The Utility of Crossover Motor Evoked Potentials

Unilateral motor cortex stimulation produces motor-evoked potentials (MEPs) on the opposite side of the body (contralateral) relative to the stimulated cerebral hemisphere. When MEPs are also recorded from the same side of the body (ipsilateral), we sometimes refer to them as “crossover” MEPs. Many people record from both sides of the body after unilateral motor cortex stimulation during brain and spine surgery, while others just record from the relevant contralateral muscles. Recording from ipsilateral muscles can serve a number of purposes. But, are these extra recordings even necessary, what do they tell us, and what are their benefits and limitations? That’s the topic of today’s post.

Basic Premise of the Motor Evoked Potential:

When we deliver anodal stimulation to the motor cortex, compound muscle action potentials (CMAPs or MEPs) are elicited from the contralateral side of the body. This is true whether the stimulation is delivered transcranially or directly to the brain. So, you stimulate the right side of the brain to record CMAP/MEPs from the left side of the body, and vice versa. Also, as the theory goes, if you deliver a high intensity stimulus, the current will arc deep into the brain and activate the corticospinal tract bilaterally, resulting in MEPs recorded from both sides of the body (see Figure 1).

Figure 1. Current flow during transcranial and direct brain stimulation via grid electrode are presented schematically. During strong transcranial stimulation, current penetrates deep in the brain, activating both corticospinal tracts. During direct brain stimulation, using a grid electrode current flow is restricted to a single corticospinal tract if one uses low current intensity. From Deletis & Sala (2008).

Crossover MEPs in Spine Surgery

In spine surgery, it is only necessary to record MEPs from the side of the body contralateral to the stimulated hemisphere; however, some folks record from both sides of the body. Let’s see what that might look like using a cervical spine surgery as an example:

Left side MEP test (anodal stimulation of right side of brain), and record from the:

  • Left deltoid
  • Left biceps
  • Left wrist extensor
  • Left triceps
  • Left hand
  • Left tibialis anterior
  • Left abductor hallucis
  • Right hand
  • Right abductor hallucis

And the exact opposite would be true for a right side MEP test.

Why would anyone add 2 (or more) extra recordings? Well, the primary purpose for adding the “ipsilateral control” recording (right hand/foot) is to check for a technical error; to ensure that you have correctly plugged your stimulating electrodes both into the patient and the hardware. So, if you were running left sided MEPs (as above) and you got better responses from the right side of the body, as opposed to the left side, then the most likely explanation is that you’ve plugged in your MEP stimulating electrodes backwards.

A secondary purpose might be to identify the extremely rare situation where the patient’s corticospinal tract does not cross over in the brainstem (MacDonald et al, 2004). If you were to stimulate the right cerebral hemisphere in one of these non-decussating patients, you would get MEPs from the right side of the body. This condition is so rare, it’s hardly even worth mentioning.

I never use these ipsilateral control recordings in spine surgery because I think of them as being cognitive noise – excessive information that provides little-to-no benefit and detracts from my ability to quickly and accurately interpret data. What I do instead of using the ipsilateral control recordings, in the rare cases where it is necessary to perform such a technical check, I simply reverse the polarity of the stimulus. If that doesn’t work, then I get up out of my chair and physically perform the check on both ends to make sure my wires are plugged in and positioned correctly. Sometimes I do both, depending on the situation.

As you will see below, it doesn’t even matter whether or not you actually record crossover MEPs in your ipsilateral control muscle recording during spine surgery, but it does matter during brain surgery!

Crossover MEPs in Brain Surgery

Just like spine surgery, one can use ipsilateral control recordings in intracranial surgery as a technical check to make sure everything is plugged in correctly; however, this is not the real reason why we chose to include the extra recordings. Rather, they are included during intracranial surgery to demonstrate focal stimulation. You always have to think about the at-risk tissue, and whether or not your stimulation and recordings are bracketing the site of risk. So, when monitoring long-track motor function, you want to stimulate the corticospinal tract on one side of the nervous system, relative to the location of the lesion/risk, and record CMAPs on the other end of the nervous system. If the system is compromised between the site of stimulation and recording, then the ability to record a CMAP should be compromised as well. That’s what I meant by bracketing the site of risk.

If you stimulate with too much intensity, you are increasing the odds that you may bypass the area of concern. So, if there is a lesion around the eloquent cortex, and you stimulate too high, your current may arc around the pathology and activate the corticospinal tract deeper into the brain and beyond the pathway (refer back to Figure 1). This would result in recording a CMAP that looks normal despite a potentially compromised pathway. The preservation of MEPs despite paralysis is a false negative.

In the case of brain surgery, you do not want to see crossover MEPs from the ipsilateral control muscles because it indicates that you may be stimulating too deep into the brain and activating the corticospinal tract bilaterally. So, absence of crossover MEPs helps to determine that you are delivering focal stimulation. During intracranial cases, and particularly when the tissue at risk is near the cortical surface, it’s more or less advisable to get suboptimal CMAP data with no crossover MEPs as opposed to better data with crossover.

Two papers included in the references (below) address the fact that high stimulation can reach as far as the lower brainstem or foramen magnum (Rothwell et al, 1994; Szelényi et al, 2007).

In spine surgery, you don’t need to limit stimulation intensity to bracket the site of risk because the stimulation will not arc below the foramen magnum and approach the site of risk. Even if both corticospinal pathways are activated in the cranium, this doesn’t matter since it’s cephalic to the site of surgery, so you’re still bracketing the site of risk. The situation is different in brain surgery. Adding the additional recordings will give you some degree of feedback regarding how localized your stimulation is; allowing you to optimize your stimulation parameters to produce more focal activation of the corticospinal tract.

Limitations and Pitfalls of Crossover MEPs

Absence of crossover MEPs from ipsilateral recording locations does not guarantee that you are bracketing the site of risk. It only serves as a tool to guide the process.

Presence of crossover MEPs from ipsilateral recording locations does not guarantee that you are bypassing the site of risk. We use anodal stimulation because it is most effective for evoking MEPs. Keep in mind that the cathode (on the other side of the brain) is quite capable of evoking MEPs, too.


Recordings from ipsilateral muscles are sometimes used in spine surgery and their main purpose is a technical check to make sure your electrodes are plugged in correctly. In spine surgery, it doesn’t matter whether or not you actually record crossover MEPs. Recordings from ipsilateral muscles are used frequently in intracranial surgery to guide optimization of stimulation parameters and demonstrate focal stimulation. As a general rule, you do not want to record crossover MEPs in intracranial surgery. One must be aware of the limitations associated with using ipsilateral recordings, particularly in intracranial surgery.

Portions of this post were originally written by Dr. Adam Doan and were included in this post with his kind permission.


  1. Deletis D, Sala F. Corticospinal tract monitoring with D- and I-waves from the spinal cord and muscle MEPs from limb muscles. In: Nuwer MR, editor. Intraoperative Monitoring of neural function. Vol. 8, Handbook of clinical neurophysiology. Amsterdam: Elsevier; 2008. p. 235–51.
  2. MacDonald DB, Streletz LJ, Al-Zayed Z, Abdool S, Stigsby B. Intraoperative neurophysiologic discovery of uncrossed sensory and motor pathways in a patient with horizontal gaze palsy and scoliosis. Clin Neurophysiol. 2004 Mar;115(3):576-82.
  3. Rothwell J, Burke D, Hicks R, Stephen J, Woodforth I, Crawford M. Transcranial electrical stimulation of the motor cortex in man: further evidence for the site of activation. J Physiol 1994;481(Pt 1):243–50.
  4. Szelényi A, Kothbauer KF, Deletis V. Transcranial electric stimulation for intraoperative motor evoked potential monitoring: Stimulation parameters and electrode montages. Clin Neurophysiol. 2007 Jul;118(7):1586-95.

13 thoughts on “The Utility of Crossover Motor Evoked Potentials

  • Reply Cody April 12, 2016 at 12:20

    Great article Dr Vogel

  • Reply Tyson Hale April 15, 2016 at 14:55

    Nice write up!

    We call it “co-activation” when we’re performing meps. Sometimes they are tough to avoid in brain surgeries (whether lesion obstructing anatomy or other reason) so we teach and make sure all of our neurosurgeons are aware of these crossover meps and what it could potentially mean. We document when this occurs and that the surgeon understands limitations.

    Sometimes we see UE cmaps without crossover but when increasing stim to elicit the LE cmaps, we’ll see the coactivation present. In these situations we’ll document 2 sets of intensities, and inform surgeon.

    Neat stuff!

    • Reply Richard Vogel April 25, 2016 at 14:30

      Hi Tyson, thanks for the comment. I actually like the phrase “co-activation” better than “crossover”. Regarding the second part of your comment, I will sometimes use multiple stimulation montages to maximize recordings for different areas of the body (i.e., upper vs lower extremity). Or, sometimes we need to place a small cup electrode, or strip electrode, on the surface of the brain to get lower extremity MEPs. I think Lanjun Guo talked about this in her recent ASNM webinar. Great points!

  • Reply Ahmed Imran April 20, 2016 at 18:07

    Nice information, Thanks.

  • Reply MD April 22, 2016 at 13:00

    Nice. Depending on who you read, 80-90% of the tracts are believed to decussate. Any concerns with the 10-20% of the tracts that remain ipsilateral? Or, are taking advantage of the fact that when we stim MEPS we are recruiting only a small percentage of the motor neuron pool anyway?

    • Reply Richard Vogel April 25, 2016 at 08:46

      Hi. Thanks for the comment. The non-decussating tracts form the anterior corticospinal tract and mostly innervate the trunk. This is generally not a concern in brain surgery, and certainly not in spine surgery; however, I will note that on one particular occasion we did perform motor cortex mapping specifically to identify the eloquent motor area for abdominal function as an access point to a tumor, and we had to be mindful of this non-decussating tract. It is also suggested that the aCST perhaps innervates the most proximal muscles of the limbs, but this unlikely relevant for neurophysiology given observations that unilateral transcranial motor stimulation does not readily trigger MEPs from ipsilateral proximal extremity muscles (such as deltoids). Finally, you are correct, most references in the literature suggest that MEP stimulation activates ~3% of the CST, which in turn activates a small percentage of the motor neuron pool indeed.

  • Reply Alan Forster April 23, 2016 at 19:22

    Completely agree with the fear of excess stimulation bypassing site of risk – worrying when the subdural strip is ‘in the way’ so surgeon pushes it to one side, and only monitoring obtainable is at higher current than I’m comfortable with – getting comments such as ‘but you’re still getting responses so that’s ok then’….!!!!!….giving false sense of security. I’ve followed habit of 1 subdural strip contact as anode, to forehead or mastoid subdermal needle as cathode: from your diagram are you using adjacent strip contacts to limit stimulation spread???
    Nice article! Thanks!

    • Reply Richard Vogel April 25, 2016 at 08:50

      Hi. Great comment, and great question. I do the same thing that you do, using surface electrodes as anode, referenced to a cathode in the scalp. I know that some people do, in fact, use adjacent electrodes in an attempt to limit spread. I don’t see the difference, given the mechanism by which we depolarize the pyramidal cells.

  • Reply Kent Rice April 25, 2016 at 11:22

    We always record bilaterally for all TcMEP polarities because of occasional unexpected results. Ensuring that only contralateral muscles are activated during TcMEP monitoring for brain surgery is probably the most practical way to minimize chances of bypassing the area at risk. However, don’t forget that the cathode is also capable of eliciting MEPs, just not as effectively as the anode. Thus, bilateral responses don’t necessarily indicate deeply penetrating current.
    Good post.

    • Reply Richard Vogel April 25, 2016 at 14:24

      Hi Kent, thanks for your comment. There was one section at the end of the post that you probably missed… “Presence of crossover MEPs from ipsilateral recording locations does not guarantee that you are bypassing the site of risk. We use anodal stimulation because it is most effective for evoking MEPs. Keep in mind that the cathode (on the other side of the brain) is quite capable of evoking MEPs, too.”

  • Reply umara May 20, 2016 at 17:14

    interesting to read thank you

  • Reply Etienne Pralong July 17, 2017 at 04:53

    Nice exemple and comment, thanks. Another exemple of normal MEP and temporary loss of motor function is lesion of premotor cortex causing hemiparesia

Leave a Reply

%d bloggers like this: