Patent No. 6493576 Method and apparatus for measuring stimulus-evoked potentials of the brain

 

Patent No. 6493576

Method and apparatus for measuring stimulus-evoked potentials of the brain (Dankwart-Eder, Dec 10, 2002)

Abstract

For measurement and display of auditory evoked potentials of the brain, pickup electrodes (A, Ref), an amplifier circuit (1 . . . 5) and an evaluation facility (7) are used to display the time trend of midlatency auditory evoked potentials (MLAEP) and simultaneously therewith, during measurement of the information signals, the time trend of early brain stem auditory evoked potentials (BAEP), which is displayed at a higher time resolution as compared to the information curve. For this purpose an amplifier circuit is used the isolation stage (8) of which is arranged in a way so that the total gain of the amplifier unit (1, 3) connected between electrode assembly (A, Ref) and isolation stage (8) amounts to between approximately 1000 and 4000. A further improvement is achieved by connecting the inputs of the amplifier circuits, which are connected to one pick up electrode (A, Ref) each and to a reference potential (ground) via an input terminating resistor, to a current generator included in an impedance measuring circuit for measuring electrode impedance.

Notes:

DESCRIPTION

This invention pertains to a method and a test assembly for measurement and display of stimulus-evoked potentials of the brain, especially for monitoring analgesia and anesthetic depth.

During surgery it has to be assured that the patient will not wake up from general anesthesia and especially that pains during surgery or other surgical manipulations will neither be perceived during surgery nor remembered postsurgically by the patient. The fright caused by an experience like this may result in a so-called post-traumatic stress syndrome. For this reason, it is of great interest to measure and display suitable parameters which determine anesthetic depth and analgesia to enable anesthetists to control anesthesia more precisely than previously and to reduce patient strain to a minimum. In particular, undesired awakenings of the patient during anesthesia have to be realized as early as possible.

We know different methods to detect wake stages during general anesthesia. The most important (1) are the so-called PRST-score, calculated from changes of blood pressure, heart rate, sweating and tear production, and (2) the isolated forearm method, during which one of the patient's forearms is isolated against anesthesia by interrupting the blood flow by means of, for example, a hemomanometer cuff. To monitor conciousness the patient will then be examined whether he is able to adhere to simple commands during surgery. Furthermore, an EEG processing method is known, which evaluates EEG frequency and amplitude changes occurring during wake and anesthetic stages. However, the PRST score is not always a reliable method to detect intraoperative wake stages. The isolated forearm method can only be applied over a short period of time and is consequently not suited to indicate motor responses of the patient during long-lasting procedures. The processed EEG and its resulting parameters (median frequency and spectral cut-off frequency) are not optimally suited for this purpose, too.

Stimulus-evoked EEG signals, such as visually evoked potentials, somato-sensory evoked potentials and auditory evoked potentials, which will undergo dose-dependent suppression during general anesthesia, are better suited to fulfill this task. With midlatency auditory evoked potentials this dose-dependent effect becomes especially obvious. Auditory evoked potentials consist of a series of positive and negative voltages generated at different sites along the auditory pathway which can be picked up by electrodes at the skull. They reflect collection, transmission and processing of acoustic information from the cochlea via the brain stem to the cerebral cortex. Early auditory evoked potentials are generated by structures of the peripheral auditory pathway and the brain stem. They give evidence of stimulus transduction and primary stimulus transmission. It is known that early auditory evoked potentials remain almost stable during anesthesia in contrast to dose-dependent suppressed midlatency auditory evoked potentials.

In general, stimulus evoked potentials are well suited for monitoring anesthetic depth as well as for recording further neurophysiological functions. Change of the time course of the measured potentials in comparison to the unchanged potentials make it possible to draw conclusions for the neurophysiologic function to be observed. Previously, it has been difficult to apply this method practically. This was due to the fact that there were no satisfying possibilities for measuring and evaluating the potentials, especially when the measurement could not be performed under laboratory conditions but, for example, in the operating room. Under suboptimal conditions many difficulties may occur. Maybe the staff is not familiar with recording of evoked potentials or the anesthetist does not know how to interpret the recorded curves. It needs experienced experts to assess the results and to operate the equipment mentioned. Furthermore, the electrodes have to be applied simply and quickly. Other problems may occur by interference due to specific instruments in the operation room and by signals of other equipment induced into the electrodes. Besides, attending measures on the patient may give rise to movement artifacts.

Amplifier circuits, normally used for recording cerebral potentials, may be a further source of problems. They normally consist of an instrumentation amplifier, which transfers the incoming signal (typically 0.5 to 10 .mu.V) via a high-pass filter, a mid-amplifier, a low-pass filter and a post-amplifier to an A/D converter coupled with an evaluation unit. Patient and analysis unit must be isolated from each other by, for example, an isolation amplifier, type BF or CF, to prevent inadmissible currents from passing through the electrodes if external voltages are applied.

In known circuits, this isolation unit is located either after the first amplifier stage, i.e. after a gain between 10 and less than 100, or after total gain of up to 10.sup.6 or after the A/D converter. This results in various disadvantages. In the first case extremely small signals pass through lines which impose a capacitive load on the driving amplifier and consequently may give rise to signal distortion. Furthermore, interference decoupling is limited, even with low driver impedance. However, if isolation is intended to be introduced after total gain, decoupling of interference will be good; indeed, the required number of components does not allow for the desired miniaturization of the circuit. Consequently, the distances to the recording sites are longer than desired for optimal recording. Besides, coupling of the output signals into the input circuit can not definitely be avoided in small sized amplifiers with high gains (from about 5000 onwards). Gains, as high as those required for measurement of evoked potentials, may cause a feedback from the amplifier output to its input, either via direct capacitive coupling or via the supply voltage. This feedback is especially disturbing if the terminating impedance of the input amplifier is high or asymmetrical. A feedback loop will change the frequency vs. time behavior of the amplifier system so that instabilities may occur which may give rise to amplifier oscillations. This is aggravated by the fact that these effects will often only give rise to minimal changes of the entire amplified signal, but will have considerable effect on the portion to be measured, which is "hidden" in the entire signal. For example, the brainstem auditory evoked potentials (BAEP) having a signal amplitude of 1 .mu.Vss cannot be recognized in the spontaneous EEG, which has an amplitude of 20 .mu.Vss to 50 .mu.Vss. The BAEP can best be measured by means of averaging, provided that the spontaneous portion of the EEG averaged over time does not correlate with BAEP and that the BAEP portions occur time-synchronous after the stimulation signal. This cannot be assured, if the time behavior of the amplifier is subject to change or non-linearities due to intermodulation distortion occur. The recording signals are distorted or disappear during averaging. Another side-effect is that common mode rejection deteriorates due to positive feedback on the input. Furthermore, the system can encounter an unfavorable operation point for a short period due to common mode interference such as mains hum, monitor interference and transients. Often these instabilities are hardly visible in a single signal sweep but show their effects after averaging. The signal to noise ratio can deteriorate considerably so that the measured signals can not be analyzed any more.

The invention is intended to avoid the problems mentioned and to provide a method, and a test assembly, which allow easy, reliable and trouble-free measurement of neurophysiological signals, so that correct measurements can be performed which can be reliably assessed and which do not require experienced experts for operating the instrument being part of the test assembly.

This task is fulfilled by the features states with the patent claims.

According to a further aspect of the invention, the time trend of the evoked potentials from the brain stem (BAEP) will be displayed with a higher resolution as compared to the information signals during measurement and display of the potentials related to the function measured (MLAEP).

It is of special importance, that the reference curve or other curve (i.e. availability of reference or other curve data) are displayed during the measurement or simultaneously with the information signals. Just as the information curve, the display is automatically controlled by the computer integrated in the evaluation unit.

The invention allows for easy and reliable evaluation of the recorded signals, even under sub-optimal conditions, e.g. during surgery. This is especially useful for monitoring anesthetic depth. As the curve of the brainstem evoked potentials does not change during anesthesia and can be evaluated reliably due to a higher time resolution, it can be taken as a standard for the quality of the recorded information signals. Deterioration of the reference curve can be assumed to be due to bad recording conditions.

The evaluation of the signals will be most favorable and informative due to the fact that according to this invention the evoked stimuli will be repeated periodically or stochastically and that the actual information curve is displayed after a predefined time or number of recording periods but has been averaged over a considerably greater period of time or a considerably higher number of previous recording periods. In this context it is useful that for averaging over a predefined number of recording periods the average values of the signals recorded over a considerably lower number of recording periods will be compared automatically with each other and will be averaged themselves whereby averages that differ greatly from the majority of the other averages will be neglected. By correlating sub-groups, artifacts can be detected more easily and can be eliminated.

It may be of special advantage, if the time trend of the variation of a characteristic peak of the midlatency auditory evoked potentials (MLAEP) and/or the latency of this peak and/or the automatically calculated quotient of these two factors is displayed.

Furthermore it is useful that the time trend of a variable instrument factor is calculated by the evaluation unit and displayed simultaneously with the time trend of the recorded values. This is an additional feature which assures good quality of the measurement.

For evaluating the evoked potentials the typical amplification should be between 10.sup.5 and 10.sup.6. According to a further aspect of the invention, the total gain of the amplifier part between the electrodes and the isolating amplifier is considerably higher than 100 but less than 10.sup.4. The post-amplifier between the isolation amplifier and the evaluation unit takes over the remaining amplification. If, for example, the gain prior to the isolation amplifier is between 1000 and 4000, a post-amplification of more than 25, but typically between 100 and 1000, is required. This allows for the best possible compromise between the highest achievable common mode rejection (gain as high as possible before isolation stage) and a minimal coupling into the input (gain as low as possible before isolation stage). With gains of about 1000, the common mode rejection ratio can amount to 150 dB.

At the same time, the input and output circuits will be decoupled after 1000 fold gain due to the isolation stage, so that instabilities of the amplifier will no longer be expected. The output signal with the highest amplitude will then be decoupled from the supply voltage of the input circuit. At the same time, the output signal will be treated as a common mode signal by the input circuit. A further improvement can be achieved if the ground of the output circuit is connected to earth. This will considerably simplify operation of the test assembly.

The direct test assembly furthermore provides the advantage that the required amplifier circuit can be optimally miniaturized without having to cope with disadvantages such as long leads to the recording sites, signal distortion, insufficient noise decoupling and noise signals, etc.

Reliability can also be improved with further steps of development by connecting an impedance circuit for recording electrode impedance to the inputs of the amplifier circuits which, after having exceeded a limit value and/or after having reached a certain difference between e.g. the active and the reference electrode, will generate a warning signal. Wrong electrode impedances which are mainly due to the contact between electrodes and skull can cause recording errors in connection with the invention-type amplifier of the invention.

As already mentioned, the invention is not only suitable for determining anesthetic depth but also for recording any visually evoked potentials. The following description of the invention is mainly based on measurements of auditory evoked potentials for monitoring anesthetic depth.