2 - Prediction of Heart Failure Admission or Death Using a Novel CMR-based Risk Score in Patients with Dilated Cardiomyopathy

Saturday, February 15, 2020

1:25 PM – 1:35 PM

Location: Canary 3 & 4

Background: The prediction of heart failure-related events in patients with dilated cardiomyopathy using CMR-based markers is of strong interest given rapid expansion of CMR use in this patient population. In this study, we aimed to demonstrate capacity for multivariable risk score development for the prediction of heart failure admission or death in this referral cohort over an intermediate duration of clinical follow-up.

Methods: A total of 737 patients with dilated cardiomyopathy clinically referred for CMR-based evaluation were prospectively enrolled in the Cardiovascular Imaging Registry of Calgary (CIROC) between January 2015 and June 2018. In addition to this referral status, the study population was required to have confirmation of LVEF≤50% by CMR, no known CAD, no prior MI, and no CMR evidence of either prior ischemic injury or acute myocarditis. All underwent routine collection of patient demographics, comorbidities and symptom scores and standardized reporting using commercial software (Acuity, Cohesic Inc) followed by automated linkage to administrative databases for the collection of laboratory, pharmacy and ICD-10 coded clinical outcome data. Patients were followed for the combined endpoint of heart failure hospital admission or death. A multivariable cox proportional hazards model was developed and was used to establish a points-based risk score for prediction of the primary outcome.

Results: The population was 28% female with a mean age of 56 ±14years. During a median follow up of 923 days (IQR 627-1227), 78 (11%) patients experienced the composite endpoint of heart failure-related hospital admission or death. A multivariable model identified 4 CMR and 5 non-CMR variables associated with the primary outcome (Figure 1). Weighting of these according to adjusted beta coefficients produced a composite risk score inclusive of up to a maximum of 26 points. ROC analysis of the risk model showed good performance with an AUC of 0.723 with Goodness of Fit Testing showing excellent agreement between expected and observed events. Stratification of Risk scores is shown in Figure 2, demonstrating a significant 2.85-fold risk (p=0.004) for those with a score of 5-15, and 9.45-fold risk for a score >15 (p< 0.0001).

Conclusion: A risk score model inclusive of CMR-based markers can identify DCM patients at elevated risk of heart failure admission or death. External validation of this risk model in a unique patient cohort is required.

Presenting Author(s)

AC

Aidan K. Cornhill, BSc

MSc Student
Stephenson Cardiac Imaging Centre

Disclosure: Disclosure information not submitted.

Primary Author(s)

AC

Aidan K. Cornhill, BSc

MSc Student
Stephenson Cardiac Imaging Centre

Disclosure: Disclosure information not submitted.

Co-Author(s)

SD

Steven Dykstra, MSc, BEng

PhD Student
Libin Cardiovascular Institute of Alberta, University of Calgary

Disclosure: Disclosure information not submitted.
YM

Yoko Mikami, MD, PhD

Core Laboratory Manager
Stephenson Cardiac Imaging Centre, Libin Cardiovascular Institute of Alberta, University of Calgary

Disclosure: Disclosure information not submitted.
JF

Jacqueline Flewitt, MSc

Biomedical Engineer
Libin Cardiovascular Institute of Alberta, University of Calgary

Disclosure: Disclosure information not submitted.
MS

Michelle Seib, RN

RN, Research Coordinator
Libin Cardiovascular Institute of Alberta, University of Calgary

Disclosure: Disclosure information not submitted.
KY

Karen Yee, MSc, MPH

Alberta Health Services

Disclosure: Disclosure information not submitted.
PF

Peter D. Faris, PhD

Director, Research Facilitation
University of Calgary

Disclosure: Disclosure information not submitted.
CL

Carmen Lydell, MD

Clinical Assistant Professor, Clinical Co-Director,
Libin Cardiovascular Institute of Alberta, University of Calgary

Disclosure: Disclosure information not submitted.
Bobby Heydari, MD, MPH

Assistant Professor
University of Calgary

Disclosure: Disclosure information not submitted.
JL

Joon Lee, PhD

University of Calgary

Disclosure: Disclosure information not submitted.
James White, MD

Professor of Medicine
Libin Cardiovascular Institute of Alberta, University of Calgary

Disclosure: Cohesic Inc. (Individual(s) Involved: Spouse/Partner, Products/Services: Diagnostic Informatics): Ownership Interest (stocks, stock options, or other ownership interest excluding diversified mutual funds)