ASNM's Updated EMG Position Statement: What Changed Since 2005, and What It Asks of the IONM Technologist

The first rewrite of the society's EMG guidance in 21 years tightens technical recommendations, questions some long-standing habits, and treats communication as part of the modality itself.

In 2005, the American Society of Neurophysiological Monitoring (ASNM) published its position statement on intraoperative electromyography (EMG).¹ For the next two decades, it served as a primary reference for how the field recorded, interpreted, and reported EMG, even as surgery changed around it.

ASNM has now released its replacement, published in the Journal of Clinical Monitoring and Computing.² Much of it codifies what experienced teams already do. Some of it is direct about practices the authors believe are falling short.

Here is what's in it, what's been removed, and what it could mean the next time you set up a case.

‍

A Brief Primer for Non-Specialists

EMG records electrical activity from muscle. The famous neurophysiologist, Sir Charles Sherrington, described the motor neuron as the “final common path” between the nervous system and muscle.³ Watching the muscle tells the monitoring team what is happening to the nerve upstream.

EMG comes in two forms:

• Spontaneous ("free-run") EMG is passive listening. When an instrument stretches, heats, or bumps a nerve, the muscle it supplies may fire. The neuromonitoring team sees and hears this in real time and alerts the surgeon.

• Triggered EMG is active. The surgeon stimulates a structure with a probe, and a muscle response confirms that a motor nerve is present and conducting.²

However, silence does not prove safety. A nerve that has been slowly compressed, lost its blood supply, or even been cleanly cut may produce no EMG activity at all.² That is why EMG is usually one part of a multimodality plan alongside motor and sensory evoked potentials.

‍

What's Changed from the 2005 Statement

Neuromonitoring Associates' Director of Research compared the new statement with its 2005 predecessor. The pattern is a document that narrows its focus in some areas and expands it in others.

‍

New clinical ground

Cranial nerve monitoring gets far more attention. Three additions stand out:

• Guidance on shaft-insulated and longer needles to limit cross-talk from neighboring muscles, such as facial muscle activity bleeding into masseter or extraocular channels.

• Triggered-EMG mapping of the fourth ventricle floor.

• Lateral spread response monitoring during microvascular decompression for hemifacial spasm.²

Spinal coverage expands to lateral transpsoas interbody fusion and to EMG-guided placement of spinal cord and dorsal root ganglion stimulator leads.²

Transabdominal motor-evoked potential studies are covered but explicitly labeled emerging and investigational. The authors advise new users to be circumspect about early claims. They recommend pairing these studies with transcranial electrical motor evoked potentials (TcMEP) for now, and they note that recordings appear likely to be confounded in patients with larger waist circumference.²

‍

A shared vocabulary, finally on paper

The statement adopts the A-, B-, and C-train nomenclature for spontaneous EMG patterns, which traces back to Romstöck and colleagues.⁴ Most neurophysiologists have spoken this language for years. It simply was not in the 2005 document.

• A-trains: sustained, high-frequency discharges. The statement considers these the most pathologic and says reporting one should include a request for a surgical pause and a TcMEP.

• B-trains: spikes and bursts, usually tied to brief nerve contact. They are useful feedback but not reliable predictors of injury.

• C-trains: irregular patterns resembling voluntary recruitment, often a sign of a lightening anesthetic.²

The statement adds a point that surprises people outside the field: a discharge's amplitude does not correlate with injury risk. It depends on how close the electrode sits to the firing motor units, not how threatened the nerve is.²

It also formalizes suprasegmentally generated EMG discharges. These are discharges recorded in muscles below the operative level during spinal cord–level surgery. The statement reports that they tend to anticipate TcMEP loss, and that discharges graded as severe have been associated with roughly a 50% risk of nearly immediate TcMEP loss.²

‍

What was dropped

Impedance-based pedicle screw testing is gone. It rarely saw real-world use.

H-reflex monitoring no longer gets its own treatment. Related reflex studies now appear only within thetransabdominal and posterior root reflex discussion.^1,2

‍

Technical Changes That Matter in the Operating Room

The low-frequency filter moves to 30 Hz

The statement notes that a 2–3 Hz low-frequency filter had historically been used for diagnostic EMG. It now recommends 30 Hz for spontaneous EMG, to limit baseline drift and far-field activity from other muscles.

For triggered EMG it gives a 3–30 Hz range, suggesting 30 Hz during nerve mapping when far-field contamination is a concern. It advises against using the 60 Hz notch filter by default, particularly for triggered EMG, because the notch can introduce ringing artifact.²

‍

The subdermal needle gets questioned directly

This may be the most operationally significant section. Standard 13 mm subdermal needles remain acceptable for triggered and evoked responses.

For spontaneous EMG, however, the authors write that the common practice of subdermal recording "needs to be seriously questioned and reconsidered." They draw on Skinner andc olleagues' work showing that surface and subdermal electrodes can miss low-amplitude burst and train discharges that intramuscular electrodes detect.^2,5

For deeper muscles, such as proximal limb and extraocular muscles, they recommend longer needles with a coated shaft and an exposed tip. When a nerve root is at high risk, they also suggest two or more channels within the same muscle.²

‍

The 8 mA pedicle screw threshold, withits caveats

The statement highlights a retrospective analysis of 8,584 lumbosacral pedicle screw stimulations in 1,179 patients. In that analysis, a triggered-EMG threshold of 8 mA or less carried 0.90 specificity for new postoperative deficits. The diagnostic odds ratio for new deficits was 4.34 (95% CI, 1.83–10.27).^2,6 A companion analysis of the same cohort identified 8 mA as the optimal alert threshold among those tested from 4 to 20 mA, although the Youden index was marginally higher at 10 mA (0.23 vs 0.22).^2,7

The same analysis reported sensitivity of only 0.32.^2,6 An alert below 8 mA is meaningful, but a screw that tests above it is not proof of safety.

The statement also still describes 10 mA as a commonly used warning threshold. It notes that screws below 5 mA are the most likely to pose real danger to the nerve root.²

Two related cautions matter just as much as the number:

• Threshold criteria assume a 0.2 ms pulse duration, because that is the basis of most published pedicle screw studies.

• Several factors can artificially raise thresholds and produce false negatives: neuromuscular blockade, chronically compressed roots, current shunting through fluid or conductive dilators, hydroxyapatite-coated screws, and stimulating the polyaxial head instead of the shaft.²

‍

Anesthesia: the concern is blockade

According to the statement, no general anesthetic agent affects EMG monitoring. The primary concern is neuromuscular blockade, and the authors' bottom line is to avoid any level of blockade during periods of neurologic risk when possible.²

Perform train-of-four monitoring in the same or a neighboring muscle group, because recovery varies between muscles. The statement also flags local and regional anesthesia as confounders:

• Lidocaine jelly on endotracheal tubes during recurrent laryngeal nerve monitoring.

• Lidocaine infiltration near the facial nerve.

• Spinal anesthesia for awake lumbar fusion.

• Erector spinae plane blocks, which may spread to motor nerves.²

‍

What This Means for Surgical Neurophysiologists

None of the points below replace your supervising physician's protocols. They are, however, worth raising at your next protocol review:

• Check your EMG channel defaults. Many systems still carry older filter settings, or share settings between EMG and MEP channels. Best practice is to use separate settings for EMG and MEP recorded from the same electrode pairs.²

• Don't assume a more sensitive display setting is helping. On some systems, 10–50 µV/division changes only the analog-to-digital converter range, not the amplification. You lose dynamic range without gaining resolution.²

• Revisit where intramuscular needles belong in your spinal and cranial setups. This is especially relevant for high-risk muscles.

• Report A-trains with a clear ask: pause, then test conduction.

• Don't take automated thresholds at face value. The statement warns that auto-thresholding software is agnostic to waveform morphology and needs human verification of the raw data.²

• Document contemporaneously: confirmation of neuromuscular recovery, surgical stages, notable EMG events, and every stimulation result with its intensity.²

‍

What This Means for Hospital and OR Leaders

For executives, the most useful section may be communication. The authors describe how a "tech-does-all-the-communicating" dynamic, or a "surgeon-is-in-charge" bias, can undermine proper recording and reporting. The result can be a paradox: lots of monitoring but limited benefit.

Their fix is a three-part model:

1. The in-room neuromonitorist must know how to record discharges dependably.

2. The supervising neurophysiologist, in the room or remote, must be ready to lead communication when reporting is incomplete, inaccurate, or ignored.

3. When the surgeon relies on a stand-alone device or is unsympathetic to full monitoring, the team should still respectfully deploy full multimodality intraoperative neurophysiological monitoring (IONM) and report the results expertly.²

The statement also reaffirms that appropriately qualified people should acquire and interpret EMG. It points to the joint ASNM, the American Clinical Neurophysiology Society, the American Association of Neuromuscular & Electrodiagnostic Medicine, and the American Society of Electroneurodiagnostic Technologists guidelines on personnel qualifications, and to credentials such as Certified in Neurophysiologic Intraoperative Monitoring at the technical level.^2,8

For leaders evaluating a program, several practical questions follow:

• Does the monitoring team use intramuscular electrodes where the evidence favors them?

• Is there professional-level oversight?

• Is documentation contemporaneous?

• Does the team have the staffing and supplies to follow updated guidance, or is it still running on 2005 defaults?

‍

The Real Update Is About Responsibility

This statement treats signal acquisition, pattern recognition, and communication as a single chain. A weak link anywhere, whether a short needle, a muted alert, or an unlogged stimulation, reduces the value of everything else.

After 21 years, that framing may be the part of the update that ages best.

‍

‍

‍

References

1. Leppanen RE. Intraoperative monitoring of segmental spinal nerve root function with free-run and electrically-triggered electromyography and spinal cord function with reflexes and F-responses. J Clin Monit Comput. 2005;19(6):437–61. doi:10.1007/s10877-005-0086-2

2. Rice K, Skinner S, Hoffman EM, Oishi T, Guo L, Leppanen R. Intraoperative electromyography (EMG): an updated position statement by the American Society of Neurophysiological Monitoring. J Clin Monit Comput. 2026. doi:10.1007/s10877-026-01489-x

3. Sherrington CS. The Integrative Action of the Nervous System. New Haven: Yale University Press; 1906.

4. Romstöck J, Strauss C, Fahlbusch R. Continuous electromyography monitoring of motor cranial nerves during cerebellopontine angle surgery. J Neurosurg. 2000;93(4):586–93. doi:10.3171/jns.2000.93.4.0586

5. Skinner SA, Transfeldt EE, Savik K. Surface electrodes are not sufficient to detect neurotonic discharges: observations in a porcine model and clinical review of deltoid electromyographic monitoring using multiple electrodes. J Clin Monit Comput. 2008;22(2):131–9. doi:10.1007/s10877-008-9114-3

6. Melachuri SR, Melachuri MK, Anetakis K, Crammond DJ, Balzer JR, Thirumala PD. Diagnostic accuracy of thresholds ≤8 mA in pedicle screw testing during lumbar spine procedures to predict new postoperative lower extremity neurological deficits. Spine. 2021;46(2):E139–45. doi:10.1097/BRS.0000000000003727

7. Melachuri SR, Melachuri MK, Mina A, Anetakis K, Crammond DJ, Balzer JR, et al. Optimal “low” pedicle screw stimulation threshold to predict new postoperative lower-extremity neurologic deficits during lumbar spinal fusions. World Neurosurg. 2021;151:e250–6. doi:10.1016/j.wneu.2021.04.022

8. López JR, Ahn-Ewing J, Emerson R, Ford C, Gale C, Gertsch JH, et al. Guidelines for qualifications of neurodiagnostic personnel: a joint position statement of the American Clinical Neurophysiology Society, the American Association of Neuromuscular & Electrodiagnostic Medicine, the American Society of Neurophysiological Monitoring, and ASET—The Neurodiagnostic Society. J Clin Neurophysiol. 2023;40(4):271–85. doi:10.1097/WNP.0000000000001004

‍

‍