Early onset of left ventricular regional asynchrony as an indicator of compromised coronary flow capacity and sub-clinical arterial stenosis

Purpose: Coronary flow capacity (CFC) has been proposed as an approach to categorize the severity levels of left ventricular (LV) ischemia using myocardial blood flow data from PET. Our study explored whether the less critical levels of CFC are associated with early signs of LV ischemia, including regional relative hypoperfusion and regional asynchrony (Asynch) in individual coronary territories.

Methods: Data were examined retrospectively for 105 pts evaluated for known/suspected CAD who underwent rest & regadenoson-stress ⁸²Rb PET/CT & coronary angiography. Rest & stress absolute myocardial blood flow (MBF) & myocardial flow reserve (MFR = stress MBF/rest MBF) were quantified from first-pass ⁸²Rb PET curves. Based on their regional stress MBF & MFR values, each patient’s 3 major coronary territories were categorized by CFC (J Am Coll Cardiol 2018;72:2642-62): CFC1 “No reduced CFC,” with MFR ≥ 2.95 & stress MBF ≥ 2.20 ml/g/min; CFC2 “Minimally reduced CFC,” with MFR < 2.95 & stress MBF < 2.20 ml/g/min; CFC3 “Mildly reduced CFC,” with MFR < 2.20 & stress MBF < 1.80 ml/g/min; CFC4 “Moderately reduced CFC,” with MFR < 1.55 & stress MBF < 1.05 ml/g/min; CFC5 “Severely reduced CFC,” with MFR < 1.00 & stress MBF < 0.80 ml/g/min. Relative perfusion values, including summed stress score (SSS), were assessed for each arterial territory by applying gender-specific ⁸²Rb normal limits. Asynchrony scores averaged visual scores of 2 independent readers who viewed systolic & diastolic rest & stress phase polar maps & phase histograms graded on a 5-point scale (0 = normal to 4 = markedly asynchronous extensive territory). All angiograms were reviewed at a core lab by expert cardiologists, and stenoses of each coronary artery were quantified by planimetry.

Results: Of the 315 arterial territories, 95 were classified as “No reduction in CFC,” 53 “Minimally reduced,” 110 “Mildly reduced,” 40 “Moderately reduced,” & 17 “Severely reduced” CFC. By multiple logistic regression, the parameter regional Asynch+regional SSS [(0.23±0.04*Asynch) + (0.24±0.04*SSS)] yielded the strongest parameter for separating CFC categories. “Normal-to-mildly reduced CFC” was distinguished from “moderately-to-severely reduced CFC” arterial territories (χ² = 100.0, p < 0.0001), with χ² higher than for either variable separately (χ² = 63.7 & χ² = 48.2). For CFC3 (“mildly reduced CFC”), the Asynch+SSS parameter was significantly different than for CFC1 & for CFC2, despite there being only subcritical stenoses averaging 30% for CFC3. CFC4 showed further worsening of Asynch+SSS, with slightly tighter stenosis of 45%. Asynch+SSS produced accuracy (ROC AUC) of (87±3%), with 70% sensitivity & 90% specificity, for identifying “moderately-to-severely reduced CFC” arterial territories, & correlated significantly with magnitude of measured stenosis (linear regression r = 0.44, p < 0.0001). ROC analysis indicated that stenosis > 72% distinguished CFC for arterial territories categorized as those with “moderately-to-severely reduced CFC.”

Conclusions: Our results indicate that CFC classification of arterial territories allows identification of LV ischemia at a subcritical degree of coronary stenosis. Further studies are warranted to determine whether the mild degrees of hypoperfusion and asynchrony seen at low levels of CFC have prognostic implications. .