Background: Despite growing recognition of attenuation correction (AC) utilizing an external radiation source, prognostic studies using AC are lacking.
Methods: Consecutive patients (n = 419) who underwent stress Tc-99m sestamibi SPECT imaging with AC but without ECG-gating, due to arrhythmia, were followed for cardiac death or non-fatal myocardial infarction (MI). A 17-segment model and the summed stress score (SSS) were used to classify both the non-AC and AC images.
Results: The 419 patients had a mean age of 71.5 +/- 11.7 years and most (70.6%) underwent pharmacologic stress. In follow-up, 35 (8.4%) patients suffered an adverse cardiac event. Patients with AC-SSS 1-3 and AC-SSS 4-8 had similar cardiac event rates (11.4% vs 10.5%, P = NS). Accordingly, AC-SSS cutoffs of 0, 1-8, and >8 were selected to classify perfusion as normal, mildly abnormal, and moderately to severely abnormal with annualized event rates of 2.1%, 10.8%, and 18.7%, respectively (P < .001). In multivariable analysis, AC-SSS >8 was the most powerful predictor of cardiac events followed by AC-SSS 1-8, history of CAD, age >75 and pharmacologic stress.
Conclusions: AC provides powerful risk stratification when added to clinical variables in patients undergoing stress Tc-99m sestamibi SPECT imaging without ECG-gating. Moreover, smaller/less severe defects on AC data are more significant than if the same defects were seen on non-AC data.