Clinical models with CAC score improve CAD risk stratification
Click Here to Manage Email Alerts
In suspected obstructive CAD, clinical models with risk factors and a coronary artery calcium score identified up to 3.5 times more patients who may not benefit from further diagnostic testing vs. a basic pretest probability model.
“Inclusion of risk factors and coronary calcium score in the initial evaluation of patients with chest pain seems to improve patient management,” Simon Winther, MD, PhD, senior consultant in the department of cardiology at Gødstrup Hospital, Denmark, told Healio. “Previous studies have demonstrated improved reclassification of patients, and this study supports this conclusion, with data showing additional improved risk stratification for MI and death compared with the standard pretest probability models.”
Incorporating CAC scoring
Winther and colleagues analyzed incidences of MI and death, stratified according to categories by the risk factor-weighted clinical likelihood model and CAC score-weighted clinical likelihood model, compared with categories by the basic pretest probability model. Researchers used cohorts of patients without previously diagnosed CAD from a Danish register (n = 41,177) and a North American randomized study (PROMISE; n = 3,952). All patients were symptomatic and were referred for diagnostic testing because of clinical indications.
Researchers found that substantially more patients were categorized into the “very low” category of less than 5% probability with the risk factor-weighted clinical likelihood model (45.2%) and CAC score-weighted clinical likelihood model (60.2%) compared with the pretest probability model (18%). In addition, the CAC score-weighted clinical likelihood model categorized more patients into the high category (2.7%) compared with the pretest probability model (0.9%) and the risk factor-weighted clinical likelihood model (0%).
The annualized event rates of MI and death were 0.51% for the risk factor-weighted clinical likelihood model (95% CI, 0.46-0.56), 0.48% for the CAC score-weighted clinical likelihood model (95% CI, 0.44-0.56) and 0.37% for the pretest probability model (95% CI, 0.31-0.44).
Comparison of the predictive power of the three models using Harrell’s C statistics demonstrated superiority for the risk factor-weighted clinical likelihood model (0.64; 95% CI, 0.63-0.65) and CAC score-weighted clinical likelihood model (0.69; 95% CI, 0.67-0.7) compared with the pretest probability model (0.61; 95% CI, 0.6-0.62).
‘These data show the ability to have reduced diagnostic testing of patients with chest pain due to improved pretest stratification,” Winther told Healio.
Winther said more research is needed regarding the optimal cutoff for deferring testing. “Europe and the U.S. have different cutoffs due to lack of evidence,” Winther said.
Dispelling ‘clouds of uncertainty’
In a related editorial, Khurram Nasir, MD, MPH, MSc, chief of cardiovascular prevention and wellness at Houston Methodist DeBakey Heart and Vascular Center and co-director of the Center for Outcomes Research at Houston Methodist Hospital, and Safi U. Khan, MD, MS, cardiology fellow at the DeBakey Center, wrote that the data provide “much needed momentum toward dispelling existing clouds of uncertainty” with a pragmatic approach.
“With significant clinical equipoise on the basis of now favorable observational data on accurately identifying lower-risk symptomatic individuals at much higher rates with both risk factor-weighted clinical likelihood and CAC score-weighted clinical likelihood models not requiring downstream testing, recent guidelines tiptoeing with class 2A recommendation for consideration of risk stratification for low-risk individuals with CAC testing, and, most importantly, an accelerated transformative journey toward value-based medicine practices requiring us to do more with less, we believe that the bells will begin to ring much louder for a well-designed clinical trial to clarify the impact of these promising strategies vs. current standard-of-care approaches not only for identifying the right patients for advanced imaging, but also for determining how these strategies will affect subsequent patient outcomes,” Nasir and Khan wrote.
Reference:
For more information:
Simon Winther, MD, PhD, can be reached at: simowint@rm.dk.
Collapse
Click Here to Manage Email Alerts