Cocaine cardiovascular toxicity.

Abstract: Cocaine abuse kills thousands every year. Preexisting coronary artery disease appears to account for many of the deaths, but often the mechanism is much more complex. There exists a widely held but utterly mistaken notion that cocaine-related deaths are due to drug overdose. Except in the case of drug couriers ("body packers") with massive drug exposure, death is not dose related, and cocaine blood levels cannot be used to predict toxicity. Most deaths occur after prolonged drug use, which initiates a series of changes at the molecular, cellular, and tissue levels. All of these changes favor sudden death. Potentially lethal myocardial alterations include hypertrophy, fibrosis, and microangiopathy. Recently it has become clear that genetic causes, such as fully or partially expressed congenital long QT syndrome, may also play a role. The relative importance of each of these factors is reviewed.

Key Words: chronic cocaine use, cocaine, cocaine-related death, sudden cardiac death 

**********

A significant percentage of the deaths reviewed by US coroners and medical examiners are related to cocaine use. The Drug Abuse Warning Network's (DAWN) survey of medical examiners reported 4,043 cocaine-related deaths in the year 2000. (1) Clinical experience strongly suggests that the number of cocaine-related deaths is rising, and probably quite substantially. Elsewhere in the world the absolute number of deaths, though miniscule by American standards, is increasing at an even faster rate. Cocaine abuse poses a very large problem in both the New World and the Old, and it does not look as if the problem will be going away any time soon.

One might suppose that given the large number of deaths, some diagnostic patterns would have emerged and that the diagnosis of cocaine-related death would be a simple matter. In fact, the cause of death in cocaine users is often difficult to determine. There are two major problems. The first is that most cocaine-related deaths are a consequence of vascular disease, and there is very little to distinguish cocaine-induced vascular disease from any other kind of naturally occurring vascular disease. The second difficulty has to do with drug tolerance and interindividual variability.

Recent studies with healthy volunteers have shown that tolerance to physiologic and psychologic effects begin immediately after the first dose is given. With the first dose, both the degree of psychologic "high" and the magnitude of physiologic responses (pulse, blood pressure) correlate with cocaine plasma levels, but even though plasma levels of cocaine rise with additional doses, the "high" does not increase or "improve," and no further increases in pulse or blood pressure occur. (2) In other words, acute tolerance emerges between the first and second dose.

A large study has shown that clinical toxicity in emergency room patients did not relate to measured blood concentrations. The individual with the highest blood concentration, 3.9 mg/L, a concentration that had in the past been considered close to fatal, had few symptoms and was discharged to police custody. Conversely, one of the few patients who died had a plasma concentration of 0.029 mg/L, a concentration below threshold needed to produce measurable changes in pulse or blood pressure in humans. (3) Similar observations regarding postmortem cocaine concentrations were made more than a decade ago (4); blood concentrations in those dying of cocaine toxicity and in those where cocaine was an incidental finding completely overlap. In fact, this same situation applies to all abused opiates and stimulants; blood concentrations and toxic manifestations do not correlate. (5)

If blood cocaine concentration is not a valid indicator of toxicity, and there is nothing particularly unique about the underlying pathology, are there any good indicators of cocaine toxicity? Probably the best indicator is confirmed history of chronic use. (6) It is important to understand that death from acute cocaine toxicity is a relatively rare event, essentially only seen in "body packers" or "body stuffers" who find themselves exposed to massive, multigram quantities of cocaine. Otherwise, the only cocaine users likely to become seriously ill or die are the chronic users. Chronic cocaine abuse sets in train a series of neurochemical and anatomic alterations that favor the occurrence of sudden death. Some of these changes are anatomic (coronary artery disease, myocardial hypertrophy), some are subcellular (upregulation of certain dopamine transporters and downregulation of others), and some are centrally mediated. (7,8) The purpose of this study was to provide a brief overview of changes that occur within the hearts of cocaine users and to explain why those changes often prove fatal.

Coronary Artery Disease

The incidence of coronary artery disease in cocaine abusers has not been studied systematically, although myocardial infarction among cocaine users is now so common that it is no longer considered reportable. Cocaine-related coronary spasm is mentioned frequently as a cause of morbidity in cocaine users, but spasm probably only becomes an issue in individuals who already have preexisting coronary artery disease. The degree of spasm produced (roughly a 10% decrease in the caliber of large epicardial vessels), even by sizable doses of cocaine given intravenously, is more or less comparable to the degree of constriction produced by cigarette smoking. (9)

Laboratory studies continue to suggest a linkage between cocaine use and accelerated atherosclerosis. Aortic vascular smooth muscle cells, for example, undergo rapid apoptosis in response to cocaine, and the degree of damage appears to be concentration-dependent. (10) Similar responses have been observed in cell cultures of human endothelial cells. (11) Whether or not cocaine really does accelerate atherosclerotic changes, cocaine users with coronary artery disease are at increased risk for infarction and sudden death. Much of the risk can be attributed to the ability of cocaine to block the reuptake of norepinephrine. Chronic use of cocaine leads to elevated blood catecholamine concentrations, primarily norepinephrine, (12) exaggerating the effects of those hormones.

Excess stimulation leads to vasoconstriction of vascular smooth muscle, and that leads to ischemia. Stimulation of [beta]-receptors also means that heart rate, and therefore myocardial work (blood pressure X heart rate) increases, raising the demand for oxygen. In the presence of preexisting coronary artery disease, the added burden may be enough to precipitate, if not infarction, then at least loss of membrane integrity, leading, in turn, to electrical instability (reentry) and ultimately the occurrence of lethal ventricular arrhythmias. There are no controlled studies, but this probably accounts for a fair percentage of reported cocaine fatalities.

Other types of coronary artery disease, such as coronary artery dissection and intimal proliferation with an appearance that very much resembles the pattern seen in chronic rejection, (13) are also recognized. However, both disorders remain rare and reportable occurrences. The latter condition may be somewhat underdiagnosed because the process involves the entire length of the diseased blood vessel, and, unless earlier angiograms are available for comparison, would not be apparent on angiography. The former condition is being diagnosed with increasing frequency. At one time, coronary artery dissection was considered almost exclusively a disease of pregnancy, but within the last few years it is being reported more and more often in stimulant abusers. (14) The left anterior descending artery is the most frequently affected vessel, and, histologically, the most common finding is a hematoma occupying the outer third of the media. This can result in complete compression of the true lumen, resulting in myocardial infarction or sudden cardiac death (SCD). In pregnancy, it is assumed that hemodynamic factors and a lytic action of tissue proteases are responsible. Why this disease should emerge in cocaine abusers remains a mystery.

Myocardial Hypertrophy

Myocardial hypertrophy is an adaptive response to increased wall stress, albeit a maladaptive one that eventually leads to heart failure. (15) Hypertrophy also leads to the occurrence of SCD, even in nondrug users. The hearts of rats chronically treated with cocaine become enlarged, containing an increased amount of collagen. Myocardium from these rats expresses genes that elevate transcription of atrial natriuretic factor, [beta]-myosin heavy chain, and skeletal [alpha]-actin. (16,17) At the same time, there is reduced expression of the sarcoplasmic reticulum proteins phospholamban, sarcoplasmic reticulum [Ca.sup.2+]-ATPase-2[alpha], and ryanodine receptors. (18) Similar changes have been observed in humans.

Cocaine-induced myocardial hypertrophy in humans has been confirmed in both clinical and autopsy studies. (19) Compared with age- and sex-matched normal control subjects, electrocardiograms of both asymptomatic cocaine users in rehabilitation and symptomatic cocaine users with chest pain are abnormal and meet voltage criteria for hypertrophy. (20,21) Echocardiographic studies confirm the existence of increased left ventricular mass. (22) In autopsy studies, the heart weights of cocaine users, regardless of the cause of death, are, on average, 10% greater than the mean weights predicted by standard nomograms. (23) Epidemiologic, clinical, and autopsy studies have repeatedly demonstrated that this type of heart disease is associated with an increased risk for sudden death (24) and that the increase is even greater in those who are overweight. (25)

It is not clear how cocaine initiates the process of hypertrophy. Cocaine acts as an oxidant, and hypertrophy may be due to direct oxidant effect of cocaine on cardiac myocytes, or it might be the result of some extrinsic stimuli, such as cocaine-induced hypertension. Sustained hypertension occurs in experimental animals treated with cocaine, but its occurrence is less predictable in humans. Most of the stimuli associated with myocardial hypertrophy exert their effect through activation of a G-coupled protein and either adenyl cyclase or phospholipase-C. (26) Activation of either of the latter activates the genes needed to make proteins required for cell growth and results in activation of cardiac fibroblasts that appear to control heart size through the production and release of various cytokines. (15) There is every reason to suppose that a similar sequence occurs in cocaine users. If cocaine-related hypertrophy was simply a response to cocaine-related hypertension, that would explain why very similar morphologic changes are seen in the hearts of cocaine users and in hypertensive hearts: enlarged myocytes, increased collagen deposition, perivascular fibrosis, (27) and medial hypertrophy of the smaller arterioles. (28,29)

Catecholamine Toxicity

The histology of cocaine users' hearts often resembles that of patients found to have heart failure secondary to pheochromocytoma. (30) The myocardial response to chronic catecholamine toxicity is well characterized. Norepinephrine "myocarditis" was observed almost as soon as pressor agents were first introduced into clinical practice. (31) Lesions are very focal, with necrotic myocytes surrounded by normal-appearing cells. Unlike ischemic necrosis, in which myofilaments remain in register, the myofilaments damaged by cat-echolamines become hypercontracted, forming eosinophilic clumps. Injured cells are gradually replaced by fibrosis that is best described as microfocal, to convey the fact that the pattern of injury cannot be related to any particular blood vessel distribution. Diffuse myocardial fibrosis favors reentry and the occurrence of arrhythmias (see Fig. 1).

Microvascular Disease

Thickening of the small intramyocardial arteries has been demonstrated in cocaine users. (32) This change may lead to supply-demand mismatch in blood flow, and it has grave clinical and therapeutic implications for hypertensives, diabetics, and patients with coronary artery disease. (33) It is speculated that cocaine-induced apoptosis somehow damages the muscular layers of the small resistance vessels (see Fig. 2), (34) though others believe that some other sort of direct oxidant action is involved. (35) Recognition of the role of apoptosis and the occurrence of microangiopathy is likely to have a major impact on treatment, since many drugs are now known to exert antiapoptotic effects, and treatment with antiapoptotic drugs might prevent the development of left ventricular hypertrophy and sudden coronary death in chronic cocaine abusers. The difficulty in treating these individuals is that cocaine users do not know they have any underlying microvascular disease, and they usually do not seek treatment.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

It is possible to measure the degree of underperfusion in those seeking treatment, and a number of new hemodynamic indices have been devised to study microvascular disease. The most useful of these indices is called the fractional flow reserve. It is determined by placing an ultrasound flow probe into a coronary artery and measuring flow and resistance at baseline and then again after maximal vasodilation. (36) The difference between the two measurements allows for accurate assessment of perfusion. (37) As the thickness of the media increases, the caliber of the vessel decreases, resulting in decreased flow and underperfusion. If the fractional flow reserve is found to be decreased, the myocardium is not receiving enough blood flow and is ischemic. Ischemic myocardium has a lower threshold for fibrillation.

Myocardial Hypertrophy and SCD

Why do these morphologic changes favor the occurrence of sudden death? It is probably for many of the same reasons that favor SCD in hypertensives. Coronary flow reserve in these individuals may be reduced by as much 30 to 50%, although the degree of flow reduction does not correspond exactly to the severity of hypertensive left ventricular hypertrophy. (38,39) In hypertrophied myocardium, blood flow is compromised not just because microvascular disease is present but also because of a failure of angiogenesis. New blood vessel formation generally lags muscle fiber hypertrophy, so that at all times, even at rest, the hypertrophic myocardium is relatively underperfused, at least when compared with muscle fibers in normal myocardium.

Another problem is that hypertrophied hearts utilize oxygen less efficiently than normal hearts. Under normal circumstances, coronary flow can increase by as much as 80%, but if the resistance vessels are abnormal and the myocytes are increased in size, the increase in reserve flow may be far less. Interestingly, this inability to respond and provide more flow is only seen when myocardial enlargement is due to pressure overload. (40) When cardiac enlargement is a consequence of exercising/training, angiogenesis keeps up with muscle growth, and there is no microvascular disease and no underperfusion. (41) In cocaine abusers, it can be presumed that the presence of hypertrophy means that the myocardium is relatively ischemic. Ischemia disrupts the normal processes of repolarization, thereby setting the stage for the occurrence of reentry arrhythmias. The process of re-entry is thought to account for the majority of lethal arrhythmias, even in non-drug users.

Cocaine and QT Prolongation

Hypertrophied myocardium is not electrically homogenous. The degree of abnormality can be quantified by measuring QT interval dispersion (QTd), defined as the difference between the lengths of the QT interval in the lead where it is maximal, with the length of the interval in the lead where it is minimal. It has been suggested that this measurement provides information about the spatial differences in myocardial recovery time, that is, how long it takes cardiomyoctes to repolarize after they have contracted in different parts of the heart. The greater the difference between the longest and shortest QT interval, the greater the degree of QT dispersion, the greater the degree of inhomogeneity, and the greater the probability of SCD. The length of the QT interval varies from lead to lead, but under normal circumstances, the difference between the longest and shortest QT interval, abbreviated as QTd, should be 49 [+ or -] 20 ms. When QT dispersion is greater than 80 ms, there is loss of synchronized depolarization, favoring the occurrence of re-entry arrhythmia. In other words, QTc prolongation and QT interval dispersion exert very much the same effect on conduction as does heart size itself. In fact, the phenomena probably just represents two sides of the same coin. (42,43)

The effects of QT dispersion are highly significant. Patients with a corrected QT interval of more than 440 ms are many times more likely to have SCD than individuals with QTc of less than 400 ms. Prolongation of either the corrected QT interval or QTd indicates that a shift toward reduced vagal modulation has occurred, itself another predictor of SCD. (42,44,45) It only makes sense that QT dispersion would be seen in individuals with enlarged fibrotic hearts containing increased amounts of collagen. The repolarization process takes longer, and re-entry is more likely. However, the occurrence of QT prolongation and QT dispersion in cocaine users is not restricted to those individuals with enlarged hearts.

QT Prolongation and SCD

Like all local anesthetic agents, cocaine is classified as a slow on-off sodium channel blocker, but it is also a fast on-off potassium blocker. As a consequence, its effects on repolarization are biphasic: At low concentrations, cocaine delays ventricular recovery, whereas at higher concentrations, cocaine accelerates it. A massive overdose with cocaine, as might be seen in a "body packer" with a ruptured packet of cocaine in their intestines, can lead to monomorphic slow ventricular tachycardia, idioventricular rhythm, or, if the dosage is sufficiently large, complete standstill. (46) In some of the cases, sodium channel blockade can be partially reversed by treatment with sodium bicarbonate. These cases are rare, or at least rarely come to medical attention.

Much more commonly, hypertrophy and QT dispersion are the underlying cause of death. Once a user stops taking cocaine, myocardial remodeling will, at some point, reverse the structural changes that placed the user at increased risk, but the process takes time. In a chronic user with an enlarged heart, cocaine can legitimately be deemed the cause of death, even when it is not present. However, when the drug is present, risk is very much greater than the risk attributable to simple left ventricular hypertrophy. Even modest doses of cocaine cause repolarization changes and QT prolongation in individual myocytes, (47) and there is no well-established threshold duration below which QT prolongation is known to be harmless. (48)

QT prolongation leads to the occurrence of torsade de pointes, a type of monomorphic ventricular tachycardia. Cocaine, by virtue of its ion channel-blocking ability, facilitates its occurrence. This comes about mainly because cocaine acts as a fast on-off potassium blocker. (49-52) Potassium channel blockade only causes torsade in those with the right set of underlying risk factors. The most obvious of those underlying factors is, of course, hypertrophy, but it is now apparent that genetic causes, such as fully or partially expressed congenital long QT syndrome (LQTS), are equally important.

Inherited LQTS is caused by mutations in [K.sup.+] channel, [Na.sup.+] ion channel, or ankyrin-B (a member of a family of adapter proteins required for, among other things, the coordinated assembly of Na/Ca exchanger) genes. (53) Acquired LQTS is much more common than the inherited form and usually occurs as an unintended side effect of treatment with a structurally diverse group of medications (eg, terfenadine, cisapride), all of which exert their primary effect on potassium conductance. The molecular basis of cocaine-induced LQTS is a block of human ether-a-go-go related gene (hERG) channels that conduct [I.sub.Kr], the rapid delayed rectifier [K.sup.+] current important for the repolarization phase of the cardiac action potential. (48) A reduction in [I.sub.Kr] prolongs the action potential duration of ventricular myocytes, lengthens the QT interval, and increases dispersion as measured by electrocardiographic recordings. The net effect is to increase the risk that torsade de pointes will occur, degenerate into ventricular fibrillation, and cause sudden death. If structural heart disease is also present, the chances for SCD are greatly increased.

Conclusions

Cocaine exerts so many deleterious effects on the heart that it is a wonder that the mortality rate is not higher. The only aspect of cocaine cardiotoxicity that seems perfectly clear, with the possible exception of individuals with heritable channelopathy and those with truly massive overdose, is that the process takes a long time to develop. Unless severe underlying heart disease is already present, cocaine-related deaths do not occur with the first exposure or even the second or third. That reality has positive implications for treatment. If scientists are finally able to assign the proper importance to each of the different effects of cocaine, it should be possible to design treatments that work. Until that happens, efforts at treatment and prevention are likely to remain unsuccessful, and the number of dead is likely to keep growing.

References

1. Substance Abuse and Mental Health Service Administration, Office of Applied Studies. Annual Medical Examiner Data from the Drug Abuse Warning Network, 2002. DAWN Series D-23, DHHS Publication (SMA) 03-3781.

2. Foltin RW, Fischman MW. Smoked and intravenous cocaine in humans: acute tolerance, cardiovascular and subjective effects. J Pharmacol Exp Ther 1991;257:247-261.

3. Blaho K, Logan B, Winbery S, et al. Blood cocaine and metabolite concentrations, clinical findings, and outcome of patients presenting to an ED. Am J Emerg Med 2000;18:593-598.

4. Karch SB, Stephens B, Ho CH. Relating cocaine blood concentrations to toxicity: an autopsy study of 99 cases. J Forensic Sci 1998;43:41-45.

5. Drummer O, Forrest AR, Goldberger B, et al. Forensic science in the dock. BMJ 2004;329:636-637.

6. Stephens BG, Jentzen JM, Karch S, et al. National Association of Medical Examiners Position Paper on the Certification of Cocaine-related Deaths. Am J Forensic Med Pathol 2004;25:11-13.

7. Wetli CV, Mash D, Karch SB. Cocaine-associated agitated delirium and the neuroleptic malignant syndrome. Am J Emerg Med 1996;14:425-428.

8. Stephens BG, Jentzen JM, Karch S, et al. Criteria for the interpretation of cocaine levels in human biological samples and their relation to the cause of death. Am J Forensic Med Pathol 2004;25:1-10.

9. Lange RA, Cigarroa RG, Yancy CW Jr, et al. Cocaine-induced coronary-artery vasoconstriction. N Engl J Med 1989;321:1557-1562.

10. Su J, Li J, Li W, et al. Cocaine induces apoptosis in primary cultured rat aortic vascular smooth muscle cells: possible relationship to aortic dissection, atherosclerosis, and hypertension. Int J Toxicol 2004;23:233-237.

11. He J, Xiao Y, Zhang L. Cocaine induces apoptosis in human coronary artery endothelial cells. J Cardiovasc Pharmacol 2000;35:572-580.

12. Mendelson JH, Mello NK, Sholar MB, et al. Temporal concordance of cocaine effects on mood states and neuroendocrine hormones. Psychoneuroendocrinology 2002; 27:71-82.

13. Simpson RW, Edwards WD. Pathogenesis of cocaine-induced ischemic heart disease: autopsy findings in a 21-year-old man. Arch Pathol Lab Med 1986;110:479-484.

14. Steinhauer JR, Caulfield JB. Spontaneous coronary artery dissection associated with cocaine use: a case report and brief review. Cardiovasc Pathol 2001;10:141-145.

15. Jaffre F, Callebert J, Sarre A, et al. Involvement of the serotonin 5-HT2B receptor in cardiac hypertrophy linked to sympathetic stimulation: control of interleukin-6, interleukin-1 beta, and tumor necrosis factor-alpha cytokine production by ventricular fibroblasts. Circulation 2004;110:969-974.

16. Tseng YT, Rockhold RW, Hoskins B, et al. Cardiovascular toxicities of nandrolone and cocaine in spontaneously hypertensive rats. Fundam Appl Toxicol 1994;22:113-121.

17. Henning RJ, Li Y. Cocaine produces cardiac hypertrophy by protein kinase C dependent mechanisms. J Cardiovasc Pharmacol Ther 2003;8:149-160.

18. Wang JF, Yan X, Min J, et al. Cocaine downregulates cardiac SERCA2a and depresses myocardial function in the murine model. Can J Physiol Pharmacol 2002;80:1015-1021.

19. Karch SB, Green GS, Young S. Myocardial hypertrophy and coronary artery disease in male cocaine users. J Forensic Sci 1995;40:591-595.

20. Nademanee K. Cardiovascular effects and toxicities of cocaine. J Addict Dis 1992;11:71-82.

21. Chakko S, Fernandez A, Mellman TA, et al. Cardiac manifestations of cocaine abuse: a cross-sectional study of asymptomatic men with a history of long-term abuse of 'crack' cocaine. J Am Coll Cardiol 1992;20:1168-1174.

22. Brickner ME, Willard JE, Eichhorn EJ, et al. Left ventricular hypertrophy associated with chronic cocaine abuse. Circulation 1991;84:1130-1135.

23. Kitzman DW, Scholz DG, Hagen PT, et al. Age-related changes in normal human hearts during the first 10 decades of life, II (maturity): a quantitative anatomic study of 765 specimens from subjects 20 to 99 years old. Mayo Clin Proc 1988;63:137-146.

24. Haider AW, Larson MG, Benjamin EJ, et al Increased left ventricular mass and hypertrophy are associated with increased risk for sudden death. J Am Coll Cardiol 1998;32:1454-1459.

25. Messerli FH, BD, Ventura HO, et al. Overweight and sudden death: Increased ventricular ectopy in cardiopathy of obesity. Arch Intern Med 1987;147:1725-1728.

26. Francis GS, McDonald KM, Cohn JN. Neurohumoral activation in pre-clinical heart failure: remodeling and the potential for intervention. Circulation 1993;87(5 Suppl): IV-90-IV-96.

27. Gonzalez A, Lopez B, Diez J. Fibrosis in hypertensive heart disease: role of the renin-angiotensin-aldosterone system. Med Clin North Am 2004;88:83-97.

28. Kern TS, Engerman RL. Microvascular metabolism in diabetes. Metabolism 1986;35(4 Suppl 1): 24-27.

29. Picano E, Pelosi G, Marzilli M, et al. In vivo quantitative ultrasonic evaluation of myocardial fibrosis in humans. Circulation 1990;81:58-64.

30. Karch SB, Billingham ME. Myocardial contraction bands revisited. Hum Pathol 1986;17:9-13.

31. Szakacs JE Cannon A. L-Norepinephrine myocarditis. Am J Clin Pathol 1958;30:425-434.

32. Karch SB, Billingham ME. Coronary artery and peripheral vascular disease in cocaine users. Coron Artery Dis 1995;6:220-225.

33. Abarquez RF Jr, Cinco JE. Microcirculation: target therapy in cardiovascular diseases: a clinical perspective. Clin Hemorheol Microcirc 2003;29:157-165.

34. Nassogne MC, Evrard P, Courtoy PJ. Selective direct toxicity of cocaine on fetal mouse neurons: teratogenic implications of neurite and apoptotic neuronal loss. Ann N Y Acad Sci 1998;846:51-68.

35. Sherman MP, Aeberhard EE, Wong VZ, et al. Effects of smoking marijuana, tobacco or cocaine alone or in combination on DNA damage in human alveolar macrophages. Life Sci 1995;56: 2201-2207.

36. Kern MJ. Coronary physiology revisited: practical insights from the cardiac catheterization laboratory. Circulation 2000;101:1344-1351.

37. Frohlich ED. State of the art lecture: risk mechanisms in hypertensive heart disease. Hypertension 1999;34:782-789.

38. Britten MB, Zeiher AM, Schachinger V. Microvascular dysfunction in angiographically normal or mildly diseased coronary arteries predicts adverse cardiovascular long-term outcome. Coron Artery Dis 2004;15:259-264.

39. Takechi S, Nomura A, Machida M, et al. Different coronary blood flow increase in left ventricular hypertrophy due to hypertension compared to hypertrophic cardiomyopathy at elevated heart rate. Hypertens Res 2003;26:789-793.

40. Gavin JB, Maxwell L, Edgar SG. Microvascular involvement in cardiac pathology. J Mol Cell Cardiol 1998;30:2531-2540.

41. Titcomb CP Jr. Matters of the athletic heart. J Insur Med 2003;35:179-182.

42. Engel G, Beckerman JG, Froelicher VF, et al. Electrocardiographic arrhythmia risk testing. Curr Probl Cardiol 2004;29:365-432.

43. Calder KK, Tomongin C, Mallon WK, et al. Manual measurement of QT dispersion in patients with acute myocardial infarction and nondiagnostic electrocardiograms. Acad Emerg Med 2002;9:851-854.

44. Sredniawa B, Musialik-Lydka A, Pasyk S. [Measurement dispersion of the QT interval and its significance in different diseases]. Pol Merkuriusz Lek 2001;11:52-55.

45. Saadeh AM. Relation between age, ventricular arrhythmia, left ventricular hypertrophy and QT dispersion in patients with essential hypertension. Acta Cardiol 2004;59:249-253.

46. Seifen E, Plunkett LM, Kennedy RH. Cardiovascular and lethal effects of cocaine in anesthetized dogs and guinea-pigs. Arch Int Pharmacodyn Ther 1989;300:241-253.

47. Perera R, Kraebber A, Schwartz MJ. Prolonged QT interval and cocaine use. J Electrocardiol 1997;30:337-339.

48. Fenichel RR, Malik M, Antzelevitch C, et al. Drug-induced torsades de pointes and implications for drug development. J Cardiovasc Electrophysiol 2004;15:475-495.

49. Karle CA, Kiehn J. An ion channel 'addicted' to ether, alcohol and cocaine: the HERG potassium channel. Cardiovasc Res 2002;53:6-8.

50. Ferreira S, et al Effects of cocaine and its major metabolites on the HERG-encoded potassium channel. J Pharmacol Exp Ther 2001;299:220-226.

51. Zhang S, Rajamani S, Chen Y, et al. Cocaine blocks HERG, but not KvLQT1 + minK, potassium channels. Mol Pharmacol 2001;59:1069-1076.

52. O'Leary ME. Inhibition of human ether-a-go-go potassium channels by cocaine. Mol Pharmacol 2001;59:269-277.

53. Wichter T, Schulze-Bahr E, Eckardt L, et al. Molecular mechanisms of inherited ventricular arrhythmias. Herz 2002;27:712-739.

Steven B. Karch, MD

From Berkeley, California.

Reprint requests to Steven B. Karch, MD, PO Box 5139, Berkeley, CA 94705. Email: [e-mail]

Accepted November 8, 2004.

Supported in part by the Royal Society of Medicine.

Presented at The Royal Society of Medicine Symposium "Profiting from cocaine: a growing addiction," London, December 12, 2003.

RELATED ARTICLE: Key Points

* The number of cocaine-related deaths continues to increase in the United States and Europe.

* Cocaine-related deaths are not dose related, and isolated blood levels do not predict toxicity.

* Chronic cocaine use leads to cardiac hypertrophy, myocardial fibrosis, and microangiopathy.

* Most cocaine-related deaths occur in chronic, not naive, drug takers.

* The cause of death is multifactorial but in some cases may involve a heritable ion channel abnormality.