1351594 - Clinical evaluation of multi-parametric T1, T2, and T1ρ with mSASHA at 3T

Late gadolinium enhancement (LGE) is widely used to visualize extracellular volume changes that occur in a variety of pathologies, such as regional replacement fibrosis or edema. Non-contrast evaluation by parametric mapping may reduce exam time, cost, and obviate contrast administration.

T_1ρ has been proposed to be more sensitive to collagen content of fibrosis-based pathology, providing complimentary information to existing T₁ and T₂ mapping, potentially discriminating causes of ECV expansion such as infiltrates or edema (1). Adiabatic T_1ρ pulses have improved robustness to B₀/B₁ inhomogeneities compared to conventional spin-lock (2,3) and may enable 3T imaging. The mSASHA sequence (4) can be extended for simultaneous T₁, T₂ and adiabatic T_1ρ mapping in a single breath-hold.

Methods:

The prototype mSASHA sequence comprises a set of images with variable T₁, T₁+T₂, and T₁+T_1ρ weighting acquired in a 15-heartbeat breath-hold with maps jointly calculated from all images (Fig 1). T_1ρ preparation was achieved with 2 adiabatic hyperbolic secant pulses with β=3.5-7.0, ƒ_max=1600-1650Hz, duration = 30 ms each, max B₁⁺ ≥15 uT (3). 

Twenty-three patients (55 ± 19 yrs, 17 male) referred for clinical CMRs were enrolled and scanned at 3T (MAGNETOM Prisma or Skyra, Siemens Healthcare, Germany). Patients had one or more diagnoses: 22% hypertrophic cardiomyopathy, 48% coronary artery disease, 9% amyloidosis, 13% myocarditis, 17% other. Imaging included non-contrast mSASHA, bSSFP cine, and LGE following 0.1 mmol/kg contrast (Gadovist, Bayer HealthCare Pharmaceuticals, USA).

Focal lesions and remote myocardium were identified on LGE and corresponding ROIs were drawn on mSASHA maps. Relative percent change between lesions and remote areas were compared across T₁, T₂ and T_1ρ maps.

Results:

mSASHA mapping was successful in all subjects without visible B₀/B₁ related artifacts. LGE-identified pathology was present in 83% of patients and 83% of LGE lesions were visible in at least one of the parametric maps. Pathological regions showed larger relative elevations in T_1ρ compared to remote than T₂ and T₁ with respective changes of 30 ± 40%, 21 ± 27% and 12 ± 11%. However, some LGE findings showed no T_1ρ changes despite elevation on T₁ or T₂. 

In one study (Fig. 2), multi-vessel chronic infarction accompanied multi-focal septal injury on LGE suspected to be of embolic origin. The latter showed marked elevations in T_1ρ and T₂ and patchy T₁ elevations. Another example of acute coronary syndrome showed diffuse patchy enhancement of the LAD territory corresponding to marked elevations in T₁ and T₂ with mildly elevated T_1ρ (Fig. 3).

Conclusion:

Non-contrast multi-parametric mSASHA with simultaneous T_1ρ, T₁ and T₂ is clinically feasible at 3T and LGE lesions generally had greater T_1ρ contrast than T₁ or T₂.  Heterogenous presentation on maps suggests this approach may sub-differentiate etiology of LGE lesions. Further optimization of T_1ρ pulses may improve its sensitivity and specificity for certain pathologies.