To evaluate the sensitivity to change of power Doppler ultrasound (PDUS) assessment of joint inflammation and the predictive value of PDUS parameters in disease activity and radiologic outcome in patients with early rheumatoid arthritis (RA).
Forty-two patients with early RA who started therapy with disease-modifying antirheumatic drugs underwent blinded sequential clinical, laboratory, and ultrasound assessment at baseline, 3 months, 6 months, and 1 year and radiographic assessment at baseline and 1 year. For each patient, 28-joint Disease Activity Score (DAS28) was recorded at each visit. The presence of synovitis was investigated in 28 joints using gray-scale ultrasonography and intraarticular power Doppler signal. Active synovitis was defined as intraarticular synovitis detected with power Doppler signal. The ultrasound joint count for active synovitis and an overall joint index for power Doppler signal were calculated. Sensitivity to change of PDUS variables was assessed by estimating the smallest detectable difference (SDD) from the intraobserver variability.
The SDD for ultrasound joint count for active synovitis and ultrasound joint index for power Doppler signal was lower than mean changes from baseline to 3 months, 6 months, and 1 year. Time-integrated values of PDUS parameters demonstrated a highly significant correlation with DAS28 after 1 year (r = 0.63, P < 0.001) and a stronger correlation with radiographic progression (r = 0.59–0.66, P < 0.001) than clinical and laboratory parameters (r < 0.5).
PDUS is a sensitive and reliable method for longitudinal assessment of inflammatory activity in early RA. PDUS findings may have a predictive value in disease activity and radiographic outcome.
The accurate assessment of joint inflammation and sensitive monitoring of disease activity in patients with rheumatoid arthritis (RA) is essential in evaluating response to treatment and disease outcome (1). In early RA, synovitis appears to be the primary abnormality responsible for structural joint damage (2). In this case, the monitoring of therapy of patients with RA should focus on synovitis.
It is known that synovial inflammation consists of periarticular vasodilatation followed by synovial proliferation, which is accompanied by angiogenesis resulting in intraarticular blood vessel formation (3). Hypervascularization and angiogenesis of the synovial membrane are considered to be primary pathogenic mechanisms responsible for the invasive behavior of rheumatoid pannus (3–5). Therefore, there is a relationship between joint inflammatory activity and synovial vascularization (6).
Joint synovitis has traditionally been assessed indirectly by means of inflammatory subjective clinical data and laboratory parameters. Imaging techniques such as magnetic resonance imaging (MRI) and musculoskeletal ultrasound (US) are playing an increasingly important role in the evaluation and monitoring of patients with chronic inflammatory arthritis. Assessment of synovial inflammatory activity by MRI has shown a close correlation with histologic findings (7, 8). In addition, MRI findings have demonstrated a predictive value in structural joint damage in early RA (9, 10). However, MRI is expensive, time consuming, and not widely available for routine clinical use in many countries.
The greater resolution of superficial musculoskeletal structures offered by high-frequency transducers has promoted an increasing use of US in rheumatic diseases (11). US is a routinely available, noninvasive, and relatively inexpensive bedside imaging method with high patient acceptability. This technique is more sensitive and reproducible than clinical evaluation in assessing joint inflammation (12–19). The main advantage of US over MRI is that all peripheral joints can be examined as many times as required at the time of consultation, which improves the accuracy of clinical evaluation. In addition, prosthetic joints do not interfere with US images.
Both color Doppler and power Doppler US (PDUS) techniques detect synovial flow, which is a sign of increased synovial vascularization (20). The presence of intraarticular color Doppler/power Doppler signal aids in distinguishing active synovitis from inactive intraarticular thickening (21–29). Color Doppler and power Doppler findings have correlated with local clinical evaluation of joint inflammatory activity (13, 22, 23), overall clinical and biologic inflammatory activity (19), MRI joint inflammatory findings (24, 25, 26), and histologic synovial vascularization (27–29) in patients with RA.
Several studies have demonstrated a significant reduction of joint inflammation evaluated by gray-scale and color Doppler or PDUS in a limited number of arthritic joints in patients treated with different methods (15, 22, 30–37). Nevertheless, to the best of our knowledge, there are no studies on the predictive value of longitudinal US joint assessment in the areas of disease activity, functional status, and radiologic progression in early RA. The purpose of this study was to demonstrate the sensitivity to change of overall PDUS joint assessment and the predictive value of sequential PDUS parameters in clinical, functional, and radiologic outcomes in patients with early RA who started treatment with disease-modifying antirheumatic drugs (DMARDs).
PATIENTS AND METHODS
The study prospectively included 42 consecutive patients (11 men, 31 women) with early RA (joint symptoms for <1 year) according to the 1987 American College of Rheumatology (formerly the American Rheumatism Association) criteria for RA (38) who were attending the outpatient rheumatologic clinic and who started therapy with DMARDs. The mean ± SD age was 53.6 ± 14.1 years (range 24–77 years) and mean ± SD disease duration was 6.8 ± 3.6 months (range 1.5–12).
The patients underwent a clinical, laboratory, and PDUS evaluation at baseline (within 24 hours of starting treatment with DMARDs), 3 months, 6 months, and 1 year. Radiographic assessment was performed at baseline and at 1 year of followup. Therapeutic decisions were made without knowledge of the US findings. The study was approved by the local ethics committee and informed consent was obtained from all patients before study entry.
Clinical evaluation was performed for all patients by the same rheumatologist (PC), who was blinded to the US and radiographic findings and who was not involved in the treatment decisions. The following data were recorded for each patient at study entry: age, sex, symptom duration, nonsteroidal antiinflammatory drugs (NSAIDs) and corticosteroids received for RA before study entry, DMARDs prescribed, extraarticular involvement of RA, and rheumatoid factor (immunoturbidimetric assay, Roche/Hitachi Systems, Barcelona, Spain, normal level: 0–15 IU/ml). Drugs received for RA, extraarticular RA involvement, and joint surgery for RA were recorded at each visit.
At each visit, 28 joints (39) including bilateral glenohumeral, elbow, wrist, metacarpophalangeal (MCP), proximal interphalangeal (PIP) of the hands, and knee joints were assessed for tenderness and swelling. Tender joint count and swollen joint count were recorded for each patient. A global pain intensity visual analog scale score (VAS pain; range 0–100 mm), a VAS score for the patient's overall assessment of disease activity (range 0–100 mm), and functional status were also recorded at each visit. Functional ability was evaluated by a self-assessment Spanish version of the Health Assessment Questionnaire (HAQ) (40).
Serum markers of inflammation, C-reactive protein (CRP) level (immunoturbidimetric assay, Roche/Hitachi Systems; normal level: 0–10 mg/liter), and erythrocyte sedimentation rate (ESR; measured by the Westergren method, VESMATIC 60, version 2.05, Menarini Laboratory, Barcelona, Spain; normal level: 10–20 mm/hour) were obtained from each patient's laboratory test within 48 hours of each visit.
Disease activity assessment.
Disease activity was assessed by calculating the 28-joint Disease Activity Score (DAS28) for each patient at each visit (41).
The patients underwent a US assessment within 30 minutes of each clinical evaluation by a single rheumatologist experienced in US (EN) who was unaware of the clinical, laboratory, and radiographic findings and who was not involved in the treatment decisions. To reduce the possibility of bias, US was performed without access to the previous visit results. The patients were asked not to talk about their clinical symptoms with the US examiner.
A systematic gray-scale PDUS examination of the 28 joints clinically investigated was carried out with a commercially available real-time scanner (Logiq 500CL; General Electric Medical Systems, Kyunngi, Korea) using multifrequency linear array transducers (7–12 MHz). The US scanning method is described in Table 1 (14, 15, 17, 19, 24, 25, 31, 34, 42, 43). Joint synovitis was defined as the presence of intraarticular effusion and/or synovial hypertrophy. The presence of synovitis was identified in each joint as hypoechoic intraarticular material according to the criteria listed in Table 1 (19, 36, 42–44). Measurements were taken at the point where most capsular or joint recess distension was observed. Distances were measured using electronic calipers.
Table 1. Ultrasonographic scanning of joints and criteria of synovitis
Posterior recess, transducer transversal to the humerus, shoulder in neutral position: maximum distance from the posterior labrum to the posterior infraspinatus and teres minor tendon (posterior capsule) >3 mm. Axillar recess, transducer longitudinal to the axilla, shoulder in 90° of abduction: maximum distance from the humeral profile to the capsule >3 mm.
Longitudinal and transversally, from the anterior recess with the joint in extension: maximum distance from the humeral capitellum or the coronoid fossa to the joint capsule >2 mm.
Wrist (radiocarpal and midcarpal joint)
Longitudinal and transversally, from the dorsal aspect with the joint in neutral position: maximum distance from the bones to the joint capsule >2 mm.
Metacarpophalangeal and proximal interphalangeal joints of hands
Longitudinal and transversally, from the dorsal, medial, and lateral view with the joint in extension: maximum distance from the articular bony margin to the joint capsule >2 mm.
Longitudinal and transversally, from the suprapatellar recess, in a supine position, with the joint in 30° of flexion: maximum anteroposterior diameter of the suprapatellar recess >4 mm. Medial and lateral parapatellar recesses, transducer transversal to the patella, in a supine position, with the knee fully extended: maximum anteroposterior diameter >2 mm.
Synovial blood flow was evaluated by power Doppler in each of the 28 joints. Power Doppler imaging was performed by selecting a region of interest that included the bony margins, articular space, and a variable view of surrounding tissues (depending on the joint size). Power Doppler parameters were adjusted at the lowest permissible pulse repetition frequency (PRF) to maximize sensitivity. This setting resulted in PRF ranging from 500 Hz to 1,000 Hz, depending on the joint scanned. Low wall filters were used. The dynamic range was 20–40 dB. Color gain was set just below the level at which color noise appeared underlying bone (no flow should be visualized at bony surface). This setting resulted in gains from 18 dB to 30 dB. Flow was additionally demonstrated in 2 planes and was confirmed by pulsed wave Doppler spectrum to exclude artifacts.
Active synovitis was defined as the presence of intraarticular synovitis with power Doppler signal. US joint count for active synovitis was obtained at each US assessment. In addition, the intraarticular power Doppler signal was graded on a semiquantitative scale from 0 to 3 (0 = absence, no intraarticular flow; 1 = mild, single-vessel signal or isolated signals; 2 = moderate, confluent vessels; 3 = marked, vessel signals in more than half of the intraarticular area) during the US examination (19, 22, 28, 30, 37). An overall US joint index for power Doppler signal (the sum of the power Doppler signal scores obtained from each joint) was calculated at each US assessment. Representative images of PDUS findings are shown in Figure 1.
Posteroanterior films of patients' hands and anteroposterior films of patients' feet were made at baseline (within 10 days of study entry) and at 1 year of followup. The radiographs were read twice, with a minimum interval of 2 weeks, in chronological order by an independent observer (AC) who was blinded to patients' identity and clinical, laboratory, and US findings. Radiologic damage was assessed according to van der Heijde and colleagues' modification of Sharp's method (45, 46). This method measures erosions (score range 0–280) and joint space narrowing (JSN; score range 0–168) in 44 different joints, and provides a sum score ranging from 0 to 448. As defined in the description of the scoring method, total scores could increase or remain stable, but could not decrease. Results were expressed as erosion score, JSN score, and total score.
Scores from the first assessment of baseline and final films were used for analysis with the clinical, laboratory, and US data. Intraobserver reliability was assessed by calculating the intraclass correlation coefficient (ICC) from both radiographic readings.
US intraobserver reliability.
Intraobserver reliability of the US examination was evaluated by recording representative images of the 28 joints from one randomly chosen visit of 20 patients on a magnetic optical disk. The stored images were blindly read and scored for power Doppler signal by the same rheumatologist who performed all US examinations (EN) a minimum of 3 months after the corresponding real-time scanning.
The DAS28, HAQ score, radiographic erosion, JSN, and total scores at 1 year along with progression in the radiographic erosion, JSN, and total scores from baseline to 1 year were considered the outcome variables.
Statistical analysis was performed using SPSS statistical software, version 8.0 (SPSS, Chicago, IL). Quantitative variables (clinical, laboratory, US, and radiographic parameters) were given as the mean ± SD and range. Correlations between clinical, laboratory, US, and radiographic parameters were analyzed by Pearson's or Spearman's rank correlation test according to the variable distribution. The course of the process variables was obtained by calculating time-integrated values using the area under the curve method (47). Any P value less than 0.05 was considered statistically significant.
Sensitivity to change of the US variables was assessed by estimating the smallest detectable difference (SDD) (48). Intraobserver variability was obtained by calculating the ICC (2-way mixed effects model, consistency definition) for joint count for active synovitis and joint index for power Doppler signal. The US intraobserver reliability was also evaluated using the unweighted kappa test and the overall agreement (defined as the percentage of observed exact agreements) for the grade of power Doppler signal in each joint. Kappa values <0.40 reflect poor agreement, values 0.40–0.75 reflect fair to good agreement, and values >0.75 reflect excellent agreement (49).
Complete followup data were obtained from 38 of the 42 patients included in the study. One patient attended only the baseline and 1-year visits, 1 patient did not attend the 6-month and 1-year visits, and 2 patients did not attend the 1-year visit. Available data from the 4 patients with incomplete followup data were analyzed.
At study entry, rheumatoid factor (RF) was positive in 30 (71.4%) patients and negative in 12 (28.6%) patients. The mean ± SD positive RF value was 135 ± 160 IU/ml (range 16–880 IU/ml).
Before study entry, 27 (64.3%) patients had received oral corticosteroids for a mean ± SD of 1.6 ± 1.4 months (range 0.2–6 months) and 36 (85.7%) patients had received NSAIDs for a mean ± SD of 4.2 ± 2.9 months (range 1–12 months). At inclusion, 41 patients started therapy with 1 DMARD and 1 patient started therapy with 2 DMARDs. Therapeutic regimens included antimalarial drugs (61.9%), methotrexate (28.6%), leflunomide (4.8%), sulfasalazine (4.8%), and gold salts (2.4%). Low doses (5–10 mg/day) of prednisone were prescribed to 30 (71.4%) patients and NSAIDs were prescribed to 28 (66.7%).
After the 1-year followup, 31 patients were taking 1 DMARD and 7 patients were taking a combination of DMARDs. RA treatment consisted of antimalarial drugs (61.5%), methotrexate (41%), leflunomide (7.7%), sulfasalazine (5.1%), gold salts (2.6%), low doses (2.5–7.5 mg/day) of prednisone (66.7%), and NSAIDs (59%).
No patient had extraarticular RA involvement at baseline or at 1 year. Joint surgery for RA was not required for any patient during the study.
Clinical, laboratory, US, and radiographic course.
Clinical, laboratory, and US parameters during followup are shown in Table 2. Intraarticular power Doppler signal was only present in joints with synovitis. US examination of the 28 joints lasted ∼20 minutes, not including documentation. Bone erosions were detected in 19 (45.2%) patients at baseline and in 23 (59%) patients after 1 year. The mean ± SD radiographic erosion and JSN scores increased from 1.7 ± 3.2 (range 0–13) and 11.2 ± 9.8 (range 0–43), respectively, at baseline to 3.8 ± 6.3 (range 0–28) and 14.5 ± 11.1 (range 0–45), respectively, after 1 year. Seventeen (43.6%) patients showed a progression in radiographic erosion score and 27 (69.2%) in JSN score at 1 year of followup.
Table 2. Clinical, laboratory, and ultrasonographic course*
Values are the mean ± SD (range). VASP = visual analog scale for pain; VASOA = visual analog scale for patient's overall assessment of disease activity; TCJ = tender joint count; SJC = swollen joint count; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; DAS28 = 28-joint Disease Activity Score; HAQ = Health Assessment Questionnaire; USJCAS = ultrasonographic joint count for active synovitis; USJIPD = ultrasonographic joint index for power Doppler signal.
42.5 ± 30.5 (0–100)
34.4 ± 29.9 (0–100)
33 ± 30.9 (0–100)
34.9 ± 32.5 (0–100)
56.5 ± 30 (0–100)
38.8 ± 36.2 (0–100)
42.7 ± 30.4 (0–100)
36.7 ± 31.8 (0–100)
4.8 ± 4.2 (0–17)
2.6 ± 3.4 (0–13)
2.1 ± 3.1 (0–13)
1.8 ± 2.4 (0–9)
5.5 ± 4.5 (0–20)
3.8 ± 3.9 (0–15)
4 ± 4.9 (0–23)
2.4 ± 3.4 (0–16)
25.5 ± 18 (4–81)
22.7 ± 16.5 (1–88)
27.3 ± 22 (5–113)
20.5 ± 12.7 (4–50)
12.8 ± 12.8 (2–65)
8.6 ± 8.6 (2–35)
14.5 ± 22.4 (2–113)
8.3 ± 8.5 (2–39)
4.6 ± 1.2 (1.9–7.2)
3.6 ± 1.3 (1.7–6.6)
3.7 ± 1.2 (1.7–7.1)
3.4 ± 1.1 (1.7–6.1)
0.9 ± 0.7 (0–2.8)
0.6 ± 0.7 (0–2.4)
0.7 ± 0.7 (0–2.4)
0.6 ± 0.6 (0–2.3)
4 ± 4.5 (0–21)
2.3 ± 2.9 (0–12)
2.8 ± 4.1 (0–19)
1.9 ± 3.7 (0–14)
5.7 ± 6.4 (0–26)
3.8 ± 5.3 (0–25)
4 ± 6.3 (0–30)
3.1 ± 6.3 (0–23)
Transversal correlation between US variables and disease activity and functional status.
The cross-sectional correlations between the US parameters and the DAS28, CRP level, and HAQ score at each visit are shown in Table 3. The US joint count for active synovitis and US joint index for power Doppler signal correlated significantly with the DAS28 and CRP level throughout the study. The correlation coefficients were higher at 6 months and 1 year than at baseline and 3 months of followup. There was a weakly significant correlation between US variables and HAQ score at 3 months, 6 months, and 1 year.
Table 3. Transversal correlation between ultrasonographic variables, DAS28, HAQ, and CRP at each visit*
US = ultrasonographic; NS = nonsignificant; see Table 2 for additional abbreviations.
Intraobserver reliability and sensitivity to change of the US assessment.
Intraobserver kappa values for the US evaluation of each joint ranged from good to excellent (κ = 0.75–1). The mean ± SD kappa value was 1 ± 0 (range 1–1) for glenohumeral power Doppler signal, 0.75 ± 0.4 (range 0.49–1) for elbow power Doppler signal, 0.94 ± 0.1 (range 0.87–1) for wrist power Doppler signal, 0.91 ± 0.2 (range 0.47–1) for MCP power Doppler signal, 1 ± 0 (range 1–1) for PIP of the hands power Doppler signal, and 0.87 ± 0.2 (range 0.73–1) for knee power Doppler signal. Intraobserver US overall agreement ranged from 98% to 100%.
Intraobserver ICC was 0.99 (95% confidence interval [95% CI] 0.99–0.99) for the US joint count for active synovitis and 0.99 (95% CI 0.98–0.99) for the US joint index for power Doppler signal. The SDD was 0.95 for the US joint count for active synovitis and 1.61 for the US joint index for power Doppler signal.
The mean ± SD change in US joint count for active synovitis was 1.7 ± 3.8 from baseline to 3 months, 1.2 ± 3.1 from baseline to 6 months, and 2.1 ± 2.3 from baseline to 1 year. The mean ± SD change in US joint index for power Doppler signal was 1.9 ± 5.3 from baseline to 3 months, 1.7 ± 3.6 from baseline to 6 months, and 2.6 ± 3.7 from baseline to 1 year.
Intraobserver reliability of the radiographic assessment.
ICCs for the baseline films were 0.95 (95% CI 0.87–1) for the erosion score, 0.86 (95% CI 0.74–0.98) for the JSN score, and 0.85 (95% CI 0.71–0.98) for the total score; ICCs for the films taken after 1 year were 0.95 (95% CI 0.89–1) for the erosion score, 0.82 (95% CI 0.65–0.99) for the JSN score, and 0.82 (95% CI 0.65–0.98) for the total score.
Longitudinal correlation between US, clinical, and laboratory parameters and outcome variables.
There was no correlation between changes in the US parameters and changes in the DAS28 throughout followup. The correlations between the clinical, laboratory, functional, and US parameters at each visit and the DAS28 and HAQ score at the following visit demonstrated that the US joint count for active synovitis and US joint index for power Doppler signal were the strongest predictive variables of disease activity at the following visit (Table 4). The VAS pain and HAQ scores were the strongest predictors of functional status at the following visit (Table 4).
Table 4. Correlation between clinical, laboratory, and ultrasonographic variables at each visit and disease activity (DAS28) and functional ability (HAQ) at the following visit*
Correlation between baseline and 3 months
Correlation between 3 and 6 months
Correlation between 6 and 12 months
NS = nonsignificant; see Table 2 for additional abbreviations.
The correlations between the time-integrated values of the clinical, laboratory, functional, and US parameters and the outcome variables are displayed in Table 5. The time-integrated values of US joint count for active synovitis and US joint index for power Doppler signal demonstrated stronger significant correlations with the progressions in radiographic erosion score, JSN score, and total score, as well as the erosion and total scores at 1 year, than did the clinical, laboratory, and functional parameters, including the DAS28. In addition, the time-integrated values of the US parameters demonstrated a highly significant correlation with the DAS28 after 1 year (Table 5). The time-integrated US values did not correlate with functional status at 1 year (Table 5). There was not a significant correlation between the baseline clinical, laboratory, functional, and US parameters and the DAS28, HAQ score, and radiographic scores at 1 year of followup.
Table 5. Correlation between time-integrated value (TIV) of clinical, laboratory, and ultrasonographic variables and disease activity (DAS28), functional ability (HAQ), and radiologic outcome at 1 year of followup*
DAS28 1 year
HAQ 1 year
Erosion score progression
JSN score progression
Total score progression
Erosion score 1 year
JSN score 1 year
Total score 1 year
JSN = joint space narrowing; NS = nonsignificant; see Table 2 for additional abbreviations.
The development of new reliable methods for assessing synovial inflammation and response to treatment in RA is a challenge in daily practice and clinical trials and a relevant research field in rheumatology. Within the last decade, there has been an increasing use of musculoskeletal US with color Doppler or power Doppler technique for evaluating joint inflammatory activity in patients with RA.
In our study, we chose a combination of gray-scale (presence of joint effusion and/or synovial hypertrophy) and power Doppler findings (presence and grade of intraarticular power Doppler signal) as US variables reflecting active rheumatoid pannus. We found a significant transversal correlation between the PDUS findings and standard measurements of RA inflammatory activity such as the DAS28 and CRP level, whereas correlations between the PDUS parameters and HAQ score were weakly to moderately significant at followup. In a previous cross-sectional study on the comparison of gray-scale and power Doppler US with global clinical and laboratory assessment of joint inflammation in RA, we also found a significant correlation between US parameters and disease activity markers such as swollen joint count, CRP level, and ESR (19). In contrast, there was no correlation between the US variables and HAQ score. This discrepancy may be due to the longest disease duration (mean ± SD 69.3 ± 58.29 months) of the patients included in the previous study. In long-standing RA, HAQ score indicates either disease activity or residual structural joint damage, whereas in early RA, functional status is more likely related to inflammatory activity.
In keeping with our results, other studies have found changes in color Doppler or PDUS to be associated with clinical and laboratory response to intraarticular corticosteroid injections (22, 33, 36), systemic corticosteroid therapy (30, 32), and biologic agents (15, 31, 34, 37) in chronic inflammatory arthritis. However, the sensitivity to change of any method should be demonstrated by calculating the intraobserver variation and the SDD between repeated measurements (50). Most of the previous longitudinal studies have not assessed the sensitivity to change of US parameters. Ribbens et al (15) and Fiocco et al (37) reported intraobserver coefficients of lower variation than the changes in power Doppler findings. We obtained a lower SDD for the US joint count for active synovitis and the US joint index for power Doppler signal than the changes in these variables from baseline to 3 months, 6 months, and 1 year.
Although changes in PDUS parameters and DAS28 score were parallel throughout the study, we did not find a significant correlation between them. Therefore, PDUS findings seem to be a measurement of disease activity independent of standard clinical and laboratory variables.
The active synovitis count and the synovial vascularization index obtained by PDUS demonstrated a stronger correlation with disease activity at the following visit than the clinical, laboratory, and functional parameters, including the DAS28. In addition, the cumulative PDUS parameters of inflammatory activity over time demonstrated a high correlation with disease activity at 1 year and demonstrated the strongest correlation with radiographic damage progression as well as radiographic erosion and total scores after 1 year of DMARD therapy in patients with early RA. Because changes in the RA treatment throughout the study were based only on clinical and laboratory parameters, a predictive value of PDUS findings in disease activity and radiologic outcome may be accepted.
Taylor et al (35) have previously evaluated the prognostic value of US in RA in a recent randomized controlled trial of anti–tumor necrosis factor α in early RA. They demonstrated that the baseline synovial vascularization detected by power Doppler in MCP joints correlated with the radiographic joint damage over the following year (35) in patients receiving only 1 DMARD (methotrexate). In contrast, we did not find a significant correlation between the baseline clinical, laboratory, functional, and PDUS variables and disease activity, functional status, and radiographic damage at 1 year. The different therapeutic regimens prescribed for the patients during the followup may explain the lack of baseline predictors in our study.
The results of both studies can reflect the pathogenic destructive role of angiogenesis in the rheumatoid synovium (3–5). Therefore, the detection of vascularization in early rheumatoid synovial proliferation by PDUS could be considered a strong predictor of disease aggressiveness, which would contribute to making treatment decisions.
Some limitations of our study should be mentioned. First, our study was conducted in accordance with daily clinical practice. Patients were treated with various DMARDs, oral corticosteroids, and NSAIDs at a variable dose during the study. Therapeutic decisions were made without knowledge of US findings. Therefore, we could not compare the predictive value of PDUS variables depending on the DMARD received, evaluate the potential role of different DMARDs in PDUS parameters, or study the effect of PDUS findings in therapeutic decisions.
Moreover, the rheumatologist performing US scanning could not be completely unaware of patient's joint signs and symptoms. To avoid as much bias as possible, US examination was carried out without light so the examiner could not see the joints well, and the patients were asked not to communicate with the US examiner.
The lack of standardization of US examination method and settings for power Doppler can limit the use of this technique in research protocols. Some different methods have been used for assessing color Doppler or power Doppler findings such as semiquantitative signal scoring (15, 22, 30, 36, 37), color pixel counting (31–35), and resistive index calculating (33, 34). We considered semiquantitative signal grading as the most suitable for clinical practice. In agreement with previous studies (15, 37), our intraobserver kappa values and ICCs were very high for PDUS parameters.
In addition, power Doppler is extremely sensitive to tissue movement, especially at low PRF, which can result in flash artifacts. However, we used pulsed Doppler spectra as proof of the presence of vessels when the images were doubtful.
In conclusion, our results suggest that in addition to current clinical and laboratory evaluation, PDUS technique is a sensitive and reliable method for longitudinal assessment of inflammatory activity in patients with early RA in daily management and clinical trials. Furthermore, PDUS inflammatory findings seem to have a predictive value in disease activity as well as radiographic outcome. The latter emphasizes the importance of taking into account PDUS findings for therapeutic decisions in early RA.
Dr. Naredo 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. Dr. Naredo.
Acquisition of data. Drs. Naredo, Palop, Cabero, Richi, and Crespo.
Analysis and interpretation of data. Drs. Naredo, Collado, and Cruz.
Manuscript preparation. Dr. Naredo.
Statistical analysis. Dr. Carmona.
ROLE OF THE STUDY SPONSOR
The funding organizations agreed to submit the manuscript and approved the content.
We would like to thank Alejandro Balsa, MD, PhD, for his assistance with the radiographic assessment. We are grateful to General Electric Medical Systems Corporation for their technical support.