Efforts to improve the predictive value of the Framingham Risk Score (FRS) to determine who will develop coronary heart disease have led to the suggestion that other variables might improve strategies for identifying high-risk patients. The coronary artery calcium score (CACS) determined by computed tomography might improve risk prediction. Greenland and colleagues sought to determine whether the FRS combined with the CACS assessment provides better prognostic information than either alone.

The South Bay Heart Watch study enrolled asymptomatic patients with at least one coronary artery risk factor. After 30 months, participants underwent further evaluation for CACS. Patients with diabetes were excluded. Coronary heart disease risk factors were obtained, as were computed tomographic scans, to determine the CACS. Patients were followed using medical records and interviews for up to 8.5 years after the computed tomographic examination, with 99 percent of participants completing the questionnaire at least once in the follow-up period. The study end points included nonfatal myocardial infarction or coronary heart disease–related death.

The study cohort included 1,029 participants. The risk of a coronary heart disease–related death or a nonfatal myocardial infarction in participants with a CACS higher than 300 was 3.9 times that of participants with a CACS of zero. The risk of a coronary heart disease–related death or nonfatal myocardial infarction in participants with the highest FRS was 14.3 times higher than that of participants with an FRS of less than 10 percent. The risk of coronary heart disease–related death or nonfatal myocardial infarction across increasing categories of CACS was 1.6 and 1.7. These same risks were 1.4 for CACS and 1.6 for FRS per standard deviation increase. Elevated CACS (greater than 300) was predictive of higher risk as FRS increased.

An FRS of greater than 20 percent significantly predicted all-cause mortality risk, which was not the case for any CACS alone. Both FRS and CACS were able to predict coronary heart disease event risk independently. The CACS significantly modified FRS prediction except when the FRS was less than 10 percent. In the higher FRS categories (10 to 15 percent, 16 to 20 percent, and 21 percent or higher), CACS improved the predictive value in 3- to 9-percent increments when the score was greater than 300. Absence of a CACS did not preclude the risk of a coronary heart disease event. Because a CACS of more than 300 was associated with a significant increase in coronary heart disease event risk compared with FRS alone in an at-risk population, CACS can be particularly helpful in assessing risk and assisting in decision making in intermediate-risk patients (whose FRS indicate a 10- to 19-percent 10-year event risk), in whom there is greater uncertainty as to optimal management.

editor’s note: In a recent medical news and perspectives article inJAMA, the author notes that “the public, spurred in part by savvy marketing by facilities with CTs, clamors for the procedure [computed tomography calcium screening].” The author goes on to comment on recent efforts to correct the lack of scoring standardization deriving from variations in manufacturers’ protocols.¹ Of greater concern than problems of standardization, however, is the public demand for a procedure that has not been determined previously to have an impact on motivation and behavior.² It is an economic and public health challenge to adjudicate between the public’s desire for a diagnostic tool and that same public’s inability to use diagnostic information to make necessary lifestyle changes. Even if the calcium score adds prognostic information, evidence regarding patient-oriented impact is lacking.—c.w.