Bispectral Analysis of the Electroencephalogram

  

 Margaret Bothwell, Tom LeBlanc, Chris Maier,

Bill Maurer, Peter Sebel, Tim Vandervoort

 

Background:

It is an unfortunate truism that, until recently, anesthesiologists have been unable to identify accurately how "deeply" a patient is anesthetized during surgery. Typical assessments of the adequacy of anesthesia include measurement of blood pressure, heart rate, and other autonomic responses such as lacrimation (tear production) and diaphoresis (sweating). Intuitively, one would argue that, since anesthesia affects the brain, there should be some way of measuring the effects of anesthetics on the brain.

Much like the heart, the brain produces low voltage electrical signals (approximately 100 m V) known as the electroencephalogram (EEG). EEG recordings were first made in the 1930s. Researchers first suggested in the 1950s that EEG recordings could be used to measure the effects of anesthetics on the brain.

There were two impediments to using the EEG to monitor anesthetic effects on the brain. Firstly, EEG recordings were typically made on paper and would generate many hundred yards of paper recordings during an anesthetic making the data impossible to interpret in a useful manner. Secondly, electrode technology was such that it took a trained technician approximately thirty minutes to apply the electrodes necessary for recording the EEG.

All complex waveforms, including the EEG, can be broken down into component sine waves by using a technique known as Fast Fourier Transform (FFT). Three items of information are obtained from each epoch (time period): frequency, power and phase. As computer technology improved, EEG monitors for use in anesthesia first became commercially available in the early 1980s. These produced a power spectrum of the EEG using only the frequency and power information from the FFT and were intended for monitoring the effects of anesthetics on the brain. Unfortunately, the monitors did not live up to their early promise and were generally abandoned.

 

History and evolution of the technology:

Bispectral analysis goes back to the 1950s when researchers were attempting to predict wave motion in the ocean. The technique of bispectral analysis involves using the phase information contained in the FFT and computing interfrequency phase relationships. Thus, it is a more complete description of the waveform yet it is computationally very intensive. Bispectral analysis was also used in some military applications, including submarine tracking (as popularized in the movie The Hunt for Red October).

In the mid 1980s, a young researcher at Harvard Medical School, Nassib Chamoun, conducted research into the use of signal analysis techniques of the electrocardiogram to identify myocardial infarction (heart attack). He became interested in the techniques of bispectral analysis and left his research to form the company now known as Aspect Medical Systems, which is devoted to developing the use of bispectral signal processing techniques in medicine.

While raising the initial venture capital for Aspect Medical Systems, it became apparent that the greatest market for these signal-processing techniques was Anesthesiology. Unlike previous EEG derivatives that have been introduced to anesthesiologists, bispectral analysis of the EEG has undergone a rigorous scientifically controlled program resulting in the development of the Bispectral Index (BIS). Complex signal processing algorithms are used to produce the BIS which is graded from 100 (awake) to 0 (electrical silence). The scale has been demonstrated in repeated, well-controlled, experiments to correlate well with the hypnotic end-points of anesthesia. For the first time, anesthesiologists have a "window into the brain."

 

The other technological breakthrough that accompanied BIS technology was the development of the ZIP-PREPÔ electrodes. Anesthesiologists are notoriously impatient individuals and do not have the technical resources to spend thirty minutes applying EEG electrodes to each patient. The ZIP-PREPÔ electrodes are self-prepping adhesive electrodes that are applied to the forehead which guarantee adequate impedance for recording the very low voltage EEG. These adhesive electrodes contain polymer tines, which break skin contact, allowing good quality electrical recordings to be made. The electrodes can be applied in less than 10 seconds by someone with no skill. Recently, the individual electrodes have been incorporated into a single strip called the sensor.

Majors players / users:

Patents for BIS technology and ZIP-PREPÔ electrodes are held by Aspect Medical Systems. They are a small, privately held, biotechnology company based near Boston. They are the only players in the field. The potential users are anesthesiologists in the operating room and physicians in intensive care units. Considering that twenty million anesthetics are given in the USA each year, the potential use of this technology in the USA alone is enormous. Early adoption has begun in major medical centers such as Massachusetts General Hospital in Boston and Grady Memorial Hospital in Atlanta.

Applications and Limitations:

The application of this technology enables anesthesiologists to monitor the hypnotic component of anesthetics, i.e. the effect of anesthetics on the brain. It allows them to dose hypnotic drugs more accurately, probably to reduce the risk of awareness during general anesthesia and to improve patient recovery.

The challenge in developing this application, and indeed its limitation, is the pervasive nature of cost constraint within the health care environment. Although anesthesiologists may see value in the technologies, in order to justify technology acquisition, hospital administrators need to be convinced that value is added to patient care. In order to overcome the constraints of capital budgeting and to convince hospital administrators the technology is useful, two techniques have been employed. Instead of the traditional method of selling equipment to the hospital on a capital basis, a fee-for-usage program has been introduced throughout the United States. This obviates the need for the hospital to develop a large capital budget to acquire the technology. Secondly, a large study has been carried out at Grady Memorial Hospital to identify the cost impact of this technology. Because of improved patient recovery and throughput as well as decreased drug utilization, implementation of this technology more than pays for itself. Thus, administrators can be convinced of the added value.

Future Development and Expectations:

Research in the applications of BIS is growing exponentially. In 1992, one paper on BIS was presented at the American Society of Anesthesiologistsí national meeting. This year there will be twenty-seven such papers. Sales of the monitoring technology in 1996 were of the order of tens of units, in 1997 hundreds of units will be sold. During 1998, adoption of the technology will be in the thousands of units. As this technology is fully adopted in anesthesia, it is likely to become the standard of care. Next year and beyond, Aspect will tailor BIS for Intensive Care Unit and conscious sedation applications and continue development efforts on cardiac technology.