Since the highly publicized death of American  basketball star, Len Bias, there has been a spate of claims  in the popular media and scientific literature concerning cocaine's extraordinary cardiotoxicity. These claims are  illustrated in the following quotations, one from a popular  Canadian paperback by a respected journalist and one  from a leading American science magazine:

...if [cocaine] was ever legalized you might as well open up a bunch of new cemeteries and funeral parlours, because you'll have a lot of bodies dropping...It causes all kinds of problems with your central nervous  system and it causes cardiac arrest. Your heart is a piece of meat and it can only take so much snorting and cranking up  before it cuts out on you. (Toronto cocaine dealer cited by Malarek, 1990, p. 124).  "I've never seen such an ability to turn off the vagus as with cocaine," Hull remarks. Indeed, patients with acute cocaine intoxication should therefore be treated "as if  they're having a heart attack — they're at  that high a risk of sudden death," he says  (Science, 1990).

Although such claims are made with authority and certitude, a review of the available scientific and medical data has led us to different conclusions. This article uses the available data. to assess, first, the risk of heart attack associated with moderate use of cocaine, second, the risk associated with heavy use, and, finally, the risk associated with "crack" and other forms of cocaine that are typically smoked. We conclude that all three risks have been seriously overstated in recent literature.

This article only considers the cardiac consequences of cocaine use. The larger issue of the lethality of cocaine, including that attributable to other physiological and behavioral mechanisms, is not systematically addressed. The fundamental social and policy implications of changing perspective on risks associated with cocaine have been considered elsewhere (Alexander, 1990).

Moderate Cocaine Use

This section on moderate cocaine use may seem pointless to some readers, since it is widely believed that moderate use is only a way station on the road to addiction. However, an earlier review (Erickson and Alexander, 1989) showed that the majority of those who have used cocaine either give it up quickly or continue to use it only occasionally without addiction. Most of the remainder use it with moderation and control any inclinations to exces-sive use that arise. Only a small minority ever require treatment for addiction. Therefore, moderate use is not impossible but, in fact, more typical than heavy use.

Moderate cocaine use has been studied through surveys of illegal cocaine users and through the legal administration of moderate doses as part of medical treatment and scientific experiments. Together, these sources of data provide strong evidence that moderate use of cocaine does not have more severe cardiotoxic consequences than other, socially acceptable forms of stimulation.

Survey Studies of Illicit Users

Survey data indicate that moderate, illicit users have few cardiac problems and, in fact, few serious medical problems of any sort. Less than 20 per cent of British Columbians who reported using cocaine more than once in their lives stated that they had experienced any negative side effects (Coordinated Law Enforcement Unit, 1987). None of a sample of 111 mostly moderate cocaine users interviewed in Ontario included chest pains or other heart symptoms on list of their reactions to cocaine, although a number mentioned other medical complications, primarily nasal irritation and inability to relax or sleep (Erickson, et al., 1987). Murphy, Reinarman, and Waldorf (1989) reported no complaints of heart problems among an American group of 21 mostly moderate cocaine users over an 11 year period, although several people in the group reported chronic nasal irritation. Mugford and Cohen (1989) studied 73 mostly moderate cocaine users in Australia with a mean age of 27. Of this group, 7% had been treated for high blood pressure at some time in the previous 5 years and 1% had been treated for heart disease. Among 160 "nondeviant" cocaine users in Amsterdam with a mean age of 30 years, Cohen (1989) found that 21% had experienced high blood pressure at some time in their lives, 11% reported having it the previous year and 6% believed their high blood pressure was an effect of cocaine use. About 4% of the subjects had suffered heart disease during their lifetimes, less than 1% in the year preceding the interview and none attributed it to cocaine.

Although the large majority of people who responded to these survey reports are moderate cocaine users, a minority were not, and the complaints concerning blood pressure tended to be strongly correlated with level of cocaine use (Cohen, 1989).

Medical Administration of Moderate Doses

Like survey studies, studies of medical administration of cocaine provide no indication that moderate doses cause cardiotoxicity.

Cocaine is routinely applied to the nasal membranes as a local anesthetic in doses of 200 mg or more in nasal surgery and not uncommonly in doses larger than 400 mg (Haddad, 1983; Moore et al., 1986; Chui, Brecht, DasGupta, and Mhoon, 1986; Gordon, 1987). These doses are comparable to those typically reported by recreational users. Gradman (1988) described 100 mg as "a usual recreational dose" and Erickson et al. (1987) reported that about 36% of those Canadian cocaine users who could estimate their use "per occasion" (where an "occasion" might include more than one dose) used less than 250 mg and 40% used 250-500 mg. The peak blood levels of cocaine following intranasal application as a local anesthetic (reported as 0.12-0.47 mg/L of blood by Van Dyke et al., 1976) are comparable to those found following intranasal doses that produce a "high" in experienced users (reported as 0.10-0.53 mg/L by Javaid et al., 1978).

A survey of plastic surgeons who had together performed 93,004 operations involving cocaine anesthetic revealed 5 deaths (or 5/1000 of 1% of the patients) and 34 severe but non-fatal reactions (4/100 of 1% of the patients) (Feehan and Mancusi-Ungaro, 1976). These data provide no information concerning the number of these deaths and  serious reactions that involved cardiotoxicity.

Chui, Brecht, DasGupta, and Mhoon (1986) reported a single non-fatal myocardial infarction in a woman who was topically anesthetized with cocaine for nasal surgery. The patient recovered well enough that the nasal surgery could be completed and no further complications were reported. She reported having had chest pains on the two times she had previously used cocaine recreationally. These authors pointed out that,

"Cocaine...has been widely used by physicians performing otolaryngologic operations since the end of the 19th century...To our knowledge, this is the first documented report of acute myocardial infarction with clinical application of cocaine in a hospitalized patient during topical anesthesia" (p. 988)

Cocaine is one of three active ingredients in "TAC", a preparation containing tetracaine, adrenaline, and cocaine that is widely used as a local anesthetic for injured children (Pierluisi and Turndrup, 1989; Tipton, DeWitt, and Eisenstein, 1989). The number of reported adverse effects is very low (Bonadio and Wagner, 1988).

Addicted patients have been chronically maintained on medically prescribed cocaine in Britain for decades (Trebach, 1982, pp. 115, 148). A recent case-study described an English woman (not an addict) who legally applied cocaine to her nasal membranes several times a day for the past 55 years. She is currently prescribed 3,150 mg of cocaine per week. Although this use strains the limits of the term "moderate", it is instructive here. At age 80, "She appears to have suffered no ill effects from the prolonged use of cocaine in physical, psychological or social terms" (Brown and Middlefell, 1989, p. 946).

Experimental Administration of Cocaine to Human Beings

Doses of cocaine at the upper limit of those that could be considered moderate have proved safe in experimental studies designed to measure various physiological reactions to the drug. For example, Fischman and Schuster (1980) injected varying doses up to 200 mg of cocaine per hour intravenously into volunteers who were recreational cocaine users. These injections were repeated on several days. Levels of cocaine in blood plasma up to 1.2 mg/L were recorded. These blood levels exceed the peak plasma levels reported in experienced cocaine snorters, injectors, and smokers as tabulated by Verebey and Gold (1988). Out of 50 subjects in Fischman and Schuster's report, only one had an adverse reaction, an "intense anxiety attack associated with muscle contraction" that lasted about 45 minutes. Other investigators (Barnett, Hawks, and Resnick, 1981; Ambre et al., 1988) have also shown that experimental subjects tolerate injected doses of cocaine that produce blood levels up to and in excess of 1.0 mg/L.

Critique

It is possible to question the relevance of the above data. The survey data can be challenged on the grounds that subjects who are willing to respond to survey questionnaires are not representative of the people who use cocaine. Similarly, the medical and laboratory studies can be questioned on the grounds that pure cocaine administered in unstimulating hospital settings may not have the same effects as adulterated cocaine administered on the street.

Whereas these objections have merit, they do not negate the essential relevance of the data. The survey data show that moderate use of cocaine does not induce heart problems in the kind of people who respond to survey questions, i.e., the majority of the population in North America. If moderate cocaine use should prove to have a greater cardiotoxicity in fringe populations, this does not so much indict the drug as the conditions which make disadvantaged people more vulnerable to cardiac damage from amounts of a stimulant that are harmless to most of the population.

Similarly, the medical and laboratory data show that moderate doses of pure cocaine do not produce cardiotoxic effects. If street drugs do prove to induce cardiotcodcity, then the harmfulness does not inhere in cocaine itself It probably inheres in the social milieu that fosters impure, black market drugs and unsafe methods of administration. The issues of cocaine cardiotoxicity in fringe populations and the effects of cocaine adulterants will be examined later in this article.

Whereas the research we have cited is far from ideal, it is better controlled than the case studies that are used to argue that moderate use of cocaine does have cardiotoxic effects. Generalizations that have been made from these studies, particularly in the popular media, do not stand up well to normal scientific skepticism.

For example, an article in the Journal of the Canadian Medical Association (Rollingher, Belzberg, and Macdonald, 1986) is frequently cited in medical literature and has attracted media attention in Canada (see Gifford-Jones, 1986). The article reports that one 24-year-old man suffered a serious heart attack that required several days hospitalization after a single dose of street cocaine, while his friends, who took similar doses, only experienced euphoria. The young man injected an unknown amount of street cocaine dissolved in tap-water directly into a vein.

Interpreting this study as evidence that cocaine is likely to produce heart attacks in moderate users requires ignoring several critical issues. For example, did this  patient really purchase cocaine? No chemical analysis of the drug of the man's blood or urine was undertaken. A Los Angeles study found that only 58.3 per cent of street "cocaine" samples contained cocaine alone. About 19.2 per cent contained no cocaine whatsoever. (Klatt et al., 1986). A review of the literature on street cocaine found an average purity of 40%. The adulterants that comprised the other 60% of the "cocaine" were classified into 6 groups: local anesthetics, sugars, stimulants, toxins (including strychnine), inert compounds, and other substances (Shannon, 1988).

A second critical issue concerns whether or not this man was really a moderate user. He "denied any previous abuse of illicit drugs" (Rollingher, Belzberg, and Macdonald, 1986, 45), but such an unconfirmed denial in this context is the weakest of evidence. Intravenous injection of cocaine is almost never chosen for a first illicit drug experience in Canada. For example, in British Columbia, 99.4 per cent of those who have used cocaine have used cannabis as well. Less than 2/10 of 1 per cent of cocaine users in British Columbia reported injection as their first method of using cocaine (Coordinated Law Enforcement Unit, 1987, 36, 49). A young man who had come to the attention of the authorities in the midst of a "War on Drugs" would have strong reasons to minimize the extent of his illegal drug use. Thus, this widely publicized story provides no firm evidence that moderate use cocaine can cause heart attacks.

There are similar methodological problems with the other North American case studies that are used to support the view that moderate cocaine use causes heart attacks and other cardiotoxicity in moderate users (Kossowsky and Lyon, 1984; Schachne, Roberts, and Thompson, 1984; Howard, Hueter, and Davis, 1985; Pasternack, Colvin, and Bauman, 1985; Isner et al., 1986; Edwards and Rubin, 1987; Haines and Sexter, 1987; Wehbie, Vidaillet, Navetta, and Peter, 1987; Ascher, Stauffer, and Gaasch, 1988; Wang et al., 1988; Chokshi, Moore, Pandian, and Isner 1989; Gadaleta, Hall, and Nelson, 1989). None of these clinical studies provide solid evidence that the cardiotoxicity was caused by cocaine rather than some adulterant, by some other drug the patient had administered, or by some other aspect of the patienes lifestyle. None provides solid evidence that any of the patients were only moderate users. Although some of these studies suggest that some of their patients were only  moderate users, most of the case information suggests that they were heavy users of cocaine and other drugs. We will again consider these studies under the heading of heavy cocaine use.

Because cocaine is a heart stimulant, it is certainly possible that even moderate doses can increase the likelihood of heart attacks in people with high blood pressure or preexisting heart damage and some case studies are consistent with this interpretation (Barth, Bray, and Roberts, 1986; Grannis, Bryant, Cafferatti, and Turner, 1988). However, this supposition would merely put cocaine in the same class as other activities that accelerate heart rate and blood pressure. Caffeine, alcohol, tobacco, sports, gambling, and sex may occasionally precipitate a heart attack in a vulnerable person, but cannot be considered extraordinary hazards.

There may well be isolated cases in which a person has an atypical reaction that is specific to cocaine and is independent of preexisting pathology, as in the case reported by Chui et al. (1986), but rare atypical reactions are characteristic of drugs in general. A familiar example is the occasionally fatal hypersensitive reaction to aspirin-like drugs (Flower, Moncada, and Vane, 1985, p. 679).

On all the grounds we have been able to review, it seems safe to conclude that moderate use of cocaine produces no risk of cardiotoxicity beyond that associated with any stimulating activity.

Paradoxically, one study of experimental animals demonstrated that a single dose of cocaine greatly reduced the risk of heart attack precipitated by an overdose of tetracaine 1-4 weeks later (Antelman, De Giovanni;and Kocan, 1989). The authors of this study explained their unexpected result by making an analogy with physical exercise. Exercise increases heart rate and blood pressure, which can produce a heart attack in people who are in tenuous health. In most cases, however, the heart attack does not occur and the person becomes more resistant to later heart attacks. In the authors' words, "These data suggest that when appropriately timed, strong sympathomimetic stimulation — whether generated by an environmental stressor or a drug — can provide long-lasting protection against the sudden cardiac death potential of local anesthetics..." (201). Although it would be premature to conclude that moderate cocaine use prevents heart attacks in human beings, this possibility warrants investigation.

Fearsome claims, such as the claim that all people who use cocaine should be treated for heart attacks whether they have one or not, or the now common claim that "there is no safe dose of cocaine" contradict the existing medical and scientific evidence.

Heavy Cocaine Use

At the other end of the spectrum of cocaine use, there are causal links between heavy, prolonged cocaine use and heart failure. However, the links are not nearly so clear cut as those proclaimed in the popular media. In this section, we first consider deaths caused by cocaine over-dose, i.e., a very large single dose of the drug. We then consider "cocaine-related" deaths, i.e., deaths in which  there is evidence of prior cocaine use but which are not necessarily due to overdose. These types of death are treated separately because the causes of heart failure appear to be different.

Cocaine Overdose

There is no doubt that overdoses of cocaine can cause illness and death, although it is questionable whether or not such deaths should be considered as cardiac events.

Cocaine overdose, which is also described as "cocaine poisoning" in the medical literature, has been well documented in human beings and experimental animals since the 19th century (Finkle Euid McClosky, 1978; Smart and Anglin, 1987). Supplementing the classical literature in this area, cocaine poisoning has been recently documented in human "body packers" who are in custody when condoms containing large amounts of cocaine burst within their bodies and in traffickers who sometimes swallow their drugs at the time of their arrest (Introma and Smialek, 1989).

Cocaine overdose victims generally become excited and confused shortly after administration of a large dose. They subsequently undergo convulsions, depression, coma, and in severe cases, death which is often attributed to "respiratory depression" or "cardiovascular collapse".

Recent research indicates that the "respiratory depression" or "cardiac collapse" which result from severe cocaine overdose do not entail myocardial infarction or other traumatic cardiac event with any appreciable frequency (Gradman, 1988; Tardiff, Gross, Wu, Stajic, and Millman, 1989). Moreover, heart failure is probably not a result of direct action of cocaine on the heart. Instead, it appears to be a secondary consequence of overtaxing the heart with convulsions, hyperthermia, and acidosis (Gradman, 1988). Deaths from cocaine overdose can be prevented by controlling convulsions, hyperthermia, and acidosis in the patient (Catravas and Waters, 1981; Jonsson, O'Meara, and Young, 1983; Derlet and Albertson, 1989).

"Cocaine-related" Deaths

Deaths in which there is evidence of prior cocaine use are frequently referred to as "cocaine-related" deaths. It has become commonplace to assume that many of the decedents in such cases are only moderate users and that many of the deaths are caused by cocaine-induced cardiotoxicity. However, the medical data on cocaine-related deaths shows that both of these assumptions are wrong. Few cocaine-related deaths involve moderate users and the majority of these deaths are probably not caused by cocaine at all, and therefore not by cocaine-induced cardiotoxicity.

Most Cocaine-Related Deaths Involve Heavy Users. The evidence cited above on moderate use makes it seem unlikely that any appreciable number of moderate users of cocaine would be found among the victims of cocaine-related deaths. Wong and Alexander (in press) tested this directly by analyzing the coroner's files on 29 cocaine-related deaths in British Columbia. The investigation was prompted by a frontpage newspaper article implying that many of the decedents were merely moderate cocaine users who had paid the ultimate penalty for their moment of self-indulgence with cocaine.

Analysis of the coroner's files revealed that 26 of the decedents were known to have been heavy or addictive users of illicit drugs and/or alcohol. Twenty three were known to have been intravenous users of cocaine, heroin, and other drugs. Thirteen showed lung damage that is typical of chronic intravenous drug users (junkie-lung"). Moreover, almost all of the decedents appeared to have lived socially deviant lifestyles. There was strong evidence of criminality, prostitution, social isolation, dysfunctional families, and psychiatric problems in the 22 cases for which the records provided lifestyle information. We concluded from our analysis that at most 3 of the decedents could have been only moderate users of cocaine and other illicit drugs.

In the following sections, we explore the possible causes of cocaine-related deaths. We first show that most cocaine-related deaths are not caused by cocaine overdose or by myocardial infarction. Next, we show that it is unsafe to conclude that cocaine is the cause of cocaine-related deaths at all. We conclude that the most likely cause of the majority of cocaine-related deaths is a chronic cumulative deterioration to which cocaine is only one of several contributing factors.

Possible Causes of Cocaine-Related Deaths: Overdose. As shown above, heart failure can result from cocaine overdose, and so an overdose death can arguably be identified as a cocaine-induced cardiac event. However, cocaine overdose is a relatively uncommon cause of cocaine-related death. Wong and Alexander (in press) found no documentation that more than one of 29 cocaine-related fatalities in British Columbia fit the syndrome of cocaine overdose described above. In 17 of the 29 cases, there was no demonstration of blood levels of cocaine' greater than 1 mg,/L. In several cases there was a interval of hours between the last dose of cocaine and the onset of the terminal distress, and the convulsions that are characteristic of cocaine overdose were observed in at most one case (In several cases there was no witness at the time of death who could have observed convulsions). Pulmonary edema was a conspicuous feature of these deaths, whereas it is not a feature that is usually associated with true cocaine overdose deaths. Tardiff, Gross, Wu, Stajic, and Millman  (1989) reported that only 4% of 935 cocaine-related deaths in New York city were due to cocaine overdose and another 12% were due to overdose of a combination of heroin and cocaine.

Possible Causes of Cocaine-Related Deaths: Heart Attack. Heart attack is a relatively uncommon cause of cocaine-related deaths and there is serious doubt about whether cocaine is actually the cause of the heart attacks that do occur. Tardiff et al. (1989) reported that only 1.1% of the 935 New York cocaine-related deaths they studied were due to acute myocardial infarctions or aortic rupture. Other authors (Sternberg et al. 1989, 522) have confirmed that the number of acute cardiac events associated with cocaine use is small compared with the millions of North Americans who have used cocaine and the thousands of heavy users. In a 1988 review Gradman noted that "Over the past five years, approximately 20 cases of cocaine-associated myocardial infarction have been reported" (Gradman, 1988, p. 141).

Although the authors of many articles on heart attacks continue to assert that cocaine is the cause of the cardiac events they have observed, others point out the evidence for such a causal inference is equivocal. For example, Haines and Sexter (1987) state "no causal relationship between cocaine use and myocardial infarction has been shown" (1326). Devenyi and McDonough (1988), Gradman (1988), and Sternbern et al. (1989) have pointed out the weakness of attributing heart attacks to cocaine when so many other factors in these heart patients could be responsible. In a thorough review, Bates (1988) states, "It is not possible to unequivocally state that cocaine abuse causes myocardial infarction" (441).

Causes of Other Cocaine-Related Deaths. If the majority of non-violent cocaine-related deaths are not caused by overdose or by cocaine-induced myocardial infarction, what is the cause?

We believe that the most parsimonious conclusion from the available evidence is that the majority of these deaths are caused by generalized deterioration that results from accumulating damage to heart, liver, lungs, brain, and other vital organs in people with unhealthy, deviant life styles. Cocaine would be only one of many causes of such deaths.

Wong and Alexander (in press) argued that most of the cocaine-related deaths they studied in British Columbia were best described as the result of a long period of deterioration resulting from a socially deviant lifestyle and chronic illness. Most of the decedents were long-time users of intravenous heroin and other illicit drugs, many had a history of alcoholism, cigarette dependency, crime, prostitution, and at least one was HIV-positive. Postmortem examinations showed that many of the decedents had "junkie lung", a cumulative lung deterioration found in long-term IV drug users and thought to be produced by impurities in street drugs (Haller, 1988; Shannon, 1988). Several had coronary artery pathology, the most common form being arteriosclerotic coronary artery disease. Others had other kinds of heart pathology. Several had liver pathologies common to chronic intravenous drug users including hepatitis and livers laden with macrophages, infiltrates and lymphocytes. In a study of deaths attrib-uted to drug abuse in Europe, Ingold (1986) reported:

"it is likely that such acute intoxication is  not the sole or principle cause of death in the cases examined, since the impaired general condition and the lesions (for example those affecting the liver and lungs) play such an important role that they alone may be the cause of death" (pp. 87-88).

As stated above, Tardiff et al (1989) found that a total of 17.1% of cocaine related deaths could be attributed either to overdose of cocaine or overdose of cocaine plus heroin or to acute cardiac events. Fifty-three per cent of the deaths were attributed to homicide, accidents, or suicide. Of the remaining deaths, a number appeared to fit the cumulative deterioration syndrome described by Wong and Alexander (See Tardiff et al.'s cases 1, 3, 4-6, 11, 17, 19.), although Tardiff et al. did not elaborate on this possibility. The cumulative deterioration hypothesis could be checked in future research by comparing cocaine-related deaths with heroin and alcohol related deaths or, even better, with the deaths of deviant people who did not use illicit drugs. Such controlled research should reveal whether chronic cocaine use in the context of a deviant life style increases the risk of heart attack.

It is possible that many non-violent cocaine-related deaths are due to cocaine-induced heart failure without acute signs (e.g., heart failure induced by transient coronary vasospasm) or to accumulating heart damage that is specifically caused by cocaine, there is no substantial evidence of this.

A great deal of recent research presupposes that cocaine-induced cardiotoxicity causes cocaine-related deaths and seeks to identify the mechanisms by which this occurs. These studies purport to demonstrate many different mechanisms including transient coronary vasospasm, narrowing of the coronary arteries through cocaine-induced atherosclerosis (Dressler, Malekzadeh and Roberts, 1990), cocaine-induced hypertrophy of pulmonary arteries (Murray, Snialek, Galle, and Albin, 1990), cocaine-induced cardiac and vascular toxicity (Wilkerson, 1988; Peng, French, and Pelikan, (1989), cocaine-induced inflammation of heart muscle walls (Virmani, Robinowitz, Smialek, and Smyth, 1988), cocaine-induced contraction  band necrosis (Tazelaar, Karch, Stephens, and Billingham, 1987; Virmani et al., 1988), and heart damage due to oxygen deficits during withdrawal from cocaine (Nadamanee, Gorelick, Josephson, Ryan, Wilkins, Robertson, Mody, and Intarachot, 1989). Most recently, focus seems to have shifted to direct depressive action of cocaine on heart muscle caused by cocaine-induced impairment of cardiac conduction (Kabas, Blanchard, Matsuyama, Long, Hoffman, Ellinwood, Smith, and Strauss, 1990; Crumb and Clarkson, 1990).

None of these studies are compelling. Because most are based on medical patients or autopsies where experimental control of unhealthy habits other than cocaine use is impossible, they tell little about the propensity of heavy and prolonged use of cocaine per se to cause the heart damage that has been found. They also tell little about the likelihood that the damage that has been found causes heart attacks.

Sudden Death

Mittleman and Wetli (1987) have advanced an argument that is essentially the mirror image of that made above. They contend that cocaine's involvement in deaths from heart failure and other diseases is greatly understated in the existing literature. They purport to have shown that 20-37% of apparently natural sudden deaths in Miami, Florida could have been caused by cocaine. However, their data suffer from all the problems of missing experimental control identified in the studies reviewed above, and cannot be taken seriously without further corroboration.

Is "Crack" a Special Case?

Although the popular media and some professionals claim that crack, free-base, and coca paste and other forms of alkaloidal cocaine used for smoking are far more damaging than cocaine hydrochloride, such claims are not supported by strong evidence. Because we have found no studies that specifically purport to show a difference in cardiotoxicity between cocaine hydrochloride and crack, we have relied here on more general reports.

Some of the most dramatic claims that crack and other forms of smokable cocaine are extraordinarily lethal lack any empirical basis. For example, USA Today attributed 563 deaths to cocaine and crack in the first six months of 1986. However, Trebach in his 1987 book, The Great Drug War wrote:

I have searched. My assistants have  searched. We have gone through many  government reports. We have quizzed government statistical experts. We have yet to discover one death in which the presence of crack was a confirmed factor...(p. 12)
All the claims about the great rise in crack deaths for the first months of 1986, such as that reported in USA Today and other major media voices, were false. Even more significant, none of the leading government drug-abuse officials, who know the claims for 1986 to be false, felt any responsibility to tell the public the truth. (p. 11)

To our knowledge, Trebach's claims have never been refuted, although they have been almost universally ignored. Two years after Trebach wrote these passages, Ostrowski (1989) reached a similar conclusion in a policy analysis written for the Washington based CATO Foundation:

In recent years, the cocaine derivative  crack has become the drug of the moment. In spite of the fact that crack is a more pure and potent form of cocaine, there is little evidence that its use has increased cocaine fatalities. The author was unable to obtain statistical information about crack fatalities in phone calls to the National Institute on Drug Abuse, the 1-800 COCAINE hotline, or the New York State Division of Substance Abuse Services. (p. 49).

There are some statements in the current medical literature that various forms of smokable cocaine are significantly more harmful than cocaine hydrochloride (Siegel, 1985; Washton, Gold, and Pottash, 1986; Honer, Gewirtz, and Turey, 1987; Isaacs, Martin, and Willoughby, 1987; Levine et al., 1987; Manschreck, Allen, and Neville, 1987; Mody et al., 1988), but most of these claims lack documentation apart from the well-established fact that smokable cocaine reaches the bloodstream faster than orally or nasally administered cocaine (but not faster than injected cocaine or smoked marijuana or tobacco). Surprisingly however, there has been only a little overt criticism of them (Levine et al., 1987). The authors of these articles seem to have taken license from the prevailing drug war context to state dramatic conclusions with little or no empirical basis.

The few data that have been collected comparing the relative dangers of cocaine hydrochloride and smokable cocaine do not seem to justify the conclusions that are drawn from it. Honer, Gewirtz, and Turey (1987) found greater frequencies of various adverse consequences in users of crack, compared to users of cocaine hydrochloride. However, these differences could only reflect the fact that different types of people use the two drugs — perhaps the relatively poor people buy crack who are already less healthy than more affluent people who can afford to spend  larger amounts of cash for cocaine hydrochloride. This obvious possibility is supported by the fact that Honer et. al's data shows a considerable difference between the users of crack and freebase cocaine. These are both forms of alkaloidal cocaine manufactured from cocaine hydrochloride and both are usually administered by smoking, however they are typically used by people of different socioeconomic levels and, presumably, different standards of health. This same critique applies to a more recent study by Ringwalt and Palmer (1989).

Smoking cocaine appears to cause respiratory damage which may ultimately overtax and damage the heart, but studies of respiratory damage so far do not establish that smoking cocaine is more dangerous than smoking anything else. Many studies report that prolonged exposure to cocaine smoke produces at least temporary respiratory problems (Itkonen, Scholl, and Glassroth, 1984; Kissner et al., 1987; Patel, Dutta, and Schonfeld, 1987; Tashkin et al., 1987; Weiss et al., 1987; Murray et al., 1988; Rebhun 1988; Eurman, Potash, Eyler, Paganussi, and Beute, 1989). Others report that the forced deep breathing and mouth-to-mouth smoke exchange sometimes carried out by free-base smokers may result in a painful chest problem called pneumomediastinum (Salzman, Kahn, and Emory, 1987; Wiener and Putnam, 1987). One study, based on a single case report, suggests that cocaine smoking may cause pneumomediastinum by direct action of cocaine, i.e., without forced deep breathing (Cristou, Turnbull, and Cline, 1990). Together, these studies only establish that inhalation of hot smoke may lead indirectly to cardiac damage, as has been shown with tobacco (Benowitz, 1988).

We have been able to find no studies showing that cocaine smoke is appreciably more harmful to the cardiopulmonary system than other kinds of smoke that people may deliberately or inadvertently inhale. A study by Tashkin, et al. (1987) did find some kinds of acute lung distress in the higher of their two levels of cocaine smokers that were not found in tobacco smokers. On the other hand, these authors found no difference between tobacco and cocaine smokers in symptoms of chronic lung damage, no indications of greater lung damage in the lighter of the two groups of cocaine smokers compared to the tobacco smokers, and found some kinds of damage in tobacco smokers that were not found in cocaine smokers. Even in this relatively well-controlled study, there was no way to control for the many adulterants that are known to be present in smokable cocaine and could well be the cause of lung damage that does occur. The possible role of adulterants as a cause of the respiratory effects of smoking cocaine has been stressed in a recent review by Ettinger and Albin (1989).

Summary: Cocaine and Heart Attacks

We have found clear evidence that moderate use of cocaine does not cause any significant risk of heart attacks, especially when cocaine is viewed in perspective as one of many stimulating drugs and activities that people may use. We cannot reach a certain conclusion concerning the cardiac effects of heavy and prolonged use of cocaine, but we believe that the best guesses from the available literature are that 1) cocaine overdose deaths involve heart failure that is secondary to more global effects including seizures, 2) most cocaine-related deaths involve heavy users, 3) most non-violent cocaine deaths do not appear to be caused by cocaine per se, but appear to be the result of cumulative deterioration of many organs resulting from unhealthy lifestyles in which cocaine is just one of many contributing factors.

In the face of these conclusions, we suggest that the officially promulgated and universally accepted belief that even moderate use of cocaine is a significant cause of heart attacks may reflect the panicky reasoning of a society in the midst of drug war, rather than a scientific or dispassionate outlook.

B.K Alexander & L.S. Wong are with Simon Fraser University, Burnaby, B.C., V5A 1S6 Canada. (604) 291- 3354.

References

Alexander, B.K. (1990). Peaceful Measures: Canada's Way Out of the War on Drugs. Toronto: University of Toronto Press.

Ambre, J., Ruo, T.I., Nelson, J., and Belknapp, S. (1988). Urinary Excretion of Cocaine, Benzoylecgonine, and Ecgonine Methyl Ester in Humans. Journal of Analytical Toxicolgy. 12:301-306.

Antelman, S.M., DeGiovanni, L.A. and Kocan, D. (1989). A Single Exposure to Cocaine or Immobilization Stress Provides Extremely Long-Lasting, Selective Protection Against Sudden Cardiac Death from Tetracaine. Life Sciences. 44: 201-207.

Ascher, E., Stauffer, J.C.E. and Gaasch, W.H. (1988). Coronary Artery Spasm, Cardiac Arrest, Transient Electrocardiographic Q Waves and Stunned Myocardium in Cocaine-associated Acute Myocardial Infarction. American Journal of Cardiology. 61:939-941.

Barnett, G., Hawks, R., and Resnick, R. (1981). Cocaine Pharmacokinetics in Humans. Journal of Ethnopharmacology. 3:353-366.

Barth, C.W.III, Bray, M. and Roberts, W.C. (1986). Rupture of the Ascending Aorta during Cocaine Intoxication. American Journal of Cardiology. 57:496.

Bates, C.K. (1988). Medical Risks of Cocaine Use. Western Journal of Medicine. 148:440-444.

Benowitz, N.L. (1988). Drug Therapy: Pharmacologic Aspects of Cigarette Smoking and Nicotine Addiction. New England Journal of Medicine. 319:1318-1330.

Bonadio, W.A. and Wagner, V. (1988). Efficacy of TAC topical anesthetic for repair of pediatric lacerations. American Journal of Disease of Children. 142:203-205.

Brown, R. Etnd Middlefell, R. (1989). Fifty-five Years of Cocaine Dependence. British Journal of Addiction. 84:946.

Catravas, J. and Waters, I. (1981). Acute Cocaine Intoxica-tion in the Conscious Dog: Studies on the Mechanism of Lethality. The Journal of Pharmacology: Exper. Ther. 217:350-356.

Chokshi, S.K, Moore, R., Pandian, N.G. and Isner, J.M. (1989). Annals of Internal Medicine. 111:1039-1040.

Chui, Y.C., Brecht, K, DasGupta, D.S., and Mhoon, E. (1986). Myocardial Infarction with Topical Cocaine Anesthesia for Nasal Surgery. Archives of Otolaryngology-Head and Neck Surgery. 112:988-990.

Coordinated Law Enforcement Unit (1987). Final Report: Cocaine: Demand Reduction Strategies. Victoria, BC: Ministry of the Attorney General.

Cohen, P. (1989). Cocaine Use in Amsterdam in Non Deviant Subcultures. Amsterdam: University of Amsterdam, Institute for Social Geography.

Community Epidemiology Work Group. (1987). Patterns and Trends of Drug Abuse in the United States and Europe. Proceedings, June 1987. Rockville, MD: National Instit9te on Drug Abuse.

Cristou, T., Turnbull, T., and Cline, D. (1990). Cardiopul-monary abnormalities after smoking cocaine. Southern Medical Journal. 83:335-338.

Derlet, R.W. and Albertson, T.E. (1989). Diazepam in the Prevention of Seizures and Death in Cocaine-intoxicated Rats. Annals of Emergency Medicine. 18:542-546.

Devenyi, P. and McDonough, MA (1988). Cocaine Abuse and Endocarditis. Annals of Internal Medicine. 109:82.

Dressler, F.A., Malekzadeh, S. and Roberts, W.C. (1990). Quantitative Analysis of Amounts of Coronrary Arterial Narrowing in pocaine Addicts. American Journal of Cardiology. 65:303-308.

Edwards, J. and Rubin, R.N. (1987). Aortic Dissection and Cocaine Abuse. Annals of Internal Medicine. 107:779-780.

Erickson, P.G. and Alexander, B.K (1989). Cocaine and addictive liability. Social Pharmacology. 3: 249-270.

Erickson, P.G., Adalf, E.M., Murray, G.F., and Smart, R.G. (1987). The Steel Drug: Cocaine in Perspective. Lexington, MA: D.C. Health.

Ettinger, N.A., Albin, R.J. (1989). A Review of the Respira-tory Effects of Smoking Cocaine. The American Journal of Medicine. 87:664-668.

Eurman, D.W., Potash, H.I., Eyler, W.R., Paganussi, P.J., Buete, G.H. (1989). Chest Pain and Dyspnea Related to "Crack" Cocaine Smoking: Value of Chest Radiography. Radiology. 172:459-462.

Feehan, H.F., Mancusi-Ungaro, A. (1976). The Use of Cocaine as a Topical Anesthetic in Nasal Surgery: A Survey Report. Plastic and Reconstructive Surgery. 57:62- 65.

Finkle, B.S. and McCloskey, KL. (1978). The Forensic Toxicology of Cocaine (1971-1976). Journal of Forensic Sciences. 23:173-189.

Fischman, M.W. and Schuster, C.R. (1980). Experimental Investigations of the Actions of Cocaine in Humans. In F.R. Jeri, ed., Cocaine 1980. Lima, Peru: Pacific Press.

Flower, R.J., Moncada, S. and Vane, J.R. (1985). Analgesic-Antipyretics and Antiinflammitory Agents: Drugs Employed in the Treatment of Gout. In A.G. Gilman et al.,eds, 1985. Goodman and Gilman's The Pharmacological Basis of Therapeutics.

Gadaleta, D., Hall, M.H. and Nelson, R.L. (1989). Cocaine-induced Aortic Dissection. CHEST. 96:1203-1204.

Gifford-Jones, W. (1986). The Doctor Game: Even a Small Dose of Cocaine Can Trigger a Heart Attack. Globe and Mail. 30 December.

Gordon, B.R. (1987). Topical Cocaine Nasal Anesthesia. Archives of Otolaryngology-Head and Neck Surgery. 113:211.

Gradman, A.H. (1988). Cardiac Effects of Cocaine: A Review. Yale Journal of Biology and Medicine. 61:137-148.

Grannis, F.W., Bryant, M.D., Cafferatti, J.D. and Turner, A.F. Acute Aortic Dissection Associated With Cocaine Abuse. Clinical Cardiology. 11:572-574.

Haddad, L.M. (1983). Cocaine. In L.M. Haddad and J.F. Winchester, eds, Clinical Management of Poisoning and Drug Overdose. Philadelphia: W.B. Saunders

Haines, J.D. and Sexter, S. (1987). Acute Myocardial Infarction Associated with Cocaine Abuse. Southern Medical Journal. 80:1326-1327.

Haller, P.R. 1988. Infections in Intravenous Drug Abusers. Postgraduate Medicine. 83(4):95-114.

Honer, W.G., Gewirtz, G. and Turey, M. (1987). Psychosis and Violence in Cocaine Smokers. Lancet 2 (8556):451.

Howard, R.E., Hueter, D.C. and Davis G.J. (1985). Acute Myocardial Infarction following Cocaine Abuse in a Young Woman with Normal Coronary Arteries. Journal of Ameri-can Medical Association. 254:95-96.

Ingold, F.R. (1986). Study of Deaths Related to Drug Abuse in France and Europe. Bulletin on Narcotics. 38:81- 89.

Introna, F. and Skmialek, J.E. (1989). The `Mini-Packer' Syndrome: Fatal Ingestion of Drug Containers in Balti-more, Maryland. American Journal of Forensic Medicine and Pathology. 10: 21-24.

Isaacs, S.O., Martin, P. and Willoughby, J.H. (1987). Crack (an extra potent form of cocaine) Abuse: A Problem of the Eighties. Oral Surgery, Oral Medicine, Oral Pathology.  63:12-16.

Isner, J.M., Estes, N.A.M., Thompson, M.R., Costanzo-Nordin, M.R., Subramanian, R., Miller, G., Katsas, G., Sweeney, K and Sturner, W.Q. (1986). Acute Cardiac Events Temporally Related to Cocaine Abuse. New England Journal of Medicine. 315:1438-1443.

Itkonen, J., Schnoll, S. Etnd Glassroth, J. (1984). Pulmo-nary Dysfunction in "Freebase" Cocaine Users. Archives of Internal Medicine. 144:2195-2197.

Javaid, J.I., Fischman, M.W., Schuster, C.R., Dekirmenjian, H., Davis, J.M. (1978). Cocaine Plasma Concentration: Relation to Physiological and Subjective Effects in Humans. Science. 202:227-228.

Jonsson, S., O'Meara, M. and Young, J.B. (1983). Acute Cocaine Poisoning: Importance of Treating Seizures and Acidosis. American Journal of Medicine. 75:1061-1064.

Kabas, J.S., Blanchard, S.M., Matsuyama, Y., Long, J.D., Hoffman, G.W., Ellinwood, E.H., Smith, P.K., and Strauss, H.C. (1990). Cocaine-mediated impairment of cardiac conducton in the dog: A potential mechanism for sudden death after cocaine. Journal of Pharmacology and Experimental Therapeutics. 252: 185-191.

Kissner, D.G., Lawrence, W.D., Selis, J.E. and Flint, A. (1987). Crack Lung: Pulmonary Disease Caused by Cocaine Abuse. American Review of Respiratory Disease. 136:1250-1252.

Klatt, E.C., Montgomery, S., Namiki, T. and Noguchi, T.T. (1986). Misrepresentation of Stimulant Street Drugs: A Decade of Experience in an Analysis Program. Journal of Toxicology: Clinical Toxicology. 24:441-450.

Kossowsky, W.A., and Lyon, A.F. (1984). Cocaine and Acut,e Myocardial Infarction: A Probable Connection. CHEST. 86:729-731.

Levine, S.R., Washington, J.M., Jefferson, M.F., Kieran, S.N., Moen, M., Feit, H. and Welch, K.M.A. (1987). "Crack" Cocaine-Associated Stroke. Neurology. 37:1849-1853.

Manschreck, T.C., Allen, D.F. and Neville, M. (1987). Freebase Psychosis: Cases from a Bahamian Epidemic of Cocaine Abuse. Comprehensive Psychiatry. 28:555-564.

Mittleman, R.E.. R.E. and Wetli, C.V. (1987). Cocaine and Sudden "Naturar Death. Journal of Forensic Sciences. 32:11-19.

Mody, C.K., Miller, H.B., McIntyre, S.K., Cobb, S.K. and Goldberg, M.A. (1988). Neurological Complications of Cocaine Use. Neurology. 38:1189-1193.

Moore, G.F., Emanuel, J.M., Ogren, T.P., Yonkers, A.J. (1986). Cocaine: Current Clinical Use and Potential Abuse. Nebraska Medical Journal. 71:317-321.

Mugford, S. and Cohen, P. (1989). Drug Use, Social Rela-tions and Commodity Con,sumption: A Study of Recre-ational Cocaine Users in Sydney, Canberra and Melbourne: A Report to the Research into Drug Abuse Advisory Committee. Department of Sociology, Australian
National University.

Murphy, S.B., Reinarman, C. and Waldorf, D. (1989). An 11-year Follow-up of a Network of Cocaine Users. British Journal of Addiction. 84:427-436.

Murray, R.J., Albin, R.J., Mergner, W. and Criner, Q.J. (1988). Diffuse Alveolar Haemorrhage Temporally Related to Cocaine Smoking. CHEST. 93:427-428.

Murray, R.J., Snialek, J.E., Galle, M., and Albin, R.J. (1990). Pulmonary artery medial hypertrophy in cocaine users without foreign particle microembolization. CHEST. 96: 1050-1053.

Nadamanee, K, Gorelick, D.A., Josephson, M.A., Ryan, M.A., Wilkins, J.N., Robertson, H.A., Mody F.V., and Intarachot, V. (1989). Myocardial ischemia during cocaine withdrawal. Annals of Internal Medicine. 111:876-880. National Institute on Drug Abuse. (1988). Data from the Drug Abuse Warning Network: Annual Data 1987. Statistical Series I, 7. Rockville, MD: National Institute on Drug Abuse.

Ostrowski, J. (1989). Thinking about Drug Legalization. Policy Analysis no. 121. Washington: CATO Institute.

Pasternack, P.F., Colvin, S.B. and Bauman, F.G. (1985). Cocaine-induced Angina Pectoris and Acute Myocardial Infarction in Patients Younger than 40 Years. American Journal of Cardiology. 55:847.

Patel, R.C., Dutta, D. and Schonfield, S.A. (1987). Free-base Cocaine Use Associated with Bronchiolitis Obliterans Organizing Pneumonia. Annals of Internal Medicine. 107:186-187.

Peng, S.K., French, W.J., and Pelikan, P.C.D. (1989). Direct cocaine cardiotoxicity demonstrated by endomyocardial biopsy. Archives of Pathology and Labora-tory Medicine. 113:842-845.

Pierluisi, G.J. and Terndrup, T.E. (1989). Influence of topical anesthesia on the sedation of pediatric emergency department patients with lacerations. Pediatric Emer-gency Care. 5:211-215.

Rebhun, J. (1988). Association of Asthma and Freebase Smoking. Annals of Allergy. 60:339-342.

Ringwalt, C.L. and Palmer, J.H. (1989). Cocaine and Crack Users Compared. Adolescence. 24:851-860.

Rollingher, I.M., Belzberg, A.S., and Macdonald, I.L. (1986). Cocaine-induced Myocardial Infarction. Canadian Medical Association Journal. 135:45-46.

Salzman, G.A., Kahn, F. and Emory, C. (1987). Pneumomediastinum after Cocaine Smoking. Southern Medical Journal. 80:1427-1429.

Schachne, J.S., Roberts, B.H. and Thompson, P.D. (1984). Coronary-Artery Spasm and Myocardial Infarction Associ-ated with Cocaine Use. New England Journal of Medicine. 310:1665-1666.

Science. (1990). A FASEB Sampler: Cocaine's Harmful Effects. Science. 248: 166.

Shannon, M. (1988). Clinical Toxicity of Cocaine Adulterants. Annals of Emergency Medicine. 17:1243-1247.

Siegal, S. (1985) New Patterns of Cocaine Use: Changing Doses and Routes. In Kozel and Adams, eds, Cocaine Use in America.

Smart, R.G. and Anglin, L. (1987). Do We Know the Lethal Dose of Cocaine? Journal of Forensic Sciences. 32(2):303- 312.

Sternberg, R.G., Winniford, M.D., Hillis, L.D., Dowling, G.P. and Buja, L.M. (1989). Simultaneous Acute Thrombosis of Two Major Coronary Arteries Following Intravenous Cocaine Use. Archives of Pathology Laboratory Medicine. 113:521-524.

Tardiff, K., Gross, E., Wu, J., Stajic, M. and Millman, Ft. (1989). Analysis of Cocaine-positive Fatalities. Journal of Forensic Sciences. 34:53-63.

Tashkin, D.P., Simmons, M.S., Coulson, A.H., Clark, V.A. and Gong, H. (1987). Respiratory Effects of Cocaine "Freebasing" among Habitual Users of Marijuana with or without Tobacco. CHEST. 92:638-644.

Tazelaar, H.D., Karch, S.B., Stephens, B.G. and Bellingham, M.E. (1987). Cocaine and the Heart. Human Pathology. 18:195.

Trebach, A.S. (1982). The Heroin Solution. New Haven: Yale University Press.

Trebach, AS. (1987). The Great Drug War. New York: Macmillan.

Tyston, GA, DeWitt, G.W., and Eisenstein, S.J. (1989). Topical TAC (tetracaine, adrenaline, cocaine) solution for local anesthesia in children: prescribing inconsistency and acute toxicity. Southern Medical Journal. 82:1344-1346.

Van Dyke, C., Barash, P.G., Jatlow, P., Byck, R. (1976). Cocaine Plasma Concentrations after Intranasal Application in Man. Science. 191:859-861.

Verebey, K and Gold, M.S. (1988) From Coca Leaves to Crack: The Effects of Dose and Routes of Administration in Abuse Liability. Psychiatric Annals. 18:513-520.

Virmani, R., Robinowitz, M., Smialek, J.E. and Smyth, D.F. (1988). Cardiovascular Effects of Cocaine: An Autopsy Study of 40 Patients. American Heart Journal. 115:1068- 1076.

Wang, T., Hadidi, F., Triana, F. and Bargout, M. (1988). Morning Report at Christ Hospital: Myocardial Infarction Associated with Use of Cocaine. American Journal of Medical Sciences. 295:569-570.

Washton, A.M., Gold, M.S. and Pottash, A.C. (1986). Crack. Journal of the American Medical Association. 256:711.

Wehbie, C.S., Vidaillet, H.J., Navetta, F.I. and Peter, R.H. (1987). Acute Myocardial Infarction Associated with Initial Cocaine Use. Southern Medical Journal. 80:933-934.

Weiss, R.D., Tilles, D.S., Goldenheim, P.D. and Mirin, S.M. (1987). Decreased Single Breath Carbon Monoxide Diffusing Capacity in Cocaine Freebase Smokers. Drug and Alcolwl Dependence. 19:271-276.

Wiener, M.D. and Putnam, C.E. (1987). Pain in the Chest of a User of Cocaine. Journal of the American Medical Association. 258:2087-2088.

Wilkerson, R.D. (1988). Cardiovascular Toxity of Cocaine. in D. Clouet, K. Asghar, and R. Brown (eds.), Mechanisms of Cocaine Abuse and Toxicity, NIDA Research Monograph 88. Rockville, Maryland: U.S. Public Health Service, National Institute on Drug Abuse.

Wong L.S. and Alexander B.K. (in press, 1990). Cocaine Related Deaths: Media Coverage in the War on Drugs. Journal of Drug Issues.

Footnotes

Winek's authoritative Drug and Chemical Blood-Level Data (1985). In British Columbia, 1 mg/L of the total of cocaine plus benzoyl ecgonine is used as a minimum lethal dose by the coroner's service. In 6 of these 17 cases combined levels of cocaine and benzoyl ecgonine, exceeded 1 mg/L.