Cocaine use is endemic in the United States. Its toxicity is substantial, rendering it one of the most frequent causes of drug-related visits to the emergency department (ED). Cocaine intoxication will frequently evoke chest pain, prompting a concern in both patients and providers that myocardial damage is imminent. Indeed, numerous studies have described both ST-segment elevation myocardial infarction (STEMI) and non–ST-segment elevation acute coronary syndrome (ACS) in temporal association with cocaine use.1–4 When a patient with overt myocardial damage presents in association with cocaine intoxication, the disposition from the ED is fairly straightforward. However, what to do with the patient who has neither troponin elevation nor an electrocardiogram (ECG) with evidence of ischemia/infarction? In other words, given a low-risk chest pain patient who has chest pain after cocaine use, do we need to change our usual way of managing the patient? If the patient chronically uses cocaine, does that change what we need to do?

Chang et al.5 have presented a retrospective analysis of a convenience sample of patients receiving a coronary computed tomography angiography (CTA) scan for evaluation for potential ACS. By removing subjects with elevated cardiac biomarkers, high-risk ST-segment changes, and other clinical factors that would preclude observation unit management, the authors selected a sample that would have a fairly low-to-moderate baseline risk of ACS and one wherein a history of chronic cocaine use and presumed obstructive coronary artery disease could truly have a measurable effect on the decision-making process in the ED. After multivariate adjustment for a host of typical risk factors, the authors reported adjusted relative risk ratios that do not support cocaine as a risk factor for an increased burden of obstructive coronary artery disease.

This is a solid, but not perfect, study. Data definition is key. The use of previous medical records to assess whether a subject was a chronic cocaine user or cocaine-naïve is problematic. The stark truth with stimulant abuse is that clinicians do not ask the question enough,6 and patients do not tell us often enough.7 The authors report that, in those subjects who had positive urine drug testing for cocaine, only 1 of 34 had denied cocaine use. This is in contrast to previous reports7,8 and to our own clinical experience; however, social normative behavior and values will vary regionally, leading to patients perhaps being more forthright about cocaine use in different environments. Although patients were directly asked by research personnel at the index visit regarding current cocaine use, the authors relied on an (exhaustive) chart review to determine the presence or absence of previous cocaine abuse. Fairly conservative assumptions were made and actions taken, such as excluding those individuals with no mention one way or the other in the previous record as to cocaine use. Likewise, cocaine use was classified as chronic unless specifically established otherwise—an assumption based on the authors’ previous work,1 but with substantial face validity in our own experience. It remains conceivable that individuals with few cocaine exposures would be classified in the same ohort as those with long-standing abuse. The potential for crossover between groups (untested, nondisclosing cocaine users classified as nonusers, as well as infrequent users cohorted with heavy abusers) would bias toward a null result. However, the authors are correct in stating that limiting inclusion to those patients receiving drug testing would result in substantial selection bias. It is this interplay between misclassification bias and selection bias that often limits the utility of studies that rely on retrospective data.

This study builds on the work of Weber et al.,8 who evaluated the safety of short-stay observation in cocaine chest pain, as well as the utility of stress testing in this setting. In this landmark study, Weber and colleagues evaluated 302 patients with cocaine chest pain (all of whom had urine drug testing) in an observation setting (42 were admitted after initial evaluation), including 158 who received stress testing with or without sestamibi imaging and 144 who presented after the institution had discontinued routine stress testing for patients with cocaine-related chest pain. All patients were observed with serial cardiac biomarker and ECG measurements for 9 hours. A remarkable effort produced a 30-day follow-up rate of 99%. Only four subjects experienced myocardial infarction within the follow-up interval, all in the setting of continued cocaine use. None of the patients placed in the observation pathway developed myocardial infarction, as contrasted with nearly 50% who were admitted. In the observation unit population, stress testing was not helpful in predicting 30-day outcomes.

Cunningham et al.9 similarly validated the safety of the approach of serial biomarkers and ECG without provocative testing. Although the follow-up rate was not as complete as in the study by Weber et al., at 1 year no myocardial infarctions were seen in patients who had completed an initial observation unit stay consisting of 9 hours of sequential biomarker assessments and ECGs (n = 175). In this cohort, 79% continued to use cocaine during the study period, and approximately one-fourth of the sample re-presented to the ED with cocaine chest pain within the year. By way of comparison, Collin et al.10 followed a young (age < 40 years) cohort of low-risk chest pain patients evaluated in the ED with a normal or nonischemic ECG (cocaine users were excluded). At 1 year, the rate of death, myocardial infarction, and revascularization was 1% (6 of 609, 95% confidence interval [CI] = 0.4% to 2.3%). Likewise, Weisenthal presented 1-year follow-up data on outcomes of ED chest pain patients from the same institution as Chang, again excluding cocaine users, stratified by TIMI score.11 One-year death, myocardial infarction, and revascularization rates ranged from 4% for patients with a TIMI score of 0% to 13% for those with a TIMI score of 2. These data suggest that long-term outcomes for patients presenting with cocaine chest pain are comparable to those of patients presenting with “conventional” chest pain.

It should be noted that Chang et al. have not presented a study that is attempting to set a practice standard. This is a study attempting to quantify the relationship between plaque burden and chronic cocaine use. A study establishing comparative efficacy between coronary CTA and routine management would require data elements not available in the present article—specifically, patient outcomes and resource utilization are required to judge the wisdom of routine coronary CTA in the setting of cocaine chest pain. Data were not presented regarding rapid discharge from the ED versus prolonged ED stay awaiting daytime coronary CTA availability. A study by the same group using coronary CTA for a similar low-risk chest pain cohort had an even split between immediate CTA and delayed CTA (285 vs. 283 subjects).12 Assuming a similar distribution of patients arriving off-hours existed in this study, it is not at all clear that this approach would save time and resources when compared to conservative management (such as serial biomarkers without objective testing).

The authors nicely summarize the existing literature base that has prompted the common wisdom that chronic cocaine use produces accelerated atherosclerosis within the coronary arteries, leading to a presumed increased risk of ACS at an earlier age. However, there is substantial selection bias in these studies, as the samples studied were those with myocardial infarction,13 those selected for coronary angiography,14 or those who were already dead.15,16 The logical construct could be therefore summarized as follows: if a patient using cocaine is sick, then he or she is more likely to have advanced/premature coronary artery disease (previous literature). If a patient using cocaine is not sick, he or she is not more likely to have advanced/premature coronary artery disease (Chang). Therefore, it seems to us that the first step is to determine whether or not a patient using cocaine is sick, and in contrast to non–cocaine-using patients suspected of having ACS, knowing the coronary artery disease status becomes a secondary, almost Bayesian, consideration. Given that patients with cocaine chest pain tend to have better short-term3 and long-term9–11 outcomes, and given what we now know in terms of the absence of a direct relationship between obstructive coronary artery disease and cocaine use,5 cocaine as a provocative factor for chest pain seems to preference toward less, rather than more, diagnostic testing.

Unanswered questions remain for risk stratification in cocaine-related chest pain. Does the absence of what is conventionally assumed to be obstructive coronary artery disease ensure short-term safety in the setting of a drug known to cause vasospasm and direct myocardial toxicity? Previous literature has suggested that the highest risk for myocardial infarction is within the first hour after use.2 With a plasma half-life of approximately 46 minutes,17 and substantial decay in hemodynamic effects beginning 2 to 3 hours after last cocaine use,18 it is reasonable to assume maximum cardiac injury would likely occur during the first few hours after use. However, given current cardiac biomarker kinetics, a bump in the biomarkers might not show by the time the coronary CTA was performed and the patient judged safe for discharge based on anatomic criteria. The authors reported, in a previous, smaller sample, a zero event rate at 30 days following coronary CTA for cocaine chest pain.19 However, only 18 subjects (of 59 studied) were discharged immediately from the ED without serial observation. With a CI that encompasses 0% to 19% (zero events in 18 trials), further safety data are required.

What will high-sensitivity troponin analysis bring to the table? Will high-sensitivity troponin approach the status of D-dimer testing in venous thromboembolism—reassuring if negative and only an abnormal result leading to further testing? Even with conventional troponin assays, there is a gray zone between the “normal” threshold and the point of optimum diagnostic accuracy for myocardial infarction.20 Will we see the same “test-creep”—more patients undergoing testing for weaker reasons with no more increase in pathologic diagnoses—that we have seen with pulmonary CTA and D-dimer?21 Or will we be able to discharge patients after a single blood test, given that their cardiac system has already run on the “crack pipe treadmill” and failed to sustain myocardial damage?

Finally, we would be remiss in our duties to our colleagues in the public health arena if we did not bring up the question of prevention and addiction treatment. Rates of recidivism and continued cocaine use are extraordinarily high in this population.9,22 A school of thought holds that an ED visit is a potential teachable moment for those patients who exhibit high-risk behaviors in regards to tobacco use,23 alcohol abuse,24 or violent behaviors.25 Does a rapid assessment and discharge plan lessen our ability to affect future patterns of cocaine use?

In summary, the present study adds substantially to the knowledge base regarding the management of chest pain in cocaine users. We know that cocaine can infrequently produce an immediate, overt, disastrous response resulting in STEMI, myocardial damage, and other end-organ damage due to hypertensive emergency, with the greatest risk in the first hours after cocaine use.2 These patients are already “at the front of the line,” and their management does not represent a particular dilemma. (With the exception of the persistent question of the risk vs. benefit of beta blockade in acute cocaine toxicity with concomitant hypertensive emergency, a question that the current literature base does not adequately address.) What Chang and colleagues have provided is evidence that chronic cocaine use does not appear to be a trump card overriding established risk assessment protocols.


  1. Top of page
  2. References
  • 1
    Hollander JE, Hoffman RS, Gennis P, et al. Prospective multicenter evaluation of cocaine-associated chest pain. Cocaine Associated Chest Pain (COCHPA) Study Group. Acad Emerg Med. 1994; 1:3309.
  • 2
    Mittleman MA, Mintzer D, Maclure M, Tofler GH, Sherwood JB, Muller JE. Triggering of myocardial infarction by cocaine. Circulation. 1999; 99:273741.
  • 3
    Feldman JA, Fish SS, Beshansky JR, Griffith JL, Woolard RH, Selker HP. Acute cardiac ischemia in patients with cocaine-associated complaints: results of a multicenter trial. Ann Emerg Med. 2000; 36:46976.
  • 4
    Weber JE, Chudnofsky CR, Boczar M, Boyer EW, Wilkerson MD, Hollander JE. Cocaine-associated chest pain: how common is myocardial infarction? Acad Emerg Med. 2000; 7:8737.
  • 5
    Chang AM, Walsh K, Shofer F, McCusker C, Litt H, Hollander J. Relationship between cocaine use and coronary artery disease in patients with symptoms consistent with an acute coronary syndrome. Acad Emerg Med. 2011; 18:19.
  • 6
    James TL, Feldman J, Mehta SD. Physician variability in history taking when evaluating patients presenting with chest pain in the emergency department. Acad Emerg Med. 2006; 13:14752.
  • 7
    Lee MO, Vivier PM, Diercks DB. Is the self-report of recent cocaine or methamphetamine use reliable in illicit stimulant drug users who present to the emergency department with chest pain? J Emerg Med. 2009; 37:23741.
  • 8
    Weber JE, Shofer FS, Larkin GL, Kalaria AS, Hollander JE. Validation of a brief observation period for patients with cocaine-associated chest pain. N Engl J Med. 2003; 348:5107.
  • 9
    Cunningham R, Walton MA, Weber JE, et al. One-year medical outcomes and emergency department recidivism after emergency department observation for cocaine-associated chest pain. Ann Emerg Med. 2009; 53:31020.
  • 10
    Collin MJ, Weisenthal B, Walsh KM, McCusker CM, Shofer FS, Hollander JE. Young patients with chest pain: 1-year outcomes. Am J Emerg Med. 2010; epub Mar 24.
  • 11
    Weisenthal BM, Chang AM, Walsh KM, Collin MJ, Shofer FS, Hollander JE. Relation between thrombolysis in myocardial infarction risk score and one-year outcomes for patients presenting at the emergency department with potential acute coronary syndrome. Am J Cardiol. 2010; 105:4414.
  • 12
    Hollander JE, Chang AM, Shofer FS, McCusker CM, Baxt WG, Litt HI. Coronary computed tomographic angiography for rapid discharge of low-risk patients with potential acute coronary syndromes. Ann Emerg Med. 2009; 53:295304.
  • 13
    Hollander JE, Shih RD, Hoffman RS, et al. Predictors of coronary artery disease in patients with cocaine-associated myocardial infarction. Cocaine-Associated Myocardial Infarction (CAMI) Study Group. Am J Med. 1997; 102:15863.
  • 14
    Kontos MC, Jesse RL, Tatum JL, Ornato JP. Coronary angiographic findings in patients with cocaine-associated chest pain. J Emerg Med. 2003; 24:913.
  • 15
    Dressler FA, Malekzadeh S, Roberts WC. Quantitative analysis of amounts of coronary arterial narrowing in cocaine addicts. Am J Cardiol. 1990; 65:3038.
  • 16
    Tardiff K, Gross E, Wu J, Stajic M, Millman R. Analysis of cocaine-positive fatalities. J Forensic Sci. 1989; 34:5363.
  • 17
    Sholar MB, Mendelson JH, Mello NK, et al. Concurrent pharmacokinetic analysis of plasma cocaine and adrenocorticotropic hormone in men. J Clin Endocrinol Metab. 1998; 83:9668.
  • 18
    Walsh SL, Stoops WW, Moody DE, Lin SN, Bigelow GE. Repeated dosing with oral cocaine in humans: assessment of direct effects, withdrawal, and pharmacokinetics. Exp Clin Psychopharmacol. 2009; 17: 20516.
  • 19
    Walsh K, Chang AM, Perrone J, et al. Coronary computerized tomography angiography for rapid discharge of low-risk patients with cocaine-associated chest pain. J Med Toxicol. 2009; 5:1119.
  • 20
    Apple FS, Wu AH, Jaffe AS. European Society of Cardiology and American College of Cardiology guidelines for redefinition of myocardial infarction: how to use existing assays clinically and for clinical trials. Am Heart J. 2002; 144:9816.
  • 21
    Kabrhel C, Matts C, McNamara M, Katz J, Ptak T. A highly sensitive ELISA D-dimer increases testing but not diagnosis of pulmonary embolism. Acad Emerg Med. 2006; 13:51924.
  • 22
    Hollander JE, Hoffman RS, Gennis P, et al. Cocaine-associated chest pain: one-year follow-up. Acad Emerg Med. 1995; 2:17984.
  • 23
    Bernstein SL, Boudreaux ED, Cabral L, et al. Nicotine dependence, motivation to quit, and diagnosis among adult emergency department patients who smoke: a national survey. Nicotine Tob Res. 2008; 10:127782.
  • 24
    Crawford MJ, Patton R, Touquet R, Drummond C, Byford S, Barrett B, Reece B. Screening and referral for brief intervention of alcohol-misusing patients in an emergency department: a pragmatic randomised controlled trial. Lancet. 2004; 364:13349.
  • 25
    Johnson SB, Bradshaw CP, Wright JL, Haynie DL, Simons-Morton BG, Cheng TL. Characterizing the teachable moment: is an emergency department visit a teachable moment for intervention among assault-injured youth and their parents? Pediatr Emerg Care. 2007; 23:5539.