It has been shown that extracellular volume (ECV) mapping provides information on the presence of diffuse fibrosis formation in patients with DCM¹. Main drawback of this method is the need of a gadolinium based contrast agent including the need for a substantial delay (> 15 min) between injection and image acquisition and possible adverse renal effects. A non-contrast enhanced method would overcome these problems. Recently, it was shown that a significantly higher T_1ρ relaxation is found in compact myocardial fibrosis after chronic myocardial infarction in patients². In previous work we showed the feasibility of native T_1ρ-mapping for the detection of diffuse myocardial fibrosis in patients with DCM³. In this work, we will evaluate the performance of T_1ρ-mapping versus ECV and native T₁, as measure of diffuse myocardial fibrosis.

Twelve patients with idiopathic DCM (5 male, 7 female, age 61 ± 10 years) underwent an MRI exam on a Philips Ingenia 1.5 T MR scanner. Eight healthy control subjects (6 male, 2 female, age 51 ± 6 years) were imaged to obtain reference values. Written informed consent was obtained from all subjects. T_1ρ-mapping was performed using a 2D, single shot T_1ρ-prepared steady-state free precession (SSFP) sequence. Five images were obtained with different spin-lock (SL) preparation times (SL = 0, 10, 20, 30, 40 ms, amplitude 500 Hz) in diastole. Other parameters: bandwidth/pixel = 723 Hz, TE/TR = 1.74/3.5 ms, resolution = 2 x 2 mm2, slice thickness = 8 mm, FOV = 288x288 mm², flip angle = 35 degrees, SENSE = 2. Corresponding T₁ maps were acquired before contrast agent injection, and in the patients also 15 minutes after contrast injection (0.2 ml/kg Gadovist), using the MOLLI 3(3)5 scheme⁴. Blood samples were acquired to determine hematocrit.

Left ventricular myocardium was delineated by manually contouring the endocardial and epicardial border on the T_1ρ-maps of both patients and healthy volunteers. The area between the contours was used as ROI to calculate mean T_1ρ-relaxation time and mean native T₁-relaxation time for each slice. In patients, the mean extracellular volume fraction was also calculated in these ROIs.

The T_1ρ relaxation time was significantly higher in the DCM patients (55.6 ± 3.0 ms), compared to the healthy control subjects (51.5 ± 1.2 ms), p < 0.005 (Figure 2). A significant correlation was found between the T_1ρ relaxation time and the Extracellular Volume fraction in patients, with a Pearson correlation coefficient r = 0.61 (p = 0.046) (Figure 3). The native T₁ relaxation time was significantly higher in the DCM patients (1157 ± 65 ms), compared to healthy control subjects (1026 ± 21 ms, p < 0.005). However, no significant correlation was found between the T_1ρ relaxation time and native T₁ values in patients (Pearson r = 0.19, p = 0.57), or between the native T₁ and Extracellular Volume fraction in patients (Pearson r = -0.11, p = 0.74), as shown in Figure 4.In this study we demonstrate the first proof of principle for native in vivo detection of diffuse myocardial fibrosis in patients with end-stage DCM using non contrast-enhanced T_1ρ –mapping. We found a significantly higher T_1ρ relaxation time in DCM patients compared to healthy control subjects, and a significant correlation between T_1ρ values and ECV in DCM patients. Myocardial native T₁ was also significantly higher in the DCM patients, which is in accordance with other studies in literature^4,5, but no correlation was found between native T₁ and T_1ρ or between native T₁ and ECV (Figure 4). Although the patients sample size is small in our study, these results suggest that the T_1ρ relaxation time is potentially a better predictor of the ECV value than the native T₁ value. Native T_1ρ –mapping requires no separate pre- and post-contrast acquisitions with mandatory delays, and removes the need for hematocrit measurement. Thus, native T_1ρ –mapping is easier to incorporate in a clinical protocol as opposed to ECV-mapping. Main limitation for the use of T_1ρ mapping for fibrosis detection is that the T_1ρ relaxation time found in DCM patients is only approximately 2 standard deviations higher compared to healthy control subjects. Although this appears to be sufficient to distinguish healthy and diseased myocardium, more contrast could improve the sensitivity of the method.In conclusion, we evaluated the endogenous detection of diffuse myocardial fibrosis with T_1ρ MRI and showed a significant relation between T_1ρ and ECV, but not between native T₁ and ECV.