Abstract 4185: Does Combined Stress, Delayed Enhancement, and Functional Cardiac Magnetic Resonance Imaging Yield Incremental Prognostic Value and Risk Stratification?
Background. Evaluation of the intermediate risk patient (Pt) using combined cardiac magnetic resonance imaging (CMR) assessment of adenosine stress perfusion (STRESS), myocardial delayed enhancement viability (MDE), and ventricular wall motion imaging (WM) is a promising clinical approach. Whether this technique yields risk stratification and incremental prognostic value over pre-CMR data is uncertain.
Methods. We identified 1,002 consecutive patients [Age 64±14, 58% male, 50% prior coronary artery disease (CAD)] undergoing combined CMR for suspicion of CAD or ischemia. STRESS first-pass gadolinium gradient echo pulse sequence, MDE, and steady state free precession WM were interpreted using 17-segment, 5 point scoring. Summed scores (SS) were calculated for each. The primary endpoint was all-cause death (ACD), the secondary endpoint cardiac death (CD). Incremental prognostic value was defined as an increased Cox proportional hazards model (CPH) global chi-square with the addition of CMR data after adjustment for clinical and historical data.
Results: Follow-up (FU) at 2.6±1.2 years after CMR (2.4% lost to FU) identified 78 ACD (8.0%) and 38 CD (3.9%). Compared to pts with normal stress CMR (STRESS, MDE, and WM all normal), pts with abnormal results had greater ACD rates over FU (5.9% vs 11%, p=0.006). After adjusting for the most predictive pre-CMR data (dyspnea, rest heart rate, age, and known renal disease; global chi-square 55, p<0.001), the addition of SS from STRESS, MDE, or WM added incremental value for prediction of ACD (chi-squares: 66, 64, 67; all p<0.001, respectively). After adjusting for pre-CMR data and WM SS, only MDE SS added further (chi-square 71, p<0.001); STRESS SS did not add to WM SS (p=0.17). CPH predicting CD identified both STRESS and SWMS as incremental predictors (chi-square 55, p<0.001)
Conclusion: STRESS, MDE and WM all add incremental value for prediction of ACD and CD over pre-CMR data and can be combined to further enhance prognostication.