Read more

March 10, 2020
2 min read
Save

CT-based algorithms may predict CV outcomes more accurately than current clinical parameters

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Automated CT-based algorithms predicted 5-year CV risk and overall survival in otherwise asymptomatic patients with greater efficacy than the Framingham risk score and BMI, according to a study published in The Lancet Digital Health.

In a cohort of 9,223 consecutive adults (mean age, 57 years; 56% women), most of whom were asymptomatic outpatients undergoing low-dose unenhanced abdominal CT for colorectal cancer screening as part of routine health maintenance, each algorithm, derived from one of five body composition measures, had greater univariate 5-year area under the curve values for overall survival compared with BMI, and some of them bested the Framingham risk score.

The univariate 5-year AUC values were 0.743 (95% CI, 0.705-0.78) for aortic calcification, 0.721 (95% CI, 0.683-0.759) for muscle density, 0.661 (95% CI, 0.625-0.697) for ratio of visceral to subcutaneous fat, 0.619 (95% CI, 0.582-0.656) for liver density, 0.646 (95% CI, 0.603-0.688) for vertebral density, 0.688 (95% CI, 0.65-0.727) for the Framingham risk score and 0.499 (95% CI, 0.454-0.544) for BMI, according to the researchers.

“This is the first study, to our knowledge, to apply a battery of validated, fully automated CT biomarkers to a large screening cohort of asymptomatic adults with long-term clinical follow-up to assess their ability to predict future adverse clinical events, such as myocardial infarction, stroke and death,” Perry J. Pickhardt, MD, chief of gastrointestinal imaging and medical director of oncological imaging at the University Wisconsin Carbone Cancer Center, and colleagues wrote. “Predictive ability of these CT biomarkers was compared with the well-established Framingham risk score and BMI. We found that the automated CT-based prediction was overall superior to the Framingham risk score and BMI.”

Moreover, during a follow-up of 8.8 years, during which 20% of patients died or had a major CV event, HRs for mortality for the highest-risk quartile compared with others using the same five CT-based measures were 4.53 (95% CI, 3.82-5.37) for aortic calcification, 3.58 (95% CI, 3.02-4.23) for muscle density, 2.28 (95% CI, 1.92-2.71) for the ratio of visceral to subcutaneous fat, 1.82 (95% CI, 1.52-2.17) for liver density and 2.73 (95% CI, 2.31-3.23) for vertebral density, compared with 2.82 (95% CI, 2.36-3.37) for the Framingham risk score and 1.36 (95% CI, 1.13-1.64) for BMI.

According to the study, multivariate combinations of CT-derived biomarkers further improved predictive efficacy over the Framingham risk score and BMI (P < .05 for AUC).

There were similar results in predicting major CV events, Pickhardt and colleagues wrote.

PAGE BREAK

“Our study shows the rich prognostic value that can be automatically derived from abdominal CT scans, incidental to the indication for imaging,” the researchers wrote. “Given the many millions of CT scans performed each year in many countries, harnessing these valuable data could identify many pre-symptomatic patients who are at high risk of future serious adverse events, potentially allowing for earlier intervention and prevention.” – by Scott Buzby

Disclosures: Pickhardt reports he is a consultant for Bracco Diagnostics and Zebra Medical Vision, and shareholder in Cellectar, Elucent and SHINE. Please see the study for all other authors’ relevant financial disclosures.