Preoperative assessment of meningioma stiffness using magnetic resonance elastography

Abstract

Object: The object of this study was to determine the potential of magnetic resonance elastography (MRE) to preoperatively assess the stiffness of meningiomas.

Methods: Thirteen patients with meningiomas underwent 3D brain MRE examination to measure stiffness in the tumor as well as in surrounding brain tissue. Blinded to the MRE results, neurosurgeons made a qualitative assessment of tumor stiffness at the time of resection. The ability of MRE to predict the surgical assessment of stiffness was tested using a Spearman rank correlation.

Results: One case was excluded due to a small tumor size. In the remaining 12 cases, both tumor stiffness alone (p = 0.023) and the ratio of tumor stiffness to surrounding brain tissue stiffness (p = 0.0032) significantly correlated with the surgeons' qualitative assessment of tumor stiffness. Results of the MRE examination provided a stronger correlation with the surgical assessment of stiffness compared with traditional T1- and T2-weighted imaging (p = 0.089), particularly when considering meningiomas of intermediate stiffness.

Conclusions: In this cohort, preoperative MRE predicted tumor consistency at the time of surgery. Tumor stiffness as measured using MRE outperformed conventional MRI because tumor appearance on T1- and T2-weighted images could only accurately predict the softest and hardest meningiomas.

Figure 1

Example images from a subject with a hard meningioma (top row) and a subject with a soft meningioma (bottom row). The left column displays the T1-weighted image, showing the location of the tumor outlined in green dots. The middle column shows a wave image from each subject. Note that the shear wavelength is longer in the hard meningioma and shorter in the soft meningioma relative to normal brain parenchyma. The right column is the calculated stiffness map in units of kPa, demonstrating that these tumors have significantly different mechanical properties.

Figure 2

Plot of tumor stiffness as measured by MRE versus the surgeons’ qualitative assessment of tumor stiffness (r=0.65, p=0.023). Scatter along the x-axis is random and for display purposes only to avoid overlapping data points.

Figure 3

Plot of the ratio of tumor stiffness to surrounding brain tissue stiffness as measured by MRE versus the surgeons’ qualitative assessment of tumor stiffness (r=0.77, p=0.0032). Scatter along the x-axis is random and for display purposes only to avoid overlapping data points.

Figure 4

Plot of MRI (T1 and T2 imaging) prediction of tumor stiffness versus the surgeons’ qualitative assessment of tumor stiffness (r=0.51, p=0.089). Scatter along the x-axis is random and for display purposes only to avoid overlapping data points.

Figure 5

Relative performance of MRI and MRE to predict stiffness of meningiomas with intermediate stiffness. a. Plot of MRI prediction of tumor stiffness versus the surgeons’ qualitative assessment of tumor stiffness for tumors with a surgical grade of 2 through 4 (r=−0.12, p=0.89). b. Plot of ratio of tumor stiffness to surrounding brain tissue stiffness as measured by MRE versus the surgeons’ qualitative assessment of tumor stiffness for tumors with a surgical grade of 2 through 4 (r=0.53, p=0.17). Scatter along the x-axis is random and for display purposes only to avoid overlapping data points.