Neurology -- Correspondence for Jääskeläinen et al., 61 (8) 1073-1078

Correspondence published:

Sevoflurane is epileptogenic in healthy subjects at surgical levels of anesthesia: Ville HK Jäntti, Eila Sonkajärvi (5 December 2003)

Reply to Jantti: Satu K. Jääskeläinen, Kaike Kaisti, and Harry Scheinin. (5 December 2003)

Sevoflurane is epileptogenic in healthy subjects at surgical levels of anesthesia 5 December 2003

Ville HK Jäntti,
Tampere University Hospital
Dept. of Clinical Neurophysiology, Tampere University Hospital, P.O. Box 2000, FIN-33521 Finland,
Eila Sonkajärvi
Send Correspondence to journal:
Re: Sevoflurane is epileptogenic in healthy subjects at surgical levels of anesthesia

Ville.Jantti{at}pshp.fi Ville HK Jäntti, et al.

In their paper, Jääskeläinen et al [1] claim that rapid increase in sevoflurane concentration itself does not cause epileptiform activity but just reflects the high concentration. After rapid increase causing epileptiform patterns, the EEG pattern typically normalizes to resemble EEG in isoflurane anesthesia.
We have studied over 100 patients and have not recorded clinical seizures, although they probably sometimes occur. Three types of epileptogenicity of sevoflurane in a healthy brain can be distinguished: 1) rapid rise with possible hyperventilation causes symmetric epileptiform EEG patterns; 2) prolonged high concentration causes similar patterns and focal or generalized seizures; and 3)during recovery from anesthesia occasional seizures not properly documented with EEG. In addition, sevoflurane enhances epileptic activity in patients with epilepsy.
The authors claim to analyze delta activity 0.1-3.1 Hz with FFT from EEG samples of 2 and 1 seconds. Spectral estimates from 1- and 2-second samples should not be statistically compared. Therefore the results on SEF and PF--including Fig 2--are questionable. It is more important to show raw EEG traces and the authors should have presented the EEG of the two propofol recordings they claim to be epileptiform. There is no reason to conclude this without proper illustration. Finally, during suppression, EEG is of relatively low amplitude but not inactive.
References
1). Jääskeläinen SK, Kaisti K, Suni L, Hinkka S, Scheinin H. Sevoflurane is epileptogenic in healthy subjects at surgical levels of anesthesia Neurology 2003; 61: 1073-1078.
2.) Jäntti V, Yli-Hankala A. Neurophysiology of anaesthesia. Suppl Clin Neurophysiol. 2000;53:84-88.

Reply to Jantti 5 December 2003

Satu K. Jääskeläinen,
Department of Clinical Neurophysiology
PoBox 52 FI-20521 Turku, Finland,
Kaike Kaisti, and Harry Scheinin.
Send Correspondence to journal:
Re: Reply to Jantti

satu.jaaskelainen{at}tyks.fi Satu K. Jääskeläinen, et al.

Our findings [1] indicate that the change of the drug concentration per se was not responsible for the occurrence and aggravation of epileptiform activity during deepening sevoflurane anesthesia. For example, the epileptiform seizure discharges in three out of eight subjects started from 10 to 25 min after the sevoflurane concentration had been stabilized to 2.0 MAC level which was carefully monitored.
During the induction phase of anesthesia, especially when mechanical hyperventilation is used, very deep anesthesia is not uncommon. 2.0 MAC anesthesia may be temporarily reached and thus, studies on the EEG events during the induction phase do not only evaluate the effects of the rapidly changing drug concentration but also of the deep level of anesthesia. The effects of these two factors cannot be separated from each other in that kind of a study design, especially if the exact depth of anesthesia is unknown.
We are sorry if the Methods section was not clear regarding the way the qEEG analyses were done. For spectral EEG signal analysis with Fast Fourier Transformation (FFT), ten 2-s epochs of artifact-free EEG were manually selected from awake recordings and at each level of anesthesia coinciding with the rCBF PET scans. The frequency content of the bursts (that were short during propofol anesthesia, ranging between 1 s and 2 s) during the burst suppression activity was analyzed from shorter epochs by combining ten 1-s epochs that started at the beginning of a burst and did not contain segments of suppression. The commercial analysis software (Persyst Insight and Prism software package) performed the Fast Fourier Transformation on these 20 s and 10 s periods of EEG signal. Regarding the very slow delta frequencies, the results of the qEEG analyses should, indeed, be viewed with caution. However, the main reason for performing this qEEG was to quantitate and illustrate, on the group level, the increase in the high frequency EEG activity (within high alfa and beta bands) at 2.0 MAC sevoflurane anesthesia compared to 1.5 MAC sevoflurane and popofol, which was very conspicuous when visually inspecting the raw EEG.
The very scarce epileptiform potentials observed in our study, a few scattered spikes in two subjects at 1.5EC50 propofol anesthesia are inconclusive regarding the possible epileptogenic properties of propofol. Thus, as stated in our Discussion, we do not claim that propofol is epileptogenic.
Finally, we agree that during suppression, the EEG signal is not totally inactive, as can be shown with DC recordings. Probably the term ‘relative inactivity’ would have been more appropriate but unfortunately, we have used the term ‘inactive’ (in a more pragmatic sense).
References
1). Jääskeläinen SK, Kaisti K, Suni L, Hinkka S, Scheinin H. Sevoflurane is epileptogenic in healthy subjects at surgical levels of anesthesia. Neurology 2003; 61: 1073-1078