乌鸦传媒

This article is a collaboration between 乌鸦传媒 and:

A risk calculator accurately identified the patients with stable chest pain who were the least likely to have abnormal results from coronary CT angiography or stress testing, an analysis from the PROMISE trial showed.

In it, 26.8% of patients who got coronary CT angiography for suspected coronary artery disease (CAD) turned out to have been at minimal risk as reflected by normal results and no clinical events over 25-month follow-up, , of Tufts Medical Center in Boston, and collaborators reported.

As a , the PROMISE model that Udelson's group developed had a fairly robust C-statistic of 0.725 for the derivation and validation subsets (95% CI 0.705-0.746), they reported online in JAMA Cardiology.

Factors incorporated in the model included younger age; female sex; racial or ethnic minority; no history of hypertension, diabetes, or dyslipidemia; no family history of premature coronary artery disease; never smoking; symptoms unrelated to physical or mental stress; and higher HDL cholesterol.

Patients in the 10% most likely to be at minimal risk were indeed associated with low rates of severely abnormal test results (1.3% for coronary CT angiography and 5.6% for functional testing). Likewise, cardiovascular death or MI was rare in this group (0.5% over 25 months).

"Deferred testing may be considered" for such minimal-risk patients, the authors suggested.

"Our tool is not intended to act as concrete decision support to test or not test patients, which would require a randomized clinical trial for validation. Rather, it is an aid that quantifies minimal risk and potentially identifies a subset of patients who are unlikely to benefit from proceeding directly to noninvasive testing."

Their PROMISE model was a better identifier of low-risk patients (AUC 0.726, 95% CI 0.710-0.742) than the Framingham (AUC 0.686, 95% CI 0.670-0.704) and Diamond-Forrester/CASS risk scores (AUC 0.606, 95% CI 0.589-0.624).

"The long-standing Diamond-Forrester algorithm now grossly overestimates the likelihood of obstructive CAD," agreed , of Mount Sinai St. Luke's Hospital in New York, and , of Cedars-Sinai Medical Center in Los Angeles.

"The Diamond-Forrester algorithm adopted three simple questions to assess chest pain: its location, precipitants, and duration. This limited approach has shaped the gathering of chest pain ever since, resulting in clinical data in PROMISE and in other contemporary databases that contain relatively sparse chest pain information," they noted in an accompanying the study.

"Most likely, the collection of more rather than less chest pain information should be encouraged ... Given the ease of collection and ability to assess clinical data in modern electronic databases, developing a consensus on standard clinical data that should be collected for risk assessment at the time of cardiac testing would be desirable."

"The ideal pretest risk-assessment tool remains to be described and validated," Rozanski and Berman concluded, while citing new tools for predicting who is on the opposite end of the spectrum. Such tools for picking out the highest-risk patients include the model and two recent risk scores from the European Society of Cardiology.

The randomized 10,003 patients with no known CAD to coronary CT angiography (n=4,633) or functional testing. Participants were recruited from 193 sites in North America and underwent noninvasive testing from 2010 to 2013.

Importantly, the model in question still requires further validation, the authors admitted.

"The ability to identify a subset of intermediate pretest probability patients with stable chest pain who might safely defer noninvasive testing is appealing given concerns about the low yield of testing in current practice and the associated costs," Udelson and colleagues nonetheless maintained.

"The concept of not testing such patients is commensurate with guideline recommendations to not test unless the pretest probability of obstructive disease is greater than 10%."