Cocaine and Chest Pain: Clinical Features and Outcome of Patients Hospitalized to Rule Out Myocardial Infarction Michael J. Gitter, MD; Steven R. Goldsmith, MD; David N. Dunbar, MD; and Scott W. Sharkey, MD
• Objective: To investigate the clinical features, electrocardiographic findings, and hospital course in patients admitted with acute chest pain temporally related to cocaine use. • Design: Retrospective data analysis. • Setting: A 485-bed county hospital. • Patients: One hundred and one consecutive patients with cocaine-related chest pain admitted to the hospital to rule out myocardial infarction. • Measurements and Main Results: The quality of the chest pain frequently suggested myocardial ischemia. Dyspnea was common (56%). The onset of chest pain occurred during cocaine use in 21 % of patients, within 1 hour of use in 37%, and after 1 hour of use in 42%. Admission electrocardiographic findings were interpreted as normal in 32% of patients; as acute myocardial injury in 8%; as early repolarization variant in 32%; as left ventricular hypertrophy in 16%; and as "other" in 12%. Forty-three percent of patients had ST-segment elevation meeting the electrocardiographic criteria for use of thrombolytic therapy, but such elevation was usually due to the early repolarization variant. The initial total creatine kinase was elevated more than 3.3 fxkat/L (200 U/L) in 43% of patients, and an elevated total creatine kinase was recorded at some time during the hospital course in 47% of patients. The creatine kinase MB fraction was less than 0.02 in all patients. Myocardial infarction was ruled out in all patients. No patient experienced in-hospital cardiovascular complications. • Conclusion: The quality of acute chest pain related to cocaine use is indistinguishable from that experienced in acute myocardial ischemia. Abnormal or normal variant electrocardiographic findings are common in patients with chest pain related to cocaine use, but nevertheless the incidence of acute myocardial infarction is low. The ST-segment and T-wave changes can mimic acute myocardial injury and are most likely normal findings in young black men that can be readily recognized in the emergency department. Most of these patients do not require admission to an intensive care unit.
C o c a i n e use is associated with acute myocardial infarction in young people (1-7). Cocaine induces both an increase in myocardial oxygen demand and a decrease in coronary blood flow (8). Although 5 million Americans are thought to use cocaine regularly, only 65 cases of cocaine-related acute myocardial infarction have been reported in the North American literature (9). Because of these reports, many patients with cocainerelated chest pain are admitted to the hospital for possible myocardial infarction (10-12). Our report describes the clinical features, electrocardiographic findings, and hospital course in consecutive patients with cocainerelated chest pain who were admitted to the hospital to rule out myocardial infarction. These observations have important implications for the management of such patients, particularly regarding the use of thrombolytic therapy. Methods Patients Hennepin County Medical Center is a 485-bed county hospital that provides acute care for patients from metropolitan Minneapolis, Minnesota. Its emergency department records 85 000 patient visits per year. Between January 1987 and August 1989, 101 consecutive patients with cocaine-related chest pain were admitted to the hospital via the emergency department to rule out myocardial infarction. All patients presented to the emergency department with acute chest pain related to cocaine use and were admitted to a monitored bed in either the cardiac care unit or the postcardiac rehabilitation center. These patients accounted for 1.4% of the admissions to these two units during this time period. A retrospective chart review was done for each admission. All emergency department and inhospital records were reviewed. Data gathered included symptoms at presentation, electrocardiographic findings, cardiac enzyme determinations, and hospital course. Patients with cocaine-related chest pain who were evaluated in the emergency department but sent home were not tracked. Urine Toxicology Toxicology screens were done at the discretion of the admitting physician and were carried out on urine samples obtained at admission. Urine was analyzed using an enzymemultiplied immunoassay technique that assessed the sample for the cocaine metabolite benzoylecgonine. Electrocardiograms
Annals of Internal Medicine. 1991;115:277-282. From Hennepin County Medical Center and the University of Minnesota, Minneapolis, Minnesota. For current author addresses, see end of text.
The initial electrocardiogram for each patient was interpreted by an experienced electrocardiographer who was blinded to the patient's identity and hospital outcome. Each electrocardiogram was analyzed to see if it met the Thrombolysis in Myocardial Infarction (TIMI) study criterion (ST-segment elevation of 0.1 mV or more in two contiguous leads) for administration of thrombolytic therapy. In addition, the electrocardiogram reader (cardiologist) placed each electrocardio©1991 American College of Physicians
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
277
Figure 1. Admission 12-lead electrocardiogram from a 27-year-old black man with cocaine-related chest pain showing the early repolarization pattern. The precordial leads show ST-segment elevation and peaked T waves simulating acute anterior myocardial infarction.
gram into one of five categories: normal; acute myocardial injury; early repolarization variant; left ventricular hypertrophy; and "other" (pericarditis, old infarction, nonspecific ST-T changes). The cardiologist's interpretation of the initial electrocardiogram was compared with the interpretation made by the emergency department staff at admission.
71 patients and were positive for cocaine metabolites in 62 (87%). Other drugs present included opiates (11 patients) and benzodiazapenes (4 patients).
Enzymes
Patients described their chest pain as pressure (46%), sharp pain (33%), or dull pain (20%). Pleuritic-type pain was present in 18% of patients. In 40% of patients, the pain radiated to the neck, shoulders, or upper extremities. The duration of the chest pain was less than 1 hour in 55% of patients, 1 to 3 hours in 32%, and more than 3 hours in 13%. The onset of chest pain occurred during cocaine use in 21% of patients, within 1 hour of use in 37%, and after 1 hour of use in 42%. Other symptoms commonly described included dyspnea (56%), diaphoresis (32%), nausea (28%), palpitations or tachycardia (14%), and weakness or paresthesias (7%).
Blood samples for analysis of total creatine kinase and creatine kinase MB fraction were obtained at admission and every 8 to 12 hours for 24 hours. Total creatine kinase activity was measured at 37 °C on a Paramax analyzer (Baxter Laboratories, Miami, Florida). The creatine kinase MB fraction was measured by agarose electrophoresis (CIBA Corning Diagnostic, Medfield, Massachusetts) and expressed as a percent of total creatine kinase. The diagnosis of acute myocardial infarction was made when a typical rise and fall of total creatine kinase occurred and was accompanied by a rise and fall of both the absolute and relative amounts of the creatine kinase MB isoenzyme (13). For diagnosis of acute myocardial infarction, the total creatine kinase had to be greater than 3.3 /Ltkat/L (200 U/L), and the creatine kinase MB fraction had to be more than 0.02. Total lactic dehydrogenase (LDH) and LDH isoenzyme determinations were made at the discretion of the admitting physician. The normal range for total LDH in our laboratory is 1.3 to 3.2 )Ltkat/L (80 to 190 U/L). Statistical Analysis Continuous variables are presented as mean ± SD. Analysis of variance was used to compare serial creatine kinase measurements. Chi-square analysis was used in the comparisons of categoric variables. Probability values are reported. Results The average age of the 101 patients was 31.5 years (range, 18-58). Seventy percent of patients were black, and 23% were female. Seventy-seven patients were admitted once, 9 patients were admitted twice, and 2 patients were admitted three times. Seventy-four percent of patients had two or more risk factors for atherosclerotic coronary artery disease. Only three patients reported a history of previous myocardial infarction. The route of cocaine use was smoking (60%), inhaling (10%), or intravenous injection (30%). The quantity of cocaine used (when documented) was less than 1 gram in 13% of patients, 1 to 2 grams in 63%, and more than 2 grams in 24%. Urine toxicology screens were done in 278
Chest Pain and Other Symptoms
Electrocardiograms The initial electrocardiograms from 100 of 101 patients were available for review by the cardiologist. These electrocardiograms were interpreted as follows: normal, 32 patients; acute myocardial injury, 8 patients; early repolarization variant, 32 patients; left ventricular hypertrophy, 16 patients; and "other," 12 patients. Sinus tachycardia (heart rate, 100/min or more) was present in 19 patients. Of the 100 patients, 43 met the TIMI electrocardiographic criteria for the administration of thrombolytic therapy (no patient actually received thrombolytic therapy). The ST-segment elevation was attributed to early repolarization variant (Figure 1) in 25 patients (58%), to left ventricular hypertrophy in 7 (16%), to acute myocardial injury (Figure 2) in 8 (19%), and to other causes (pericarditis, old infarction) in 3 (7%). Both ST-segment elevation and T-wave inversion were present in the 8 patients thought to have an acute myocardial injury pattern (Figure 2). Of these 8 patients, 6 had a twodimensional echocardiogram, which, in each case, revealed normal wall motion, and 1 had cardiac catheterization that revealed normal coronary arteries and normal wall motion. The electrocardiogram was normal in 19 of 78 men
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
(24%) and 13 of 22 women (59%) (P = 0.002). An early repolarization pattern was present in 31 of 78 men (40%) and in 1 of 22 women (5%) (P = 0.002). Findings consistent with left ventricular hypertrophy were present in 13 of 78 men (17%) and in 3 of 22 women (14%) (P > 0.2). Only 6 of these patients (38%) had a history of hypertension. An acute myocardial injury pattern was present in 8 of 78 men (10%) and in 0 of 22 women (0%) (P = 0.12). The electrocardiogram lead distribution of ST-segment elevation in the 43 patients who met the TIMI criteria for administration of thrombolytic therapy is shown in Table 1. The precordial leads V2 and V3 were involved in 88% and 86% of these patients, respectively. The limb leads were rarely involved. Seven patients showed ST-segment depression of 0.1 mV or more. In 6 patients, the ST-segment depression occurred in the setting of left ventricular hypertrophy with repolarization changes. In 1 patient, the ST-segment depression was unexplained and did not resolve on subsequent electrocardiograms. Fifty-nine patients had serial electrocardiograms that were available for analysis. Of these 59 patients, 51 showed no electrocardiographic evolution. Eight patients showed new or increased T-wave inversion. No patient developed Q waves. Interpretation of the initial electrocardiogram by the emergency department is compared with the interpretation by the cardiology department in Table 2. The emergency department provided no written interpretation for 14 electrocardiograms. The emergency department interpreted 21 electrocardiograms as normal, and no electrocardiograms were interpreted as representing acute myocardial injury. Creatine Kinase Enzymes Creatine kinase enzyme measurements were available in 98 of the 101 patients. Ten patients (10%) had only a single creatine kinase determination. Forty-one patients had two creatine kinase determinations, and 47 patients had three or more creatine kinase determinations. For the entire 98 patients, the mean initial creatine kinase level was 7.2 /xkat/L (range, 0.9 to 124.4 /xkat/L) and the peak creatine kinase level was 7.9 /xkat/L (range,
Table 1. Electrocardiographic Leads with ST-Segment Elevation in 43 Patients Who Met the Thrombolysis in Myocardial Infarction Study Electrocardiographic Criteria for Administration of Thrombolytic Therapy* Electrocardiographic Lead
Patients, n (%)
Limb Leads 1
AVL AVR 2 3 AVF Precordial leads
v, v2 v3 v4 v5 v6
0 0 4(9) 1(2) 5(12) 12 (28) 38 (88) 37 (86) 14 (33) 9(21) 4(9)
* ST-segment elevation > 0.1 mV.
0.9 to 164.8 /xkat/L). The creatine kinase level was elevated 3.3 /xkat/L or more at some time during hospitalization in 46 (47%) patients. Nine patients (9%) had at least one creatine kinase measurement that exceeded 16.7 /xkat/L (1000 U/L). For the 88 patients who had serial creatine kinase measurements, the mean initial creatine kinase level was 7.3 ± 14.8 /xkat/L and the mean second creatine kinase level was 6.0 ± 13.5 /xkat/L (P = 0.004). The initial creatine kinase level was also the peak level in 76 of 88 (86%) patients who had serial measurements. The relation between the initial creatine kinase measurement and the second creatine kinase measurement is presented in Figure 3. However, data are not shown for 8 patients with a markedly elevated creatine kinase. The paired (initial and second) creatine kinase measurements for these patients are as follows: 30.4 and 18.8 /xkat/L; 18.5 and 11.8 /xkat/L; 124.4 and 118.5 /xkat/L; 8.7 and 19.3 /xkat/L; 42.0 and 26.0 /xkat/L; 19.4 and 26.1 /xkat/L; 21.3 and 11.8 /xkat/L; and 50.1 and 38.5 /xkat/L. Sixty-four patients had an LDH determination (mean, 2.9 ± 0.9 /xkat/L [range, 1.4 to 6.5 /xkat/L]). For the 9 patients with a total creatine kinase level exceeding 16.7 /xkat/L, the total LDH ranged from 1.9 to 5.2 /xkat/L (113 to 310
Figure 2. Admission 12-lead electrocardiogram from a 25-year-old black man with cocaine-related chest pain showing acute myocardial injury. The precordial leads V2 to V5 demonstrate ST-segment elevation and T-wave inversion simulating acute anterior myocardial infarction. Subsequent electrocardiograms showed no evolution of the ST-segment and T-wave changes. 15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
279
Table 2. Admission Electrocardiogram Interpretations by the Cardiology and Emergency Departments Interpretation Cardiology Department
Emergency Department
Normal (n = 32)
Normal: 15 (47%)* Anterior ST-segment elevation: 4 (13%) Poor R-wave progression or Q waves: 3 (9%) Nonspecific T-wave inversion: 3 (9%) Left venticular hypertrophy: 3 (9%) High QRS voltage: 1 (3%) Missing: 3 (9%)
Acute myocardial injury (n = 8)
Anterior ST-segment elevation and T-wave inversion: 4 (50%) Widespread T-wave inversions: 3 (37%) Missing: 1 (13%)
Early repolarization variant (n = 32)
ST-segment or J-point elevation: 10 (31%) Early repolarization variant: 6 (19%) Normal: 6(19%) Left ventricular hypertrophy: 4 (13%) Nonspecific ST-T wave changes; 3 (9%) Missing: 3 (9%)
Left ventricular hypertrophy (n - 16)
Left ventricular hypertrophy: 7 (44%) Nonspecific T-wave inversions: 3 (19%) ST-segment depression: 1 (6%) ST-segment elevation and T-wave inversion: 1 (6%) Nonspecific ST and T changes: 1 (6%) Missing: 3(19%)
"Other" (n = 12) Nonspecific T-wave changes: 5 (42%) (flat or inverted) Old inferior infarction: 3 (25%) Pericarditis: 2 (17%) Old anterior infarction: 1 (8%) Nonspecific ST-segment depression: 1 (8%)
Nonspecific T-wave inversion: 3 (25%) ST-segment elevation: 2 (17%) Nonspecific ST and T changes: 1 (8%) Inferior Q waves: 1 (8%) "Ischemia": 1 (8%) Missing: 4 (33%)
* Percent of expressed n for the cardiology interpretations.
U/L). The LDH isoenzyme distributions were measured in 2 patients whose peak creatine kinase values were 49.3 /xkat/L and 50.1 ^kat/L. The total LDH was 4.5 /xkat/L (LD,, 13%; LD2, 22%) and 3.6 /xkat/L (LD,, 17%, LD2, 27%), respectively, in these 2 patients. The creatine kinase MB fraction was less than 0.02 in all patients. No patient had enzymatic evidence of acute myocardial infarction and no patient developed in-hospital cardiovascular complications. Hospital Costs The average duration of hospital stay in our series of patients was 1.3 days (range, 1 to 6 days). The average hospital charge for each patient was $1860 (range, $960 to $6080) in 1987, $1840 (range, $990 to $5700) in 1988, and $2005 (range, $1255 to $6600) in 1989. Discussion We have described clinical features and outcomes in 101 consecutive patients with acute chest pain related to cocaine use who were admitted to a monitored hospital bed to rule out a myocardial infarction. In many of these patients, the quality of the chest pain was typical of myocardial ischemia (pressure with radiation into the neck or arms). These symptoms, together with many case reports describing an association between cocaine use and acute myocardial infarction, likely influenced 280
our emergency department physicians to liberally admit these patients to the hospital. Even though the chest pain was typical of myocardial ischemia and often lasted more than 1 hour, the incidence of acute myocardial infarction was 0% in this series, a rate considerably lower than the usual incidence (25% to 30%) of acute myocardial infarction in patients hospitalized with acute chest pain unrelated to cocaine use (14). In our own institution, the incidence of acute myocardial infarction in patients admitted with chest pain unrelated to cocaine use is 17%. The cause of the cocaine-related acute chest pain is unknown, but the pain is not necessarily due to myocardial ischemia. A recent study showed that administration of cocaine to humans causes an increase in heart rate and blood pressure and a simultaneous decrease in coronary blood flow (8). Despite these potentially deleterious physiologic effects, none of the patients in that study developed chest pain or electrocardiographic evidence of myocardial ischemia. Cocaine use can cause acute rhabdomyolysis, and it is possible that the chest pain is due to a toxic effect of the drug on thoracic skeletal muscle (15, 16). Acute chest pain associated with an elevated level of creatine kinase of skeletal muscle origin and thought to represent thoracic skeletal muscle injury has been reported in 6 patients (17). In our series, 47% of patients had an elevated skeletal muscle creatine kinase level. It is important to note that
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
the initial creatine kinase level was also the peak level in most (86%) of our patients. This pattern is not usually seen with acute myocardial infarction because the peak level is usually seen on the second or third sample obtained (that is, about 12 to 24 hours after admission) (18). Confirmed cases of cocaine-related acute myocardial infarction have been associated with a typical rise and fall in the serum creatine kinase level (5, 7). In our series, initial electrocardiographic findings may well have influenced the decision to admit the patient to the hospital. Only 21% of the initial electrocardiograms were interpreted as normal by the emergency department staff. The most common abnormalities noted by both the emergency department and cardiology staff were ST-segment elevation or T-wave inversion, or both. In one study of patients with cocaine-related medical problems, the presence of nonspecific electrocardiographic abnormalities was the only variable identified that distinguished admitted from nonadmitted patients (10). The high frequency of early repolarization changes in these patients is problematic because these changes can simulate acute anterior wall myocardial infarction (19-21). In our study, a trained electrocardiographer identified 8 of 100 patients with cocaine-related chest pain as having electrocardiograms consistent with an acute anterior myocardial infarction. In these 8 patients, electrocardiograms shared two common features: STsegment elevation in the precordial leads and terminal T-wave inversions in the precordial leads (Figure 2). This combination has also been reported with the early repolarization variant (20, 22, 23). The T-wave inversions are generally most marked in the midprecordial leads (V3 and V4) and may become more pronounced during excitement and hyperventilation (20, 24, 25). Myocardial ischemia is unlikely to be the cause of the electrocardiographic findings seen in our patients because multiple recordings showed no resolution of the ST-T wave changes over hours to days. Patients with the most abnormal electrocardiogram generally received a detailed cardiac evaluation and none had evidence of ischemic heart disease. Thus, ST-segment and T-wave changes must be interpreted cautiously in this patient population. The ratio of men to women was 3.4:1 in our series. The explanation for this imbalance is unknown. Between 1985 and 1987, the ratio of men to women evaluated in our emergency department with a primary diagnosis of acute cocaine intoxication was 2.3:1. It is possible that the incidence of cocaine-related chest pain is greater in men than in women. Gender seems to be an important variable with other complications of cocaine use. Seizures, for example, are more common in women (26), whereas most (93%) acute myocardial infarctions associated with cocaine use have occurred in men (9). The known greater frequency of early repolarization changes in men may have also selected men for admission to the hospital (20). Race was also an important variable in our study. Seventy percent of our patients were black. During a similar time period, only 23% of all patients evaluated in our emergency department with acute cocaine intoxication were black. Blacks may be more susceptible to cocaine-induced chest pain than whites. Furthermore, it
is well known that healthy young black men frequently exhibit electrocardiographic findings similar to those seen in our study (27-29). The presence of these electrocardiographic findings may have selected these patients for admission to the hospital. Certain limitations of our study require discussion. The exclusion of myocardial infarction in these patients was based on analysis of total creatine kinase and creatine kinase MB isoenzyme levels. Serial creatine kinase measurements were not done in ten patients. Although myocardial infarction may have been overlooked in these patients, all had a creatine kinase MB fraction of less than 0.02. Myocardial infarction may also have been missed in patients who had a markedly elevated total creatine kinase level due to rhabdomyolysis. In this setting, a rise in the creatine kinase MB fraction due to a small myocardial infarction could have been "hidden" by the very high total creatine kinase of skeletal muscle origin. Nine patients had a total creatine kinase level in excess of 16.7 /ikat/L. The total LDH was mildly elevated in four of these patients. The LDH isoenzyme distribution (obtained in two of the four patients) was normal. Nonetheless, myocardial infarction cannot be ruled out with certainty in these patients. Only 59% of our patients had serial electrocardiograms, and some patients may have had ischemic changes that escaped our detection. We were also unable to document the number of patients with cocaine-related chest
Figure 3. The relation between the initial and the second creatine kinase (CK) level for 80 patients. The paired creatine kinase measurements for 8 patients with markedly elevated levels are not shown (see text). The normal total creatine kinase level in our laboratory is 0 to 3.3 /xkat/L (0 to 200 U/L).
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4 Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
281
pain who were sent home from the emergency department. The outcome of these patients remains unknown. In a recent study done in two hospitals in Bronx, New York, Amin and colleagues (11) noted that 31% of acute myocardial infarctions were associated with the use of cocaine. However, the average time from cocaine use to the onset of chest pain was 18 hours. The apparent association between cocaine and acute myocardial infarction may simply reflect a high prevalence of cocaine use in the community and may not necessarily indicate a cause-and-effect relation. In many cardiac care units, a similar observation could be made concerning alcohol. These investigators also reported a several-fold increase in the number of hospital admissions for cocaine-related chest pain between 1985 and 1989. As in our study, Amin and colleagues (11) found that patients with cocaine-related chest pain but no myocardial infarction were young in age, had an elevated total creatine kinase level, and had onset of chest pain within 1 hour of cocaine use. In conclusion, although acute chest pain is a common complication of cocaine use, the incidence of acute myocardial infarction is very low when the initial electrocardiogram is normal or a normal variant. Any young patient presenting to the emergency department with acute chest pain should be questioned about cocaine use. Electrocardiographic findings can be the same as those seen with acute anterior myocardial infarction, but electrocardiographic abnormalities may represent normal variants observed in young black men rather than manifestations of myocardial ischemia. Most of these patients can be safely managed in a nonintensive care unit. Some may not require hospital admission. Thrombolytic therapy should be considered only when the diagnosis of acute myocardial infarction is unequivocal. Acknowledgments: The authors thank Tina Otto, Krista Groff, Maureen Adams, and James M. Kaufmann, PhD, for contributing to the preparation of this manuscript. Requests for Reprints: Scott W. Sharkey, MD, Hennepin County Medical Center, Cardiology Division, 701 Park Avenue, Minneapolis, MN 55415. Current Author Addresses: Dr. Gitter: Mayo Clinic, 200 First Street SW, Rochester, MN 55905. Drs. Goldsmith, Dunbar, and Sharkey: Hennepin County Medical Center, Cardiology Division, 701 Park Avenue, Minneapolis, MN 55415.
5.
6. 7. 8. 9. 10. 11. 12. 13.
14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
References 1. Cregler LL, Mark H. Medical complications of cocaine abuse. N Engl J Med. 1986;315:1495-500. 2. Pasternack PF, Colvin SB, Baumann FG. Cocaine-induced angina pectoris and acute myocardial infarction in patients younger than 40 years. Am J Cardiol. 1985;55:847. 3. Cregler LL, Mark H. Relation of acute myocardial infarction to cocaine abuse. Am J Cardiol. 1985;56:794. 4. Zimmerman FH, Gustafson GM, Kemp HG Jr. Recurrent myocar-
282
27. 28.
29.
dial infarction associated with cocaine abuse in a young man with normal coronary arteries: evidence for coronary artery spasm culminating in thrombosis. J Am Coll Cardiol. 1987;9:964-8. Smith HW 3d, Lieberman HA, Brody SL, Battey LL, Donahue BC, Morris DC. Acute myocardial infarction temporally related to cocaine use. Clinical, angiographic, and pathophysiological observations. Ann Intern Med. 1987;107:13-8. Mathias DW. Cocaine-associated myocardial ischemia. Review of clinical and angiographic findings. Am J Med. 1986;81:675-8. Isner JM, Estes NA 3d, Thompson PD, Costanzo-Nordin MR, Subramanian R, Mitter G, et al. Acute cardiac events temporally related to cocaine abuse. N Engl J Med. 1986;315:1438-43. Lange RA, Cigarroa RG, Yancy CW Jr, Willard JE, Popma JJ, Sills MN, et al. Cocaine-induced coronary artery vasoconstriction. N Engl J Med. 1989;321:1557-62. Lange RA, Flores EP, Digarroa RG, Hillis LD. Cocaine-induced myocardial ischemia and infarction. Cardiology. 1990;8:74-9. Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med. 1990;88:325-31. Amin M, Gabelman G, Karpel J, Buttrick P. Acute myocardial infarction and chest pain syndromes after cocaine use. Am J Cardiol. 1990;66:1434-7. Tokarski GF, Paganussi P, Urbanski R, Carden D, Foreback C, Tomlanovich MC. An evaluation of cocaine-induced chest pain. Ann Emerg Med. 1990;19:1088-92. Lee TH, Rouan GW, Weisberg MC, Brand DA, Cook EF, Acampora D, et al. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Med. 1987;106:181-6. Lee TH, Rouan GW, Weisberg MC, Brand DA, Acampora D, Stasiulewicz C, et al. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol. 1987;60:219-24. Herzhich BC, Arsura EL, Pagala M, Grob D. Rhabdomyolysis related to cocaine abuse. Ann Intern Med. 1988;109:335-6. Roth D, Alarcon FJ, Fernandez J A, Preston RA, Bourgoignie JJ. Acute rhabdomyolysis associated with cocaine intoxication. N Engl J Med. 1988;319:673-7. Rubin RB, Neugarten J. Cocaine-induced rhabdomyolysis masquerading as myocardial ischemia. Am J Med. 1989;86:551-3. Lee TH, Goldman L. Serum enzyme assays in the diagnosis of acute myocardial infarction: recommendations based on a quantitative analysis. Ann Intern Med. 1986;105:221-33. Osher HL, Wolff L. Electrocardiographic pattern simulating acute myocardial injury. Am J Med Sci. 1953;226:541-5. Goldberger AL. Myocardial Infarction: Electrocardiographic Differential Diagnosis. 3d edition. St. Louis, Missouri: C. V. Mosby; 1984:167-78, 242-7. Chelton LG, Burchell HB. Unusual RT segment deviations in the electrocardiograms of normal persons. Am J Med Sci. 1955;230:5460. Wiener L, Rios JC, Massumi RA. T-wave inversion with elevated RS-T segment simulating myocardial injury. Am Heart J. 1964;67: 684-8. Blackman NS, Kuskin L. Inverted T waves in the precordial electrocardiogram of normal adolescents. Am Heart J. 1964;67:304-12. Thompson P. The electrocardiogram in the hyperventilation syndrome. Am Heart J. 1943;25:372. Goldberger AL. Recognition of ECG pseudo-infarct patterns. Mod Concepts Cardiovasc Dis. 1980;49:13-8. Pascual-Leone A, Dhuna A, Altafullah I, Anderson DC. Cocaineinduced seizures. Neurology. 1990;40:404-7. Thomas J, Harris E, Lassiter G. Observations on the T wave and ST segment changes in the precordial electrocardiogram of 320 young Negro adults. Am J Cardiol. 1960;5:468-72. Littmann MC. Persistence of the juvenile pattern in the precordial leads of healthy adult Negroes with report of electrocardiographic survey on three hundred Negro and two hundred white subjects. Am Heart J. 1946;32:370-82. Goldman MJ. RS-T segment elevation in the mid and left precordial leads as a normal variant. Am Heart J. 1953;46:817-20.
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
• Objective: To investigate the clinical features, electrocardiographic findings, and hospital course in patients admitted with acute chest pain temporally related to cocaine use. • Design: Retrospective data analysis. • Setting: A 485-bed county hospital. • Patients: One hundred and one consecutive patients with cocaine-related chest pain admitted to the hospital to rule out myocardial infarction. • Measurements and Main Results: The quality of the chest pain frequently suggested myocardial ischemia. Dyspnea was common (56%). The onset of chest pain occurred during cocaine use in 21 % of patients, within 1 hour of use in 37%, and after 1 hour of use in 42%. Admission electrocardiographic findings were interpreted as normal in 32% of patients; as acute myocardial injury in 8%; as early repolarization variant in 32%; as left ventricular hypertrophy in 16%; and as "other" in 12%. Forty-three percent of patients had ST-segment elevation meeting the electrocardiographic criteria for use of thrombolytic therapy, but such elevation was usually due to the early repolarization variant. The initial total creatine kinase was elevated more than 3.3 fxkat/L (200 U/L) in 43% of patients, and an elevated total creatine kinase was recorded at some time during the hospital course in 47% of patients. The creatine kinase MB fraction was less than 0.02 in all patients. Myocardial infarction was ruled out in all patients. No patient experienced in-hospital cardiovascular complications. • Conclusion: The quality of acute chest pain related to cocaine use is indistinguishable from that experienced in acute myocardial ischemia. Abnormal or normal variant electrocardiographic findings are common in patients with chest pain related to cocaine use, but nevertheless the incidence of acute myocardial infarction is low. The ST-segment and T-wave changes can mimic acute myocardial injury and are most likely normal findings in young black men that can be readily recognized in the emergency department. Most of these patients do not require admission to an intensive care unit.
C o c a i n e use is associated with acute myocardial infarction in young people (1-7). Cocaine induces both an increase in myocardial oxygen demand and a decrease in coronary blood flow (8). Although 5 million Americans are thought to use cocaine regularly, only 65 cases of cocaine-related acute myocardial infarction have been reported in the North American literature (9). Because of these reports, many patients with cocainerelated chest pain are admitted to the hospital for possible myocardial infarction (10-12). Our report describes the clinical features, electrocardiographic findings, and hospital course in consecutive patients with cocainerelated chest pain who were admitted to the hospital to rule out myocardial infarction. These observations have important implications for the management of such patients, particularly regarding the use of thrombolytic therapy. Methods Patients Hennepin County Medical Center is a 485-bed county hospital that provides acute care for patients from metropolitan Minneapolis, Minnesota. Its emergency department records 85 000 patient visits per year. Between January 1987 and August 1989, 101 consecutive patients with cocaine-related chest pain were admitted to the hospital via the emergency department to rule out myocardial infarction. All patients presented to the emergency department with acute chest pain related to cocaine use and were admitted to a monitored bed in either the cardiac care unit or the postcardiac rehabilitation center. These patients accounted for 1.4% of the admissions to these two units during this time period. A retrospective chart review was done for each admission. All emergency department and inhospital records were reviewed. Data gathered included symptoms at presentation, electrocardiographic findings, cardiac enzyme determinations, and hospital course. Patients with cocaine-related chest pain who were evaluated in the emergency department but sent home were not tracked. Urine Toxicology Toxicology screens were done at the discretion of the admitting physician and were carried out on urine samples obtained at admission. Urine was analyzed using an enzymemultiplied immunoassay technique that assessed the sample for the cocaine metabolite benzoylecgonine. Electrocardiograms
Annals of Internal Medicine. 1991;115:277-282. From Hennepin County Medical Center and the University of Minnesota, Minneapolis, Minnesota. For current author addresses, see end of text.
The initial electrocardiogram for each patient was interpreted by an experienced electrocardiographer who was blinded to the patient's identity and hospital outcome. Each electrocardiogram was analyzed to see if it met the Thrombolysis in Myocardial Infarction (TIMI) study criterion (ST-segment elevation of 0.1 mV or more in two contiguous leads) for administration of thrombolytic therapy. In addition, the electrocardiogram reader (cardiologist) placed each electrocardio©1991 American College of Physicians
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
277
Figure 1. Admission 12-lead electrocardiogram from a 27-year-old black man with cocaine-related chest pain showing the early repolarization pattern. The precordial leads show ST-segment elevation and peaked T waves simulating acute anterior myocardial infarction.
gram into one of five categories: normal; acute myocardial injury; early repolarization variant; left ventricular hypertrophy; and "other" (pericarditis, old infarction, nonspecific ST-T changes). The cardiologist's interpretation of the initial electrocardiogram was compared with the interpretation made by the emergency department staff at admission.
71 patients and were positive for cocaine metabolites in 62 (87%). Other drugs present included opiates (11 patients) and benzodiazapenes (4 patients).
Enzymes
Patients described their chest pain as pressure (46%), sharp pain (33%), or dull pain (20%). Pleuritic-type pain was present in 18% of patients. In 40% of patients, the pain radiated to the neck, shoulders, or upper extremities. The duration of the chest pain was less than 1 hour in 55% of patients, 1 to 3 hours in 32%, and more than 3 hours in 13%. The onset of chest pain occurred during cocaine use in 21% of patients, within 1 hour of use in 37%, and after 1 hour of use in 42%. Other symptoms commonly described included dyspnea (56%), diaphoresis (32%), nausea (28%), palpitations or tachycardia (14%), and weakness or paresthesias (7%).
Blood samples for analysis of total creatine kinase and creatine kinase MB fraction were obtained at admission and every 8 to 12 hours for 24 hours. Total creatine kinase activity was measured at 37 °C on a Paramax analyzer (Baxter Laboratories, Miami, Florida). The creatine kinase MB fraction was measured by agarose electrophoresis (CIBA Corning Diagnostic, Medfield, Massachusetts) and expressed as a percent of total creatine kinase. The diagnosis of acute myocardial infarction was made when a typical rise and fall of total creatine kinase occurred and was accompanied by a rise and fall of both the absolute and relative amounts of the creatine kinase MB isoenzyme (13). For diagnosis of acute myocardial infarction, the total creatine kinase had to be greater than 3.3 /Ltkat/L (200 U/L), and the creatine kinase MB fraction had to be more than 0.02. Total lactic dehydrogenase (LDH) and LDH isoenzyme determinations were made at the discretion of the admitting physician. The normal range for total LDH in our laboratory is 1.3 to 3.2 )Ltkat/L (80 to 190 U/L). Statistical Analysis Continuous variables are presented as mean ± SD. Analysis of variance was used to compare serial creatine kinase measurements. Chi-square analysis was used in the comparisons of categoric variables. Probability values are reported. Results The average age of the 101 patients was 31.5 years (range, 18-58). Seventy percent of patients were black, and 23% were female. Seventy-seven patients were admitted once, 9 patients were admitted twice, and 2 patients were admitted three times. Seventy-four percent of patients had two or more risk factors for atherosclerotic coronary artery disease. Only three patients reported a history of previous myocardial infarction. The route of cocaine use was smoking (60%), inhaling (10%), or intravenous injection (30%). The quantity of cocaine used (when documented) was less than 1 gram in 13% of patients, 1 to 2 grams in 63%, and more than 2 grams in 24%. Urine toxicology screens were done in 278
Chest Pain and Other Symptoms
Electrocardiograms The initial electrocardiograms from 100 of 101 patients were available for review by the cardiologist. These electrocardiograms were interpreted as follows: normal, 32 patients; acute myocardial injury, 8 patients; early repolarization variant, 32 patients; left ventricular hypertrophy, 16 patients; and "other," 12 patients. Sinus tachycardia (heart rate, 100/min or more) was present in 19 patients. Of the 100 patients, 43 met the TIMI electrocardiographic criteria for the administration of thrombolytic therapy (no patient actually received thrombolytic therapy). The ST-segment elevation was attributed to early repolarization variant (Figure 1) in 25 patients (58%), to left ventricular hypertrophy in 7 (16%), to acute myocardial injury (Figure 2) in 8 (19%), and to other causes (pericarditis, old infarction) in 3 (7%). Both ST-segment elevation and T-wave inversion were present in the 8 patients thought to have an acute myocardial injury pattern (Figure 2). Of these 8 patients, 6 had a twodimensional echocardiogram, which, in each case, revealed normal wall motion, and 1 had cardiac catheterization that revealed normal coronary arteries and normal wall motion. The electrocardiogram was normal in 19 of 78 men
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
(24%) and 13 of 22 women (59%) (P = 0.002). An early repolarization pattern was present in 31 of 78 men (40%) and in 1 of 22 women (5%) (P = 0.002). Findings consistent with left ventricular hypertrophy were present in 13 of 78 men (17%) and in 3 of 22 women (14%) (P > 0.2). Only 6 of these patients (38%) had a history of hypertension. An acute myocardial injury pattern was present in 8 of 78 men (10%) and in 0 of 22 women (0%) (P = 0.12). The electrocardiogram lead distribution of ST-segment elevation in the 43 patients who met the TIMI criteria for administration of thrombolytic therapy is shown in Table 1. The precordial leads V2 and V3 were involved in 88% and 86% of these patients, respectively. The limb leads were rarely involved. Seven patients showed ST-segment depression of 0.1 mV or more. In 6 patients, the ST-segment depression occurred in the setting of left ventricular hypertrophy with repolarization changes. In 1 patient, the ST-segment depression was unexplained and did not resolve on subsequent electrocardiograms. Fifty-nine patients had serial electrocardiograms that were available for analysis. Of these 59 patients, 51 showed no electrocardiographic evolution. Eight patients showed new or increased T-wave inversion. No patient developed Q waves. Interpretation of the initial electrocardiogram by the emergency department is compared with the interpretation by the cardiology department in Table 2. The emergency department provided no written interpretation for 14 electrocardiograms. The emergency department interpreted 21 electrocardiograms as normal, and no electrocardiograms were interpreted as representing acute myocardial injury. Creatine Kinase Enzymes Creatine kinase enzyme measurements were available in 98 of the 101 patients. Ten patients (10%) had only a single creatine kinase determination. Forty-one patients had two creatine kinase determinations, and 47 patients had three or more creatine kinase determinations. For the entire 98 patients, the mean initial creatine kinase level was 7.2 /xkat/L (range, 0.9 to 124.4 /xkat/L) and the peak creatine kinase level was 7.9 /xkat/L (range,
Table 1. Electrocardiographic Leads with ST-Segment Elevation in 43 Patients Who Met the Thrombolysis in Myocardial Infarction Study Electrocardiographic Criteria for Administration of Thrombolytic Therapy* Electrocardiographic Lead
Patients, n (%)
Limb Leads 1
AVL AVR 2 3 AVF Precordial leads
v, v2 v3 v4 v5 v6
0 0 4(9) 1(2) 5(12) 12 (28) 38 (88) 37 (86) 14 (33) 9(21) 4(9)
* ST-segment elevation > 0.1 mV.
0.9 to 164.8 /xkat/L). The creatine kinase level was elevated 3.3 /xkat/L or more at some time during hospitalization in 46 (47%) patients. Nine patients (9%) had at least one creatine kinase measurement that exceeded 16.7 /xkat/L (1000 U/L). For the 88 patients who had serial creatine kinase measurements, the mean initial creatine kinase level was 7.3 ± 14.8 /xkat/L and the mean second creatine kinase level was 6.0 ± 13.5 /xkat/L (P = 0.004). The initial creatine kinase level was also the peak level in 76 of 88 (86%) patients who had serial measurements. The relation between the initial creatine kinase measurement and the second creatine kinase measurement is presented in Figure 3. However, data are not shown for 8 patients with a markedly elevated creatine kinase. The paired (initial and second) creatine kinase measurements for these patients are as follows: 30.4 and 18.8 /xkat/L; 18.5 and 11.8 /xkat/L; 124.4 and 118.5 /xkat/L; 8.7 and 19.3 /xkat/L; 42.0 and 26.0 /xkat/L; 19.4 and 26.1 /xkat/L; 21.3 and 11.8 /xkat/L; and 50.1 and 38.5 /xkat/L. Sixty-four patients had an LDH determination (mean, 2.9 ± 0.9 /xkat/L [range, 1.4 to 6.5 /xkat/L]). For the 9 patients with a total creatine kinase level exceeding 16.7 /xkat/L, the total LDH ranged from 1.9 to 5.2 /xkat/L (113 to 310
Figure 2. Admission 12-lead electrocardiogram from a 25-year-old black man with cocaine-related chest pain showing acute myocardial injury. The precordial leads V2 to V5 demonstrate ST-segment elevation and T-wave inversion simulating acute anterior myocardial infarction. Subsequent electrocardiograms showed no evolution of the ST-segment and T-wave changes. 15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
279
Table 2. Admission Electrocardiogram Interpretations by the Cardiology and Emergency Departments Interpretation Cardiology Department
Emergency Department
Normal (n = 32)
Normal: 15 (47%)* Anterior ST-segment elevation: 4 (13%) Poor R-wave progression or Q waves: 3 (9%) Nonspecific T-wave inversion: 3 (9%) Left venticular hypertrophy: 3 (9%) High QRS voltage: 1 (3%) Missing: 3 (9%)
Acute myocardial injury (n = 8)
Anterior ST-segment elevation and T-wave inversion: 4 (50%) Widespread T-wave inversions: 3 (37%) Missing: 1 (13%)
Early repolarization variant (n = 32)
ST-segment or J-point elevation: 10 (31%) Early repolarization variant: 6 (19%) Normal: 6(19%) Left ventricular hypertrophy: 4 (13%) Nonspecific ST-T wave changes; 3 (9%) Missing: 3 (9%)
Left ventricular hypertrophy (n - 16)
Left ventricular hypertrophy: 7 (44%) Nonspecific T-wave inversions: 3 (19%) ST-segment depression: 1 (6%) ST-segment elevation and T-wave inversion: 1 (6%) Nonspecific ST and T changes: 1 (6%) Missing: 3(19%)
"Other" (n = 12) Nonspecific T-wave changes: 5 (42%) (flat or inverted) Old inferior infarction: 3 (25%) Pericarditis: 2 (17%) Old anterior infarction: 1 (8%) Nonspecific ST-segment depression: 1 (8%)
Nonspecific T-wave inversion: 3 (25%) ST-segment elevation: 2 (17%) Nonspecific ST and T changes: 1 (8%) Inferior Q waves: 1 (8%) "Ischemia": 1 (8%) Missing: 4 (33%)
* Percent of expressed n for the cardiology interpretations.
U/L). The LDH isoenzyme distributions were measured in 2 patients whose peak creatine kinase values were 49.3 /xkat/L and 50.1 ^kat/L. The total LDH was 4.5 /xkat/L (LD,, 13%; LD2, 22%) and 3.6 /xkat/L (LD,, 17%, LD2, 27%), respectively, in these 2 patients. The creatine kinase MB fraction was less than 0.02 in all patients. No patient had enzymatic evidence of acute myocardial infarction and no patient developed in-hospital cardiovascular complications. Hospital Costs The average duration of hospital stay in our series of patients was 1.3 days (range, 1 to 6 days). The average hospital charge for each patient was $1860 (range, $960 to $6080) in 1987, $1840 (range, $990 to $5700) in 1988, and $2005 (range, $1255 to $6600) in 1989. Discussion We have described clinical features and outcomes in 101 consecutive patients with acute chest pain related to cocaine use who were admitted to a monitored hospital bed to rule out a myocardial infarction. In many of these patients, the quality of the chest pain was typical of myocardial ischemia (pressure with radiation into the neck or arms). These symptoms, together with many case reports describing an association between cocaine use and acute myocardial infarction, likely influenced 280
our emergency department physicians to liberally admit these patients to the hospital. Even though the chest pain was typical of myocardial ischemia and often lasted more than 1 hour, the incidence of acute myocardial infarction was 0% in this series, a rate considerably lower than the usual incidence (25% to 30%) of acute myocardial infarction in patients hospitalized with acute chest pain unrelated to cocaine use (14). In our own institution, the incidence of acute myocardial infarction in patients admitted with chest pain unrelated to cocaine use is 17%. The cause of the cocaine-related acute chest pain is unknown, but the pain is not necessarily due to myocardial ischemia. A recent study showed that administration of cocaine to humans causes an increase in heart rate and blood pressure and a simultaneous decrease in coronary blood flow (8). Despite these potentially deleterious physiologic effects, none of the patients in that study developed chest pain or electrocardiographic evidence of myocardial ischemia. Cocaine use can cause acute rhabdomyolysis, and it is possible that the chest pain is due to a toxic effect of the drug on thoracic skeletal muscle (15, 16). Acute chest pain associated with an elevated level of creatine kinase of skeletal muscle origin and thought to represent thoracic skeletal muscle injury has been reported in 6 patients (17). In our series, 47% of patients had an elevated skeletal muscle creatine kinase level. It is important to note that
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
the initial creatine kinase level was also the peak level in most (86%) of our patients. This pattern is not usually seen with acute myocardial infarction because the peak level is usually seen on the second or third sample obtained (that is, about 12 to 24 hours after admission) (18). Confirmed cases of cocaine-related acute myocardial infarction have been associated with a typical rise and fall in the serum creatine kinase level (5, 7). In our series, initial electrocardiographic findings may well have influenced the decision to admit the patient to the hospital. Only 21% of the initial electrocardiograms were interpreted as normal by the emergency department staff. The most common abnormalities noted by both the emergency department and cardiology staff were ST-segment elevation or T-wave inversion, or both. In one study of patients with cocaine-related medical problems, the presence of nonspecific electrocardiographic abnormalities was the only variable identified that distinguished admitted from nonadmitted patients (10). The high frequency of early repolarization changes in these patients is problematic because these changes can simulate acute anterior wall myocardial infarction (19-21). In our study, a trained electrocardiographer identified 8 of 100 patients with cocaine-related chest pain as having electrocardiograms consistent with an acute anterior myocardial infarction. In these 8 patients, electrocardiograms shared two common features: STsegment elevation in the precordial leads and terminal T-wave inversions in the precordial leads (Figure 2). This combination has also been reported with the early repolarization variant (20, 22, 23). The T-wave inversions are generally most marked in the midprecordial leads (V3 and V4) and may become more pronounced during excitement and hyperventilation (20, 24, 25). Myocardial ischemia is unlikely to be the cause of the electrocardiographic findings seen in our patients because multiple recordings showed no resolution of the ST-T wave changes over hours to days. Patients with the most abnormal electrocardiogram generally received a detailed cardiac evaluation and none had evidence of ischemic heart disease. Thus, ST-segment and T-wave changes must be interpreted cautiously in this patient population. The ratio of men to women was 3.4:1 in our series. The explanation for this imbalance is unknown. Between 1985 and 1987, the ratio of men to women evaluated in our emergency department with a primary diagnosis of acute cocaine intoxication was 2.3:1. It is possible that the incidence of cocaine-related chest pain is greater in men than in women. Gender seems to be an important variable with other complications of cocaine use. Seizures, for example, are more common in women (26), whereas most (93%) acute myocardial infarctions associated with cocaine use have occurred in men (9). The known greater frequency of early repolarization changes in men may have also selected men for admission to the hospital (20). Race was also an important variable in our study. Seventy percent of our patients were black. During a similar time period, only 23% of all patients evaluated in our emergency department with acute cocaine intoxication were black. Blacks may be more susceptible to cocaine-induced chest pain than whites. Furthermore, it
is well known that healthy young black men frequently exhibit electrocardiographic findings similar to those seen in our study (27-29). The presence of these electrocardiographic findings may have selected these patients for admission to the hospital. Certain limitations of our study require discussion. The exclusion of myocardial infarction in these patients was based on analysis of total creatine kinase and creatine kinase MB isoenzyme levels. Serial creatine kinase measurements were not done in ten patients. Although myocardial infarction may have been overlooked in these patients, all had a creatine kinase MB fraction of less than 0.02. Myocardial infarction may also have been missed in patients who had a markedly elevated total creatine kinase level due to rhabdomyolysis. In this setting, a rise in the creatine kinase MB fraction due to a small myocardial infarction could have been "hidden" by the very high total creatine kinase of skeletal muscle origin. Nine patients had a total creatine kinase level in excess of 16.7 /ikat/L. The total LDH was mildly elevated in four of these patients. The LDH isoenzyme distribution (obtained in two of the four patients) was normal. Nonetheless, myocardial infarction cannot be ruled out with certainty in these patients. Only 59% of our patients had serial electrocardiograms, and some patients may have had ischemic changes that escaped our detection. We were also unable to document the number of patients with cocaine-related chest
Figure 3. The relation between the initial and the second creatine kinase (CK) level for 80 patients. The paired creatine kinase measurements for 8 patients with markedly elevated levels are not shown (see text). The normal total creatine kinase level in our laboratory is 0 to 3.3 /xkat/L (0 to 200 U/L).
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4 Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
281
pain who were sent home from the emergency department. The outcome of these patients remains unknown. In a recent study done in two hospitals in Bronx, New York, Amin and colleagues (11) noted that 31% of acute myocardial infarctions were associated with the use of cocaine. However, the average time from cocaine use to the onset of chest pain was 18 hours. The apparent association between cocaine and acute myocardial infarction may simply reflect a high prevalence of cocaine use in the community and may not necessarily indicate a cause-and-effect relation. In many cardiac care units, a similar observation could be made concerning alcohol. These investigators also reported a several-fold increase in the number of hospital admissions for cocaine-related chest pain between 1985 and 1989. As in our study, Amin and colleagues (11) found that patients with cocaine-related chest pain but no myocardial infarction were young in age, had an elevated total creatine kinase level, and had onset of chest pain within 1 hour of cocaine use. In conclusion, although acute chest pain is a common complication of cocaine use, the incidence of acute myocardial infarction is very low when the initial electrocardiogram is normal or a normal variant. Any young patient presenting to the emergency department with acute chest pain should be questioned about cocaine use. Electrocardiographic findings can be the same as those seen with acute anterior myocardial infarction, but electrocardiographic abnormalities may represent normal variants observed in young black men rather than manifestations of myocardial ischemia. Most of these patients can be safely managed in a nonintensive care unit. Some may not require hospital admission. Thrombolytic therapy should be considered only when the diagnosis of acute myocardial infarction is unequivocal. Acknowledgments: The authors thank Tina Otto, Krista Groff, Maureen Adams, and James M. Kaufmann, PhD, for contributing to the preparation of this manuscript. Requests for Reprints: Scott W. Sharkey, MD, Hennepin County Medical Center, Cardiology Division, 701 Park Avenue, Minneapolis, MN 55415. Current Author Addresses: Dr. Gitter: Mayo Clinic, 200 First Street SW, Rochester, MN 55905. Drs. Goldsmith, Dunbar, and Sharkey: Hennepin County Medical Center, Cardiology Division, 701 Park Avenue, Minneapolis, MN 55415.
5.
6. 7. 8. 9. 10. 11. 12. 13.
14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.
References 1. Cregler LL, Mark H. Medical complications of cocaine abuse. N Engl J Med. 1986;315:1495-500. 2. Pasternack PF, Colvin SB, Baumann FG. Cocaine-induced angina pectoris and acute myocardial infarction in patients younger than 40 years. Am J Cardiol. 1985;55:847. 3. Cregler LL, Mark H. Relation of acute myocardial infarction to cocaine abuse. Am J Cardiol. 1985;56:794. 4. Zimmerman FH, Gustafson GM, Kemp HG Jr. Recurrent myocar-
282
27. 28.
29.
dial infarction associated with cocaine abuse in a young man with normal coronary arteries: evidence for coronary artery spasm culminating in thrombosis. J Am Coll Cardiol. 1987;9:964-8. Smith HW 3d, Lieberman HA, Brody SL, Battey LL, Donahue BC, Morris DC. Acute myocardial infarction temporally related to cocaine use. Clinical, angiographic, and pathophysiological observations. Ann Intern Med. 1987;107:13-8. Mathias DW. Cocaine-associated myocardial ischemia. Review of clinical and angiographic findings. Am J Med. 1986;81:675-8. Isner JM, Estes NA 3d, Thompson PD, Costanzo-Nordin MR, Subramanian R, Mitter G, et al. Acute cardiac events temporally related to cocaine abuse. N Engl J Med. 1986;315:1438-43. Lange RA, Cigarroa RG, Yancy CW Jr, Willard JE, Popma JJ, Sills MN, et al. Cocaine-induced coronary artery vasoconstriction. N Engl J Med. 1989;321:1557-62. Lange RA, Flores EP, Digarroa RG, Hillis LD. Cocaine-induced myocardial ischemia and infarction. Cardiology. 1990;8:74-9. Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med. 1990;88:325-31. Amin M, Gabelman G, Karpel J, Buttrick P. Acute myocardial infarction and chest pain syndromes after cocaine use. Am J Cardiol. 1990;66:1434-7. Tokarski GF, Paganussi P, Urbanski R, Carden D, Foreback C, Tomlanovich MC. An evaluation of cocaine-induced chest pain. Ann Emerg Med. 1990;19:1088-92. Lee TH, Rouan GW, Weisberg MC, Brand DA, Cook EF, Acampora D, et al. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Med. 1987;106:181-6. Lee TH, Rouan GW, Weisberg MC, Brand DA, Acampora D, Stasiulewicz C, et al. Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol. 1987;60:219-24. Herzhich BC, Arsura EL, Pagala M, Grob D. Rhabdomyolysis related to cocaine abuse. Ann Intern Med. 1988;109:335-6. Roth D, Alarcon FJ, Fernandez J A, Preston RA, Bourgoignie JJ. Acute rhabdomyolysis associated with cocaine intoxication. N Engl J Med. 1988;319:673-7. Rubin RB, Neugarten J. Cocaine-induced rhabdomyolysis masquerading as myocardial ischemia. Am J Med. 1989;86:551-3. Lee TH, Goldman L. Serum enzyme assays in the diagnosis of acute myocardial infarction: recommendations based on a quantitative analysis. Ann Intern Med. 1986;105:221-33. Osher HL, Wolff L. Electrocardiographic pattern simulating acute myocardial injury. Am J Med Sci. 1953;226:541-5. Goldberger AL. Myocardial Infarction: Electrocardiographic Differential Diagnosis. 3d edition. St. Louis, Missouri: C. V. Mosby; 1984:167-78, 242-7. Chelton LG, Burchell HB. Unusual RT segment deviations in the electrocardiograms of normal persons. Am J Med Sci. 1955;230:5460. Wiener L, Rios JC, Massumi RA. T-wave inversion with elevated RS-T segment simulating myocardial injury. Am Heart J. 1964;67: 684-8. Blackman NS, Kuskin L. Inverted T waves in the precordial electrocardiogram of normal adolescents. Am Heart J. 1964;67:304-12. Thompson P. The electrocardiogram in the hyperventilation syndrome. Am Heart J. 1943;25:372. Goldberger AL. Recognition of ECG pseudo-infarct patterns. Mod Concepts Cardiovasc Dis. 1980;49:13-8. Pascual-Leone A, Dhuna A, Altafullah I, Anderson DC. Cocaineinduced seizures. Neurology. 1990;40:404-7. Thomas J, Harris E, Lassiter G. Observations on the T wave and ST segment changes in the precordial electrocardiogram of 320 young Negro adults. Am J Cardiol. 1960;5:468-72. Littmann MC. Persistence of the juvenile pattern in the precordial leads of healthy adult Negroes with report of electrocardiographic survey on three hundred Negro and two hundred white subjects. Am Heart J. 1946;32:370-82. Goldman MJ. RS-T segment elevation in the mid and left precordial leads as a normal variant. Am Heart J. 1953;46:817-20.
15 August 1991 • Annals of Internal Medicine • Volume 115 • Number 4
Downloaded From: https://annals.org/ by a Glasgow Univ Library User on 09/17/2018
