Behnam Aghabeigi studies patients with chronic facial pain

Facial arthromyalgia (FAM) or the temporomandibular joint pain dysfunction syndrome is a common condition in which patients complain of pain and tenderness in one or both temporomandibular joints (TMJ), often with limitation of jaw opening informs Dr. Behnam Aghabeigi Birmingham. The condition is four times more common in females than males and there are many reports linking these symptoms to adverse life events, stress or the lack of emotional support. This condition can occur independently or with other non-muscular non-joint pain in the face (atypical facial pain, AFP) or teeth (atypical odontalgia, AO). These are also commonly related to idiopathic head, neck and back pain, irritable bowel and pruritus. The facial pains are best controlled with tricyclic antidepressants even in the absence of depression. Recently we have shown that these patients also have impaired excretion of conjugated tyramine, a biological trait marker seen in endogenous depression suggesting a common metabolic disturbance predisposes to both pain and depression.

However, the precise underlying biochemical mechanisms leading to both pain and joint dysfunction remain to be established. In an attempt to account for the joint pain and dysfunction Dr. Ben Aghabeigi Birmingham’s attention was drawn to studies claiming to demonstrate that emotional stress and pain in animals were associated with an increased generation of free radical@-’ and by the observation that stress induced damage to the gastric mucosa was related to free radical production. ‘,i” Furthermore, there have been reports that free radical activity in synovial fluid from the knee joints of rheumatoid patients correlates with the severity of the disease.” A free radical is any molecule or atom that contains one or more unpaired electrons rendering it highly reactive. Most biological molecules such as O2 or H,O are non-radicals, containing only paired electrons.

In addition to causing pain in animals, in vitro experiments have shown that free radicals depolymerise hyaluronic acid producing lower synovial fluid viscosity,” which might impair lubrication and lead to meniscal hesitation and clicking, as originally proposed by Toller.i3 There has also been evidence that free radicals are associated with cartilage damage14 and that they can stimulate bone resorption.” Furthermore, the demonstration of the presence of eicosanoids in various inflammatory joint diseases,” which could be the product of a free radical and or neuropeptide synovitis, would fit their known role as one of the important mediators of chronic algesia and hyperalgesia. Therefore we have examined the possibility that FAM may, in part, result from the inappropriate production of free radicals in susceptible individuals.

Three parameters of free radical generation were measured in patients presenting with overt symptoms of FAM and/or a history of idiopathic oro-facial pain (AFP and AO): (I) Systemic free radical activity was studied by a comparison of the free radical production of 2,3-dihydroxybenzoic acid (DHB) from an oral dose of aspirin as opposed to the normal aspirin metabolic product 2,5-DHB.17

Intra-articular free radical activity was investigated by two methods.

One was a thiobarbituric acid (TBA) assay of saline TMJ aspirates to test for intra-articular production of lipid peroxidation products” and (III) the second was measurement of the production of intra-articular hyperalgesic eicosanoids PGE,, LTB, and 15-HETE.

Materials and Methods used by Dr. Ben Aghabeigi Birmingham


Three sets of patients were recruited for this study. Systemic free radical activity was studied in the first group of patients who were diagnosed as having chronic FAM and/or other idiopathic oro-facial pain of more than 3 months duration. Intra-articular free radical activity was studied in groups II and III which comprised patients with unilateral symptoms of TMJ pain which had been unresponsive to 12 weeks tricyclic antidepressant therapy and were undergoing TMJ arthroscopy under general anaesthesia. All the subjects gave their informed consent and none had any other joint disease or known or suspected history of allergy to aspirin. Ethical approval was obtained for all procedures.

Group I (systemic free radical activity): 10 pain patients (age range 26-64, mean 41.8 + 11; 9 females, 1 male) and 10 healthy, age- and sex-matched volunteers with no prior history of idiopathic pain were recruited as controls (age range 29-60, mean 42.129.6). These patients and control subjects had 10 ml of venous blood drawn in heparinised tubes and voided their bladders to provide a urine sample. Each subject was then administered an oral dose of 1.2 g of aspirin and after 2 h repeat blood and urine samples were collected. The blood samples were centrifuged immediately and the plasma and urine samples stored at – 70°C until assayed for 2,3-DHB.

Group II consisted of 18 patients (age range 22-49, mean 33.2+ 8.1; 13 females, 5 males). Two hours before arthroscopy the patients were administered 1.2 g of Aspirin orally in order to ensure equilibration between the plasma and synovial fluid. At arthroscopy 1 ml of normal saline was injected into the joint spaces bilaterally, allowed time to mix with the synovial fluid and aspirated through the same needle. Specimens with overt contamination with blood were discarded. The aspirate volumes were determined, 50 ul removed for haemoglobin assay and the remainder was centrifuged immediately before the supernatants were stored at -70°C. A venous blood sample was drawn into heparinised tubes at the same time as the synovial aspirates were collected, centrifuged and the plasma stored at -70°C until assayed for lipid peroxidation products by TBA assay.

Group III consisted of 15 patients (age range 15-41, mean 28.3 +7.4; 9 females, 6 males). Synovial aspirates were collected as described above and retained for hyperalgesic eicosanoid analysis, specifically prostaglandin E2 (PGE2), leukotriene B, (LTB,) and 15-hydroxyeicosatetraenoic acid ( 15HETE). These subjects did not receive aspirin because of its potential inhibitory effect on eicosanoid production.

Biochemical analyses

(a) Measurement of plasma and urine Measurement of plasma and urine 2,3- and 2,5-dihydroxybenzoic acid (DHB) was by a modification of the methods of Grootveld and Halliwell” using HPLC coupled to electrochemical detection. All samples were used immediately or stored at – 70°C until analysed.

(i) Plasma sample preparation for the DHB assay. Samples (1 ml) were acidified with 50 ul of 1 M HCI and 20 ul of an internal standard of 3,4-DHB (100 PM) added before the samples were extracted with 10 ml of ethyl acetate. The samples were then centrifuged at 1500 g for 10 min before the upper organic layer was decanted into a clean dry tube and reduced to dryness under a stream of air in a water bath at 40°C. Once dry the samples were redissolved in 200 nl of mobile phase and 50 ~1 of 1 M HCl was added prior to analysis.

(ii) Urine sample preparation for DHB assay. Samples (5 ml) were acidified with 1 ml 1 M HCl before being extracted twice with 10 ml of ethyl acetate. The combined ethyl acetate extracts were taken to dryness under a stream of air in a water bath at 40°C. The residues were reconstituted in 250 ul of 0.2 M HCl and subsequently diluted 1 in 20 or 1 in 50 with 0.2 M HCl before final analysis.

(iii) HPLC purification and electrochemical detection of DHBs. After the initial extraction of both the plasma and the urine samples, the HPLC purification of 2,3- and 2,5-DHBs were the same, with a minor modification to the detection system to facilitate a single run determination of the urinary levels of 2,5-DHB due to the large amounts of this aspirin metabolite found in urine. HPLC separations were run under isocratic conditions using a SpectraPhysics SP8800 pump with a Brownlee 5 pm ODS reverse phase column (250 x4.6 mm) coupled to an EDT instruments LCA 16 electrochemical detector equipped with a glassy carbon working electrode and a Ag/AgCl reference electrode operated in the oxidation mode. The mobile phase was 30 mM sodium titrate/27.7 mM sodium acetate buffer (pH 4.75) at a flow rate of 1 ml min- i. The mobile phase was sparged with helium gas and the eluent monitored electrochemically at an electrode potential of + 0.6 V.

When the urine samples were analysed it was found to be possible to obtain quntifiable peaks equivalent to 2,3- and 2,5-DHBs in a single run when the recorder sensitivity was decreased lo-fold after the elution of the 2,3-DHB peak.

Synovial analyses for lipid peroxidation

(b) Thiobarbituric acid test

A modified method of Rowley et al.” was used in this study; briefly, 125 yl of sample-plasma or synovial Auid- was added to 250 ul of TBA solution (1% w/v in 50 mM sodium hydroxide), 250 ul hydrochloric acid (25% v/v) and 200 ul of water; 125 ul of water was used in the place of sample as a negative control. The tubes were tightly capped and heated at 100°C for 1 h, after which they were allowed to cool to room temperature before being extracted into 1.5 ml l-butanol with vigorous mixing for 2 min. The samples were then centrifuged at 1500 g at 4°C for 15 min and the absorbance of upper organic layer determined at 532 nm.

(c) Haemoglobin measurement

Haemoglobin levels in the samples were quantified using a commercial calorimetric assay (Sigma Chemical Co.). Some of the samples were found to have low levels of haemoglobin which could only be measured by increasing the volume of the sample in the assay system from 20 ul to 50 ~1.

(d) Salicylate assat,

Salicylate was measured in both joint aspirate and venous blood by a modified method of Grootveld and Halliwelli7 using high performance liquid chromatography (HPLC) and UV detection. Samples (200 ~1) were treated with 25 ul of 1 M HCl before being extracted with 10 ml of diethyl ether. After separation, the ether layer was evaporated in a water bath at 40°C and the residue was dissolved in 225 ulof the HPLC mobile phase containing 5% (v/v) 1 M HCl. Samples not assayed immediately were stored at -70°C until used. Separation was done on a SpectraPhysics SP8800 HPLC pump operated in isocratic mode using a Shandon 5 m ODS reverse-phase column (250 x 4.6 mm). The mobile phase was 30 mM sodium titrate/27.7 mM sodium acetate buffer (pH 4.75) and methanol (94: 6) at a flow rate of 1 ml min-‘. The mobile phase was sparged with helium gas and the eluent monitored at 254 nm.

(e) Measurement of eicosanoids

Synovial aspirates from symptomatic and symptomfree joints were analysed for the presence of PGE,, LTB, and 15-HETE using commercial radioimmunoassay kits (Amersham International). The assays were conducted according to the manufacturer’s instruction.

Statistical analysis

A student’s r-test was used for analysing the parametric data in patients vs controls and symptomatic vs asymptomatic joints.


Group I

Healthy control subjects and patients presenting with chronic idiopathic oro-facial pain did not have statistically different circulating levels of the principle 2,5-DHB metabolite of aspirin indicating that the metabolic factors governing aspirin clearance were not different between the two groups. However, the circulating levels of 2,3-DHB, the suggested product of free radical activity,” was significantly elevated in the pain patients, whereas 5 out of 10 of the control subjects were found to have no detectable levels of this compound. The urine concentrations of both metabolites did not differ between the groups.

Group II

The yield of aspirate ranged from 500 ul to 1050 ul, there being no significant volumetric difference between the symptomatic and symptom-free joints. There was no significant difference in the levels of TBA-RS between the synovial fluids from the symptomatic and symptomless joints. Approximately half of the samples had haemoglobin contamination, but the contribution to the measured levels of TBA-RS did not significantly alter the analysis of the data. The synovial fluid volume was calculated using a concentration volume equation based on the plasma to TMJ aspirate salicylate ratio.ig This ratio was not significantly different between the symptomatic and symptomless joints, reflecting the absence of any difference in synovial fluid volume between painful and pain-free joints.

Group III

The presence of high levels of 15-HETE, but unmeasurable amounts of both PGE, and LTB, were found in both symptomatic and symptom-free synovial aspirates (Table 2). There was no statistical difference between the levels of 15-HETE in the synovial fluids from symptom-free and painful joints.

Discussion with Aghabeigi dentist

We have measured the levels of 2,3-DHB in both the circulation and in the urine of a group of 10 patients attending a clinic for chronic idiopathic oro-facial pain, before and after the administration of 1.2 g of aspirin and these levels have been compared to ageand sex-matched control subjects (Table 1). There was no indication that either group had differences in their innate abilities to metabolise Aspirin, as assessed by the production of 2,5-DHB, the major product of mammalian metabolism of aspirin. However, there were significantly higher levels of the free radical generated 2,3-DHB in the plasma of the patient group as compared to the control subjects. It has been proposed that the production of 2,3-DHB from an oral dose of aspirin results from a free radical reaction with salicylate.” This would support the hypothesis that patients with chronic idiopathic pain which is commonly stress related appear to generate increased free radicals or have some impairment in their scavenging metabolism.

The urinary levels of 2,3-DHB was higher in patients as compared to the controls. Although this was not statistically significant, it could be biologically important. We are not competent to comment further on the complexity of urinary free radical metabolism, which probably required some form of clearance measurement. During the past decade, saline aspirates of the upper joint space of the TMJ have been analysed for the presence of various mediators of pathological conditions.20-24 In this study we have also analysed saline aspirates, from patients presenting with a history of chronic FAM who were undergoing arthrostopic examination, for the potential to generate, in vivo, free radicals and intra-articular eicosanoids. We believe that this approach is fraught with difficulties, especially as the volumetric yield from a collection of TMJ aspirate is variable, in our case ranging from 500 ul to 1050 ~1. The volume collected is probably dependent on operator technique and the patency of the joint space.

However, the mean volume of the aspirates collected in this study is much higher than previously reported” or than that collected in a pilot study done in our laboratory.” Arthroscopic joint trauma, giving rise to haemoglobin and cellular contamination of some of the samples was an indication of an important source of error. This was controlled by the preoperative aspirin and by ensuring that centrifugation of the aspirates was carried out immediately, and would also explain, at least in part, previously reported unusually high levels of inflammatory mediators that are not appropriate to this condition.21 This is an important observation that has not previously been commented on by other workers, who have found levels of neuropeptides, prostaglandins and leukotrienes to be higher in TMJ synovial fluid than those found in the synovial fluid collected from the inflamed knee joint in gout.

Although measurement of TBA-RS is one of the most widely used single assays for determining lipid peroxidation indicating free radical activity, it has been criticised for its lack of specificity when applied to human body fluids. i9 It is interesting that we have also found TBA-RS in the saline aspirates from both symptomatic and non-symptomatic TMJs of patients with FAM and that there is no statistical difference between the sides. This suggests that other as yet unknown algesic factors contribute to the localisation of the pain and dysfunction. An alternative influence may be the suspension of masticatory activity by a patient who has fasted overnight prior to the general anaesthetic.

An interesting piece of supporting evidence for the involvement of free radicals in the pathogenesis of FAM is our demonstration of high intra-articular concentrations of the hyperalgesic mediator 15-hydroxyeicosatetraenoic acid ( 15-HETE), whose synthesis involves the free radical mediated process of lipid peroxidation of arachidonic acid, in synovial fluid. We were unable to demonstrate the presence of either prostaglandin E2 (PGE,) or leukotriene B4 (LTB,). It is worth repeating that the eicosanoid levels found by previous investigators seem to be artifactually raised even when compared to severe inflammatory disease in other joints.“j It is of importance that hyperalgesia induced by 15-hydroperoxyeicosatetraenoic acid ( 15-HPETE) in an experimental animal can substantially prolong the algesic effect of substance P(SP) producing a chronic pain model not dissimilar to FAM. This is not inhibited by non-steroidal anti-inflammatory analgesics apart from dipyrone.

Furthermore, a SP antagonist can block this effect. These findings correlate with other studies which have identified neuropeptides in the synovial fluid from the TMJ27,28 and our own observations2’ which have demonstrated that the TMJ capsule is not only richly innervated by SP neuronal tissue, but also other neurogenic peptides including calcitonin gene-related peptide, neuropeptide Y and vasoactive intestinal polypeptide. One of the major clinical problems in controlling FAM is the poor response to non-steroidal anti-inflammatory analgesics, which would correlate with the role of hyperalgesic 15-HPETE as being more important than the prostaglandins such as PGE,. As stated there were no significant differences between the symptomatic and symptom-free joints with respect to TBA-RS, 15-HETE or synovial fluid volume.

Unfortunately, as it was not ethically possible to obtain saline aspirates from the joints of healthy age- and sex-matched pain-free adults, one can only speculate that these levels found represent the pathological process. This absence of difference is not completely surprising considering that a systemic biochemical disorder would be reflected in both joints at the ends of a single bone. Furthermore, the mirror imaging of inflammatory responses in other paired joints in the body which lack the unique anatomical and functional characteristics of TMJ has been attributed to neurophysiological influences.30 However, the presence of potential pain mediators in the symptomless joints also suggests the importance of other factors such as local neuropeptide or cytokine release which may be dependant on asymmetrical masticatory function and bruxism, or personality factors which influence central modulation of the pain experience.

For more information about Behnam Aghabeigi visit here :–publications.html