5964Hybrid PET/MR imaging for the prediction of left ventricular (LV) recovery after revascularisation of chronic total occluded coronaries

Abstract

Listen

Translate article

Abstract Background Percutaneous coronary intervention (PCI) of chronic total occlusion (CTO) represents one of the major challenges in interventional cardiology. Physicians are still reluctant in referring for PCI, assuming non viability of the myocardium subtended by the CTO. Data are controversial in assessing the improvement of left ventricular (LV) wall motion after revascularisation and the prognostic value of viability testing to guide patient selection. Purpose The aim of this study was to determine, whether hybrid fluorodeoxyglucose positron emission tomography/magnetic resonance (FDG PET/MR) imaging allows a more accurate prediction of LV regional wall motion recovery after successful PCI of CTOs in comparison to PET or MR alone. Methods We enrolled 49 consecutive symptomatic patients with CTO and evidence of wall motion abnormality in the corresponding CTO-territory. All patients underwent hybrid FDG PET/MR imaging as semi-quantitative assessment of myocardial viability - glucose metabolism in PET and late gadolinium enhancement (LGE) transmurality in MR – prior of PCI of the CTO. Follow-up MRI was performed in 23 patients 3–6 months after successful revascularisation to evaluate wall motion changes. Results We assessed viability in 124 myocardial segments subtended by a CTO in 23 patients with successful PCI who underwent serial imaging. Segments with wall motion abnormality at baseline (n=80) were analysed. Most of these segments (n=54, 68%) were concordantly assessed viable by PET and MR, conversely only 2 (2%) segments were assessed non-viable by both imaging techniques. However, almost one third of the segments showed discordant patterns of viability either PET not viable/ MR viable (3 (4%) segments) or PET viable/ MR not viable (21 (26%) segments): particularly the latter revealed a significant wall motion improvement (p=0.033). The combination of PET and MR showed a fair accuracy in predicting myocardial segments with wall motion improvement after CTO revascularisation (PET/MR area under ROC curve (AUC) 0.72, SE 0.07, p=0.002), which was superior to MR-LGE (AUC=0.66, SE 0.09) and FDG-PET (AUC=0.58, SE 0.10) alone (Figure). Comparisons of ROC curves Conclusion Hybrid PET/MR imaging prior to successful CTO showed a better performance than PET or MR alone in predicting regional improvement of disturbed wall motion. The complimentary information derived from both modalities may particularly help to identify small amounts of viable epicardial myocardium within large scars which can improve contractility after CTO-revascularisation.