Abstract
Takayasu arteritis is an idiopathic large-vessel vasculitis that affects young adults and children and can lead to ischemia and end-organ damage. Vascular imaging is crucial for diagnosis, assessment of disease extent, and management of the disease. Here we critically review evidence for the clinical use of the different imaging modalities: conventional angiography, magnetic resonance imaging, computed tomography, Doppler ultrasound, and 18fluorodeoxyglucose positron emission tomography. We thereby focus on their clinical applicability, challenges, and specific use in children.
Takayasu arteritis (TA) is an idiopathic granulomatous vasculitis of the aorta, its major branches, and the pulmonary arteries. TA predominantly affects young women under the age of 40 years. Disease onset in children is far less frequent but observed in up to 20–30% of patients.1,2,3 Clinical presentation is highly variable and results from systemic and local vascular inflammation as well as organ dysfunction secondary to ischemia. Most patients have a relapsing remitting disease course, and eventually achieve remission after a variable disease duration. TA disease burden is high and mortality substantial.4,5
The precise etiology of TA remains poorly understood. Current knowledge is extrapolated from mouse models, surgery specimens from patients with TA, and genome-wide association studies.6 Evidence suggests that both the innate and adaptive immune systems are involved in the pathogenesis of TA.7 Genetic studies reveal an association with classes I and II HLA loci, in particular the HLA-B52 allele, as well as genes involved in immune-regulatory and inflammatory pathways.7,8 Inflammatory infiltrates, consisting of macrophages and lymphoid cells, expand through the vasa vasorum to the adventitia and media, possibly affecting all 3 tunics.9 This leads to vessel wall edema and degeneration of smooth muscle and elastic components with laminar necrosis and elastic fiber fragmentation, and ultimately to fibrosis and arterial remodeling.10 Macroscopically, these processes are reflected by wall thickening and may result in arterial stenosis, dilatation, or aneurysm formation, all of which are again directly reflected by the clinical features with strong effect on prognosis.
The diagnosis of TA is based on clinical criteria and abnormalities on vascular imaging and is supported by laboratory findings. Characteristic vascular abnormalities include circumferential wall thickening, multiple arterial lesions, and/or luminal abnormalities such as stenosis or aneurysms. Occlusion and occasionally arterial dissection may be seen. Lesions are typically found close to the origin of the aortic branches and are often segmental with a patchy distribution.11 Given the important role of imaging in TA, efforts were undertaken to describe angiographic patterns.12,13,14,15 Various angiographic patterns with distinct clinical presentation have been identified in different ethnicities.11,12,13,15 However, these angiography-based classifications of TA still require validation regarding outcome prediction.
Therapeutic management consists of medical and vascular interventions. Corticosteroids remain the mainstay of immunosuppressive medication.16 As relapses are frequent, the use of corticoid-sparing agents, including conventional disease-modifying antirheumatic drugs, such as methotrexate, azathioprine, mycophenolate mofetil, and cyclophosphamide, for severe disease has been recommended upfront.17,18 Recent advances in the understanding of underlying disease pathophysiology led to the increasing use of pathway targeting agents such as tumor necrosis factor, interleukin (IL)-6, or Janus kinase inhibitors with improved TA disease control, particularly in treatment-refractory patients.17,19,20 Revascularization may be indicated to treat symptomatic organ ischemia or life-threatening vascular lesions.10
Imaging is required for diagnosis, assessment of disease extent, and management of TA. Distinguishing between acute active and chronic inactive disease is crucial to guide clinical care and allow for personalized treatment with appropriate tapering of immunosuppressive therapy. A delicate balance should be pursued between prevention of irreversible damage resulting from uncontrolled inflammation and the exposure to the potentially life-threatening side effects from unnecessary immunosuppressive therapy. However, correct assessment of TA disease activity remains challenging. Symptoms may be nonspecific (ie, headaches) and can also result from prior vascular damage rather than active inflammation. Biomarkers that accurately assess TA disease activity are lacking. Acute-phase reactants (APRs) such as C-reactive protein and erythrocyte sedimentation rate are routinely used in clinical practice; however, they are nonspecific and may not allow for differentiation between clinically active and inactive disease.21 In addition, APRs rapidly normalize after treatment with IL-6 inhibitors, further complicating assessment of TA disease activity in tocilizumab-treated patients.22
Recent advances in radiological imaging have the potential to improve the management of TA by adequate and correct visualization of all arterial sequalae and by detection of early vascular inflammatory changes. TA disease state changes should be detected to facilitate prompter diagnosis and recognition of disease flares. However, the value of subtle imaging anomalies remains unclear and may result in overtreatment. More accurate monitoring of disease activity would allow for appropriate tailoring of immunosuppressive therapy. Given the lifelong and serial follow-up required, especially in children, the optimal imaging modality for the diagnosis and management of TA should be noninvasive and nonirradiating, easily accessible, and inexpensive, allowing for repeat exams and comparisons. It should quantify disease extent and accurately assess disease activity as well as treatment response.
The aim of this review is to provide an update on the available imaging methods in adult- and childhood-onset TA including conventional angiography (CA), magnetic resonance imaging (MRI), computed tomography (CT), Doppler ultrasound (US), and 18fluorodeoxyglucose positron emission tomography (18F-FDG PET), and to discuss their advantages and challenges (Table 1). The terms MRI and CT also encompass the specific angiography techniques MR angiography and CT angiography.
Methods
Literature search was conducted using PubMed combining the following search terms: “Takayasu arteritis,” “large vessel vasculitis,” “childhood,” “pediatric,” “imaging,” “angiography,” “MRI,” “CT,” “US,” and “18F-FDG PET.” All types of articles (full papers, review articles, recommendations, case reports, abstracts) were looked at and screened for relevance. An article was considered relevant if it included data on clinical presentation, imaging modality, and interpretation. Additional articles were identified through the reference lists from the initial search and authors’ knowledge of the literature. Supplementary Table 1 (available with the online version of this article) provides information about the quality of the cited studies.
Conventional angiography
CA (intraarterial digital subtraction angiography) provides images of the arterial lumen (Figure 1). Its strengths include an excellent spatial resolution and the possibility of visualization of any collateral vascularization.23,24 CA may also be useful to accurately measure central artery pressure in the presence of arterial stenosis in all extremities. Data on CA in children are scarce and limited to descriptions of imaging modalities within case series or case reports of childhood-onset TA.25–33
Historically, CA is still considered the gold standard for evaluation of the arterial vasculature, but as the arterial wall is not visualized, early inflammatory disease changes may therefore be missed and diagnosis delayed.34 The lack of information about the vascular wall also increases the difficulty to distinguish TA from other diseases causing vascular narrowing, such as chronic wall fibrosis, leaving only pattern recognition of affected arterial segments as being supportive of TA diagnosis.35 Finally, CA is invasive, with high exposure to radiation and risk of potential procedural complications. As a result of these disadvantages and the wide availability of MRI and CT, CA is not recommended in clinical routine and is usually restricted to angiographic imaging prior to revascularization procedures, especially in children.18,23,25,29,36 Overall, quality of evidence regarding CA in TA is low, as most data in both adults and children originate from retrospective cohorts or case series/reports.
Magnetic resonance imaging
MRI provides valuable information on disease extent in all vascular territories by visualization of luminal abnormalities (stenosis, occlusion, aneurysm; Figure 2 and Figure 3). MRI allows also for depiction of arterial wall lesions (thickening, edema). Principally used sequences in TA are T1- and T2-weighted fast spin echo (black blood imaging) and delayed contrast-enhanced sequences (Figure 2). In black blood imaging using inversion recovery technique, the inversion pulse nulls the blood so there is enough difference between the T1 or T2 signal of myocardium and blood for an optimal visualization of the myocardium and vascular walls.
T2-weighted imaging is sensitive to water content and depicts mural edema that may appear as a bright T2 signal37,38 in higher contrast with the black aspect of blood. T1-weighted imaging also allows visualization of the thickened intima and media, permitting evaluation of TA disease extent and quantification of vessel stenosis and dilation (Figure 4). Delayed contrast enhancement reflects increased vascularity and/or excessive leakage of contrast out of the vasa vasorum, suggestive of more acute disease and allowing for identification of inflammation and/or fibrosis in the arterial wall. Owing to its capability to detect signs of vascular inflammation such as arterial wall thickening and edema, MRI permits recognizing TA in the early prestenotic phase.
However, its utility for monitoring disease activity during follow-up is less clear. Although disease activity on contrast-enhanced MRI has been shown to correlate with clinical disease status and APRs in some small prospective and retrospective case series,37,39,40,41 it remains difficult to differentiate active from inactive disease on MRI as specificity of the presence of edema or postcontrast arterial enhancement for active disease is still debated.42,43 Tso et al prospectively evaluated 77 MRI studies with standardized T2-weighted sequences from 24 patients with TA and found a high frequency of vessel wall edema not only in patients considered clinically active (MR images consistent with edema in 94% of patients) but also in > 50% of patients considered clinically inactive or having an uncertain clinical disease activity status.43 Some concerns were raised about results because the study protocol did not require a contrast agent. However, similar findings were presented in a more recent prospective, observational cohort directly comparing standardized MRI and 18F-FDG PET imaging with blinded clinical disease activity assessment in 65 patients with large-vessel vasculitis (LVV), of whom 30 were with TA.44 Notably, in 51% of studies, both MRI and PET were interpreted as active imaging while patients were considered in clinical remission. This raises the question of whether the imaging abnormalities observed in clinically inactive patients represent subclinical vascular inflammation or nonspecific vascular remodeling.44 A further challenge regarding interpretation is the ongoing difficulty to adequately assess disease activity in TA because reliable outcome measures of ongoing vessel inflammation are lacking.
Clinical disease activity scores and biological biomarkers have their limitations; thus, lack of correlation between imaging and clinico-biological disease activity assessment may also point to the limitations related to these latter ones. In addition, the appearance of vascular lesions on imaging is usually delayed compared to biological inflammation. To make it even more complicated, there have been reports of patients who were considered in clinical remission and inactive on imaging but were found to have histologic evidence of ongoing active vasculitis (from vascular surgery or autopsy).22,45 Thus, correlation with histopathology would help to address these issues, but these findings are rarely available.22,43
The value of delayed contrast enhancement (DCE) in identifying patients with active or stable TA was previously reported in an observational cohort study using standardized imaging protocols.46 While DCE of the arterial wall (aortic vessel or pulmonary artery) could be observed in patients with active TA, it was not observed in patients with TA with stable disease. Moreover, stenotic lesions were comparable in active and stable patients with TA, but DCE was more present in patients with active than those with stable disease, demonstrating the potential of DCE to detect arterial vessel compromise before anatomical changes occur. These results are inspiring for future studies about the value of dynamic contrast-enhanced techniques in early identification of active phases of the illness and for subsequent monitoring of treatment effects.46
In children, data on MRI are limited to retrospective cohorts and case reports describing the use of MRI for diagnosis and follow-up monitoring of pediatric TA.47–52 Aluquin et al published a study on 3 children with both clinically and radiologically active TA at diagnosis. Follow-up MRI after 7 and 12 months showed reduced wall thickness and DCE in the 2 patients with clinical amelioration, while the third patient, who presented ongoing fatigue and limp claudication, did not demonstrate any improvement on MRI.53 The authors concluded that MRI findings correlated with clinical and serological disease activity, even though they noted that imaging improvement lagged 1 year behind clinical response to treatment in 1 child.53 This likely reflects the difficulty of disease activity assessment during follow-up MRI, similar to that in the adult population.
MRI may be used during pregnancy; however, gadolinium has been shown to cross the placenta and its long-term effects on the fetus are currently unknown.54
Disadvantages of MRI include reduced visualization of smaller vessels and short focal lesions, and potential overestimation of the severity of arterial narrowing, all of which are related to limitations of MRI spatial resolution.55 Arterial wall calcifications can also be missed.55 CT may then be advantageous given its improved spatial resolution and ability to depict calcifications. Pseudostenosis, an MRI artifact mimicking arterial stenosis, may be identified in the distal subclavian artery.56 Practical limitations are the high costs, duration of the procedure, and the requirement of contrast media and sedation in young children, all of which restrict availability, especially in low-income countries.57 Despite these limitations, the lack of invasiveness and of radiation delivery makes MRI highly appealing for serial assessments of TA.58 Hence, it is currently widely used in TA (especially in children) and has been proposed as the first imaging modality of choice for suspected TA by the recent European Alliance of Associations for Rheumatology recommendations on imaging of LVV36 and the European consensus-based recommendations for the diagnosis and treatment of rare pediatric vasculitides.18
Computed tomography
Similar to MRI, CT evaluates anatomical changes of the arterial lumen and wall and localizes the extent of the vascular lesions with good spatial resolution24,59 (Figure 1). Precontrast CT may demonstrate concentric arterial thickening, which appears attenuated compared with the lumen, and mural calcifications.60 While transmural aortic calcifications are characteristic of chronic TA, calcifications limited to the inner aspect of the aortic lumen are suggestive of atherosclerosis.60 The arterial phase CT may reveal mural thickening with inhomogeneous enhancement, which possibly reflects the vascularization of the tunica media.24 Postenhanced CT (delayed, venous phase) typically demonstrates a “double ring” appearance, with an inner ring with low enhancement (likely representing the thickened intima) and an outer ring with high enhancement (representing the florid active inflammation and vascularization in the media and adventitia).24,60 This double ring enhancement pattern has been suggested to be useful to evaluate treatment efficacy.60,61,62
Various reconstructed images including maximum intensity projection (MIP), curved planar reformation, and volume-rendered images allow more detailed evaluation of luminal changes24,63; MIP and volume-rendered images are also useful to visualize small vessel changes.63 Advanced CT technology with new-generation machines permits faster and higher-resolution scanning64,65 and accurate assessment of the severity of aorto-ostial lesions, particularly of coronary arteries and distal lesions,64,65,66 with low-dose radiation exposure.67,68,69,70 However, radiation exposure remains the main concern in children, especially considering that the lifetime risk of radiation exposure is cumulative.71 Epidemiologic studies provide clear evidence that the organ doses from a common CT scan (2 or 3 scans, resulting in a dose between 30 and 90 mSv) increase the risk of cancer, especially in children, who are more radiosensitive compared to adults.72,73 Hence, CT is rarely used, and whenever possible, MRI should be preferred to avoid repeated radiation exposure.18 Pediatric data are therefore scarce and restricted to case reports and small series.26,29,33,48,74–78 As previously mentioned, the main disadvantages are the radiation delivery and the need for contrast media. Similar to MRI, the contrast bolus has to be timed perfectly in order to overcome high concentration venous contrast that interferes with the evaluation of the subclavian and brachiocephalic vessels.
The quality of the cited studies evaluating CT in TA ranges from high-level evidence for prospective studies using standardized imaging protocols60,61,64 to low-level evidence for retrospective cohort studies and case series/reports.
Doppler ultrasound
Doppler US is inexpensive and noninvasive, and does not involve radiation exposure. It is useful for the visualization of the arterial wall, measurement of intima-media thickness, and anatomic study of vascular stenosis or aneurysms. The “macaroni sign,” a long, homogeneous, midechoic, concentric arterial wall thickening, is considered characteristic of TA.79,80 In contrast, atherosclerotic plaques are inhomogeneous, irregular, and often calcified.55
B-mode US shows wall morphology (ie, increased thickness). Doppler US provides valuable information regarding altered blood flow characteristics, thereby helping to detect TA in a prestenotic phase.81 In addition, it can also indirectly measure arterial stiffness, which is often increased in adults and children with TA.81,82 Contrast-enhanced US improves the visualization of the vascular lumen, the parietal vasa vasorum, and arterial vessel wall perfusion.80,83,84,85 Using standardized contrast-enhanced US in a prospective cohort of patients with TA, Ma et al found significantly higher carotid artery wall thickness and more severe neovascularization in patients with active vs inactive TA.86 Consequently, carotid artery wall thickness and severe vascularization have been suggested as a potential marker of disease activity in adult patients with TA86 and the grade of vascular inflammation to correlate with disease activity observed on 18F-FDG PET.87 Contrast-enhanced US can also detect reduction of vessel wall thickness as a response to therapy81,86; thus, it may be a useful tool for diagnosis and monitoring of treatment response in TA patients with supra-aortic vessel involvement.
In children with TA, Doppler US is increasingly used, as it is noninvasive and nonirradiant. Doppler US has been shown to be an efficient imaging tool for diagnosis and follow-up management of some pediatric TA cases. It may be routinely used in the presence of an experienced pediatric radiologist, as well as in determining vessel involvement for those vessels that can be accessed by Doppler US.18,29,49,88 Challenges include the lack of pediatric radiologists with expertise in US imaging of vasculitis. Moreover, supra-aortic vessel involvement, which is well covered by Doppler US, is less common in children with TA compared to adults.1 Pitfalls of US include the investigator-dependent quality of the exam and the limited accessibility to study vessel anatomy according to acoustic technical limits; an example is the descending thoracic aorta, particularly in children.89 Disadvantages of contrast-enhanced US include its limitation to 1 vessel due to short examination time during contrast infusion, higher costs, and longer examination times compared to conventional US.
Given the lack of radiation, Doppler US may be safely used for diagnosis and follow-up monitoring of TA in pregnant women.
18Fluorodeoxyglucose positron emission tomography
18F-FDG PET detects vascular areas with increased metabolic activity (18F-FDG uptake by metabolically active, inflammatory cells of the vascular wall). Lesions are considered active when the tracer uptake within the vascular wall is higher than the tracer uptake of the liver. In early disease, activation of inflammatory cells precedes the morphologic changes.90,91 18F-FDG PET is therefore widely used for diagnosis of suspected TA, where it has been shown to have a good sensitivity92 (Figure 5).
However, its value for follow-up evaluation is uncertain. While a partial or complete reduction of tracer uptake during treatment has been reported in some studies,93,94,95,96 the correlation between tracer uptake and disease activity markers was poor in others.44 In all these studies, definition of disease activity was based on the presence of clinical features attributable to vasculitis, and in all but one additionally on the presence of elevated APRs.93,94,95,96 Grayson et al investigated the utility of 18F-FDG PET (standardized imaging sequences) as an imaging biomarker in a prospective, blinded study of 56 patients with LVV (26/56 with TA and among these, 5 children).97 Notably, 58% of PET scans in patients considered in clinical remission were interpreted as active vasculitis.97 Increased 18F-FDG uptake may persist for several years despite good response to treatment, as it is not specific to active TA but can also be induced by a high metabolic rate during healing processes, fibrotic remodeling, and other inflammatory vascular diseases such as atherosclerosis.98,99 This results in ambiguity as to whether tracer uptake in patients with clinically inactive TA represents subclinical vasculitis or nonspecific changes related to vascular damage. Interestingly, the global burden of 18F-FDG uptake during clinical remission has been associated with future clinical relapse, suggesting that the variable 18F-FDG uptake observed in clinically inactive patients reflects subclinical vasculitis.97 Overall, the value of 18F-FDG PET regarding evaluation of treatment response and prognosis remains limited and its definite role in the management of TA has yet to be determined.98
The combination of 18F-FDG PET with CT or MRI allows more precise localization of pathologic changes.100 A recent retrospective study investigating the role of 18F-FDG PET/MRI in patients with aortitis for assessment of disease activity monitoring during immunosuppressive therapy suggested the value of this hybrid imaging complementary to clinical and laboratory markers.101 Of note, assessment of disease activity status between PET/MRI, clinical, and laboratory evaluation differed in 25% of cases. The utility of novel imaging modalities such as diffusion-weighted whole-body imaging with background body signal suppression remains to be validated.58,102 Currently, 18F-FDG PET plays a minor role in disease management of pediatric TA; aside from a few case reports,103,104 5 children were included in the prospective longitudinal study mentioned above.97 To date, 18F-FDG PET is not recommended for routine monitoring of disease activity in children due to the high radiation dose. It may be used on a case-to-case basis to assess disease activity when disease status remains unclear despite clinical, laboratory, and radiological assessment. When available, PET/MRI should be preferred over PET/CT.18
Joint procedural recommendations, published in 2018 by the European Association of Nuclear Medicine, Society of Nuclear Medicine and Molecular Imaging, and the PET Interest Group, provide guidelines for imaging specialists and clinicians for the request, performance, and result interpretation of FDG-PET imaging in patients with suspected LVV.105 The statements conclude that FDG-PET/CT (angiography) plays an important role in the diagnosis of LVV, but also highlight the various open issues associated with diagnosis and disease monitoring, the lack of standardized protocols, and technical challenges.105 PET is usually considered very sensitive in depicting active TA, but can adequately visualize neither pulmonary arteries106 nor small vessels due to limited spatial resolution.93 Other disadvantages of PET include the high costs, the limited availability, and the significant exposure to ionizing radiation, especially if combined with CT.
Conclusion
In summary, imaging modalities including MRI, CT, Doppler US, and 18F-FDG PET reliably detect signs of suggestive vessel inflammation and allow recognition of TA even in the early prestenotic phase. However, despite advances in imaging techniques, there is no clear correlation with TA disease activity during follow-up monitoring; thus, the combination of clinical, laboratory, and imaging evaluation remains crucial for disease management and dosage of immunosuppressive medication. While MRI and CT both provide good visualization of vascular lesions and disease extent, MRI is often preferred and particularly recommended in children to minimize exposure to radiation. Doppler US is useful to diagnose and monitor TA in adults and children when affected vessels can adequately be assessed. Currently, the use of CA is principally limited to angiographic imaging prior to revascularization procedures. More recently, imaging was investigated as an outcome measure in an observational cohort and incorporated as an outcome measure regarding treatment response,107,108,109 again highlighting the importance of a combined clinico-biological and radiological evaluation for disease activity assessment and disease management.
In the future, novel imaging techniques such as multimodality imaging systems, contrast-enhanced US, or the use of vasculitis/immune cell–specific tracers should be investigated.36,105 To date, very few studies have evaluated the role of imaging in the prediction of angiographic and disease progression. However, imaging and novel imaging techniques may help to address the ongoing issues of accurate differentiation between active inflammatory lesions and nonspecific vascular remodeling observed during inactive disease phases, reliably monitor response to immunosuppressive therapy, and predict angiographic and disease progression.
Footnotes
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication November 23, 2021.
- Copyright © 2022 by the Journal of Rheumatology
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.
- 28.
- 29.↵
- 30.
- 31.
- 32.
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.
- 51.
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.
- 76.
- 77.
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵