SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

The anatomic basis for joint disease localization in polymyalgia rheumatica (PMR) is poorly understood. This study used contrast-enhanced and fat suppression magnetic resonance imaging (MRI) to evaluate the relationship between synovial and extracapsular inflammation in PMR and early rheumatoid arthritis (RA).

Methods

Ten patients with new-onset PMR and 10 patients with early RA underwent dynamic contrast-enhanced MRI and conventional MRI of affected metacarpophalangeal (MCP) joints. Synovitis and tenosynovitis were calculated based on the number of enhancing voxels, initial rate of enhancement, and maximal enhancement of gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA). Periarticular bone erosion and bone edema were scored according to the OMERACT (Outcome Measures in Rheumatology Clinical Trials) scoring system in both groups. The degree of extracapsular Gd-DTPA enhancement was assessed in both conditions using semiquantitative scoring.

Results

No significant differences were seen in the volume of synovitis (P = 0.294), degree of flexor tenosynovitis (P = 0.532), periarticular erosions (P = 0.579), or degree of bone edema (P = 0.143) between RA and PMR joints. However, despite comparable degrees of synovitis, the proportion of MCP joints showing extracapsular enhancement was higher in the PMR group (100%) than in the RA group (50%) (P = 0.030). One PMR patient, but none of the RA patients, had bone edema at the capsular insertion.

Conclusion

Despite degrees of synovitis and tenosynovitis comparable with those in RA, PMR-related hand disease is associated with prominent extracapsular changes, suggesting that inflammation in these tissues is more prominent than joint synovitis, which is common in both conditions. This suggests that the anatomic basis for joint disease localization differs between RA and PMR.

Rheumatoid arthritis (RA) is a chronic inflammatory polyarthritis characterized by prominent synovial joint inflammation with progressive joint destruction. Polymyalgia rheumatica (PMR) is also characterized by synovial joint involvement, but compared with RA it is associated with prominent muscle stiffness, absence of joint destruction, and a good prognosis (1). The use of magnetic resonance imaging (MRI) in RA has confirmed the primacy of synovitis and the relationship between synovitis and erosions (2, 3). In PMR, however, MRI has not led to a consensus regarding an anatomic basis for its rheumatic manifestations, with some studies suggesting that PMR-associated joint disease represents a synovial-based disease that can be distinguished from RA on the basis of either shoulder joint bursitis (4) or hand tenosynovitis (5). Using fat-suppressed MRI techniques to look at shoulder and hand involvement in PMR, we have previously noted that edema in the extracapsular tissues was more common in PMR than in RA (6). One possible explanation for such differences might be a relatively abrupt-onset synovitis in PMR with nonspecific extension of the inflammatory processes into the adjacent soft tissues; alternatively, the primary site of inflammation in PMR might be capsular based (6).

Contrast agent gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) in dynamic contrast-enhanced MRI (DCE-MRI) allows not only the estimation of the volume of synovitis, but also the degree and kinetics of enhancement (7–9), including the initial rate of enhancement (IRE) and maximal enhancement (ME), both of which reflect synovial tissue vascularity and permeability. In addition, using contrast enhancement with fat suppression allows an assessment of extracapsular tissue edema, which may reflect an inflammatory process at that site. This study investigated the relationship between synovial and extracapsular inflammation in hand disease–related PMR and early RA.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Twenty patients were recruited from the early arthritis clinic. The study had approval from the local ethics committee, and all patients provided their informed consent. Ten consecutive patients fulfilling the 1987 revised criteria for RA of the American College of Rheumatology (formerly, the American Rheumatism Association) (10) (5 men and 5 women, mean age 55 years [range 28–72 years]) with clinical metacarpophalangeal (MCP) joint synovitis and 10 patients diagnosed as having PMR (5 men and 5 women, mean age 74 years [range 68–83 years]) fulfilling the criteria described by Bird et al (11) with recent onset of MCP joint swelling were eligible for study entry. The mean duration of symptoms was 4.9 months for the RA patients (range 2–11 months) and 10 weeks for the PMR patients (range 3–24 weeks). All of the RA patients but none of the PMR patients were seropositive for rheumatoid factor. One of the PMR patients had received oral corticosteroids but remained symptomatic, whereas all RA patients were treatment naive. Both groups had polyarticular joint disease (mean swollen joint count 13 in the RA group and 7.4 in the PMR group). Only 2 of the PMR patients had pitting edema in the dorsum of the hand. The mean C-reactive protein level was 56 mg/liter (range 8–134) in the RA group and 46 mg/liter (range <5–115) in the PMR group. Exclusion criteria included any contraindication to undergoing MRI. The small joints of the hands were used rather than the shoulder, since the imaging resolution obtained in the hand is comparatively greater, and confounding factors such as degenerative supraspinatus tendinitis, which is very common in the shoulder, were avoided.

MRI.

MRI of the clinically most swollen hand was performed using a Philips 1.5T Gyroscan ACS-NT whole-body scanner (Philips Medical Systems, Best, The Netherlands). Patients were placed in the prone position with the arm extended above the head, and a circular surface coil was placed on the dorsum of the hand over the second-through-fifth MCP joints. A total of 80 joints were studied (40 in each group). The sequences performed were as follows: T1-weighted spin-echo on coronal and axial planes; axial DCE-MRI and T1 spectral presaturation with inversion recovery (SPIR) fat-suppressed pulse sequences post–Gd-DTPA administration in the coronal and axial planes. All scoring was performed in a randomized manner with scorers blinded to the clinical information.

Imaging processing of synovitis and tenosynovitis

For the quantitative assessment of synovitis and tenosynovitis, DCE-MRI data were analyzed using commercial software (Analyze; Mayo Clinic, Rochester, MN) and software developed in house (7, 8) to calculate values of IRE and ME on a pixel-by-pixel basis as well as the number of enhancing voxels (estimation of synovial volume) (Figure 1), as previously described in detail (7). An assessor experienced in automated MRI scoring (LAR) carried out the data analysis. The intrarater reliability of this system was as follows: intraclass correlation coefficient (ICC) 0.994 (95% confidence interval [95% CI] 0.992–0.996) for the IRE calculation and ICC 0.994 (95% CI 0.992–0.996) for the ME calculation.

thumbnail image

Figure 1. Image representing 1 of 6 axial slices from a dynamic contrast-enhanced magnetic resonance imaging data set acquired in the right hand of a patient with rheumatoid arthritis, with superimposed color data showing values of maximal enhancement (ME) across the synovial space following gadolinium diethylenetriaminepentaacetic acid enhancement. The yellow pixels represent high ME values, while the red pixels represent lower values. The edge of synovitis was defined manually around the metacarpophalangeal (MCP) head in the case of joint synovitis and around the flexor tendons in the cases of tenosynovitis (arrows). There is synovitis around all MCP joints, although this is more marked on the second and third joints. There is also inflammation around tendons 2 and 3 (whitenumbers), as represented by the surrounding yellow coloration. The blue numbers represent the 8 segments that were divided automatically by the software within the joint synovitis.

Download figure to PowerPoint

Bone changes

Bone erosions and bone edema were identified according to the recommendations of the OMERACT (Outcome Measures in Rheumatology Clinical Trials) group (12). Images were scored by 1 experienced reader (PGC) using the semiquantitative methods described in the validated OMERACT RA-MRI scoring system (12). In addition, the location of bone edema with regard to the joint margin or at the capsule insertion was documented as previously reported (13). The intraobserver reliabilities for the presence or absence of bone erosions (94.30%, κ = 0.87) and bone edema (97.60%, κ = 0.60) were completely concordant. For the semiquantitative analysis, the ICCs were 0.98 and 0.65 for erosions and edema, respectively (14).

Pattern of extracapsular enhancement

Extracapsular soft tissue edema and enhancement were scored in a blinded manner on T1 SPIR post–Gd-DTPA coronal sequences by 1 experienced observer (DM) and defined as “synovial” where maximum in the joint cavity and tendon sheaths or as “extracapsular” where changes were apparent outside the synovial cavity adjacent to the joint capsule, as previously described (14). First, each MCP joint was scored as being predominantly synovial or extracapsular based on the presence of extrasynovial Gd-DTPA enhancement. Then, a global score was assigned to each patient on the basis of the overall enhancement (i.e., the overall pattern of all 4 joints). The intrareader reliabilities for synovial and extracapsular enhancement were κ = 0.847 and κ = 0.828, respectively.

Statistical analysis

Mixed between-within subjects analysis of variance (ANOVA) was performed to test for differences between the DCE-MRI parameters in both groups. Nonparametric tests were used when appropriate. All statistical analyses were carried out using SPSS 12.0 (SPSS, Chicago, IL).

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

After a mean followup of 6 years, 7 patients in the PMR group had clinical remission of their disease, were not receiving treatment, and were discharged from our service. Three PMR patients have died (2 of cerebrovascular disease and 1 of bronchial carcinoma), whereas all patients in the RA group have persistent disease requiring ongoing treatment.

Synovitis and tenosynovitis.

There was no significant difference in the total volume of MCP synovitis between the PMR and the RA groups (mean ± SEM voxel count 28,593 ± 3,411 and 24,351 ± 1,940, respectively; t = 1.08, P = 0.294) (mean ± SEM IRE [arbitrary units] 850 ± 136 and 754 ± 100, respectively; t = 0.57, P = 0.576) (mean ± SEM ME [arbitrary units] 58,508 ± 8,059 and 49,953 ± 4,940, respectively; t = 0.91, P = 0.377), although the absolute numbers of voxels were higher in the PMR group. The same pattern was observed in the tendon analysis, with no significant differences for the degree of flexor tenosynovitis between both groups (mean voxel count t = 0.64, P = 0.532; mean IRE t = 0.68, P = 0.505; mean ME t = 0.89, P = 0.383).

Periarticular bone erosion and bone edema.

MRI-determined bone erosions were seen in 80% of the RA patients (n = 8 patients, 28 erosions) and 80% of the PMR patients (n = 8 patients, 20 erosions). There was no significant difference in the number of erosions between the groups (total number of erosions per patient [z = −0.692, P = 0.529 by Mann-Whitney U test]), although a trend toward more erosions in the RA group was observed. Likewise, the size of the erosions was comparable between the groups (total erosion score [z = −0.577, P = 0.579 by Mann-Whitney U test]).

Bone edema was seen in 60% of the RA patients (n = 6 patients, 20 regions of bone edema) and 20% of the PMR patients (n = 2 patients, 7 regions of bone edema) (continuity-corrected χ2 = 1.88, P = 0.171). The degree of bone edema was comparable between the groups (z = −1.67, P = 0.143 by Mann-Whitney U test). With regard to location, bone edema was seen at the periarticular margin in all of the RA patients. However, bone edema related to the distal capsule insertion was seen in 1 of the PMR patients but in none of the RA patients (Figure 2E).

thumbnail image

Figure 2. AC, T1-weighted post–gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) sequences of the metacarpophalangeal (MCP) joints of 3 of the study patients with rheumatoid arthritis (RA). A, High signal within the second, third, and fourth MCP joints, corresponding to synovitis. Also shown is extracapsular enhancement on the radial aspect of the second MCP joint (asterisk). B and C, High signal within the joint cavity representative of synovitis (arrows). Arrowhead in C indicates a periarticular erosion. D–F, T1-weighted post–Gd-DTPA sequences showing the MCP joints of 3 patients with polymyalgia rheumatica (PMR). D, Soft tissue enhancement seen at extrasynovial sites (asterisks). E, Capsular lesion on the radial aspect of the proximal phalanx of the third MCP joint. The lesion is located away from the joint margin at the site of the capsule insertion (arrow), something not noted in the RA cohort. Arrowhead indicates extrasynovial enhancement typically seen in PMR joints. Asterisks indicate extrasynovial soft tissue enhancement. F, Periarticular erosion in the radial aspect of the second MCP joint at the same location as seen in RA (arrow). There is, however, prominent extracapsular Gd-DTPA enhancement typical of PMR (asterisks).

Download figure to PowerPoint

Pattern of extracapsular soft tissue enhancement.

The overall pattern of enhancement was extracapsular in 80% of the PMR patients (n = 8) but in only 20% of the RA patients (n = 2) (continuity-corrected χ2 = 5.00, P = 0.025). For the total joint score, the pattern was extracapsular in 31 of 40 PMR joints and in 10 of 40 RA joints (continuity-corrected χ2 = 20.01, P < 0.0001). At the individual MCP joint level, between-subjects two-way ANOVA indicated that although there was no significant difference in voxel count between the groups, joints with extracapsular enhancement had significantly higher voxel counts than those without extracapsular enhancement (P < 0.0001) in both groups. However, for matched degrees of synovitis, the proportion of MCP joints showing extracapsular enhancement was higher in the PMR group (100%) than in the RA group (50%) (continuity-corrected χ2 = 4.701, P = 0.030) (Figure 3).

thumbnail image

Figure 3. Proportion of MCP joints that exhibited extracapsular enhancement, according to the amount of synovitis present, measured by voxel count (arbitrary units). See Figure 2 for definitions.

Download figure to PowerPoint

Only 2 of the PMR patients had clinical evidence of pitting edema. These patients had an extracapsular pattern of joint enhancement on MRI, but so did the majority of patients without pitting edema.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

The purpose of this study was to investigate whether PMR-associated hand disease differed from RA with respect to synovial compartment inflammation or extracapsular inflammation. Despite nonsignificant differences in the degree of synovitis in both conditions, we noted that there was a much greater degree of Gd-DTPA enhancement in the extracapsular tissues in the PMR patients. Our findings are concordant with those of some studies (6), but not with those of others (4, 5). We believe that these apparent contradictions can be explained in part by the MRI sequences used, the resolution of the imaging, and the criteria for patient selection. The majority of the studies claiming that PMR was primarily bursal- and tenosynovial-based were performed using non–fat-suppressed MRI techniques, making it difficult to evaluate extracapsular changes. The second issue was that previous studies showing extracapsular edema were unable to quantify the degree of synovitis, so the relationship between intra- and extracapsular disease could not be accurately assessed (6). Furthermore, it is difficult to compare RA and PMR, since the latter may have an abrupt onset with severe disease. Therefore, it was possible that the extracapsular changes reflected merely nonspecific extension of the inflammatory process (13).

We addressed these issues by using fat suppression and DCE-MRI in early PMR and in a cohort of patients with severe RA-related MCP disease. A greater but nonsignificant magnitude of synovitis was noted in the present study in the PMR group. In addition, the pattern of enhancement of extracapsular edema was very diffuse and was evident in tissue well away from inflamed synovium. As previously reported, bone edema adjacent to the joint capsule was also noted, which supports the idea of capsular-based inflammation in early PMR (13); however, our samples are small, and larger numbers would need investigating to understand the real meaning of these findings.

RA usually has an insidious onset compared with PMR, which often has an abrupt onset; hence, more synovitis and tenosynovitis may be expected in PMR at clinical presentation. In this study we chose RA patients with comparable degrees of clinically evident swelling in their MCP joints so that a fair MRI comparison could be made between diseases. We were also interested in determining whether extracapsular enhancement correlated with the presence of pitting edema typical of hand involvement in PMR. However, this pattern was also seen in the majority of the remaining PMR patients without pitting edema. Since extracapsular enhancement is also seen in shoulder disease in PMR (6), we postulate that these changes outside the joint may be contributing to the severe joint stiffness seen in this condition.

A surprising finding was that MRI-determined erosion and bone edema were equally common in both groups, and indeed it has been recognized that PMR may on occasion exhibit radiographically detectable erosion (15). This likely reflects the fact that irrespective of the trigger, synovitis will lead to the same pattern of MRI-determined erosion and edema. The exquisite sensitivity of PMR to steroids is likely to lead to complete suppression of the inflammatory process and may prevent the progression of MRI lesions to radiographically evident disease. However, in RA, erosion formation is more likely to occur, since disease tends to become chronic and is harder to suppress (3). Long-term followup of our patients showed that all the RA patients continue to have persistent disease necessitating disease-modifying therapy, but the entire PMR cohort has had complete remission of their disease with steroids alone.

In conclusion, our results provide evidence that hand joint disease in early PMR shows amounts of synovitis, tenosynovitis, and MRI-determined bone damage comparable with those in early RA, while extracapsular inflammation is more common in PMR. Further imaging and histopathologic studies are needed to elucidate the meaning of these findings and to determine whether they represent an event independent of synovitis.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Dr. McGonagle 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. Marzo-Ortega, Tan, Conaghan, Pease, McGonagle.

Acquisition of data. Marzo-Ortega, Rhodes, O'Connor, Radjenovic, Pease, McGonagle.

Analysis and interpretation of data. Marzo-Ortega, Rhodes, Tan, Tanner, Conaghan, Hensor, O'Connor, Radjenovic, Emery, McGonagle.

Manuscript preparation. Marzo-Ortega, Tan, Conaghan, O'Connor, Radjenovic, Pease, Emery, McGonagle.

Statistical analysis. Hensor.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

We thank Sister Claire Brown for help with study coordination, and we also thank the staff at the MRI Unit at the Leeds General Infirmary.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES
  • 1
    Salvarani C, Cantini F, Boiardi L, Hunder GG. Polymyalgia rheumatica and giant cell arteritis. N Engl J Med 2002; 347: 26171.
  • 2
    McGonagle D, Conaghan PG, O'Connor P, Gibbon W, Green M, Wakefield R, et al. The relationship between synovitis and bone changes in early untreated rheumatoid arthritis: a controlled magnetic resonance imaging study. Arthritis Rheum 1999; 42: 170611.
  • 3
    Conaghan PG, O'Connor P, McGonagle D, Astin P, Wakefield RJ, Gibbon WW, et al. Elucidation of the relationship between synovitis and bone damage: a randomized magnetic resonance imaging study of individual joints in patients with early rheumatoid arthritis. Arthritis Rheum 2003; 48: 6471.
  • 4
    Salvarani C, Cantini F, Olivieri I, Barozzi L, Macchioni L, Niccoli L, et al. Proximal bursitis in active polymyalgia rheumatica. Ann Intern Med 1997; 127: 2731.
  • 5
    Cantini F, Salvarani C, Olivieri I, Barozzi L, Macchioni L, Niccoli L, et al. Remitting seronegative symmetrical synovitis with pitting oedema (RS3PE) syndrome: a prospective follow up and magnetic resonance imaging study. Ann Rheum Dis 1999; 58: 2306.
  • 6
    McGonagle D, Pease C, Marzo-Ortega H, O'Connor P, Gibbon W, Emery P. Comparison of extracapsular changes by magnetic resonance imaging in patients with rheumatoid arthritis and polymyalgia rheumatica. J Rheumatol 2001; 28: 183741.
  • 7
    Tan AL, Tanner SF, Conaghan PG, Radjenovic A, O'Connor P, Brown AK, et al. Role of metacarpophalangeal joint anatomic factors in the distribution of synovitis and bone erosion in early rheumatoid arthritis. Arthritis Rheum 2003; 48: 121422.
  • 8
    Radjenovic A, Conaghan PG, Ridgway JP, O'Connor PJ, Meaney JF, Gibbon WW, et al. A fully automated method for measuring treatment efficacy in rheumatoid arthritis using quantitative analysis of dynamic Gd-DTPA enhanced MRI [abstract]. Arthritis Rheum 1999; 42 Suppl 9: S81.
  • 9
    Reece RJ, Kraan MC, Radjenovic A, Veale DJ, O'Connor PJ, Ridgway JP, et al. Comparative assessment of leflunomide and methotrexate for the treatment of rheumatoid arthritis, by dynamic enhanced magnetic resonance imaging. Arthritis Rheum 2002; 46: 36672.
  • 10
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 11
    Bird HA, Esselinckx W, Dixon AS, Mowat AG, Wood PH. An evaluation of criteria for polymyalgia rheumatica. Ann Rheum Dis 1979; 38: 4349.
  • 12
    Ostergaard M, Peterfy C, Conaghan P, McQueen F, Bird P, Ejbjerg B, et al. OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging Studies: core set of MRI acquisitions, joint pathology definitions and the OMERACT RA-MRI scoring system [published erratum appears in J Rheumatol 2004;31:198]. J Rheumatol 2003; 30: 13856.
  • 13
    McGonagle D, Gibbon W, O'Connor P, Green M, Pease C, Ridgway J, et al. An anatomical explanation for good-prognosis rheumatoid arthritis. Lancet 1999; 353: 1234.
  • 14
    Brown AK, Conaghan PG, Quinn MA, Karim Z, Wakefield RJ, Hensor EM, et al. The reliability of magnetic resonance imaging in the assessment of low disease activity rheumatoid arthritis [abstract]. Rheumatology (Oxford) 2006; 45 Suppl 1: i101.
  • 15
    Paice EW, Wright FW, Hill AG. Sternoclavicular erosions in polymyalgia rheumatica. Ann Rheum Dis 1983; 42: 37983.