Journal of
Oral Rehabilitation
Journal of Oral Rehabilitation 2014 41; 250--256
Cytokines in healthy temporomandibular joint synovial fluid € ELER*§, K . D . K R I S T E N S E N * † , P . A L S T E R G R E N ‡ , P . S T O U S T R U P * , A . K US T . H E R L I N ¶ & T . K . P E D E R S E N * § *Section of Orthodontics, Faculty of Health Sciences, Aarhus University, Aarhus, Denmark, †Specialist Oral Health Center for Western Norway, Stavanger, Norway, ‡Orofacial Pain Unit, Department of Orofacial Pain and Jaw Function, Malm€o University, Malm€o, Sweden, §Department of Maxillofacial Surgery, Aarhus University Hospital, Aarhus, and ¶
Department of Pediatrics, Aarhus University Hospital, Aarhus, Denmark
SUMMARY Analysis of temporomandibular joint (TMJ) synovial fluid may elucidate the aetiology of temporomandibular disorders and arthritic conditions, as well as the inflammatory mechanisms involved. Knowledge about healthy synovial fluid is necessary to understand TMJ pathologies. We aimed to quantify the proinflammatory cytokines interleukin (IL)-1b, IL-2, IL-6 and tumour necrosis factor (TNF), and the anti-inflammatory cytokines IL-10 and interferon (IFN)-c in healthy TMJ synovial fluid to serve as reference values for future studies on TMJ pathologies. Twenty healthy, young adult volunteers without temporomandibular dysfunction were included. Bilateral synovial fluid samples were obtained using the push-pull technique with hydroxocobalamin described by Alstergren in 1999. Cytokines were quantified with Luminex multiplex assays and compared using nonparametric statistical analysis. No serious adverse effects were reported. Of 40 possible
Introduction Synovial fluid is viscous and has lubricating, metabolic and regulatory functions within synovial joints (1). Cytokines and growth factors are present in synovial fluid and are important regulatory factors for the immune and metabolic systems in the joint. Cytokines are secreted by immunocompetent cells in the synovial lining, by chondrocytes or originate from plasma. The cytokines may be categorised as proinflammatory © 2014 John Wiley & Sons Ltd
samples, 14 fulfilled the strict sampling criteria and were included in the analysis. Cytokine values (reported as medians with interquartile ranges) were as follows: TNF, 23 (13–37) pg mL1; IL-2, 18 (0–22) pg mL1; and INF-c, 10 (0–47) pg mL1. IL-1b, IL-6 and IL-10 were almost undetectable. In addition, TNF and INF-c cytokine levels correlated. We demonstrated that TNF was consistently detected and IFN-c and IL-2 sporadically detected in the TMJ synovial fluid of healthy individuals using the hydroxocobalamin method and a multiplex assay. The cytokines IL-10, IL-1b and IL-6 were barely detectable in this sample of healthy TMJs. KEYWORDS: arthrocentesis, intra-articular injections, temporomandibular joint, synovial fluid, cytokines, adult Accepted for publication 16 January 2014
(e.g. interleukin (IL)-1b, IL-2, IL-6 and tumour necrosis factor (TNF, formerly known as TNF-a) or antiinflammatory (e.g. IL-10 and interferon (IFN)-c) according to their predominant tissue-specific effects (2). In inflammatory joint diseases or injury, the cytokine balance in synovial fluid is altered, favouring proinflammatory cytokines (1). Inflammation of the temporomandibular joint (TMJ) may cause cartilage and bone tissue degeneration and have a detrimental effect on cartilage and doi: 10.1111/joor.12146
CYTOKINES IN HEALTHY TMJ SYNOVIAL FLUID bone tissue formation and growth through an interaction between endochondral ossification and the synovial pathology (3). Mechanisms underlying the TMJ pathology may be investigated by analysis of inflammatory mediators in the synovial fluid because inflammation is a major contributor to pain and degradation in articular tissues and may result in severe growth impairment in developing individuals (4). Proinflammatory cytokines such as IL-1b, TNF, IL-2, IL-6, IL-8 and IL-12 have been associated with various TMJ disorders including closed lock, osteoarthritis and internal derangement (5–7). Also, in rheumatoid arthritis (RA) patients, IL-1b, IL-1 receptor antagonist, soluble IL-1 receptor type II, TNF, TNF soluble receptor II, serotonin and prostaglandin E2 may be increased (8) and associated with TMJ pain and radiological signs of disease and progression (9). However, many studies have used small patient samples and even smaller samples of healthy controls, and the methodology used did not enable assessment of the true synovial fluid concentration. The aim of this study was to characterise healthy TMJ synovial fluid and quantify the proinflammatory cytokines IL-1b, IL-2, IL-6 and TNF as well as the anti-inflammatory cytokines IL-10 and IFN-c to serve as references for future studies on TMJ pathologies.
Materials and methods Patients Twenty healthy, young adults, eight females and 12 males, with a median (25th/75th percentile) age of 23 (22/25) years voluntarily agreed to participate in the present study. The inclusion criterion was a statement from the subjects that their craniofacial area was healthy. Exclusion criteria were subjective signs of temporomandibular dysfunction (TMD), as assessed by a modified version of the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) criteria (10), although subjects with non-painful joint clicking or bruxism were allowed. Subjects with ongoing pain disorders and currently taking medication were excluded. Clinical assessment A functional clinical assessment of the TMJ and related structures was performed to verify normal © 2014 John Wiley & Sons Ltd
joints and mandibular function. A RDC/TMD (10) questionnaire modified with an assessment of dentofacial asymmetries, and soft tissue function was used because it was considered suitable for our patient group. The questionnaire has been used previously in several studies, for example (11, 12), and comprised a self-assessment of the present TMJ status where the subjects answered questions addressing areas like pain or symptoms related to the TMJ area, facial and masticatory muscles, TMJ function, chewing ability, etc. The clinical examination comprised clinically relevant parameters, for example joint clicking, mandibular range of motion measurements and deviations, and muscle tenderness. All subjects gave their informed consent before their inclusion in the present study, which was approved by the Danish Scientific Ethical Committee (approval number: 20100002). Temporomandibular joint fluid sampling Two experienced and specially trained operators each placed a needle in the superior TMJ compartment bilaterally after disinfecting the skin with 70% ethanol and 5% chlorhexidine, and an auriculotemporal nerve block was applied. Temporomandibular joint synovial fluid samples were obtained with a push-pull technique, and the amount of recovered synovial fluid in each sample was quantified with the hydroxocobalamin method, as described by Alstergren et al. (13, 14). Briefly, a washing solution consisting of 22% hydroxocobalamin (Behepanâ 1 mg mL1*) in physiological saline (sodium chloride 9 mg mL1) was used as a washing solution. Each TMJ was washed with a total of 4 mL washing solution (13) through a stop-cock syringe. One millilitre of washing solution was injected slowly, the valve was turned and then as much fluid as possible was aspirated back (14). The injection/aspiration was repeated a total of four times for each joint leaving the same cannula inside the joint. If aspiration of the washing solution was possible and the resistance in the syringe was minor during injection, then the needle tip was considered to be placed within the joint space. The aspirates were evaluated for blood contamination (no, minor, clear or excessive) and centrifuged at 1500 g at 4 °C for 10 min (14). Then, samples from
*Kabi Pharmacia, Uppsala, Sweden.
251
252
K . D . K R I S T E N S E N et al. each joint were aliquoted into five Eppendorf tubes and frozen at 80 °C for later spectrophotometry and Luminex cytokine analyses. Also, washing solution from each subject was frozen for later spectrophotometry analysis. Spectrophotometry The absorbance at 350 nm was measured for each sample and for each washing solution (spectrophotometer: Shimadzo UV-160A†) with a capillary tube system consisting of a quartz capillary tube (07 mm in diameter, 3 lL/sample) and a capillary tube holder†. The absorbance of the aspirates and washing solutions were measured four times in the spectrophotometer, and the mean value was used. The dilution detection limit of the washing solution in the aspirate by this method was 09% (13). Cytokine quantification IL-1b, IL-2, IL-6, IL-10, TNF and IFN-c were analysed with a high-sensitivity human cytokine kit for simultaneous multianalyte detection with Luminex technology and instrumentation (Lincoplex kit‡; Bio-Plex 200 System apparatus§). The minimum detection concentrations in the assays were 006 (IL-1b), 016 (IL-2), 010 (IL-6), 015 (IL-10), 005 (TNF) and 029 (IFN-c) pg mL1, according to the manufacturer’s data. The intra-assay coefficients of variation were 311 (IL-1b), 427 (IL-2), 351 (IL-6), 331 (IL-10), 349 (TNF) and 488 (IFN-c). Undetectable levels were set to ‘0’ in the statistical analysis. The synovial fluid concentration of each cytokine was calculated using the formula: CSF ¼
volume >05 mL, no or only minor blood contamination and no haemolysis) were included in the analysis (14). Statistics Nonparametric statistical analysis was used. The median, 25th and 75th percentiles and number of observations were reported for descriptive statistics of the variables. For analytical statistics, the Spearman’s rank correlation test was used to calculate the significance of the correlation between separate variables, and the Wilcoxon’s rank-sum test was used for intra-group differences. A P < 005 was considered significant.
Results A total of 14 samples from 40 TMJs fulfilled the sample quality criteria (14). These 14 samples were obtained from 10 subjects. The reasons for not fulfilling the sample criteria were blood contamination (n = 13), a dilution factor >098 (n = 9) or too low recovery of synovial fluid (n = 4). The distribution of the recovered volumes of synovial fluid is shown in Fig. 1. Significantly more synovial fluid was recovered from female subjects compared to males (females, median: 172 lL, IQR: 44–277, n = 7 and males, median: 56 lL, IQR: 37–85, n = 7; P = 0048). As shown in Table 1, TNF was detected in all joints with
CASP AbsAsp 1 AbsWash
where CSF = synovial fluid concentration of a given cytokine, CAsp = aspirate concentration of the given cytokine measured by the Luminex assay, AbsAsp = aspirate absorbance and AbsWash = washing solution absorbance. Samples that fulfilled the sample quality criteria (dilution factor 0’: percentage of samples with detectable cytokine levels (n = 14)
IL-6 IL-10 TNF IL-1b IL-2 IFN-c
Median (pg mL1)
IQR (pg mL1)
Range (pg mL1)
>0 (%)
0 0 23 0 18 10
0 0 13–37 0 0–22 0–47
0–101 0–263 32–98 0–56 0–455 0–178
71 71 100 143 714 50
a median value of 23 pg mL1, but ranged from 32 to 98. IL-2 was found in 71% of joints with a median value of 18 pg mL1, ranging from 0 to 455. IFN-c was found in half of the joints with a median value of 10 pg mL1, ranging from 0 to 178. The median IL1b, IL-6 and IL-10 levels were zero. IL-1b was only detected in 2 and IL-6 and IL-10 in 1 of the 14 joints. Regarding the two joints with useful samples with non-symptomatic clicking, the values were as follows: joint 1: TNF: 358 pg mL1, IL-1b: 382 pg mL1, IL2: 179 pg mL1, IFN-c: 469 pg mL1 and joint 2: TNF: 124 pg mL1, IL-2: 09 pg mL1. There was a significant correlation between the TNF and IFN-c concentrations in TMJ synovial fluid (rs = 078, n = 14, P < 0001) (Fig. 2), but not between any of the other cytokines.
Fig 2. Scatter plot showing the relation between the IFN-c and TNF concentrations in the TMJ synovial fluid of 14 healthy individuals (rs = 078, n = 14, P < 0001). © 2014 John Wiley & Sons Ltd
Discussion The results of the present study indicate that TNF is consistently detected in the TMJ synovial fluid of healthy individuals and that the cytokines IFN-c and IL-2 are sporadically detected using the hydroxocobalamin (B12) method and a multiplex assay. The levels should be considered as homeostatic. The cytokines IL-10, IL-1b and IL-6 were not frequently detectable in the healthy individuals, suggesting those cytokines are more specific for inflammatory environments. The bilateral samples obtained from four patients were not systematically different from the unilateral samples. The individual bilateral levels differed and showed a similar variation in the same individual. Homeostatic cytokine levels thus seem to vary intra-individually. Alstergren et al. previously reported a hydroxocobalamin dilution method as an accurate and reproducible technique to determine the true synovial fluid concentration of a mediator after a joint washing, even in low sample volumes (15). In this study, synovial samples from the healthy TMJ were obtained, and the true synovial fluid concentration of cytokines was determined by the hydroxocobalamin dilution sampling method. This is one of the first reports using this methodology in healthy individuals, and our results could provide a basis for a standard reference for future TMJ studies on, for example, osteoarthritis, JIA or RA patients. In healthy individuals, a lower proportion of samples fulfilling the sample quality criteria may be expected compared to sampling from patients with TMJ arthritides due to the small volume of synovial fluid present in the non-inflamed joint. Synovial fluid in the healthy TMJ is a thin viscous gel-film lubricating the joint surfaces, which makes sampling difficult. The sample quality criteria applied in the present study and as determined by Alstergren et al. (14) set a high standard for acceptable sample quality, as no contamination was tolerated to detect the expected low levels of cytokines. Furthermore, the amount of recovered synovial fluid in healthy individuals seems to vary, for example as indicated by the difference between males and female; thus, not all healthy individuals are suitable for TMJ synovial fluid sampling. Therefore, we observed that 35% of the samples from healthy donors fulfilled our sample quality criteria, which was expected but considerably less than the 52–83% of samples found applicable for cytokine
253
254
K . D . K R I S T E N S E N et al. studies in patients with various TMDs (14, 16–18). Still, a reasonable proportion of samples from healthy individuals fulfilled the criteria, and the true synovial fluid concentrations were determined from them. The method can be used in healthy subjects, but non-useful samples should be expected. However, even though the method is considered safe and without large risks, it is still an intra-articular injection and brief post-operative symptoms can be expected. Also, more serious side effects as infection could potentially occur, so keeping the number of participants as low as possible with the aim of the study in mind is advised. The reason for choosing the hydroxocobalamin as an external standard was that it enables measurement of the total volume of synovial fluid recovered, and therefore, true synovial fluid cytokine concentrations can be calculated. There are a number of other TMJ synovial fluid sampling methods that have been used in previous studies of cytokine content in the joint aspirate with either no compensation for the joint washing (6, 7, 19) or the ratio between the measured cytokine concentration and the total protein content of the aspirate (5, 20). The method with hydroxocobalamin as an external marker enables the quantification of the total amount of synovial fluid and subsequent measurement of the dilution factor by spectrophotometry of the joint aspirate. It thereby enables determination of the true synovial fluid content in the joint with a detection limit of
Oral Rehabilitation
Journal of Oral Rehabilitation 2014 41; 250--256
Cytokines in healthy temporomandibular joint synovial fluid € ELER*§, K . D . K R I S T E N S E N * † , P . A L S T E R G R E N ‡ , P . S T O U S T R U P * , A . K US T . H E R L I N ¶ & T . K . P E D E R S E N * § *Section of Orthodontics, Faculty of Health Sciences, Aarhus University, Aarhus, Denmark, †Specialist Oral Health Center for Western Norway, Stavanger, Norway, ‡Orofacial Pain Unit, Department of Orofacial Pain and Jaw Function, Malm€o University, Malm€o, Sweden, §Department of Maxillofacial Surgery, Aarhus University Hospital, Aarhus, and ¶
Department of Pediatrics, Aarhus University Hospital, Aarhus, Denmark
SUMMARY Analysis of temporomandibular joint (TMJ) synovial fluid may elucidate the aetiology of temporomandibular disorders and arthritic conditions, as well as the inflammatory mechanisms involved. Knowledge about healthy synovial fluid is necessary to understand TMJ pathologies. We aimed to quantify the proinflammatory cytokines interleukin (IL)-1b, IL-2, IL-6 and tumour necrosis factor (TNF), and the anti-inflammatory cytokines IL-10 and interferon (IFN)-c in healthy TMJ synovial fluid to serve as reference values for future studies on TMJ pathologies. Twenty healthy, young adult volunteers without temporomandibular dysfunction were included. Bilateral synovial fluid samples were obtained using the push-pull technique with hydroxocobalamin described by Alstergren in 1999. Cytokines were quantified with Luminex multiplex assays and compared using nonparametric statistical analysis. No serious adverse effects were reported. Of 40 possible
Introduction Synovial fluid is viscous and has lubricating, metabolic and regulatory functions within synovial joints (1). Cytokines and growth factors are present in synovial fluid and are important regulatory factors for the immune and metabolic systems in the joint. Cytokines are secreted by immunocompetent cells in the synovial lining, by chondrocytes or originate from plasma. The cytokines may be categorised as proinflammatory © 2014 John Wiley & Sons Ltd
samples, 14 fulfilled the strict sampling criteria and were included in the analysis. Cytokine values (reported as medians with interquartile ranges) were as follows: TNF, 23 (13–37) pg mL1; IL-2, 18 (0–22) pg mL1; and INF-c, 10 (0–47) pg mL1. IL-1b, IL-6 and IL-10 were almost undetectable. In addition, TNF and INF-c cytokine levels correlated. We demonstrated that TNF was consistently detected and IFN-c and IL-2 sporadically detected in the TMJ synovial fluid of healthy individuals using the hydroxocobalamin method and a multiplex assay. The cytokines IL-10, IL-1b and IL-6 were barely detectable in this sample of healthy TMJs. KEYWORDS: arthrocentesis, intra-articular injections, temporomandibular joint, synovial fluid, cytokines, adult Accepted for publication 16 January 2014
(e.g. interleukin (IL)-1b, IL-2, IL-6 and tumour necrosis factor (TNF, formerly known as TNF-a) or antiinflammatory (e.g. IL-10 and interferon (IFN)-c) according to their predominant tissue-specific effects (2). In inflammatory joint diseases or injury, the cytokine balance in synovial fluid is altered, favouring proinflammatory cytokines (1). Inflammation of the temporomandibular joint (TMJ) may cause cartilage and bone tissue degeneration and have a detrimental effect on cartilage and doi: 10.1111/joor.12146
CYTOKINES IN HEALTHY TMJ SYNOVIAL FLUID bone tissue formation and growth through an interaction between endochondral ossification and the synovial pathology (3). Mechanisms underlying the TMJ pathology may be investigated by analysis of inflammatory mediators in the synovial fluid because inflammation is a major contributor to pain and degradation in articular tissues and may result in severe growth impairment in developing individuals (4). Proinflammatory cytokines such as IL-1b, TNF, IL-2, IL-6, IL-8 and IL-12 have been associated with various TMJ disorders including closed lock, osteoarthritis and internal derangement (5–7). Also, in rheumatoid arthritis (RA) patients, IL-1b, IL-1 receptor antagonist, soluble IL-1 receptor type II, TNF, TNF soluble receptor II, serotonin and prostaglandin E2 may be increased (8) and associated with TMJ pain and radiological signs of disease and progression (9). However, many studies have used small patient samples and even smaller samples of healthy controls, and the methodology used did not enable assessment of the true synovial fluid concentration. The aim of this study was to characterise healthy TMJ synovial fluid and quantify the proinflammatory cytokines IL-1b, IL-2, IL-6 and TNF as well as the anti-inflammatory cytokines IL-10 and IFN-c to serve as references for future studies on TMJ pathologies.
Materials and methods Patients Twenty healthy, young adults, eight females and 12 males, with a median (25th/75th percentile) age of 23 (22/25) years voluntarily agreed to participate in the present study. The inclusion criterion was a statement from the subjects that their craniofacial area was healthy. Exclusion criteria were subjective signs of temporomandibular dysfunction (TMD), as assessed by a modified version of the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) criteria (10), although subjects with non-painful joint clicking or bruxism were allowed. Subjects with ongoing pain disorders and currently taking medication were excluded. Clinical assessment A functional clinical assessment of the TMJ and related structures was performed to verify normal © 2014 John Wiley & Sons Ltd
joints and mandibular function. A RDC/TMD (10) questionnaire modified with an assessment of dentofacial asymmetries, and soft tissue function was used because it was considered suitable for our patient group. The questionnaire has been used previously in several studies, for example (11, 12), and comprised a self-assessment of the present TMJ status where the subjects answered questions addressing areas like pain or symptoms related to the TMJ area, facial and masticatory muscles, TMJ function, chewing ability, etc. The clinical examination comprised clinically relevant parameters, for example joint clicking, mandibular range of motion measurements and deviations, and muscle tenderness. All subjects gave their informed consent before their inclusion in the present study, which was approved by the Danish Scientific Ethical Committee (approval number: 20100002). Temporomandibular joint fluid sampling Two experienced and specially trained operators each placed a needle in the superior TMJ compartment bilaterally after disinfecting the skin with 70% ethanol and 5% chlorhexidine, and an auriculotemporal nerve block was applied. Temporomandibular joint synovial fluid samples were obtained with a push-pull technique, and the amount of recovered synovial fluid in each sample was quantified with the hydroxocobalamin method, as described by Alstergren et al. (13, 14). Briefly, a washing solution consisting of 22% hydroxocobalamin (Behepanâ 1 mg mL1*) in physiological saline (sodium chloride 9 mg mL1) was used as a washing solution. Each TMJ was washed with a total of 4 mL washing solution (13) through a stop-cock syringe. One millilitre of washing solution was injected slowly, the valve was turned and then as much fluid as possible was aspirated back (14). The injection/aspiration was repeated a total of four times for each joint leaving the same cannula inside the joint. If aspiration of the washing solution was possible and the resistance in the syringe was minor during injection, then the needle tip was considered to be placed within the joint space. The aspirates were evaluated for blood contamination (no, minor, clear or excessive) and centrifuged at 1500 g at 4 °C for 10 min (14). Then, samples from
*Kabi Pharmacia, Uppsala, Sweden.
251
252
K . D . K R I S T E N S E N et al. each joint were aliquoted into five Eppendorf tubes and frozen at 80 °C for later spectrophotometry and Luminex cytokine analyses. Also, washing solution from each subject was frozen for later spectrophotometry analysis. Spectrophotometry The absorbance at 350 nm was measured for each sample and for each washing solution (spectrophotometer: Shimadzo UV-160A†) with a capillary tube system consisting of a quartz capillary tube (07 mm in diameter, 3 lL/sample) and a capillary tube holder†. The absorbance of the aspirates and washing solutions were measured four times in the spectrophotometer, and the mean value was used. The dilution detection limit of the washing solution in the aspirate by this method was 09% (13). Cytokine quantification IL-1b, IL-2, IL-6, IL-10, TNF and IFN-c were analysed with a high-sensitivity human cytokine kit for simultaneous multianalyte detection with Luminex technology and instrumentation (Lincoplex kit‡; Bio-Plex 200 System apparatus§). The minimum detection concentrations in the assays were 006 (IL-1b), 016 (IL-2), 010 (IL-6), 015 (IL-10), 005 (TNF) and 029 (IFN-c) pg mL1, according to the manufacturer’s data. The intra-assay coefficients of variation were 311 (IL-1b), 427 (IL-2), 351 (IL-6), 331 (IL-10), 349 (TNF) and 488 (IFN-c). Undetectable levels were set to ‘0’ in the statistical analysis. The synovial fluid concentration of each cytokine was calculated using the formula: CSF ¼
volume >05 mL, no or only minor blood contamination and no haemolysis) were included in the analysis (14). Statistics Nonparametric statistical analysis was used. The median, 25th and 75th percentiles and number of observations were reported for descriptive statistics of the variables. For analytical statistics, the Spearman’s rank correlation test was used to calculate the significance of the correlation between separate variables, and the Wilcoxon’s rank-sum test was used for intra-group differences. A P < 005 was considered significant.
Results A total of 14 samples from 40 TMJs fulfilled the sample quality criteria (14). These 14 samples were obtained from 10 subjects. The reasons for not fulfilling the sample criteria were blood contamination (n = 13), a dilution factor >098 (n = 9) or too low recovery of synovial fluid (n = 4). The distribution of the recovered volumes of synovial fluid is shown in Fig. 1. Significantly more synovial fluid was recovered from female subjects compared to males (females, median: 172 lL, IQR: 44–277, n = 7 and males, median: 56 lL, IQR: 37–85, n = 7; P = 0048). As shown in Table 1, TNF was detected in all joints with
CASP AbsAsp 1 AbsWash
where CSF = synovial fluid concentration of a given cytokine, CAsp = aspirate concentration of the given cytokine measured by the Luminex assay, AbsAsp = aspirate absorbance and AbsWash = washing solution absorbance. Samples that fulfilled the sample quality criteria (dilution factor 0’: percentage of samples with detectable cytokine levels (n = 14)
IL-6 IL-10 TNF IL-1b IL-2 IFN-c
Median (pg mL1)
IQR (pg mL1)
Range (pg mL1)
>0 (%)
0 0 23 0 18 10
0 0 13–37 0 0–22 0–47
0–101 0–263 32–98 0–56 0–455 0–178
71 71 100 143 714 50
a median value of 23 pg mL1, but ranged from 32 to 98. IL-2 was found in 71% of joints with a median value of 18 pg mL1, ranging from 0 to 455. IFN-c was found in half of the joints with a median value of 10 pg mL1, ranging from 0 to 178. The median IL1b, IL-6 and IL-10 levels were zero. IL-1b was only detected in 2 and IL-6 and IL-10 in 1 of the 14 joints. Regarding the two joints with useful samples with non-symptomatic clicking, the values were as follows: joint 1: TNF: 358 pg mL1, IL-1b: 382 pg mL1, IL2: 179 pg mL1, IFN-c: 469 pg mL1 and joint 2: TNF: 124 pg mL1, IL-2: 09 pg mL1. There was a significant correlation between the TNF and IFN-c concentrations in TMJ synovial fluid (rs = 078, n = 14, P < 0001) (Fig. 2), but not between any of the other cytokines.
Fig 2. Scatter plot showing the relation between the IFN-c and TNF concentrations in the TMJ synovial fluid of 14 healthy individuals (rs = 078, n = 14, P < 0001). © 2014 John Wiley & Sons Ltd
Discussion The results of the present study indicate that TNF is consistently detected in the TMJ synovial fluid of healthy individuals and that the cytokines IFN-c and IL-2 are sporadically detected using the hydroxocobalamin (B12) method and a multiplex assay. The levels should be considered as homeostatic. The cytokines IL-10, IL-1b and IL-6 were not frequently detectable in the healthy individuals, suggesting those cytokines are more specific for inflammatory environments. The bilateral samples obtained from four patients were not systematically different from the unilateral samples. The individual bilateral levels differed and showed a similar variation in the same individual. Homeostatic cytokine levels thus seem to vary intra-individually. Alstergren et al. previously reported a hydroxocobalamin dilution method as an accurate and reproducible technique to determine the true synovial fluid concentration of a mediator after a joint washing, even in low sample volumes (15). In this study, synovial samples from the healthy TMJ were obtained, and the true synovial fluid concentration of cytokines was determined by the hydroxocobalamin dilution sampling method. This is one of the first reports using this methodology in healthy individuals, and our results could provide a basis for a standard reference for future TMJ studies on, for example, osteoarthritis, JIA or RA patients. In healthy individuals, a lower proportion of samples fulfilling the sample quality criteria may be expected compared to sampling from patients with TMJ arthritides due to the small volume of synovial fluid present in the non-inflamed joint. Synovial fluid in the healthy TMJ is a thin viscous gel-film lubricating the joint surfaces, which makes sampling difficult. The sample quality criteria applied in the present study and as determined by Alstergren et al. (14) set a high standard for acceptable sample quality, as no contamination was tolerated to detect the expected low levels of cytokines. Furthermore, the amount of recovered synovial fluid in healthy individuals seems to vary, for example as indicated by the difference between males and female; thus, not all healthy individuals are suitable for TMJ synovial fluid sampling. Therefore, we observed that 35% of the samples from healthy donors fulfilled our sample quality criteria, which was expected but considerably less than the 52–83% of samples found applicable for cytokine
253
254
K . D . K R I S T E N S E N et al. studies in patients with various TMDs (14, 16–18). Still, a reasonable proportion of samples from healthy individuals fulfilled the criteria, and the true synovial fluid concentrations were determined from them. The method can be used in healthy subjects, but non-useful samples should be expected. However, even though the method is considered safe and without large risks, it is still an intra-articular injection and brief post-operative symptoms can be expected. Also, more serious side effects as infection could potentially occur, so keeping the number of participants as low as possible with the aim of the study in mind is advised. The reason for choosing the hydroxocobalamin as an external standard was that it enables measurement of the total volume of synovial fluid recovered, and therefore, true synovial fluid cytokine concentrations can be calculated. There are a number of other TMJ synovial fluid sampling methods that have been used in previous studies of cytokine content in the joint aspirate with either no compensation for the joint washing (6, 7, 19) or the ratio between the measured cytokine concentration and the total protein content of the aspirate (5, 20). The method with hydroxocobalamin as an external marker enables the quantification of the total amount of synovial fluid and subsequent measurement of the dilution factor by spectrophotometry of the joint aspirate. It thereby enables determination of the true synovial fluid content in the joint with a detection limit of
