Novice EEG interpreters are often and understandably overwhelmed by the sheer amount of data before them and may balk at the whole enterprise as “information overload.” An orderly routine of inspecting the EEG facilitates accurate and thorough review.
Calibration
Prior to beginning to interpret the EEG, one should first ensure that the study was appropriately calibrated. Before running an EEG recording, the technologist first passes a known voltage through the system to ensure that each amplifier for each channel handles this known signal identically; if there is a problem at this stage, there may be a problem with amplification that could affect interpretation of the EEG. Then, biocalibration is performed as an additional step to ensure the fidelity of the cerebral signal; this commonly involves having the patient open and close their eyes.
Orientation and Nomenclature
Electrode Placement
Familiarity with the nomenclature of EEG is important prior to commencing interpretation. Please see , which demonstrates the standard International 10-20 electrode site placement strategy. The nomenclature “10-20” represents standard intervals of measurement of either 10 or 20 percent for positioning electrodes over the anterior–posterior dimension between the nasion (point at the bridge of the nose) and inion (prominent bump on the back of the head representing lowest point of the skull), and between the auricular (ear) positions. Each electrode site is represented with a letter and a given number. The letter represents which lobe of the brain the electrode site overlies (i.e., F is frontal, T is temporal, P is parietal, and O is occipital). Odd-numbered electrode sites are on the left side of head, even-numbered electrodes are on the right side, “z”-labeled sites are in the midline from anterior to posterior, and “A” or auricular sites are on the mastoid processes/ears.
The International 10-20 electrode placements. Showing a longitudinal bipolar montage. Copyright 2013. Mayo Foundation for Medical Education and Research. All rights reserved. Courtesy of Dr. Jeffrey W. Britton, MD.
Montages
Once the calibration has been reviewed and found to accurately display both known voltages and patient EEG data, one should note the EEG montage. A montage is a standardized arrangement and selection of channel pairs and chains for display and review. There are many different montages used for various purposes, but they are divided into two types: bipolar and referential. One common montage is the longitudinal bipolar montage (aka the “double banana” because the electrode configuration appears like two bananas laid front to back over each of the brain hemispheres; see ). In a bipolar montage, neighboring electrodes are paired to one another, either anterior to posterior (longitudinal bipolar) or side to side (transverse bipolar), which is a good way to try to localize EEG potentials. Alternatively, referential montages link each exploring active electrode to a distant reference. Common referential choices include the vertex (Cz electrode), the mastoid process (either individual ears, as shown in , or a mathematical derivation of both sites), or a common average reference. Another commonly used montage in the evaluation of epilepsy is the Laplacian or source derivation montage, where each active recording electrode is compared with a mathematical weighted average of the surrounding electrodes.
Ipsilateral ear referential montage. EEG electrode placement using the International 10-20 electrode placement system. Copyright 2013. Mayo Foundation for Medical Education and Research. All rights reserved. Figure courtesy of Jeffrey W. Britton, MD. (more...)
Frequency Bandwidths
Another important convention of EEG is the notation for different waveform frequency bandwidths, or bands. Frequency of EEG waveforms is important because the predominant frequencies vary by which state (e.g., alert wakefulness, drowsiness, sleep) the patient is in and may confer developmental and pathologic significance. EEG waveform frequency bands are expressed as follows: 1 to 3 cycles per second (Hz) are delta, 4 to 7 Hz are theta, 8 to 12 Hz are alpha, and 13 Hz and higher are beta. Frequencies above 25 Hz are not commonly encountered in the scalp EEG but may be seen arising directly from the cortical surface during intracranial recordings; these frequencies are termed gamma and are divided into low gamma (25–70 Hz) and high gamma (>70 Hz). The term “ripples” (generally >100 Hz) are thought to reflect epileptiform discharges (see ).
Ripples from intracranial EEG recordings. (Figure courtesy of Greg Worrell, MD, and Ben Brinkmann, PhD, Mayo Systems Electrophysiology Laboratory, Mayo Clinic Rochester).
Clinical Approach
There are different opinions as to whether one should review the clinical history prior to EEG interpretation. Some experts prefer to know the patient history prior to interpreting the EEG, so that the likelihood of a potential abnormality can be interpreted within its appropriate clinical context. Others think that knowing the history biases the interpretation and may lead to “overcalling” or “undercalling” questionable findings. There is potential value in both approaches, and one solution is to first read through the tracing without the history of the patient, and then take a second pass after reviewing the history.
