Prediction tool combining clinical probability, ultrasound confirms GCA from mimics
Click Here to Manage Email Alerts
Key takeaways:
- The HAS-GCA score combines a validated clinical scoring tool with ultrasound scans.
- An international study to validate the HAS-GCA score is under way with a larger cohort.
A new metric reliably confirms giant cell arteritis in suspected cases by combining a validated clinical scoring tool with ultrasonography, according to data in The Lancet Rheumatology.
The prediction tool, called the HAS-GCA score, provides an accurate post-test probability for low-, intermediate- and high-risk patients and could enable “correct and rapid confirmation” of GCA in fast-track clinics, the researchers wrote.
“At the moment predicting GCA is quite cumbersome, and diagnosing GCA is based on taking biopsies of temporal arteries,” Bhaskar Dasgupta, MD, FRCP, head of rheumatology at Southend University Hospital, in the United Kingdom, told Healio. “You have to track down a surgeon and try and get the surgeon to do a temporal artery biopsy in a timely way. It's an invasive procedure with sampling errors.”
To reliably and quickly exclude GCA from mimics in fast-track clinics, Dasgupta and colleagues combined emerging point-of-care ultrasound techniques with an existing GCA prediction tool, the Southend Giant Cell Arteritis Probability Score (SGCAPS). According to the researchers, SGCAPS stratifies patients into risk categories based on demographics, symptoms, physical signs and C-reactive protein concentration in blood, as well as the presence or absence of relevant alternative diagnoses. To evaluate this combination as a prediction tool, they conducted a prospective, inception cohort study of 229 patients evaluated at seven European GCA fast-track clinics from October 2019 through June 2022.
Data for the SGCAPS calculation were collected at baseline. Sonographers scanned temporal and axillary arteries for hypoechoic non-compressive halo signs, which are a “standard in the diagnostic work-up” of GCA, the researchers wrote. Multivariable logistic regression compared the GCA prediction with 6-month confirmed diagnoses that incorporated symptoms, laboratory tests, ultrasound and other tests at clinician discretion.
Three ultrasound scores were tested: halo count, or the total of halo-positive arterial segments; halo score, or the sum of the temporal artery halo and axillary artery halo scores; and the OMERACT GCA Score.
According to the researchers, the most accurate prediction model combined SGCAPS and halo count score, which researchers named the HAS-GCA score. The model rated 74% (n = 169) of patients as either low or high probability of GCA, and only two patients in each group were misclassified. The tool yielded an optimism-adjusted C statistic of 0.969 (95% CI, 0.952-0.99).
Dasgupta said it was a “very pleasant” surprise that halo count outdid halo score, as halo count is “actually much easier to assess.”
“The more complex the measurement, the more difference there is likely to be between your measurement and my measurement,” he said. “The simpler the measurement, the more likely that it’s repeatable.”
An international study to validate the HAS-GCA score is underway. The procedures used in this study will be replicated across a cohort likely triple the size of the current analysis, or approximately 600 suspected GCA cases, according to Dasgupta.
He added that “the big advantage of doing a validation study” extends beyond the validation itself.
“Not only do we want to actually validate our results, but the whole process actually will spread out the practice of doing ultrasound,” Dasgupta said. “Lots of GCA units who are not traditionally doing ultrasound may now take up ultrasound because, ultimately, what they want is to have a good prediction tool.”
He added that if all goes well, the HAS-GCA score could potentially “spread across the world.”
“I’m really very, very optimistic that this is good practice,” Dasgupta said. “I can’t really see any obstacles to it, except for the fact that GCA units need people who are familiar with the expertise of doing ultrasound. It’s not a very expensive investigation compared to many other imaging techniques.”
Collapse
Read more about
Click Here to Manage Email Alerts