Dorsal Column Steerability with Dual Parallel Leads using Dedicated Power Sources: A Computational Model

The overall goal of the following experiment is to assess the media lateral field steerability of a multi-source spinal cord stimulation system in spinal cord stimulation. Concordance of stimulation induced paresthesia over painful body regions is a necessary condition for therapeutic efficacy. Since patient pain patterns can be unique, a common stimulation configuration is the placement of two leads in parallel in the dorsal epidural space.

This construct provides flexibility in steering stimulation, current media laterally over the dorsal column to achieve better pain. Paraesthesia overlap. Assessing stimulation steerability is achieved by first creating a finite element model of the human spinal cord to determine the potential fields created by epidural spinal cord stimulation electrodes As a second step, the potentials within the dorsal column are calculated for a variety of cathodic currents delivered from two types of spinal cord stimulation systems.

Multisource, where each implanted contact has a dedicated current source and single source where there is only one current source connected to multiple contacts. Next, the potentials are used to calculate the predicted volume of activation in the dorsal column for each system type. In order to compare the media lateral resolution of fiber recruitment in each system, results show that the multi-source system can target more central points of stimulation in the dorsal column than a single source system and the mean steering step for media lateral steering is 20 microns for multisource systems versus one millimeter for single source systems.

A 50 fold improvement. The ability to center stimulation regions in the dorsal column with high resolution may allow for better optimization of paresthesia pain overlap in patients. The main advantage of using multisource stimulation technique over a single source technique is that the stimulation field can be finally adjusted to selectively stimulate the target neuron in those dose column in each individual patient, which is basically maximizing therapeutic possibility while minimizing side effects.

The mathematical model defined here is a finite element model or FEM of the low thoracic spinal cord and its surrounding environment. The FEM model includes the white and gray matter of the spinal cord, cerebral spinal fluid, dura, epidural space tissue, vertebral bone, and two cylindrical multi contact leads. Each multi contact lead consists of eight cylindrical platinum iridium contacts, separated by one millimeter lengths of insulating polymer.

Eight contact leads are now considered the clinical standard in spinal cord stimulation. The use of eight contacts evolved from earlier devices that had only two or four contacts. Studies have suggested that a greater number of contacts and thereby a greater span of spinal cord coverage is more robust at maintaining therapy over time.

In the face of lead migration, a very common problem, the leads are positioned dorsally atop the dura, separated by 2.6 millimeters. For simplicity. In the simulation, pairs of leads are positioned symmetrically relative to the midline of the spinal cord.

In the model, the thickness of the cerebral spinal fluid layer between the contacts and the dorsal surface of the spinal cord is specified to be 3.2 millimeters. The electrical resistive used in the model are given here. The CSF white matter and dura have the lowest resistivity.

The epidural fat and vertebral bone have high resistivity depicted here is the mesh of the FEM model for the spinal cord and multi contact lead. The volume is meshed with over 100 million nodes with a high density mesh in the region close to where the electrodes are located. Only the high density parts of the mesh are shown here.

The mesh is segmented into sections of variable node density. The separation of nodes near the contacts is less than or equal to 300 microns. The node separation of the insulator dura and spinal cord is less than or equal to 750 microns.

The node density of the epidural space is less than or equal to 3000 microns, and the node separation of the vertebral bone is less than or equal to 5, 000 microns. The 3D model of spinal cord geometry is created using a combination of features from relevant literature sources. Dorsal column fibers are placed on a regular grid of 200 microns in the media lateral direction, 100 microns in the dorsal ventral direction and projected into the stro coddle direction.

Once the leads are positioned within the model, two types of stimulators are implemented by defining the currents for two parallel contacts. For a single source system, there is a single cathode and anode that can be connected to any set of contacts when implemented. In the model, only the polarity of a contact is defined and the current through any contact is dependent upon the number of contacts connected to any pole.

For example, if a single contact is connected to the cathode, that contact sinks all of the cathode current. However, if two contacts are connected to the cathode pole of the source, then they split the current in half 50%each. Of course, in the mathematical model, the impedance of every contact is identical, which is not typically true in clinical application.

This implies that the current delivery to each contact in a single source system is not controlled. For our model investigations, we used three possible methods to deliver current In the first method. The leftmost cathode has all the current in the second, the two cathodes each deliver 50%of the current, and in the third, the rightmost cathode delivers all the current For the multi-source system, each contact is defined to have its own current source controllable in 1%incremental current changes between the contacts.

In other words, if the total current delivered to the two contacts is 10 milliamps, the current to each contact is specified to be any fraction of the total, so long as the sum of the currents through each contact equals 10 milliamps. For example, if the left contact delivers 6.8 milliamps, the right contact would then deliver 3.2 milliamps. The 100 fractional splits of current are programmed in this manner.

The cathode is the primary contact involved in neural stimulation. Activation of the cathartic contact causes current flow toward the contact electric current flows along the axon, but also through the inside of the axon and through the axon membrane. At the nodes of rvi, these currents lead to depolarization in the middle of the axon and hyperpolarization at the lateral parts of the axon.

The mathematical function that models the depolarization and hyperpolarization is called the activating function. The activating function is an approximation of the change in the transmembrane potentials. When extracellular stimulating current is applied to neural tissues in a given electrode and fiber geometry, the activating function is then applied to the spinal cord, dorsal column fibers.

The region of activation is defined as the locus of fibers. In the model where the activating function exceeds a predetermined threshold, the activating function is used to predict regions of activation in the dorsal columns. The central point of stimulation is defined and calculated as the geometric oid of the three dimensional region of activation To determine stimulation amplitude, one contact from the left column lead is specified to be a cathode.

In a monopolar configuration, the stimulation amplitude is then iteratively increased until the first fiber is activated. This is always a dorsal column.Fiber. This first activation is assumed to correlate to the first perception of paresthesia by a patient in a clinical setting.

In the model, the current is then increased to 1.4 times the current needed to activate the first fiber, and the OID of the resulting region of activation is calculated. OIDs of all 100 steering steps starting at 100 to zero and going through zero to 100 are computed with the amplitude determined as in the previous step. The average resolution of OID change is OID location range divided by current steps.

The finite element model assumes two parallel leads, place 2.6 millimeters apart of eight contacts each. A single source device that provides a single shared power source for all contacts can target three central points of stimulation. When shifting stimulation media laterally, this corresponds to a step size of one millimeter on average with a two millimeter lead separation, a device with a dedicated power source for each contact can target 100 central points medial laterally in the dorsal column.

When fractionalizing current in 1%increments, this corresponds to a step size of 0.02 millimeters for 1%steps when steering stimulation medial laterally between dual leads. The computational model presented here predicts that a device with independent current sources for each contact can target more central points of stimulation in the dorsal column than a single source system. As a result of this, the resolution of adjustment of the central point of stimulation is 20 microns with the multi-source system and approximately 50 fold increase in resolution compared to single source systems, The ability to steer stimulation reason in the doer column with a high resolution may allow for better optimization of procedure pain overlap.

In patient that is in a given patient, the reason of activation in those of color may be focused to maximize coverage of painful area while minimizing side effect. Therefore, the use of a fractionized current from independent source enables precise steering of stimulation on target nerves in the Doser column during the spinal cord stimulation therapy.