E. Shufman et al., ELECTRO-ENCEPHALOGRAPHY SPECTRAL-ANALYSIS OF HEROIN-ADDICTS COMPARED WITH ABSTAINERS AND NORMAL CONTROLS, Israel journal of psychiatry and related sciences, 33(3), 1996, pp. 196-206
Previous studies have shown that opiates slow the EEG and, in high dos
es, reduce the threshold of seizure activity. The present work looks a
t computerized EEG analyses (in the Fast Fourier Transform - FFT - met
hod) of heroin addicts, recent abstainers and normal controls, with th
e aim of comparison and delineation of group characteristics. Examinat
ions of 60 taped EEG recordings were performed: 20 subjects were curre
nt heroin users, 20 were recent abstainers and 20 were normal controls
. Statistical analysis was performed for the relative frequency of wav
e bands. To amplify the known findings of slowing in heroin users, spe
cific ratios were calculated: the alpha ratio (namely, the 8.0-9.5 Hz
to 9.5-12.0 Hz ratio) and the delta to low alpha ratio. The specificit
y and sensitivity of the FFT method were evaluated through the use of
discriminant analysis. The EEG was also recorded on conventional paper
and evaluated by a neurologist. Results: The addicts had a higher alp
ha ratio. The abstainers had a slowing of their alpha waves, a high in
cidence of delta waves and a delta to low alpha ratio that was relativ
ely high. As a function of time from the beginning of abstinence, this
ratio decreased. Abstainers for more than 80 days and controls had id
entical EEGs. The FFT method had 50% sensitivity for the addicts and 7
0% for the abstainers, while the specificity was 55%. In contrast, the
neurologist's reading of the EEG had a very low sensitivity (25% for
the addicts, 30% for the abstainers and 20% for the controls) and a 70
% specificity.