A Study of the Diagnostic Value of Doppler Ultrasound Score in a Single Wrist Joint to Assess the Activity of Rheumatoid Arthritis

Dear Editor,

Rheumatoid arthritis (RA) is a generalized, chronic inflammatory disease involving multiple joints, characterized by joint swelling, tenderness, tendonitis, tenosynovitis, bone destruction, and it usually affects the surrounding joints, mainly including the small joints of the hands and feet [1]. The 2018 Chinese Guidelines for the Diagnosis and Treatment of Rheumatoid Arthritis recommend the clinical use of 28 disease activity scores in 28 joints (DAS28) to evaluate patients' mobility [2]. Ultrasound, on the other hand, is more sensitive and reproducible than clinical assessment of joint inflammation [3]. Power Doppler ultrasound (PDUS) can capture the changes in blood flow signals, and on the basis of gray scale ultrasound (GSUS), determine whether there is inflammation at the hyperplasia site, and can diagnose RA more accurately and early through the joint examination of the wrist joint by GSUS and PDUS [4]. Nowadays, many studies at home and abroad use multi‐joint ultrasound scores to diagnose and assess RA mobility, but the establishment of single‐wrist ultrasound score as a standard for evaluating RA mobility is rarely reported. This study was approved by the Research Institution Review Committee of the First People's Hospital of Nantong, and the subjects voluntarily participated in the clinical study and signed the informed consent form.

RA patients who were admitted to the Department of Rheumatology and Immunology of Nantong First People's Hospital from 2021 to 2022 were randomly selected, a total of 100 RA patients were enrolled and divided into two groups based on disease activity: low/medium mobility and high mobility. Musculoskeletal ultrasonography was performed by using the French acoustic AIXPLORER ultrasound instrument and probe SL15‐6 (frequency 6 ~ 15 MHz). Scan the mid‐wrist, radial, and ulnar wrist of both wrist joints for diagnosis of synovitis, and scan the volar wrist flexor carpi radialis tendon, flexor pollicis longus tendon, flexor pollicis longus tendon, and dorsal carpal 1st to 6th chamber tendon for tendonization/tenosynovitis. Scan of the single wrist joint with an overall score, bone erosion, and joint effusion. The operator took pictures of all the scanned parts, and two sonographers with 5 years of experience in musculoskeletal imaging diagnosis could not obtain other clinical indicators of the patient, and the semi‐quantitative score of the more severe lateral wrist joint under ultrasound conditions was completed. Synovitis (S) and tendon/tenosynovitis (T) were assessed with gray‐scale score (GS) and energy Doppler score (PD), respectively, and bone erosion and joint effusion were assessed with GS. The Verio 3.0T high‐field magnetic resonance imaging instrument of Siemens was used to perform MR noncontrast scan + enhanced examination on the wrist joint. The wrist with the higher VAS score was scanned by the VAS score [1]. The MR non‐contrast scanning sequence selected the coronal turbo spin echo (TSE) and fat suppression (FS). TI‐TSE sequence: repetition time (TR) 500 ms; echo time (TE) 24 ms; Layer thickness 2 mm; Layer spacing 2 mm; Field of view (FOV): 180 mm × 180 mm. T2‐de3d sequence: TR 12.79 ms; TE 4.57 ms； Layer thickness 0.5 mm; Layer spacing 0 mm; FOV 180 mm × 180 mm. Axial Pd‐FS sequence: TR 2500 ms; TE 9.8 ms； Layer thickness 2 mm; Layer spacing 2 mm; FOV 120 mm × 120 mm [2]. MR enhancement was injected with a dose of 0.2 mL/kg and a flow rate of 2.0 mL/s Gadolinium glumethylamine acetate (Gd‐DTPA), and the injection site was the elbow vein, and the coronal T1‐vibe‐FS image was obtained. T1‐vibe‐FS sequence: TR 14.7 ms; TE 6.06 ms； Layer thickness 0.5 mm; Layer spacing 0 mm; FOV 180 mm × 180 mm. Two radiologists with 5 years of experience in RA osteoarthritic imaging will perform a semi‐quantitative score of the patient's magnetic resonance images, and if there is a disagreement, the agreement will be completed. Scored according to the RAMRIS system criteria [5], synovitis scores for the distal ulnar‐radial, radial wrist, and interosseous‐metacarpophalangeal joints of the wrist; Bone marrow edema and bone erosion scores were performed on the distal ulna, distal radius, base of metacarpal bone, and carpal bone, and the scores were added to obtain RAMRIS. The RA uni‐wrist ultrasound score, unilateral hand and wrist ultrasound score, and RAMRIS were consistent among the two observers, and the ICC values were 0.8, 0.8, and 0.7, respectively (p < 0.001).

In our study, including 47 in the high activity group (DAS 28 > 5.1) and 53 in the low and intermediate activity group (DAS‐28 ≤ 5.1). The incidence of RA synovitis was 58 cases, and the incidence rate was 58%, among which the midline of the wrist was the most common (45/100, 45%), and the radial side of the wrist (30/100, 30%) and the ulnar side of the wrist (29/100, 20%) were the most common incidences. There were 48 cases of tendon/tenosynovitis, with an incidence rate of 48%, among which the extensor tendon of the carpi radialis was the most common (16/100, 16%), followed by the flexor tendon of the third finger (15/100, 15%) and the flexor tendon of the second finger (14/100, 14%). There were 7 cases of bone erosion, with an incidence rate of 7%; There were 2 cases of joint effusion, with an incidence rate of 2%. There were statistically significant differences between the synovitis scores, mid‐wrist synovitis, wrist radial synovitis, wrist ulnar synovitis, tendon/tenosynovitis scores, 3rd finger flexor tendon tenosynovitis, 4th finger flexor tendon tenosynovitis, extensor carpi radialis breviferum tendon and extensor carpi radialis longus tenosynovitis, and total ultrasound scores, as shown in Table 1. The ultrasound performance of the RA high activity group is shown in Figure 1. The RA patient had a CRP of 55.77 mg/L, an ESR of 53 mm/h, five painful joints, five swollen joints, a DAS28 score of 5.5, synovitis, tenosynovitis, and bone erosion in the ultrasound image of the patient, and a single‐wrist ultrasound score of 14. The synovium and tendon of the wrist joint of RA patients were examined by GSUS, PDUS, and magnetic resonance imaging, respectively, and the differences of S, T, GS, PD, total ultrasound score, synovitis score (RAMRIS), bone marrow edema score (RAMRIS), bone erosion score (RAMRIS) and RAMRIS between the two groups were statistically significant (p < 0.05). The total ultrasound score, S score, T score, GS score, and PD score of RA were positively correlated with DAS28, and the correlation was statistically significant (rho = 0.447, 0.404, 0.221, 0.391, 0.461, p < 0.05). Within the ultrasound scores, the S score was positively correlated with the GS score and PD score (rho = 0.564, 0.822), the T score was positively correlated with the GS score and PD score (rho = 0.816, 0.550), and the GS score was positively correlated with the PD score (rho = 0.783), and the correlation was statistically significant. The DAS28 score of RA was positively correlated with wrist RAMRIS (rho = 482, p < 0.001). The area under the ROC curve of the unilateral wrist ultrasound score and the unilateral hand and wrist ultrasound score to determine different RA mobility and 95% CI were 0.814 (0.724–0.903) and 0.842 (0.762–0.921), respectively, and there was no significant difference between them (p = 0.396). The single wrist joint ultrasound score and RAMRIS areas under the ROC curve were 0.814 and 0.852, respectively, and there was no significant difference between them (p = 0.513).

TABLE 1.

Comparison of ultrasound image characteristics between the RA high‐mobility group and the low and intermediate mobility group.

Image features	Medium to low activity group	High mobility group	Z	p
Synovitis score (S) (points)	1 (0, 2)	6 (0, 11)	−4.609	< 0.001 a
Midline of the wrist	0 (0, 2)	2 (0, 4)	−3.037	0.002 a
Radial side of the wrist	0 (0, 0)	2 (0, 4)	−5.018	< 0.001 a
Ulnar side of the wrist	0 (0, 0)	2 (0, 4)	−5.212	< 0.001 a
Tendon/tenosynovitis score (T) (points)	0 (0, 2)	2 (0, 6)	−2.985	0.003 a
Wrist flexor tendons	0 (0, 0)	0 (0, 0)	−1.032	0.302
Flexor pollicis longus tendon	0	0 (0, 0)	−1.509	0.131
Flexor tendon of the 2nd finger	0 (0, 0)	0 (0, 0)	−1.012	0.311
Flexor tendon of the 3rd finger	0 (0, 0)	0 (0, 1)	−2.788	0.005 a
Flexor tendon of the 4th finger	0 (0, 0)	0 (0, 1)	−2.907	0.004 a
Flexor tendon of the 5th finger	0 (0, 0)	0 (0, 0)	−1.546	0.122
Extensor pollicis brevis tendon and abductor pollicis longus tendon	0 (0, 0)	0 (0, 0)	−0.151	0.880
Extensor carpi radialis brevis tendon and extensor carpi radialis longus tendon	0 (0, 0)	0 (0, 1)	−2.137	0.033 a
Extensor pollicis longus tendon	0 (0, 0)	0 (0, 0)	−0.666	0.505
Extensor tendons of the index fingers and extensor tendons of the fingers	0 (0, 0)	0 (0, 0)	−0.984	0.325
Ulnar carpi extensor tendon	0 (0, 0)	0 (0, 0)	−1.661	0.097
Bone erosion score (score)	0 (0, 0)	0 (0, 0)	−1.288	0.198
Joint effusion score (points)	0 (0, 0)	0 (0, 0)	−0.100	0.921

^a p < 0.05 was statistically significant.

FIGURE 1.

Male, 72‐year‐old with confirmed high‐mobility RA.

RA is characterized by persistent synovitis [6], Synovial inflammation can be used as an indicator or predictor of systemic inflammatory response [7]. In this study, it was found that the incidence of synovitis is higher than that of tenosynovitis, bone erosion, and joint effusion, and the incidence of tendon/tenosynovitis is lower in the low‐ to medium‐active group (37%). This suggests that the use of ultrasound to assess synovial inflammation and assess disease activity is of important clinical value. In this study, the diagnostic performance of ultrasound scores and RAMRIS in assessing the mobility of RA patients was compared, and the area under the ROC curve of single wrist ultrasound score and RAMRIS was 0.814 and 0.852, respectively, and the diagnostic performance of RAMRIS was higher, and there was no significant difference between the two. Although the diagnostic power of the two scores is similar, the composition of the scores is different; the wrist ultrasound score is composed of synovitis, tenosynovitis, bone erosion, and joint effusion and is quantitatively counted for tendon/tenosynovitis, while RAMRIS is the sum of synovitis, bone erosion, and bone marrow edema scores, and only qualitative statistics are used for tendon/tenosynovitis. On the one hand, ultrasound is cheap, real‐time, and convenient, while magnetic resonance is expensive and less available. On the other hand, bone marrow edema is a reliable predictor of arthritic disease progression [8], bone marrow edema cannot be detected by ultrasound, whereas magnetic resonance imaging is sensitive in assessing joint inflammation and can detect bone marrow edema [9]. Due to the difference in imaging ability between the two and the different focus of detection, ultrasound cannot be used to completely replace MRI, so the two can be combined, and their respective advantages can be used to diagnose ultrasound and magnetic resonance when conditions permit, so as to obtain better disease evaluation results. In this study, we also compared the single wrist score, unilateral hand, and wrist scores, and there was no significant difference between them, with an area under the ROC curve of 0.814 and 0.842, respectively (p = 0.396), which further confirmed the possibility of reducing the hand joint score, shortening the examination time without reducing the diagnostic rate of mobility. DAS28 can be affected by external factors such as environment and psychology, and while DAS28 does not include imaging assessments, ultrasound provides a visual picture of disease progression and is more meaningful than laboratory results. Patients with RA often undergo ultrasound of both hands and wrists, and the vast majority of RA ultrasound scores are multi‐articular; for example, Nam J et al. successfully predicted the progression of RA using a US32‐joint protocol [10]. In this study, considering the cumbersomeness of US examination of multiple joints, we performed US examination of both wrists and selected unilateral wrist joints with more severe US images for scoring, which simplified the ultrasound scoring and saved the time and cost of examination of PIP and MCP in both hands. Disease activity can be assessed more rapidly, and screening for active RA can be facilitated.

Our study also has certain limitations. First, the sample size of this study is small, and there may be selection bias. Secondly, there is a lack of a healthy control group in this study. Thirdly, other joint scores were not included in this study. Fourth, pathological results were not used as the gold standard to evaluate the diagnostic efficacy of different methods.

In conclusion, our study demonstrated that the ultrasound score of a single wrist joint is closely related to rheumatoid arthritis activity and has diagnostic potential for RA activity.

Author Contributions

Zijing Chu, Zhixing Zhou, Muhammad Asad Iqbal contributed to the conception and design of the study. Zhongxin Zhang, Xian Wang wrote sections of the manuscript. All authors contributed to the article and approved the submitted version.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.