TY - JOUR T1 - Regional versus global PET function and perfusion computations for detecting cardiac ischemia JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1636 LP - 1636 VL - 62 IS - supplement 1 AU - Andrew Van Tosh AU - J. Jane Cao AU - Christopher Palestro AU - Kenneth Nichols Y1 - 2021/05/01 UR - http://jnm.snmjournals.org/content/62/supplement_1/1636.abstract N2 - 1636Objectives: Because myocardial ischemia is not only produced by arterial stenosis but also by diffuse epicardial coronary disease & microvascular coronary disease, coronary flow capacity (CFC) has been advanced as an approach to categorize the severity of left ventricular (LV) ischemia by PET myocardial blood flow (MBF) measurements. The objective of this investigation was to compare global versus regional CFC computations for identifying pts with cardiac ischemia. Methods: Data were examined retrospectively for 231 pts evaluated for known/suspected CAD who underwent rest & regadenoson-stress82Rb PET/CT. Rest & stress absolute MBF & myocardial flow reserve (MFR = stress MBF/rest MBF) were quantified from first-pass 82Rb PET time activity curves. CFC categorized the 3 main arterial territories (LAD, LCX & RCA) from regional MFR & stress MBF values, according to a proposed classification system(J Am Coll Cardiol 2018;72:2642-62), based on which we defined “abnormal CFC” as either moderately reduced (MFR > 1.27-1.60 & stress MBF > 0.83-1.09 ml/g/min) or severely reduced (MFR ≤ 1.27 & stress MBF ≤ 0.83 ml/g/min). These same criteria were used to categorize overall global LV CFC. Relative perfusion values, including summed stress score (SSS), were assessed for each arterial territory by applying gender-specific 82Rb normal limits. Global SSS was computed as the sum of the 3 territorial SSS values. Asynchrony was assessed as systolic phase contraction bandwidth (BW), the % of the R-R interval spanning 95% of LV contractions. BW was assessed both globally & regionally for each of the 3 main arterial territories. A subgroup of 106 pts also had x-ray contrast arteriography. Their angiograms were reviewed at a core lab by expert cardiologists, who quantified coronary artery stenosis by planimetry. Results: Inter-rater agreement of global CFC values with maximum regional CFC values demonstrated “good agreement” (κ = 0.78). Categorizing the 231 pts by abnormal global CFC & any abnormal individual regional CFC agreed significantly (χ2 = 128.4, p < 0.0001), with inter-rater agreement considered to represent “good agreement” ( κ = 0.72). However, 23 pts had abnormal regional CFC but normal global CFC (McNemar’s Δ = 10%, p < 0.0001). Differences were not significant in identifying pts with asynchronous regional BW & those with abnormal regional CFC (Δ = -1%, p =1.00), but more pts had global asynchrony than had abnormal global CFC (Δ = 28%, p < 0.0001). Similarly, differences were not significant in identifying pts for whom regional SSS was > 4 & those with abnormal regional CFC (Δ = 5%, p = 0.13), but more pts had global SSS > 4 than had abnormal global CFC (Δ = 29%, p < 0.0001). In the subgroup of 106 pts who underwent quantitative arteriography, identification of pts with a specific artery with stenosis > 70% was similar when identified by any abnormal regional CFC (Δ = 2%, p = 0.63), but more pts had an arterial stenosis > 70% than had abnormal global CFC (Δ = 23%, p < 0.0001). Conclusions: Anatomic (>70% stenosis) & functional (regional perfusion abnormalities & LV asynchrony) evidence of myocardial ischemia correspond more closely to regional than global abnormalities of CFC. This suggests that territorial, rather than global, CFC values may potentially have greater value for identifying risk for future cardiovascular events. ER -