The Value and Limitation of Electroencephalography

ELECTROENCEPHALOGRAPHY, as an aid to diagnosis in clinical neurology, has beeni in operation for the past twelve years or more. Hans Berger's classical researches in the electrical potentials of the human brain recorded from the scalp were published between 1929-1937. This work received little attention, probably for the reason that it was published in psychiatric journals which are not as a rule read by physiologists. Since Adrian and Matthews in Cambridge studied the normal rhythms in 1934-35, a vast amount of physiological and clinical research has been done. In 1935 Gibbs, Davis and Lennox described the electrical changes associated with epilepsy. Another important landmark was Walter's description of a method for the localisation of brain tumours. Electroencephalography has been in use in the Royal Victoria Hospital since April, 1949, and more than a thousand records have been made on our machine, which is a four-channel machine manufactured by the Marconi Instruments Company. The number of channels or pairs of leads varies, with each type of machine from three to eight or more. The ideal number is probably about six, as any more than this number places too great a strain on the technician controlling the machine at the time of recording. The head is covered at equal spaces with a number of electrodes, usually four longitudinal antero-posterior lines and three transverse lines of electrodes. Good contact is made by cleansing the skin with spirit and covering the electrodes with saline pads. The electrodes are held in position by a rubber harness similar to that used in ladies' hairdressing establishments. It may be remarked at this junction that electroencephalography is an entirely painless procedure and there is no risk to life. Recording is made by pens writing directly onto moving paper. Each pen is fed by one channel, and the record resulting is a measure of the changing potential differences between two electrodes amplified approximately two million times. This is called bipolar recording. In unipolar recording a common electrode is used, usually attached to the ear. In our machine, then, four channels simultaneously record from usually five electrodes arranged serially in a straight line. The connections from the electrodes are so arranged that in a series of three electrodes connected to two channels, should a discharge occur, near to the central or common electrode, then a "phase reversal" in the records of the two pens will be apparent. When this phase …