When Does Anesthetic Fade Begin To Effect Your Motor Evoked Potentials?

Click here for the journal article: An evaluation of anesthetic fade in motor evoked potential monitoring in spinal deformity surgeries.


Though not well researched, in IONM it is known that transcranial motor evoked potentials (TcMEPs) deteriorate gradually in surgeries with general anesthesia. In other words, the threshold to get a TcMEP response will increase during long surgeries, such that it takes higher and higher stimulation to evoke a response. This is caused “anesthetic fade” and is independent of surgical maneuvers. This paper attempts to document the time course for anesthetic fade.


This is a retrospective study of 159 patients (31 male) who had spinal deformity surgery.

Patients were intubated with propofol, sevoflurane and rocuronium. Once intubation was complete the sevoflurane was turned off and no further neuromuscular blockade was given. Total intravenous anesthesia consisted of a target-controlled propofol infusions (TCI; 1.5-5 µg/ml) and remifentanil infusions (0.1-0.5 µg/kg/min).

Baseline TcMEPs were set once the train-of-four (TOF) reached >90%. The amplitude of the TcMEP response in the abductor digiti minimi (ADM) and abductor hallucis (AH) was measured at every hour up to 6 hours following onset of the propofol infusion/intubation. Response amplitudes from the left and right homologous muscles for a given time period were averaged. For statistical analysis all TcMEP measurements were compared to the baseline amplitude

MEP anesthetic fade Results:

The TcMEP responses from the foot faded by 10% four hours after initial propofol infusion. The TcMEP responses from the hand faded by 16% five hours after infusion was begun. The subsequent hour(s) had a cumulatively greater statistically significant decreases in amplitude. At 6 hours, on average, the hand responses had faded by 21% and the foot responses by 39%.

In this study there was an average 55 min interval between the beginning of the protocol infusion and the surgical incision. This mean that they began to first experience anesthetic fade 3 hours into surgery.


TcMEP anesthetic fade begins in the lower extremities 4 hours after the onset of propofol infusion and in the upper extremities 5 hours after onset of propofol infusion. It cumulatively worsens with time.



This paper defines anesthetic fade as that in which “stimulation threshold increases because the waveform amplitude decreases with the accumulation of propofol.” And yet one item they never discuss throughout the paper is the stimulation intensity. How did they determine it? Was it supramaximal? Was it ever adjusted? They note that they used a train of 5-7 pulses. Were the pulses changed within a surgery? Why not use stimulation increase as an additional measure of fade?

motor evoked stimulation

TcMEP Parameters Used In This Study

One assumes that the stimulation intensity variable is kept constant to allow for a comparison of amplitude response across propofol infusion time. So once the baseline was set, to document fade the intensity needed to remain unchanged. Or despite their stated definition did they functionally define fade independent of stimulation increase and instead defined it as a decrease in supramaximal responses regardless of stimulation intensity? This is not clear.

As a retrospective study it is easy to imagine that in individual cases the TcMEP stimulation may have been occasionally adjusted when fade set in. Judging from the error bars on their amplitude responses it seems likely there were patients that experienced enough fade to warrant stimulation increases to differentiate the fade from a surgical alert.

Questions about alarmAlarm Criteria

In surgeries that go longer than 4-5 hours it may be that a strict alarm criterion of % loss of amplitude or absolute increase in stimulation intensity are not the best alarm criteria. Rather a contextual alarm (as advocated by David MacDonald in which you take into the account many factors such as the timing, laterality and gradual vs sudden nature of the decrease) may be more appropriate.


For cases where the surgery is below T1, the upper extremity TcMEPs are monitored as a control. This study documents that the lower extremity TcMEPs are affected earlier and experience more fade than the upper extremities. This complicates the interpretation of TcMEPs. If our control responses are less affected by fade than they are a less than perfect control for distinguishing anesthetic fade.

This also leads into the natural question of why are the lower extremities more susceptible to anesthetic fade? The paper posits it may have to do with differences in corticospinal drive mechanisms. This is a vague comment and this point could use more research.

Sevoflurane + Rocuroniumanesthetia intubating

Just a small side note. They reported on average it took them 113 minutes after propofol infusion to establish TcMEP baselines. This time may have been related to the fact that 44 patients were given Sevoflurane in addition to rocuronium for intubation. Woloszcuzuk-Geicka et.al’s (2007) showed the potentiation of sevoflurane on rocuronium. Using sevoflurane with rocuronium results in a longer recovery for neuromuscular junction. As the current paper only set baselines once twitches recovered this may have contributed to a delay in getting TcMEP baselines.

Future Experiments

An obvious extension of this would to be to repeat this study comparing different halogenated gases in adults.  An ideal sample pool would be late-adolescent idiopathic scoliosis patients.

patient deformityIn this study they had a range of patients including idiopathic and neuromuscular scoliosis. There is no mention if they looked at fade across deformity types. It would be interesting to document if fade is different in the different patient groups. If fade is affected by different corticospinal drive mechanisms, then you would expect to experience it differently across different congenital and acquired conditions.

Another item that would have been interesting for them to examine would be, were there any asymmetries (i.e. left versus right) in the fade? Did they have any clinical correlations or trends?

One regret they do name in the paper is that they were unable to do a propofol dose to anesthetic fade comparison as target-controlled infusions were used. This should be a straightforward study to initially do on an animal model.

Thinking About Anesethetic FadeTake Away

For cases with a propofol infusion, if you account for setup time, you have a 2-3-hour window of stable TcMEP responses before you will begin to experience anesthetic fade. And expect the fade to occur first in the lower extremities.


Have you ever had any issues with anesthetic fade? What time points do you usually begin to see it? How do you differentiate fade from a loss of responses?

Leave a Comment