Acute pressure changes in the brain are measurable with MR Elastography: Initial feasibility in an in vivo porcine model
Arvin Arani1, Paul Min1, Nikoo Fattahi1, Nicholas M Wetjen1, Clifford Jack1, Kendall H Lee1, Richard L Ehman1, and John Huston III1

1Mayo Clinic, Rochester, MN, United States

Synopsis

Hydrocephalus is a common medical condition that results from obstruction to the flow of cerebral spinal fluid (CSF) or resorption of CSF. No non-invasive method offers direct measurement of intracranial pressure (ICP). Magnetic resonance elastography (MRE) is capable of non-invasively measuring brain tissue stiffness in-vivo, and may act as a surrogate to ICP. The objective of this study was to investigate the impact of ICP on brain stiffness using MRE in a porcine model. This study shows that MRE brain stiffness changes directly correspond to changes in ICP, motivating future investigation into shunt-therapy monitoring in a patient population.

Purpose

Hydrocephalus is a common medical condition that results from obstruction to the flow of cerebral spinal fluid (CSF) or resorption of CSF. The clinical symptoms can be confused with many other medical conditions and no noninvasive method offers direct measurement of intracranial pressure (ICP). Shunt tube placement has been shown to be an effective therapy in many cases. This procedure involves improving CSF drainage, in turn reducing ICP. A non-invasive measurement or surrogate to ICP may help with predicting patient requiring shunting and monitoring shunt-therapy outcomes. Magnetic resonance elastography (MRE) is a non-invasive imaging technique capable of measuring brain tissue stiffness, and may act as a surrogate to ICP (1). The objective of this study was to investigate the impact of ICP on brain stiffness using MRE in a porcine model.

Methods

Two pediatric catheters were implanted in the left and right lateral ventricles of 3 pigs using a previously described intra-parenchymal drug delivery system technique (2) for targeting (Figure 1). The CSF was drained post-surgery and a saline line was connected to one catheter and a pressure monitor was connected to the second catheter. The pressure in the ventricles was systematically increased by raising the saline bag a certain distance above the pig’s head. MRE was conducted post-surgery at 60Hz, 90Hz, 120Hz, and 150Hz at different ICP values ranging between 0-55 mmHg for each pig. To establish a baseline stiffness value, MRE was conducted pre-surgery at the same vibration frequencies. Image acquisition included a modified spin-echo echo planar imaging sequence to acquire MRE data with the following imaging parameters: TR/TE = 3600-3800/46-62 ms; FOV = 24 cm; 72x72; 35 contiguous 3-mm-thick axial slices; vibration frequency matched motion-encoding gradient; motion-encoding in all 3 directions; and 8 phase offsets. MRE post-processing involved: 1) Applying a [5x5x5] cubic smoothing filter, 2) calculating the curl of the displacement images, 3) calculating the stiffness (direct inversion). The mean and median magnitude of the complex shear modulus |G*| was reported.

Results

The anatomical change in ventricular size with increased pressure has been shown in Figure 2 for pig #1. The apparent change in the mean and median of |G*| over the entire brain volume post-surgery and after CSF drainage is shown in Figure 3, for all vibration frequencies tested. For pigs 1 and 2 the ICP was reduced to 0 mmHg, which resulted in decrease in |G*|, which is especially noticeable at higher frequencies. The mean and median |G*| over the entire brain volume has been plotted in Figure 4, as a function of ICP and vibration frequency. Sample elastograms illustrating the observed changes in brain stiffness over the span of the experiment for all three pigs has been shown in Figure 5.

Discussion

This study demonstrated that changes in brain stiffness correspond with variations in ICP in a porcine model. In the first two pigs, drainage of CSF was shown to decrease the mean brain stiffness by 1.2-1.7 kPa at a vibration frequency of 150Hz. Proper drainage of CSF in pig #3 was not possible due to surgical complications. This resulted in a baseline ICP of 26 mmHg instead of 0 mmHg, showing no apparent change in |G*| between pre-surgical and post-surgical scans. From Figure 4 it appears that there is a threshold over which brain stiffness does not increase with elevated ICP. In this small sample size this appears to be between 20-24 mmHg. Once, past this threshold the brain seems to stiffen with increased ICP. From a mechanical standpoint, increases in tissue stiffness result when a tissue is compressed such that the relationship between stress and strain becomes non-linear causing a rise in stiffness. The 20-24 mmHg range may represent this transition in the porcine brain. Alternatively, from Figure 5 it becomes clear that the distribution of stiffness is not uniform throughout the entire brain, which may indicate that only small regions of tissue are increasing in stiffness with acute changes in pressure and that these do not become significant enough to affect the entire brain volume until higher ICP values are reached.

Conclusion

Brain MRE is able to measure changes in total stiffness across the entire brain volume both due to CSF drainage and with increased ICP. This study motivates future investigation into the changes in brain stiffness at different time points (pre and post shunt), and in different brain regions, in order to evaluate the potential role of brain MRE for therapy monitoring.

Acknowledgements

This research was supported by Theodore W. Batterman Family Foundation and R01 grants EB001981.

References

1. Hatt A et al. AJNR American journal of neuroradiology 2015;36(10):1971-1977.

2. Kim I et al. J Neurosci Methods 2014;227:29-34.

Figures

Figure 1: A) Catheter trajectory planning using high resolution T1-weighted images (oblique-coronal view (left) and axial view (right)). B) Post-surgery catheter implantation. C) Stereotactic frame (STF) and MRE driver positioning. D) Post surgery MRI image of catheter placement.

Figure 2: Anatomical images of ventricles under different pressure levels post catheter implantation in pig #1.

Figure 3: Apparent change in |G*| post-surgery (|G*baseline|-|G*post-surgery|) as a function of vibration frequency.

Figure 4: Mean and median |G*| as a function of frequency and intracranial pressure for all 3 pigs.

Figure 5: Elastograms (|G*|) of pre-surgery, post-surgery at baseline and post-surgery at maximum intracranial pressure (ICP) in a 3 pigs and at a 120Hz vibration frequency.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
4437