During the 20th century, there have been recurrent episodes of cocaine abuse in the United States that have achieved epidemic proportions. Although the total number of people using the drug has decreased in the past decade, cocaine-related biomedical and psychosocial problems remain a major public health problem in the United States and many other countries. The expanding list of cocaine-related toxic effects, acute and chronic, includes reproductive dysfunction, hepatic necrosis, and pulmonary disease. Abuse of and dependence on cocaine may also be associated with disorders of sexual function, sleep, anxiety, and mood, as well as with delirium and psychotic disorders. Many of these problems begin during intoxication with cocaine, but some are most prominent during withdrawal.
Polydrug abuse, which involves abuse of and dependence on cocaine as well as dependence on alcohol, opiates, and nicotine, appears to be increasing in several subgroups of the population, including minority women of childbearing age. One prevalent pattern of such abuse is the concurrent use of cocaine and heroin. The combination of cocaine and heroin, administered intravenously, is called a “speedball,” and deaths associated with the use of this combination are increasing. During 1993–1994, abusers of intravenous cocaine and heroin accounted for a major new group of persons with human immunodeficiency virus (HIV) infection in several large metropolitan areas in the United States and elsewhere. Those who abuse both cocaine and heroin intravenously are at very high risk for HIV infection as a consequence of needle sharing and the combined immunosuppressive effects of the two drugs.
The 1993 National Household Survey on Drug Abuse estimated that 4.5 million Americans used cocaine in 1992. Approximately 1.3 million persons reported frequent cocaine use, defined as use at least monthly. The Drug Abuse Warning Network reported 30,900 cocaine-related visits to the emergency room during the third quarter of 1992, most of which were for cardiovascular, cerebrovascular, and gastrointestinal problems induced by the drug. Concurrent cigarette smoking and cocaine use may also have serious adverse effects on cardiac function. Recent data indicate that cocaine-related injuries are a major cause of death among young adults in New York City.
Cocaine use and abuse are part of a spectrum of substance-abuse disorders that have been classified according to severity by the American Psychiatric Association, which has enumerated the distinguishing clinical features and medical consequences of cocaine-related problems that practicing physicians may encounter. The diagnostic criteria for cocaine intoxication specified in the fourth edition of the association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) are shown in Table 1. Severe cocaine intoxication is associated with potentially life-threatening consequences that require judicious medical management. There are no specific pharmacologic treatments for cocaine intoxication, because the behavioral symptoms and signs (such as psychomotor agitation or retardation) may be diametrically opposite from the physiologic ones (such as elevated or lowered blood pressure). Since cocaine-related deaths and cocaine-induced cardiovascular, cerebrovascular, and gastrointestinal disorders usually occur within 60 to 120 minutes after cocaine use, it is important to retain and monitor all patients who are treated in emergency departments. Fortunately, the drug has a relatively short half-life, and the most severe symptoms and signs of cocaine intoxication decrease within one to two hours after it is used.
The diagnostic criteria for cocaine abuse and dependence are specified in the sections of the DSM-IV listing the criteria for substance abuse (Table 2) and substance dependence (Table 3). In patients with cocaine abuse, drug use is less frequent and less intense than in patients with cocaine dependence. The latter condition is diagnosed partly on the basis of evidence of tolerance for cocaine, compulsive drug acquisition and use, and withdrawal symptoms. The diagnostic criteria specified in the DSM-IV for cocaine withdrawal are shown in Table 4. The dysphoric mood states included among these criteria have also been described by the term “crash” by a number of clinical investigators. However, no symptoms or signs of cocaine withdrawal were noted in several studies of inpatients who were carefully monitored after the cessation of cocaine use.
The inconsistencies in the magnitude and severity of withdrawal symptoms and signs may be due to differences in the doses of cocaine used and the frequency of use. Often cocaine abusers do not know the concentration or purity of the cocaine they have purchased and may not later remember how often they used it. Even though they may report using the same amount of cocaine by the same route of administration at different times, the actual concentration of drug may vary greatly. It has long been known that the onset and severity of symptoms and signs of withdrawal vary widely after a person stops using the psychoactive substance, even when the dose is known, as it is in the case of beverage alcohol. One major factor that could affect the severity of symptoms of cocaine withdrawal is that cocaine-related stimuli are absent in an inpatient setting, whereas neighborhood drug suppliers and cocaine-using acquaintances may be present in outpatient treatment environments.
Pharmacologic Effects and Mechanisms of Abuse and Dependence
Cocaine is a stimulant as well as a local anesthetic with potent vasoconstrictor properties. It induces a complex pattern of subjective effects that have been described as intense euphoria and alertness, increased confidence and strength, heightened sexual feelings, and indifference to concerns and cares. Yet these sensations rapidly change to their antithesis, so that despondency, dejection, and despair prevail. Euphoria is restored by the next dose of the drug, and it is this rapid alternation between ecstasy and dejection that purportedly leads to the “binge” pattern of cocaine use, in which the drug is used repeatedly at short intervals until either the supply or the user is exhausted. Cocaine, like many drugs, produces both pleasure and dysphoria, so the simplistic notion of “reward” does not accurately describe its effects. Instead, cocaine has reinforcing effects, which may be defined as any effect, positive, negative, or both, that maintains the behavior that leads to the continued administration of the drug. The reinforcing properties of cocaine appear to be related to the rapid onset and brief duration of its stimulant effects.
Cocaine can be used orally, intranasally (by insufflation), by inhalation, by intravenous injection, or by the smoking or inhalation of coca paste, cocaine freebase, or “crack” cocaine (freebase prepared with sodium bicarbonate). The onset of salient subjective effects is determined in part by the route of administration and frequency of use. Mood enhancement occurs very rapidly after smoking crack cocaine and after intravenous use, but less rapidly after intranasal administration. Cocaine rapidly increases the heart rate and blood pressure in a dose-related manner. These subjective and physiologic effects are not sustained, however, and cocaine abusers report using the drug repeatedly and frequently during a binge. Cocaine has a relatively short plasma half-life (approximately 60 minutes) and is metabolized primarily by plasma esterases; its metabolites are excreted in urine. It is often used sequentially or simultaneously with other abused drugs. Combinations of cocaine and alcohol produce a metabolite, cocaethylene, that has cardiovascular effects similar to those of cocaine alone. Cocaethylene may also lengthen cocaine-induced euphoria and reduce the unpleasant symptoms of withdrawal.
The accelerating frequency of cocaine abuse has economic and social consequences that are well documented. The seemingly evanescent pleasures of using the drug are accompanied by the risk of death from cardiac arrhythmia, respiratory depression, and convulsions.2 Disorders of cerebral blood flow and perfusion defects due to the potent vasoconstrictive effects of the drug on cerebral arteries are also associated with chronic cocaine abuse.
The neurobiologic mechanisms underlying the effects of cocaine are not well understood. There is considerable evidence, however, that the initiation and continuation of cocaine use are associated with the effects of the drug on the dopaminergic and neuroadrenergic modulation of central nervous system function. Studies in animals suggest that the mesocorticolimbic dopaminergic pathways are important mediators of cocaine’s reinforcing properties. Species may differ, however, with respect to the distribution and interaction of neurons with dopamine receptors and subtypes. Dopaminergic systems are probably not homogeneous. Recent studies of drug self-administration in rodents indicate that dopamine D3 receptors contribute to its reinforcing properties. To date, five dopamine-receptor subtypes have been identified, and their role in inducing and perpetuating cocaine abuse and dependence in humans remains to be determined.
A “dopamine depletion” hypothesis has been advanced to explain the occurrence of symptoms such as depression and anergia after the cessation of cocaine use. Alterations in dopaminergic function after protracted cocaine use may also result in hyperprolactinemia, an indication of impaired dopaminergic regulation of prolactin secretion. Dysregulation of the dopamine system may contribute to immunologic as well as neuroendocrine disorders in cocaine-dependent men and women. The drug also affects serotonergic functions in the central nervous system, and cocaine-induced changes in serotonergic activity may underlie disorders of sleep and wakefulness that occur during chronic cocaine use and during withdrawal.
Cocaine also affects neuroendocrine systems that modulate responsivity to stress and sexual behavior. The drug induces a rapid increase in corticotropin secretion in humans and rhesus monkeys. Cocaine-induced stimulation of corticotropin secretion in rodents was inhibited by the administration of antiserum to corticotropin-releasing hormone. This increased secretion of corticotropin may be associated in humans with the reinforcing properties of cocaine and its rapid enhancement of perceived pleasure and diminution of concern about environmental stressors. Cocaine also stimulates the secretion of luteinizing hormone in rhesus monkeys and humans.43 This effect may be related to the perception of enhanced sexual interest and responsivity after the drug is used. Altered neuroendocrine function during protracted cocaine abuse may result in serious disorders, including altered stress and immune responses, as well as impairment of reproductive function in both sexes.
Important advances have been made in understanding the neuropharmacologic concomitants of cocaine use by humans, subhuman primates, and rodents, but the specific neurochemical and neurophysiologic bases of cocaine reinforcement remain to be determined. Although neuroadrenergic, dopaminergic, and serotonergic functions of the central nervous system are directly affected by cocaine, it is likely that interactions between these neurochemical systems are of considerable importance in understanding the reinforcing properties of cocaine.
Pharmacotherapy for Cocaine Abuse and Dependence
Cocaine abuse and dependence may be initiated and perpetuated by a wide range of physiologic, behavioral, and sociocultural factors. Recent treatment-center evaluations of urban, economically disadvantaged cocaine abusers revealed that 73 percent had one or more personality disorders. Accordingly, the use of existing medications and the development of new ones should be based on accurate diagnosis and a consideration of the multiple interactive processes that may have contributed to the genesis and continuation of cocaine abuse and dependence.
A number of drugs have been used to treat cocaine-related problems, in part because of the postulated role of antecedent disorders in the genesis of chronic abuse, as well as the neurobiologic consequences of abuse and dependence. For example, clinical observations of dysphoric mood changes during cocaine intoxication and withdrawal (Table 1 and Table 4) have led to the use of antidepressant drugs to treat cocaine abuse and dependence. However, the diversity of patterns of use and routes of administration, as well as the concurrent abuse of other drugs, complicates judgments about the efficacy of medication. The conclusions of studies of drugs in persons who use cocaine intranasally (by insufflation) may not apply to persons who inhale cocaine (as crack) or use intravenous cocaine, alone or in combination with opiates. Moreover, drugs that appear to be effective in cocaine abusers may not be useful in treating cocaine dependence.
The evaluation of drug efficacy must be based on the severity of illness. The Addiction Severity Index is one test that facilitates the grading of substance-abuse disorders and is important for research on the treatment of cocaine abuse and dependence. Judgments about drug efficacy can only be made in relation to ratings of the biobehavioral severity of cocaine abuse or dependence. This position was clearly articulated in a recent review of research and clinical perspectives on the treatment of cocaine abuse and dependence that was prepared for the National Institute on Drug Abuse.51 This report emphasized the need to pay more attention to research design, the analysis of treatment methods, research on treatment, and the specification of important diagnostic and demographic characteristics of patients participating in studies of pharmacologic or behavioral treatment.
Many approaches to the treatment of cocaine abuse and dependence have also been used in treating patients with alcoholism and other substance-abuse disorders. There has been interest in comparing the effectiveness of drug therapy with that of psychotherapy, as well as with behavioral treatments of cocaine abuse and dependence. In this review we primarily discuss pharmacotherapeutic approaches to the management of cocaine abuse and dependence, but some recent advances in behavioral treatment are also described. The rationale for the use of various classes of medications is presented in the following sections, with data on treatment outcome. Any assessment of the efficacy of therapy for cocaine abuse and dependence is complicated by the high frequency of abuse of other substances.
Dysphoric moods (Table 4) are frequently reported after the cessation of cocaine use. Clinical observations of depression have provided a rationale for exploring the effectiveness of antidepressant drugs in treating cocaine abuse. Although depression may precede or follow cocaine use, it was hypothesized that ameliorating symptoms of depression might decrease such use. Desipramine, an antidepressant, has been used for both cocaine detoxification and the maintenance of abstinence. A 1984 report that this drug was safe and effective in treating cocaine abuse led to a placebo-controlled, double-blind assessment of the relative efficacy of desipramine in 72 cocaine-dependent persons treated at an outpatient clinical facility. The results indicated that desipramine was beneficial, but in subsequent trials it was found not to be effective in treating patients with cocaine dependence. The assessment of this treatment is complicated by the fact that many cocaine abusers often abuse other substances, such as opiates, and there is considerable disagreement about the effectiveness of desipramine for patients who abuse or are dependent on opiates. At present, desipramine appears to be most effective for persons with diagnosed cocaine abuse (Table 2) who have antecedent or consequent symptoms of severe depression. Desipramine is often ineffective in treating cocaine dependence (Table 3). The results of two recent clinical trials indicated that fluoxetine is ineffective in treating cocaine dependence with or without concurrent dependence on opiates. Other antidepressant drugs, including imipramine and trazodone, have also been used, but they have more adverse effects than desipramine.
Drugs Affecting Dopaminergic Function
The rationale for the use of dopamimetic drugs to treat cocaine abuse and dependence is based in part on the effects of cocaine on dopamine-transporter systems; cocaine blocks the reuptake of dopamine and acts as an indirect dopamine agonist. Moreover, chronic exposure to cocaine is postulated to affect dopaminergic function in the brain adversely (dopamine depletion). A less toxic drug similar to cocaine could be substituted for cocaine in a manner analogous to the use of methadone in opiate-abuse treatment. Unfortunately, the relative effectiveness of dopamimetic drugs in treating cocaine abuse and dependence is often severely compromised by adverse effects, including gastrointestinal disorders such as nausea and abdominal pain, headaches, cardiovascular instability, hypertension, and psychosis-like illness. A number of agonists and antagonists for highly selective dopamine D1, D2, and D3 receptors have also been developed, and although preclinical studies show many to be effective in reducing cocaine use, none are approved for clinical use.
Bromocriptine, a dopamimetic drug used to treat hyperprolactinemia, has been evaluated in both open and placebo-controlled trials. Bromocriptine was reported to decrease the craving for cocaine during detoxification and to reduce dysphoria during both detoxification and abstinence. The dopamine agonist methylphenidate was found to increase rather than decrease the craving for cocaine. Amantadine has also been used in the detoxification of cocaine-dependent patients. An encouraging report of the effectiveness of amantadine in reducing cocaine craving was not confirmed by a double-blind, placebo-controlled study. Both open and placebo-controlled, double-blind studies designed to compare amantadine with desipramine as a treatment for patients in methadone-maintenance programs who also had cocaine abuse or dependence indicated that both drugs increased the time during which patients remained in the programs.
Mazindol inhibits the binding of cocaine to dopamine transporters in the brain. In controlled studies, mazindol neither attenuated the subjective effects associated with intravenous cocaine use in abusers of the drug nor reduced cocaine-induced craving. Bupropion, an antidepressant, was no more effective than placebo in treating cocaine dependence in patients in methadone-maintenance programs. Flupentixol, a dopamine-receptor antagonist, has antidepressant effects at low doses and neuroleptic effects at high doses. Initial open-label studies suggest that flupentixol may be useful in treating cocaine abuse and dependence.
Tryptophan and tyrosine, dietary amino acid precursors of dopamine, have been administered in open-label studies for both cocaine detoxification and the maintenance of abstinence. However, there is no evidence that either is effective for the treatment of cocaine abuse and dependence. In a placebo-controlled, double-blind study, levodopa and carbidopa together did not attenuate symptoms of abstinence after the cessation of cocaine use.
Opioid Antagonists and Mixed Agonist–Antagonists
One rationale for the use of opioid antagonists (such as naltrexone) or opioid mixed agonist–antagonists (such as buprenorphine) to treat cocaine abuse and dependence is that these conditions may be preceded or accompanied by abuse of or dependence on opiates. Endogenous opioid systems in the brain may also be involved in the reinforcing effects of other abused substances, including cocaine and alcohol. Compliance with naltrexone therapy for opiate abuse and dependence has been poor, however, even when there is no cocaine abuse or dependence. No clinical trials of the effectiveness of naltrexone in treating cocaine abuse and dependence have been conducted as yet, but naltrexone appeared more effective than methadone in reducing the abuse of cocaine by opiate-dependent patients. Naltrexone has usually been ineffective in reducing cocaine self-administration in preclinical studies.
In contrast, the opioid-agonist properties of buprenorphine appear to make it more acceptable to patients than naltrexone, even though the two drugs antagonize opioid effects equally well. Buprenorphine is currently under review by the Food and Drug Administration (FDA) for the treatment of opioid dependence, and initial clinical trials were encouraging. In preclinical studies, buprenorphine reduced cocaine use significantly. Clinical studies of inpatients have demonstrated that buprenorphine is safe in combination with cocaine. Ongoing clinical trials suggest that buprenorphine reduces cocaine abuse as well as opiate abuse in patients who are dependent on both drugs according to the criteria in the third edition of the Diagnostic and Statistical Manual of Mental Disorders, revised.
Carbamazepine is an anticonvulsant drug used to treat seizure disorders. In an open trial, 200 to 800 mg of carbamazepine two to four times daily suppressed the craving for cocaine and decreased cocaine use, but carbamazepine was no more effective than placebo in several subsequent clinical trials.
Psychotherapy and Behavioral Therapy
A number of cognitive and behavioral therapies have been recommended to prevent relapse in patients with cocaine abuse or dependence. These include aversion therapy, network therapy, behavioral treatment, exposure to cocaine-related cues, contingency-based contracting, and cognitive therapy. Although no specific cognitive or behavioral treatment is uniquely effective in patients with cocaine abuse or dependence, the systematic application of behavioral concepts to drug-abuse treatment is relatively new. One 24-week behavioral study indicated that giving patients incentives (in the form of vouchers exchangeable for retail items) to provide cocaine-free urine specimens significantly improved compliance and outcomes. The effectiveness of behavior-related contingency management appears to be enhanced by having patients’ significant others participate in the treatment program. Therapeutic endeavors are likely to benefit when there is psychosocial support. In psychotherapy, mutual trust and confidence in the therapist are important ingredients of success. Having the patient take part in selecting the therapeutic regimen may also facilitate treatment.
New Medications — Promises and Problems
The prospects for developing more specific drugs to treat cocaine-related disorders should be greatly improved by advances in research on the neurobiology of cocaine and by the availability of preclinical models of drug abuse that are useful in evaluating medications. For example, clarifying the relative contributions of the various subtypes of dopamine receptors to the reinforcing effects of cocaine should lead to more specific pharmacologic interventions. Identifying the molecular structures of glycoprotein dopamine transporters and discovering unique ligands, such as the polyamine spermine, that bind to these transporters provide a promising approach to the development of medications. In addition, the development of specific antibodies for the catalyzed degradation and immune binding of cocaine may also contribute to more effective medications for the treatment of cocaine abuse. On the basis of these discoveries, it is reasonable to anticipate that molecular pharmacology will contribute to the successful development of specific cocaine antagonists. However, the effectiveness of such antagonists as compared with that of less toxic cocaine-like agonists remains to be determined.
Theoretically, a safe and effective antagonist would block the reinforcing properties of cocaine and decrease cocaine use. However, it is also possible that patients will use higher doses of cocaine in an effort to surmount the pharmacologic blockade. Empirically, the treatment of opiate abuse and dependence with antagonists has been disappointing. With few exceptions, the FDA-approved selective opioid antagonist naltrexone has not proved effective in treating opioid abuse, because of poor compliance. Patients usually discontinue antagonist therapy in the anticipation of resuming illicit-drug use. In the treatment of opiate abuse, an opiate agonist, methadone, and a partial agonist, buprenorphine, have been most effective.
Whatever drug is used to treat cocaine abuse and dependence, adequate evaluation will require using appropriate and realistic criteria for efficacy. However, defining efficacy has been an ongoing source of divisive polemics, and the prospects of a general consensus are remote. Traditionally, total abstinence has been the accepted standard for success, and any result less than abstinence has been regarded as a failure of treatment. Yet, in other areas of medicine it is recognized that most treatments have relative rather than complete efficacy. For example, a partial reduction of angina pectoris is very important to the patient, even though there may not be a total remission of symptoms. Similarly, in treating substance-abuse disorders, any reduction in drug use may create an opportunity for the patient to be more receptive to counseling and job-skills training and to learn constructive alternatives to drug acquisition and use. Moreover, any reduction in intravenous drug use, needle use, and needle sharing should lower the risk of HIV infection. The clinical reality is that most abusers of cocaine and other substances continue to use some drugs during maintenance treatment, and a 30 to 60 percent reduction in drug use should benefit society as well as the patient. Future evaluations of new medications for the treatment of cocaine abuse and dependence should include the use of quantitative urine screening to ascertain the degree of relative abstinence from cocaine.
As we have noted, the evaluation of any treatment to reduce cocaine abuse and dependence is complicated by the pervasiveness of polydrug abuse. Developing effective medications to treat polydrug abuse is increasingly important, because many persons do not abuse a single substance in order to achieve a salient effect. Paradoxically, drugs with differing or even opposite effects on mood and behavior may be used, for reasons that are poorly understood.
At present, no drug therapy is uniquely effective in treating cocaine abuse and dependence. A number of medications that were initially developed to treat other disorders have been used to treat the biobehavioral disorders often associated with cocaine abuse and dependence. Unfortunately, none are considered to be highly effective for either cocaine detoxification or the maintenance of abstinence. However, the limitations of the medications currently available should not lead investigators to give up their search, because medication is only one element of humane and comprehensive therapy for persons with drug dependence. There is compelling evidence that many cocaine abusers have had major psychological and psychosocial impairments that contributed to and may have been compounded by subsequent problems of drug dependence. These impairments include cognitive and learning disorders, interpersonal and social problems, and legal and financial difficulties. The recent report of the first randomized clinical trial in which the efficacy of the interaction between pharmacotherapy and psychotherapy was evaluated in the treatment of cocaine abuse stressed the “heterogeneity among cocaine abusers and the need to develop specialized treatment for clinically distinct subgroups.” It is axiomatic that treatment for persons with cocaine abuse or dependence should be selected on the basis of all biomedical and psychosocial factors associated with the patients’ illness.
Persons who abuse cocaine and also have antecedent or concurrent depressive disorders may benefit from treatment with an antidepressant drug, such as desipramine. It is unlikely, however, that patients dependent on cocaine will remain abstinent as a function of treatment with either antidepressant or anxiolytic agents. Cocaine-dependent persons who take that drug by inhalation (as crack cocaine) often have major psychosocial, cognitive, and legal problems and may require referral to an intensive program of psychosocial treatment. In contrast, persons who abuse or depend on cocaine and administer the drug intranasally (by insufflation) may enter remission after being referred to specialists in psychiatry or addiction medicine. The most seriously ill patients with cocaine abuse or dependence are those who use the drug intravenously, either by itself or with heroin or morphine, and are at high risk for myocardial infarction, stroke, and the acquired immunodeficiency syndrome. These patients need the most intensive medical and psychiatric observation and should be referred to programs that provide multimodal treatment for polydrug abuse and dependence. Developing safe and effective medications to treat cocaine abuse and dependence is an ongoing challenge. However, as is true of good medical practice generally, the treatment of drug abuse or dependence relies on a triad of compassion, psychosocial enrichment, and safe and effective pharmacotherapy.
Supported in part by grants (KO5-DA 00064, KO5-DA 00101, DA 06116, and DA 04059) from the National Institute on Drug Abuse, National Institutes of Health.
Jack H. Mendelson, M.D., and Nancy K. Mello, Ph.D.
From Harvard Medical School, Boston, and McLean Hospital, Belmont, Mass.
Address reprint requests to Dr. Mendelson at McLean Hospital, 115 Mill St., Belmont, MA 02178.
1. Musto DF. The American disease: origins of narcotic control. New Haven, Conn.: Yale University Press, 1973.
2. Mendelson JH, Mello NK. Cocaine and other commonly abused drugs. In: Isselbacher KJ, Braunwald E, Wilson JD, Martin JB, Fauci AS, Kasper DL, eds. Harrison’s principles of internal medicine. 13th ed. New York: McGraw-Hill, 1994:2429-33.
3. Gfroerer JC, Brodsky MD. Frequent cocaine users and their use of treatment. Am J Public Health 1993;83:1149-1154.
4. National Institute on Drug Abuse. Epidemiologic trends in drug abuse. Vol. 1. Highlights and executive summary. Washington, D.C.: Government Printing Office, 1995.
5. Pillai R, Nair BS, Watson RR. AIDS, drugs of abuse and the immune system: a complex immunotoxicological network. Arch Toxicol 1991;65:609-617.
6. Chaisson RE, Bacchetti P, Osmond D, Brodie B, Sande MA, Moss AR. Cocaine use and HIV infection in intravenous drug users in San Francisco. JAMA 1989;261:561-565.
7. Donahoe RM, Falek A. Neuroimmunomodulation by opiates and other drugs of abuse: relationship to HIV infection and AIDS. In: Bridge TP, Mirsky AF, Goodwin FK, eds. Psychological, neuropsychiatric, and substance abuse aspects of AIDS. Vol. 44 of Advances in biochemical psychopharmacy. New York: Raven Press, 1988:145-57.
8. Schoenbaum EE, Hartel D, Selwyn PA, et al. Risk factors for human immunodeficiency virus infection in intravenous drug users. N Engl J Med 1989;321:874-879.
9. Preliminary estimates from the 1993 National Household Survey on Drug Abuse. Advance report no. 7. Rockville, Md.: Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 1994:29.
10. Preliminary estimates from the Drug Abuse Warning Network (DAWN). Advance report No. 2. Rockville, Md.: Substance Abuse and Mental Health Services Administration, Office of Applied Studies, 1993.
11. Cregler LL, Mark H. Medical complications of cocaine abuse. N Engl J Med 1986;315:1495-1500.
12. Hollander JE. The management of cocaine-associated myocardial ischemia. N Engl J Med 1995;333:1267-1272.
13. Isner JM, Estes NAM III, Thompson PD, et al. Acute cardiac events temporally related to cocaine abuse. N Engl J Med 1986;315:1438-1443.
14. Moliterno DJ, Willard JE, Lange RA, et al. Coronary-artery vasoconstriction induced by cocaine, cigarette smoking, or both. N Engl J Med 1994;330:454-459.
15. Marzuk PM, Tardiff K, Leon AC, et al. Fatal injuries after cocaine use as a leading cause of death among young adults in New York City. N Engl J Med 1995;332:1753-1757.
16. Diagnostic and statistical manual of mental disorders. 4th ed.: DSM-IV. Washington, D.C.: American Psychiatric Association, 1994.
17. Dackis CA, Gold MS. New concepts in cocaine addiction: the dopamine depletion hypothesis. Neurosci Biobehav Rev 1985;9:469-477.
18. Gawin FH, Kleber HD. Abstinence symptomatology and psychiatric diagnosis in cocaine abusers: clinical observations. Arch Gen Psychiatry 1986;43:107-113.
19. Weddington WW, Brown BS, Haertzen CA, et al. Changes in mood, craving, and sleep during short-term abstinence reported by male cocaine addicts: a controlled, residential study. Arch Gen Psychiatry 1990;47:861-868.
20. Satel SL, Price LH, Palumbo JM, et al. Clinical phenomenology and neurobiology of cocaine abstinence: a prospective inpatient study. Am J Psychiatry 1991;148:1712-1716.
21. Mendelson JH, Mello NK. Biologic concomitants of alcoholism. N Engl J Med 1979;301:912-921.
22. Freud S. Cocaine papers. Byck R, ed. New York: Stonehill, 1974.
23. Mello NK. A behavioral analysis of the reinforcing properties of alcohol and other drugs in man. In: Kissin B, Begleiter H, eds. The pathogenesis of alcoholism, biological factors. Vol. 7. New York: Plenum Press, 1983:133-98.