ADA Presentation for HART CAD & CVE

Performance of Clinical/Proteomic Biomarker Panels to Predict Coronary Artery Disease Presence or Cardiovascular Prognosis in Patients with and without Diabetes Mellitus

American Diabetes Association (ADA) Scientific Sessions

June 24, 2018

Highlights & Key Findings

  • An estimated 23.4 million, or 9.1 percent of American Adults are diagnosed with diabetes, with an estimated 7.6 million, or 3.1 percent, of American adults with undiagnosed diabetes. Additionally, about 81.6 million, or 33.9 percent, of American adults have prediabetes or abnormally high blood sugar with or without symptoms.


  • The presence of diabetes increases the risk of Cardiovascular Disease (CVD) twofold, compared with the general population. On average, cardiovascular events occur approximately 15 years earlier in patients with diabetes.


  • Patients with diabetes mellitus (DM) are at increased risk for prevalent coronary artery disease (CAD) and incident major adverse cardiac events (MACE), such as cardiovascular death, myocardial infarction, and stroke. Prediction of risk in those with DM may be challenging.


  • Two biomarker panels were used to compare accuracy of these panels in patients with and without DM. The HART CAD panel includes 2 clinical variables (male sex and prior coronary angioplasty, with or without stent placement (prior PCI)) and 4 biomarkers: adiponectin, apolipoprotein C-1, kidney injury molecule-1 (KIM 1), and midkine. The HART CVE panel includes 4 biomarkers (amino-terminal pro-B type natriuretic peptide, KIM-1, osteopontin, and tissue inhibitor of matrix metalloproteinase-1 (TIMP-1)). These panels were derived in 167 DM and 482 non-DM patients; performance of each was validated in a second cohort of patients (N=278), with and without DM.

  • In patients with DM, the HART CAD panel had excellent performance for prediction of ≥ 70% coronary obstruction, with area under the curve (AUC) of 0.81 (p<0.001), sensitivity 85%, specificity 69%, positive predictive value (PPV) 90%, and negative predictive value (NPV) 58% for detection of CAD. These results were comparable to those patients without DM.

  • In patients with DM, the HART CVE panel was highly predictive of incident 1-year MACE: AUC of 0.80 (p=0.002), sensitivity 78%, specificity 69%, PPV 47%, and NPV of 90%, again, comparable to those without DM. In Kaplan-Meier analyses, an elevated HART CVE score predicted shorter time to first MACE (p<0.001).


  • The HART CAD and HART CVE biomarker panels effectively predict prevalent CAD and risk for incident MACE in at-risk patients with DM.

Scroll Down for the ADA Presentation for HART CAD & CVE

The 2018 ADA Presentation for HART CAD & CVE