Combination drug strategy in recent-onset rheumatoid arthritis suppresses collagen I degradation and is associated with retardation of radiological progression.

OBJECTIVE: To assess whether serum C-terminal cross-linking telopeptide of type I collagen (ICTP), a marker of type I collagen degradation, has any additional value in the assessment of treatment effect in patients with early rheumatoid arthritis (RA). METHODS: A total of 182 patients were randomized to treatment either with three disease-modifying anti-rheumatic drugs (DMARDs) and low-dose prednisolone (COMBI) or with a single DMARD with or without low-dose prednisolone (SINGLE). We investigated the prognostic value of serum ICTP level for the progression of joint destruction in X-rays (Larsen's score) from baseline to 2 years. RESULTS: There was a statistically significant decrease in serum ICTP levels from baseline to 1 year. At 6 months, the serum ICTP level was lower in the COMBI patients compared to that of the SINGLE cases (p = 0.008, after adjustment for baseline ICTP). When grouping the patients according to serum ICTP tertiles at 6 months, there was a statistically significant trend for increasing median change in Larsen score from baseline to 2 years from lowest to highest ICTP tertile in the SINGLE patients [p = 0.008, after adjustment for 28-joint Disease Activity Score (DAS28) score and RF status at baseline], while in the COMBI, the change remained low in all ICTP tertiles. CONCLUSIONS: The COMBI strategy for recent-onset RA results in early suppression of type I collagen degradation, which is reflected in radiological joint damage at 2 years. Serum ICTP at 6 months may be useful for identifying those RA patients whose treatment should be intensified to prevent further joint damage.     

Application of markers of collagen metabolism in serum and synovial fluid for assessment of disease process in patients with rheumatoid arthritis.

OBJECTIVE--To assess the potential of markers of collagen metabolism to reflect disease processes in rheumatoid arthritis (RA). METHODS--Serum (S) and synovial fluid (SF) from 59 patients with RA, and a knee joint effusion and serum from 90 control subjects were studied with radioimmunoassays for the aminoterminal propeptides of type I and type III procollagens (PINP and PIIINP, respectively). The breakdown of type I collagen was quantified with a radioimmunoassay for the cross linked carboxyterminal telopeptide of type I collagen (ICTP). RESULTS--About 50% of the patients had increased S-ICTP and S-PIIINP values, whereas S-PINP was increased in only 20% of the patients. The mean SF:S ratios of these markers varied between 4 (for ICTP) and 340 (for PIIINP), indicating that markers of collagen metabolism are formed locally and then released into the circulation. SF-PINP and SF-PIIINP correlated with each other (rs = 0.86, p < 0.001) and with SF-ICTP (rs = 0.69, p < 0.001, and rs = 0.65, p < 0.001, respectively). SF-ICTP was clearly related to radiographic findings in the corresponding knee joint, patients with gross bone deformation having the greatest SF-ICTP concentrations. S-ICTP and S-PIIINP also correlated with conventional markers of disease activity, such as C reactive protein and joint swelling score. CONCLUSION--Markers of collagen metabolism both in serum and synovial fluid can be measured to provide an assessment of disease process in patients with RA. ICTP and PIIINP are the most informative.     

Prediction of disease progression in early rheumatoid arthritis by ICTP, RF and CRP. A comparative 3-year follow-up study

OBJECTIVE: To test the predictive value of the cross-linked carboxyterminal telopeptide of type I collagen (ICTP; a marker of type I collagen degradation), rheumatoid factor (RF) and C-reactive protein (CRP) for disease progression in patients with early rheumatoid arthritis (RA) METHOD: We tested the value of baseline values of RF, CRP and ICTP for the prediction of radiological joint progression over 3 yr in 63 consecutive patients with early RA who were treated with the 'saw-tooth strategy'. RESULTS: Age- and sex-adjusted risks as odds ratios (95% confidence intervals) of elevated serum ICTP, RF positivity and increased CRP for progressive joint disease (defined as an increase of > 20 in Larsen's index on radiographs of the hands and feet) were 3.9 (1.3, 11.9), 3.9 (1.0, 15.5) and 2.6 (0.9, 7.5), respectively. Better prediction was achieved when the tests were used in combination, and where there was both elevated ICTP and positive RF the odds ratio was 9.1 (2.5, 32.9). This test combination showed good sensitivity and specificity (71 and 77%, respectively), with a positive predictive value of 65% and a likelihood ratio of 3.1. CONCLUSION: This kind of risk profile, in which the tests used reflect different aspects of the disease process, may be useful in early disease assessment to find patients who will need the most active drug therapy.     

Detritus synovitis in chronic polyarthritis: a clinical and operation histologic evaluation

In rheumatoid arthritis (RA) joint inflammation is due to two processes: 1) the underlying inflammatory process (UIP) characterized by a lymphoplasmacellular infiltration of the synovial tissue, as well as pannus formation, and 2) the detritogenic synovitis (DS), a synovial response to articular wear products from cartilage and bone (detritus) that induces a preferentially fibrinous inflammation. In order to estimate the role of DS in the clinical presentation of such joints, 40 patients with RA undergoing knee-joint surgery on 48 occasions were evaluated for clinical parameters, radiological stage (Larsen), and histopathological characteristics of UIP and DS. The clinical parameters were comparable in knee joints with predominantly UIP or DS. However, DS was regularly seen in knees with advanced destruction according to Larsen's stages 4 to 5, while UIP occurres in joints even without radiological damage. In conclusion, it is assumed that the poor response of patients with advanced RA to so-called long-term drug therapy may be in part explained by the modifying influence of joint detritus on the underlying "rheumatoid" inflammatory process.     

Involvement of fibroblast growth factor-2 in joint destruction of rheumatoid arthritis patients.

OBJECTIVE: To investigate the effect of the synovial fluid from knee joints of rheumatoid arthritis (RA) patients with different severities of joint destruction on osteoclastogenesis and bone resorption. METHODS: Synovial fluid was harvested from the knee joints of 59 RA patients and 37 ostcoarthritis (OA) patients. RA patients with Larsen's knee grade 1-3 were classified as mild RA (n = 30) and those with grade 4 or 5 as severe RA (n = 29). Cytokine concentrations in synovial fluid were measured by ELISA. Osteoclastogenesis was measured by tartrate-resistant acid phosphatase (TRAP)-positive multinucleated cell (MNC) formation in a co-culture of mouse osteoblastic cells and bone marrow cells, and bone resorption by 45Ca release from pre-labelled cultured neonatal mouse calvariae. RESULTS: The synovial fluid of severe RA patients significantly stimulated TRAP-positive MNC formation and 45Ca release compared to those of mild RA and OA patients. Among the bone-resorptive cytokines fibroblast growth factor-2 (FGF-2), tumour necrosis factor alpha (TNF-alpha), interleukin-1alpha (IL-1alpha), IL-6 and soluble IL-6 receptor (sIL-6R), only FGF-2 concentration in the synovial fluid was positively correlated to Larsen's grade, and severe RA patients showed significantly higher FGF-2 concentrations than mild RA patients. Osteoclastogenesis in a co-culture system which was stimulated by the synovial fluid of severe RA patients was significantly inhibited by a neutralizing antibody against FGF-2 and this inhibition was stronger than antibodies against other cytokines. CONCLUSION: The increase in endogenous FGF-2 levels in the synovial fluid of RA patients may play a role in the joint destruction by inducing osteoclastogenesis.     

Leucocyte migration inhibition with collagen type I and collagen type III in rheumatoid arthritis and degenerative joint diseases

The leucocytes of 22 patients with classic rheumatoid arthritis were investigated with the direct leucocyte migration inhibition technique in agarose as described by Clausen. The leucocytes of 9 patients with degenerative joint diseases and of 9 healthy persons served as controls. Collagen type I and collagen type III were used as antigens. The leucocytes of 22 patients with rheumatoid arthritis showed in 15 cases a migration inhibition with collagen type III and in 10 cases with collagen type I. The inhibition with collagen type III was stronger and more frequent than with collagen type I. The frequent leucocyte migration inhibition with collagen type III in classic rheumatoid arthritis seems to be an expression of possible cell mediated reactivity against that type of collagen which appears especially in rheumatoid synovial tissue.     

Synovial fluid acidosis correlates with radiological joint destruction in rheumatoid arthritis knee joints

Metabolic variables in samples of synovial fluid (SF) from 33 rheumatoid arthritis (RA) knees were analyzed. These variables were correlated with radiological destruction and SF proteoglycan concentrations. SF acidosis correlated with radiological involvement (rs = 0.62, p less than 0.002), but not with proteoglycan concentrations. A weak correlation was found between SF acidosis and granulocyte concentrations (rs = 0.37, 0.02 less than p less than 0.05). Metabolic variables covaried as expected. Samples from right and left knees from the same patients correlated regarding cellular and metabolic variables. Our results indicate insufficient nutritional supply in RA joints with increasing radiological involvement.     

Increased type I collagen degradation correlates with disease severity in rheumatoid arthritis.

OBJECTIVES--To assess the extent and clinical significance of type I collagen degradation in rheumatoid arthritis (RA). METHODS--Serum samples from 90 consecutive patients with RA from a cross-sectional population based study and 90 age- and sex-matched controls were analysed with the new assay of cross-linked carboxyterminal telopeptide of type I collagen (ICTP). RESULTS--Patients with RA had significantly higher concentrations of ICTP than the controls. ICTP correlated strongly with measures of impairment in RA, such as the erosive state of joint disease (ES) (r = 0.57, p < 0.001) and Keitel function test (KFT) (r = 0.49, p < 0.001), and more weakly with various disease activity markers. When erythrocyte sedimentation rate (ESR), ES or KFT were used as indicators of disease severity among the patients with disease duration over five years, ICTP distinguished the more serious RA from milder cases. CONCLUSIONS--Elevated serum concentrations of ICTP are common in RA and are associated with signs of aggressive disease.     

Type I collagen degradation does not diminish with RA disease duration

OBJECTIVE: To assess the relation between type I collagen degradation and the duration of rheumatoid arthritis (RA). METHODS: The serum concentrations of cross linked carboxyterminal telopeptide of type I collagen (ICTP) measured earlier in a community based series (90 patients) and a hospital based series (59 patients) were re-evaluated with reference to the duration of RA. RESULTS: The serum ICTP showed a positive correlation with the duration of the disease in the hospital based series (r(s)=0.40, p<0.01) but not in the community based one (r(s)=0.18, p=0.10). CONCLUSIONS: Type I collagen degradation predominantly reflecting pathological bone destruction does not seem to diminish in longlasting RA.     

Levels of Rheumatoid Factor Isotypes, Metalloproteinase-3 and Tissue Inhibitor of Metalloproteinase-1 in Synovial Fluid from Various Arthritides

When synovial effusion is the only symptom, it is often difficult to make an exact diagnosis of the arthritic disease. To distinguish various types of arthritis with synovial effusion, we measured the levels of matrix metalloproteinase-3 (MMP-3, Stromelysin), tissue inhibitor of metalloproteinase-1 (TIMP-1) and rheumatoid factor (RF) isotypes in synovial fluid (SF) from patients with rheumatoid arthritis (RA), osteoarthritis (OA), pyogenic arthritis (PA), pseudogouty arthritis (PG), gouty arthritis (GA) and traumatic arthritis (TA). SF was aspirated from the knee joint or the ankle joint. Levels of IgG-, IgM- and IgA-RF isotypes were measured by ELISA. Levels of MMP-3 and TIMP-1 in SF were simultaneously determined by a one-step EIA system. Levels of IgG-RF, IgM-RF and MMP-3 in SF from RA patients were significantly higher than those in OA, PA, PG, GA and TA. However, IgA-RF in SF from RA patients, when compared with PA and GA, did not show a significantly increased level. In addition, TIMP-1 in SF from RA, when compared with PA and TA, also has not shown a significantly increased level. Therefore, in addition to analysing clinical data, measurements of IgG-RF, IgM-RF and MMP-3 in SF may contribute in distinguishing RA from other arthritic diseases. Received: 18 January 1999 / Accepted: 15 June 1999     

The clinical role of IL-23p19 in patients with rheumatoid arthritis

OBJECTIVE: To determine the clinical implications of the over-expression of synovial and circulating interleukin (IL)-23p19 and the correlation between IL-23p19 and other cytokines such as IL-17, tumour necrosis factor (TNF)alpha, and IL-1beta in rheumatoid arthritis (RA). METHODS: Synovial fluid (SF) and sera of 22 patients with RA were obtained during knee arthrocentesis and stored at -20 degrees C. Tender/swollen joint counts, 100-mm visual analogue scale (VAS), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), rheumatoid factor (RF), and antibodies to cyclic citrullinated peptide (anti-CCP Ab) were measured. Bony erosions were determined by X-rays. Serum and SF IL-23p19, IL-17, TNFalpha, and IL-1beta concentrations were measured by sandwich enzyme-linked immunosorbent assay (ELISA). RESULTS: The concentration of IL-23p19 correlated with the concentration of IL-17 in SF and sera, and with the concentrations of TNFalpha and IL-1beta in sera. SF IL-23p19 concentration was higher in patients who had bony erosions than those who had not. However, there was no correlation between IL-23p19 concentrations and other clinical parameters of RA. CONCLUSION: Upregulated IL-23p19 in SF might be involved in joint destruction in RA through interplay with other cytokines such as IL-17, TNFalpha, and IL-1beta.     

Detection of stromelysin in synovial fluid and serum from patients with rheumatoid arthritis and osteoarthritis

Summary Stromelysin levels were measured using a one-step sandwich immunoassay in synovial fluid (SF) obtained from 31 patients with rheumatoid arthritis (RA) (31 samples) and 13 patients with osteoarthritis (OA) (13 samples) and in serum from 81 patients with RA (106 samples), 12 with OA (14 samples), 12 with gouty arthritis (gout) (14 samples), and 8 with osteoporosis (OP) (14 samples) to identify differences in the levels in these diseases as well as correlations with clinical parameters in RA. SF stromelysin levels were significantly higher in RA than in OA, and rose with increasing joint destruction in the former. No significant correlations were found between the SF stromelysin level in RA and various clinical parameters, except for the volume of SF which showed a correlation. Serum levels of stromelysin were highest in RA, gout, OA, and osteoporosis in decreasing order, and in RA were correlated with the Steinbrocker Stage. A significant correlation was also found between the serum stromelysin level and number of swollen joints, and correlations with the Lansbury index, ESR, CRP, WBC and Plt. The stromelysin level in SF was thought to be a useful parameter of local joint involvement and that in serum of the severity of systemic joint inflammation.     

Knee and/or hip joint destruction in rheumatoid arthritis is associated with HLA-DRB1*0405 in Japanese patients

To determine the prognostic factors for knee and/or hip joint destruction in rheumatoid arthritis (RA) patients, we typed 379 RA patients for HLA-DRB1 alleles and analysed the antigen frequencies. The DRB1^*0405 antigen frequency in RA patients who underwent total knee replacement and/or total hip replacement was significantly higher than in those who did not have replacements, which meant that DRB1^*0405 was associated with knee and/or hip joint destruction. This finding may be of value for predicting knee and/or hip joint destruction in RA.     

Impaired glutathione reductase activity and levels of collagenase and elastase in synovial fluid in rheumatoid arthritis

OBJECTIVE: To test the activity of elastase, collagenase and glutathione reductase in the synovial fluid (SF) of patients with rheumatoid arthritis (RA) and in patients with osteoarthritis (OA); to correlate the elastase and collagenase activity with the glutathione reductase activity, which is important for the inactivation of oxygen free radicals. METHODS: 24 patients affected by osteoarthrosis and 24 patients affected by rheumatoid arthritis took part in the study. We measured elastase activity towards the substrate metoxysuccinyl-alanyl-alanyl-prolyl-valyl-p-nitroanilide (MeOSuc-ala-ala-proval-p-NA) which is highly specific for elastase, and insensitive to the other serine proteases, such as cathepsin G; collagenase activity was measured using [14C]-acetylated collagen as the substrate. Glutathione reductase activity was measured following the oxidation of nicotinamide adenine dinucleotide phosphate reduced (NADPH) in the presence of oxidized glutathione (GSSG). RESULTS: The concentrations of elastase, collagenase and glutathione reductase were statistically higher in patients with RA than in patients with OA. Moreover, in the SF of patients with RA we found positive correlation between enzyme activity levels. CONCLUSION: These results confirm a high activity of collagenase and elastase in the SF of patients with RA, which is about 30 times higher than that found in the SF of patients with OA. These data underline the synergic action of these enzymes in the pathogenesis of joint damage. RA patients also exhibit higher levels of glutathione reductase, which is important for the detoxification pathway of oxygen free radicals. However, compared with findings for collagenase and elastase, the increase in glutathione reductase is only three times higher than level found in the SF of OA patients. The limited increase in glutathione reductase activity during the inflammatory process might lead to an insufficient protective effect at the joint level in rheumatoid arthritis.     

Intraarticular release and accumulation of defensins and bactericidal/permeability-increasing protein in patients with rheumatoid arthritis

OBJECTIVE: Defensins and bactericidal/permeability-increasing protein (BPI) are the components of the azurophilic granules of polymorphonuclear cells (PMNC) maintaining antimicrobial protection. Both these substances have been suggested to interact with the host immune system rather than merely kill invading pathogens. We assessed concentrations of BPI and a-defensins in synovial fluid (SF) and matching blood samples of patients with rheumatoid arthritis (RA). METHODS: Matching samples of SF and blood were collected from 67 patients with RA (aged 21-73 yrs) with acute joint effusion. Blood samples from 22 healthy individuals made up a control group. Concentrations of BPI and human neutrophil peptides (HNP 1-3) were measured by ELISA. The results were related to radiological signs of destructive arthritis, duration of the disease, and laboratory markers of inflammation. RESULTS: BPI and HNP concentrations in SF were 10-60 times higher than in matching blood samples (p < 0.0001). Strong correlations between BPI and HNP concentrations were found in both blood and SF. In SF, BPI and HNP concentrations correlated to white blood cell (WBC) count (p < 0.001), and were associated with erosive joint disease (p < 0.05). In contrast, WBC count, serum C-reactive protein, or rheumatoid factor were not significantly correlated to the BPI or HNP concentrations. Serum BPI concentrations were moderately but significantly increased in RA patients compared in blood to controls (p < 0.05). CONCLUSION: BPI and HNP are accumulated in the synovial cavity of patients with RA. Significant correlation between joint erosion and local occurrence of BPI and HNP suggests participation of these molecules in regulation of the destructive course of RA.     

Quantitative analysis of pyridinium crosslinks of collagen in the synovial fluid of patients with rheumatoid arthritis using high-performance liquid chromatography

The pyridinium crosslinks are important and definite biomarkers of mature hard tissue collagen degradation. A gradient ion-paired reversed-phase high-performance liquid chromatographic method was used for the simultaneous determination of both crosslinks in synovial fluid (SF) samples of 25 patients with rheumatoid arthritis (RA). The mean±SD levels of pyridinoline (Pyd) and deoxypyridinoline (Dpyd) in SF were 107.7±182.3 nmol/1 and 4.8±8.3 nmol/1, respectively. The Pyd/Dpyd ratio, which indicates the amount of Pyd released from cartilage rather than bone, amounted to 30.8±29.5. This value is significantly higher than in urine or serum of the same patients. These data suggest increased destruction of joint cartilage in patients with RA and the release of collagen II fragments in SF. In addition, the levels of the crosslinks in SF reflect considerable interindividual variation, indicating substantial individual differences in the amount of collagenous material that is degraded.     

Plasma monocyte chemoattractant protein 1 is a marker for joint inflammation in rheumatoid arthritis

OBJECTIVE: Monocyte chemoattractant protein 1 (MCP-1) level in plasma is described as a marker for joint inflammation in rheumatoid arthritis (RA). METHODS: MCP-1 in plasma and synovial fluid (SF) was quantified by ELISA in 36 RA patients with synovitis of the knee at Day 1 and 30. Disease activity was assessed by the swollen joint count, Ritchie Articular Index (RAI), global assessment, pain on visual analog scale, Health Assessment Questionnaire, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP). RESULTS: By linear regression analysis plasma MCP-1 levels correlated significantly with the swollen joint count (Day 1: R = 0.47, p = 0.005; Day 30: R = 0.53, p < 0.001) and the RAI (Day 1: R = 0.37, p = 0.03; Day 30: R = 0.41, p = 0.01). The correlations of swollen joint count and RAI with ESR and CRP were significant only on Day 30 for the ESR (R = 0.40, p = 0.02). No association was found between plasma MCP-1 levels and the ESR/CRP levels. MCP-1 levels in plasma in RA patients were elevated compared to controls (p < 0.001) and MCP-I levels in SF were higher than in plasma (p < 0.001). No correlation was found between SF MCP-1 levels and in vitro migration of mononuclear cells towards SF. MCP-1 appears to participate in the disease process in RA, and plasma MCP-1 may be useful in monitoring joint inflammation.     

Elastolytic activity of human monocytes from synovial fluid and blood of patients with arthritis. Relations to levels of interleukin 6 and soluble interleukin 2 receptor

Synovial fluid (SF) and blood from 24 patients with non-traumatic, sterile hydarthron were examined for monocyte elastolysis (M?E) and for levels of interleukin 6 (IL-6) and of soluble interleukin 2 receptor (sIL-2R). Six patients had osteoarthrosis (OA) and 18 patients had inflammatory hydarthron (IH), 10 of whom had rheumatoid arthritis (RA). Blood M?E was lower in OA than in IH, both measured as basal M?E activity and after in vitro stimulation with immune complexes and phorbol myristate acetate (PMA). SF M?E was higher than M?E in blood (p less than 0.01). This increase in SF M?E could be mimicked in vitro by prestimulation of blood M? with low levels of IC. SF IL-6 and sIL-2R were also elevated (p less than 0.01). All three parameters correlated to the degree of joint inflammation evaluated by SF leucocyte level, complement activation, blood C Reactive Protein, and to the clinical evaluation of the joint. The increase in SF M?E, IL-6 and sIL-2R in patients with IH, points to a stimulation of M? and lymphocytes in the joint.     

Colony stimulating factor occurs in both inflammatory and noninflammatory synovial fluids

Synovial fluids (SF) from patients with osteoarthritis (OA) and rheumatoid arthritis (RA) and various other arthritides were examined for the presence of colony stimulating factors (CSF). CSF was found in 7 of 13 (54%) SF from OA patients and in 8 of 12 (67%) SF from RA patients. It was also found in SF from patients with other arthropathies including 5 of 5 samples from patients with septic arthritis. Inhibition studies employing monospecific antisera indicated that in both RA and OA, CSF was of the macrophage type (M-CSF). While CSF was found in both inflammatory and noninflammatory effusions, significantly greater numbers of colonies were stimulated by RA SF than by OA SF and in general greater numbers of colonies correlated with higher SF leukocyte counts. Our data suggest that CSF as well as other cytokines may be involved in the perpetuation of joint destruction that occurs in various rheumatological conditions.     

Synovial proliferation differentially affects hypoxia in the joint cavities of rheumatoid arthritis and osteoarthritis patients

This study was performed to investigate whether synovial proliferation (SP) differentially affects hypoxia in the joint cavities of rheumatoid arthritis (RA) and osteoarthritis (OA) patients. Thirty RA and 42 OA patients who underwent synovitis assessment were classified into two groups based on the presence or absence of SP, as revealed by musculoskeletal ultrasonography. Synovial fluids (SFs) from the knee joints were analyzed for interleukin (IL)-8, pO(2), and white blood cell counts and blood samples were analyzed for erythrocyte sedimentation rate (ESR). No difference was found between the OA patients with and without SP in terms of SF oxygen tension (SF pO(2)) or IL-8 level, whereas the RA patients had significantly lower SF pO(2) levels in their knee joints than did the OA patients with SP, and the RA patients had higher levels of IL-8 in their joints than did the OA patients. The counts of infiltrated immune cells in the SF and tissues were much higher for patients with RA and SP than for those with OA and SP. The ESRs were not found to be correlated with SP in OA patients but were negatively correlated with SF pO(2) levels in RA patients. We conclude that ultrasonographically detected SP in OA patients does not generate a more hypoxic SF than that found in OA patients without SP. The SFs from RA patients with SP are hypoxic, which indicates that SP may have different impacts on hypoxia in the joint cavities of RA and OA patients.