Dorsal Column Mapping for Intramedullary Spinal Cord Tumor Surgery

Surgical treatment of intramedullary spinal cord tumors carries significant risk for morbidity. In order to extract the tumor, the surgeon must perform a laminectomy, open the dura, and cut into the spinal cord (called a myelotomy). Neurologic function is usually monitored during these procedures with a multi-modality approach that includes transcranial electrical motor evoked potentials (MEP), somatosensory evoked potentials (SSEP), electromyography (EMG), electroencephalography (EEG) and train-of-four (TOF). For monitoring motor function, we generally record compound muscle action potentials (CMAP) from the periphery, as well as corticospinal volleys (D-waves) from an epidural electrode placed along the midline of the spinal cord.

As I always say, monitoring is very different from mapping, and these procedures also require the neurophysiologist to map (identify and preserve) certain structures in order to maximize surgical outcomes. In order to preserve somatosensory function, the dorsal columns must be mapped so that the myelotomy can be performed safely. Whenever possible, the best place to perform a myelotomy is along the physiologically-silent region between the gracile fasciculi on the posterior spinal cord. This usually coincides with the location of the dorsomedian raphe (Gonzalez et al., 2013).

874px-Spinal_cord_tracts_-_English.svg

The dorsal columns transmit proprioceptive info (vibration, joint position, light touch), and damage to these structures often results in sensory ataxia and/or painful dysesthesias. Post-operative dorsal column dysfunction is normally observed in 45-50% of patients undergoing resection of intramedullary tumor; however, dorsal column mapping reduces the incidence of dysfunction to approximately 9% (Mehta et al., 2012). So, the technique serves an important role in optimizing surgical outcomes.

There are three common methods for mapping the dorsal columns. I have used all three, successfully; however, I also have my favorite, go-to method (Technique #2), which I have used for years – even before it was published. The techniques are listed below, and the methods/settings/etc are the same as what you will find in the literature. As always, do not perform these tests unless you have been adequately trained to do so.

References from Introduction:

  1. Mehta AI, et al. (2012). Dorsal column mapping for intramedullary spinal cord tumor resection decreases dorsal column dysfunction. Journal of Spinal Disorders and Techniques, 25, 205–209.
  2. Gonzalez AA, Shilian P, Hsieh P. (2013). Spinal cord mapping. Journal of Clinical Neurophysiology, 30(6), 604-612.

Technique #1:  Retrograde

Theory: 

Antidromic sensory nerve action potentials (SNAPs) can be obtained by stimulating the spinal cord and recording from a peripheral nerve.  Thus, stimulation of a gracilis tract will result in a SNAP recording from the ipsilateral posterior tibial or peroneal nerve.

Technique:

Give surgeon a side-by-side bipolar electrode with 2-3 mm inter-electrode distance.  Begin on one side of spinal cord about 5 mm from midline.  Make sure to orient the electrode tips parallel to the long axis of the spinal cord with cathode caudal to anode.  After each stim/record, have the surgeon move closer to midline in 1 mm increments until no response is recorded.  Then, do the other side.

The area not responding to right and left stimulation corresponds to the functional midline (dorsomedian raphe) and selected for the location of the midline myelotomy.

Stimulation:

  • Frequency: 1 Hz (monophasic square wave pulse)
  • Intensity: 3-8 mA.
  • Pulse width: 200 µsec

Recording:

  • Time Base: 100 msec
  • Filters: 10-30 (HP) and 1500-2000 (LP) Hz
  • Sensitivity: 1-5 µV
  • # Averages: 50-100
  • Montage:
    • LPTN-/LPTN+ (or LPER-/LPER+)
    • RPTN-/RPTN+ (or RPER-/RPER+)

Technique #1 References:

  1. Quinones-Hinojosa et al (2002). Spinal cord mapping as an adjunct for resection of intramedullary tumors: Surgical technique with case illustrations. Neurosurgery, 51, 1199-1207.
  2. Gonzalez et al (2013). Spinal cord mapping. Journal of Clinical Neurophysiology, 30, 604-612.

Technique #2: Anterograde – Simulation of spinal cord and recording from scalp

Theory: 

Because stimulation of a gracilis tract is similar to peripheral nerve stimulation, SSEPs can be obtained by stimulating the spinal cord and recording from the scalp.  Using the C3’-C4’ (active-reference) montage, a change in SSEP polarity occurs as stimulation from the left to right dorsal column is performed.

Surgeon uses a side-by-side bipolar stimulating probe to activate the ascending dorsal columns of the spinal cord.

Surgeon uses a side-by-side bipolar stimulating probe to activate the ascending dorsal columns of the spinal cord.

Technique:

Give surgeon a side-by-side bipolar electrode with 2-3 mm inter-electrode distance.  Begin on one side of spinal cord about 5 mm from midline.  Make sure to orient the electrode tips parallel to the long axis of the spinal cord with cathode cephalic to anode.

When stimulating the left dorsal column, a positive response will typically be recorded on the C3’-C4’ channel, similar to stimulation originating from the left posterior tibial nerve. As you move from left to right across the functional midline, the positive polarity response flips into negative polarity.

The area not responding to right and left stimulation (flat SSEP) corresponds to the functional midline (dorsal median raphe) and selected for the location of the midline myelotomy.

Stimulation:

  • Frequency: 3.17 Hz
  • Intensity: 0.2-1.0 mA
    • Start at 0.2 mA (Going higher too soon can spread current and contaminate results).
  • Pulse width: 300 µsec

Recording:

  • Time base: 5-10 msec (depends on cervical vs thoracic).
  • Filters: 30-300 Hz.
  • Sensitivity: 0.5-2.0 µV/div
  • # Averages: up to 400 may be needed.
  • Montage:
    • CP3-CP4 channel will exhibit phase reversal.
    • CPz-Fpz channel is control and will not exhibit phase reversal.
  • Latency: Cervical was P10 at C2-5 level in paper, P11-12 at C6.

Table:  Ranges reported by Nair et al (2014).

Spinal Level Mapping Latency (msec)
C2 9.9
C3 10.0-10.6
C4 11.1-11.2
C5 11.3-12.0
C7 9.5
T2 12.2
T3 13.4
T7 15.3
T8 15.6-18.4
T9 19.7
T10 16.6

Notes:

Original work used a hand-made, 8-contact mini-electrode (8-bar-style contact, consisting of 2 mm of exposed 40 gauge stainless steel wires, with 1-mm inter-contact spacing) placed on the dorsal spinal cord, perpendicular on the longitudinal axis of the spine.  Dorsal columns were stimulated by activating adjacent leads.

Use of a micro-electrode is delicate and time-consuming, regardless of whether it is used for recording or stimulation. This is due to the fact that sustained good contact with the pulsatile dorsal cord of the 8-contact electrode usually requires several re-adjustments of the position, irrigation, and sometimes even replacement of the electrode due to defective contacts. Also, presence of prominent vasculature of the dorsal cord may be an additional obstacle in easy and safe placement of the recording electrode. Finally, the mini-electrode is prohibitively expensive, without being reusable.

I have successfully mapped the dorsal columns with a bipolar hand-held probe (I prefer side-by-side, but concentric works just as well). A recent paper by Nair (2014), confirms the utility of this technique.

Technique #2 References:  

  1. Gonzalez et al (2013). Spinal cord mapping. Journal of Clinical Neurophysiology, 30: 604.
  2. Nair D, Kumaraswamy VM, Braver D, Kilbride RD, Borges LF, Simon MV. (2014). Dorsal column mapping via phase reversal method: The refined technique and clinical applications. Neurosurgery. 2014 Jan 19. [Epub ahead of print]
  3. Simon et al (2012).Phase Reversal of Somatosensory Evoked Potentials Triggered by Gracilis Tract Stimulation: Case Report of a New Technique for Neurophysiologic Dorsal Column Mapping. Neurosurgery, 70: E783-E788.

Technique #3: Anterograde – Peripheral stimulation and recording from spinal cord

Theory:

Orthodromic sensory potentials can be obtained by stimulating the posterior tibial nerve and recording from an electrode array placed over the dorsal columns of the spinal cord.  The recorded volleys are the product of electrical signals of lower extremity sensory information traveling via the gracillis tract within the dorsal columns.

Technique:

Requires a specialty electrode – 8-contact micro-electrode (8 bar-style contacts, consisting of 2 mm of exposed 40 gauge stainless steel wires, with 1-mm contact spacing).  See notes regarding electrode.  Have surgeon place the electrode on the dorsal spinal cord, perpendicular to the longitudinal axis of the spine.

Stimulation is performed using similar stimulation parameters as those used for SSEP stimulation. Stimulation is performed from each tibial (or other) nerve individually, which will produce distinct and recordable responses in the ipsilateral spinal cord. The response is about 10 msec of multi-spike activity.

Because of the somatotopic representation of lower limbs being closer to the midline, the highest response that can be elicited is also located closest to the neurophysiological midline, with a gradual reduction in amplitude observed as recording is performed away from the midline. The neurophysiological midline is then located in the electrode adjacent to the highest responses obtained from both left and right posterior tibial nerve stimulations.

Stimulation:

  • Frequency: 17 Hz.
  • Intensity: 40 mA.
  • Pulse width: 100-200 µsec (300 is probably fine).

Recording:

  • Time base: 5-10 msec (depends on cervical vs thoracic).
  • Filters: 20-2000 Hz.
  • Sensitivity: 5-10 µV/division
  • #Averages: ~100.
  • Montage:
    • 1-8
  • Latency: Responses were ~30 msec in the Krzan chapter.

Technique #3 References:

  1. Krzan MJ (2002). Intraoperative neurophysiological mapping of the spinal cord’s dorsal columns. In: Deletis V, Shils JL, editors. Neurophysiology in Neurosurgery. New York: Academic Press; pp 153.
  2. Gonzalez et al (2013). Spinal cord mapping. Journal of Clinical Neurophysiology, 30: 604.

That’s it for now. Feel free to leave comments!

Dr. Rich Vogel is board-certified intraoperative neurophysiologist working for Safe Passage Neuromonitoring. He started the Neurologiclabs website and blog to connect with others in the field of neuromonitoring.

NeuroLogicLabs

12 thoughts on “Dorsal Column Mapping for Intramedullary Spinal Cord Tumor Surgery

  • Reply diane bouchard April 28, 2015 at 05:27

    i will read this with great pleasure, i am asked to do this by my surgeon, did it once but with a monopolar stimulator and found it not so reproduceable. i will read carefully and try again.

    i will let you know of my success :))

    thanks

  • Reply John N. Gardi April 30, 2015 at 17:31

    Rich,

    I suggest a fourth option – stimulate dorsal tracts with a 1.0 mm ball-tipped electrode and look for peripheral muscle responses in all the normal muscle groups used for MEPs. It is suspected that by doing so one is invoking sensory feedback, reflex-like activity in motor tracts and eventually in post-synaptic activation of peripheral, lower motor neurons. It works like a champ and I have used it with a lot of success in select cases where dorsal column stimulators are placed in the midline as well as these more challenging intra-medullary tumor cases. Of course, the beauty is you get immediate feedback as the surgeon tracks the electrode horizontally across the dorsal columns just like a CMAP (albeit 1.0-2.0 ms longer latency). Also you can use a multi-pulse stimulus as well, but in most cases the response is very sensitive. Most cases I got responses to single pulse stimulation at 1.0-2.0 volts. And it follows the somatotopic layout of the dorsal columns where S1-3 activity patterns appear closest to the midline. The more interesting and potentially difficult result to contend with is what happens when a subsurface tumor has distorted the true position of the midline or has distorted dorsal tracts deep to surface. For these reasons I insist that the surgeon not only re-map at various longitudinal levels along the intact dorsal tracts, but also ask them to continue to map within the myelotomy and until he or she reaches the tumor.

    John Gardi

  • Reply Dr. Nishanth Sampath MD October 16, 2015 at 06:51

    Dear Rich,

    I am an intraoperative neurophysiologist from India.

    I really enjoyed reading this write-up.

    Could you recommend a good brand of 8-contact micro-electrode?

    Thanks,
    Nishanth

    • Reply Richard Vogel October 18, 2015 at 12:44

      To my knowledge, there are no companies making this electrode at the moment. The ones in the literature are hand-made. I think Ad-Tech used to make one, but it wasn’t profitable, so it was discontinued. Please let me know if you find one.

      • Reply Dr. Nishanth Sampath MD October 27, 2015 at 08:20

        I am badly in need of one such microelectrode array. Could you help me out with the same?

        Or could you guide me through as to how to make such electrodes by myself? If I succeed, I can contribute to a blog “How to make an in-house multi-contact electrode for dorsal column mapping?”

        Thanks,
        Dr. Nishanth

        • Reply Richard Vogel October 27, 2015 at 15:59

          Hi. If you look at the Gonzalez paper, cited above, there is a description and image on page 6.

  • Reply Aiman Mahfouz, DABNM, CNIM August 4, 2016 at 00:21

    I had the good fortune of reading your article . It is well-written and contained sound, practical advice. You pointed out several things that I will remember for my next case, I look forward to reading your next informative work. Thank you.

  • Reply Carmine October 12, 2016 at 20:16

    This article is really interesting. I have bookmarked it.
    Do you allow guest posting on your page ? I can write high quality articles
    for you. Let me know.

  • Reply Choi Young Doo February 10, 2017 at 01:30

    Hello. I’m a neuromonitorist in SNUH, South Korea.
    We are testing dorsal column mapping these days.
    Methods are similar with ours.

    If you don’t mind.
    Can i just see of your mapping signal figure?
    i’m waiting your answer pleasely.

    thanks.

Leave a Reply

%d bloggers like this: