Psychiatry Research: ~euro~maging 35:95-105 Elsevier
95
Quantitative EEG Correlates of Crack Cocaine Dependence Kenneth I?. Alper, Robert J. Chabot, Anthony E. Roy John Received
H. Kim, Leslie S. Prichep, and
May 18, 1990; revised version received August 17, 1990; accepted August 19, 1990.
Abstract. Evidence dependence
for a distinctive syndrome of neuroadaptation has accumulated from behavioral, neurophysiological,
in cocaine and preclin-
ical and clinical pharmacological studies. The authors report on the results of a preliminary investigation of the quantitative electroencephalographic (QEEG) correlates of severe DSM-III-R crack cocaine dependence in seven patients abstinent from cocaine for 1 to 68 days. The major QEEG finding was increased absolute and relative alpha power. Increased alpha power has also been reported in multiple previous studies of depressed patients. This series of crack-dependent patients showed significant depressive morbidity; four patients attempted suicide subsequent to initiating their use of crack and the group mean (31 SD) Beck Depression Scale score was 18.9 (AI 6.5). These results complement other studies that support the concept of neuroadaptation to chronic cocaine exposure. Prospective studies correlating QEEG measures with subsequent response to pharmacological interventions for cocaine dependence should be considered. Key Words. Cocaine, crack, electroencephalogram.
Although the acute effects of cocaine on the human electroencephalogram (EEG) were described by Hans Berger over 50 years ago (Berger, 1937), no study has yet appeared describing the EEG in chronic cocaine dependence. This may be due in part to the belated recognition of a syndrome of tolerance and withdrawal associated with cncaine &-j not_ ~&IJ& cQ&nc denendc?ncc? __d-____ me* -__, in ___ 1980 ____ j)&M-_r_ll __~ _-____-__ RI -I -a ctina-_-. nostic entity (American Psychiatric Association, 1980). The current epidemic of crack cocaine dependence (Gawin and Ellinwood, 1988) lends special relevance to the investigation of this disorder. Phenomenological evidence indicates the existence of a characteristic progression of abstinence symptoms in withdrawing addicts (Gawin and Kleber, 1986). Decreased regional cerebral blood flow (KBF), measured by positron emission tomography (PET), has been noted in abstinent cocaine abusers (Volkow et al., 1988). Pharmacological interventions selected for their interaction with neurotransmitter systems subserving reward-mediated behavior such as flupen-
Ya.nnmeh ..“..L.“l‘.
I? ‘q,W., Al...ar I..
M n I.X.Y.)
in I?ee,.nroh 111 I.I”“...“L.
Arr;rtnnt ‘1.,“1I+...a.
Pmf,wcrw. 1 ‘“lrlll”I,
Rr\hrti I\““w.+
1 4.
f-hcahnt “l.UY”L,
PhIS,_., l-J 1
is &g&t~
Professor; and Anthony H. Kim, M.D., is Fellow in Alcohol and Substance Abuse, Department of Psychiatry, New York University Medical Center. Leslie S. Prichep, Ph.D., is Associate Professor of Psychiatry and Assistant Director of Brain Research Laboratories, and E. Roy John, Ph.D., is Professor of Psychiatry and Director of Brain Research Laboratories, Nathan S. Kline Research Institute, Orangeburg, NY, and Department of Psychiatry, New York University Medical Center. (Reprint requests to Dr. K.R. Alper, Brain Research Laboratories, 8th floor, Old Bellevue Administration Building, First Avenue at 27th St., New York, NY 10016, USA.) 0~65-~781~~~~3.50
@ 1990 Elsevier Scientific Publishers Ireland Ltd.
thixol (Gawin et al., 1989a), desipramine (Gawin et al., 19896). and bromocriptmt (Giannini et al., 1989) have shown evidence of efficacy in treating the anhedonia and craving of the withdrawal state. Collectively. the available behavioral, neurophysiological, and pharmacological evidence appears to support the concept of a syndrome of neuroadaptation in cocaine dependence (World Health Organization. I98 I : Dackis and Gold, 1985; Gawin and Kleber, 1986). The quantitative EEG (QEEG) correlates of cocaine withdrawal are of interest for several reasons. As a reflection of central nervous system function, QEEG features may provide an approach to the investigation of the neurophysiological basis of the cocaine withdrawal syndrome. The substantial prevalence of cocaine abuse in psychiatric populations (Gawin and Kleber, 1986; Susser et al., 1989) has made an understanding of the QEEG correlates of cocaine dependence necessary to resolve the possible effects of cocaine exposure from the abnormalities seen in other psychiatric disorders (John et al.. 1988). The discovery of characteristic QEEG profiles associated with certain classes of psychopharmacological agents has been applied successfully toward the development of new psychiatric drugs (Itil, 1983). It is possible that an understanding of the QEEG correlates of cocaine withdrawal might provide a contribution to the screening and development of compounds directed at the anhedonia and craving of the withdrawal state. In the present study, we report on the EEG in seven individuals with severe crack cocaine dependence. Due to concern over the possibly confounding effects of psychiatric and other comorbidity related to substance abuse, subjects were selected who lacked a psychiatric history before their crack use, had never been dependent on substances other than cocaine, and were not intoxicated with crack or other substances at the time of the examination. While these subjects may not represent a typical cross-section of the crack-abusing population as a whole, they could safely be considered to be “hard core” and more or less exclusive users of crack cocaine. Methods Subjects. Seven patients were included in this study. Four patients came from an inpatient unit at Bellevue dedicated to the treatment of substance dependence. Three subjects were outpatients who presented themselves for treatment of crack cocaine dependence to the Bellevue Emergency Psychiatric Service. The four inpatients were off medication for at least 7 days before the EEG session. Of these four inpatients, two were tested 7 days, one 20 days, and one 68 days after admission to the hospital. The three outpatients were off cocaine for at least 24 hours before the EEG session; two reported having abstained from cocaine use for 1 day and another for 3 days before the EEG testing. The procedures for verification of patients’ reports about their length of time off crack and their use of other substances are described below. Table 1 lists variables describing demographic and clinical characteristics and cocaine history of the patient group. Criteria for inclusion in the study were as follows: (1) Fulfilling the criteria for DSM-HZ-R (American Psychiatric Association, 1987) dependence on cocaine in the form of crack for at least 1 year. Those with a history of ever having met the criteria for DSM-III-R dependence on any other substance were excluded. All subjects reported that they strongly preferred crack to any other substance of abuse. Subjects who reported alcohol intake confined their drinking to periods of crack use and used alcohol to modify unpleasant subjective activating effects of
Table 1. Subject characteristics (n = 7) AgeWI sex Race Years since first intranasal cocaine use Years since first crack cocaine use Estimated weekty crack cocaine us8 during 6-~nth preceding el~troenceph~~ram (g/w~k) Days since most recent crack use Hamilton Rating Scale for Depression score
24.6 f 4.5 4 males 3 females 4 black, 2 Hispanic, 1 white 4.4 rir:1.6 2.0 f 1.6 period
Beck Depression Inventory score Number of subjecta reporting suicide attempts before beginning crack use Number of subjects reporting suicide attempts after beginning crack use Number of subjects se&supporting at the time of beginning crack use Number of subjects self-supporting at the time of intake into the study
8.5 f 8.3 15.3 (range l-68) 13.1 xt 6.5
18.9 f 6.5 017 417 417 l/7
Note. Dataare presentedas meanf SD.
crack. No subject actively sought or purchased cannabis in the 60 days preceding the study, although three did use cannabis occasionally when it was available. (2) Negative urine screening for amphetamine, barbiturates, methamphetamine, opiates, methadone, tricyclic antidepressants (amitriptyline is a commonly abused drug in the community from which subjects were selected), benzodiazopines, phencyclidine, and propoxyphene at the time of EEG. (3) No use of cocaine or other drugs of abuse for at least 24 hours preceding the study. Outpatients were asked to abstain from crack use for at least 24 hours before the study but were told that the EEG would be done, and that they would receive payment ($20) in any event if they attended the intake appointment. Evidence from previous studies of cocaine users suggests that subjects can be reliable about their substance use if no adverse consequences follow disclosure of illicit drug use (Gawin et al., 1989~). For outpatients, it was required that they reside with an individual (in all cases this turned out to be a parent) who did not abuse drugs and who could verify that the patient had remained at home and apparently not used crack for at least 24 hours preceding the study. In practice, the pattern of binges associated with crack use was obvious to family members. No subject had received treatment with a psychotropic agent for at least 60 days before the study. Inpatients were evaluated at least 7 days after admission, and had not been treated with any psychotropic agent while in the hospital. Urine testing was used to confirm the absence of illicit drug use in the hospital. (4) No history of head trauma or a neurological or medical condition known to affect the EEG and no history of ever having used any drug by the intravenous route. A history of seizures limited to periods of cocaine intoxication was also grounds for exclusion. (5) No history of psychiatric hospit~ization or having been treated with psychotropic medication before using crack. EEG Data Acquisition. Eyes-closed, resting EEG data were collected from the 19 monopolar electrodes of the International lo/20 System, referred to linked earlobes. A differential eye channel was used for the detection of eye movement. The EEG amplifiers had a bandpass from 0.5 to 30 Hz (3 dB points), with a 60-Hz notch filter. The data were digitized at 100 Hz with 1Zbit resolution. Twenty minutes of continuous EEG data were collected for subsequent editing and analysis.
3x
EEG Data Analysis. All data acqutsttion and analyses were performed using the methodology of neurometrics on a Cadwell Spectrum 32 (software revision ~3.1). Quantitative features are extracted, log-transformed to obtain normal (Gaussian) distributions, ageregressed, and z-transformed relative to population norms. The reader is referred elsewhere for a general overview of the methodology (John et al., 1983, 1988) and for articles addressing specific topics such as the validity of the use of log transformation of QEEG data to obtain Gaussianity (Gasser et al.. 1982; Gken and Chiappa, 1988) development of the normative data base (John et al.. 1987). construction of age regression equations (John et al.. 1980, 1988). studies of test-retest reliability (Kaye et al., 1981; John et al., 1983; Fein et al., 1984), and independent replications of the QEEG norms (Matouse’k and Petersen, 1973; Gasser et al., 1983; Jonkman et al.. 1985: Yingling et al., 1986; Alvarez et al., 1987; Harmony et al., 1987: John et al., 1989). The neurometric adult EEG norms have been derived from a group of 150 normal subjects, aged 17 to 90 (John et al., 1988). On the basis of the distribution statistics obtained from these normative data, it is possible to calculate z-values or standard scores for these features (proportional to the probability of abnormality) in the evaluation of index subjects. The data were edited using computerized artifact-detection algorithms combined with visual inspection to obtain 24 epochs (2.5 set each, a total of 1 min) of artifact-free data from 20 min of continuous EEG. Power spectral analysis was performed using the Fast Fourier Transform (FFT). For each of the 19 monopolar derivations, absolute and relative (%) power and mean frequency were computed for the delta ( 1.5-3.5 Hz), theta (3.5-7.5 Hz), alpha (7.5-12.5 Hz), and beta (12.5-25 Hz) frequency bands. Overall absolute power and mean frequency for the entire spectrum from I .5 to 25 H7 were also computed
Results Clinical Data. Four of seven subjects attempted suicide after they began to use crack. The mean 21-item Beck Depression Inventory score of 18.9 also suggests significant depressive morbidity (Beck, 1967). A trend for correlation between Hamilton (24 item) and Beck scores, commonly seen in studies of depressed patients (Hamilton, 1987) was not evident in this study (r = 0.17, p > 0.10).Of the four patients who had been economically self-supporting (defined as residing in a household for a period of 1 year in which the patient’s legitimately earned income accounted for at least one-half of the rent and household expenses) immediately before the initiation of their use of crack, only one remained legitimately selfsupporting at the time of evaluation. The substance abuse histories of the individual subjects revealed that while none of them were naive to other drugs of abuse such as ethanol and cannabis, they greatly preferred, sought, and used crack relative to other substances. All subjects used intranasal cocaine before beginning to use crack and reported an escalation of cocaine intake after using crack, a phenomenon that has been reported elsewhere (Gawin and Ellinwood, 1988; Verebey and Gold, 1988). EEG Data. Fig. 1 presents topographic maps of group mean z scores for absolute and relative power and mean frequency for each of the four frequency bands. It should be recalled that the color-coded values displayed designate the probability of a given value being within the confidence limits of EEG measurements obtained from normal individuals. The values are age-regressed and take into account the topographic variations of normative EEG data. The major trend is an absolute and relative increase in alpha power. Values for
99 Fig. 1, Topographic head maps for group mean z scores for absolute power (top row), relative power (middle row), and mean frequency (bottom row) DELTA
Abso
THETA
ALPHA
BETA
TOTAL
1.
Power
Color coding denotes the value of the group mean z score, which is proportional lo the probability that the data obtained from the 7 index subjects differ from the reference population of 150 normal adults. In estimates of the significance of a group mean z score, it is necessary to take the number of index subjects in the group into account. In the above figure, for example, the estimate of the critical value denoting significance at the level of p = 0.05 is f 1.96/\j7 = f 0.74 and is designated by the horizontal arrows to the right of the vertical color scale.
excess alpha are significant at the 0.05 level at 17 of 19 leads for absolute power and 18 of 19 leads for relative power. Inspection of the data of individual subjects reveals that the relative alpha excess is present in each subject and is apparently consistent in each member of the group. Fig. 2 presents examples of high-amplitude alpha seen in study subjects, which are compared with a representative epoch obtained from a normal control. Absolute delta power is significantly decreased in five of the six frontal leads. A relative but not absolute “hyperfrontal” distribution of alpha power is due to decreased delta rather than significant topographic variability of increased alpha. There appears to be a weak trend for increased absolute beta power, which reaches significance in 4 of 19 leads. Relative power in all frequencies except alpha is decreased as a result of the widespread alpha excess. There is a trend for increased mean EEG frequency greater in the anterior leads for delta and theta but not alpha and beta. Discussion The apparent effect of crack on the lives of this small group of subjects is impressive. Suicide attempts in four of seven subjects after they began to use crack and the mean Beck score of 18.9 suggest significant depressive morbidity. A trend for a correlation between Beck and Hamilton scores, which‘has been reported in previous studies of depression (Hamilton, 1987), was not present in this study. This may suggest a phe-
%i vemca, SCale of lne eiecrroencephalograniEEGi ,s 50 sir i‘m. The stib!r?c; on the f&R is a normal maie. aged 24 years z-i!e subject x the .middie li d aged 28 years abstrnent from cocaine for 7 days The sublect on the nghf !s a crack-dependent male, aged 23. abstinent fron n cocaine for 20 days
3-r .&veCde;’
.-i;:e
Fig. 2. Representative 2.5set EEG epochs obtained from a normal subject (left) and 2 crack-dependent subjects #I554 NON-USER (middle & right) CRACK USER PT. fi8443
101
nomenological distinction between the anhedonia and dysphoria of cocaine withdrawal and the full constellation of clinical symptomatology seen in affective illnesses. The clinical data suggest a dramatic decline in the functioning of individual crack users. The study subjects had all initially presented themselves to the emergency psychiatric service and may possibly reflect the more severe end of the full clinical spectrum of crack abuse. Nonetheless, three of four subjects who had supported themselves by legitimately earned income before they began to use crack were no longer self-supporting after becoming dependent on crack. It is unlikely that the effects of acute cocaine intoxication influenced the results reported here. The half-life of cocaine is on the order of 40 min (Fischman, 1987) and subjects were off cocaine for at least 24 hours before the EEG, with four subjects abstinent at least 7 days. The alpha increase was present in individual patients who had been abstinent for 7 or more days (Fig. 2). Furthermore, the acute effect of cocaine appears to be an increase in beta (Berger, 1937; Herning et al., 1985), and not alpha as noted in this study of crack cocaine withdrawal. The finding of increased alpha noted in this study of cocaine abusers is remarkably similar to the results reported by Struve and his colleagues for chronic cannabis abusers (Struve et al., 1989; Straumanis et al., 1990). Several possibilities exist. The results reported here may reflect a history of cannabis use rather than crack exposure. The very low use of cannabis relative to crack in the study population argues against this explanation. Another possibility is that cannabis- and cocainedependent subjects actually do have similarities in their QEEG profiles. Such a finding would bear an intriguing relevance to arguments for hypothetical “common final pathways” in substance use disorders (Gardner et al., 1988; Waldrop, 1989). However, consideration of QEEG evidence in the context of theories of common end effects in substance use disorders must take into account the QEEG correlates of alcohol dependence, which consist of low overall absolute power and increased relative power in beta (Funkhouser et al,, 1953; Naitoh, 1973; Jones and Holmes, 1976; Gabrielli et al., 1982; John et al., 1988) and differ from the findings for crack cocaine dependence and for cannabis dependence. The classic concept of alpha varying inversely with cerebral activation (Gevins et al., 1979) would appear to put the alpha increase noted in this study in accord with the observation of decreased rCBF in cocaine abuse (Volkow et al., 1988). Increased alpha might represent a correlate of a hypothetical depression in cerebral metabolism. This interpretation, however, may not explain the mean EEG frequency findings reported here. In studies involving the measurement of QEEG and rCBF in normal individuals, a positive correlation of overall mean frequency with rCBF has been observed (Obrist et al., 1963; Ingvar et al., 1976; Sulg et al., 1981). On the basis of these studies, the expected result in a state of cerebral metabolic depression would be a reduction in overall mean EEG frequency. Overall mean EEG frequency was not decreased in this study and actually tended to be increased in the anterior leads. Interpretation of QEEG data in terms of a hypothetic~ alteration of metabolism in cocaine withdrawal will require more information on metabolic correlates of the EEG. This issue may be addressed productively by obtaining simultaneous PET and QEEG in normal individuals and in subjects with cocaine dependence.
102
The finding of increased alpha is of interest in view of the results of QEEG investigations of two psychiatric conditions thought to predispose to cocaine abuse: depression and attention deficit hyperactivity disorder (ADHD) (Gittelman et al., 1985; Khantzian, 1985; Gawin and Kleber, 1986; Nunes et al., 1988). Increased alpha appears to be a consistent finding in studies of depression @chaffer et al., 1983; von Knorring et al., 1983; Brenner et al., 1986; John et al., 1988; Pollock and Schneider. 1989, 1990). Considerable evidence has accumulated in the pharmaco-EEG literature for a characteristic antidepressant QEEG profile (Itil, 1983). Diminution of alpha is a prominent feature of this putative antidepressant QEEG profile. A clinically effective antidepressant, mianserin, was selected for development on the basis of the similarity of its QEEG effects to those of antidepressants in preclinical studies (It& 1983). Tricyclic antidepressant medication has been used with success in cocaine dependence (Gawin et al., 19896) and also in ADHD (Wins~rg et al,, 1972; Rapoport et al., 1974; Garfinkelet al., 1983; Donnelly et al., 1986; Biederman et al., 1989). In a recent trial of buproprion for ADHD (Simeon et al., 1984, 1986), increased alpha power at baseline correlated with subsequent response to buproprion. It may prove rewarding to investigate the relationship of the apparent alpha increase in crack cocaine dependence to subsequent response to pharmacological interventions, particularly those with antidepressant effects. References Alvarez, A.; Pascual, R.; and Valdez, P. U.S. EEG developmental equations confirmed for Cuban schoolchildren. Electroencephalography and Clinical Neurophysiology. 67:330-332, 1987. American Pspychiatric Association. DSM-III: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC: APA, 1980. American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of dental Disorders. 3rd ed., revised. Washington, DC: APA, 1987. Beck, A. Measurement of depression: The Depression Inventory. In: Depression: Causes and Treatment. Philadelphia: University of Pennsylvania, 1967. Chapter 12, pp. 186-207. Berger, H. Electroencephalogram of man. Archiv ftir Psychiatric tind Nervenkrankheiten, 106:577-584, 1937. Biederman, J.; Baldessarini, R.; Wright, V.; Knee, D.; and Harmatz, J. A double-blind placebo controlled study of desipramine in the treatment of ADD: I. Efficacy. Journal of the American Academy of Child and Adolescent Psychiatry, 28(5):777-784, 1989. Brenner, R.; Ulrich, R.; Spiker, D.; Sclabassi, R.; Reynolds, C.; Marin, R.; and Boller, F. Computerized EEG spectral analysis in elderly normal, demented and depressed subjects. Electroencephalograp~y and Clinical Neurophysiology, 64:433-492,1986. Dackis, C., and Gold, M. New concepts in cocaine addiction: The dopamine depletion hypothesis. Neuroscience and Biobehavioral Reviews, 9:469477, 1985. Donnelly, M.; Zametkin, A.; Rapoport, .I.; Ismond, D.; Weingartner, II.; Lane, E.; Oliver, J.; Linnoila, M.; and Potter, W. Treatment of childhood hyperactivity with desipramine: Plasma drug concentration, cardiovascular effects, plasma and urinary catecholamine levels, and clinical response. Clinical Pharmacology and Therapeutics, 39:72-8 1, 1986. Fein, G.; Galin, D.; Yingling, C.; Johnstone, J.; and Nelson, M. EEG spectra in 9-13-yearold boys are stable over I-3 years. ~lect~oencephalo~aphy and Clinical Neurophysiology, 58517-518, 1989.
103 Fischman, M. Cocaine and ~phetamines. In: Meltzer, H.Y., ed. Psychopharmacology: The Third Generation of Progress. New York: Raven Press, 1987. Chapter 166, pp. 1543-1553. Funkhouser, J.; Nagler, B.; and Walke, N. The electro-encephalogram of chronic alcoholism. Southern Medical Journal, 46(5):423-428, 1953. Gabrielli, W.; Mednick, S.: Volavka, J.; Pollock, V.; Schulsinger, F.; and It& T.M. Electroencephalograms in children of alcoholic fathers. Psychophysiology, 19:404-407,1982. Gardner, E.; Paredes, W.; Smith, D.; Donner, A.; Milling, C.; Cohen, D.; and Morrison, D. Facilitation of brain stimulation reward by 6Qetrahydrocannabinol. Psychopharmacology, 96:142-144,1988. Garfinkel, B.; Wender, P.H.; Sloman, L.; and G’Neill, I. Tricyclic ~tidepr~s~t and methylphe~date treatment of attention deficit disorder in children. Journal of the American Academy of Child and Adolescent Psychiatry, 22(4):343-348, 1983. Gasser, T.; Bather, P.; and Moths, J. Transformation towards the normal distribution of broadband spectral parameters of the EEG. Electroencephalography and Clinical Neurophysiology, 53: 119-124, 1982. Gasser, T.; Moths, J.; Lenard, H.; Bather, P.; and Verleger, R. The EEG of mildly retarded children: Developmental, classificatory and topographic aspects. Electroencephalography and Clinical Neurophysiolo~, 55:131-144, 1983. Gawin, F.; Allen, D.; and Humblestone, 3. Outpatient treatment of “crack” cocaine smoking with ~upent~ol decanoate. Archives of General Psychiatry, 46~322-325, 1989,. Gawin, F., and Ellinwood, E. Cocaine and other stimulants. New England Journal of Medicine, 318(18):1173-1182, 1988. Gawin, F., and Kleber, H. Abstinence symptomatology and psychiatric diagnosis in cocaine abusers. Archives of General Psychiatry, 43: 107-l 13, 1986. Gawin, F. H.; Kleber, H.; Byck, R.; Rounsaville, B.; Kosten, T.; Jatlow, P.; and Morgan, C. Desipramine facilitation of initial cocaine abstinence. Archives of General Psychiatry, 46: l17121, 1989b. Gevins, A. S.; Zeitlin, G.M.; Doyle, J.C.; Yingling, CD.; Schaffer, R.E.; Callaway, E.; and Yeager, CL. El~troen~ph~o~~ correlates of higher cortical function. Science, 203:665668,1979. Giannini, A. J.; Felts, D.J.; Feather, J.; and Sullivan, B. Bromocriptine and amantadine in cocaine detoxification. Psychiatry Research, 29: 1l-16, 1989. Gittelman, R.; Mannuua, S.; Shenker, R.; and Bonagura, N. Hyperactive boys almost grown up. Archives of General Psychiatry, 42:937-947, 1985. Hamilton, M. Assessment of psychopathology. In: Hindmarch, I., and Stonier, P., eds. Human Psychophar~acology~ Measures and Methods. New York: John Wiley & Sons, 1987. pp. I-17. Harmony, T.; Alvarez, A.; Pascual, R.; Ramos, A.; Marosi, E.; Leon, A.E.D.D.; Valdes, P.; and Becker, J. EEG maturation in children with different economic and psychosocial characteristics. International Journal of Neuroscience, 31: 103-I 13, 1987. Herning, R.; Jones, R.; Hooker, W.D.; Mendelson, J.; and Blackwell, L. Cocaine increases EEG beta: A replication and extension of Hans Berger’s historic experiments. Electroencephalography and Clinical Neurophysiology, 601470477, 1985. Ingvar, D.; Sjiilund, B.; and Ardii, A. Correlation between dominant EEG frequency, cerebral oxygen uptake and blood flow. Electroen~ephalography and Clinical Neurophysiology, 41~268-276, 1976. Itil, T. The discovery of ~tidepressant drngs by computer-an~yzed human cerebral bioelectrical potentials (CEEG). Progress in Neurobiology, 20~185-249, 1983. John, E.R.; Ahn, H.; Prichep, L.; Trepetin, M.; Brown, D.; and Kaye, H. Developmental equations for the electroencephalogram. Science, 210:1255-1258, 1980. John, E.R.; Prichep, L.; Ahn, H.; Easton, P.; Fridman, J.; and Kaye, H. Neurometric evaluation of cognitive dysfunctions and neurological disorders in children. Progress in Neurobiology, 21~239-290, 1983.
John, E.; Prichep, L. S.; Ahn, H.; Kaye, H.; Brown, D.; Easton, P.; Karmel, B.L.; ‘I oro, A . and Thatcher, R. Neurometric Evaluation of Brain Function in Normal and l.earninic Disabled Children. Ann Arbor, Ml: University of Michigan Press, 1989. pp. i-2 19. John, E.; Prichep. L.; and Easton, P. Normative data banks and neurometrics: Basic concepts, methods and results of norms constructions. In: Remond, A., cd. ~undbo~~k oj Electroencephalography and ~~in~~ul Neurophysiolog~~~:Vol I II. Computer .4nuiysi.s of zhrp EEG and Other Neurophysiologi~a~ Signals. Amsterdam: Elsevier, 1987. pp. 449-495. John, E.; Prichep, L.: and Easton, P. Neurometrics: Computer assisted differential diagnosis of brain dysfunctions. Science, 392:162-169, 1988. Jones, F., and Holmes, D. Alcoholism, alpha production and biofeedback. -Journal’ o~# Consulting and Clinical Psycholog.v, 44(2):224-228,
1976.
Jonkman, E.; Poortvliet, D.; Veering, M.; deWeerd, A.; and John, E. ‘The use of neurometrics in the study of patients with cerebral ischemia. ~~e~troen~epha~~~graph~ and Clinical ~europh,vsioiogy, 5 1333-34 1. 1985. Kaye, H.; John, E,; Ahn, H.; and Prichep, L. Neurometric evaluation of learmng disabled children. tnternat~onal Journal of Neuroscience, i 3: 15-25, 198 1. Khantzian, E. The self-medication hypothesis of addictive disorders: Focus on heroin and cocaine dependence. American Journal of Psychiatry, 142: 1259-I 264, 1985. Matousek, M., and Petersen, I. Norms for the EEG. In: Kellaway, P., and Petersen, I., eds. Automation of Clinical Electroencephalographv. New York: Raven Press, 1983. pp. 75102. Naitoh, P. The value of electroencephalography in alcoholism. Annuls of the New York Academy of Sciences, 2 15:303-320. 1973.
Nunes, E.; Quitkin, F.M.; and Klein, D. Psychiatric diagnosis in cocaine abuse. Ps_vchiatry Research, 28:105-l 14, 1988.
Obrist, W.D.; Sokoloff, I..; Lassen, N.A.; Lane, M.H.; Butler, R.N.; and Feinberg, 1. Relation of EEG to cerebral blood flow and metabolism in old age. Electroencephalography and Clinical Neurophysiology,
15:610-619, 1963.
Oken, B., and Chiappa, K. Short-term encephalography
variability
in EEG frequency analysis. f?/ecfro-
and Clinical Neurophysiology, 69: 191-198, 1988.
Pollock, V., and Schneider, L. Topographic electroencephalographic alpha in recovered depressed elderly. Journal ofAbnormal Ps,vchology, 98(3):268-273, 1989. Pollock, V., and Schneider, L. Quantitative. waking EEG research on depression. Biological Ps~vchiatr.v,27:757-780,
1990.
Rapoport, J.; Quinn, P.; Bradbard, G.; Riddle, D.; and Brooks, E. Imipramine and methylphenidate treatments of hyperactive boys. Archives ofGeneral Psychiatry, 30:789-793. 1974.
Schaffer. C.; Davidson, R.; and Saron, C. Frontal and parietal electroencephalogram asymmetry in depressed and nondepressed subjects. Biological Psychiatry, 18(7):753-762. 1983.
Simeon, J.; Ferguson, H.; and Fleet, J.V.W. Bupropion effects in attention deficit and conduct disorders. Canadian Journal of Psychiatry, 31:581-585, 1986. Simeon, J.; Ferguson, B.; Knott, V.; Rochon, R.; and Fleet, J.V.W. Pharmaco-EEG and clinical effects of bupropion in children with attention deficit and conduct disorders. Clinical Neuropharmacology, 7: 134-135, 1984. Straumanis, J.; Struve, F.; Patrick, G.; and Raz, Y. Quantitative EEG and cognitive evoked potentials in chronic marijuana users. Electroencephalography and Clinical Neurophysiology, 75( 1):S145, 1990.
Struve, F.; Straumanis, J.; Patrick, G.: and Price, L. Topographic mapping of quantitative EEG variables in chronic heavy marijuana users: Empiricai findings with psychiatric patients. Chnical Elertroen~epha1ograph.v. 20( I ):6-23. 1989.
105 Sulg, I.; Sotaniemi, K.; Tolonen, U.; and Hokkanen, E. Dependence between cerebral metabolism and blood flow as reflected in the quantitative EEG. Advances in Biological Psychiatry, 6:102-108, 1981.
Susser, E.; Struening, E.; and Conover, S. Psychiatric problems in homeless men. Archives of General Psychiatry, 46:845-850,
1989.
Verebey, K., and Gold, M. From coca leaves to crack: The effects of dose and routes of administration in abuse liability. Psychiatric Annals, 18(9):513-520, 1988. Volkow, N.; Mullani, N.; Gould, K.L.; Adler, S.; and Krajewski, K. Cerebral blood flow in chronic cocaine users: A study with positron emission tomography. British Journal of Psychiatry, 152:641-648, 1988.
von Knorring, L.; Perris, C.; Goldstein, L.; Kemali, D.; Monaakhov, K.; and Vacca, L. Intercorrelations between different computer-based measures of the EEG alpha amplitude and its variability over time and their validity in differentiating healthy volunteers from depressed patients. Advances in Biological Psychiatry, 13: 172-181, 1983. Waldrop, M. NIDA aims to fight drugs with drugs. Science, 245:1443-1444, 1989. Winsberg, B.; Bailer, I.; Kupietz, S.; and Tobias, J. Effects of imipramine and dextroamphetamine on behavior of neuropsychiatrically impaired children. American Journal of Psychiatry, 128:1425-1431, 1972.
World Health Organization. Nomenclature and classification of drug- and alcohol-related problems: A WHO memorandum. Bulletin of the World Health Organization, 59(2):225-242, 1981.
Yingling, C.; Galin, D.; Fein, G.; Peltzman, D.; and Davenport, L. Neurometrics does not detect “pure” dyslexics. Electroencephalography and Clinical Neurophysiology, 63:426-430, 1986.
95
Quantitative EEG Correlates of Crack Cocaine Dependence Kenneth I?. Alper, Robert J. Chabot, Anthony E. Roy John Received
H. Kim, Leslie S. Prichep, and
May 18, 1990; revised version received August 17, 1990; accepted August 19, 1990.
Abstract. Evidence dependence
for a distinctive syndrome of neuroadaptation has accumulated from behavioral, neurophysiological,
in cocaine and preclin-
ical and clinical pharmacological studies. The authors report on the results of a preliminary investigation of the quantitative electroencephalographic (QEEG) correlates of severe DSM-III-R crack cocaine dependence in seven patients abstinent from cocaine for 1 to 68 days. The major QEEG finding was increased absolute and relative alpha power. Increased alpha power has also been reported in multiple previous studies of depressed patients. This series of crack-dependent patients showed significant depressive morbidity; four patients attempted suicide subsequent to initiating their use of crack and the group mean (31 SD) Beck Depression Scale score was 18.9 (AI 6.5). These results complement other studies that support the concept of neuroadaptation to chronic cocaine exposure. Prospective studies correlating QEEG measures with subsequent response to pharmacological interventions for cocaine dependence should be considered. Key Words. Cocaine, crack, electroencephalogram.
Although the acute effects of cocaine on the human electroencephalogram (EEG) were described by Hans Berger over 50 years ago (Berger, 1937), no study has yet appeared describing the EEG in chronic cocaine dependence. This may be due in part to the belated recognition of a syndrome of tolerance and withdrawal associated with cncaine &-j not_ ~&IJ& cQ&nc denendc?ncc? __d-____ me* -__, in ___ 1980 ____ j)&M-_r_ll __~ _-____-__ RI -I -a ctina-_-. nostic entity (American Psychiatric Association, 1980). The current epidemic of crack cocaine dependence (Gawin and Ellinwood, 1988) lends special relevance to the investigation of this disorder. Phenomenological evidence indicates the existence of a characteristic progression of abstinence symptoms in withdrawing addicts (Gawin and Kleber, 1986). Decreased regional cerebral blood flow (KBF), measured by positron emission tomography (PET), has been noted in abstinent cocaine abusers (Volkow et al., 1988). Pharmacological interventions selected for their interaction with neurotransmitter systems subserving reward-mediated behavior such as flupen-
Ya.nnmeh ..“..L.“l‘.
I? ‘q,W., Al...ar I..
M n I.X.Y.)
in I?ee,.nroh 111 I.I”“...“L.
Arr;rtnnt ‘1.,“1I+...a.
Pmf,wcrw. 1 ‘“lrlll”I,
Rr\hrti I\““w.+
1 4.
f-hcahnt “l.UY”L,
PhIS,_., l-J 1
is &g&t~
Professor; and Anthony H. Kim, M.D., is Fellow in Alcohol and Substance Abuse, Department of Psychiatry, New York University Medical Center. Leslie S. Prichep, Ph.D., is Associate Professor of Psychiatry and Assistant Director of Brain Research Laboratories, and E. Roy John, Ph.D., is Professor of Psychiatry and Director of Brain Research Laboratories, Nathan S. Kline Research Institute, Orangeburg, NY, and Department of Psychiatry, New York University Medical Center. (Reprint requests to Dr. K.R. Alper, Brain Research Laboratories, 8th floor, Old Bellevue Administration Building, First Avenue at 27th St., New York, NY 10016, USA.) 0~65-~781~~~~3.50
@ 1990 Elsevier Scientific Publishers Ireland Ltd.
thixol (Gawin et al., 1989a), desipramine (Gawin et al., 19896). and bromocriptmt (Giannini et al., 1989) have shown evidence of efficacy in treating the anhedonia and craving of the withdrawal state. Collectively. the available behavioral, neurophysiological, and pharmacological evidence appears to support the concept of a syndrome of neuroadaptation in cocaine dependence (World Health Organization. I98 I : Dackis and Gold, 1985; Gawin and Kleber, 1986). The quantitative EEG (QEEG) correlates of cocaine withdrawal are of interest for several reasons. As a reflection of central nervous system function, QEEG features may provide an approach to the investigation of the neurophysiological basis of the cocaine withdrawal syndrome. The substantial prevalence of cocaine abuse in psychiatric populations (Gawin and Kleber, 1986; Susser et al., 1989) has made an understanding of the QEEG correlates of cocaine dependence necessary to resolve the possible effects of cocaine exposure from the abnormalities seen in other psychiatric disorders (John et al.. 1988). The discovery of characteristic QEEG profiles associated with certain classes of psychopharmacological agents has been applied successfully toward the development of new psychiatric drugs (Itil, 1983). It is possible that an understanding of the QEEG correlates of cocaine withdrawal might provide a contribution to the screening and development of compounds directed at the anhedonia and craving of the withdrawal state. In the present study, we report on the EEG in seven individuals with severe crack cocaine dependence. Due to concern over the possibly confounding effects of psychiatric and other comorbidity related to substance abuse, subjects were selected who lacked a psychiatric history before their crack use, had never been dependent on substances other than cocaine, and were not intoxicated with crack or other substances at the time of the examination. While these subjects may not represent a typical cross-section of the crack-abusing population as a whole, they could safely be considered to be “hard core” and more or less exclusive users of crack cocaine. Methods Subjects. Seven patients were included in this study. Four patients came from an inpatient unit at Bellevue dedicated to the treatment of substance dependence. Three subjects were outpatients who presented themselves for treatment of crack cocaine dependence to the Bellevue Emergency Psychiatric Service. The four inpatients were off medication for at least 7 days before the EEG session. Of these four inpatients, two were tested 7 days, one 20 days, and one 68 days after admission to the hospital. The three outpatients were off cocaine for at least 24 hours before the EEG session; two reported having abstained from cocaine use for 1 day and another for 3 days before the EEG testing. The procedures for verification of patients’ reports about their length of time off crack and their use of other substances are described below. Table 1 lists variables describing demographic and clinical characteristics and cocaine history of the patient group. Criteria for inclusion in the study were as follows: (1) Fulfilling the criteria for DSM-HZ-R (American Psychiatric Association, 1987) dependence on cocaine in the form of crack for at least 1 year. Those with a history of ever having met the criteria for DSM-III-R dependence on any other substance were excluded. All subjects reported that they strongly preferred crack to any other substance of abuse. Subjects who reported alcohol intake confined their drinking to periods of crack use and used alcohol to modify unpleasant subjective activating effects of
Table 1. Subject characteristics (n = 7) AgeWI sex Race Years since first intranasal cocaine use Years since first crack cocaine use Estimated weekty crack cocaine us8 during 6-~nth preceding el~troenceph~~ram (g/w~k) Days since most recent crack use Hamilton Rating Scale for Depression score
24.6 f 4.5 4 males 3 females 4 black, 2 Hispanic, 1 white 4.4 rir:1.6 2.0 f 1.6 period
Beck Depression Inventory score Number of subjecta reporting suicide attempts before beginning crack use Number of subjects reporting suicide attempts after beginning crack use Number of subjects se&supporting at the time of beginning crack use Number of subjects self-supporting at the time of intake into the study
8.5 f 8.3 15.3 (range l-68) 13.1 xt 6.5
18.9 f 6.5 017 417 417 l/7
Note. Dataare presentedas meanf SD.
crack. No subject actively sought or purchased cannabis in the 60 days preceding the study, although three did use cannabis occasionally when it was available. (2) Negative urine screening for amphetamine, barbiturates, methamphetamine, opiates, methadone, tricyclic antidepressants (amitriptyline is a commonly abused drug in the community from which subjects were selected), benzodiazopines, phencyclidine, and propoxyphene at the time of EEG. (3) No use of cocaine or other drugs of abuse for at least 24 hours preceding the study. Outpatients were asked to abstain from crack use for at least 24 hours before the study but were told that the EEG would be done, and that they would receive payment ($20) in any event if they attended the intake appointment. Evidence from previous studies of cocaine users suggests that subjects can be reliable about their substance use if no adverse consequences follow disclosure of illicit drug use (Gawin et al., 1989~). For outpatients, it was required that they reside with an individual (in all cases this turned out to be a parent) who did not abuse drugs and who could verify that the patient had remained at home and apparently not used crack for at least 24 hours preceding the study. In practice, the pattern of binges associated with crack use was obvious to family members. No subject had received treatment with a psychotropic agent for at least 60 days before the study. Inpatients were evaluated at least 7 days after admission, and had not been treated with any psychotropic agent while in the hospital. Urine testing was used to confirm the absence of illicit drug use in the hospital. (4) No history of head trauma or a neurological or medical condition known to affect the EEG and no history of ever having used any drug by the intravenous route. A history of seizures limited to periods of cocaine intoxication was also grounds for exclusion. (5) No history of psychiatric hospit~ization or having been treated with psychotropic medication before using crack. EEG Data Acquisition. Eyes-closed, resting EEG data were collected from the 19 monopolar electrodes of the International lo/20 System, referred to linked earlobes. A differential eye channel was used for the detection of eye movement. The EEG amplifiers had a bandpass from 0.5 to 30 Hz (3 dB points), with a 60-Hz notch filter. The data were digitized at 100 Hz with 1Zbit resolution. Twenty minutes of continuous EEG data were collected for subsequent editing and analysis.
3x
EEG Data Analysis. All data acqutsttion and analyses were performed using the methodology of neurometrics on a Cadwell Spectrum 32 (software revision ~3.1). Quantitative features are extracted, log-transformed to obtain normal (Gaussian) distributions, ageregressed, and z-transformed relative to population norms. The reader is referred elsewhere for a general overview of the methodology (John et al., 1983, 1988) and for articles addressing specific topics such as the validity of the use of log transformation of QEEG data to obtain Gaussianity (Gasser et al.. 1982; Gken and Chiappa, 1988) development of the normative data base (John et al.. 1987). construction of age regression equations (John et al.. 1980, 1988). studies of test-retest reliability (Kaye et al., 1981; John et al., 1983; Fein et al., 1984), and independent replications of the QEEG norms (Matouse’k and Petersen, 1973; Gasser et al., 1983; Jonkman et al.. 1985: Yingling et al., 1986; Alvarez et al., 1987; Harmony et al., 1987: John et al., 1989). The neurometric adult EEG norms have been derived from a group of 150 normal subjects, aged 17 to 90 (John et al., 1988). On the basis of the distribution statistics obtained from these normative data, it is possible to calculate z-values or standard scores for these features (proportional to the probability of abnormality) in the evaluation of index subjects. The data were edited using computerized artifact-detection algorithms combined with visual inspection to obtain 24 epochs (2.5 set each, a total of 1 min) of artifact-free data from 20 min of continuous EEG. Power spectral analysis was performed using the Fast Fourier Transform (FFT). For each of the 19 monopolar derivations, absolute and relative (%) power and mean frequency were computed for the delta ( 1.5-3.5 Hz), theta (3.5-7.5 Hz), alpha (7.5-12.5 Hz), and beta (12.5-25 Hz) frequency bands. Overall absolute power and mean frequency for the entire spectrum from I .5 to 25 H7 were also computed
Results Clinical Data. Four of seven subjects attempted suicide after they began to use crack. The mean 21-item Beck Depression Inventory score of 18.9 also suggests significant depressive morbidity (Beck, 1967). A trend for correlation between Hamilton (24 item) and Beck scores, commonly seen in studies of depressed patients (Hamilton, 1987) was not evident in this study (r = 0.17, p > 0.10).Of the four patients who had been economically self-supporting (defined as residing in a household for a period of 1 year in which the patient’s legitimately earned income accounted for at least one-half of the rent and household expenses) immediately before the initiation of their use of crack, only one remained legitimately selfsupporting at the time of evaluation. The substance abuse histories of the individual subjects revealed that while none of them were naive to other drugs of abuse such as ethanol and cannabis, they greatly preferred, sought, and used crack relative to other substances. All subjects used intranasal cocaine before beginning to use crack and reported an escalation of cocaine intake after using crack, a phenomenon that has been reported elsewhere (Gawin and Ellinwood, 1988; Verebey and Gold, 1988). EEG Data. Fig. 1 presents topographic maps of group mean z scores for absolute and relative power and mean frequency for each of the four frequency bands. It should be recalled that the color-coded values displayed designate the probability of a given value being within the confidence limits of EEG measurements obtained from normal individuals. The values are age-regressed and take into account the topographic variations of normative EEG data. The major trend is an absolute and relative increase in alpha power. Values for
99 Fig. 1, Topographic head maps for group mean z scores for absolute power (top row), relative power (middle row), and mean frequency (bottom row) DELTA
Abso
THETA
ALPHA
BETA
TOTAL
1.
Power
Color coding denotes the value of the group mean z score, which is proportional lo the probability that the data obtained from the 7 index subjects differ from the reference population of 150 normal adults. In estimates of the significance of a group mean z score, it is necessary to take the number of index subjects in the group into account. In the above figure, for example, the estimate of the critical value denoting significance at the level of p = 0.05 is f 1.96/\j7 = f 0.74 and is designated by the horizontal arrows to the right of the vertical color scale.
excess alpha are significant at the 0.05 level at 17 of 19 leads for absolute power and 18 of 19 leads for relative power. Inspection of the data of individual subjects reveals that the relative alpha excess is present in each subject and is apparently consistent in each member of the group. Fig. 2 presents examples of high-amplitude alpha seen in study subjects, which are compared with a representative epoch obtained from a normal control. Absolute delta power is significantly decreased in five of the six frontal leads. A relative but not absolute “hyperfrontal” distribution of alpha power is due to decreased delta rather than significant topographic variability of increased alpha. There appears to be a weak trend for increased absolute beta power, which reaches significance in 4 of 19 leads. Relative power in all frequencies except alpha is decreased as a result of the widespread alpha excess. There is a trend for increased mean EEG frequency greater in the anterior leads for delta and theta but not alpha and beta. Discussion The apparent effect of crack on the lives of this small group of subjects is impressive. Suicide attempts in four of seven subjects after they began to use crack and the mean Beck score of 18.9 suggest significant depressive morbidity. A trend for a correlation between Beck and Hamilton scores, which‘has been reported in previous studies of depression (Hamilton, 1987), was not present in this study. This may suggest a phe-
%i vemca, SCale of lne eiecrroencephalograniEEGi ,s 50 sir i‘m. The stib!r?c; on the f&R is a normal maie. aged 24 years z-i!e subject x the .middie li d aged 28 years abstrnent from cocaine for 7 days The sublect on the nghf !s a crack-dependent male, aged 23. abstinent fron n cocaine for 20 days
3-r .&veCde;’
.-i;:e
Fig. 2. Representative 2.5set EEG epochs obtained from a normal subject (left) and 2 crack-dependent subjects #I554 NON-USER (middle & right) CRACK USER PT. fi8443
101
nomenological distinction between the anhedonia and dysphoria of cocaine withdrawal and the full constellation of clinical symptomatology seen in affective illnesses. The clinical data suggest a dramatic decline in the functioning of individual crack users. The study subjects had all initially presented themselves to the emergency psychiatric service and may possibly reflect the more severe end of the full clinical spectrum of crack abuse. Nonetheless, three of four subjects who had supported themselves by legitimately earned income before they began to use crack were no longer self-supporting after becoming dependent on crack. It is unlikely that the effects of acute cocaine intoxication influenced the results reported here. The half-life of cocaine is on the order of 40 min (Fischman, 1987) and subjects were off cocaine for at least 24 hours before the EEG, with four subjects abstinent at least 7 days. The alpha increase was present in individual patients who had been abstinent for 7 or more days (Fig. 2). Furthermore, the acute effect of cocaine appears to be an increase in beta (Berger, 1937; Herning et al., 1985), and not alpha as noted in this study of crack cocaine withdrawal. The finding of increased alpha noted in this study of cocaine abusers is remarkably similar to the results reported by Struve and his colleagues for chronic cannabis abusers (Struve et al., 1989; Straumanis et al., 1990). Several possibilities exist. The results reported here may reflect a history of cannabis use rather than crack exposure. The very low use of cannabis relative to crack in the study population argues against this explanation. Another possibility is that cannabis- and cocainedependent subjects actually do have similarities in their QEEG profiles. Such a finding would bear an intriguing relevance to arguments for hypothetical “common final pathways” in substance use disorders (Gardner et al., 1988; Waldrop, 1989). However, consideration of QEEG evidence in the context of theories of common end effects in substance use disorders must take into account the QEEG correlates of alcohol dependence, which consist of low overall absolute power and increased relative power in beta (Funkhouser et al,, 1953; Naitoh, 1973; Jones and Holmes, 1976; Gabrielli et al., 1982; John et al., 1988) and differ from the findings for crack cocaine dependence and for cannabis dependence. The classic concept of alpha varying inversely with cerebral activation (Gevins et al., 1979) would appear to put the alpha increase noted in this study in accord with the observation of decreased rCBF in cocaine abuse (Volkow et al., 1988). Increased alpha might represent a correlate of a hypothetical depression in cerebral metabolism. This interpretation, however, may not explain the mean EEG frequency findings reported here. In studies involving the measurement of QEEG and rCBF in normal individuals, a positive correlation of overall mean frequency with rCBF has been observed (Obrist et al., 1963; Ingvar et al., 1976; Sulg et al., 1981). On the basis of these studies, the expected result in a state of cerebral metabolic depression would be a reduction in overall mean EEG frequency. Overall mean EEG frequency was not decreased in this study and actually tended to be increased in the anterior leads. Interpretation of QEEG data in terms of a hypothetic~ alteration of metabolism in cocaine withdrawal will require more information on metabolic correlates of the EEG. This issue may be addressed productively by obtaining simultaneous PET and QEEG in normal individuals and in subjects with cocaine dependence.
102
The finding of increased alpha is of interest in view of the results of QEEG investigations of two psychiatric conditions thought to predispose to cocaine abuse: depression and attention deficit hyperactivity disorder (ADHD) (Gittelman et al., 1985; Khantzian, 1985; Gawin and Kleber, 1986; Nunes et al., 1988). Increased alpha appears to be a consistent finding in studies of depression @chaffer et al., 1983; von Knorring et al., 1983; Brenner et al., 1986; John et al., 1988; Pollock and Schneider. 1989, 1990). Considerable evidence has accumulated in the pharmaco-EEG literature for a characteristic antidepressant QEEG profile (Itil, 1983). Diminution of alpha is a prominent feature of this putative antidepressant QEEG profile. A clinically effective antidepressant, mianserin, was selected for development on the basis of the similarity of its QEEG effects to those of antidepressants in preclinical studies (It& 1983). Tricyclic antidepressant medication has been used with success in cocaine dependence (Gawin et al., 19896) and also in ADHD (Wins~rg et al,, 1972; Rapoport et al., 1974; Garfinkelet al., 1983; Donnelly et al., 1986; Biederman et al., 1989). In a recent trial of buproprion for ADHD (Simeon et al., 1984, 1986), increased alpha power at baseline correlated with subsequent response to buproprion. It may prove rewarding to investigate the relationship of the apparent alpha increase in crack cocaine dependence to subsequent response to pharmacological interventions, particularly those with antidepressant effects. References Alvarez, A.; Pascual, R.; and Valdez, P. U.S. EEG developmental equations confirmed for Cuban schoolchildren. Electroencephalography and Clinical Neurophysiology. 67:330-332, 1987. American Pspychiatric Association. DSM-III: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC: APA, 1980. American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of dental Disorders. 3rd ed., revised. Washington, DC: APA, 1987. Beck, A. Measurement of depression: The Depression Inventory. In: Depression: Causes and Treatment. Philadelphia: University of Pennsylvania, 1967. Chapter 12, pp. 186-207. Berger, H. Electroencephalogram of man. Archiv ftir Psychiatric tind Nervenkrankheiten, 106:577-584, 1937. Biederman, J.; Baldessarini, R.; Wright, V.; Knee, D.; and Harmatz, J. A double-blind placebo controlled study of desipramine in the treatment of ADD: I. Efficacy. Journal of the American Academy of Child and Adolescent Psychiatry, 28(5):777-784, 1989. Brenner, R.; Ulrich, R.; Spiker, D.; Sclabassi, R.; Reynolds, C.; Marin, R.; and Boller, F. Computerized EEG spectral analysis in elderly normal, demented and depressed subjects. Electroencephalograp~y and Clinical Neurophysiology, 64:433-492,1986. Dackis, C., and Gold, M. New concepts in cocaine addiction: The dopamine depletion hypothesis. Neuroscience and Biobehavioral Reviews, 9:469477, 1985. Donnelly, M.; Zametkin, A.; Rapoport, .I.; Ismond, D.; Weingartner, II.; Lane, E.; Oliver, J.; Linnoila, M.; and Potter, W. Treatment of childhood hyperactivity with desipramine: Plasma drug concentration, cardiovascular effects, plasma and urinary catecholamine levels, and clinical response. Clinical Pharmacology and Therapeutics, 39:72-8 1, 1986. Fein, G.; Galin, D.; Yingling, C.; Johnstone, J.; and Nelson, M. EEG spectra in 9-13-yearold boys are stable over I-3 years. ~lect~oencephalo~aphy and Clinical Neurophysiology, 58517-518, 1989.
103 Fischman, M. Cocaine and ~phetamines. In: Meltzer, H.Y., ed. Psychopharmacology: The Third Generation of Progress. New York: Raven Press, 1987. Chapter 166, pp. 1543-1553. Funkhouser, J.; Nagler, B.; and Walke, N. The electro-encephalogram of chronic alcoholism. Southern Medical Journal, 46(5):423-428, 1953. Gabrielli, W.; Mednick, S.: Volavka, J.; Pollock, V.; Schulsinger, F.; and It& T.M. Electroencephalograms in children of alcoholic fathers. Psychophysiology, 19:404-407,1982. Gardner, E.; Paredes, W.; Smith, D.; Donner, A.; Milling, C.; Cohen, D.; and Morrison, D. Facilitation of brain stimulation reward by 6Qetrahydrocannabinol. Psychopharmacology, 96:142-144,1988. Garfinkel, B.; Wender, P.H.; Sloman, L.; and G’Neill, I. Tricyclic ~tidepr~s~t and methylphe~date treatment of attention deficit disorder in children. Journal of the American Academy of Child and Adolescent Psychiatry, 22(4):343-348, 1983. Gasser, T.; Bather, P.; and Moths, J. Transformation towards the normal distribution of broadband spectral parameters of the EEG. Electroencephalography and Clinical Neurophysiology, 53: 119-124, 1982. Gasser, T.; Moths, J.; Lenard, H.; Bather, P.; and Verleger, R. The EEG of mildly retarded children: Developmental, classificatory and topographic aspects. Electroencephalography and Clinical Neurophysiolo~, 55:131-144, 1983. Gawin, F.; Allen, D.; and Humblestone, 3. Outpatient treatment of “crack” cocaine smoking with ~upent~ol decanoate. Archives of General Psychiatry, 46~322-325, 1989,. Gawin, F., and Ellinwood, E. Cocaine and other stimulants. New England Journal of Medicine, 318(18):1173-1182, 1988. Gawin, F., and Kleber, H. Abstinence symptomatology and psychiatric diagnosis in cocaine abusers. Archives of General Psychiatry, 43: 107-l 13, 1986. Gawin, F. H.; Kleber, H.; Byck, R.; Rounsaville, B.; Kosten, T.; Jatlow, P.; and Morgan, C. Desipramine facilitation of initial cocaine abstinence. Archives of General Psychiatry, 46: l17121, 1989b. Gevins, A. S.; Zeitlin, G.M.; Doyle, J.C.; Yingling, CD.; Schaffer, R.E.; Callaway, E.; and Yeager, CL. El~troen~ph~o~~ correlates of higher cortical function. Science, 203:665668,1979. Giannini, A. J.; Felts, D.J.; Feather, J.; and Sullivan, B. Bromocriptine and amantadine in cocaine detoxification. Psychiatry Research, 29: 1l-16, 1989. Gittelman, R.; Mannuua, S.; Shenker, R.; and Bonagura, N. Hyperactive boys almost grown up. Archives of General Psychiatry, 42:937-947, 1985. Hamilton, M. Assessment of psychopathology. In: Hindmarch, I., and Stonier, P., eds. Human Psychophar~acology~ Measures and Methods. New York: John Wiley & Sons, 1987. pp. I-17. Harmony, T.; Alvarez, A.; Pascual, R.; Ramos, A.; Marosi, E.; Leon, A.E.D.D.; Valdes, P.; and Becker, J. EEG maturation in children with different economic and psychosocial characteristics. International Journal of Neuroscience, 31: 103-I 13, 1987. Herning, R.; Jones, R.; Hooker, W.D.; Mendelson, J.; and Blackwell, L. Cocaine increases EEG beta: A replication and extension of Hans Berger’s historic experiments. Electroencephalography and Clinical Neurophysiology, 601470477, 1985. Ingvar, D.; Sjiilund, B.; and Ardii, A. Correlation between dominant EEG frequency, cerebral oxygen uptake and blood flow. Electroen~ephalography and Clinical Neurophysiology, 41~268-276, 1976. Itil, T. The discovery of ~tidepressant drngs by computer-an~yzed human cerebral bioelectrical potentials (CEEG). Progress in Neurobiology, 20~185-249, 1983. John, E.R.; Ahn, H.; Prichep, L.; Trepetin, M.; Brown, D.; and Kaye, H. Developmental equations for the electroencephalogram. Science, 210:1255-1258, 1980. John, E.R.; Prichep, L.; Ahn, H.; Easton, P.; Fridman, J.; and Kaye, H. Neurometric evaluation of cognitive dysfunctions and neurological disorders in children. Progress in Neurobiology, 21~239-290, 1983.
John, E.; Prichep, L. S.; Ahn, H.; Kaye, H.; Brown, D.; Easton, P.; Karmel, B.L.; ‘I oro, A . and Thatcher, R. Neurometric Evaluation of Brain Function in Normal and l.earninic Disabled Children. Ann Arbor, Ml: University of Michigan Press, 1989. pp. i-2 19. John, E.; Prichep. L.; and Easton, P. Normative data banks and neurometrics: Basic concepts, methods and results of norms constructions. In: Remond, A., cd. ~undbo~~k oj Electroencephalography and ~~in~~ul Neurophysiolog~~~:Vol I II. Computer .4nuiysi.s of zhrp EEG and Other Neurophysiologi~a~ Signals. Amsterdam: Elsevier, 1987. pp. 449-495. John, E.; Prichep, L.: and Easton, P. Neurometrics: Computer assisted differential diagnosis of brain dysfunctions. Science, 392:162-169, 1988. Jones, F., and Holmes, D. Alcoholism, alpha production and biofeedback. -Journal’ o~# Consulting and Clinical Psycholog.v, 44(2):224-228,
1976.
Jonkman, E.; Poortvliet, D.; Veering, M.; deWeerd, A.; and John, E. ‘The use of neurometrics in the study of patients with cerebral ischemia. ~~e~troen~epha~~~graph~ and Clinical ~europh,vsioiogy, 5 1333-34 1. 1985. Kaye, H.; John, E,; Ahn, H.; and Prichep, L. Neurometric evaluation of learmng disabled children. tnternat~onal Journal of Neuroscience, i 3: 15-25, 198 1. Khantzian, E. The self-medication hypothesis of addictive disorders: Focus on heroin and cocaine dependence. American Journal of Psychiatry, 142: 1259-I 264, 1985. Matousek, M., and Petersen, I. Norms for the EEG. In: Kellaway, P., and Petersen, I., eds. Automation of Clinical Electroencephalographv. New York: Raven Press, 1983. pp. 75102. Naitoh, P. The value of electroencephalography in alcoholism. Annuls of the New York Academy of Sciences, 2 15:303-320. 1973.
Nunes, E.; Quitkin, F.M.; and Klein, D. Psychiatric diagnosis in cocaine abuse. Ps_vchiatry Research, 28:105-l 14, 1988.
Obrist, W.D.; Sokoloff, I..; Lassen, N.A.; Lane, M.H.; Butler, R.N.; and Feinberg, 1. Relation of EEG to cerebral blood flow and metabolism in old age. Electroencephalography and Clinical Neurophysiology,
15:610-619, 1963.
Oken, B., and Chiappa, K. Short-term encephalography
variability
in EEG frequency analysis. f?/ecfro-
and Clinical Neurophysiology, 69: 191-198, 1988.
Pollock, V., and Schneider, L. Topographic electroencephalographic alpha in recovered depressed elderly. Journal ofAbnormal Ps,vchology, 98(3):268-273, 1989. Pollock, V., and Schneider, L. Quantitative. waking EEG research on depression. Biological Ps~vchiatr.v,27:757-780,
1990.
Rapoport, J.; Quinn, P.; Bradbard, G.; Riddle, D.; and Brooks, E. Imipramine and methylphenidate treatments of hyperactive boys. Archives ofGeneral Psychiatry, 30:789-793. 1974.
Schaffer. C.; Davidson, R.; and Saron, C. Frontal and parietal electroencephalogram asymmetry in depressed and nondepressed subjects. Biological Psychiatry, 18(7):753-762. 1983.
Simeon, J.; Ferguson, H.; and Fleet, J.V.W. Bupropion effects in attention deficit and conduct disorders. Canadian Journal of Psychiatry, 31:581-585, 1986. Simeon, J.; Ferguson, B.; Knott, V.; Rochon, R.; and Fleet, J.V.W. Pharmaco-EEG and clinical effects of bupropion in children with attention deficit and conduct disorders. Clinical Neuropharmacology, 7: 134-135, 1984. Straumanis, J.; Struve, F.; Patrick, G.; and Raz, Y. Quantitative EEG and cognitive evoked potentials in chronic marijuana users. Electroencephalography and Clinical Neurophysiology, 75( 1):S145, 1990.
Struve, F.; Straumanis, J.; Patrick, G.: and Price, L. Topographic mapping of quantitative EEG variables in chronic heavy marijuana users: Empiricai findings with psychiatric patients. Chnical Elertroen~epha1ograph.v. 20( I ):6-23. 1989.
105 Sulg, I.; Sotaniemi, K.; Tolonen, U.; and Hokkanen, E. Dependence between cerebral metabolism and blood flow as reflected in the quantitative EEG. Advances in Biological Psychiatry, 6:102-108, 1981.
Susser, E.; Struening, E.; and Conover, S. Psychiatric problems in homeless men. Archives of General Psychiatry, 46:845-850,
1989.
Verebey, K., and Gold, M. From coca leaves to crack: The effects of dose and routes of administration in abuse liability. Psychiatric Annals, 18(9):513-520, 1988. Volkow, N.; Mullani, N.; Gould, K.L.; Adler, S.; and Krajewski, K. Cerebral blood flow in chronic cocaine users: A study with positron emission tomography. British Journal of Psychiatry, 152:641-648, 1988.
von Knorring, L.; Perris, C.; Goldstein, L.; Kemali, D.; Monaakhov, K.; and Vacca, L. Intercorrelations between different computer-based measures of the EEG alpha amplitude and its variability over time and their validity in differentiating healthy volunteers from depressed patients. Advances in Biological Psychiatry, 13: 172-181, 1983. Waldrop, M. NIDA aims to fight drugs with drugs. Science, 245:1443-1444, 1989. Winsberg, B.; Bailer, I.; Kupietz, S.; and Tobias, J. Effects of imipramine and dextroamphetamine on behavior of neuropsychiatrically impaired children. American Journal of Psychiatry, 128:1425-1431, 1972.
World Health Organization. Nomenclature and classification of drug- and alcohol-related problems: A WHO memorandum. Bulletin of the World Health Organization, 59(2):225-242, 1981.
Yingling, C.; Galin, D.; Fein, G.; Peltzman, D.; and Davenport, L. Neurometrics does not detect “pure” dyslexics. Electroencephalography and Clinical Neurophysiology, 63:426-430, 1986.
