Meeting NewsPerspective

Vascular ultrasound highly sensitive, specific for large-vessel GCA diagnosis

CHICAGO — Vascular ultrasound, when performed by experienced sonographers, demonstrated high sensitivity and specificity for the diagnosis of large-vessel giant cell arteritis, according to data presented here.

Researchers noted that these findings suggest the imaging modality has implications for first-line use in patients with suspected large-vessel disease.

“In this disease subset of GCA patients, there is an unmet need for earlier recognition and earlier diagnosis,” Berit Dalsgaard Nielsen, MD, a doctoral fellow at Aarhus University in Denmark, said during a press conference. “We performed this study to look into the performance of ultrasound. New EULAR recommendations recommend GCA patients have an early imaging test and patients who present with cranial symptoms, which is typical of GCA when it effects the cranial arteries, ultrasound for the temporal arteries is the recommended imaging test.”

Nielsen added that “For patients who do not present with cranial symptoms, different modalities can be used for confirmation of large vessel involvement, but no priority for the modalities is given because there are no studies that compare [them], and studies on diagnostic accuracy for large vessel involvement are sparse.”

Nielsen and colleagues performed clinical and laboratory examinations, as well as ultrasound and PET scans prior to treatment initiation, among 90 patients with suspected GCA. Patients were considered for inclusion if they were aged 50 years; had CRP>15 mg/L or ESR>40 mm/h; had either cranial symptoms, new-onset claudication, protracted constitutional symptoms or polymyalgia rheumatica symptoms. The researchers excluded patients with recent or ongoing glucocorticoid or DMARD treatment, those with a previous diagnosis of GCA or polymyalgia rheumatica, or large vessel inflammation mimicking large-vessel GCA.

The control group was comprised of patients whose clinical diagnosis was not GCA (n = 34), and the case group included patients with GCA with large vessel involvement (n = 46). Ultrasound was performed by experienced sonographers who were blinded to the results of patients’ PET scans.

Axillary or carotid artery ultrasound was positive among 36 of 46 patients with large-vessel GCA; no control patients had a positive large vessel ultrasound, yielding a specificity of 100% and a sensitivity of 78%. Excluding carotid artery assessment from the analysis only decreased sensitivity to 76%. Using an intima-media thickness cutoff of 1.0 mm in the axillary artery, 70% sensitivity and 93% specificity was obtained. The agreement between sonographers' assessment of the ultrasound images was almost perfect, Nielsen said.

“We think these results implicate that ultrasound should not only be a first-line imaging test in patients presenting with cranial symptoms, but also in patients suspected of GCA presenting with constitutional symptoms,” Nielsen said. “If this examination is included in the standard examinations in the fast-track clinics, it may overcome the delay in diagnosis and patients can be treated earlier. It may also spare the unneeded examinations performed in these patients.”– by Stacey L. Adams

Reference:
Nielsen BD. Abstract 2905. Presented at ACR/ARHP Annual Meeting, Oct. 20-24, 2018; Chicago.

Disclosure: Nielsen reports no relevant financial disclosures.

CHICAGO — Vascular ultrasound, when performed by experienced sonographers, demonstrated high sensitivity and specificity for the diagnosis of large-vessel giant cell arteritis, according to data presented here.

Researchers noted that these findings suggest the imaging modality has implications for first-line use in patients with suspected large-vessel disease.

“In this disease subset of GCA patients, there is an unmet need for earlier recognition and earlier diagnosis,” Berit Dalsgaard Nielsen, MD, a doctoral fellow at Aarhus University in Denmark, said during a press conference. “We performed this study to look into the performance of ultrasound. New EULAR recommendations recommend GCA patients have an early imaging test and patients who present with cranial symptoms, which is typical of GCA when it effects the cranial arteries, ultrasound for the temporal arteries is the recommended imaging test.”

Nielsen added that “For patients who do not present with cranial symptoms, different modalities can be used for confirmation of large vessel involvement, but no priority for the modalities is given because there are no studies that compare [them], and studies on diagnostic accuracy for large vessel involvement are sparse.”

Nielsen and colleagues performed clinical and laboratory examinations, as well as ultrasound and PET scans prior to treatment initiation, among 90 patients with suspected GCA. Patients were considered for inclusion if they were aged 50 years; had CRP>15 mg/L or ESR>40 mm/h; had either cranial symptoms, new-onset claudication, protracted constitutional symptoms or polymyalgia rheumatica symptoms. The researchers excluded patients with recent or ongoing glucocorticoid or DMARD treatment, those with a previous diagnosis of GCA or polymyalgia rheumatica, or large vessel inflammation mimicking large-vessel GCA.

The control group was comprised of patients whose clinical diagnosis was not GCA (n = 34), and the case group included patients with GCA with large vessel involvement (n = 46). Ultrasound was performed by experienced sonographers who were blinded to the results of patients’ PET scans.

Axillary or carotid artery ultrasound was positive among 36 of 46 patients with large-vessel GCA; no control patients had a positive large vessel ultrasound, yielding a specificity of 100% and a sensitivity of 78%. Excluding carotid artery assessment from the analysis only decreased sensitivity to 76%. Using an intima-media thickness cutoff of 1.0 mm in the axillary artery, 70% sensitivity and 93% specificity was obtained. The agreement between sonographers' assessment of the ultrasound images was almost perfect, Nielsen said.

“We think these results implicate that ultrasound should not only be a first-line imaging test in patients presenting with cranial symptoms, but also in patients suspected of GCA presenting with constitutional symptoms,” Nielsen said. “If this examination is included in the standard examinations in the fast-track clinics, it may overcome the delay in diagnosis and patients can be treated earlier. It may also spare the unneeded examinations performed in these patients.”– by Stacey L. Adams

Reference:
Nielsen BD. Abstract 2905. Presented at ACR/ARHP Annual Meeting, Oct. 20-24, 2018; Chicago.

Disclosure: Nielsen reports no relevant financial disclosures.

    Perspective
    Anthony M. Sammel

    Anthony M. Sammel

    Congratulations to Dr. Nielsen’s group on this important study.

    These new imaging studies in giant cell arteritis show that we have options to risk-stratify patients before deciding who may or may not need a biopsy.

    Each of the imaging modalities has benefits and drawbacks. The main benefit of ultrasound is cost. Infrastructure costs less than for PET/CT and MRI. The challenge with ultrasound is that it is highly operator-dependent. For PET/CT and MRI the interpretation of the scan is reporter dependent, but the scan itself is standardized. If there are concerns regarding interpretation, the scan may be reviewed at a later time. This is in contrast to ultrasound where both acquisition and interpretation is operator-dependent and performed in real-time. There are two steps for inaccuracies to arise.

    The challenge with making a definitive diagnosis using imaging lies with the implications if we get it wrong. If we incorrectly reassure a GCA patient that they don’t have the condition and withhold or stop treatment, they are at risk of going blind. If we over-diagnose the condition, then patients without GCA may end up on corticosteroids for months or years. It’s critical to get the diagnosis right.

    In centers where the sonographers are very experienced in both performing and interpreting the ultrasound in GCA patients, I think it is a very useful test. Dr. Nielsen’s study would support this. My concern is that in centers where it’s not commonly performed, some patients may come to harm from misdiagnosis. It is instructive to look at the TABUL study, which was performed in the United Kingdom and published in 2016. It included 360 patients suspected of having giant cell arteritis and the sensitivity, compared with biopsy, was 73%. This means that 27 out of every 100 patients with biopsy-positive GCA may be missed with ultrasound. Those 27 patients are at risk of going blind if treatment is withheld. Looking more broadly at the ultrasound literature, some centers report very good diagnostic accuracy and others less so. The TABUL study included a broad cross-section of centers across the country and may best reflect real-world practice.

    The other benefit of PET scan, which we discussed this morning [at the press conference] is that we have the opportunity to detect both alternative mimicking conditions and inflammation in the aorta. In the Unites States, American Heart Association-endorsed imaging guidelines recommend that GCA patients undergo cross-sectional imaging of their chest to assess for aortic aneurysm and inflammation. This changes the cost equation of ultrasound and makes the discordance with PET/CT less marked.

    In summary, the choice of imaging test will depend on the expertise and resources of individual centers.

    • Anthony M. Sammel, MBBS
    • Rheumatologist, North Shore Hospital, Sydney

    Disclosures: Sammel reports no relevant financial disclosures.

    See more from American College of Rheumatology Annual Meeting