“Abnormal” electrocardiograms in patients with cocaine-associated chest pain are due to “normal” variants

The Journal

of Emergency

Medicine,

Vol12.

199-205.1994 sciel& Ltd Printed in the USA. AU rights racrvcd 0736-4679/94$6.00 + .oo

cop&ht

Pergamon

No 2.

DD

0 i9!34tier

07364679(93)E4012-7

Weeted TcqpicS

Cardiobgy

Commentary

“ABNORMAL” ELECTROCARDIOGRAMS IN PATIENTS WITH COCAINE-ASSOCIATED CHEST PAIN ARE DUE TO “NORMAL” VARIANTS Judd E. Hollander, MD,*t Michael Lozano, MD,t Phillip Fairweather, Paul Gennis, MD,t Gerard X. Brogan, MD, FACEP,* David Cooling, and E. John Gallagher, MD, FACEPt

rvm,t Evan Goldstein, hm,t Henry C. Thode, Pm,*

MD,*

*University Hospital, SUNY Health Sciences Center, Stony Brook, and tBronx Municipal Hospital Center, Albert Einstein College of Medicine, Bronx, New York Reprint Address: Judd E. Hollander, MD, Department of Emergency Medicine, University Hospital, L4-515, SUNY, Stony Brook, Stony Brook, NY 11794-7400

0 Abstract -“Abnormal” electrouudiograms are found in 56% to 84k of p8t&!nt8 with cocaine-a88ociated clmt pain. TM8 8tudy wn8 ddgned to aseem whether these findings can be explabd by Ynormd” variations in young patients’ elcetroerrdloguns. This crom-aectional study waa couducted in a mtmidpal hospital emergency department and w8k-inclhk.EWtoryandreuultsofmektroa&ogram for consecutive p8thta with c-ted chest pain, aged 18 to 35 years, were compared to normal controls m8tched for age, race, rind gender. Bs underwent detdled adyda by two phyaichns blinded to both the 8tudy protocol and the hypothesis. Interpbynichn condiagno8i8 wa8 8ub8tMcord8nce for elcetrocudioyphk tid. There were 112 patienti enrolled, 56 in each group. There was no aig&hxmt difference found in tbe mean frequency of -raplie diagnoses between the co-ted chest pin patients aad controls. The eariy repohldon vari8nt wa8 common. In conclusion, %ormal” vnrhths (J point and ST segment elevations) account for many of the “abnormal” electrocardiograms observed ill young patient8 with coadne-amodated clle8t pain. Farther study is needed to define the prevalence of these “normal” varhtions, and to determine if standard deetnnrpdioorrpbk criteria for thrombolysis apply to young patients. Cl Keywords-cocaine; thromboly&; ek&oardi

INTRODUCTION

Recreationalcocaineuse has reachedepidemicproportionswithin the pastdecade.It hasbeenestimated that 25 to 40 million Americanshaveusedcocaineat leastonce(l), and that 5 million useit regularly(2). Consequently,emergencydepartment (ED) physicians have witnessedan increasein cocaine-related medical complaints (3), with chest pain accounting for up to 40010 of suchcomplaints(4). Myocardial infarction secondaryto cocainewas first reportedin 1982(5), and over 100caseshave appearedin the literature since (6,7). Three retrospective (8-10) and two prospective(11,12) series haveevaluatedpatientswith cocaine-associated chest pain and havefound an incidenceof myocardial infarction ranging from 0% to 31% (median = 6%). In patientspresentingto the ED with chestpain (unrelatedto cocaine),the initial electrocardiogramhas historically been consideredthe best predictor of myocardialinfarction (13). Electrocardiograms in patients with cocaineassociatedchestpain havebeenreportedto be abnormal in 56% to 84% of patients (8-10). An ST segment elevation has been noted in 32% to 37% of patientswith cocaine-associated chestpain (9,lO). In

angina; myoaudhl infarction; ogrsm; early repolarlzation

RECEIVED: 19MaylW3; FINALSUBMISSIONRECXIVBD: ACCEPTED: 27 September1993

20August1993;

199

200

J. E. Hollander et al.

oneseries(lo), 43% of patientsmet standardelectrocardiographiccriteria for thrombolytic administration, despitethe fact that none sustaineda myocardial infarction. Potential explanationsfor the high prevalenceof abnormalelectrocardiograms in the absenceof myocardialinfarction are a direct electrical effect of cocaineon the electrocardiogram,transient ischemia,or a high prevalenceof the earlyrepolarization variant in this typically younggroupof patients. This studywasdesignedto assesswhetherthe electrocardiogramsof patientswith cocaine-associated chest pain are significantly different from those of other youngpatients.

METHODS

The study group consistedof consecutivepatients, aged18to 35, who presentedto the Bronx Municipal Hospital ED or walk-in clinic with a chief complaint of chestpain unexplainedby radiographicabnormality or local trauma, and cocaineusewithin the past week. One week was chosenas our cutoff because cocaine-inducedmyocardial infarction has beenreported severaldays after most recent cocaine use (14), and cocaine-inducedST segmentelevationhas beenreportedfor up to severalweeksafter cessation of cocaineuse (15). Control patients consistedof medicalcenteremployeesandpatientswho presented to the ED or walk-in clinic. Patients were matched for age,race, and gender.Control patientswereexcluded from the study if they had a history of any heart disease,a seriousmedical problem requiring admission,any potentialcardiopulmonarycomplaint (i.e., chestpain, shortnessof breath),or if they admitted to cocaineusewithin the past week.All participantsprovidedinformed consent. Patientscompletedan anonymousquestionnaire. Demographicfeaturessuchasthe patient’sage,race, and genderwererecorded.The patient’schief complaint, medications,the presenceor absenceof cardiac risk factors (hypertension,hypercholesterolemia, diabetesmellitus, tobaccouse, family history of prematureatheroscleroticheart disease),and cocaineusewereobtained.All patientshad an electrocardiogram performed by a trained electrocardiographtechnicianor physicianusingthe Burdick E600 Electrocardiographsystem. The electrocardiogramswere separatedfrom the data sheets,and all identifying information was removed. Interpretationswereperformedby two experienced ED faculty members. The two physicians were from a different institution and were blinded

to both the study protocol and the hypothesis.The electrocardiogramswere evaluatedwith respectto rate, rhythm, axis, intervals, chamberenlargement, presenceor absenceof bundlebranchblock, ST segment elevation,ST segmentdepression,J point elevation, T waveinversion,and Q wavesin eachof the 12standardelectrocardiographicleads.Elevationsor depressionswere quantitatedin each lead. Patients with cocaine-associated chestpain and ST segment elevationshad repeatelectrocardiogramsto evaluate the persistenceof the ST segmentelevations.If the physicianmanagingthe casefelt the chestpain was potentially ischemic, the patient had serial cardiac isoenzymesto rule out myocardialinfarction. Each electrocardiogramwasclassifiedinto one of six categoriesusinga closedquestionformat: (a)normal (no electrocardiographicevidenceof ischemia); (b) nonspecific(accepteddeviation from the norm with the lowest likelihood of ischemia, such as an invertedT wavein III or sinustachycardia);(c) early repolarization variant (elevatedtakeoff of the ST segmentat the J point of the QRS complexgreater than 1 mm relativeto the isoelectricline); (d) abnormal but nondiagnosticof myocardial ischemia(prolonged PR, QRS, QTc intervals, bundle branch blocks, left ventricular hypertrophywith strain); (e) ischemia or prior infarction (ST depressionmore than 0.1 mV measured80 msec from the J point, invertedT wavesmore than 0.3 mV, or Q wavesat least 30 msec in duration); (f) suggestiveof acute myocardialinfarction (ST elevationgreaterthan 0.1 mV measured80 ms from the J point, with or without reciprocalST depressions). Two additional questionswere askedat the time of the electrocardiograminterpretation:(a) Is there evidencefor earlyrepolarizationabnormalitiesin this electrocardiogram?(b) Givena clinical scenarioconsistent with acute ischemia and this electrocardiogram, would you considerthe use of thrombolytics (in the absenceof contraindications)?TIM1 criteria (16)wereusedby the physiciansto evaluatethe presenceof thrombolytic criteria. Dichotomousanswers wererecordedfor eachof the two additional questions. Resultsof eachphysician’sevaluationarereported individually. Discrepanciesbetweenphysicianswere not resolvedto obtain a singlediagnosis.Interphysician concordancewas assessed by groupingthe electrocardiographicdiagnosesas follows: (a) normal, categorya; (b) nondiagnostic,categoriesb through d; and (c) ischemic, categoriese and f. The mean frequenciesreported are the averagepercentageof the specifiedvariablesbetweenthe two physicians. Individual valuesarereportedin tabular form.

Cocaine-Associated EKG “Abnormalities”

201

came-associated chestpain had heart rateslessthan 60 beats/mm (P = NS). All but one patient was in normal sinus rhythm. One patient with cocaineassociatedchest pain was in atrial fibrillation. No patient had atrial or ventricular prematurecontractions. Both physiciansfound the axisto be normal in 109of the 112patientswith no differencesbetween casesandcontrols. The meanPR interval (both 0.15, P = NS) and the mean QRS interval (0.090 vs. 0.096,P = NS) werenot different betweencasesand controls. No patient had a completebundle branch block. The QTc interval was longer in the patients with cocaine-associated chestpain (0.41 vs. 0.39, P c 0.001). The final electrocardiographicdiagnosticclassificationsof casesand controls werenot different (see Table 2). Comparison of the final electrocardiographicclassificationof physician1 and physician2 revealedno difference among casesand controls. When the diagnosticcategoriesweregroupedto assessconcordance,interphysician concordancewas found to be substantial(kappa = 0.70).Comparison of the mean frequencyof electrocardiographicdiagnosesbetweencasesand controls, respectively,revealed: normal, 8% vs. 14%; nondiagnostic,70% vs. 75%; and ischemic,22% vs. 13%. J point elevationand ST segmentelevationwere common in both casesand controls (seeTable 3). Mean frequencyof J point elevationwas45% among cases,and40% amongcontrols(P = NS). The mean frequencyof ST segmentelevationwas 35% among casesand 30% among controls, that is, casesand controlswerenot different. No patieutdemonstrated reversibleST segmentelevations.The prevalenceof early repolarixationasassessed by independentquestioning revealedno difference among cases(mean

Statistical Analysis

Continuousdatawereanalyzedusingtwo-tailed Student t tests. Comparison of casesand controls for categoricalvariableswasperformedusing(X-square or Fisher’sExact Tests,whereappropriate.Interphysicianconcordancewasassessed usingthe kappastatistic (17); P < 0.05 was consideredsignificant for all tests. If one assumesa baselineprevalenceof 15%, this studyhad a powerof 8OVoto detectan approximately two-fold increasein the prevalenceof electrocardiogramsconsistentwith ischemiaor infarction among patientswith cocaine-associated chestpain relativeto controls.

RESULTS

One hundred twelve patients were enrolled, 56 in eachgroup. The patientsweresimilar with respectto age, race,and gender.Cardiac risk factors werenot different exceptfor smoking (seeTable 1). Patients using cocaineweremore likely to be regularlysmokersof tobacco(80% vs. 52%, p < 0.001). Forty-seven of the 56 patients with cocaineassociatedchestpain provided information with respectto their last use of cocaine.Patientswith cocaine-associatedchest pain presentedshortly after their last useof cocaine(median,223minutes; interquartile range25Voto 75%, 120to 680minutes). The mean ( f SEM) heart rate was faster in the patients presenting with cocaine-associatedchest pain than in the controls (85 * 3 vs. 70 l 2, P < 0.0001).Twelve controls and eight patientswith co-

Table 1. Demofjmphlc and Hietorlcal Control Group8

Chamcterl8tlce

Cocaine/Chest Pain

Number yPY1 Slack Hispanic Caucasian Sex (male) Cardiac risk factors Hypertension Hypercholesterolemia Family history Diabetes mellltus Tobacco use

*z?* 22 20 14 42

(39%) (36%) (25%) (75%)

8 (14%) 2 (4%) 8 (14%) 4: (80%)

of Caee and

Controls

P Value

56 28.1 22 20 14 42

(39%) (30%) (25%) (75%)

3 (5%) 1 (2%) 5 (9%) 2 (52%)

NS

NS g $ 0.001

J. E. Hollander et al.

202 Tabie 2. Eiectrocardiographlc

Diagnosis of Each Physician

for Cases and Controls

Physician 1 Control Normal Nonspecific Early repolarization Abnormal, not diagnostic ischemic Acute infarction

5 17 22 2 10 0

Cocaine/Chest Pain

(9%) (30%) (39%) (4%) (16%)

of Eiectrocardiographic

9 (16%) 25 (45%) 6 (14%) 10 (16%) 3 (5%) 1 (2%)

ST elevation ST depression J point elevation T wave inversion Q waves

9 (16%) 2 (4%) 23 (42%) 21 (36%) 1 (2%)

7 (13%) 17 (30%) 7 (13%) 6 (14%) 14 (25%) 3 (5%)

likely to have sustaineda myocardial infarction at initial presentation. The prevalenceof electrocardiographic criteriafor thrombolytic administration was also not different among cases(mean frequency, 12%) and controls (meanfrequency,5%) for eitherphysician(seeTable 4). Interphysicianconcordancein evaluatingwhether the electrocardiogramsmet criteria for thrombolytic administrationwasmoderate(kappa = 0.48). DISCUSSION

The recent epidemic of cocaineuse in the United Stateshasresultedin anincreasednumberof patients presentingto EDs with cocaine-associated medical complaints.Chestpain is the most common of these complaints, occurring up to 40% of the time (4). Although aortic dissection(18), pulmonary infarction (19), and pneumothorax (20) have been reported, myocardial infarction is the most common seriouscauseof chestpain that needsto be excluded. Therefore,manypatientsareadmittedto the hospital to be observedand to rule out myocardialinfarction. The exact incidenceof myocardial infarction in patients presenting with cocaine-associatedchest pain is unclear, with rangesfrom 0% to 31% reported in the literature (8-12). Many clinical and computer algorithms for the identification of lowrisk patients presentingto the ED with chest pain Abnonnaiities

Among Case8 and Physician 2

Physician 1 Controls

Cocaine/Chest Pain

Control

2 (4%) 22 (39%) 21 (36%) 3 (5%) 7 (13%) 1 w44

frequency, 36%) and controls (mean frequency, 36%) for eitherphysician(seeTable 4). Although isolated T-wave abnormalities were common (Table 3), invertedT wavesin all the right precordial leads (Vl-V3) were infrequently noted. Physician 1 noted these findings in three patients with cocaine-associated chestpain and in two controls (P = NS). Physician2 noted thesefindings in two patientswith cocaine-associated chestpain and in onecontrol (P = NS). Of the 17 patientswith cocaine-associated chest pain who had electrocardiogramsclassifiedas ischemia or infarction by physician1, 11(65%) had myocardialinfarction ruled out by serialCPK-MB isoenzymes,and an additional four patients(24%) had a singlenegativeCPK-MB. The remainingtwo patients without cardiacisoenzymeswerenot felt to haveischemicchestpain by the physicianmanagingthe case and weredischargedfrom the ED. Of the eight patients with cocaine-associated chest pain who had electrocardiogramsclassifiedas ischemiaor infarction by physician2, 5 (63%) had myocardial infarction ruled out by serial CPK-MB isoenzymes.Two additional patients (25%) had one negativeCPKMB, and one patient was felt to have nonischemic chestpain by the physician managingthe case.All of the patientswith electrocardiogramsclassifiedas ischemiaor infarction by either physician who did not haveserialcardiacisoenzyrnes had follow-up obtainedbetween12daysand 4 months.They werenot Tabio 3. Prevalence Controls

Physician 2

Cocaine/Chest Pain 9 (16%) 1 (2%) 23 (42%) 24 (43%) 1 Pw

Controls 25 (45%) 1 (2%) 22 (39%) 12 (21%) 0

Cocaine/Chest Pain 30 (54%) 4 VW 27 (46%) 10 (16%) 0

Cocaine-Associated EKG “Abnormalities”

203

Tsble 4. Provancs ofEsllyftepokrlzrtlonMrJ

Ekcttwa~nphlc Thrombotyt&CritwhAmongC8sosmdControls Physician 1 Controls

Early repolarization Thrombolytic criteria

Physician 2

CocainelChest Pain

27 (48%) 3 (5%)

(unrelatedto cocaine)havebeenpresentedin the past decade(13,21-23).Critical to ail of theseprotocols was the initial electrocardiogram.Although a multitude of clinical criteria havebeenexamined,the electrocardiogramis generallyconsideredthe besttest to identify low-risk patients(13,24,25). However, the value of the electrocardiogramfor identifying ischemia in patients with cocaineassociatedchestpain is unclear.Electrocardiograms havebeenreportedto be abnormal in 56% to 84% of patients with cocaine-associated chest pain (810,12). In one seriesof 42 patients with cocaineassociatedchestpain and normal or nonspecificelectrocardiograms,19% of patientshad elevatedCPKMB isoenxymes(11). On the other hand, in a series of 101patientswith cocaine-associated chestpain in which therewereno myocardialinfarctions, 43% of patients had electrocardiogramsthat met standard criteria for the administrationof thrombolytics (10). From the presentseries,a mean frequencyof 46% of electrocardiogramswere considerednormal or nonspecific. The majority of the remaining cases wereinterpretedas eitherearly repolarixation,ischemia, or infarction (seeTable 2). Thesedata are in accordwith previousseries(a-10,12).The electrocardiogram appearsto haveboth poorer sensitivityand specificity for predicting myocardial infarction in this patient population than in older patients with ischemic chest pain secondaryto atherosclerotic heartdisease. Age and race are important considerationsfor proper interpretation of the electrocardiogram.Series of patientswith chestpain unrelatedto cocaine have frequently excludedfemale patients under 40 yearsof age(22)and male patientsunder30 yearsof age(22,23).The meanagesof patientsin theseseries were 56 to 62 years,whereaspatientsin the clinical seriesof cocaine-associated chest pain have mean agesof 28 to 35 years(8-12). The “normal” electrocardiographicvariations found in young patients(J point and ST segmentelevationsdueto earlyrepolarixation or T-wave inversionssecondaryto the persistent juvenile pattern) may make the interpretation of myocardialischemiamore difficult in this patient population.

23 (41%) 8 (11%)

Controls 13 (23%) 3 (5%)

Cocaine/Chest Pain 18 (32%) 7 (13%)

Early repolarizationwas first describedin 1936 and has beenshown to occur in 1% to 2% of healthy young military recruits (27), with an increasedprevalencereportedin youngblack men (2833). We found a high incidenceof J point and ST segmentelevationin both casesand controls(seeTable 3), with most of theseelectrocardiographic“abnormalities” attributed to early repolarization (see Table 4). The prevalenceof early repolarixationin this series(23% to 48%) is similar to the 25% to 35% reportedin other seriesof predominantlyblack patients(30-33). The early repolarizationvariant has beenshown to mimic the electrocardiographicfindings of acute myocardialischemia(28,29).In this series,eachphysicianclassified15% of the electrocardiogramsas ischemicand an additional 2Voto 7!70as suggestiveof acutemyocardial infarction (seeTable 2). Five percent to 13% of electrocardiogramsevenmet the currently acceptedcriteria for thrombolytic administration. ST segmentelevation secondaryto transient ischemiain the patientswith cocaine-associated chest pain was unlikely becauserepeatelectrocardiograms revealedno change,and most patientswith cocaineassociatedchestpain and electrocardiogramsinterpretedas ischemiaor infarction had myocardial infarction ruled out by serial cardiac isoenxymes. Despitean increaseprevalenceof electrocardiograms consistentwith ischemiaor infarction in the patients with cocaine-associated chestpain relativeto controls (22% vs. 13%), thesepatientswerenot found to be statistically different from the age-, race-, gendermatched controls. This study had a power of 80% to detectan approximatelytwo-fold increasein the prevalenceof electrocardiogramsconsistentwith ischemia or infarction among patients with cocaineassociatedchestpain relative to controls. Although we cannotexcludethe possibility that a lesserdifferenceexists,this differencewould be unlikely to have clinical significance. Although not statisticallysignificant, a history of hypertensionwas more common in the patientswith cocaine-associated chestpain (seeTable 1). Patients with hypertensionas well as chronic cocaineusers (even in the absenceof hypertension)have been

(26)

204

J. E. Hollander et al.

noted to have an increased prevalence of left ventricular hypertrophy (34). Left ventricular hypertrophy is associated with electrocardiographic ST and Twave abnormalities. It is possible that some of the difference between the two groups in this study may be accounted for on the basis of hypertension. Our data demonstrate that the high incidence of “abnormal” electrocardiograms previously reported in clinical series of patients with cocaine-associated chest pain is common in a significant proportion of healthy patients under 35 years of age. Many of these observed “abnormalities” are due to the ST segment and J point elevations common in this age group. Early repolarization is found in approximately onethird of patients and may be confused with ischemia. Although isolated T-wave abnormalities were not rare, T-wave inversions in Vl-V3 consistent with the persistent juvenile pattern were present in fewer than 5% of patients. Electrocardiographic criteria for thrombolytic use were met in 5 070to 13 (90of cases. Because thrombolytic therapy has demonstrated value in young patients with acute myocardial infarction (35) and some evidence of success in patients with cocaineinduced myocardial infarction (36-38), this is an important observation for both groups of patients. This study is limited by the fact that we were unable to confirm the control group had not used cocaine. Occult use of cocaine by this group would inhibit the ability to demonstrate a significant difference between the two groups. However, it is unlikely that a significant percentage of the control group was engaging in occult cocaine use. The data

were collected anonymously to increase patient reliability. Other than smoking, cardiac risk factors were equally distributed between patients and controls. This is similar to previous reports of patients with cocaine-associated chest pain (6). In addition, the presence of a faster heart rate and QTc interval prolongation (9) in the patients with cocaine-associated chest pain relative to controls may add some credence to our claim that the two groups differed with respect to cocaine use. As current electrocardiographic criteria for thrombolysis misclassifies young patients with cocaine-associated chest pain, we recommend adoption of a cautious policy regarding thrombolytic use in patients with cocaine-associated chest pain and ST segment elevation. Patients with potentially ischemic chest pain should be treated with oxygen, establishment of intravenous access, aspirin (6), and nitroglycerin (39). When possible, comparison with prior electrocardiograms should be performed. If the chest pain and the ST segment elevations persist and are known to be unchanged from old electrocardiograms, it may be best to withhold thrombolysis and pursue diagnostic cardiac catheterization. If the ST segment elevations are known to be new in the presence of continued chest pain, it may be reasonable to give thrombolytic agents in the absence of contraindications. Further study is needed to better determine the prevalence of ST segment elevation in patients under 35 years of age and to determine whether the currently accepted criteria for thrombolytic administration should apply to this group of patients.

REFERENCES 1. National Institute of Drug Abuse. National household survey on drug abuse. Population estimates, 1988. DHHS number ADM (89-1636). 1989. 2. Cregler LL. Mark H. Relation of acute myocardial infarction to recreational cocaine abuse. Am J Card&l. 1985;56:794. 3. Goldfrank LR. Hoffman RS. The cardiovascular effects of cocaine. Ann Emerg Med. 1991;20:165-75. 4. Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med. 1990, 88:325-31. 5. Coleman DL, Ross TF, Naughton JL. Myocardial ischemia and infarction related to recreational cocaine use. West J Med. 1982;136:444-6. 6. Hollander JE, Hoffmann RS. Cocaine induced myocardial infarction: an analysis and review of the literature. J Emerg Med. 1992;10:169-77. 7. Minor RL, Scott BD, Brown DD, Winniford MD. Cocaine induced myocardial infarction. In patients with normal coronary arteries. Ann Intern Med. 1991;115:797-806. 8. Amin M, Gabelman G, Karpel J, Buttrick P. Acute myocardial infarction and chest pain syndromes after cocaine use. Am J Cardiol. 1990$6:1434-7. 9. Zimmerman JL, Dellinger RP, Majid PA. Cocaine associated chest pain. Ann Emerg Med. 1991;20:61 l-5.

10. Gitter MJ, Goldsmith SR, Dunbar DN, Sharkey SW. Cocaine and chest pain: clinical features and outcome of patients hospitalized to rule out myocardial infarction. Ann Intern Med. 1991;115:277-82. 11. 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. 12. Hollander JE, Hoffmann RS, Genms P, et al. Prospective multicenter evaluation of cocaine associated chest pain. Vet Hum Toxicol. 1992;47:324. Abstract. 13. Lee TH, Cook EF, Weisberg M, Sargent RK, Wilson C, Goldman L. Acute chest pain in the emergency room: identification and examination of low-risk patients. Arch Intern Med. 1985; 145:65-9. 14. Del Aguila C, Rosman H. Myocardial infarction during cocaine withdrawal. Ann Intern Med. 1990;112:712. 15. Nademanee K, Gorelick DA, Josephson MA, et al. Myocardial ischemla during cocaine withdrawal. AM Intern Med. 1989; 111:876-80. 16. The TIM1 Study Group. The thrombolysis in myocardial infarction (TIMI) trial: phase I findings. N Engl J Med. 1985; 312:932. 17. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33: 159-74.

Cocaine-Associated EKG “Abnormalities” 18. Fisher A, Hohoyd BR. Cocaine-associated dissection of the thoracic aorta. J Emerg Med. 1992;10:723-7. 19. Delaney K, Hoffman RS. Pulmonary infarction associated with crack cocaine use in a previously healthy 23 year old woman. Am J Med. 1991;91:92-4. 20. Seaman ME. Barotrauma related to inhalational drug use. J Emerg Med. 1990;8:141-9. 21. Goldman L, Weinberg M, Weisberg M, et al. A computer derived protocol to aid in the diagnosis of emergency room patients with acute chest pain. N Engl J Med. 1982;307:588%. 22. Pozen MW, D’Agostino RB, Selker HP, Sytkowski PA, Hood WB. A predictive instrument to improve coronary care unit admission practices in acute ischemic heart disease. N Engl J Med. 1984;310:1273-8. 23. Goldman L, Cook EF, Brand DA, et al. A computer protocol to predict myocardial infarction in Emergency Department patients with chest pain. N Engl J Med. 1988;318:797-803. 24. Brush JE, Brand DA, Acampora D, Chalmer B, Wackers FJ. Use of the initial electrocardiogram to predict in hospital complications of acute myocardial infarction. N Engl J Med. 1985; 312:1137-41. 25. Slater DK, Hlatky MA, Mark DB, Harrell FE, Pryor DB, Califf RM. Outcome in suspected acute myocardial infarction with normal or minimally abnormal admission electrocardiographic findings. Am J Cardiol. 1987;60:766-70. 26. Shipley RA, Hallaran WR. The four lead electrocardiogram in two hundred normal men and women. Am Heart J. 1936; 11:325-45. 27. Parisi A, Be-ckmann C, Lancaster M. The spectrum of ST segment elevation in the electrocardiograms of healthy adult men. J Electrocardiol. 1971;4:136-44. 28. Wasserburger RH, Alt WJ, Lloyd CJ. The normal RS-T segment elevation variant. Am J Cardiol. 1% 1; 184: 184-92.

205 29. Goldman MJ. RS-T segment elevation in mid and left precordial leads as a normal variant. Am Heart J. 1953;46:817-20. 30. Gottschalk CW, Craige E. A comparison of the precordial S-T and T waves in the electrocardiograms of 600 healthy young negro and white adults. Southern Med J. 1956,49:453-7. 3 1. Grusin H. Peculiarities of the African’s electrocardiogram and the changes observed in serial studies. Circulation. 1954;9: 860-7. 32. Thomas J. Harris E, Lassiter G. Observations on the T wave and S-T segment changes in the electrocardiogram of 320 young negro adults. Am J Cardiol. 1960,5:468-72. 33. Seriki 0, Smith AJ. The electrocardiograms of young Nigerians. Am Heart J. 1966;72:153-7. 34. Brickner ME. Willard JE. Eichhom EJ. Black J. Gravbum PA. Left ve&icular hypehophy ass&ted with &ro& cocaine abuse. Circulation. 1991;&4:1130-5. 35. Chouhan L, Hajar HA, Pomposiello JC. Comparison of thrombolytic therapy for acute myocardial infarction in patientsaged <35and >55years.AmJCardiol. 1993;71:157-9. 36. Hadjimiltiades S, Covalesky V, Manno BS, Haaz WS, Mmtz GS. Coronary arteriographic findings in cocaine abuseinduced myocardial infarction. Cathet Cardiovasc Diagn 1988; 14:33-6. 37. Smith HWB, Liberman HA, Brody SL, et al. Acute myocardial infarction temporally related to cocaine use. Clinical, angiographic, and pathophysiologic observations. Ann Intern Med. 1987;107:13-8. 38. Pate1 R, Haider B. Ahmed S, Regan TJ. Cocaine related myocardial infarction: high prevalence of occlusive coronary thrombi without significant obstructive atherosclerosis. Circ 1988;78(suppl II)II-436. Abstract. 39. Hollander JE. Hoffman RS, G&s P, et al. Nitroglycerin in the treatment of cocaine associated chest pain: clinical safety and efficacy. Ann Emerg Med. 1993;22:945.