Does evaluation of the ligamentous compartment enhance diagnostic utility of sacroiliac joint MRI in axial spondyloarthritis?

Machine Learning Pipeline for Predicting Bone Marrow Edema Along the Sacroiliac Joints on Magnetic Resonance Imaging

OBJECTIVE

We aimed to develop and validate a fully automated machine learning (ML) algorithm that predicts bone marrow edema (BME) on a quadrant level in sacroiliac (SI) joint magnetic resonance imaging (MRI).

METHODS

A computer vision workflow automatically locates the SI joints, segments regions of interest (ilium and sacrum), performs objective quadrant extraction, and predicts presence of BME, suggestive of inflammatory lesions, on a quadrant level in semicoronal slices of T1/T2-weighted MRI scans. Ground truth was determined by consensus among human readers. The inflammation classifier was trained using a ResNet18 backbone and five-fold cross-validated on scans of patients with spondyloarthritis (SpA) (n = 279), postpartum individuals (n = 71), and healthy subjects (n = 114). Independent SpA patient MRI scans (n = 243) served as test data set. Patient-level predictions were derived from aggregating quadrant-level predictions, ie, at least one positive quadrant.

RESULTS

The algorithm automatically detects the SI joints with a precision of 98.4% and segments ilium/sacrum with an intersection over union of 85.6% and 67.9%, respectively. The inflammation classifier performed well in cross-validation: area under the curve (AUC) 94.5%, balanced accuracy (B-ACC) 80.5%, and F1 score 64.1%. In the test data set, AUC was 88.2%, B-ACC 72.1%, and F1 score 50.8%. On a patient level, the model achieved a B-ACC of 81.6% and 81.4% in the cross-validation and test data set, respectively.

CONCLUSION

We propose a fully automated ML pipeline that enables objective and standardized evaluation of BME along the SI joints on MRI. This method has the potential to screen large numbers of patients with (suspected) SpA and is a step closer towards artificial intelligence-assisted diagnosis and follow-up.

Advances in the Evaluation of Peripheral Enthesitis by Magnetic Resonance Imaging in Patients With Psoriatic Arthritis

Enthesitis is a key disease manifestation in patients with psoriatic arthritis (PsA) that considerably contributes to pain, lower physical function, and reduced quality of life. Clinical assessment of enthesitis lacks sensitivity and specificity, and therefore better methods are urgently needed. Magnetic resonance imaging (MRI) allows detailed assessment of the components of enthesitis, and consensus-based validated MRI scoring systems exist. These include the Outcome Measures in Rheumatology (OMERACT) Heel Enthesitis MRI Scoring System (HEMRIS) method, which assesses the entheses of the heel region in a detailed manner, and the OMERACT MRI Whole-Body Score for Inflammation in Peripheral Joints and Entheses (MRI-WIPE) method, which provides an overall assessment of the inflammatory burden in the peripheral entheses and joints in the entire body using whole-body MRI. At an MRI workshop at the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) 2022 meeting in Brooklyn, the MRI appearances of peripheral enthesitis were described, as were the scoring methods. The utility of MRI for improved assessment of enthesitis was demonstrated with examples of patient cases. Clinical trials in PsA that evaluate enthesitis by MRI as a key endpoint should include the presence of MRI enthesitis as an inclusion criterion, and apply validated MRI outcomes to assess the effect of therapeutics on enthesitis are recommended.

The bone marrow side of axial spondyloarthritis

Spondyloarthritis (SpA) is characterized by the infiltration of innate and adaptive immune cells into entheses and bone marrow. Molecular, cellular and imaging evidence demonstrates the presence of bone marrow inflammation, a hallmark of SpA. In the spine and the peripheral joints, bone marrow is critically involved in the pathogenesis of SpA. Evidence suggests that bone marrow inflammation is associated with enthesitis and that there are roles for mechano-inflammation and intestinal inflammation in bone marrow involvement in SpA. Specific cell types (including mesenchymal stem cells, innate lymphoid cells and γδ T cells) and mediators (Toll-like receptors and cytokines such as TNF, IL-17A, IL-22, IL-23, GM-CSF and TGFβ) are involved in these processes. Using this evidence to demonstrate a bone marrow rather than an entheseal origin for SpA could change our understanding of the disease pathogenesis and the relevant therapeutic approach.

Axial Skeleton Bone Marrow Changes in Inflammatory Rheumatologic Disorders

Magnetic resonance imaging (MRI) of the axial skeleton, spine, and sacroiliac (SI) joints is critical for the early detection and follow-up of inflammatory rheumatologic disorders such as axial spondyloarthritis, rheumatoid arthritis, and SAPHO/CRMO (synovitis, acne, pustulosis, hyperostosis, and osteitis/chronic recurrent multifocal osteomyelitis). To offer a valuable report to the referring physician, disease-specific knowledge is essential. Certain MRI parameters can help the radiologist provide an early diagnosis and lead to effective treatment. Awareness of these hallmarks may help avoid misdiagnosis and unnecessary biopsies. A bone marrow edema-like signal plays an important role in reports but is not disease specific. Age, sex, and history should be considered in interpreting MRI to prevent overdiagnosis of rheumatologic disease. Differential diagnoses-degenerative disk disease, infection, and crystal arthropathy-are addressed here. Whole-body MRI may be helpful in diagnosing SAPHO/CRMO.

Differences in topographical location of sacroiliac joint MRI lesions in patients with early axial spondyloarthritis and mechanical back pain

Background

Early diagnostics of axial spondyloarthritis (axSpA) remains a challenge. Traditional imaging one-plane sacroiliac joint (SIJ) MRI assessment is used. By introducing a two-plane assessment system, the objective was to analyse the differences in SIJ MRI changes in early axSpA compared with changes in patients with mechanical back pain (MBP) by exploring the differences in volume and location.

Methods

MRIs in the early diagnostic state of 25 axSpA patients (mean age 31.3 years) and 59 MBP patients (mean age 32.3 years) were included. The MRIs were assessed by two readers regarding the distribution of bone marrow edema (BME) in 14 joint portions and structural changes in six joint portions in addition to SIJ anatomical variations and lumbar spine disc degeneration.

Results

AxSpA patients had a significantly higher overall BME sumscore (volume) of 25.1 compared to MBP patients 6.8, p < 0.005. The MBP group had the highest prevalence (66%) and sumscore (5.7) in the middle anterior sacrum. The axSpA group had significantly higher prevalence and sumscores in all joint portions except the three cartilaginous anterior sacral joint portions, including the ligamentous compartments (prevalence 40–60% compared to 8–15%, p both < 0.005). The axSpA group had also a significantly higher prevalence of erosions and fatty marrow disposition, but there were no differences in the prevalence of anatomical variations except the bipartite iliac bony plate.

Conclusions

AxSpA patients demonstrated a widespread distribution of both inflammatory and structural changes, including high BME occurrence in the ligamentous joint portions whereas patients with MBP had the highest occurrence of BME in the middle anterior sacrum. These findings may help differentiate axSpA patients from other back pain conditions in the early diagnostic phase.

Australian Consensus Statements for the Assessment and Management of Non-radiographic Axial Spondyloarthritis

Background

The understanding of non-radiographic axial spondyloarthritis (nr-axSpA) has accelerated over the last decade, producing a number of practice-changing developments. Diagnosis is challenging. No diagnostic criteria exist, no single finding is diagnostic, and other causes of back pain may act as confounders.

Aim

To update and expand the 2014 consensus statement on the investigation and management of non‐radiographic axial spondyloarthritis (nr-axSpA).

Methods

We created search questions based on our previous statements and four new topics then searched the MEDLINE and Cochrane databases. We assessed relevant publications by full-text review and rated their level of evidence using the GRADE system. We compiled a GRADE evidence summary then produced and voted on consensus statements.

Results

We identified 5145 relevant publications, full-text reviewed 504, and included 176 in the evidence summary. We developed and voted on 22 consensus statements. All had high agreement. Diagnosis of nr-axSpA should be made by experienced clinicians, considering clinical features of spondyloarthritis, blood tests, and imaging. History and examination should also assess alternative causes of back pain and related conditions including non-specific back pain and fibromyalgia. Initial investigations should include CRP, HLA-B27, and AP pelvic radiography. Further imaging by T1 and STIR MRI of the sacroiliac joints is useful if radiography does not show definite changes. MRI provides moderate-to-high sensitivity and high specificity for nr-axSpA. Acute signs of sacroiliitis on MRI are not specific and have been observed in the absence of spondyloarthritis. Initial management should involve NSAIDs and a regular exercise program, while TNF and IL-17 inhibitors can be used for high disease activity unresponsive to these interventions. Goals of treatment include improving the frequent impairment of social and occupational function that occurs in nr-axSpA.

Conclusions

We provide 22 evidence-based consensus statements to provide practical guidance in the assessment and management of nr-axSpA.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40744-021-00416-7.

Imaging in Axial Spondyloarthritis: What is Relevant for Diagnosis in Daily Practice?

PURPOSE OF REVIEW

To explore how imaging may assist diagnosing axial spondyloarthritis in rheumatology practice.

RECENT FINDINGS

A diagnosis of axial spondyloarthritis is based on pattern recognition by synthesizing clinical, laboratory, and imaging findings. In health care settings providing low threshold access to advanced imaging, sacroiliac joint MRI is the preferred imaging modality in clinically suspected axial spondyloarthritis. In daily routine, the optimum protocol to assess suspected inflammatory back pain combines sacroiliac joint and spine MRI fitting a 30-min slot. Contextual assessment of concomitant structural and active MRI lesions is key to enhance diagnostic utility. In women with postpartum back pain suggestive of axial spondyloarthritis, recent reports advocate waiting 6-12 months after delivery before acquiring sacroiliac joint MRI. Major unmet needs are consistent MRI protocols, standardized training modules on how to evaluate axial MRI, and timely dissemination of imaging advances into mainstream practice both in rheumatology and in radiology. In rheumatology practice, MRI has become indispensable to help diagnose early axial spondyloarthritis. However, major gaps in training and knowledge transfer to daily care need to be closed.

Morphological characteristics of sacroiliac joint MRI lesions in axial spondyloarthritis and control subjects

OBJECTIVES

To investigate sacroiliac joint(SIJ) MRI inflammation, structural and degenerative lesion characteristics in patients with axial spondyloarthritis(axSpA) and various control groups.

METHODS

Patients with axSpA(n = 41) and lumbar disc herniation(n = 25), women with(n = 46) and without(n = 14) post-partum(childbirth within 4-16 months) buttock/pelvic pain, cleaning assistants(n = 26), long-distance runners(n = 23) and healthy men(n = 29) had MRI of the SIJs prospectively performed. MRI lesions were assessed on 9 slices covering the cartilaginous compartment by two experienced readers according to the definitions of the Spondyloarthritis Research Consortium of Canada(SPARCC) SIJ inflammation and structural scores, and were evaluated according to depth and extent. Other morphological characteristics were also analysed.

RESULTS

Total depth scores for bone marrow oedema(BME) and fat lesion(FAT) and total extent score for erosion were statistically significantly highest in axSpA, while scores for sclerosis were numerically highest in women with post-partum pain. Maximum BME depth > 10mm was frequently and exclusively found in axSpA and post-partum women(39% vs 14-17%) while FAT depth > 5mm was predominantly found in axSpA(76% vs 0-10%). Erosions were primarily seen in axSpA, especially when extensive(≥4 or confluent; 17% vs 0%). Capsulitis was absent in non-axSpA groups. BME and FAT in the ligamentous compartment were primarily found in axSpA(17/22% vs 0/2% in non-axSpA groups). In non-axSpA, osteophytes(axSpA vs non-axSpA: 0% vs 3-17%) and vacuum phenomenon(7% vs 30-66%) were more frequent, and the joint space was wider(mean(SD) 1.5(0.9)mm vs 2.2(0.5)mm).

CONCLUSIONS

FAT depth > 5mm, but not BME depth > 10mm, could almost differentiate axSpA patients from all other groups. When excluding post-partum women, BME >5mm and erosion were highly specific for axSpA.

Performance of magnetic resonance imaging in the diagnosis of axial spondyloarthritis: a systematic literature review

OBJECTIVES

To summarize the evidence on the performance of MRI for the diagnosis of axial SpA.

METHODS

This was a systematic literature review of all studies from January 2013 to March 2017 including adult patients with clinically suspected axial SpA undergoing MRI. Studies from a previously published systematic literature review up to January 2013 were also included.

RESULTS

Thirty-one studies were included. Six studies demonstrated good sensitivity and specificity for SI joint (SIJ) bone marrow oedema (BMO). Specificity was increased by the presence of other structural lesions alongside BMO, particularly erosions or fat infiltration. Four studies addressed the utility of SIJ fat infiltration, finding good sensitivity but poor specificity. SIJ erosions showed good specificity in five studies. Studies addressing high T1 signal in the SIJ, fluid signal in the SIJ, ankylosis, sclerosis, capsulitis, backfill and vacuum phenomenon reported limited diagnostic value. In the spine, four studies reported moderate sensitivity and specificity for corner inflammatory lesions, and four reported poor sensitivity and specificity for spinal fat infiltration. Five studies evaluated the added value of spinal MRI over SIJ MRI alone, with variable results depending on the cohort. Six studies addressed the effect of acquisition parameters on diagnostic accuracy: fat-saturated T2-weighted imaging and short tau inversion recovery (STIR) imaging showed comparable utility in identifying BMO. Three studies showed that gadolinium was of minimal added value in the detection of BMO.

CONCLUSIONS

These results confirmed the diagnostic utility of MRI in axial SpA. Performance varied according to the characteristics of the cohort and the number and combination of MRI lesions considered.

Imaging of sacroiliitis: Current status, limitations and pitfalls

The clinical need to diagnose sacroiliitis at an earlier stage has led to the sacroiliac joints being more frequently imaged, particularly with magnetic resonance imaging (MRI). This review outlines the imaging approach to sacroiliitis, emphasizing the imaging protocols, diagnostic criteria, limitations and potential mimics of MRI examination. The value of imaging-guided intervention in sacroiliac joint disease is also briefly outlined.

Diffusion-weighted imaging versus short tau inversion recovery sequence: Usefulness in detection of active sacroiliitis and early diagnosis of axial spondyloarthritis

Objective

To compare the utility of Diffusion weighted imaging (DWI) with short tau inversion recovery (STIR) sequence in the diagnosis of early axial spondyloarthritis (SpA).

Methods

Three hundred and five patients with chronic back pain were recruited consecutively from 3 rheumatology centers. Clinical, radiological and blood parameters were recorded. Patients with back pain duration no more than 3 years were classified as having early disease. STIR sequence and DWI of the sacroiliac joints were obtained and assessed using the Spondyloarthritis Research Consortium of Canada (SPARCC) method. The Assessment in Spondyloarthritis international Society definition was used to define positive STIR and DWI. Results were compared to expert diagnosed axial SpA.

Results

When compared to STIR sequence, DWI had similar sensitivity (STIR 0.29, DWI 0.30) and specificity (STIR 0.97, DWI 0.92) in diagnosing sacroiliitis. However, STIR sequence had better reliability (STIR 0.78, DWI 0.61). In early disease group, DWI was not better than STIR sequence in detecting active sacroiliitis (sensitivity DWI vs STIR: 0.34 vs 0.36; specificity DWI vs STIR: 0.93 vs 0.93; positive predictive value DWI vs STIR: 0.92 vs 0.92; negative predictive value DWI vs STIR: 0.36 vs 0.37). Using the Assessment in SpondyloArthritis international Society (ASAS) classification criteria, 67/98 patients with early disease (sensitivity 0.91 specificity 0.90) and 221/305 overall (sensitivity 0.90; specificity 0.92) were classified as axial SpA. Among the expert diagnosed axial SpA patients who did not meet the ASAS criteria, only 2 had positive DWI.

Conclusion

DWI and STIR have similar sensitivity in diagnosing axSpA in early disease. However, the use of DWI is limited by poorer reliability when compared with STIR.

The Role of Imaging in Diagnosing Axial Spondyloarthritis

Imaging has a central role in the diagnosis, management, and follow-up of patients with axial spondyloarthritis (axSpA). For the early diagnosis of axSpA, magnetic resonance imaging is of utmost relevance. While no novel imaging techniques were developed during the past decade, improvements to the existing modalities have been introduced. This report provides an overview of the applications and limitations of the existing imaging modalities.

The role of MRI in the evaluation of spondyloarthritis: a clinician's guide

Plain radiography remains the benchmark for diagnostic evaluation of spondyloarthritis although MRI has much great sensitivity because it can detect a variety of inflammatory lesions as well as fat metaplasia. So, it is the imaging modality of choice when pelvic radiographs are equivocal and especially when important treatment decisions have to be made. Appropriate imaging includes the use of T1-weighted and short tau inversion recovery sequences of the sacroiliac joint in the tilted coronal plane. If there is localized spinal symptomatology, sagittal scans of the spine may also be helpful. However, routine spinal imaging is not recommended. Current consensus designates a positive MRI for classification purposes as requiring the presence of two definite subchondral inflammatory lesions on a single coronal slice or the presence of one such lesion on two consecutive coronal slices. However, such inflammatory lesions can occur in healthy individuals and in those with nonspecific back pain. Erosions are more specific, and their presence can enhance confidence in the diagnosis. MRI, together with CRP, can be helpful in selecting which patients without radiographic sacroiliitis are most likely to respond to tumor necrosis factor inhibitor therapy. The role of MRI in monitoring of patients with SpA remains unclear although it may be helpful in excluding other sources of back pain. A major unanswered question is whether MRI may be useful in predicting relapse following withdrawal of TNFi in patients who have achieved sustained remission.