Abstract 2842: CMR-Determined Regional Myocardial Perfusion Uptake Rate Predicts Events in Women Without Obstructive Coronary Artery Disease: The NHLBI-Sponsored Women’s Ischemia Syndrome Evaluation (WISE) Study
Introduction: Previously, we showed that low uptake of gadolinium contrast by magnetic resonance myocardial perfusion imaging (MPI) predicted adverse events in women without significant angiographic obstructive coronary artery disease (CAD). Others have indicated that gradients in the transmural perfusion pattern may be present, but these signals can be contaminated by blood-pool-myocardial susceptibility artifacts.
Hypothesis: We postulate that base to apex gradients in normalized MPI uptake slope may better predict adverse events in women without significant epicardial CAD.
Methods: Women (n=116), mean age 57±11 yrs, with symptoms suggestive of acute myocardial ischemia and without significant obstructive CAD as assessed by angiography (<70% stenosis) underwent MR MPI. During follow-up (34±16 months) time to first adverse event (death, myocardial infarction or hospitalization for worsening anginal symptoms) or first serious adverse event (death or myocardial infarction) was analyzed using basal and mid-level MPI data of normalized uptake-slope and uptake level, together with ejection fraction (EF).
Results: Adverse events occurred in 27 (23%), and serious adverse events occurred in 6 (5%). By univariate Cox regression modeling, parameters predictive of adverse events were the average MPI uptake level, the average MPI uptake slope and EF (p<0.005 for each). Assessment of the MPI uptake level requires establishment of normal values for each site, and when restricting analysis to self-normalized variables, myocardial uptake slope and EF were combined to predict adverse events (p<0.001). Further, the normalized uptake slope at the mid-LV was a better predictor than the basal-LV, (p<0.001). A positive value of the Cox model regression variable for the uptake slope at the mid-LV and EF identified high-risk patients. High vs. low risk patients experienced annualized event rates of 9% vs. 3%, p<0.001, and serious adverse events at annualized rates of 2.3% vs. 0.4%, respectively.
Conclusions: Among women with suspected myocardial ischemia but no significant obstructive CAD, mid-LV measures of MPI normalized uptake slope better predicted prognosis compared to basal-LV measures. Combination with EF further improved prediction of prognosis.