Deep Learning Algorithm of the SPARCC Scoring System in SI Joint MRI.

BACKGROUND: The Spondyloarthritis Research Consortium of Canada (SPARCC) scoring system is a sacroiliitis grading system. PURPOSE: To develop a deep learning-based pipeline for grading sacroiliitis using the SPARCC scoring system. STUDY TYPE: Prospective. POPULATION: The study included 389 participants (42.2-year-old, 44.6% female, 317/35/37 for training/validation/testing). A pretrained algorithm was used to differentiate image with/without sacroiliitis. FIELD STRENGTH/SEQUENCE: 3-T, short tau inversion recovery (STIR) sequence, fast spine echo. ASSESSMENT: The regions of interest as ground truth for models' training were identified by a rheumatologist (HYC, 10-year-experience) and a radiologist (KHL, 6-year-experience) using the Assessment of Spondyloarthritis International Society definition of MRI sacroiliitis independently. Another radiologist (YYL, 4.5-year-experience) solved the discrepancies. The bone marrow edema (BME) and sacroiliac region models were for segmentation. Frangi-filter detected vessels used as intense reference. Deep learning pipeline scored using SPARCC scoring system evaluating presence and features of BMEs. A rheumatologist (SCWC, 6-year-experience) and a radiologist (VWHL, 14-year-experience) scored using the SPARCC scoring system once. The radiologist (YYL) scored twice with 5-day interval. STATISTICAL TESTS: Independent samples t-tests and Chi-squared tests were used. Interobserver and intraobserver reliability by intraclass correlation coefficient (ICC) and Pearson coefficient evaluated consistency between readers and the deep learning pipeline. We evaluated the performance using sensitivity, accuracy, positive predictive value, and Dice coefficient. A P-value <0.05 was considered statistically significant. RESULTS: The ICC and the Pearson coefficient between the SPARCC scores from three readers and the deep learning pipeline were 0.83 and 0.86, respectively. The sensitivity in identifying BME and accuracy of identifying SI joints and blood vessels was 0.83, 0.90, and 0.88, respectively. The dice coefficients were 0.82 (sacrum) and 0.80 (ilium). DATA CONCLUSION: The high consistency with human readers indicated that deep learning pipeline may provide a SPARCC-informed deep learning approach for scoring of STIR images in spondyloarthritis. EVIDENCE LEVEL: 1 TECHNICAL EFFICACY: Stage 2.

A deep neural network for MRI spinal inflammation in axial spondyloarthritis.

OBJECTIVE: To develop a deep neural network for the detection of inflammatory spine in short tau inversion recovery (STIR) sequence of magnetic resonance imaging (MRI) on patients with axial spondyloarthritis (axSpA). METHODS: A total 330 patients with axSpA were recruited. STIR MRI of the whole spine and clinical data were obtained. Regions of interests (ROIs) were drawn outlining the active inflammatory lesion consisting of bone marrow edema (BME). Spinal inflammation was defined by the presence of an active inflammatory lesion on the STIR sequence. The 'fake-color' images were constructed. Images from 270 and 60 patients were randomly separated into the training/validation and testing sets, respectively. Deep neural network was developed using attention UNet. The neural network performance was compared to the image interpretation by a radiologist blinded to the ground truth. RESULTS: Active inflammatory lesions were identified in 2891 MR images and were absent in 14,590 MR images. The sensitivity and specificity of the derived deep neural network were 0.80 ± 0.03 and 0.88 ± 0.02, respectively. The Dice coefficient of the true positive lesions was 0.55 ± 0.02. The area under the curve of the receiver operating characteristic (AUC-ROC) curve of the deep neural network was 0.87 ± 0.02. The performance of the developed deep neural network was comparable to the interpretation of a radiologist with similar sensitivity and specificity. CONCLUSION: The developed deep neural network showed similar sensitivity and specificity to a radiologist with four years of experience. The results indicated that the network can provide a reliable and straightforward way of interpreting spinal MRI. The use of this deep neural network has the potential to expand the use of spinal MRI in managing axSpA.

MRI lesions in SpA: a comparison with noninflammatory back pain using propensity score adjustment method.

Background: Magnetic resonance imaging (MRI) is important in the management of axial spondyloarthritis (SpA). However, many MRI lesions are not exclusive to axial SpA. Further characterization of these lesions may lead to better clinical decisions. Objective: The objective of this study was to compare the frequency of individual spinal MRI lesions between axial SpA and noninflammatory back pain. The factors associated with individual lesions in participants with axial SpA were also determined. Design: This was a cross-sectional observational study. Methods: MRI lesions in 447 participants with axial SpA and 122 participants with noninflammatory back pain were compared using the propensity score adjustment method. Individual lesions included discovertebral lesions (DVL), Modic type 1 lesions, DVL without Modic type 1 lesions, facet joint lesions, costovertebral joint lesions, corner inflammatory lesions (CIL), and fatty corner lesions (FCL). The factors associated with the lesions were determined using regression analyses. Results: Among participants with axial SpA, 81.9% were HLA-B27-positive, 55.0% had radiographic axial SpA, and 60.5% had radiographic features of spinal damage (mSASSS >2). Almost half (48.6% in axial SpA versus 31.1% in noninflammatory back pain) had inflammatory lesions on spinal MRI. In propensity score matching with noninflammatory back pain, axial SpA had an increased occurrence of DVL without Modic type 1 lesion (OR = 3.43, p = 0.01), costovertebral lesion (OR = 11.89, p = 0.02), number of CIL (B = 1.19, p < 0.001), and number of FCL (B = 3.33, p < 0.001). Similar associations were found in the regression models in the radiographic axial SpA subgroup: DVL without Modic type 1 lesion (OR = 2.46, p = 0.001), costovertebral lesion (OR = 3.86, p < 0.001), number of CIL (B = 1.13, p < 0.001), and FCL (B = 2.29, p < 0.01). Conclusion: MRI lesions including DVL without Modic type 1, costovertebral joint lesions, CIL, and FCL were more specific in axial SpA.

Clinical, radiological, and magnetic resonance imaging characteristics of axial spondyloarthritis with late onset.

We aimed to investigate the clinical, diagnostic, and imaging features of patients with late onset axial spondyloarthritis (SpA) with initial symptom manifestation aged over 45 years. Participants with axial SpA were consecutively recruited. Clinical, demographic, blood, and imaging parameters were compared between the groups with early (≤45 years) and late onset (>45 years) at a cross-sectional level. Logistic regressions were used to determine the independent associations with axial SpA with late onset. A total of 455 participants were recruited. Among them, 70 (15.4%) had late onset disease. Multivariate analyses showed that axial SpA with late onset was associated with higher C-reactive protein based ankylosing spondylitis disease activity index (ASDAS-CRP) (B = 0.10; P = .04), higher intensity of spinal inflammation as measured by maximum apparent diffusion coefficient (spinal ADC max) (B = 0.27; P = .03) and mean ADC (spinal ADC mean) (B = 0.30; P = .004), lower modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS) (B = -0.12; P = .02), more tender joint count (B = 0.12; P = .02), and fewer inflammatory back pain (IBP) (OR = 0.26; P < .001). Axial SpA with late onset had higher clinical disease activity, higher intensity of spinal MRI inflammation, less radiographic damage, and more tender joint count. There was also less inflammatory back pain, which could make the diagnosis more difficult.

Deep learning algorithms for magnetic resonance imaging of inflammatory sacroiliitis in axial spondyloarthritis.

OBJECTIVE: The aim of this study was to develop a deep learning algorithm for detection of active inflammatory sacroiliitis in short tau inversion recovery (STIR) sequence MRI. METHODS: A total of 326 participants with axial SpA, and 63 participants with non-specific back pain (NSBP) were recruited. STIR MRI of the SI joints was performed and clinical data were collected. Region of interests (ROIs) were drawn outlining bone marrow oedema, a reliable marker of active inflammation, which formed the ground truth masks from which 'fake-colour' images were derived. Both the original and fake-colour images were randomly allocated into either the training and validation dataset or the testing dataset. Attention U-net was used for the development of deep learning algorithms. As a comparison, an independent radiologist and rheumatologist, blinded to the ground truth masks, were tasked with identifying bone marrow oedema in the MRI scans. RESULTS: Inflammatory sacroiliitis was identified in 1398 MR images from 228 participants. No inflammation was found in 3944 MRI scans from 161 participants. The mean sensitivity of the algorithms derived from the original dataset and fake-colour image dataset were 0.86 (0.02) and 0.90 (0.01), respectively. The mean specificity of the algorithms derived from the original and the fake-colour image datasets were 0.92 (0.02) and 0.93 (0.01), respectively. The mean testing dice coefficients were 0.48 (0.27) for the original dataset and 0.51 (0.25) for the fake-colour image dataset. The area under the curve of the receiver operating characteristic (AUC-ROC) curve of the algorithms using the original dataset and the fake-colour image dataset were 0.92 and 0.96, respectively. The sensitivity and specificity of the algorithms were comparable with the interpretation by a radiologist, but outperformed that of the rheumatologist. CONCLUSION: An MRI deep learning algorithm was developed for detection of inflammatory sacroiliitis in axial SpA.

Both ASDAS and ADC are associated with spinal mobility in active axial spondyloarthritis: A comparison between early and later disease.

OBJECTIVE: Using diffusion-weighted imaging (DWI)-derived apparent diffusion coefficient (ADC), we aimed to determine the relationship between intensity of spinal inflammation and mobility in patients with axial spondyloarthritis (SpA) in early and later stages of active disease. The Ankylosing Spondylitis Disease Activity Score (ASDAS) was also used for a more comprehensive evaluation. METHODS: Participants with axial SpA and back pain were recruited from 10 rheumatology centers. Clinical, biochemical and radiological parameters were collected. Short tau inversion recovery (STIR) sequence magnetic resonance imaging (MRI) and DWI of the spine and sacroiliac (SI) joints were performed. ADC maps were generated. Participants were examined for Bath Ankylosing Spondylitis Metrology Index (BASMI). Linear regression models were used to determine associations between BASMI and various clinical, radiological, and MRI parameters in participants with active inflammation on spinal ADC maps. RESULTS: One-hundred and twenty-seven participants were included in the analyses. Multivariate linear regression showed that mean ADC spine (ß = .16; P = .03), ASDAS-C-reactive protein (CRP) (ß = .29, P < .001), and ASDAS-erythrocyte sedimentation rate (ESR) (ß = .25, P < .01) were associated with BASMI. In participants with duration of back pain ≤3 years, mean spine ADC (ß = .37; P = .03), ASDAS-CRP (ß = .44; P = .01), and ASDAS-ESR (ß = .42; P = .01) were associated with BASMI after adjustment for confounding factors. In participants with duration of back pain >3 years, only ASDAS-CRP (ß = .25; P < .01) and ASDAS-ESR (ß = .20; P = .20) were associated with BASMI. CONCLUSION: Intensity of inflammation and clinical disease activity were independently associated with impairment of spinal mobility. The associations were stronger in early (≤3 years) than later disease.

Statin-associated anti-3-hydroxy-3-methyl-glutaryl-coenzyme A reductase (HMGCR) myopathy: Imaging findings on thigh-muscle magnetic resonance imaging (MRI) in six patients.

INTRODUCTION/AIMS: Anti-3-hydroxy-3-methyl-glutaryl-coenzyme A reductase (HMGCR) myopathy is a rare but serious complication of statin use. The aim of this study was to describe the imaging features of statin-associated anti-HMGCR myopathy on thigh muscle magnetic resonance imaging (MRI). METHODS: Thigh muscle MRI images of six patients with statin-associated anti-HMGCR myopathy were reviewed for muscle edema, fatty replacement, and fascial edema in four compartments of each thigh: gluteal, anterior, medial, and posterior. Signal intensity ratio (SIR) was calculated in 17 muscles in both thighs on T2-weighted fat-suppressed images. Intracompartmental comparison of T2 SIRs of different muscles were performed. RESULTS: All patients demonstrated bilateral symmetrical pan-compartmental muscle edema. Three patients had fatty infiltration, involving gluteal and/or posterior compartments. In the anterior compartment, rectus femoris and vastus lateralis most frequently demonstrated the highest T2 SIR. In the posterior compartment, semimembranosus most frequently demonstrated the highest T2 SIR. The long head of the biceps femoris always had a higher T2 SIR than the short head. DISCUSSION: Statin-associated anti-HMGCR myopathy commonly demonstrates bilateral symmetrical pan-compartmental edema on thigh muscle MRI, with anterolateral predilection in the anterior compartment, and greater involvement of the semimembranosus with relative sparing of the short head of the biceps femoris in the posterior compartment. These observations can contribute to the diagnosis of statin-associated anti-HMGCR myopathy.

Diagnostic utility of whole spine and thoracic spine MRI corner inflammatory lesions in axial spondyloarthritis.

BACKGROUND: The presence of ⩾3 corner inflammatory lesions has been proposed as the definition of a positive spinal magnetic resonance imaging (MRI) for axial spondyloarthritis (axSpA), but subsequent studies showed inconclusive findings. Our objective was to evaluate whether locations of corner inflammatory lesions (CILs) would affect the diagnostic utility of MRI in axSpA. METHOD: Two groups were consecutively recruited from eight rheumatology centers in Hong Kong. The 'axSpA' group included 369 participants with a known diagnosis of axSpA. The 'non-specific back pain' (NSBP) control group consisted of 117 participants. Clinical, biochemical, and radiological parameters were collected and all patients underwent MRI of the spine and sacroiliac joints. CILs were assessed based on their locations (cervical, thoracic or lumbar) to determine the optimal cutoff for diagnosis. RESULTS: The cutoff of ⩾5 whole spine CILs (W-CILs) and ⩾3 thoracic spine CILs (T-CILs) had comparable specificity to MRI sacroiliitis. Of 85/369 axSpA patients without sacroiliitis on conventional radiograph or MRI, 7 had ⩾5 W-CILs and 11 had ⩾3 T-CILs. Incorporating the proposed cutoffs into Assessment of SpondyloArthritis international Society axSpA criteria, ⩾5 W-CILs and ⩾3 T-CILs had similar performance when added to the imaging criteria for sacroiliitis (sensitivity 0.79 versus 0.80, specificity 0.92 versus 0.91). CONCLUSION: Spinal MRI provided little incremental diagnostic value in unselected axSpA patients. However, in patients without sacroiliitis on MRI or radiographs, 8-13% might be diagnosed by spinal MRI. Thoracic and whole spine MRI had similar diagnostic performance using the proposed cutoff of ⩾5 W-CILs and ⩾3 T-CILs.

Intensity of spinal inflammation is associated with radiological structural damage in patients with active axial spondyloarthritis.

OBJECTIVE: The aim was to investigate the relationship between the intensity of spinal inflammation using the apparent diffusion coefficient (ADC) and radiographic progression in axial SpA. METHODS: This is a cross-sectional study of participants with axial SpA and back pain. Clinical, biochemical and radiological parameters were collected. The ankylosing spondylitis disease activity score (ASDAS)-CRP was determined. Radiographic progression was represented by the modified Stoke ankylosing spondylitis spine score (mSASSS). MRI with short tau inversion recovery (STIR) and diffusion-weighted imaging sequences were performed simultaneously. Inflammatory lesions on STIR were used for the Spondyloarthritis Research Consortium of Canada (SPARCC) MRI indexes and as references in outlining regions of interest in ADC maps to produce mean (ADCmean) and maximal (ADCmax) ADC values. Univariate and multivariate linear regression analyses were used to determine independent associations between ADC and radiographic progression. RESULTS: The 84 participants with identifiable lesions on spinal ADC maps recruited were characterized by a mean (s.d.) age of 45.01 (13.68) years, long disease duration [13.40 (11.01) years] and moderate clinical disease activity [ASDAS-CRP 2.07 (0.83)]. Multivariate regression analysis using ADCmean as the independent variable showed that age (regression coefficient [B] = 0.34; P = 0.01), male sex (B = 0.25; P = 0.04) and ADCmean (B = 0.30; P = 0.01) were positively associated with mSASSS. Multivariate regression analysis using ADCmax as the independent variable showed a tendency for ADCmax to be associated with mSASSS (B = 0.21; P = 0.07). CONCLUSION: The intensity of spinal inflammation as determined by ADC is associated with radiographic progression in participants with active axial SpA.

MRI inflammation of facet and costovertebral joints is associated with restricted spinal mobility and worsened functional status.

OBJECTIVE: To investigate the association of spinal inflammation on MRI in patients with various clinical, functional and radiological outcomes in patients with axial spondyloarthritis (SpA). METHODS: Three hundred and ninety-seven participants with axial SpA and back pain were recruited from 10 rheumatology centres. Clinical, biochemical and radiological parameters were collected and participants underwent MRI of the spine. MRI features including inflammatory lesions of facet joints and costovertebral joints, corner inflammatory lesions, and spondylitis were assessed. BASFI, Bath Ankylosing Spondylitis Disease Activity Index, Bath Ankylosing Spondylitis Global Index, BASMI and modified Stoke Ankylosing Spondylitis Spinal Score were measured. Multivariate linear regression models were used to determine the associations between MRI parameters and various clinical, functional and radiological outcomes. RESULTS: BASMI and BASFI correlated well with inflammatory features in spinal MRI. Multivariate analysis showed that lumbar facet joint inflammation was independently associated with BASMI (regression coefficient (ß) = 0.12, P < 0.001), lumbar spinal flexion (ß = 0.13, P = 0.00), lateral spinal flexion (ß = 0.09, P = 0.04), tragus-to-wall distance (ß = 0.16, P < 0.001) and BASFI (ß = 0.14, P = 0.01). Costovertebral joint inflammation was also associated with BASMI (ß = 0.08, P = 0.05). CONCLUSION: Inflammatory lesions of facet and costovertebral joints in MRI are associated with restriction in spinal mobility and functional impairment. These important yet commonly overlooked lesions should be reviewed in clinical practice in patients with SpA.

Clinical Associations of Uveitis in Axial Spondyloarthritis Group and Ankylosing Spondylitis Group: Do They Represent the Same Disease?

OBJECTIVE: The aim of this study was to determine the prevalence and associated factors for uveitis in ethnic Chinese patients with axial spondyloarthritis (SpA) and ankylosing spondylitis (AS). METHODS: This was a cross-sectional study. Patients fulfilling the Assessment of SpondyloArthritis international Society axial SpA criteria were recruited consecutively from 3 rheumatology centers in Hong Kong from March 2014 to July 2017. Clinical and biochemical parameters were collected. History of uveitis was inquired from both history and medical records. All patients received lumbosacral spine x-rays and whole-spine and sacroiliac joint magnetic resonance imaging. Patients were defined as axial SpA if they fulfilled the Assessment of SpondyloArthritis international Society criteria and AS if they fulfilled the modified New York criteria. Clinical and radiological findings were compared between patients with and without uveitis in the 2 groups. Factors associated with uveitis were identified with univariate analyses and multivariate logistic regression analyses. RESULTS: Among 252 patients, 67 patients (26.6%) had a history of uveitis. The male-to-female ratio was 55.4 to 44.6. Disease duration was 12.3 ± 11.7 years. In the axial SpA group, multivariate regression showed that older age (odds ratio [OR], 1.05; p = 0.01), human leukocyte antigen B27 positivity (OR, 11.79; p = 0.01), and history of inflammatory bowel disease (OR, 9.74; p = 0.04) were positively associated with uveitis. In the AS group, multivariate regression showed that back pain duration (OR, 1.05; p = 0.01) and male sex (OR, 3.46; p = 0.03) were associated with uveitis. CONCLUSIONS: Axial SpA represents a spectrum of diseases. Its clinical associations with uveitis should be distinguished from those of traditional AS.

ASDAS is associated with both the extent and intensity of DW-MRI spinal inflammation in active axial spondyloarthritis.

Objective: To investigate the relationship between Ankylosing Spondylitis Disease Activity Score (ASDAS) and intensity of spinal inflammation measured by apparent diffusion coefficient (ADC) in MRI in participants with active axial spondyloarthritis (SpA). Methods: Participants with axial SpA and back pain were recruited. Clinical, demographic, biochemical and imaging data were collected. ASDAS was calculated based on C reactive protein (CRP) and erythrocyte sedimentation rate (ESR). Inflammatory lesions were identified in short tau inversion recovery images and the corresponding ADC maps to determine the maximum apparent diffusion coefficient (ADCmax), normalised maximum ADC, mean apparent diffusion coefficient (ADCmean) and normalised mean ADC by two independent readers. Spondyloarthritis Research Consortium of Canada (SPARCC) spine and sacroiliac (SI) joint MRI indexes were determined. Univariate and multivariate linear regression models were used to determine the associations between of ASDAS with ADC values, SPARCC spine and SI MRI scores. Results: Eighty-two participants had identifiable ADC lesions. Multivariate analyses using ADCmax and SPARCC spine MRI as independent variables showed associations with ASDAS-CRP (ADCmax: B=0.27, p=0.02; SPARCC: B=0.32, p=0.01) and ASDAS-ESR (ADCmax: B=0.24, p=0.03; SPARCC: B=0.36, p<0.01); using ADCmean and SPARCC spine MRI as independent variables also showed an association with ASDAS-ESR (ADCmean: B=0.22, p=0.05; SPARCC: B=0.36, p<0.01) and a tendency to associate with ASDAS-CRP (ADCmean: B=0.21, p=0.07; SPARCC: B=0.34, p<0.01). Conclusion: ASDAS is associated with both the extent and the intensity of spinal inflammation in patients with detectable inflammatory lesions. Our results showed that ASDAS is an objective disease assessment tool. Trial registration number: HKUCTR-2087.

Fatty corner lesions in T1-weighted magnetic resonance imaging as an alternative to sacroiliitis for diagnosis of axial spondyloarthritis.

BACKGROUND: A fatty corner lesion (FCL) is a well-demarcated fat infiltration in the corner of a vertebral body on T1 magnetic resonance imaging (MRI) sequence. It has been reported to be useful in the diagnosis of axial spondyloarthritis (axSpA). Our objective is to systematically evaluate the diagnostic accuracy of FCLs in tertiary centre patients with chronic back pain. METHOD: Two hundred and thirty eight axSpA patients and 62 non-axSpA patients with back pain were recruited from five rheumatology centres. Clinical, biochemical, and radiological parameters were collected and all patients underwent a MRI of the spine and sacroiliac (SI) joints. FCLs in vertebral bodies from C4 to L5 were scored. The number and location of FCLs were clustered together to determine an optimal combination for diagnosis. Results were compared with expert diagnosis as the "gold standard". RESULTS: FCLs of the anterior whole spine (AUC 0.62; p = 0.003) and anterior thoracic spine (AUC 0.64; p = 0.001) had diagnostic significance. Incorporating at least 5 whole spine FCLs into the imaging criteria of the Assessment of SpondyloArthritis international Society (ASAS) criteria for axSpA yielded a sensitivity of 91.6% and specificity of 91.9%. Similarly, applying at least 3 anterior thoracic FCLs to the imaging criteria of the ASAS axial SpA criteria yielded a sensitivity of 92.0% and specificity of 93.5%. CONCLUSION: FCLs could be used to diagnose axial SpA. The presence of at least 3 anterior thoracic FCLs in T1-weighted MRI spine suggests a diagnosis of axial SpA without additional MRI of the SI joints. TRIAL REGISTRATION: The cohort has been registered in the clinical trial registry of the University of Hong Kong (HKUCTR-2087).

Apparent Diffusion Coefficient as an Imaging Biomarker for Spinal Disease Activity in Axial Spondyloarthritis.

Background A quantifiable imaging measure to gauge the intensity of individual inflammatory lesions in axial spondyloarthritis (SpA) has not been well established. Previous studies have shown that diffusion-weighted (DW) MRI reflects disease activity in axial SpA. Purpose To determine the association between apparent diffusion coefficient (ADC) at MRI of discovertebral lesions and disease activity in individuals with axial SpA. Materials and Methods In this prospective study, 243 study participants (mean age ± standard deviation, 43.2 years ± 13.5) with back pain who fulfilled the Assessment of SpondyloArthritis International Society criteria for SpA were recruited from four rheumatology centers between April 2014 and March 2018. There were 132 men (mean age, 41.4 years ± 13.3) and 111 women (mean age, 45.3 years ± 13.4). Clinical, biochemical, and radiologic parameters were collected. All participants underwent whole-spine MRI by using a short inversion time inversion-recovery sequence and DW imaging. Two independent readers identified the presence of discovertebral lesions. ADCs were measured and normalized with normal bone marrow. Regression analysis was performed to determine association between the mean, maximum, and normalized mean and maximum ADCs of the discovertebral lesions and disease activity and functional parameters (Bath Ankylosing Spondylitis Disease Activity Index [BASDAI], Bath Ankylosing Spondylitis Functional Index [BASFI], and Bath Ankylosing Spondylitis Global Index [BASGI]). Results Ninety-one discovertebral lesions (five cervical, 61 thoracic, 25 lumbar) were present in 55 of the 243 study participants (22.6%). After adjusting for confounding factors, increased maximum ADC was independently associated with increased BASFI (regression coefficient [ß] = 1.94 [×10-3 mm2/sec], P = .04). Increased normalized maximum ADC was independently associated with BASDAI question 2 (ie, back pain score) (ß = 0.45, P = .01), mean stiffness score (ß = 0.41, P = .04), and BASGI (ß = 0.43, P = .04). Increased normalized mean ADC was independently associated with BASDAI question 2 (ß = 0.61, P = .04). Conclusion Apparent diffusion coefficients at MRI of discovertebral lesions were associated with disease activity, functional impairment, and patient global assessment in axial spondyloarthritis. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Guermazi and Roemer in this issue.

Ultrasound-MRI Fusion for Targeted Biopsy of Myopathies.

OBJECTIVE. The purpose of this article is to describe the use of ultrasound-MRI fusion imaging to guide precise and targeted muscle biopsy in patients with suspected myopathies. CONCLUSION. Ultrasound-MRI fusion-guided muscle biopsy allows targeted sampling of tissues with active inflammatory changes and facilitates diagnosis of myopathies.

Development and validation of an imaging and clinical scoring system to predict early mortality in spontaneous ruptured hepatocellular carcinoma treated with transarterial embolization.

PURPOSE: To develop and validate a scoring system using a combination of imaging and clinical parameters to predict 30-day mortality in ruptured HCC (rHCC) patients after transarterial embolization (TAE). METHODS: 98 consecutive patients with rHCC who underwent abdominal CT and subsequent TAE between January 2007 and December 2016 were retrospectively reviewed. The CT scans were reviewed by two radiologists blinded to the patient outcome. Clinical parameters including serum bilirubin, albumin, INR, creatinine, and hemoglobin were recorded. Independent risk factors for 30-day mortality after TAE were identified using multivariate binary logistic regression, for development of a scoring system. The scoring system was then validated in 20 patients between January 2017 and May 2018. RESULTS: In the development cohort, bilobar tumor distribution (OR = 29.6), clinical parameters of bilirubin > 2.5 mg/dL (OR = 5.9), and albumin < 30 g/L (OR = 4.1) were independent predictors for 30-day mortality. A 6-point score was derived and yielded area-under-the-receiver-operating-characteristic-curve (AUC) of 0.904. A score ≥ 4 resulted in sensitivity of 80.5% and specificity of 91.2% for 30-day mortality. In the validation cohort, AUC for 30-day mortality was 0.939. A score ≥ 4 resulted in sensitivity of 81.2% and specificity of 88.9%. In both development and validation cohorts, the proposed scoring system was better than biochemical components of Child-Pugh score and serum bilirubin to predict 30-day mortality. CONCLUSION: Imaging and clinical parameters can be combined into a scoring system to accurately predict 30-day mortality after TAE in rHCC patients. The score may help identify and counsel high-risk patients.