Methods

This is a retrospective cross-sectional study of a cohort of patients who had CAC scoring at the Cleveland Clinic Foundation between 2003 and 2011. The cohort was identified through an interrogation of the imaging data system, and data were collected through a retrospective chart review. The electronic medical records were reviewed to ascertain their co-morbidities, lipid profile results, blood pressure, and lifestyle habits including smoking at the time of CAC testing. In an effort to model the population from which the PCE are derived and applied, we opted to exclude patients with a known history of ASCVD, type 2 diabetes mellitus, age 40 years or less, age greater than 75 years, and patients who were not African-American or white. Furthermore, we excluded all patients on lipid-lowering therapy. This study was approved by the Institutional Review Board at the Cleveland Clinic Foundation.

Imaging was performed on a variety of scanner models over time, including standard 16-, 40-, and 64-slice scanners, dual-source scanners, and extended range single-source scanners from different vendors. Images were reconstructed at 3-mm slice thickness. Per institutional protocol, radiation exposure (dose-length product and effective radiation dose) was monitored in all patients and limited to as low as reasonably achievable; whenever possible, axial mode with prospective electrocardiographic triggering using a tube voltage of 120 kVp was used, and tube current was carefully selected on the basis of patient size, which models guidelines later recommended by the Society for Cardiovascular Computed Tomography. Scan length was limited to coverage of the heart only. Quantitative CAC scores were calculated using the method proposed by Agatston et al. If more than one CAC measurement was available for a participant during the study period, the initial measurement was used.

The 10-year risk for an ASCVD event was calculated using the published formula generated by the PCE set by the 2013 AHA/ACC Cardiovascular Risk Assessment guidelines. These guidelines categorize patients into low (10-year ASCVD risk <5%), intermediate (10-year ASCVD risk <5% but <7.5%), and high (10-year ASCVD risk ≥7.5%) risk categories. The treatment algorithm to determine the need and type of treatment for ASCVD was based on these guidelines, which use the risk assessment as determined by the PCE. In summary, adults aged ≥21 years with an low-density lipoprotein-C (LDL-C) ≥190 mg/dl and adults with diabetes between the ages of 40 to 75 years should be treated with a high-intensity statin unless contraindicated (treatment category 1[T1]); adults aged 40 to 75 years with an LDL-C of 70 to 189 mg/dl without clinical ASCVD or diabetes and an estimated 10-year ASVD risk ≥7.5% should also be treated with a moderate- or high-intensity statin (treatment category 2 [T2]); for adults aged 40 to 75 years with an LDL-C of 70 to 189 mg/dl without clinical ASCVD or diabetes and an estimated 10-year ASVD risk between 5% and 7.5%, it is reasonable to treat consider treatment with a moderate-intensity statin (treatment category 3 [T3]) and finally, for adults aged 40 to 75 years with an LDL-C of 70 to 189 mg/dl without clinical ASCVD or diabetes and an estimated 10-year ASVD risk <5%, routine statin treatment is not recommend unless supported by other factors (treatment not necessary [T4]).

The cohort was also categorized based on their CAC score, subjects were categorized into the following strata: low (0 or <100 Agatston units [AU]), intermediate (100 to 299 AU), and high (≥300 AU). The use of these cutpoints for categorization were based from the 2013 ACC/AHA cardiovascular risk guideline document which outlined the expert opinion thresholds for use of CACS when risk-based decisions regarding initiation of pharmacologic therapy are uncertain. A CACS score of 300 AU or higher or ≥75th percentile or higher for age, gender, and ethnicity is considered high risk and supportive for initiation of statin therapy if additional factors need to be considered. The cutpoints within the guidelines are largely based on the published results of the South Bay Heart Watch study and the subsequent Multi-Ethnic Study of Atherosclerosis (MESA), which suggested that coronary calcium predicted coronary heart disease in a multiethnic American cohort. In the MESA study, in comparison to a CAC score of 0, those with a CAC score between 101 and 300 had an increased adjusted risk of a coronary event by a factor of 7.73 and those who had a CAC score >300 had an increased risk by a factor of 9.67.

Continuous and categorical variables were described as means ± SD or counts and percentages, respectively. Cross tabulation of risk categories based on the PCE and CAC absolute values were performed to describe the number and percentage of subjects who were classified into potential treatment arms appropriately and potential implications on treatment. Comparisons between groups were done using the independent t test for continuous variables and chi-square test or Fisher’s exact test, as applicable, for categorical ones. Agreement between the different risk scoring systems was calculated based on Cohen’s Kappa coefficients. A value between 0.01 and 0.2 represents slight agreement, and a value between 0.21 and 0.4 represents fair agreement. Analyses were performed using SPSS, version 20.0 (IBM, Armonk, New York). p Value ≤0.05 was used to indicate significance of tests.