Results from an analysis of patients presenting at 3 NYC-based centers suggests CTA scans combined with clinical questionnaire data yielded an 83% accuracy rate for diagnosing COVID-19 in stroke patients.
New data from a trio of New York City hospitals indicates computed tomography angiogram (CTA) scans could help identify stroke patients with coronavirus disease 2019 (COVID-19) earlier than standard nasal swab tests.
An analysis of lung scans from more than 50 stroke patients, results of the study indicate CTA scans combined with responses on COVID-19 symptom questionnaires led to an 83% accuracy rate, which could contribute to earlier detection and improve care for these patients.
“In combination with symptoms, CTA scan analysis is relatively accurate in diagnosing COVID-19, even compared to the nasal swab test. Since this analysis is much faster and at no extra cost, we hope it could be incorporated as a rapid diagnosis tool for patients with acute stroke,” said lead investigator Charles Esenwa, MD, MS, an assistant professor and a stroke neurologist at the Albert Einstein College of Medicine in the Bronx, New York City, in a statement from the American Heart Association. “In addition, accurately diagnosing COVID-19 within hours, rather than the sometimes days wait-time to receive the results from nasal swab tests could help protect both patients and medical professionals.”
With evaluation of lung apices using CTA scans for providing the first opportunity to screen for COVID-19 in patients with acute ischemic stroke, Esenwa and a team sought to retrospectively analyze the use of these evaluations as a COVID-19 screening measure. With this in mind, investigators pooled data from 3 Montefiore Health System hospitals in New York City, which yield a population of 110 patients with CTA scans of the head and neck for possible inclusion.
After eliminating patients who did not undergo an RT-PCR, those who did not present with acute ischemic stroke, and the application of further inclusion criteria, investigators identified a cohort of 57 patients for inclusion in their final analysis. For the purpose of analysis, CTA images were assessed by fellowship-trained radiologists blinded to the patient’s COVID-19 status using the COVID-19 reporting and data system (CO-RADS) scheme.
Of the 57 patients included in the final analysis, 30 were considered COVID-19-positive and 27 were considered COVID-19-negative. Upon initial analysis, 20 patients in the positive group and 2 patients in the negative group had findings on CTA images considered highly or very highly suspicious using CO-RADS (P <.001).
When used in isolation, a CO-RADS designation of highly or very highly suspicious for COVID-19 pneumonia yielded a sensitivity of 0.67, specificity of 0.93, positive predictive value of 0.19, negative predictive value of 0.99, and accuracy of 0.92 for the diagnosis of COVID-19. When used in combination with self-reported clinical symptoms, sensitivity of the lung assessments increased to 0.83.
“Early diagnosis via CT scans has helped our center protect other patients and staff through early isolation, and it has also allowed us to start early supportive care for those suspected of having stroke who are COVID-19 positive,” Esenwa added, in the aforementioned statement.
Investigators pointed out multiple limitations in their study. These included a lack of universal standard for diagnosing COVID-19, the exclusion of patients without RT-PCR, and the possibility for increased diagnostic accuracy due to location of the study centers.
This study, “Utility of Apical Lung Assessment on Computed Tomography Angiography as a COVID-19 Screen in Acute Stroke,” was published in Stroke.