Introduction
Giant cell arteritis (GCA) is the most common form of large vessel vasculitis [
1] and a potentially life-threatening inflammatory condition. When suspected, GCA requires prompt clinical evaluation and fast diagnostic imaging tests for an early and accurate diagnosis [
2]. Due to its high specificity, the gold standard for its diagnosis has traditionally been considered a temporal artery (TA) biopsy. However, ultrasound (US) has emerged as a rapid and non-invasive imaging tool to detect signs of GCA. After more than 25 years since it was first described by Schmidt et al. [
3], US has displaced TA biopsy being nowadays the preferred method to detect GCA. Moreover, the recently published European League Against Rheumatism (EULAR) guidelines recommends TA and axillary arteries (AA) US exam as the first-line imaging test in centers with expertise when patients are presenting with cranial symptoms of GCA [
4]. The diagnosis of GCA can be confirmed with US positive finding not requiring additional complementary exams. The halo sign is the most important US finding in GCA and it is defined as a homogeneous, hypoechoic wall thickening, well delineated towards the luminal side, visible in two perpendicular planes, most commonly concentric in transverse scan [
5]. A recent meta-analysis by EULAR showed a pooled sensitivity of 77% and specificity of 96% for the halo sign [
6]. Furthermore, the compression sign, which is defined as a thickened arterial wall that remains visible upon compression, is also useful for GCA screening [
5].
In recent years, very high-resolution transducers have been available for the US exam providing new tools to assess GCA patients as the measurement of the intima-media thickness (IMT) by B-Mode without Doppler. In this context, cut-off values for TA and AA have been published, with excellent sensitivities and specificities (> 97%) for the diagnosis of GCA in subjects with a high pre-test probability [
7‐
9]. More recently, the Halo Score has been proposed as a quantitative measure to determine the extent of vascular inflammation by US, incorporating the IMT of each halo [
10]. According to these findings, the Halo Score correlates with systemic markers of inflammation and risk for ocular ischemia, and a high Halo Score (≥ 10) may help to confirm GCA diagnosis with high specificity (> 95%). This novel scoring system has also been validated in routine care showing good diagnostic accuracy [
11,
12].
However, IMTs above the cut-off values that mimic a positive halo and compression sign have been also described in patients not having GCA [
13] leading to false positives in patients with atherosclerosis [
14], intimal hyperplasia on biopsy [
15], or in several disorders as amyloidosis [
16], vasculitis mimics, or lymphoma [
17]. Thus, the trend towards the utilization of IMT cut-off values to diagnose GCA needs to be evaluated in the clinical context of patients with elevated cardiovascular risk (CVR). It has been described how a positive compression sign or IMTs above the traditional cut-off values may be present in patients without GCA in the presence of atherosclerosis or elevated CVR [
15]. Besides, quantitative US tools like the Halo Score, relying on the measurement of each artery IMT, may be influenced by the CVR, so an accurate interpretation of the score needs to be taken into account in this context.
The primary objective of this study was to evaluate the impact of CVR on the diagnostic accuracy of the US Halo Score in patients with suspected GCA.
Methods
Patients
This was a retrospective cross-sectional study including patients referred to a US fast track clinic (FTC) [
2] at our Academic Center for screening of possible GCA over a 2-year period (from June 2019 to June 2021). Patients with suspected GCA are referred to this US clinic for examination within 24 h per protocol. For the purposes of this study, only consecutive patients with GCA suspicion were included. The study was performed in routine clinical practice conditions.
Data collection
The following variables were collected from the electronic health records: demographics, presenting symptoms (headache, scalp tenderness, jaw claudication, visual symptoms and ocular ischemia diagnosis by an ophthalmologist, fever, polymyalgia rheumatica, and constitutional symptoms), previous use of glucocorticoids, and laboratory variables as C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), hemoglobin, and platelets. TA biopsy data were also included if available. The gold standard for GCA diagnosis was the clinical confirmation by the treating clinician after a 6 months follow-up. The clinical GCA diagnosis could be dismissed according to the clinician criteria, even in patients fulfilling ACR 1990 criteria or with positive imaging tests, if another more reasonable diagnosis was established.
Cardiovascular risk stratification
To assess the CVR, the following variables were collected: body weight, height and body mass index, history of acute myocardial infarction, acute coronary syndromes, transient ischemic attack, stroke, aortic aneurism, peripheral artery disease, diabetes mellitus with or without organ damage, estimated glomerular filtration rate (eGFR) using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula, cholesterol level (total cholesterol, high-density lipoprotein cholesterol and low-density lipoprotein cholesterol), systolic and diastolic blood pressure, and smoking habit (current or previous smoker). The European Society of Cardiology (ESC) Guidelines on CV Disease Prevention in clinical practice were applied to define four different categories of CVR [
18]. We calculated the risk score of each patient using the available ESC CVD Risk Calculator app for mobile devices (
https://www.escardio.org/Education/ESC-Prevention-of-CVD-Programme/Risk-assessment/esc-cvd-risk-calculation-app), and subjects were classified as very high, high, moderate, or low CVR. Patients with high or very high CVR were compared with patients with low or moderate CVR to determine differences in the diagnostic accuracy of the Halo Score.
Ultrasound assessment
The three TA segments (common superficial TA, its parietal and frontal branches) and extracranial (carotid, subclavian and distal axillary) arteries were bilaterally evaluated by US in all patients within 24 h per protocol (excluding weekends with delays up to 48 h). The exam was performed in a supine position, by a single experienced ultrasonographer (JMC) using an EsaoteMyLab8 (Esaote, Genoa) with a hockey stick 12–18 MHz high frequency transducer for the temporal arteries and an 8–14 frequency transducer for extracranial arteries. The distal axillary arteries were scanned from the axillary fossa. The focus was positioned at 5 mm below the skin for the TA and 2–3 cm for the axillary arteries. The pulse repetition frequency was 2–3 kHz. The color box was set at an angle between sound waves and artery < 60°. The presence or absence of a halo, compression, stenosis, or occlusion for cranial arteries and the presence of a halo for extracranial arteries were evaluated. The IMT was measured in gray scale mode in all the arteries included in the protocol if technically possible due to the maximum frequency probe (18 MHz). Doppler color mode was only used to better delineate the lumen of the vessel to improve the visibility of the IMT in unclear cases. The extent of vascular inflammation was quantified according to the Halo Score, identifying the maximum IMT in each artery and calculating the composite score according to predefined cut-off values. The halo grade scores of the axillary arteries were multiplied by a factor of 3. The Halo score values could range from 0 to 48 [
10]. The presence of a halo and/or compression sign in temporal arteries or the presence of a halo in extracranial arteries in the absence of atherosclerosis was considered sufficient for a positive US examination. The ultrasonographer was not blinded to the clinical information of the patient.
Statistical analysis
Quantitative data were described as mean (standard deviation, SD) and qualitative variables as absolute frequency (percentages). Chi-square test or Fisher’s exact test were used to analyze differences between proportions; Student’s t test was used for comparison between means. Criterion validity was evaluated using receiver operating characteristic (ROC) curves with GCA clinical diagnosis as external criterion and construct validity was determined by Spearman’s rank correlation coefficient (rho). All tests were two-sided; p values < 0.05 were considered statistically significant. The SPSS software (version 23.0; IBM, USA) was used for statistical analysis.
Ethical approval
This study was performed in accordance with the ethical standards of the responsible committee on human experimentation and the Helsinki Declaration of 1975, as revised in 1983. Research ethics committee approval for the protocol was obtained prior to commencing the study (RHEUM0322). The committee determined that written informed consent was not mandatory.
Discussion
In this observational cross-sectional study, we evaluated for the first time the impact of the CVR evaluation on the diagnostic accuracy of the most used US quantitative index, the Halo Score. Our main finding is that patients with high and very high CVR show higher US Halo Score affecting the diagnostic accuracy of this US index.
IMT cut-off values for GCA diagnosis have been previously described [
7‐
9,
19], but the impact of the individually CVR on the diagnostic accuracy of US has not been well determined. Similar to our results, De Miguel et al evaluated the IMT of carotid and TA in 40 patients with high CVR [
14]. They found that atherosclerotic disease with a carotid IMT > 0.9 mm increases the IMT of TA and might mimic the halo sign. They proposed a cut-off of TA IMT > 0.34 mm in at least two branches to minimize false positives in a GCA diagnosis, achieving high specificity (> 95%). In this study, the authors did not evaluate the impact of atherosclerosis or CVR in the extracranial (p.e axillary) artery IMT. More recently, Martire et al. [
20] investigated the performance of previously proposed IMT cut-off values [
7], scanning the temporal and axillary arteries in a cohort of non-GCA. They found that CVR was associated with higher IMT values, especially at superficial temporal artery level, showing IMT values higher than the proposed cut-off values in 21% of healthy subjects with high/very high CVR, and in only 3.2% of subjects with low/moderate CVR.
To our knowledge, this is the first study specifically design to evaluate the diagnostic performance of the US Halo Score for GCA in different CVR subgroups of patients. According to our findings, the diagnostic accuracy of the US Halo Score is reduced in patients with high/very high CVR in comparison with patients with low/moderate CVR. A possible explanation is that the Halo Score is based on the measurement of the halo (or increased IMT), which would be influenced by the CVR of each individual patient. Our results are in line with other previously published by Martire et al. [
20], although they studied a slightly different population (healthy subject without GCA or PMR), in contrast with a suspected GCA population in our cohort. In agreement with de Miguel et al. results [
14], a numerically higher IMT of TA were found in patients with high/very high CVR, although we did not specifically include atherosclerosis in our analysis. Subjects without GCA with high/very high CVR had numerically higher IMT of TA and significantly higher IMT of extracranial arteries than those with low/moderate CVR (Table
2). Similarly, GCA patients with high/very high CVR had higher IMT of TA. Surprisingly, among patients with GCA, IMT of extracranial arteries was higher in those with low/moderate CVR. However, these results are explained by the fact that patients with GCA and extracranial involvement are significantly younger than those with TA involvement and, therefore, they had lower CVR. Our results highlight the importance of appropriate interpretation of IMT values when assessing patients with suspected GCA. The ultrasonographer should be cautious when assessing patients for GCA screening and be able to discard the disease even with IMT values above the diagnostic cut-offs if patients have a low pre-test probability and elevated CVR.
The main strength of the present study is that it is the first to evaluate the impact of CVR on the diagnostic accuracy of the Halo Score for GCA diagnosis, in a relatively large cohort of patients with suspected GCA evaluated in a well stablished FTC. This study has several limitations including the retrospective design and the limited data for some ancillary studies as TAB which was only performed per clinician criteria, consistent with real-world clinical practice. Second, the ultrasonographer was not blinded to clinical data. Third, intra and inter-observer reliability was not investigated.
In summary, high and very high CVR may influence the diagnostic accuracy of the US Halo Score leading to false-positive findings in these patients. Higher IMT values may be found in cranial and extracranial arteries in subjects with high/very high CVR without GCA. Thus, CVR should be taken into consideration in the US vascular assessment of patients with suspected GCA. These results need to be confirmed in additional cohorts to determine the need of a modified US Halo Score in patients with high and very high CVR.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.