Background
Takayasu arteritis (TAK) is a systemic vasculitis mainly involving the aorta and its major branches and can cause occlusive or aneurysmal degeneration [
1]. The prevalence of TAK in Japan was estimated at 40 per 1 million of population, with a female predominance [
2]. The prevalence of TAK in European studies is lower than that in Japan, being 4.7–13.2 per 1 million of population [
3‐
5].
Despite this rarity, based on a number of previous studies [
1,
6‐
11], knowledge regarding TAK has much improved since its first report in Japan in 1908 [
12]. In particular, the vascular manifestations of TAK have been well reported in previous studies [
1,
6‐
11]. Although there might be some discrepancies among different ethnic populations, the largest scale observational study conducted in Japan [
1] reported that local symptoms and findings attributable to vascular involvement were most commonly observed in the cervicobrachial area. Accordingly, the majority of the patients exhibited vascular involvements of the aortic arch or its major branches, which supply the cervicobrachial area. Furthermore, the most common type of angiographic involvement according to Hata’s classification [
13] was type I (branches of the aortic arch).
Although the vascular manifestations of TAK are well known, the extravascular manifestations of TAK remain unclear. Some recent studies have addressed spondyloarthritis (SpA) [
14,
15] and inflammatory bowel disease (IBD) [
16,
17] in TAK patients. However, only sporadic case reports regarding other extravascular manifestations such as erythema nodosum [
18‐
20] and uveitis [
21,
22] have been reported. Moreover, peripheral arthritis has occasionally been observed in TAK patients, but data concerning these manifestations are not well established. Therefore, we aimed to identify the characteristics of extravascular manifestations of TAK. Furthermore, we evaluated the association between vascular and extravascular manifestations of TAK.
Methods
Study population
Two independent cohorts from two tertiary referral hospitals in South Korea were included. All study subjects were diagnosed with TAK and encoded M314 according to the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) between January 2012 and October 2017. All patients met the 1990 American College of Rheumatology (ACR) classification criteria for TAK [
23]. Computed tomography (CT) scans encompassing the entire aorta and its branches as a diagnostic evaluation were present for all patients. Patients who only underwent localized imaging work-up such as carotid duplex sonography or neck angiography were excluded.
Electronic medical records of the study subjects were reviewed, and information pertaining to age, sex, the presence of hypertension (HTN), clinical symptoms and signs, laboratory data at initial presentation including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), and imaging data were collected. Regarding clinical symptoms and signs, those attributable to vascular involvement such as claudication, decreased pulse, and blood pressure difference > 10 mmHg between arms, bruit, and carotidynia were reviewed. The CT scans were utilized to assess the distribution of vascular lesions and categorize them according to Hata’s classification [
13].
This study was approved by the Institutional Review Board of Asan Medical Center in Seoul, South Korea (IRB No. 2017-0857). The requirement for informed consent was waived due to the retrospective nature of the study.
The presence of peripheral arthritis, axial arthritis (sacroiliitis), recurrent oral ulcers, erythema nodosum, IBD (Crohn’s disease (CD) and ulcerative colitis (UC)), and uveitis were determined from the electronic medical records. A radiologist confirmed the presence of sacroiliitis from the CT scans that were taken for the diagnostic work-up of TAK.
For patients with peripheral arthritis, the number and distribution of joint involvements and positivity of rheumatoid factor (RF), anti-cyclic citrullinated peptide (anti-CCP) antibody, and antinuclear antibody (ANA) (if the laboratory test was performed) were reviewed. The number of joint involvements was classified as monoarthritis (1 joint involved), oligoarthritis (2–3 joints involved), and polyarthritis (≥4 joints involved). The distribution of joint involvement was assessed by the size (small joints, large joints, or both), symmetry, and location (upper extremities, lower extremities, or both) of the joints involved. For patients with sacroiliitis, positivity of HLA-B27 (if the laboratory test was performed) and whether they fulfilled the radiologic 1984 modified New York criteria [
24] were reviewed.
Statistical analysis
We described both vascular and extravascular manifestations. Continuous variables were expressed as the mean ± standard deviation (SD) and median (interquartile range (IQR)) for a normal and a non-normal distribution, respectively. Categorical variables were expressed as number (%). To evaluate the association between vascular and extravascular manifestations, we performed logistic regression analysis. Univariate analysis was performed for each variable, and variables determined statistically significant (p < 0.05) were included in a multivariate analysis.
Discussion
In this study, we determined that extravascular manifestations of TAK were not rare, with a prevalence of 19.0%. In particular, arthritis (11.9%) was the most common extravascular manifestation. To the best of our knowledge, this is the first study describing the extravascular manifestations of TAK in detail.
Sacroiliitis was present in 7.1% TAK patients. Considering that the global prevalence of SpA is approximately 1% [
25], the proportion of TAK patients with sacroiliitis appears high. Although not clear to date, there might be a shared genetic background between TAK and sacroiliitis, which could play a central role in their co-occurrence. Further study regarding this issue could aid in elucidating the pathophysiologic background of these diseases. We discovered that the characteristics of sacroiliitis in TAK differed from SpA patients in general, being predominantly female (89.5%) and exhibiting a low incidence of HLA-B27 positivity (14.3%) [
25‐
27]. This finding is consistent with a previous study [
15] that also demonstrated female predominance and low HLA-B27 positivity. Furthermore, the presence of SpA features other than sacroiliitis in these patients was relatively low (peripheral arthritis 15.8%, enthesitis 5.3%, uveitis 5.3%, IBD 5.3%) compared to SpA patients in general (peripheral arthritis 39.8–58.0%, enthesitis 37.8–50.0%, uveitis 8.5–27.0%, IBD 1.8–11.0%) [
28]. Considering that aortitis is one of the cardiovascular manifestations observed in ankylosing spondylitis and other forms of SpA [
29], it can be challenging to differentiate TAK patients with sacroiliitis from SpA patients with aortitis. The aforementioned differences in the characteristics of sacroiliitis between these two patient types could provide clues in distinguishing one from another.
Peripheral arthritis was observed in 6.0% of TAK patients. This manifestation was characterized by an asymmetric oligoarthritis pattern, mostly involving the large joints. This was similar to the pattern observed in SpA (asymmetric oligoarthritis with lower extremity predominance) [
30] except that an even distribution between the upper and lower extremities was observed.
Interestingly, the most common type of vessel involvement in TAK patients with arthritis (axial and/or peripheral arthritis) was type IIB, differing from the total study population where type V was most frequently observed. Furthermore, in the multivariate logistic regression analysis evaluating the association between arthritis and vascular manifestations, type IIB was associated with arthritis approximately three times higher than nontype IIB. Therefore, TAK patients with type IIB should be carefully monitored for arthritis.
In our study, the prevalence of IBD (2.6%) in TAK patients appeared low compared to previous reports (5.8–8.3%) [
16,
31]. However, considering that the prevalence of CD and UC are 0.01% and 0.03%, respectively, in the general Korean population [
32], a prevalence of 2.6% IBD in our TAK patients is obviously higher than that of the general population. This is consistent with a previous study reporting a higher prevalence of IBD in TAK patients than that in the general population [
16,
31].
Recurrent oral ulcers (8.6%) were the second most common extravascular manifestation in our study, whereas erythema nodosum (1.5%) and uveitis (0.7%) were rarely observed. Recurrent oral ulcers, erythema nodosum, and monoarthritis or oligoarthritis are manifestations that are also observed in Behcet’s disease, as well as inflammation of large vessels [
33]. Thus, it may be confusing to distinguish TAK from Behcet’s disease. However, a notable difference between TAK and Behcet’s disease is the absence of genital ulcers. In our data, none of the TAK patients exhibited genital ulcers, which have the highest discriminatory value in the International Study Group for Behcet’s Disease criteria [
34]. Furthermore, occlusion or stenosis due to homogeneous concentric wall thickness favors TAK, whereas thrombotic occlusion or a solitary aneurysm is more likely in Behcet’s disease [
33].
The present study has some limitations. First, this study was retrospective. Although not many, there were some missing data concerning clinical symptoms and signs. Therefore, there might be other extravascular manifestations that we have not identified. Second, only Korean TAK patients were included, and other ethnic populations could display different results. Thus, further studies with subjects from different ethnic populations are required. Despite these limitations, considering that TAK is such a rare disease, our results are strengthened by the relatively large number of patients.