Abstract
3388
Introduction: Attenuation corrected (AC) myocardial perfusion imaging (MPI) has been demonstrated to improve the specificity of detecting coronary artery disease. Total perfusion deficit (TPD) reflects the total extent and severity of perfusion defects, and have been shown to correlate well with visual interpretation of MPI. We sought to evaluate and compare the predictive value of AC stress TPD (AC-sTPD) and noncorrected (NC) stress TPD (NC-sTPD) by analyzing MPI studies in REFINE SPECT (Registry of Fast Myocardial Perfusion Imaging with Next Generation SPECT).
Methods: From the REFINE SPECT Registry, a total of 4,870 patients with SPECT/CT performed at Yale New Haven Hospital were included (age: 64 ± 12 years, 45% female, body mass index [BMI]: 30.0 ± 6 kg/m2, 8% prior myocardial infarction, 6% prior coronary artery bypass surgery, 11% prior percutaneous coronary intervention, 26% diabetes, 64% hypertension, 53% hyperlipidemia). Stress TPD values were generated by standard analysis. The association between AC-sTPD or NC-sTPD and major adverse cardiovascular event (MACE) defined as the composite end point of mortality, nonfatal myocardial infarction or late coronary revascularization (>90 days after SPECT) was evaluated with time to event and Cox regression analyses. Receiver Operator Characteristics (ROC) curves were analyzed to evaluate the ability of NC-sTPD or AC-sTPD for predicting MACE.
Results: During a median follow-up of 25 months, 485 patients (10%) experienced MACE. In the overall cohort AC-sTPD had a higher receiver-operator-characteristic area under the curve (ROC-AUC, 0.629) than NC-sTPD (0.602; P = 0.03) for detection of MACE. When stratifying patients based on median sTPD, patients with higher than median AC-sTPD (>2.3 %), had worse MACE-free survival when compared to patients with lower than median NC-sTPD (hazard ratio [HR]: 2.01, 95% confidence interval [CI]: 1.67 – 2.42, p <0.001). Similarly, patients with higher than median NC-sTPD (>2.2 %), had worse MACE-free survival when compared to patients with lower than median NC-sTPD (HR: 1.68, 95% CI: 1.40 – 2.02, p <0.001). In the overall cohort AC-sTPD did not provide incremental prognostic value beyond NC-sTPD (net reclassification index -0.01 [95% CI: -0.04 – 0.03]). When restricting the analysis to patients with BMI ≥ 35 kg/m2 (n=996), patients with higher than median AC-sTPD, had worse MACE-free survival when compared to patients with lower than median NC-sTPD (Figure 1A, HR: 1.91, 95% CI: 1.21 – 3.02, p = 0.005). On the other hand, when restricting the analysis to patients with BMI ≥ 35 kg/m2, patients with higher than median NC-sTPD, had no difference in MACE-free survival when compared to patients with lower than median NC-sTPD (Figure 1B, HR: 1.12, 95% CI: 0.71 – 1.76, p=0.62). In patients with BMI ≥ 35 kg/m2, AC-sTPD had higher ROC-AUC when compared to NC-sTPD for the detection of MACE (Figure 1C, 0.611 versus 0.526, p=0.02). In patients with BMI ≥ 35 kg/m2 AC-sTPD provided significant incremental prognostic value beyond NC-sTPD (net reclassification index 0.19 [95% CI: 0.17 – 0.35]).
Conclusions: MPI quantification using AC-sTPD showed higher ROC-AUC for the prediction of MACE in patients undergoing stress MPI when compared to NC-sTPD. This improvement was most pronounced in patients with BMI ≥ 35 kg/m2. AC-sTPD might be a useful prognostic marker for the prediction of adverse events, especially in patients with BMI ≥ 35 kg/m2.