Differences in synovial tissue infiltrates between anti–cyclic citrullinated peptide–positive rheumatoid arthritis and anti–cyclic citrullinated peptide–negative rheumatoid arthritis




To compare synovial tissue infiltrates from patients with anti–cyclic citrullinated peptide (anti-CCP)–positive rheumatoid arthritis (RA) with those from patients with anti-CCP–negative RA.


Synovial tissue samples were obtained arthroscopically from the inflamed knee joints of 57 patients with RA (34 of whom were anti-CCP positive) and examined for several histologic features along with immunohistologic expression of cell markers. Joint damage was assessed using the Kellgren/Lawrence (K/L) scale (range 0–4) on standard anteroposterior radiographs. In 31 patients (18 of whom were anti-CCP positive), synovial tissue was available from an earlier time point, allowing analysis of temporal changes.


Synovial tissue from anti-CCP–positive patients was characterized by a higher mean number of infiltrating lymphocytes (61.6 versus 31.4/high-power field [hpf] [400×]; P = 0.01), less extensive fibrosis (mean score of 1.2 versus 2.0; P = 0.04), and a thinner synovial lining layer (mean score of 2.1 versus 3.3; P = 0.002) compared with synovial tissue from anti-CCP–negative patients. Anti-CCP–positive patients expressed more CD3, CD8, CD45RO, and CXCL12. More anti-CCP–positive patients had a K/L score >1 compared with anti-CCP–negative patients. The difference in the mean lymphocyte counts was already present a mean of 3.8 years before the index biopsy (76.7 lymphocytes/hpf and 26.7 lymphocytes/hpf in anti-CCP–positive patients and anti-CCP–negative patients, respectively; P = 0.008) and was independent of disease duration and K/L score.


Synovitis in patients with anti-CCP–positive RA differs from that in patients with anti-CCP– negative RA, notably with respect to infiltrating lymphocytes, and is associated with a higher rate of local joint destruction.

Rheumatoid arthritis (RA) is a chronic autoimmune disease affecting the joints. In recent years, the importance of antibodies against cyclic citrullinated peptide (anti-CCP) has been convincingly demonstrated. Citrullination of peptides takes place in many forms of inflammation, but antibodies against these peptides have been observed primarily in RA.

Anti-CCP antibodies have been demonstrated to be predictive of both the progression of undifferentiated arthritis toward RA (1) as well as the progression toward erosive disease in patients with RA (2). Anti-CCP antibodies have also been linked to the pathogenesis of RA, with its relationship to HLA–DR4 and class II major histocompatibility complex molecules being involved in the recognition of antigens and the activation of T cells (3). Because the specificity for RA of CCP autoantibodies is better than that of IgM–rheumatoid factor (RF), measurement of CCP autoantibodies has become an important diagnostic tool (4).

Although the clinical features of anti-CCP–positive and anti-CCP–negative patients with RA are the same at baseline (5), progression to erosive disease occurs predominantly in anti-CCP–positive patients. Therefore, we wanted to investigate whether synovial infiltrates from anti-CCP–positive patients with RA differ from those from anti-CCP–negative patients with RA. To this end, we collected synovial tissue samples from patients with RA by arthroscopy and analyzed the synovial infiltrates from anti-CCP–positive and anti-CCP–negative patients to determine whether the morphologic and cellular features differed. In a subgroup of patients for whom synovial tissue samples obtained at a previous time point were available, we also studied temporal changes and the effect of secondary osteoarthritis (OA).


We examined synovial tissue specimens obtained from all patients with an established diagnosis of RA according to American College of Rheumatology (formerly, the American Rheumatism Association) criteria (6) who underwent therapeutic arthroscopic lavage of an inflamed knee between 2002 and 2005. Patient sera were obtained at the time of the index arthroscopy and were tested for the presence of anti-CCP antibodies, using the anti-CCP2 antibody enzyme-linked immunosorbent assay (Immunoscan RA Mark 2; Euro-Diagnostica, Arnhem, The Netherlands), according to the manufacturer's instructions, with a cutoff value of 25 units.

Standard anteroposterior (AP) knee radiographs were obtained in all patients and were included in the analysis if they were obtained within 3 months before or after the synovial tissue sampling. Radiographs were scored for the severity of OA, using the Kellgren/Lawrence (K/L) scale for joint space narrowing, osteophytes, and sclerosis (7). This score was divided into 2 categories depending on the absence of OA (K/L grades 0 and 1) or the presence of OA (K/L grades 2–4). Before arthroscopy, a Disease Activity Score in 28 joints (DAS28) (8) and a knee score (9) were obtained by a trained research nurse. The knee score (range 0–7) encompassed knee tenderness (range 0–3, where 0 = no tenderness, 1 = tenderness when asked for, 2 = tenderness on pressure, and 3 = tenderness and wincing), knee swelling (range 0–3, where 0 = no swelling, 1 = little swelling, 2 = moderate swelling, and 3 = abundant swelling), and patient-assessed knee pain on a 100-point visual analog scale (VAS), where 0 = no pain and 100 = maximal pain.

Synovial tissue analysis.

Synovial tissue biopsy specimens were obtained from all patients in a standardized manner (10) and were analyzed histologically and by immunohistochemistry.

Histologic analysis.

Synovial specimens were embedded in paraffin, sectioned at 3 μm, and stained with hematoxylin and eosin. Coded sections were scored independently by 2 observers (MO and IB) for the following histologic features: synovial lining layer thickness (mean number of layers) from 6 randomly selected sites; mean number of infiltrating lymphocytes, plasma cells, and neutrophils from 3 randomly selected high-power fields (hpf [400×]); vascularity, expressed as the mean number of blood vessels with detectable endothelial cells, perivascular infiltrates, and ectopic germinal centers in 3 hpf (200×); and the presence of fibrosis and fibrin (scored on a scale of 0–4, where 0 = absent and 4 = abundantly present). Differences between the observers were resolved by mutual agreement.

Immunohistochemical analysis.

Immediately after harvesting, synovial tissue was collected en bloc in a mold, embedded in Tissue-Tek OCT (Miles, Elkhart, IN), and stored in liquid nitrogen (−180°C) until sectioned. Five-micrometer sections were cut on a cryostat (Leica, Rijswijk, The Netherlands), mounted on glass slides (Star Frost; Knittelgläser, Braunschweig, Germany), and stored at −70°C until immunohistochemical analysis was performed in a single session. Serial sections were stained with the following monoclonal antibodies: anti-CD3 (M0740), anti-CD4 (F0818), anti-CD8 (M0707), anti-CD19 (M0740), anti-CD68 (M0718), anti-CD138 (M7228) (all from Dako, Glostrup, Denmark), anti-CXCL12 (MAB350), anti-CXCR4 (MAB170), anti–tumor necrosis factor α (anti-TNFα; MAB610) (all from R&D Systems, Abingdon, UK), anti-CD38 (34768; Becton Dickinson, Mountain View, CA), rabbit anti-human interleukin-1 (IL-1) (LP712; Genzyme, Cambridge, MA), and anti–IL-18 (D043-3; MBL, Nagoya, Japan). All antibodies were mouse anti-human unless indicated otherwise.

The immunohistochemical staining procedure was performed as follows. Slides were warmed up to room temperature, fixed in acetone (99.5%; Merck, Darmstadt, Germany) for 10 minutes, and then air-dried for 20 minutes. After each step, the samples were washed with phosphate buffered saline (PBS; Apotheek LUMC, Leiden, The Netherlands), and all incubations were performed at room temperature. Endogenous peroxidase activity was blocked with 1% hydrogen peroxide (Merck) in PBS containing 0.1% sodium azide (Merck) for 20 minutes. The monoclonal antibodies were diluted in PBS with 1% bovine serum albumin (BSA; ICN Biomedicals, Aurora, OH) and incubated for 60 minutes. For control sections, PBS, matching isotype, and conjugate controls were applied.

Detection of the monoclonal antibodies was performed using affinity-purified horseradish peroxidase (HRP)–conjugated goat anti-mouse IgG (Dako), goat anti-mouse IgG2a (Dako), goat anti-rabbit–HRP (BioSource International, Camarillo, CA), and swine anti-goat–HRP (BioSource International). The biotinylated tyramide/streptavidin–HRP amplification system (NEN Life Science Products, Boston, MA) was used to enhance the HRP staining. The HRP-conjugated antibodies were diluted in PBS/BSA (1%), with 10% normal human serum (Bloedbank LUMC, Leiden, The Netherlands) as blocking serum, and incubated for 30 minutes. Next, the biotinyl tyramide was diluted in dilution buffer (NEN Life Science Products) and incubated for 30 minutes. HRP activity was detected using hydrogen peroxide as a substrate and aminoethylcarbazole (Sigma) as a dye. After washing with distilled water, the sections were counterstained with Mayer's hemalum (Merck) and mounted with Kaiser's glycerol gelatine (Merck).

Semiquantitative scoring of inflammation.

Stained sections were coded and then analyzed in a random manner. All areas of each biopsy section were scored by 2 independent observers (MO and NL), who were blinded to clinical data. At least 2 samples per patient per time point were semiquantitatively scored for all markers of inflammation on a scale of 1–4, where 1 represents the lowest level of expression and 4 represents the highest level of expression. The scoring system was calibrated for each marker separately, because some markers are more abundantly expressed than others. All differences between the observers were resolved by mutual agreement. Interobserver agreement reached 90% and differed by 1 point at most.

Previously obtained synovial biopsy specimens.

In order to correct for a confounding effect of secondary OA on lymphocyte infiltration and to evaluate the stability of synovial features, we searched our database for previous synovial biopsy specimens from our patients. Biopsy samples had to have been obtained from the same inflamed knee via arthroscopy. Again, for each synovial biopsy specimen, we matched an AP knee radiograph obtained within 3 months of the biopsy.

Statistical analysis.

Statistical analyses were performed using Student's t-test, Wilcoxon's signed rank test, nonparametric correlation analysis, and the chi-square test, where appropriate. All statistical analyses were done with SPSS version 11.5 software (SPSS, Chicago, IL). P values less than 0.05 were considered significant.


A total of 57 patients with RA were included in this study; 34 of these patients were anti-CCP positive, and 23 were anti-CCP negative. Characteristics of the patients are shown in Table 1. The anti-CCP–positive and anti-CCP–negative patient groups did not differ with respect to age and disease duration. More anti-CCP–positive patients than anti-CCP–negative patients were IgM-RF positive. All patients had been treated with disease-modifying antirheumatic drugs (DMARDs), including methotrexate and sulfasalazine. The use of biologic agents was comparable between both groups (8 and 9 patients in the anti-CCP–negative group and the anti-CCP–positive group, respectively, had been treated with TNFα-blocking agents [P = 0.38]). Anti-CCP–positive patients had a higher mean DAS28 and a higher VAS score for general well-being (P = 0.02 and P = 0.03, respectively). Peripheral blood leukocyte and peripheral blood lymphocyte counts were higher in anti-CCP–negative patients (P = 0.04 and P = 0.01, respectively).

Table 1. Characteristics of the anti-CCP–positive and anti-CCP–negative patients with RA*
CharacteristicAll patients (n = 57)Anti-CCP positive (n = 34)Anti-CCP negative (n = 23)P
  • *

    Except where indicated otherwise, values are the mean ± SD. Anti-CCP = anti–cyclic citrullinated peptide; RA = rheumatoid arthritis; RF = rheumatoid factor; DMARDs = disease-modifying antirheumatic drugs; VAS = visual analog scale; K/L = Kellgren/Lawrence.

  • P value refers to the combination of K/L score 0–1 and K/L score 2–4.

Anti-CCP titer552 ± 790912 ± 85420 ± 0.4<0.001
No. of men/no. of women19/387/2712/110.02
Age, years56.7 ± 14.257.5 ± 12.655.6 ± 16.60.91
Disease duration, years9.2 ± 7.59.8 ± 7.18.4 ± 8.30.33
Erosive disease, no. of patients4027130.07
RF positive, no. of patients3831 7<0.001
No. of previous DMARDs3.4 ± 2.43.5 ± 2.23.2 ± 2.60.42
Current oral steroid use, no. of patients 6 2 40.18
Current methotrexate use, no. of patients3018120.65
Disease Activity Score in 28 joints4.14 ± 1.074.48 ± 1.053.70 ± 0.950.02
Erythrocyte sedimentation rate, mm/hour32.4 ± 25.334.2 ± 25.930 ± 250.38
VAS score for general well-being (range 0–100)42 ± 2247 ± 2235 ± 200.03
K/L score in index knee (range 0–4)1.2 ± 1.11.3 ± 1.01.0 ± 1.30.19
K/L score in contralateral knee (range 0–4)0.8 ± 1.00.8 ± 0.90.8 ± 1.10.74
K/L score 0–1, no. of patients3819190.07
K/L score 2–4, no. of patients1713 4 
Peripheral blood leukocytes, 109/liter8.26 ± 3.37.24 ± 1.79.67 ± 4.40.04
Peripheral blood lymphocytes, 109/liter1.79 ± 0.631.55 ± 0.352.06 ± 0.80.01
No. of previous injections in index knee3.00 ± 3.582.00 ± 1.744.36 ± 4.850.18
No. of previous injections in contralateral knee1.59 ± 3.291.14 ± 1.692.18 ± 4.630.94

Anti-CCP–positive patients had a higher number of infiltrating lymphocytes (Figures 1 and 2) and stronger expression of CD3, CD8, CXCL12, and CD45RO by immunohistochemical analysis (Table 2). The difference between groups in the number of infiltrating lymphocytes remained demonstrable after correction for disease activity: with a DAS28 <4, the numbers of infiltrating lymphocytes were 37/hpf in anti-CCP–negative patients (n = 14) and 102/hpf in anti-CCP–positive patients (n = 8) (P = 0.004); with a DAS28 <3.5, the numbers of infiltrating lymphocytes were 37/hpf in anti-CCP–negative patients (n = 8) and 112/hpf in anti-CCP–positive patients (n = 6) (P = 0.003).

Figure 1.

Numbers of infiltrating lymphocytes and fibrosis scores in anti–cyclic citrullinated peptide (anti-CCP)–positive patients with rheumatoid arthritis (RA) and anti-CCP–negative patients with RA. A, Number of infiltrating lymphocytes at index arthroscopy (n = 57 patients). ∗ = P = 0.01. B, Number of infiltrating lymphocytes at previous arthroscopy (n = 31 patients). ∗ = P = 0.008. C, Extent of fibrosis (n = 57 patients). ∗ = P = 0.04. Data are presented as box plots, where the boxes represent the 25th to 75th percentiles, the lines within the boxes represent the median, and the lines outside the boxes represent the 10th and 90th percentiles.

Figure 2.

Microscopic features of synovial tissue from anti–cyclic citrullinated peptide (anti-CCP)–positive (A and C) and anti-CCP–negative (B and D) patients with rheumatoid arthritis, showing lymphocyte infiltration (A and B) and fibrosis (C and D). (Hematoxylin and eosin stained; original magnification × 400.)

Table 2. Differences in synovial markers and knee scores between anti-CCP–positive RA and anti-CCP–negative RA*
CharacteristicAll patients (n = 57)Anti-CCP positive (n = 34)Anti-CCP negative (n = 23)P
  • *

    Except where indicated otherwise, values are the mean ± SD. Anti-CCP = anti–cyclic citrullinated peptide. See Patients and Methods for a description of the scoring methods.

Synovial fluid aspirated, ml42.7 ± 78.027.0 ± 21.762.4 ± 113.40.38
Synovial fluid leukocytes, 109/mm310.1 ± 8.7410.6 ± 6.49.4 ± 11.90.24
Knee score (range 0–7)4.06 ± 1.34.01 ± 1.24.17 ± 1.40.68
Histologic markers    
 Synovial lining layer thickness2.65 ± 1.502.05 ± 1.103.3 ± 1.620.002
 Lymphocytes/high-power field48.5 ± 39.561.6 ± 44.231.4 ± 24.00.010
 Plasma cells8.6 ± 16.110.9 ± 19.95.8 ± 9.10.77
 Neutrophils7.35 ± 13.16.2 ± 13.28.8 ± 13.20.21
 Vascularity19.8 ± 8.921.0 ± 10.418.3 ± 6.70.27
 Perivascular infiltrates, no. of patients10640.98
 Germinal centers, no. of patients10820.09
 Fibrosis1.52 ± 1.291.17 ± 1.141.96 ± 1.360.04
 Fibrin1.71 ± 1.551.72 ± 1.601.70 ± 1.520.96
Immunohistologic markers    
 CD32.57 ± 0.962.90 ± 1.002.09 ± 0.660.02
 CD42.44 ± 1.102.53 ± 1.342.32 ± 0.640.34
 CD82.04 ± 1.132.47 ± 1.261.41 ± 0.440.008
 CD191.06 ± 1.191.38 ± 1.370.59 ± 0.660.21
 CD383.00 ± 0.823.09 ± 0.972.86 ± 0.550.20
 CD45RO2.94 ± 0.703.19 ± 0.702.59 ± 0.540.02
 CD682.85 ± 0.622.72 ± 0.583.05 ± 0.650.29
 CD1382.50 ± 1.112.69 ± 1.222.23 ± 0.910.23
 Interleukin-12.35 ± 0.762.31 ± 0.892.41 ± 0.540.79
 Interleukin-82.41 ± 0.622.44 ± 0.632.36 ± 0.640.98
 Interleukin-182.89 ± 0.743.00 ± 0.662.73 ± 0.850.48
 Tumor necrosis factor α2.92 ± 0.572.97 ± 0.582.85 ± 0.580.46
 CXCL122.74 ± 1.153.09 ± 0.992.23 ± 1.230.04
 Transforming growth factor β3.00 ± 0.763.00 ± 0.843.00 ± 0.670.94
 Fibrin2.37 ± 0.882.50 ± 0.882.18 ± 0.900.39

Anti-CCP–negative patients showed an increased extent of fibrosis (Figures 1 and 2) and a thicker synovial lining layer (Table 2 and Figure 3). There was no difference in vascularity between the 2 groups (P = 0.27). Perivascular infiltrates were present in 6 anti-CCP–positive patients and in 4 anti-CCP–negative patients (P = 0.98). More anti-CCP–positive patients had ectopic germinal centers than did anti-CCP–negative patients (8 versus 2; P = 0.09). For 55 of 57 patients, we were able to evaluate radiographs and synovial tissue at the same time point (±3 months). There was no difference in absolute K/L scores between the groups, but more anti-CCP–positive patients had a K/L score >1 (Table 1).

Figure 3.

Microscopic features of synovial tissue from anti–cyclic citrullinated peptide (anti-CCP)–positive (A and C) and anti-CCP–negative (B and D) patients with rheumatoid arthritis. A and B, Hematoxylin and eosin–stained sections, showing the small synovial lining layer (arrow in A, and area between the arrows in B). C and D, Immunohistochemically stained sections, showing infiltration of CD3+ cells. (Original magnification × 400 in A and B; × 250 in C and D.)

For 31 patients (18 of whom were anti-CCP positive), we were able to study synovial tissue obtained a mean ± SD of 3.78 ± 3.18 years before the index sample was obtained. There was no difference between the groups in the time interval between both biopsies (P = 0.32). The mean ± SD lymphocyte count at the time of the earlier arthroscopy was 76.7 ± 64.6 in anti-CCP–positive patients compared with 26.7 ± 24.9 in anti-CCP–negative patients (P = 0.008) (Figure 1). The K/L score at this time point did not differ between groups (1.05 versus 0.6 in 17 anti-CCP–positive patients and 10 anti-CCP–negative patients, respectively; P = 0.24). Five anti-CCP–positive patients compared with 1 anti-CCP–negative patient had a K/L score >1 (P = 0.07). In the subgroup of patients with a disease duration of <1 year at the time of the index arthroscopy, the mean numbers of infiltrating lymphocytes were 74.1/hpf in anti-CCP–positive patients (n = 5) and 11.8/hpf in anti-CCP–negative patients (n = 5) (P = 0.008). The K/L scores in these 2 groups of patients with a short disease duration were not different (P = 0.31).

Only 2 anti-CCP–negative patients had a synovial lymphocyte count >60/hpf, whereas 14 anti-CCP–positive patients had a count ≥60/hpf. Using a cutoff value for infiltrating lymphocytes of 60/hpf, the sensitivity for the presence of anti-CCP antibodies in all 57 patients was 87%, with a specificity of 55%. In a correlation analysis, we observed no relationship between the numbers of synovial lymphocytes and disease duration or K/L score in the study group as a whole. In the anti-CCP–positive subgroup, a correlation between disease duration and the K/L score was observed at both time points (Spearman's rho 0.499 and 0.635, respectively [P = 0.01 and P = 0.006, respectively]). In all patients, the histologic lymphocyte count correlated with the semiquantitative immunohistology scores for CD3 (ρ = 0.456), CD8 (ρ = 0.558), CD38 (ρ = 0.659), CD19 (ρ = 0.661), CD45RO (ρ = 0.478), and CXCL12 (ρ = 0.461) (P < 0.025 for all markers). For immunohistochemical markers, the closest correlation was found between CD3-positive and CD8-positive cells, with Spearman's rho values of 0.888 and 0.819 for all patients and for anti-CCP–positive patients, respectively (P < 0.001 for both).


The aim of the present study was to analyze and compare the synovial tissue characteristics in anti-CCP–positive and anti-CCP–negative patients with RA. We observed significant differences in synovial tissue infiltrates from patients with anti-CCP–positive versus anti-CCP–negative disease, particularly in the number of lymphocytes (CD3, CD8, CD45RO), as well as in the extent of fibrosis and synovial lining layer thickness. In those patients for whom synovial tissue samples were obtained at 2 different time points, this difference was already present a mean of 3.78 years before the index synovial biopsy, at a time when secondary OA was less advanced. Disease duration and the K/L score did not influence synovial lymphocyte infiltration. Of note, the difference in lymphocyte infiltration was already present in patients with a disease duration of <1 year, indicating that lymphocyte infiltration is associated with anti-CCP–positive disease and is not a consequence of secondary OA.

Although anti-CCP–positive patients had a higher mean DAS28 compared with anti-CCP–negative patients, the difference in lymphocyte infiltration remained demonstrable in patients with a lower DAS28. Other investigators have shown that the cellular infiltrate is indistinguishable between involved and uninvolved joints in RA (11, 12); therefore, a difference in local synovitis is unlikely to influence lymphocyte infiltration. Furthermore, because our patients underwent knee arthroscopy for therapeutic reasons and both the knee scores as well as the numbers of previous knee injections were similar, we believe that our patient groups are comparable in terms of (previous) episodes of knee synovitis.

Strikingly, the number of leukocytes (especially lymphocytes) in peripheral blood was decreased in anti-CCP–positive patients. Because the level of CXCL12, a chemokine involved in the attraction of leukocytes, was increased in these patients, we postulate a preferential homing of leukocytes in the inflamed joints of anti-CCP–positive patients with RA.

The lack of a difference in the number of CD4 cells between anti-CCP–positive and anti-CCP–negative patients with RA may be caused by macrophages that also express CD4, because we observed no difference in CD68 expression. CD8 cells have been shown to be involved in the formation of ectopic germinal centers in RA synovial tissue, which have a critical role in antibody formation (13). Ectopic germinal centers have also been shown to produce more inflammatory cytokines (14). In the present study, both germinal centers and CD8 cells were more frequently observed in anti-CCP–positive patients.

To our knowledge, this study is the first to demonstrate differences between anti-CCP–positive and anti-CCP–negative RA. Previous synovial tissue studies revealed no difference between IgM-RF–positive and IgM-RF–negative patients (15). Anti-CCP antibodies and IgM-RF have been shown to be equally predictive for the progression of erosive disease and are detectable years before clinical signs of RA become apparent (16). CCP antibodies can be directed against citrullinated fibrin, a protein commonly observed in RA synovial tissue and fluid. Although the exact epitope against which CCP antibodies are directed has not yet been identified, the production of fibrin in the synovial joint and subsequent citrullination attributable to inflammation may be partly responsible for perpetuation of inflammation by activating antigen-presenting cells and antibody formation. Our observation of accelerated joint destruction in patients with a lymphocytic infiltrate is consistent with an earlier study by Kraan et al (17), who showed an association between joint damage and synovial expression of T cells (CD3+), granzyme B–positive cytotoxic cells, and fibroblast-like synoviocytes.

A limitation of our study is the retrospective design, which hampers the correction for confounding effects such as DMARD use or differences between the patient groups. The observed difference in sex is not supported in the literature by more aggressive disease or increased occurrence of CCP antibodies in women. As to the lymphotoxic effect of steroids, we observed no differences between groups in the number of previous joint injections or oral steroid use.

Because bony erosions in the knee are uncommon in RA (only 3 patients, all of whom were anti-CCP positive, showed erosions in the knee) and may not be the best parameter to monitor joint destruction in the knee, we evaluated joint destruction by K/L scores. These scores are infrequently used in RA studies, because they were originally developed for the monitoring of OA. Nonetheless, cartilage degradation is a hallmark of joint destruction in RA, characterized by joint space narrowing on plain radiographs. More severe primary OA was described as the cause of more synovial inflammation (18), but our data showed no relationship between OA and the lymphocyte count.

RA can be considered a heterogeneous disease, with different synovial infiltration patterns observed in several ways (19, 20). The results of our study add to this knowledge and are consistent with the observation that distinct genetic and environmental risk factors underlie anti-CCP–positive and anti-CCP–negative disease (21, 22), indicating that separate pathogenic pathways are involved in anti-CCP–positive and anti-CCP–negative disease. However, our study also leaves unanswered the question of whether the expression of CCP antibodies is the cause or consequence of increasing numbers of infiltrating lymphocytes.

In conclusion, our study demonstrates marked differences in the results of synovial infiltration analysis between anti-CCP–positive and anti-CCP–negative RA, independent of disease duration and K/L score, reflecting differences in underlying pathophysiologic mechanisms.


Dr. van Oosterhout had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Van Oosterhout, van Laar.

Acquisition of data. Van Oosterhout, Levarht.

Analysis and interpretation of data. Van Oosterhout, Bajema, Levarht, Toes, Huizinga, van Laar.

Manuscript preparation. Van Oosterhout, Bajema, Toes, Huizinga, van Laar.

Statistical analysis. Van Oosterhout.