Predictive value of myocardial perfusion single-photon emission computed tomography and the impact of renal function on cardiac death.

BACKGROUND Patients with chronic kidney disease (CKD) have worse cardiovascular outcomes than those without CKD. The prognostic utility of myocardial perfusion single-photon emission CT (MPS) in patients with varying degrees of renal dysfunction and the impact of CKD on cardiac death prediction in patients undergoing MPS have not been investigated. METHODS AND RESULTS We followed up 1652 consecutive patients who underwent stress MPS (32% exercise, 95% gated) for cardiac death for a mean of 2.15+/-0.8 years. MPS defects were defined with a summed stress score (normal summed stress score <4, abnormal summed stress score>or=4). Ischemia was defined as a summed stress score >or=4 plus a summed difference score >or=2, and scar was defined as a summed difference score <2 plus a summed stress score >or=4. Renal function was calculated with the Modified Diet in Renal Disease equation. CKD (estimated glomerular filtration rate <60 mL . min(-1) . 1.73 m(-2)) was present in 36%. Cardiac death increased with worsening levels of perfusion defects across the entire spectrum of renal function. Presence of ischemia was independently predictive of cardiac death, all-cause mortality, and nonfatal myocardial infarction. Patients with normal MPS and CKD had higher unadjusted cardiac death event rates than those with no CKD and normal MPS (2.7% versus 0.8%, P=0.001). Multivariate Cox proportional hazards models revealed that both perfusion defects (hazard ratio 1.90, 95% CI 1.47 to 2.46) and CKD (hazard ratio 1.96, 95% CI 1.29 to 2.95) were independent predictors of cardiac death after accounting for risk factors, left ventricular dysfunction, pharmacological stress, and symptom status. Both MPS and CKD had incremental power for cardiac death prediction over baseline risk factors and left ventricular dysfunction (global chi(2) 207.5 versus 169.3, P<0.0001). CONCLUSIONS MPS provides effective risk stratification across the entire spectrum of renal function. Renal dysfunction is also an important independent predictor of cardiac death in patients undergoing MPS. Renal function and MPS have additive value in risk stratisfying patients with suspected coronary artery disease. Patients with CKD appear to have a relatively less benign prognosis than those without CKD, even in the presence of a normal scan.