Characterization of myocardial T₁ and partition coefficient as a function of time after gadolinium delivery in healthy subjects

Background

Diffuse myocardial fibrosis is associated with myocardial infarction [1], heart failure [2] and dilated cardiomyopathy [3]. Conventional T₁-weighted late gadolinium enhancement (LGE) imaging highlights focal scaring in contrast to remote reference tissue, but it cannot detect global changes in T₁ associated with diffuse fibrosis. Quantitative T₁ imaging permits assessment of diffuse fibrosis by eliminating the use of reference tissue, but the dependence of the derived partition coefficient (lambda) on the time post-contrast injection (t_post) is not well established.

Purpose

Determine blood and myocardial T₁ values as a function of t_post and the resulting dependence of the blood-tissue partition coefficient.

Methods

Nine healthy subjects (22.0±5.5 yrs, 6 male) were imaged using a Siemens Avanto 1.5T MRI. T₁ mapping was performed on a mid-ventricular short-axis slice using a custom saturation recovery single-shot TrueFISP sequence at baseline and one-minute intervals for 15 minutes following a bolus injection of gadopentetate dimeglumine (0.1 mmol/kg). At each time point, one "no-saturation" image and nine images with varying saturation recovery times spanning the cardiac cycle were acquired during a single breath-hold.

The myocardium was divided into 18 segments and mean values were fitted to a 3 parameter saturation recovery curve to determine T₁ values for each segment at every time point. Blood T₁ values were computed using a region of interest within the left ventricular cavity. Lambda was computed using {lambda=[R₁(myocardium_post) - R₁(myocardium_pre)]/[R₁(blood_post) - R₁(blood_pre)]}, where R₁=1/T₁.

Results

Figure 1 shows myocardial T₁, blood T₁, and lambda values averaged over all segments and subjects as a function of t_post. Average within-subject standard deviations of T₁ and lambda for t_post from 3-15 min were 34.1 ms and 0.046 respectively. Linear regression for lambda and t_post (3-15 min) shows an increase in lambda of 0.001 min^-1 (R²=0.75). Quantitative T₁ imaging is likely to be added to a clinical protocol following LGE imaging (t_post 10-15 min), where T₁ values increase by 5.9±1.6% and lambda increase by 1.1±2.7%.

Figure 1

T₁ (myocardium, blood) and lamba following contrast injection.

Conclusion

Saturation recovery SSFP T₁ mapping can be performed in a single breath-hold with derived blood-tissue partition coefficient (lambda) values in good agreement with previous measurements³. In the post-LGE window of 10-15 min after contrast bolus, derived lambda values show less time dependence than myocardial T₁.