Abstract

Objectives . To investigate if disease assessment by contrast-enhanced dynamic and static magnetic resonance imaging (MRI) and quantitative nanocolloid (NC) scintigraphy gives useful additional information in early rheumatoid arthritis (RA).

Methods . Twenty-seven patients with early RA (disease duration ≤12 months) were followed up for 1 yr and 24 of them for 2 yrs with contrast-enhanced MRI and NC scintigraphy of the wrist joint. Synovial inflammation was assessed by measuring time-dependent enhancement rates (E-rate) from dynamic MRI scans and technetium 99m -labelled nanocolloid ( 99m Tc-NC) uptake from scintigraphy scans. Synovial membrane hypertrophy, bone oedema and erosions were semiquantitatively scored according to the Outcome Measures in Rheumatology Clinical Trials RA-MRI scoring system from static MR images. Response to the treatment was evaluated based on whether or not ≥50% improvement was achieved in the tender and swollen joint scores and the Health Assessment Questionnaire score, with normal C-reactive protein (CRP) or erythrocyte sedimentation rate (ESR) levels. Progression of the erosion score on wrist MRI was evaluated as the outcome.

Results . The baseline MRI bone oedema score (ρ= 0.67), MRI synovitis score (ρ= 0.57), ESR (ρ= 0.56), CRP (ρ= 0.48), E-rate (ρ= 0.47) and 99m Tc-NC uptake (ρ= 0.45) were related with the change in the MRI erosion score from baseline to 2 yrs (ρ= Spearman's correlation). In the multivariate logistic regression model, the bone marrow oedema score was the only baseline variable that predicted erosive progression at 2 yrs’ follow-up (OR 4.2, 95% CI 1.3–13.8). The median (interquartile range) change in the erosion score from baseline to 2 yrs was 0 (0, 0) and 4 (2, 5) in the patients with ( n = 9) and without ( n = 15) a persistent clinical response over the 2 yrs, respectively ( P = 0.001). The non-responders who presented with erosive progression from 1 yr to 2 yrs had higher MRI synovitis scores, bone oedema scores, E-rate and 99m Tc-NC uptake at 1-yr follow-up than the non-responders without progressive bone damage.

Conclusion . The degree of local synovial inflammation at baseline, evaluated by dynamic and static MRI and quantitative NC scintigraphy, is closely related to the progression of wrist joint erosions during the first 2 yrs of the disease. Furthermore, at follow-up, if no persistent clinical response is achieved, these imaging methods may help to predict future erosiveness and help in clinical therapeutic decision making.

Introduction

Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease, which involves the synovial joints and often results in irreversible destruction of articular cartilage and subchondral bone. However, RA is a highly heterogeneous disease with wide variation in disease activity and progression. Accurate assessment of disease activity and means for outcome prediction at presentation are needed in order to reduce or prevent joint damage with modern intensive therapy. In early disease, clinical examination, biochemical assessments and conventional radiography alone are not sensitive and specific enough to discriminate between self-limiting disease and persistent erosive arthritis. Frequent assessments of disease activity and response to therapy are crucial for long-term management of RA [ 1 ].

Magnetic resonance imaging (MRI) has been shown to be a highly sensitive technique for the detection of inflammatory soft tissue proliferation, bone oedema and early erosions, and since the implementation of MRI into the clinical practice, numerous cross-sectional papers concerning the MRI-detectable features of RA have been published [ 12 , 2–11 ]. However, few longitudinal studies using multimodality imaging methods, such as MRI and scintigraphy, for the detection of inflammatory changes in RA exist [ 12 , 13 ]. A 2-yr prospective observational study following a group of early RA patients with dynamic and static MRI and NC scintigraphy of the wrist joint was conducted to explore the value of these methods for disease monitoring and prognostication of erosive disease.

Methods

Patients

A series of 28 patients with early RA fulfilling the revised American College of Rheumatology criteria for RA [ 14 ] underwent contrast-enhanced MRI, NC scintigraphy, and laboratory and clinical examinations at baseline. Details of patient demographics have been presented previously [ 15 ]. In brief, the patient group at baseline consisted of 21 females and 7 males, with a median age of 51 yrs (range, 21–71). The median duration of symptoms was 5 months (range, 1–12). Eighteen (64%) of the patients were rheumatoid factor (RF) positive. Swollen joint count, tender joint count and Health Assessment Questionnaire score (HAQ) [ 16 ] were recorded. In addition, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level and IgM rheumatoid factor titre were measured. Medication, imaging parameters and clinical parameters are presented in Table 1 .

T able 1.

Data on medication, MRI and NC scintigraphy parameters of the wrist, and clinical and laboratory findings at baseline, and at 1 yr and 2 yrs of follow-up

  Baseline ( n = 28)   At 1 yr ( n = 27)   At 2 yrs ( n = 24)  Pa 
DMARD, no (%) b 24 (86) 26 (96) 22 (92)  
Single DMARD     
    Sulfasalazine, 1.5–3 g/day 13  
     Methotrexate, 5–15 mg/week –  
     Hydroxychloroquine, 300 mg/day  
    Intramuscular gold  
    Combination DMARD * 18 16  
    Predisolone 7–10 mg/day, no (%) b 15 (54) 9 (33) 7 (29)  
MRI data c     
    Erosion score 2 (0, 3) 2 (1, 6) 3 (2, 9) 0.001 
    Synovitis score 5.5 (2.5, 6.5) 3.5 (2, 6.5) 3 (2, 6) 0.148 
    Edema score 1.5 (0, 4) 0 (0, 3) 0.5 (0, 2) 0.203 
    E-rate max 1.12 (0.54, 1.59) 0.65 (0.37, 0.95) 0.61 (0.29, 1.25) 0.468 
    E-rate average 0.71 (0.36, 1.21) 0.44 (0.26, 0.73) 0.40 (0.19, 0.88) 0.503 
NC scintigraphy data c     
     99m Tc-NC uptake max 244 (163, 356) 173 (155, 264) 177 (147, 296) 0.377 
     99m Tc-NC uptake average 154 (122, 229) 131 (119, 172) 131 (116, 176) 0.446 
Clinical and laboratory data c     
    Swollen joint count 8 (4, 14) 2 (0, 6) 2 (0, 5) 0.001 
    Tender joint count 13 (4, 20) 2.5 (0, 6) 2 (1, 4) 0.001 
    HAQ 0.63 (0.28, 0.97) 0.38 (0, 0.63) 0.25 (0, 0.88) 0.035 
    CRP (mg/l) 15 (0, 28) 0 (0, 7) 0 (0, 6) 0.001 
    ESR 25 (14, 54) 8 (5, 15) 7 (2, 17) 0.001 
  Baseline ( n = 28)   At 1 yr ( n = 27)   At 2 yrs ( n = 24)  Pa 
DMARD, no (%) b 24 (86) 26 (96) 22 (92)  
Single DMARD     
    Sulfasalazine, 1.5–3 g/day 13  
     Methotrexate, 5–15 mg/week –  
     Hydroxychloroquine, 300 mg/day  
    Intramuscular gold  
    Combination DMARD * 18 16  
    Predisolone 7–10 mg/day, no (%) b 15 (54) 9 (33) 7 (29)  
MRI data c     
    Erosion score 2 (0, 3) 2 (1, 6) 3 (2, 9) 0.001 
    Synovitis score 5.5 (2.5, 6.5) 3.5 (2, 6.5) 3 (2, 6) 0.148 
    Edema score 1.5 (0, 4) 0 (0, 3) 0.5 (0, 2) 0.203 
    E-rate max 1.12 (0.54, 1.59) 0.65 (0.37, 0.95) 0.61 (0.29, 1.25) 0.468 
    E-rate average 0.71 (0.36, 1.21) 0.44 (0.26, 0.73) 0.40 (0.19, 0.88) 0.503 
NC scintigraphy data c     
     99m Tc-NC uptake max 244 (163, 356) 173 (155, 264) 177 (147, 296) 0.377 
     99m Tc-NC uptake average 154 (122, 229) 131 (119, 172) 131 (116, 176) 0.446 
Clinical and laboratory data c     
    Swollen joint count 8 (4, 14) 2 (0, 6) 2 (0, 5) 0.001 
    Tender joint count 13 (4, 20) 2.5 (0, 6) 2 (1, 4) 0.001 
    HAQ 0.63 (0.28, 0.97) 0.38 (0, 0.63) 0.25 (0, 0.88) 0.035 
    CRP (mg/l) 15 (0, 28) 0 (0, 7) 0 (0, 6) 0.001 
    ESR 25 (14, 54) 8 (5, 15) 7 (2, 17) 0.001 

a Significance is from the Friedman test. Data are b number of patients, with percentage in brackets and c median, with interquartile range in brackets.

*Methotrexate combined with one or more other DMARDs.

MRI, magnetic resonance imaging; NC, nanocolloid; DMARD, disease-modifying drugs; E-rate, enhancement rate; HAQ, Health Assessment Questionnaire; CRP, C-reactive protein; ESR, Erythrocyte sedimentation rate.

Altogether, 27 and 24 patients underwent the same examinations again after 1 yr and 2 yrs of follow-up, respectively. Response to the treatment at follow-up was defined as ≥50% improvement in the tender and swollen joint scores and HAQ, and normal CRP or ESR.

At 1 yr, one patient had dropped out because of a refusal to undergo MRI and NC scintigraphy examinations. At 2 yrs, additional three patients had dropped out for the following reasons: two patients could not be reached and one was pregnant. At 1-yr follow-up, all but one patient were taking one or more disease-modifying anti-rheumatic drugs (DMARDs). At 2 yrs, two patients were not on DMARD medication. Anti-tumour necrosis factor-alpha agents were not used by any of the patients.

All patients involved in this study gave their written informed consent to the protocol, which had been approved by the local ethics committee.

Imaging and assessment of nanocolloid scintigraphy

All patients received an intravenous injection of 555 MBq of 99m Tc-labelled nanocolloid (Nanocoll®, Nycomed Amersham Sorin, Saluggia, Italy), and anterior spot views of the hands were acquired 1 h later using an Elscint 409 ECT camera (Haifa, Israel) equipped with a low-energy general-purpose collimator. Matrix size was 128× 128 and acquisition time was 10 min. The quantification of 99m Tc-NC uptake was done as described in detail previously [ 15 ]. In brief, the images were transferred into a Hermes processing system (Nuclear Diagnostics, Hagerstad, Sweden) and filtered using a low-pass filter with a cut-off frequency of 0.32 cycles per centimetre [ 17 ]. The data were normalized by rescaling the images so that the mean counts/pixel in the normal radius was 100. The maximum and average uptake rates in the wrist were calculated ( 99m Tc-NC uptake max and 99m Tc-NC uptake average ). At the time of the initial baseline analysis, the quantitative data of eight study patients were lost because of a data storage failure, and we were thus able at that time to analyse the scans of only 20 patients. However, the lost baseline data were later restored, and we were able to conduct the quantification of all our patients. The results of these missing scans were now added to the baseline data. The quantification procedures were made by the same nuclear medicine specialist (R.T.) who was blinded to the clinical evaluation, laboratory results and MRI.

Magnetic resonance imaging

Imaging of the dominant or clinically more affected wrist was done using an open-configuration 0.23T MRI scanner (Philips Medical Systems MR-technologics Finland, Vantaa, Finland). The detailed method and the parameters of MR scanning have been previously described [ 15 ]. Static short-tau inversion recovery (STIR) coronal, T1 spin echo coronal, T2 fast spin-echo (FSE) axial and T1 3D gradient echo (GRE) coronal sequences were obtained. Dynamic gadolinium–diethylenetriaminepenta-acetic acid (Gd–DTPA)-enhanced (Magnevist 469mg/ml, Schering AG) scans were obtained in the coronal plane using a three-dimensional (3D) GRE technique. The dynamic acquisition sequence consisted of 12 continuous image slices and was obtained in 69 s. A pre-contrast scan was obtained, after which 15 ml of contrast was injected i.v. through a cannula into the opposite forearm. This injection was given over a period of 15–25 s with a subsequent flush of 10 ml of normal saline, followed by the first post-contrast sequence. Four post-contrast sequences were obtained, each of 69 s duration, with a delay of 1 s between them. After dynamic imaging, static post-contrast coronal T1w 3D GRE images were obtained. At the 1-yr and 2-yr follow-up examinations, the axial T2 FSE sequence was replaced by a post-contrast axial T1w 3D sequence because of its better delineation of synovial hypertrophy and bone erosions.

Assessment of dynamic MRI scans

Analysis of the dynamic data was performed using VIA 2.0 software (Philips Medical Systems MR-technologics Finland, Vantaa, Finland). Region of interest (ROI) circles (5–25 mm 2 ) were placed over the region of maximal synovial enhancement within three different areas of the wrist. The ROI measurements of the baseline, 1-yr and 2-yr MRI scans were performed by the same radiologist (K.P.), who was blinded to the NC scintigraphy measurements, clinical data and laboratory results. Three curves were obtained, plotting the mean pixel intensity of the ROI circle against the time following the gadolium injection. For quantitative characterization of these curves, the rate of enhancement per second after the first post-contrast sequence (69 s) was calculated as follows:  

formula
where SI 0 is the signal intensity before the contrast injection, and SI t is the signal intensity reached after completing the first post-contrast sequence (69 s). The highest E-rate value of each wrist was presented as the maximal E-rate (E-rate max ). Of the three curves, an average E-rate value was also calculated and presented as the average E-rate (E-rate average ).

Scoring of static MRI images

At baseline and at the 1-yr and 2-yr follow-up examinations, the scoring of synovial hypertrophy, bone oedema and bone erosions of the wrist joint were done independently by the same two observers (K.P. and J.V.) by reading the STIR images (bone oedema) and the static T1 3D GRE images obtained before and after contrast enhancement (synovitis and erosions) according to the Outcome Measures in Rheumatology Clinical Trials (OMERACT) group RA-MRI Scoring (RAMRIS) system [ 18 ]. Synovitis was scored on a 0–3 scale at three different locations: radioulnar joint, radiocarpal joint and intercarpal–carpometacarpal joints (total maximum score 9). A score of 0 is normal, with no enhancement or enhancement up to the thickness of normal synovium, while the scores from 1 to 3 (mild, moderate, severe) refer to increments of one-third of the presumed maximum volume of enhancing tissue in the synovial compartment. The carpal bones, distal radius, distal ulna and metacarpal base (15 locations) were scored separately for bone oedema (scored 0–3 based on the volume of oedema: 1: 1–33%; 2: 34–66%; 3: 67–100%) and bone erosions (scored 0–10, based on the proportion of eroded bone compared with “assessed bone volume”: 0: no erosion; 1: 1–10% of bone eroded; 2: 11–20%, etc.). The maximum score for bone oedema was 45 and that for bone erosions 150. The metacarpophalangeal joints were not evaluated, as they were not completely covered in the image sets.

Both readers were blinded to the clinical and laboratory parameters, and scored the 1-yr and 2-yrs MRI scans without reference to the baseline scans. After independent readings, however, an additional consensus reading was performed with reference to the baseline scans, to achieve maximum accuracy in scoring the bone erosions and oedema.

Statistical methods

The association between the baseline parameters and the change of erosion scores from baseline to 2 yrs were analysed using Spearman's rank correlation coefficients (ρ). The baseline variables that were significantly related to erosive progression, were then incorporated into a multivariate regression model (forward stepwise). Change in the erosion score of two or more was chosen as the cut-off value. Friedman's test was used to assess the change in the variables over the follow-up. Mann–Whitney's U test was used to explore the variable differences between the groups obtained based on erosive progression and the response to treatment. Intraclass correlation coefficients (ICC) were calculated to investigate the inter-observer reliability of scoring static MRI scans [ 19 ]. The level of P < 0.05 was considered statistically significant. SPSS 11.0 was used to conduct analyses.

Results

Disease evolution and response to the treatment

Medication, imaging parameters and clinical measures of inflammation at baseline, 1 yr and 2 yrs are presented in Table 1 . In the whole group, the erosion scores progressed during the follow-up. Swollen joint count, tender joint count, HAQ and laboratory parameters improved significantly from baseline to 1 yr.

A persistent response to the treatment was shown by nine patients out of 24 (38%) throughout the 2 yrs of follow-up. The median change in the erosion score was 0 [interquartile range (IQR) 0, 0] in these patients, compared with 4 (IQR 2, 5) in the patients with an inadequate response ( P = 0.001) ( Fig. 1 ). No significant differences in age, sex or medication were seen between these groups. In the group of responders, only one patient out of nine (11%) presented new erosions. In the group of non-responders, 13 patients out of 15 (87%) presented new/progressive erosions from baseline to 1-yr follow-up. From 1-yr to 2-yrs follow-up, nine non-responders out of 15 (60%) had continuing progression of bone damage, while six non-responders (40%) had stopped erosive progression. The patients who presented with erosive progression from 1 yr to 2 yrs had higher MRI synovitis scores, oedema scores, 99m Tc-NC uptake and E-rate at 1-yr follow-up than the patients without progressive bone damage ( Table 2 ).

F ig . 1.

A boxplot figure showing the erosion progression scores of patients with a persistent clinical response ( n = 9) and those with an inadequate clinical response ( n = 15) during the 2-yr follow-up.

F ig . 1.

A boxplot figure showing the erosion progression scores of patients with a persistent clinical response ( n = 9) and those with an inadequate clinical response ( n = 15) during the 2-yr follow-up.

T able 2.

One-year follow-up parameters of the clinical non-responders ( n = 15), divided based on the progression of erosions during the following year

 Progression of MRI erosion score from 1 to 2 yrs  
 
 
 
  No ( n = 6)   Yes ( n = 9)  Pa 
One-year parameters    
MRI data    
    Erosion score 2 (0.5, 6.5) 6 (3.5, 7) 0.272 
    Synovitis score 3.5 (1.75, 3.5) 6.5 (5.75, 8.5) 0.036 
    Edema score 0 (0, 1.5) 5 (1, 8.5) 0.026 
    E-rate max 0.59 (0.31, 0.88) 0.94 (0.78, 1.67) 0.050 
    E-rate average 0.39 (0.17, 0.47) 0.78 (0.70, 1.33) 0.001 
NC scintigraphy data    
     99m Tc-NC uptake max 158 (149, 202) 269 (210, 359) 0.012 
     99m Tc-NC uptake average 121 (110, 131) 191 (135, 235) 0.012 
Clinical and laboratory data    
    Swollen joint count 2 (1, 5) 5 (0.5, 7.5) 0.328 
    Tender joint count 5 (1, 17) 5 (0.5, 9) 0.529 
    HAQ 0.13 (0, 0.88) 0.5 (0.38, 0.63) 0.388 
    CRP (mg/l) 4 (0, 6) 5 (0, 28) 0.529 
    ESR 8 (2, 13) 8 (6, 18) 0.689 
 Progression of MRI erosion score from 1 to 2 yrs  
 
 
 
  No ( n = 6)   Yes ( n = 9)  Pa 
One-year parameters    
MRI data    
    Erosion score 2 (0.5, 6.5) 6 (3.5, 7) 0.272 
    Synovitis score 3.5 (1.75, 3.5) 6.5 (5.75, 8.5) 0.036 
    Edema score 0 (0, 1.5) 5 (1, 8.5) 0.026 
    E-rate max 0.59 (0.31, 0.88) 0.94 (0.78, 1.67) 0.050 
    E-rate average 0.39 (0.17, 0.47) 0.78 (0.70, 1.33) 0.001 
NC scintigraphy data    
     99m Tc-NC uptake max 158 (149, 202) 269 (210, 359) 0.012 
     99m Tc-NC uptake average 121 (110, 131) 191 (135, 235) 0.012 
Clinical and laboratory data    
    Swollen joint count 2 (1, 5) 5 (0.5, 7.5) 0.328 
    Tender joint count 5 (1, 17) 5 (0.5, 9) 0.529 
    HAQ 0.13 (0, 0.88) 0.5 (0.38, 0.63) 0.388 
    CRP (mg/l) 4 (0, 6) 5 (0, 28) 0.529 
    ESR 8 (2, 13) 8 (6, 18) 0.689 

Data are median, with interquartile range in brackets.

a Significance is from the Mann–Whitney test.

Detection of erosions

At baseline, MRI-detectable bone erosions were found in 21 patients (75%). At 1 yr and 2 yrs, erosions were found in 22 out of 27 patients (81%) and 20 out of 24 patients (83%), respectively. Four patients had no erosions in their baseline and follow-up scans. Of the patients with erosions at baseline, all but one still had erosions at 1-yr and 2-yr follow-ups. At baseline, 1 yr, and 2 yrs, MRI detected 64, 88 and 104 erosive bones (out of a total of 420, 405 and 360 wrist bones), respectively.

Association between the baseline variables and bone damage

Spearman's rank correlation tests were conducted to evaluate the correlation of the baseline variables with the change in the erosion score from baseline to 2 yrs. As shown in Table 3 , erosive development on MRI correlated with the baseline bone oedema score, synovitis score, ESR, CRP, E-rate average and 99m Tc-NC uptake average . Age, sex, medication (precence of DMARDs or prednisone at baseline), swollen or tender joint count, HAQ and RF at baseline did not correlate with the change in the bone erosion score from baseline to 2 yrs. When the baseline variables that were associated with the change in the bone erosion score at 2 yrs were incorporated into a multivariate logistic regression model, the bone marrow oedema score was the only baseline variable that predicted erosive progression at 2 yrs’ follow-up [odds ratio (OR) 4.2, 95% confidence interval (CI) 1.3–13.8].

T able 3.

Baseline parameters related to the change in MRI erosion score from baseline to 2 yrs

Baseline parameters ρ P 
Edema score 0.67 0.001 
Synovitis score 0.57 0.004 
E-rate average 0.47 0.023 
99m Tc-NC uptake average 0.45 0.028 
CRP (mg/l) 0.48 0.020 
ESR 0.56 0.004 
Baseline parameters ρ P 
Edema score 0.67 0.001 
Synovitis score 0.57 0.004 
E-rate average 0.47 0.023 
99m Tc-NC uptake average 0.45 0.028 
CRP (mg/l) 0.48 0.020 
ESR 0.56 0.004 

ρ, Spearman's correlation coefficients.

Site-specific association between bone oedema and erosions

Site-specific analysis of the 15 evaluated carpal bones was performed. The sites of all patients were combined, assuming that each finding of bone oedema is an independent occurrence. Only the sites with oedema but without erosions at baseline were chosen. Logistic regression analysis revealed significant associations between bone oedema at baseline and erosion at 1 and 2 yrs in MR scans (OR 28.0, 95% CI 11.7–67.1 and OR 14.9, 95% CI 6.3–34.9).

Validation of the MRI scoring system

The following ICCs of the separate baseline MRI readings were obtained: MRI synovitis score 0.87 (95% CI: 0.74, 0.94), bone oedema score 0.93 (0.85, 0.97) and erosion score 0.91 (0.82, 0.96). At the 1- and 2-yr follow-up examinations, the corresponding ICCs were: MRI synovitis 0.93 (0.85, 0.97) and 0.82 (0.63, 0.92), bone oedema 0.86 (0.71, 0.93) and 0.89 (0.76, 0.95), and erosions 0.71 (0.45, 0.86) and 0.84 (0.66, 0.93), respectively.

Discussion

The results of the present study indicate that the patients at high risk for erosive progression on wrist MRI have high local inflammatory activity at baseline, which can be reliably detected in contrast-enhanced dynamic and static MRI and quantitative NC scintigraphy ( Fig. 2 ). Furthermore, at follow-up, active erosive disease can be detected with these methods. Our results support the existing data on the importance of MRI in disease monitoring and the prognostication of erosive disease together with the contribution of quantitative NC scintigraphy, a method not previously documented in the follow-up of early RA.

F ig . 2.

Serial MRI and NC scintigraphy scans of a female patient born in 1948. (A) Baseline coronal T1w MR image scan of the right wrist shows a low-intensity signal in the trapezium and scaphoideum consistent with marrow oedema. (B) At 1 yr, a small erosion is visible at the distal end of the scaphoideum (white arrow). Bone oedema has disappeared. (C) At 2 yrs, the erosion is still present. (D) A baseline contrast-enhanced image shows massive synovial enhancement in the intercarpal compartment of the joint (arrows). (E) At 1 yr, marked reduction in synovial hypertrophy is noted, and some locally enhancing synovial hypertrophy is still present next to the trapezium bone (arrow). (F) At 2 yrs, more enhancing tissue is present compared with the 1-yr image (arrow). (G) Serial NC scintigraphy images from the same patient show intense uptake in the right wrist at baseline ( 99m Tc-NC uptake max/average = 416/239), weak uptake at 1 yrs ( 99m Tc-NC uptake max/average = 212/130) and moderate uptake at 2 yrs ( 99m Tc-NC uptake max/average = 269/141). Note that more intense uptake is detected in the left wrist compared with the right side at 2 yrs. (H) Dynamic MRI shows a steep curve at baseline, indicating intense and rapid enhancement, a flat curve (slow and weak enhancement) at 1 yr, and an intermediate curve at 2 yrs.

F ig . 2.

Serial MRI and NC scintigraphy scans of a female patient born in 1948. (A) Baseline coronal T1w MR image scan of the right wrist shows a low-intensity signal in the trapezium and scaphoideum consistent with marrow oedema. (B) At 1 yr, a small erosion is visible at the distal end of the scaphoideum (white arrow). Bone oedema has disappeared. (C) At 2 yrs, the erosion is still present. (D) A baseline contrast-enhanced image shows massive synovial enhancement in the intercarpal compartment of the joint (arrows). (E) At 1 yr, marked reduction in synovial hypertrophy is noted, and some locally enhancing synovial hypertrophy is still present next to the trapezium bone (arrow). (F) At 2 yrs, more enhancing tissue is present compared with the 1-yr image (arrow). (G) Serial NC scintigraphy images from the same patient show intense uptake in the right wrist at baseline ( 99m Tc-NC uptake max/average = 416/239), weak uptake at 1 yrs ( 99m Tc-NC uptake max/average = 212/130) and moderate uptake at 2 yrs ( 99m Tc-NC uptake max/average = 269/141). Note that more intense uptake is detected in the left wrist compared with the right side at 2 yrs. (H) Dynamic MRI shows a steep curve at baseline, indicating intense and rapid enhancement, a flat curve (slow and weak enhancement) at 1 yr, and an intermediate curve at 2 yrs.

Although conventional radiography has been considered as the golden standard for the assessment of joint damage in RA [ 20 , 21 ], MRI has been shown to have higher sensitivity in the monitoring of erosive progression [ 12 , 13 , 22 ]. In an established RA follow-up study, 78% of the new radiographic erosions could be visualized 2 yrs earlier by MRI than by conventional radiography [ 23 ]. Longitudinal studies have demonstrated the relationship between MRI-detectable inflammation, bone oedema and subsequent MRI-detectable bone damage [ 9 , 24 , 25 ]. Since MRI has been shown to be the most sensitive method for the detection of joint damage, erosive development on wrist MRI was chosen as the outcome measure of this study.

The most powerful baseline variable with a significant association with the change in the erosion score at 2 yrs was bone marrow oedema. Furthermore, in the multivariate regression model, it was the only baseline variable that predicted erosive progression. Bone marrow oedema (i.e. increased water content in trabecular bone) is, however, a non-specific, reversible finding, which is frequently seen in trauma, tumours, osteoarthritis and RA. In RA, bone oedema has been interpreted as a pre-erosive lesion, and it has been reported to associate with a six-fold risk for erosion occurrence on MRI at the same site 1 and 6 yrs later [ 26 , 27 ]. Our results of the site-specific analysis are similar, but show an even greater risk for erosive development if bone oedema is present at baseline.

The amount of hypertrofied synovial tissue, for which the synovitis score is a semi-quantitative estimate, was a weaker, but still significant, baseline variable that correlated with the development of erosions. In accordance with this result, previous follow-up studies have shown the baseline MRI synovitis or the MRI-determined synovial membrane volume to associate with erosive progression [ 10 , 24 ].

For the evaluation of synovitis, bone oedema and erosions, we used the semiquantitative scoring system developed by the OMERACT group. A previous multicentre reliability exercise showed acceptable levels of agreement between the assessments of change in MRI synovitis, bone erosions and bone oedema using this method [ 28 ]. In our study, the inter-observer reliability of scoring MRI changes at follow-up with the OMERACT RAMRIS scoring system was acceptable and similar to that in the baseline readings. At 1 yr, however, the reliability for scoring erosions was somewhat poorer, probably due to the increase of small erosive changes.

Increased synovial hyperaemia due to angiogenesis is an important feature of RA pathogenesis [ 29 ]. Bone damage has been shown to associate with the presence of angiogenetic factors (e.g. vascular endothelial growth factor) [ 30 ]. We used dynamic MRI for estimating local synovial hyperaemia. The technique involves obtaining images at short intervals after the injection of contrast agent [ 2 , 3 ]. E-rate indicating the speed and intensity of the diffusion of contrast agent in inflamed tissue can be calculated from these images. The early enhancement rate has been shown to relate to the number, size and permeability of synovial vessels as well as to the volume of the synovial membrane [ 3 , 8 ]. A few reports concerning dynamic MRI in the follow-up of early RA patients have been published. Huang et al . [ 31 ] reported the baseline E-rate to predict the development of erosions on static MRI scans 1 yr later. We found the baseline E-rate average to associate with the change in the erosion score at 2-yr follow-up. However, the baseline E-rate max , i.e. the signal intensity measured from the small area that presents the most intense Gd–DTPA enhancement, failed to show a significant association with the change in the MRI erosion score (Spearman's ρ= 0.32, P = 0.14). These findings imply that more accurate information about local inflammation is achieved by calculating the E-rate from several measurements of different regions of the wrist. The reproducibility of only one measurement may not be sufficient. Huang et al . [ 31 ] reported 10% variability in the initial rate of enhancement in blinded measurements. Our baseline E-rate measurements showed similar results with ICC of 0.92 (95% CI 0.84–0.96) [ 15 ].

To our knowledge, assessment of local disease activity in the follow-up of RA patients by quantitative NC scintigraphy has not been studied before. Intense 99m Tc-NC uptake seems to associate with erosive progression on MRI. Previous studies have shown this method to be sensitive in detecting patients with active joint disease caused by various different disorders [ 32–34 ]. The examination is cheaper than MRI, and the scanning of the whole body in the same setting is possible. Disadvantages include poor specificity, the lack of detailed visualization of exact anatomic structures, which makes the evaluation of erosions and bone oedema impossible, and the use of a 99m Tc-labelled compound, which exposes the patient to a small amount of radiation.

Baseline HAQ and swollen and tender joint counts did not correlate with the erosive progression on MRI. The absence of association between the swollen or tender joint count at baseline and erosive progression in a single joint has also been reported in other longitudinal MRI studies [ 10 , 24 ].

Nine patients out of 24 (38%) showed a persistent response to the treatment throughout the 2 yrs’ follow-up. Only one of them presented new erosions during the follow-up. Thus, persistent erosive progression seems unlikely in patients with a persistent clinical response. However, the significance of this finding is influenced by the small number of patients in this subgroup. A previous study with a larger number of patients showed that progression of erosions in hand radiographs can occur in some patients despite clinical remission [ 35 ]. The wide range of erosion progression scores among patients with an insufficient clinical response reflects the heterogeneity in the erosiveness of the disease, ranging from persistent non-erosive to persistent erosive disease. In these cases, the evaluation of local disease activity by quantitative imaging methods may be useful for predicting future erosiveness as shown in Table 2 . During follow up, bone marrow oedema with thick and intensively enhancing synovitis in MRI or intensively uptaking 99m Tc-NC in scintigraphy seems to imply a later erosive development.

The present series represents the clinical treatment practice from the late 1990s in a regional secondary clinic in a university hospital which serves a population of 270 000 inhabitants. Thus, our clinics admit patients with the whole disease spectrum from mild to severe RA. At baseline, one-third of the patients were treated with some DMARD combination, including methotrexate, which seems a rather conservative approach compared with the modern aggressive treatment strategies. The treatment was guided by clinical parameters and at the end of the 2-yr follow-up, two-thirds of the patients were on a combination. Our analysis showed that a clinically defined sustained treatment response is associated with non-progressive bone damage. This kind of information, however, is always obtained more or less retrospectively. According to our results, a more intensive treatment with, e.g. biological agents, should be considered if high inflammatory activity can be documented as a form of bone oedema and/or high 99m Tc-NC uptake in patients on DMARDs. However, these findings must be interpreted cautiously, because no attempt to standardize DMARD therapy was made in this observational study, which was not designed to study any specific treatment. Further case-controlled studies are needed to explore this topic. Nevertheless, for monitoring early RA, MRI and NC scintigraphy seem to produce sensitive information about the inflammatory activity of the disease, which is useful in clinical decision making.

In summary, the degree of local synovial inflammation at baseline, evaluated by dynamic and static MRI and quantitative NC scintigraphy, is associated with erosive progression on wrist MRI in early RA patients. Bone oedema is the baseline variable most closely associated with erosive development. Furthermore, during follow-up, these methods may help to find the patients at risk for erosive development if no persistent clinical response is achieved.

graphic

The authors have declared no conflicts of interest.

References

1
Lee
DM
Weinblatt
ME
Rheumatoid arthritis
Lancet
 , 
2001
, vol. 
358
 (pg. 
903
-
11
)
2
Reiser
MF
Bongartz
GP
Erlemann
R
, et al.  . 
Gadolinium-DTPA in rheumatoid arthritis and related diseases: first results with dynamic magnetic resonance imaging
Skeletal Radiol
 , 
1989
, vol. 
18
 (pg. 
591
-
97
)
3
Konig
H
Sieper
J
Wolf
KJ
Rheumatoid arthritis: evaluation of hypervascular and fibrous pannus with dynamic MR imaging enhanced with Gd-DTPA
Radiology
 , 
1990
, vol. 
176
 (pg. 
473
-
7
)
4
Jorgensen
C
Cyteval
C
Anaya
JM
Baron
MP
Lamarque
JL
Sany
J
Sensitivity of magnetic resonance imaging of the wrist in very early rheumatoid arthritis
Clin Exp Rheumatol
 , 
1993
, vol. 
11
 (pg. 
163
-
8
)
5
Rominger
MB
Bernreuter
WK
Kenney
PJ
Morgan
SL
Blackburn
WD
Alarcon
GS
MR imaging of the hands in early rheumatoid arthritis: preliminary results
Radiographics
 , 
1993
, vol. 
13
 (pg. 
37
-
46
)
6
Ostergaard
M
Gideon
P
Sorensen
K
, et al.  . 
Scoring of synovial membrane hypertrophy and bone erosions by MR imaging in clinically active and inactive rheumatoid arthritis of the wrist
Scand J Rheumatol
 , 
1995
, vol. 
24
 (pg. 
212
-
18
)
7
McQueen
FM
Stewart
N
Crabbe
J
, et al.  . 
Magnetic resonance imaging of the wrist in early rheumatoid arthritis reveals a high prevalence of erosions at four months after symptom onset
Ann Rheum Dis
 , 
1998
, vol. 
57
 (pg. 
350
-
6
)
8
Ostergaard
M
Stoltenberg
M
Lovgreen-Nielsen
P
Volck
B
Sonne-Holm
S
Lorenzen
I
Quantification of synovitis by MRI: correlation between dynamic and static gadolinium-enhanced magnetic resonance imaging and microscopic and macroscopic signs of synovial inflammation
Magn Reson Imaging
 , 
1998
, vol. 
16
 (pg. 
743
-
54
)
9
McGonagle
D
Conaghan
PG
O’Connor
P
, et al.  . 
The relationship between synovitis and bone changes in early untreated rheumatoid arthritis: a controlled magnetic resonance imaging study
Arthritis Rheum
 , 
1999
, vol. 
42
 (pg. 
1706
-
11
)
10
Ostergaard
M
Hansen
M
Stoltenberg
M
, et al.  . 
Magnetic resonance imaging-determined synovial membrane volume as a marker of disease activity and a predictor of progressive joint destruction in the wrists of patients with rheumatoid arthritis
Arthritis Rheum
 , 
1999
, vol. 
42
 (pg. 
918
-
29
)
11
Cimmino
MA
Innocenti
S
Livrone
F
Magnaguagno
F
Silvestri
E
Garlaschi
G
Dynamic gadolinium-enhanced magnetic resonance imaging of the wrist in patients with rheumatoid arthritis can discriminate active from inactive disease
Arthritis Rheum
 , 
2003
, vol. 
48
 (pg. 
1207
-
13
)
12
Klarlund
M
Ostergaard
M
Jensen
KE
Madsen
JL
Skjodt
H
Lorenzen
I
Magnetic resonance imaging, radiography, and scintigraphy of the finger joints: one year follow up of patients with early arthritis
Ann Rheum Dis
 , 
2000
, vol. 
59
 (pg. 
521
-
28
)
13
Backhaus
M
Burmester
GR
Sandrock
D
, et al.  . 
Prospective two year follow up study comparing novel and conventional imaging procedures in patients with arthritic finger joints
Ann Rheum Dis
 , 
2002
, vol. 
61
 (pg. 
895
-
904
)
14
Arnett
FC
Edworthy
SM
Bloch
DA
, et al.  . 
The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis
Arthritis Rheum
 , 
1988
, vol. 
31
 (pg. 
315
-
24
)
15
Palosaari
K
Vuotila
J
Takalo
R
, et al.  . 
Contrast-enhanced dynamic and static MRI correlates with quantitative 99Tcm-labelled nanocolloid scintigraphy. Study of early rheumatoid arthritis patients
Rheumatology
 , 
2004
, vol. 
43
 (pg. 
1364
-
73
)
16
Fries
JF
Spitz
P
Kraines
RG
Holman
HR
Measurement of patient outcome in arthritis
Arthritis Rheum
 , 
1980
, vol. 
23
 (pg. 
137
-
45
)
17
Starck
S-Å
Carlsson
S
Digital filtering of bone scans: an ROC study
Nucl Med Commun
 , 
1997
, vol. 
18
 (pg. 
98
-
104
)
18
Ostergaard
M
Peterfy
C
Conaghan
P
, et al.  . 
OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging Studies. Core set of MRI acquisitions, joint pathology definitions, and the OMERACT RA-MRI scoring system
J Rheumatol
 , 
2003
, vol. 
30
 (pg. 
1385
-
6
)
19
Shrout
PE
Fleiss
JL
Intraclass correlations: uses in assessing rater reliability
Psychol Bull
 , 
1979
, vol. 
86
 (pg. 
420
-
8
)
20
Brower
AC
Use of the radiograph to measure the course of rheumatoid arthritis. The gold standard versus fool
Arthritis Rheum
 , 
1990
, vol. 
33
 (pg. 
316
-
24
)
21
Watt
I
Basic differential diagnosis of arthritis
Eur Radiol
 , 
1997
, vol. 
7
 (pg. 
344
-
51
)
22
McQueen
FM
Stewart
N
Crabbe
J
, et al.  . 
Magnetic resonance imaging of the wrist in early rheumatoid arthritis reveals progression of erosions despite clinical improvement
Ann Rheum Dis
 , 
1999
, vol. 
58
 (pg. 
156
-
63
)
23
Ostergaard
M
Hansen
M
Stoltenberg
M
, et al.  . 
New radiographic bone erosions in the wrists of patients with rheumatoid arthritis are detectable with magnetic resonance imaging a median of two years earlier
Arthritis Rheum
 , 
2003
, vol. 
48
 (pg. 
2128
-
31
)
24
Conaghan
PG
O’Connor
P
McGonagle
D
, 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
, vol. 
48
 (pg. 
64
-
71
)
25
Quinn
MA
Conaghan
PG
O’Connor
PJ
, et al.  . 
Very early treatment with infliximab in addition to methotrexate in early, poor-prognosis rheumatoid arthritis reduces magnetic resonance imaging evidence of synovitis and damage, with sustained benefit after infliximab withdrawal: results from a twelve-month randomized, double-blind, placebo-controlled trial
Arthritis Rheum
 , 
2005
, vol. 
52
 (pg. 
27
-
35
)
26
McQueen
FM
Benton
N
Crabbe
J
Robinson
E
Yeoman
S
McLean
L
Stewart
N
What is the fate of erosions in early rheumatoid arthritis? Tracking individual lesions using X rays and magnetic resonance imaging over the first two years of disease
Ann Rheum Dis
 , 
2001
, vol. 
60
 (pg. 
859
-
68
)
27
McQueen
FM
Benton
N
Perry
D
, et al.  . 
Bone edema scored on magnetic resonance imaging scans of the dominant carpus at presentation predicts radiographic joint damage of the hands and feet six years later in patients with rheumatoid arthritis
Arthritis Rheum
 , 
2003
, vol. 
48
 (pg. 
1814
-
27
)
28
Conaghan
P
Lassere
M
Ostergaard
M
, et al.  . 
OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging Studies. Exercise 4: an international multicenter longitudinal study using the RA-MRI Score
J Rheumatol
 , 
2003
, vol. 
30
 (pg. 
1376
-
9
)
29
Walsh
DA
Angiogenesis and arthritis
Rheumatology
 , 
1999
, vol. 
38
 (pg. 
103
-
12
)
30
Ballara
S
Taylor
PC
Reusch
P
, et al.  . 
Raised serum vascular endothelial growth factor levels are associated with destructive change in inflammatory arthritis
Arthritis Rheum
 , 
2001
, vol. 
44
 (pg. 
2055
-
64
)
31
Huang
J
Stewart
N
Crabbe
J
, et al.  . 
A 1-year follow-up study of dynamic magnetic resonance imaging in early rheumatoid arthritis reveals synovitis to be increased in shared epitope-positive patients and predictive of erosions at 1 year
Rheumatology
 , 
2000
, vol. 
39
 (pg. 
407
-
16
)
32
Vorne
M
Lantto
T
Paakkinen
S
Salo
S
Soini
I
Clinical comparison of 99Tcm-HMPAO labelled leucocytes and 99Tcm-nanocolloid in the detection of inflammation
Acta Radiol
 , 
1989
, vol. 
30
 (pg. 
633
-
37
)
33
Liberatore
M
Clemente
M
Iurilli
AP
, et al.  . 
Scintigraphic evaluation of disease activity in rheumatoid arthritis: a comparison of technetium-99m human non-specific immunoglobulins, leucocytes and albumin nanocolloids
Eur J Nucl Med
 , 
1992
, vol. 
19
 (pg. 
853
-
7
)
34
Adams
BK
Al Attia
HM
Khadim
RA
Al Haider
ZY
99Tc(m) nanocolloid scintigraphy: a reliable way to detect active joint disease in patients with peripheral joint pain
Nucl Med Commun
 , 
2001
, vol. 
22
 (pg. 
315
-
8
)
35
Molenaar
ET
Voskuyl
AE
Dinant
HJ
Bezemer
PD
Boers
M
Dijkmans
BA
Progression of radiologic damage in patients with rheumatoid arthritis in clinical remission
Arth Rheum
 , 
2004
, vol. 
50
 (pg. 
36
-
42
)

Comments

0 Comments