Optimizing TI selection in LGE imaging: pharmacokinetically standardized synthetic LGE and clinical cohort validation of a fixed 400 ms TI for LGE quantification
摘要
Purpose To evaluate the stability of late gadolinium enhancement (LGE) quantification across inversion times (TIs) around the null point (TInull) using pharmacokinetically generated synthetic LGE (sLGE) and to validate a fixed TI of 400 ms against conventional null-based LGE. Methods In the first cohort (n = 115), a two-compartment pharmacokinetic model was fitted to dynamic T1 maps acquired after gadolinium administration to predict T1 at 10 min, and sLGE images were simulated from the predicted T1 maps. LGE proportion (LGEp) was quantified from TInull − 60 to TInull + 100 ms, and LGEp was compared. In the second cohort (n = 80), LGEp with TI fixed at 400 ms (LGE400) was compared with conventional LGE using a Look-Locker–derived TI (LGEnull), and these images were visually compared. Results LGEp on sLGE was stable from TInull to TInull + 100 ms, with no significant differences across this range, and LGEp decreased significantly to approximately zero below TInull (p < 0.05). LGEp on sLGE with TI = 400 ms showed strong agreement with that from TInull to TInull + 30 ms (r = 0.979 to 1.0). In the second cohort, LGEp correlated strongly between LGE400 and LGEnull (r = 0.85), with minimal bias on Bland–Altman analysis. In participants without amyloidosis, the correlation further improved (r = 0.96), and both imaging approaches showed similar LGE findings. Conclusion Pharmacokinetically standardized sLGE showed stable LGE proportion above the myocardial null point and reduced LGE proportion below it. Under the evaluated 10-min post-contrast 2D inversion recovery-LGE conditions, fixed TI = 400 ms provided LGE quantification comparable to Look-Locker–determined TI in most patients.