Mar 25, 2025

Public workspaceThe Injectrode - A Truly Injectable Electrode for Dorsal Root Ganglion Stimulation to Treat Pain Protocol

DOI
https://dx.doi.org/10.17504/protocols.io.261ge5b8dg47/v1
The Injectrode - A Truly Injectable Electrode for Dorsal Root Ganglion Stimulation to Treat Pain Protocol
  • Ashley N Dalrymple1,2,3,
  • Jordyn E Ting2,
  • Rohit Bose2,4,
  • Lee Fisher2,
  • Douglas Weber1,2
  • 1Carnegie Mellon University;
  • 2University of Pittsburgh;
  • 3University of Utah;
  • 4Case Western Reserve University
  • Ashley N Dalrymple: Collection of data and creation of code;
  • Jordyn E Ting: Collection of data and creation of code;
  • Rohit Bose: Collection of data and creation of code;
  • Lee Fisher: Conception of protocol
  • Douglas Weber: Conception of protocol and collection of data
Brian P Balo
Icon indicating open access to content
QR code linking to this content
DOI: https://dx.doi.org/10.17504/protocols.io.261ge5b8dg47/v1
External link: https://iopscience.iop.org/article/10.1088/1741-2552/ac2ffb
Protocol Citation: Ashley N Dalrymple, Jordyn E Ting, Rohit Bose, Lee Fisher, Douglas Weber 2025. The Injectrode - A Truly Injectable Electrode for Dorsal Root Ganglion Stimulation to Treat Pain Protocol. protocols.io https://dx.doi.org/10.17504/protocols.io.261ge5b8dg47/v1
Manuscript citation:
Ashley N Dalrymple et al 2021 J. Neural Eng. 18 056068. DOI: 10.1088/1741-2552/ac2ffb
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: September 19, 2024
Last Modified: March 25, 2025
Protocol Integer ID: 108078
Keywords: Injectrode, biomaterials, dorsal root ganglion, electrical stimulation, neuromodulation, injectable electrode for dorsal root ganglion stimulation, dorsal root ganglion stimulation, afferent fiber recruitment by dorsal root ganglion, injectable electrode, injectable polymer electrode, pain protocol, fibers due to stimulation, evoked compound action potential, electrode, injectrode, stimulation, afferent fiber recruitment, traditional cylindrical stainless steel electrode
Funders Acknowledgements:
NIH/NIBIB
Grant ID: U18EB029251
Disclaimer
DISCLAIMER – FOR INFORMATIONAL PURPOSES ONLY; USE AT YOUR OWN RISK

The protocol content here is for informational purposes only and does not constitute legal, medical, clinical, or safety advice, or otherwise; content added to protocols.io is not peer reviewed and may not have undergone a formal approval of any kind. Information presented in this protocol should not substitute for independent professional judgment, advice, diagnosis, or treatment. Any action you take or refrain from taking using or relying upon the information presented here is strictly at your own risk. You agree that neither the Company nor any of the authors, contributors, administrators, or anyone else associated with protocols.io, can be held responsible for your use of the information contained in or linked to this protocol or any of our Sites/Apps and Services.
Abstract
This is a protocol for an experiment to compare afferent fiber recruitment by dorsal root ganglion (DRG) stimulation using an injectable polymer electrode (Injectrode‱) and a more traditional cylindrical stainless steel electrode. This protocol will result in data describing the recruitment rate and evoked compound action potentials (ECAP) of Aα, Aβ, and Aδ fibers due to stimulation at different pulse widths by the stainless steel electrode versus the Injectrode. This protocol will also result in data on the impedance values of both electrodes.
Image Attribution
Ashley N Dalrymple et al 2021 J. Neural Eng. 18 056068. DOI: 10.1088/1741-2552/ac2ffb
Materials
ABCDEFGH
Electronic EquipmentVital Signs MonitoringPre-surgery ToolsSurgical ToolsSurgical AccessoriesImplant ToolsDrugsMisc. Hardware/Docs
Ripple Stim headstageMonitorHeating bagDisposable scalpelMarkerSurgical microscopeKetamine or dexdomitorElectrical tape
Ripple GrapevinePulse oximeterSurgical mats on tableScissors - small pointyBeakers (saline, water)MicrorulersIsofluraneScissors
Ripple recording headstageTemperature probeO2 tank full, hooked upScissors - small round tipStainless steel bowl (trash)SalineDrug Dosage sheet
Ripple cablesRespiration monitorLubeScissors - large round tipCotton ballsEuthasolNotes doc
Magnetic holders for headstagesExtra heart rate montiorGlovesForceps - small straight (x2)Gauze, largeCamera
Nerve cuffsExtra respiration monitorTurnikeyForceps - small curved (x2)Gauze, smallStreaming setup
Injectrode material StethoscopeIV cathetersForceps - normal teethSyringes (for irrigation)Calipers
Injectrode injectorEmergency vent bagPRN adaptersForceps - normal no teethSuction beaker (enzyme)
DRG electrodesSurgical tapeNeedle driverSuction tube with beaker plug
EMG electrodesLine for IV bagPeriosteal elevatorSuction large tube to vacuum
Cooner wireClippersRongeurs (S, L)Suction tip
Electrode for stim testingVacuumHooksSuction stylet
Surgical lightMichel clipsSaugers
Towels and saline in incubatorMichel clip toolBone wax
Magnetic board on tableCauterizerCotton buds
Ties for legsCautery return electrodeSutures 3-0 (X4)
Cautery gel (signa gel)
Cautery BP tip
Piezo drill
Drill bit
Towel clips
Microscissors
Microspatula
Wooden handled bone scraper
Blunt pokers
Troubleshooting
Experiment Model (Animal Anesthetization)
Anesthetized adult male cats with ketamine (intramuscular, 10 mg kg-1) and acepromazine (intramuscular, 0.1 mg kg-1) or dexdomitor (intramuscular, 0.04 mg kg-1).
Intubated cat and administered isoflurane (inhalation, 2%-2.5%) for the duration of the experiment.
Used antisedan (intramuscular, 1:1 dexdomitor dose) to reverse the effects of dexdomitor following the administration of isoflurane.
Administered atropine (intramuscular, 0.05 mg kg-1) to reduce saliva output.
Shaved the back, limbs, and ears of the cat.
Continuously monitored vital signs including heart rate, SpO2, core temperature, respiratory rate, and ETCO2 throughout the experiment.
Continually administered IV fluids (saline, 0.9% sodium chloride) throughout the experiment.

NOTE: Ventilator Settings Used:
Flow 2 L/min
Respiration Rate 8 breaths/min
PIP 7
Assist -0.3
Implant Nerve Cuffs
NOTE: Implanted the nerve cuff electrodes in the hind limb ipsilateral to the stimulated dorsal root ganglion (DRG) around the sciatic, common peroneal, and tibial nerves (Figures 1 and 2).

Placed the cat on the table in supine position.
Lifted and tied the hind limb ipsilateral to the stimulated DRG to the bar of a stereotaxic frame set up on the table.
Made an incision behind the knee.
Dissected tissue bluntly to locate sciatic, common peroneal, and tibial nerves.
Placed one spiral nerve cuff electrode with five contacts (3 mm diameter, contacts 4 mm apart, center-to- center; Ardiem Medical, Indiana, PA, USA) made of silicone with platinum contacts on the sciatic nerve. (Figure 2)
Shorted the proximal, center, and distal contacts together to use as a reference electrode. The second and fourth contact were used for recording. (created two tripolar configurations)
Placed one spiral nerve cuff electrode with three contacts (2 mm diameter, contacts 4 mm apart, center-to-center) around both the tibial nerve and the common peroneal nerve (Figure 2). Shorted the proximal and distal contacts together to create a tripolar configuration with the center electrode used for recording.
Took photos of cuff placement.
Tested stimulation through each electrode on each cuff to verify function. Determined the motor threshold for each electrode contact.

NOTE: Motor threshold is determined by observing twitches in the hindlimb.
Sutured incision closed.  

NOTE: Can staple or suture the incision closed.

Figure 1: Diagram of cat hind limb with sciatic, common peroneal, and tibial nerve locations denoted for nerve cuff placement.

Implant Electrodes
Injectrode Information:
Injectrodes were manufactured by Neuronoff, Inc. (Cleveland, OH, USA) using a similar polymer-conductor variant of the Injectrode as described previously by Trevathan et al (2019). Briefly, we mixed two parts of Pt-curing silicone elastomers (World Precision Instruments, FL, USA) with metallic silver particles (Sigma-Aldrich, MO, USA) and loaded the mixture into a syringe that was calibrated to hold 10 µL of Injectrode material. An insulated silver wire (AS-766-36; Cooner Wire Company, Chatsworth, CA, USA) with de-insulated ends was placed inside the syringe such that it became embedded in the Injectrode material upon curing on one end, while the other end was connected to the stimulator for DRG stimulation (Figure 2). Based on measurements taken from explanted Injectrodes, the lengths and diameters of the Injectrodes ranged from 4.0–6.0 (4.4 ± 0.7) mm and 1.3–3.1 (1.8 ± 0.6) mm, respectively. These parameters correspond to an average surface area of 30.8 ± 12.5 mm2.
Stainless Steel Electrode Information:
We manufactured stainless steel DRG electrodes in-house (University of Pittsburgh) using 1 mm diameter hypodermic tubing (McMaster-Carr, Elmhurst, IL, USA) cut to a length of 4.5 mm and crimped to a Teflon-insulated lead wire (AS632; Cooner Wire Company, Chatsworth, CA, USA). The surface area of the stainless steel electrode was 15.7 mm2, which is 1.96 times smaller than the average surface area of the Injectrode.
After we implanted the nerve cuffs, we repositioned the cat to prone.
Used cautery to create an incision along spinal processes L6-L7.
Implanted electrode through either a partial laminectomy or a burr hole.
Partial Laminectomy
Perform a partial laminectomy of L6 and L7 to expose the DRG.
Place the electrodes on the dorsal surface of the crown of the DRG for stimulation.
Burr Hole
Expose the contralateral side of the DRG through a partial laminectomy as noted above.
Guide the burr hole by matching the locations of the DRG on the exposed contralateral side.
Drill a burr hole through the laminae overlying the DRG (Figure 2).
For Stainless Steel Electrode:
Place vertically in the hole.
For Injectrode:
Inject into the hole by placing the tip of the blunt needle into the hole (implantation video available on the corresponding SPARC dataset: DOI: 10.26275/qjai-jsin)
Filled the open cavity with 5 mL of saline to keep the tissue from drying.
Placed a stimulation ground electrode (AS636, Cooner Wire Company, Chatsworth, CA, USA) with ∼5 cm exposure in between the skin and lumbodorsal fascia ipsilateral to the DRG electrodes for monopolar stimulation.
Took photos of spinal cord, DRG, and electrode placement with transparent ruler and reference in place.
Measured Injectrode with calipers.

Figure 2: Experimental setup. (a) Exposure of dorsal root ganglion (DRG) by either a partial laminectomy or a burr hole. Inset: Injectrode delivered through a burr hole over the DRG. (b) An Injectrode or a stainless steel electrode were placed on top of a DRG for delivering stimulation. Antidromic evoked compound action potentials were recorded using spiral nerve cuffs placed on the sciatic, tibial, and common peroneal nerves. Inset: five-contact nerve cuff placed on sciatic nerve. Cuff contacts were 4 mm apart. Gray nerve cuff contacts (1, 3, 5 for sciatic or 1, 3 for tibial and common peroneal) were shorted together and used as a reference. Green nerve cuff contacts (2, 4 for sciatic or 2 for tibial and common peroneal) were used for recording. P = proximal; D = distal. Inset: Scaled illustration of the stainless steel electrode and Injectrode.

Data Acquisition and Stimulation
Determined the motor threshold by delivering a train of five stimulation pulses to the electrode in the DRG.
Recorded electroneurogram (ENG) signals using Grapevine Neural Interface Processor (Ripple, Salt Lake City, UT, USA) and a single-ended headstage (Surf S2).
Filtered the ENG using a 0.1 Hz high pass filter and a 7.5 kHz low pass filter, followed by digitization at 30 kS s−1.
Delivered monopolar stimulation using a high current ECOG + Stim front end (Ripple, Salt Lake City, UT, USA) or an A-M Systems 4100 Stimulator (A-M Systems, Sequim, WA, USA) at 58 Hz using biphasic, symmetric, cathode-leading pulses.
Stimulation pulse widths were either 80, 150, or 300 μs per phase depending on desired stimulation tests.


Constructed recruitment curves for all cats, electrodes, and DRG locations by varying the stimulation amplitude from below sensory threshold up to motor threshold in steps of 20–250 μA (for coarse and fine steps).
Visualized responses to select the stimulation amplitude range for the recruitment curves.
Interpolated over a 1 ms window to blank stimulation artifacts.
Filtered the ENG using a second-order high-pass filter with a cutoff frequency of 300 Hz, and stimulation-triggered average the responses to the 600 stimulus pulses.
Randomized the order of stimulation amplitudes in the recruitment curve. Repeated 600 pulses at each amplitude with 5–10 s pauses between changes in amplitude.
Repeated steps 19 and 20 for each pulse width.
Electrochemical Impedance Study (EIS)
Conduct after stimulation trials have been run.
Used a three-electrode setup:
Working Electrode: The DRG stimulation electrode (either stainless steel or Injectrode)
Counter Electrode: A platinum wire (CHI115, CH Instruments, Inc., Austin, TX, USA; 32 mm long, 0.5 mm diameter) placed distally in between the skin and muscle of the leg ipsilateral to the active (working) electrode.
Reference Electrode: An Ag|AgCl wire (CHI111, CH Instruments, Inc., Austin, TX, USA; 40 mm long, 0.5 mm diameter) placed in between the skin and muscle of the back near the active electrode.
Used a potentiostat (CompactStat, Ivium Technologies, Eindhoven, The Netherlands) to perform EIS measurements.
Tested EIS frequency values ranging from 1 to 100 000 Hz at eight points per decade and a peak-to-peak voltage of 25 mV.
After Data Collection
Following testing with the electrodes delivered through the burr hole, removed the lamina to confirm the location of the Injectrode.
Injected 86 mg kg-1 euthasol intravenously (IV) to euthanize animal.

NOTE: Euthosol is made of 390 mg ml-1 pentobarb and 50 mg ml-1 phenytoin (anti-convulsant).
Confirmed that animal is deceased.
ENG Analysis
Used MATLAB code to analyze data (can find code on corresponding SPARC data repository: (insert link to repository)).
Determined the presence of an evoked compound action potential (ECAP) at a particular stimulation amplitude by comparing the root-mean-square (RMS) amplitude of the ENG to a threshold value (shown in Ayers et al 2016, Nanivadekar et al 2019).
Used a bootstrapping method to verify consistency of the evoked responses:
For each stimulation amplitude, we generated a stimulus-triggered average of the ENG on each nerve cuff electrode using a random sample of 80% of the 600 total stimulus repetitions. We repeated this step 200 times.
Rectified and smoothed the average waveforms by calculating the RMS of each averaged ENG signal using a 100 μs sliding window with a 33 μs overlap with the previous window.
From each averaged RMS ENG signal, we calculated the ECAP threshold. This is defined as one standard deviation above the upper limit of the 99% confidence interval of the baseline RMS amplitude during a 1 ms period before each stimulation pulse. The threshold is the lowest stimulation amplitude that evoked a response in any of the four nerve cuff electrodes.
Determined if an ECAP is detected. An ECAP is considered detected if 95% of the subsets of random samples were supra-threshold during the same 100 μs window of the RMS ENG signal.
Confirmed the accuracy of the automated approach by visually identifying stimulation thresholds.
Calculated the conduction velocity of the ECAPs using the ENG from the sciatic nerve cuffs by following a procedure described in Fisher et al 2014:
Measure the time difference between the peaks of the first, short-latency ECAP recorded from the proximal and distal tripoles in the five-contact cuff (contacts 2 and 4).
Contacts 2 and 4 are 8 mm apart, center-to-center. Divide that distance by the time difference (dt) between peaks of the ECAPs to obtain the conduction velocity for the lowest latency ECAP.
If multiple ECAPs are present, calculate the conduction velocity of the longer latency ECAP(s):
Determine the distance between the DRG electrode and the proximal tripole using the conduction velocity of the first ECAP and the latency between the onset of the stimulus and the peak of that ECAP.
Divide the distance between the stimulating and recording electrodes by the latency of the later ECAP(s) to obtain their conduction velocity.
Used the calculated distance between the DRG electrode and cuff electrode, and the known conduction velocity ranges for primary afferent axons, Aα (80–120 m s−1), Aβ (35–80 m s−1), or Aδ fibers (5–35 m s−1) to define latency windows from the onset of stimulation for each fiber type.
Used these latency windows to find the ECAP threshold for the different fiber types (Aα, Aβ, and Aδ) and construct recruitment curves for the different fiber type thresholds and each pulse width.
Depicted the peak-to-peak amplitude of the ECAP as a function of stimulation charge (current amplitude × pulse width).
Determined the recruitment rate for the different fiber types by fitting a line to the linear region of the recruitment curves using least squares regression (MATLAB's polyfit function, first order). The recruitment rate is the slope of the best-fit line.

NOTE: ECAPs from Aδ fibers were only detected when stimulation amplitude was very high (close to motor threshold) in the corresponding paper (Dalrymple et al 2021). This may lead to not collecting enough data points to create recruitment curves for Aδ fibers.
Compared the recruitment rates across all trials by normalizing all slope values to the slope of the recruitment curve obtained for the stainless steel electrode and a pulse width of 300 μs.

NOTE: 300 μs is the pulse width typically used in clinical applications of DRG stimulation (Liem et al 2013, Deer et al 2017, Kent et al 2018, Graham et al 2019).
Statistical Methods
Used the Shapiro–Wilk test to test for normality and Levene's test to assess the homogeneity of variance.


Used a student's t-test to compare threshold and recruitment rate means between the L6 and L7 DRG.
Performed a two-way analysis of variance (ANOVA) to test for effects of material (Injectrode or stainless steel) and fiber type (Aα, Aβ, or Aδ) on ECAP thresholds.
Performed two-way ANOVAs to test for significant effects of material and pulse width (80, 150, or 300 μs) on ECAP thresholds and recruitment rate.
Used the Bonferroni correction for multiple comparisons to adjust the resulting p-value.

NOTE: ANOVA and the post-hoc tests were performed using SPSS Statistics (version 26; IBM, Armonk, NY, USA).
Used a paired t-test to compare ECAP thresholds over time using Excel (version 2108; Microsoft Corporation, Redmond, WA, USA).
Performed equivalence testing using the two one-sided t-test (TOST) method (Lakens 2017) in MATLAB using the TOST function (Rastogi 2017) to compare the thresholds of the Injectrode versus the stainless steel electrodes across fiber types and over time.

NOTE: A p-value ⩽0.05 indicated significance in this analysis.
Protocol references
Acknowledgements
We would like to thank the Staff at Magee Women’s Research Institute Animal Facility and Rachel Pitzer for their assistance with animal care and monitoring. We would also like to thank Ritesh Kumar for his technical assistance, as well as the Cohen-Karni lab at CMU for lending us the electrodes for performing the electrochemical measurements.