Collateral presence and extent do not predict myocardial viability and ischemia in chronic total occlusions: A stress-CMR study

Background: Well-developed collaterals are assumed as a marker of viability and ischemia in chronic total occlusions (CTO). We aim to correlate viability and ischemia with collateral presence and extent in CTO patients by cardiac magnetic resonance (CMR). Methods: Multicentre study of 150 CTO patients undergoing stress-CMR, including adenosine if normal systolic function, high-dose-dobutamine for patients with akinetic/>2 hypokinetic segments and EF =35%, otherwise low-dose-dobutamine (LDD); all patients underwent late gadolinium enhancement (LGE) imaging. Viability was defined as mean LGE transmurality <= 50% for adenosine, as functional improvement for dobutamine-stress-test, ischemia as >= 1.5 segments with perfusion defects outside the scar zone. Results: Rentrop 3/CC 2 defined well-developed (WD, n = 74) vs poorly-developed collaterals (PD, n = 76). Viability was equally prevalent in WD vs PD: normo-functional myocardium with <= 50% LGE in 52% vs 58% segments, p = 0.76, functional improvement by LDD in 48% vs 52%, p = 0.12. Segments with none, 1-25%,26-50%,51-75% LGE showed viability by LDD in 90%,84%,81%,61% of cases, whilst in 12% if 76-100% LGE (p < 0.01). There was no difference in WD vs PD for ischemia presence (74% vs 75%, p = 0.99) and extent (2.7 vs 2.8 segments, p = 0.77). Conclusions: In a large cohort of CTO patients, presence and extent of collaterals did not predict viability and ischemia by stress-CMR. Scar extent up to 75% LGE was still associated with viability, whereas ischemia was undetectable in 25% of patients, suggesting that the assessment of CTO patients with CMR would lead to a more comprehensive evaluation of viability and ischemia to guide revascularization.