Electric stimulation in the brain cavity wall to reduce its associated symptoms. A generic approach applied in central post-stroke pain.

Millions of people worldwide suffer from neurological symptoms that are associated with an abnormal brain cavity (aBC). This aBC is the anatomical result of neuronal loss, mostly after stroke, brain trauma, surgery or infection. The symptomatology is dependent on the location and extent of the aBC.

We hypothesised that electrical stimulation of the aBC wall can alleviate aBC-associated symptoms. This is reasonable, as there is both experimental and clinical evidence that the aBC wall contains viable tissue and that this tissue can be modulated by electrical stimulation.

Our goal was to test this generic hypothesis in a rat model for central post-stroke pain (CPSP), which is a neuropathic pain syndrome that can develop after stroke, typically but not exclusively affecting the sensory thalamus (ST), then named thalamic syndrome. We chose this disorder because it (1) is associated with an aBC; (2) is poorly understood and this work might lend more insight into its mechanisms; (3) currently lacks effective treatment options; and (4) is probably associated with thalamic hyperactivity, which could be treated more effectively by electrical stimulation ES than hypoactivity.

In the presented work, we first reviewed the 5 published rat models and 4 published mouse models for CPSP, and critically appraised the face (symptom similarity), construct (underlying mechanism similarity) and predictive (similarity in effective and non-effective treatments) validity, as well as the robustness and reproducibility. These models mainly rely on sensory changes in response to thermal (heat and cold) and mechanical stimuli, which are presumed to be similar to those observed in CPSP patients. Although every author claims his model to be valid, much room for improvement exists.

Next, we have attempted to replicate and refine 3 of the published CPSP rat models, all lesioning the ST.

In the haemorrhagic model, in which collagenase (a blood-vessel degrading enzyme) is injected in the ST, we found an optimal protocol in terms of injected dose and volume and target coordinates, that evoked contralesional hypersensitivity for cold and heat and decreased motor performance, but did not yield mechanical hypersensitivity (vs. baseline, sham-operated control rats and ipsilesional limb results). However, we could not replicate the thermal hypersensitivity in a larger group with longer follow-up.

In the excitotoxic model, kainate (a potent excitatory amino acid causing neuronal overactivation and eventually cell death) is injected in the ST. When injected directly, we observed very high early mortality due to status epilepticus, anorexia and adipsia. When injected via a previously implanted cannula, mortality was reduced but only motor impairment and no sensory changes were observed (vs. baseline, sham-operated control rats and ipsilesional limb results). Importantly, the presence of seizures unblinded the observer from day 1 until 2 weeks post-injection.

In the electrolytic CPSP rat model, an ST lesion was made electrolytically (by temporarily inserting electrodes and applying a lesioning current). We found a long-lasting motor deficit but no consistent changes in mechanical and thermal sensibility (vs. baseline, sham-operated control rats and ipsilesional limb results)

As this motor deficit was a consistent, long-lasting and easily measurable aBC-associated symptom, we tested our generic hypothesis focused on this symptom.

First, we implanted 3 linear twisted-bipolar electrodes targeting the ventral aBC wall and electrically stimulated electrolytically lesioned rats during motor tests. In a group analysis, we found no improvement of the motor impairment with electrical stimulation ON vs. OFF.

Next, we implanted a custom-made flexible electrode with 7 electrode contacts that opens flower-wise when pushed through a cannula that connects the brain surface with the electrolytic aBC. Electrical stimulation via this electrode resulted in a significant improvement of the motor performance in 2/8 rats (in 2/6 rats that showed a significant motor deficit vs. pre-lesioning) vs. OFF, as assessed with the rotarod test. No motor improvement was observed with electrical stimulation ON vs. OFF when using the ladder rung walking test.

Next, positron emission tomography after injecting a radioactively labelled glucose analogue in rats with electrolytic ST lesions resulted in decreased tracer uptake in the vicinity of the aBC, extending to the ipsilateral primary and secondary sensorimotor cortices and in increased tracer uptake in the central pontine grey and its direct environment vs. sham-operated rats. Moreover, we have measured electrical signals (local field potentials; LFPs) from the aBC wall. LFPs recorded from the aBC wall appear similar to those recorded from intact brain tissue at the same anatomical location, and display a correlation with paw-specific movement.

Further, we have developed and validated new techniques for electrode tip localisation and electrode implantation, based on computed tomography (CT) imaging. They resulted in similar localisation accuracy but higher cost- and time-efficiency, but not in increased implantation accuracy, compared to conventional techniques, respectively. We have also defined the optimal protocol for iodine immersion to increase the internal brain contrast in ex vivo rat brain CT imaging.

Lastly, we have conducted a prospective clinical study in patients with deep brain stimulation (DBS) for obsessive-compulsive disorder. The use of rechargeable stimulators in this very specific population appeared to be effective, applicable and safe and diminished the need for stimulator replacements. We have also conducted histopathological examinations on a brain donated by a 66-year old female with 11 years of DBS in the subthalamic nucleus for Parkinson’s disease, providing the longest histopathological follow-up of DBS in this target, for this indication and of a patient in whom microelectrode recording (MER) was applied. Interestingly, although we confirmed the histopathological changes around the DBS electrodes as observed prior in studies with shorter follow-up, we did not detect significant changes induced by MER electrode passing.

Taken together, our work has laid an interesting and solid foundation for the hypothesis of aBC wall electrical stimulation, which, once optimised in terms of stimulation parameters, stimulation area and electrodes, has the potential of becoming a therapeutic neurosurgical option in patients suffering from aBC-associated symptoms.