 4.44)
      0.77

      (−5.47; 7.02)
      − 2.34 (−9.42; 4.75)
      Evidence of DJD− 0.65

      (−10.30; 9.01)
      − 1.57

      (−8.13; 5.00)
      − 1.09 (−9.33; 7.15)
      ADDwR, anterior disc displacement with reduction; ADDwoR, anterior disc displacement without reduction; β, regression coefficient; CI, confidence interval; DJD, degenerative joint disease; ID, internal derangement; MFIQ, mandibular function impairment questionnaire; NA, not available; VASm, visual analogue scale during mandibular movement (and/or function); VASr, visual analogue scale at rest.
      a Additional treatment in the conservative, non-surgical intervention group with arthrocentesis during the follow-up (n = 6).
      b Additional treatment in the arthrocentesis group during the follow-up (n = 2).
      In the full multivariable model, there was no fixed interaction between intervention and time and/or the random effect of time did not significantly improve the model (VASm (P = 0.971); VASr (P = 0.888); MFIQ (P = 0.298)). In this model, VASm was significantly lower in the arthrocentesis group compared to the NS group over the entire follow-up (−10.23 mm (95% CI −17.86; −2.60)) (Table 3). Of the secondary outcome measures, VASr was significantly lower in the arthrocentesis group (−8.39 mm (95% CI −13.70; −3.08)). However, the MFIQ score did not differ significantly between the treatment groups (−2.41 (95% CI −8.61; 3.78)) after adjustment for confounders (Table 3).
      In the full multivariable model, the overall presence of ID or DJD had no significant influence on the outcome measures when compared to patients with, respectively, no clinical signs of ID or radiological signs of DJD (Table 3).

      Discussion

      The aim of this study was to evaluate the efficacy of arthrocentesis as the initial treatment for TMJ arthralgia compared to non-surgical intervention during long-term follow-up (≥5 years). In addition to the previous report,
      • Vos L.M.
      • Huddleston Slater J.J.R.
      • Stegenga B.
      Arthrocentesis as initial treatment for temporomandibular joint arthropathy: a randomized controlled trial.
      where promising results were reported for the 6-month follow-up, the findings of the present study demonstrate that arthrocentesis as the initial treatment is more efficacious in reducing TMJ arthralgia over a period of ≥ 5 years than conservative, non-surgical intervention.
      LMM analyses in this study allowed the accurate longitudinal evaluation of treatment efficacy. Results from these analyses are a better representation of treatment efficacy than results from comparing solely the outcome variables of the individual follow-up moments (Fig. 2; Table 2). This is because the LMM allowed the inclusion of the time of latest follow-up (which differed between subjects) and the baseline scores as predictors in the statistical model. Comparison of the outcome variables at the latest follow-up, however, showed a smaller spread in the arthrocentesis group, indicating a more predictable treatment outcome compared to NS (Table 2).
      The study findings support current evidence that arthrocentesis may be beneficial for patients with ID and/or DJD.
      • Guarda-Nardini L.
      • de Almeida A.
      • Manfredini D.
      Arthrocentesis of the temporomandibular joint: systematic review and clinical implications of research findings.
      • Monje-Gil F.
      • Nitzan D.
      • González-Garcia R.
      Temporomandibular joint arthrocentesis. Review of the literature.
      More importantly, the timing of arthrocentesis is paramount to optimize the treatment outcome. Performing interventions as soon as possible enhances the immediate joint recovery process. The response to arthrocentesis is more evident in acute cases of TMJ arthralgia compared to chronic ones.
      • Emshoff R.
      • Rudisch A.
      Determining predictor variables for treatment outcomes of arthrocentesis and hydraulic distention of the temporomandibular joint.
      However, traditional conservative care encompasses a prolonged treatment period, putting patients more at risk of developing chronic pain in the event that those treatment methods are insufficient in reducing symptoms. Therefore, a reconsideration of the current therapeutic strategies in favour of arthrocentesis could potentially be beneficial for patients.
      The important role of performing minimally invasive treatment procedures early is supported by a recent systematic review, in which the authors concluded that such procedures are more beneficial than conservative care in the short term (≤5 months) and intermediate term (6 months–4 years).
      • Al-Moraissi E.A.
      • Wolford L.M.
      • Ellis E.
      • Neff A.
      The hierarchy of different treatments for arthrogenous temporomandibular disorders: a network meta-analysis of randomized clinical trials.
      The current study provides further evidence that arthrocentesis is superior in improving patient-reported outcomes in the long term. On the other hand, another recent systematic review investigating the influence of the timing of arthrocentesis on patient outcomes, in relation to conservative care, concluded that in addition to the favourable outcomes of arthrocentesis in general, the optimal results were obtained when arthrocentesis was performed within 3 months after conservative care.
      • Li D.
      • Wong N.
      • Li S.
      • McGrath C.
      • Leung Y.
      Timing of arthrocentesis in the management of temporomandibular disorders: an integrative review and meta-analysis.
      The authors noted, however, that the evidence was of low to moderate quality due to the low availability of well-designed studies and the heterogeneity between studies in diagnoses, treatment modalities, and techniques used. Definitive conclusions regarding the optimal timing of arthrocentesis may be drawn once different timings are compared in a single clinical trial in the future.
      Interestingly, mandibular function, expressed as the MFIQ score, did not differ significantly between the two treatment groups in the specific study model. Seemingly, the patients who underwent arthrocentesis did not experience an improvement in performing daily activities related to their jaw compared to the patients who received NS, despite having less pain. It is hypothesized that factors other than pain may also play a role in the patients’ perception of mandibular function, such as mouth opening restrictions due to mechanical obstructions, or the psychosocial profile (i.e., pain-avoiding behaviour and fear).
      Additional treatment during the follow-up was not associated with better treatment outcomes in the full multivariable model. Remarkably, 10 (26%) patients who were allocated to the NS group required additional treatment, six of whose outcome measurements were recorded afterwards, whereas only two (6%) patients allocated to the arthrocentesis group received additional treatment. These findings are in line with the results of Tatli et al.,
      • Tatli U.
      • Belidayi M.
      • Ekren O.
      • Salimov F.
      Comparison of the effectiveness of three different treatment methods for temporomandibular joint disc displacement without reduction.
      who reported the superiority of arthrocentesis over splint therapy after 6 months for the treatment of anterior disc displacement (ADD) without reduction (success rate of 93% for arthrocentesis versus 60% for splint therapy).
      Neither age nor sex was associated with the degree of TMJ arthralgia or functional impairment in the current full multivariable model. Furthermore, the degree of DJD seen on radiographic imaging was not significantly associated with TMJ arthralgia, which is consistent with previous findings.
      • de Leeuw R.
      • Boering G.
      • van der Kuijl B.
      • Stegenga B.
      Hard and soft tissue imaging of the temporomandibular joint 30 years after diagnosis of osteoarthrosis and internal derangement.
      • de Leeuw R.
      • Boering G.
      • Stegenga B.
      • de Bont L.
      Radiographic signs of temporomandibular joint osteoarthrosis and internal derangement 30 years after nonsurgical treatment.
      This is because observed radiographic changes in bony structures may have also been the result of remodelling in the context of joint recuperation, indicating a return to a homeostatic state of the articular structures.
      • Stegenga B.
      Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
      The presence of ID was not statistically associated with TMJ arthralgia in the multivariable model, which is also consistent with the literature.
      • Ohlmann B.
      • Rammelsberg P.
      • Henschel V.
      • Kress B.
      • Gabbert O.
      • Schmitter M.
      Prediction of TMJ arthralgia according to clinical diagnosis and MRI findings.
      The main strength of this study is that it is one of only a few randomized controlled trials that have been performed on the efficacy of arthrocentesis compared to non-surgical intervention, specifically with a long-term follow-up. Other strengths of the study include a thorough study design, using the intra-articular injection procedure as a diagnostic tool, and the relatively large number of study patients. Furthermore, appropriate statistical models for longitudinal analyses were constructed.
      This study has some limitations. No magnetic resonance imaging (MRI) scans were performed, which may have affected the diagnostic testing of ID. According to the DC/TMD,
      • Schiffman E.L.
      • Ohrbach R.R.
      • Truelove E.L.
      • Look J.
      • Anderson G.
      • Goulet J.
      • List T.
      • Svensson P.
      • Gonzalez Y.
      • Lobbezoo F.
      • Michelotti A.
      • Brooks S.L.
      • Ceusters W.
      • Drangsholt M.
      • Ettlin D.
      • Gaul C.
      • Goldberg L.J.
      • Haythornthwaite J.A.
      • Hollender L.
      • Jensen R.
      • John M.T.
      • de Laat A.
      • de Leeuw R.
      • Maixner W.
      • van der Meulen M.
      • Murray G.M.
      • Nixdorf D.R.
      • Palla S.
      • Petersson A.
      • Pionchon P.
      • Smith B.
      • Visscher C.M.
      • Zakrzewska J.
      • Dworking S.F.
      Diagnostic Criteria for Temporomandibular Disorders (DC/TMD) for clinical and research applications: recommendations of the International RDC/TMD Consortium Network* and Orofacial Pain Special Interest Group.
      definitive diagnosis of ID, with the exception of ADD without reduction with limited mouth opening, requires an MRI to achieve adequate diagnostic sensitivity (i.e., ≥70%) and specificity (i.e., ≥95%). However, the participants were included in this study several years before the general consensus regarding optimal diagnostic testing using imaging was described in the current DC/TMD. More importantly, an MRI-confirmed displaced disc is not exclusive to a pathological condition; it is present in approximately 30% of asymptomatic patients
      • Stegenga B.
      Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
      and does not influence the treatment choice or outcome.
      • Honda K.
      • Yasukawa Y.
      • Fujiwara M.
      • Abe T.
      • Urade M.
      Causes of persistent joint pain after arthrocentesis of temporomandibular joint.
      • Emshoff R.
      • Rudisch A.
      Are internal derangement and osteoarthrosis linked to changes in clinical outcome measures of arthrocentesis of the temporomandibular joint.
      In the present study, the psychosocial profiles of the study patients over time were not recorded. This could have distorted the results, since psychosocial factors are known to be associated with temporomandibular pain.
      • Ohlmann B.
      • Rammelsberg P.
      • Henschel V.
      • Kress B.
      • Gabbert O.
      • Schmitter M.
      Prediction of TMJ arthralgia according to clinical diagnosis and MRI findings.
      • Rollman G.B.
      • Gillespie J.M.
      The role of psychosocial factors in temporomandibular disorders.
      However, due to the duration of follow-up, answering additional extensive questionnaires on the phone many years post-treatment could have resulted in an excessive burden for the study participants. Furthermore, the trial may be limited in its generalizability to other clinics and general populations with temporomandibular disorders. Patient inclusion through a diagnostic intra-articular injection allowed the accurate selection of patients suffering from pain that primarily originated from the joint.
      • Tjakkes G.H.E.
      • TenVergert E.M.
      • De Bont L.G.M.
      • Stegenga B.
      The effect of intra-articular injection of Ultracain in the temporomandibular joint in patients with preauricular pain: a randomized prospective double-blind placebo-controlled crossover study.
      However, since many patients often experience myogenous symptoms concurrently with arthralgia, some arthralgia patients with more prominent myogenous symptoms may have been excluded from the study. Moreover, in the non-surgical intervention group, any additional physical or splinting therapy was given based on the response to the initial soft diet advice, which resulted in treatment heterogeneity within the control group. The authors chose to design the study in such a manner, as this is believed to be an accurate representation of the clinical reality, where conservative therapies are mostly patient-tailored and given based on clinical presentation.
      In conclusion, arthrocentesis was found to be more efficacious as an initial treatment in reducing TMJ arthralgia than non-surgical intervention over follow-up of ≥ 5 years, although mandibular function was similar after both treatments. Performing arthrocentesis at an earlier stage to treat TMJ arthralgia allows immediate initiation of joint recuperation, which is sustained over the long term. Verification of the current study results is mandatory to substantiate this conclusion.

      Ethical approval

      Approval to conduct this study was given by the Institutional Review Board of UMCG (METc 2008/197).

      Funding

      This study was funded by ZorgOnderzoek Nederland en Medische Wetenschappen (Project number ZonMw 170 991 006).

      Acknowledgements

      The authors would like to acknowledge Prof. Dr Boudewijn Stegenga, who is no longer with us, for his role as principal investigator during the study with 6 months of follow-up.

      Competing interests

      None.

      Patient consent

      Signed informed consent was obtained from all study patients prior to the commencement of any study-related procedures.

      Trial registration

      This trial is registered in the Netherlands Trial Register (NL1444).

      Appendix A. Supplementary material

      References

        • Marklund S.
        • Wanman A.
        Incidence and prevalence of temporomandibular joint pain and dysfunction. A one-year prospective study of university students.
        Acta Odontol Scand. 2007; 65: 119-127https://doi.org/10.1080/00016350601083554
        • Tanaka E.
        • Detamore M.S.
        • Mercuri L.G.
        Degenerative disorders of the temporomandibular joint: etiology, diagnosis, and treatment.
        J Dent Res. 2008; 87: 296-307https://doi.org/10.1177/154405910808700406
        • Bertram S.
        • Rudisch A.
        • Innerhofer K.
        • Pümpel E.
        • Grubwieser G.
        • Emshoff R.
        Diagnosing TMJ internal derangement and osteoarthritis with magnetic resonance imaging.
        J Am Dent Assoc. 2001; 132: 753-761https://doi.org/10.14219/jada.archive.2001.0272
        • Stegenga B.
        Osteoarthritis of the temporomandibular joint organ and its relationship to disc displacement.
        J Orofac Pain. 2001; 15: 193-205
        • Kim Y.K.
        • Kim S.G.
        • Kim B.S.
        • Lee J.Y.
        • Yun P.Y.
        • Bae J.H.
        • Oh J.S.
        • Ahn J.M.
        • Kim J.S.
        • Lee S.Y.
        Analysis of the cytokine profiles of the synovial fluid in a normal temporomandibular joint: preliminary study.
        J Craniomaxillofac Surg. 2012; 40: 337-341https://doi.org/10.1016/j.jcms.2012.02.002
        • Wang X.D.
        • Zhang J.N.
        • Gan Y.H.
        • Zhou Y.H.
        Current understanding of pathogenesis and treatment of TMJ osteoarthritis.
        J Dent Res. 2015; 94: 666-673https://doi.org/10.1177/0022034515574770
        • Hamada Y.
        • Kondoh T.
        • Holmlund A.B.
        • Yamamoto M.
        • Horie A.
        • Saito T.
        • Ito K.
        • Seto K.
        • Sekiya H.
        Inflammatory cytokines correlated with clinical outcome of temporomandibular joint irrigation in patients with chronic closed lock.
        Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2006; 102: 596-601https://doi.org/10.1016/j.tripleo.2005.11.010
        • Vos L.M.
        • Huddleston Slater J.J.R.
        • Stegenga B.
        Arthrocentesis as initial treatment for temporomandibular joint arthropathy: a randomized controlled trial.
        J Craniomaxillofac Surg. 2014; 42: e134-e139https://doi.org/10.1016/j.jcms.2013.07.010
        • Gulen H.
        • Ataoglu H.
        • Haliloglu S.
        • Isik K.
        Proinflammatory cytokines in temporomandibular joint synovial fluid before and after arthrocentesis.
        Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009; 107: e1-e4https://doi.org/10.1016/j.tripleo.2009.02.006
        • Nitzan D.
        • Svidovsky J.
        • Zini A.
        • Zadik Y.
        Effect of arthrocentesis on symptomatic osteoarthritis of the temporomandibular joint and analysis of the effect of preoperative clinical and radiological features.
        J Oral Maxillofac Surg. 2017; 75: 260-267https://doi.org/10.1016/j.joms.2016.08.017
        • Murphy M.K.
        • MacBarb R.F.
        • Wong M.E.
        • Athanasiou K.A.
        Temporomandibular disorders: a review of etiology, clinical management, and tissue engineering strategies.
        Int J Oral Maxillofac Implants. 2013; 28: e393-e414https://doi.org/10.11607/jomi.te20
        • Vos L.
        • Stegenga B.
        • Stant A.
        • Quik E.
        • Huddleston Slater J.
        Cost effectiveness of arthrocentesis compared to conservative therapy for arthralgia of the temporomandibular joint: a randomized controlled trial.
        J Oral Facial Pain Headache. 2018; 32: 198-207https://doi.org/10.11607/ofph.1457
        • Tjakkes G.H.E.
        • TenVergert E.M.
        • De Bont L.G.M.
        • Stegenga B.
        The effect of intra-articular injection of Ultracain in the temporomandibular joint in patients with preauricular pain: a randomized prospective double-blind placebo-controlled crossover study.
        Clin J Pain. 2007; 23: 233-236https://doi.org/10.1097/AJP.0b013e31802f0950
        • Ahmad M.
        • Hollender L.
        • Anderson Q.
        • Kartha K.
        • Ohrbach R.K.
        • Truelove E.L.
        • John M.T.
        • Schiffman E.L.
        Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD): development of image analysis criteria and examiner reliability for image analysis.
        Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009; 107: 844-860https://doi.org/10.1016/j.tripleo.2009.02.023
        • Schiffman E.L.
        • Ohrbach R.R.
        • Truelove E.L.
        • Look J.
        • Anderson G.
        • Goulet J.
        • List T.
        • Svensson P.
        • Gonzalez Y.
        • Lobbezoo F.
        • Michelotti A.
        • Brooks S.L.
        • Ceusters W.
        • Drangsholt M.
        • Ettlin D.
        • Gaul C.
        • Goldberg L.J.
        • Haythornthwaite J.A.
        • Hollender L.
        • Jensen R.
        • John M.T.
        • de Laat A.
        • de Leeuw R.
        • Maixner W.
        • van der Meulen M.
        • Murray G.M.
        • Nixdorf D.R.
        • Palla S.
        • Petersson A.
        • Pionchon P.
        • Smith B.
        • Visscher C.M.
        • Zakrzewska J.
        • Dworking S.F.
        Diagnostic Criteria for Temporomandibular Disorders (DC/TMD) for clinical and research applications: recommendations of the International RDC/TMD Consortium Network* and Orofacial Pain Special Interest Group.
        J Oral Facial Pain Headache. 2014; 38: 6-27https://doi.org/10.11607/jop.1151
        • Stegenga B.
        • de Bont L.G.
        • de Leeuw R.
        • Boering G.
        Assessment of mandibular function impairment associated with temporomandibular joint osteoarthrosis and internal derangement.
        J Orofac Pain. 1993; 7: 183-195
        • Kahan B.
        • Morris T.
        Adjusting for multiple prognostic factors in the analysis of randomised trials.
        BMC Med Res Methodol. 2013; 1399https://doi.org/10.1186/1471-2288-13-99
        • Bates D.
        • Mächler M.
        • Bolker B.
        • Walker S.
        Fitting linear mixed-effects models using lme4.
        J Stat Softw. 2015; 67: 1-48
        • Guarda-Nardini L.
        • de Almeida A.
        • Manfredini D.
        Arthrocentesis of the temporomandibular joint: systematic review and clinical implications of research findings.
        J Oral Facial Pain Headache. 2021; 35: 17-29https://doi.org/10.11607/ofph.2606
        • Monje-Gil F.
        • Nitzan D.
        • González-Garcia R.
        Temporomandibular joint arthrocentesis. Review of the literature.
        Med Oral Patol Oral Cir Bucal. 2012; 17: e575-e581https://doi.org/10.4317/medoral.17670
        • Emshoff R.
        • Rudisch A.
        Determining predictor variables for treatment outcomes of arthrocentesis and hydraulic distention of the temporomandibular joint.
        J Oral Maxillofac Surg. 2004; 62: 816-823https://doi.org/10.1016/j.joms.2003.12.020
        • Al-Moraissi E.A.
        • Wolford L.M.
        • Ellis E.
        • Neff A.
        The hierarchy of different treatments for arthrogenous temporomandibular disorders: a network meta-analysis of randomized clinical trials.
        J Craniomaxillofac Surg. 2020; 48: 9-23https://doi.org/10.1016/j.jcms.2019.10.004
        • Li D.
        • Wong N.
        • Li S.
        • McGrath C.
        • Leung Y.
        Timing of arthrocentesis in the management of temporomandibular disorders: an integrative review and meta-analysis.
        Int J Oral Maxillofac Surg. 2021; 50: 1078-1088https://doi.org/10.1016/j.ijom.2021.01.011
        • Tatli U.
        • Belidayi M.
        • Ekren O.
        • Salimov F.
        Comparison of the effectiveness of three different treatment methods for temporomandibular joint disc displacement without reduction.
        Int J Oral Maxillofac Surg. 2017; 46: 603-609https://doi.org/10.1016/j.ijom.2017.01.018
        • de Leeuw R.
        • Boering G.
        • van der Kuijl B.
        • Stegenga B.
        Hard and soft tissue imaging of the temporomandibular joint 30 years after diagnosis of osteoarthrosis and internal derangement.
        J Oral Maxillofac Surg. 1996; 54: 1270-1280https://doi.org/10.1016/s0278-2391(96)90480-2
        • de Leeuw R.
        • Boering G.
        • Stegenga B.
        • de Bont L.
        Radiographic signs of temporomandibular joint osteoarthrosis and internal derangement 30 years after nonsurgical treatment.
        Oral Surg Oral Med Oral Pathol Oral Radiol. 1995; 79: 382-392https://doi.org/10.1016/S1079-2104(05)80233-1
        • Ohlmann B.
        • Rammelsberg P.
        • Henschel V.
        • Kress B.
        • Gabbert O.
        • Schmitter M.
        Prediction of TMJ arthralgia according to clinical diagnosis and MRI findings.
        Int J Prosthodont. 2006; 19: 333-338
        • Honda K.
        • Yasukawa Y.
        • Fujiwara M.
        • Abe T.
        • Urade M.
        Causes of persistent joint pain after arthrocentesis of temporomandibular joint.
        J Oral Maxillofac Surg. 2011; 69: 2311-2315https://doi.org/10.1016/j.joms.2011.02.025
        • Emshoff R.
        • Rudisch A.
        Are internal derangement and osteoarthrosis linked to changes in clinical outcome measures of arthrocentesis of the temporomandibular joint.
        J Oral Maxillofac Surg. 2003; 61: 1162-1167https://doi.org/10.1016/S0278-2391(03)00676-1
        • Rollman G.B.
        • Gillespie J.M.
        The role of psychosocial factors in temporomandibular disorders.
        Curr Rev Pain. 2000; 4: 71-81https://doi.org/10.1007/s11916-000-0012-8