Exploration of acute gout diagnosis based on ultrasound viscoelastic imaging: quantitative parameter analysis and clinical validation Authors: Jiesi Zhang, Xiao Chen, Peipei Li, Mei Zhao, Li Li, Yuquan Ye Published: 2025 ABSTRACT ObjectivesTo explore the optimal quantitative parameters and clinical application value of ultrasound viscoelastic imaging in the differential diagnosis of acute gout (AG) and non-AG.MethodsThis study enrolled 102 patients presenting with acute joint swelling and pain, and divided them into the AG group and non-AG group. Differences in viscoelastic quantitative parameters, including the shear wave velocity (C), viscosity coefficient (V) and dispersion coefficient (D), between the two groups were compared. Parameters with significant differences were included in generalized estimating equations (GEE) analysis to screen out the optimal parameters for predicting AG, and receiver operating characteristic (ROC) curves were plotted to evaluate the diagnostic efficacy of each parameter alone and in combination.ResultsThe AG group included 45 patients, and the non-AG group included 57 patients. There were statistically significant differences in Cmean, Cmax, Cmin, CSD, Vmean, Vmax, VSD, Dmean, Dmax, and DSD between the two groups (all P < 0.05). GEE analysis showed that the parameters finally included in the equation were Cmean and Dmax, and the equation was Logit(P) = (−11.186) + (3.456 × Cmean) + (0.191 × Dmax). ROC curve analysis revealed that among single parameters, Cmean had the highest AUC of 0.836 (95% CI 0.774∼0.897), with an optimal cutoff value of 2.36 m/s. The AUC of the GEE model was 0.885 (95% CI 0.833∼0.938), with an optimal cutoff value of 0.41. The AUC of the GEE model combining multiple parameters was superior to that of single-parameter diagnosis (P = 0.033).ConclusionAmong ultrasound viscoelastic parameters, Cmean and Dmax are effective non-invasive quantitative indicators for diagnosing AG, and the diagnostic efficacy of combined multi-parameter diagnosis is superior to that of single-parameter diagnosis. Introduction Gout is a crystal-related arthropathy caused by the deposition of monosodium urate (MSU), which is directly associated with hyperuricemia resulting from purine metabolism disorders and/or reduced uric acid excretions. With the progression of the disease, inflammation may lead to permanent damage and destruction of articular cartilage and bone. Without effective and timely treatment, gout can result in adverse outcomes such as disability, loss of work capacity, and loss of self-care ability. With the development of social economy and changes in people’s dietary structure, the prevalence of gout has been increasing year by year and tends to be younger. The global prevalence of gout ranges from 0.1% to 6.8% (1). Currently, the internationally recognized gold standard for the diagnosis of gout is the detection of urate crystals via joint aspiration (2). However, for patients with a small amount of joint effusion or involvement of small joints, joint aspiration is technically challenging (3). Additionally, needle aspiration biopsy is an invasive procedure with potential risks of complications. Therefore, it is crucial to explore a non-invasive and reliable diagnostic method. In recent years, imaging techniques have made significant progress in the diagnosis and evaluation of gout. The 2015 gout classification criteria issued by the American College of Rheumatology and the European League Against Rheumatism (ACR/EULAR) incorporated ultrasound and dual-energy computed tomography (DECT) as new effective tools for gout diagnosis (4). High-frequency ultrasound is cost-effective and convenient, allowing real-time dynamic observation of lesions, which makes it suitable for routine screening and efficacy follow-up. Specific ultrasound signs (double contour sign, tophus) have high specificity and high positive predictive value for early gout, but their sensitivity is relatively limited (5, 6). Therefore, exploring a new non-invasive, simple technique that can improve the sensitivity of ultrasound diagnosis has become the direction of our research. Shear wave elastography (SWE) is a non-invasive dynamic imaging technique capable of quantifying changes in tissue elasticity (stiffness). It reflects differences in tissue elasticity by measuring the propagation velocity of directional shear waves emitted by ultrasonic pulses, thereby acquiring quantitative information regarding the elasticity of biological tissues. However, current SWE techniques treat tissues as homogeneous pure elastic bodies, while living tissues are heterogeneous and complex biomechanical structures that exhibit both elastic solid properties and viscous fluid properties. Owing to the viscous nature of biological tissues, the propagation velocity of shear waves within them varies with frequency, a phenomenon known as dispersion during propagation (7). Based on conventional shear wave elastography, ultrasound viscoelastic imaging incorporates the dispersion factor. Utilizing a novel rheological model (the VOIGT model) (8), it quantifies the differences in shear wave velocities at different frequencies to calculate two new quantitative parameters: the viscosity coefficient (V) and the dispersion coefficient (D), which can reflect the viscous characteristics of tissues. Simultaneously, it can also obtain the shear wave velocity(C) and elastic modulus (E), which are capable of reflecting the elastic (stiffness) characteristics of tissues(9).Ultrasound viscoelastic imaging is an emerging technique that has garnered increasing attention in recent years. It has demonstrated several advantages in the assessment of liver fibrosis and liver grafts (10–12), non-invasive evaluation of chronic kidney disease (13), and differentiation of benign and malignant breast lesions (14–16). Currently, there is a paucity of studies on the application of viscoelastography in gout. This study aimed to explore the application value of multiple ultrasonic shear wave viscoelastic parameters in the diagnosis of acute gout and identify the optimal quantitative parameters. Materials and methods Patients This retrospective study analyzed 102 outpatients and inpatients (involving 151 joints) who presented with joint swelling and pain at a single center from April 2024 to March 2025. According to the corresponding clinical guidelines, the patients were divided into the acute gout (AG) group and the non-AG group. Inclusion criteria: (1) All patients had acute onset of joint swelling and pain involving ≥ 1 joint within 3 days before ultrasound examination; this could be either the first episode of joint swelling and pain or an acute exacerbation of chronic joint disease. (2) All patients underwent grayscale ultrasound and ultrasonic shear wave viscoelastography examinations. (3) Grayscale ultrasound showed synovial hypertrophy in the swollen and painful joints. Exclusion criteria: (1) Pregnant or lactating women; (2) Patients with more than two concurrent joint diseases; (3) Patients who had taken uric acid-lowering drugs or hormonal therapeutic drugs within the past one month; (4) Patients with complicated malignant tumors; (5) Patients with cognitive or intellectual impairment and poor clinical compliance. Grouping criteria: All patients in the acute gout (AG) group met the 2015 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) gout classification criteria (4). Patients in the non-AG group were compliance with the diagnostic criteria of the American College of Rheumatology guidelines or pathologic diagnosis; and normal serum uric acid concentration. All patients were independently diagnosed by two physicians with more than 10 years of clinical experience in rheumatology and immunology. Patients in the AG group were strictly diagnosed in accordance with the 2015 ACR/EULAR gout classification criteria, while those in the non-AG group were diagnosed based on the ACR diagnostic criteria for the corresponding disease or pathological diagnosis results. For patients with uncertain diagnosis (e.g., atypical clinical manifestations, ambiguous laboratory indicators), their grouping was collectively determined through departmental case discussions. Patients with two or more concurrent joint diseases were excluded in accordance with the exclusion criteria (a total of 3 such patients were excluded), as concurrent diseases might interfere with the analysis of viscoelastic parameters (Figure 1). Flowchart of patient selection in this study. This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of Hebei General Hospital (No. 2025-247), and registered at the China Clinical Trial Registration Centre (ChiCTR2500103688). All ultrasound examinations were performed as part of routine clinical practice, with prior informed consent obtained from patients or their legal guardians. Sample size calculation The sample size of this study was estimated based on pilot study data (n = 20). Taking the core viscoelastic parameter Cmean (mean shear wave velocity) as the primary outcome measure, the effect size was set to 0.8, the significance level α = 0.05 (two-tailed), and the power (1−β) = 80%. Calculations were performed using G*Power 3.1 software, and the results showed that the minimum required total sample size was 84 cases. A total of 102 patients were finally included in this study, which exceeded the minimum sample size requirement and ensured sufficient statistical power for the comparison of key parameters. Instrumentation A Mindray Resona A20 diagnostic ultrasound system (Shenzhen, China) equipped with a linear array transducer (frequency range: 3–15 MHz) was used. Ultrasound examinations Gray-scale ultrasound and viscoelastic ultrasound examinations were performed on all subjects by the same ultrasonographer. Before the examination, patients were in a resting state, and their positions were adjusted according to the examination site with the affected joints fully exposed. The ultrasound examination method for each joint was conducted with reference to the ESSR Musculoskeletal Ultrasound: Technical Guidelines (17). Grayscale scanning was performed on the swollen and painful joints, with observation and documentation of synovial thickness, joint effusion, bone erosion, double contour sign, and tophus. Subsequently, based on gray-scale ultrasound, the maximum longitudinal section of synovial thickening in the joint was obtained for ultrasonic shear wave viscoelastography examination. When acquiring images, the probe was gently placed in contact with the patient’s skin, and stabilized with minimal pressure. The acquisition frame covered the entire lesion area. To reduce interference from respiratory movement, patients were instructed to remain calm and hold their breath for 3–5 s. Then, the viscoelastic examination was activated to enter the four-panel mode, which simultaneously displayed the gray-scale image, shear wave elasticity images, viscosity images and reliability map (RLB) (Figure 2). The system integrated quality control modes including motion stability index (M-STB) and RLB to improve data stability and reliability. M-STB was graded on a scale of 1–5 stars, which dynamically evaluated motion interference between the biomedical sensor and the subject’s skin. Images were considered to be acquired under stable conditions when M-STB ≥ 4 stars. The RLB displayed areas with high reliability in green and areas with low reliability in purple. The RLB index indicated the reliability of the image as a percentage, and images were deemed qualified when the RLB index > 90%. All physicians performing ultrasound examinations and parameter analysis were completely blinded to the patients’ clinical information and did not refer to any clinical data during the operation, including demographic characteristics, medical history, laboratory results, joint aspiration findings, clinical diagnoses, and treatment plans. Schematic diagram of viscoelastic images (measurement of shear wave velocity and viscosity coefficient in the first metatarsophalangeal joint of the right foot). (a) (top left) Shear wave elasticity image; (b) (top right) Viscosity coefficient image; (c) (bottom left) Grayscale ultrasound image; (d) (bottom right) Reliability map (RLB). Image analysis Synovial hypertrophy was assessed on single grayscale images, primarily characterized by hypoechoic areas within the joint. In this study, the definition of synovial thickening refers to the EULAR-OMERACT semi-quantitative scoring system(18), Grade 0 = No synovial hypertrophy (SH) independently of the presence of effusion; Grade 1 = Minimal hypoechoic SH up to the level of the horizontal line connecting bone surfaces between the metacarpal head and the proximal phalanx; Grade 2 = Moderate hypoechoic SH extending beyond joint line but with the upper surface concave (curved downwards) or hypertrophy extending beyond the joint line but with the upper surface flat; Grade 3 = Severe hypoechoic SH with or without effusion extending beyond the joint line but with the upper surface convex (curved upward). Patients with a synovial grade of 1–3 are considered to have synovial hypertrophy and included in this study, while those with Grade 0 (no synovial thickening) are excluded. Images were analyzed under the four-panel mode of viscoelastic imaging (Figures 2, 3). On the gray-scale ultrasound image, region of interest (ROI) was delineated along the edge of the thickened synovial, and the corresponding ROI was synchronously displayed on the shear wave elasticity and viscosity images. The system automatically acquired quantitative parameters: shear wave elastic parameters (reflecting tissue stiffness) including the mean, maximum, minimum, and standard deviation of shear wave velocity (C, m/s); viscosity imaging parameters (reflecting tissue viscosity) including the mean, maximum, minimum, and standard deviation of viscosity coefficient (V, Pa⋅s) and dispersion coefficient [D, (m/s)/kHz]. A total of 12 parameters were obtained. Image acquisition and data measurement were repeated 5 times, and the mean values were recorded and calculated. Schematic diagram of viscoelastic images (measurement of shear wave velocity and dispersion coefficient in the first metatarsophalangeal joint of the right foot). (a) (top left) Shear wave elasticity image; (b) (top right) Dispersion coefficient image; (c) (bottom left) Grayscale ultrasound image; (d) (bottom right) Reliability map (RLB). To verify the reliability of ultrasound viscoelastic parameter measurement, a subset of images was selected for intra-rater and inter-rater reliability assessment. A total of 20 joint ultrasound images were randomly chosen, with 10 from the acute gout (AG) group and 10 from the non-AG group. Twelve viscoelastic parameters (including shear wave velocity, viscosity coefficient, and dispersion coefficient-related parameters) were independently and repeatedly measured by two physicians who had more than 5 years of professional experience in musculoskeletal ultrasound. The intraclass correlation coefficient (ICC) was adopted as the statistical indicator to evaluate intra-rater reliability and inter-rater reliability. Statistical analysis All statistical analyses were performed using SPSS 27.0 software (IBM Corp, Armonk, NY, USA) and MedCalc 22.0 software (MedCalc Software Ltd, Ostend, Belgium). Categorical data were expressed as n (%), and comparisons between groups were conducted using the chi-square (χ2) test. Continuous data were tested for normality via the Shapiro-Wilk test. Data conforming to a normal distribution were presented as X¯ ± S, and comparisons between groups were performed using the independent samples t-test.Data with a skewed distribution were expressed as M (P25, P75), and comparisons between groups were conducted using the Mann-Whitney U test. For the assessment of measurement reliability (intra-rater and inter-rater reliability of viscoelastic parameters), the intraclass correlation coefficient (ICC) was used. With gout as the dependent variable and each quantitative parameter of ultrasound viscoelastic imaging as the independent variable, generalized estimating equations (GEE) was used to screen for independent influencing factors of gout among the ultrasound viscoelastic parameters. Receiver operating characteristic (ROC) curves were plotted to analyze the diagnostic efficacy of each parameter alone and in combination for acute gout. The DeLong test was used to compare the areas under the curves (AUC). A two-tailed P-value of less than 0.05 was considered to indicate statistically significant differences. Results Participant characteristics A total of 45 patients (involving 70 joints) were enrolled in the gout group, aged 21∼78 years, with a mean age of (47.71 ± 13.43) years, including 38 males and 7 females. The non-gout group included 57 patients (involving 81 joints), aged 28∼75 years, with a mean age of (48.58 ± 15.05) years, including 42 males and 15 female. There were no statistically significant differences in the gender composition, age, and distribution of joint sizes between the two groups (χ2 = 1.721, P = 0.190; t = −0.303, P = 0.762; χ2 = 3.802, P = 0.149). The non-gout group consisted of 36 cases of rheumatoid arthritis, 12 case of osteoarthritis, 5 case of ankylosing spondylitis, 3 case of psoriatic arthritis, and 1 case of dermatomyositis. Comparisons of BMI, disease duration, and synovial thickness between the two groups showed no statistically significant differences (all P > 0.05). The serum uric acid level in the AG group was higher than that in the non-AG group, with a statistically significant difference (t = 10.249, P < 0.001) (Table 1). TABLE 1Baseline characteristics of patients.Participant characteristicsAG group (n = 45)Non-AG group (n = 57)Statistical valueP-valueAge (years)47.71 ± 13.4348.58 ± 15.05−0.3030.762*Male (case)38 (84.44%)42 (73.68%)1.7210.190#BMI (kg/m2)24.32 ± 1.3923.97 ± 1.401.2580.211*Disease duration (months)17.27 ± 9.8220.54 ± 12.42−1.4460.151*Serum uric acid (μmol/L)455.18 ± 96.36283.01 ± 65.7410.249<0.001*Number of involved joints (case)n = 70n = 81––Large joints (hip joint, knee joint, shoulder joint)15 (21.43%)29 (35.80%)3.8020.149#Medium joints (wrist joint, ankle joint, elbow joint)21 (30.00%)21 (25.93%)Small joints (hand joints, foot joints)34 (48.57%)31 (38.27%)AG, acute gout; NAG, non-acute gout; BMI, body mass index. *Independent samples t-test. #Chi-square(χ2). Baseline characteristics of patients. AG, acute gout; NAG, non-acute gout; BMI, body mass index. *Independent samples t-test. #Chi-square(χ2). Comparative of grayscale ultrasound images between the AG group and non-AG group Statistical analysis was performed on the ultrasound signs between the AG and non-AG groups, including synovial thickness, joint effusion, bone erosion, double contour sign, and tophus. The results showed that there were no statistically significant differences in synovial thickness or bone erosion between the two groups (P > 0.05), while significant statistical differences were observed in joint effusion, double contour sign, and tophus between the two groups (P < 0.05). Using positive double contour sign or positive tophus as the diagnostic criterion, the sensitivity, specificity, AUC of ultrasound for diagnosing AG were 61.2%, 88.9%, and 0.752 (95% CI 0.670–0.833), respectively (Supplementary Table 1). Reliability of viscoelastic parameter measurement The results of reliability assessment showed that the intra-rater ICC of the 12 ultrasound viscoelastic parameters ranged from 0.83 to 0.91, and the inter-rater ICC ranged from 0.79 to 0.89. According to the commonly used ICC reliability evaluation criteria (ICC > 0.75 indicates good reliability), both intra-rater and inter-rater reliability of the parameter measurements in this study were classified as good, confirming the stability and credibility of the viscoelastic parameter measurement results (Supplementary Table 2). Comparative of ultrasound viscoelastic parameters between the AG group and non-AG group A total of 12 parameters were obtained in this study. Among them, the comparison of Cmean, Cmax, Cmin, CSD, Vmean, Vmax, VSD, Dmean, Dmax, DSD between the AG group and non-AG group showed statistically significant differences (all P < 0.05); while the differences in Vmin and Dmin between the two groups were not statistically significant (P > 0.05) (Table 2). TABLE 2Comparison of ultrasound viscoelastic parameters between AG group and non-AG group.ParametersAG groupNon-AG groupStatistical valueP-valueNumber of involved joints (case)7081––Cmean (m/s)2.69 (2.44, 3.08)2.13 (1.78, 2.49)−7.103<0.001▲Cmax (m/s)4.43 (3.83, 5.37)3.29 (2.77, 4.06)−5.685<0.001▲Cmin (m/s)1.32 ± 0.541.08 ± 0.582.5420.012*CSD (m/s)0.59 (0.45, 0.80)0.45 (0.29, 0.68)−3.0300.002▲Vmean (Pa.s)0.88 (0.50, 1.30)0.64 (0.43, 0.80)−2.9650.003▲Vmax (Pa.s)3.18 (1.76, 4.59)1.79 (1.30, 3.16)−3.3640.001▲Vmin (Pa.s)0.00 (0.00, 0.00)0.00 (0.00, 0.00)−0.2600.795▲VSD (Pa.s)0.64 (0.41, 1.01)0.39 (0.27, 0.61)−4.079<0.001▲Dmean [(m/s)/kHz]5.43 (4.05, 6.64)4.09 (3.01, 6.36)−2.4930.013▲Dmax [(m/s)/kHz]18.12 (10.68, 20.00)10.91 (8.09, 15.05)−3.960<0.001▲Dmin [(m/s)/kHz]0.00 (0.00, 1.15)0.00 (0.00, 1.40)−0.4160.677▲DSD [(m/s)/kHz]3.17 (1.97, 3.81)1.95 (1.51, 2.80)−3.931<0.001▲AG, acute gout; NAG, non-acute gout; C, shear wave velocity; V, viscosity coefficient; D, dispersion coefficient. ▲Mann-Whitney U test. *Independent samples t-test. Comparison of ultrasound viscoelastic parameters between AG group and non-AG group. AG, acute gout; NAG, non-acute gout; C, shear wave velocity; V, viscosity coefficient; D, dispersion coefficient. ▲Mann-Whitney U test. *Independent samples t-test. Generalized estimating equations (GEE) analysis of ultrasound viscoelastic parameters for predicting AG With gout as the dependent variable and the ultrasound viscoelastic parameter (Cmean, Cmax, Cmin, CSD, Vmean, Vmax, VSD, Dmean, Dmax, DSD) which showed statistically significant differences between the AG group and non-AG group as the independent variables, GEE analysis was performed. The results showed that Cmean and Dmax were independent risk factors for predicting AG (both OR > 1, P < 0.05). Based on the results of the GEE analysis, a multi-parameter combined diagnostic equation was established as follows: Logit(P) = (−11.186) + (3.456 × Cmean) + (0.191 × Dmax) (Table 3). TABLE 3Generalized estimating equations (GEE) analysis results of ultrasound viscoelastic parameters for predicting acute gout (AG).ParametersBSE.Wald P OR95% CI for ORLowerUpperCmean3.4560.72822.550<0.00131.6837.610131.912Dmax0.1910.04815.686<0.0011.2101.1011.330Constant−11.1861.83337.231<0.0010.000––C, shear wave velocity; D, dispersion coefficient; B, unstandardized regression coefficient; SE, standard error; OR, odds ratio. Generalized estimating equations (GEE) analysis results of ultrasound viscoelastic parameters for predicting acute gout (AG). C, shear wave velocity; D, dispersion coefficient; B, unstandardized regression coefficient; SE, standard error; OR, odds ratio. Exploration of optimal ultrasound viscoelastic parameters Among the single parameters, Cmean showed the highest diagnostic efficacy, with an AUC of 0.836 (95% CI 0.774∼0.897). The optimal cut-off value was 2.36 m/s (Figures 4, 5), at which the sensitivity and specificity for diagnosing AG were 81.4% and 71.6%, respectively. The AUC of the GEE model was 0.885 (95% CI 0.833∼0.938), with an optimal cutoff value of 0.41, and the corresponding sensitivity and, specificity for diagnosing AG were 87.1% and 76.5%, respectively (Table 4). Ultrasound viscoelastic image of the left elbow joint in a patient with acute gout (AG). A 69-year-old male patient. Ultrasound viscoelastic imaging showed that the thickened synovium of the left elbow joint had a Cmean of 2.98 m/s (>cut-off value of 2.36 m/s) and a Dmean of 8.47 (m/s)/kHz [>cutoff value of 4.35 (m/s)/kHz]. Ultrasound viscoelastic image of the right wrist joint in a patient with rheumatoid arthritis. A 60-year-old male patient. Ultrasound viscoelastic imaging showed that the thickened synovium of the right wrist joint had a Cmean of 2.04 m/s (