Recovery behavior of the hydrocephalic brain after shunt surgery

Introduction:: Hydrocephalus is a condition in which the ventricular system in the brain enlarges due to an imbalance in cerebrospinal fluid (CSF) regulation. CSF shunt surgery is a common treatment option for hydrocephalus patients. However, predicting shunt outcomes and understanding the mechanism of shunt failure remains controversial and challenging. Medical history and clinical symptoms alone are not reliable indicators to evaluate hydrocephalic conditions and shunt outcomes. Even in patients with similar clinical conditions, neurosurgeons cannot predict shunt outcomes. The uncertainty in shunt outcomes may be due to the complex behavior of the brain when exposed to elevated intracranial pressure (ICP) resulting from hydrocephalus, as well as the unpredictable recovery behavior of brain tissue after shunting. Physiological and biomechanical functions of the brain, and consequently brain disorders such as hydrocephalus, are time-dependent. Hydrocephalus patients often experience extended periods of elevated ICP, rather than an acute onset of symptoms. Additionally, shunt failure may occur months or even years after the initial surgery. Hence, long-term evaluation of brain recovery behavior is not only important but also necessary for a better understanding of the complexities underlying hydrocephalus and its management. This study aims to evaluate the long-term recovery behavior of the hydrocephalic brain after shunt surgery. To achieve this goal, we calculated changes in volume, strain, and creep of CSF and brain, as well as ICP, exerted force from CSF on the brain, brain stiffness, intracranial compliance (ICC), and intracranial elastance (IE) non-invasively using fluid-structure interaction (FSI) simulation over 18 months after shunting.


Materials and Methods:: Among the 39 adult patients with communicating hydrocephalus who were evaluated, 14 patients did not require any shunt revisions or adjustments to the valve over 18 months after shunt surgery. Therefore, only 14 patients who had shunting with improved outcomes without any changes in valve adjustment were included in this study. All patients underwent brain magnetic resonance imaging (MRI) at 8 stages: before shunting and 1, 2, 3, 6, 9, 12, and 18 months after shunting. To compare the deviation of the hydrocephalic condition from the normal state, we additionally enrolled 16 healthy subjects. CSF volume was measured in healthy subjects and patients at each stage using image reconstruction process techniques. Additionally, ICP and CSF velocity were calculated non-invasively for healthy subjects and patients at all stages using FSI simulation. A poro-viscoelastic model was used to define the brain material and CSF was considered an incompressible Newtonian fluid. The input data for the FSI simulations included a 3D model of the head substructures and CSF velocity in the lateral ventricles, obtained from DICOM files and cine-PC MRI data, respectively, which were provided for all healthy subjects and patients in all stages. ICP was measured experimentally using invasive ICP monitoring, and CSF velocity in the aqueduct of Sylvius was measured using cine-PC MRI for all patients before shunting to compare with calculated ICPs and CSF velocities for data validation. Notably, the differences between the calculated and experimental ICPs and CSF velocities were acceptable, being less than 4.4% and 1.0%, respectively.


Results, Conclusions, and Discussions:: Hydrocephalus patients experienced a 16.9% increase in brain volume and significant reductions in both ICP by 70.4% and CSF volume by 80.0% after 18 months of shunting. The changes in brain volume, ICP, and CSF volume reached a stable condition at different time points: 15, 6, and 12 months after shunting, respectively. Unlike CSF, the brain's volumetric strain exhibited a notable increase (threefold) following shunt surgery, whereas the brain's volumetric creep decreased fourfold from 0.020 s⁻¹ to 0.005 s⁻¹. The residual strain values, representing the difference in strain between treated patients and healthy subjects, were -0.69 and 0.15 for the brain and CSF, respectively. The maximum brain stiffness of 2.5 N/ml was observed in the first month after shunting (first consolidation phase), and no significant changes were observed in subsequent months. In the third consolidation phase, after the fifteenth month, the viscous component of the brain could overcome and dampen the applied load on the brain. Contrary to the short-term approach, our long-term calculation indicated a non-monotonic trend between the changes in CSF volume and ICP after shunt surgery. The ICC in the first month after shunting was 14.7 ml/cm H₂O, and ICC increased oscillatory in the subsequent months, whereas IE exhibited non-monotonic changes. Additionally, no significant correlation was found between ICP and ICC, or ICP and IE after shunt surgery. This study also demonstrates the long-term effects of the viscous damping component of the brain on the recovery behavior of the hydrocephalic brain after shunt surgery.

We showed that the brain tissue requires two months to adapt to the ICP reduction after shunting, which may be the relaxation time of the brain tissue after elevated ICP removal. Hence, we suggested a gradual reduction of ICP after shunt surgery using making adjustments to the shunt valve pressure levels. The findings of the present study hold the potential to enhance the treatment strategies and improve shunt outcomes for hydrocephalus patients. The study shed light on the biomechanical properties of brain tissue in the context of shunt treatment, offering valuable insights into the underlying causes and mechanisms of shunt failure.


Acknowledgements (Optional): : We thank the Margaret Hackett Family Program (MHFP)

References (Optional): :