- Email: [email protected]
Myeloperoxidase and C-reactive protein in patients with cocaine-associated chest pain
American Journal of Emergency Medicine 31 (2013) 664–669
Contents lists available at SciVerse ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Myeloperoxidase and C-reactive protein in patients with cocaine-associated chest pain Katie O'Conor a, Anna Marie Chang MD a, Alan H.B. Wu PhD b, Judd E. Hollander MD a,⁎ a b
Department of Emergency Medicine, University of Pennsylvania, Philadelphia, PA Department of Laboratory Medicine, University of California at San Francisco
a r t i c l e
i n f o
Article history: Received 7 September 2012 Received in revised form 31 October 2012 Accepted 23 November 2012
a b s t r a c t Background: Myeloperoxidase (MPO) and C-reactive protein (CRP) are markers of inflammation and elevated levels have been found in patients with acute coronary syndrome (ACS) unrelated to cocaine. We evaluated the utility of MPO and CRP for diagnosis of ACS and the prediction of 30-day adverse cardiovascular events in patients with cocaine-related chest pain. Methods: This is a secondary analysis from a prospective cohort study of ED patients who received evaluation for ACS. Structured data collection at presentation included demographics, chest pain history, laboratory results, and electrocardiographic data. Our primary outcome was diagnosis of ACS at index visit and 30-day adverse events. As a secondary analysis, we provide data on a matched cohort without cocaine use. Results: Baseline data and CRP were available for 95 cocaine users; 82 had MPO data also. Patients had a mean age of 46.6 (SD 8.1) years, 90% were black, and 62% were male. Acute coronary syndrome occurred in 7% of cocaine users. With respect to diagnosis of ACS, the area under the curve was poor for both MPO (0.65; 95% confidence interval [CI]: 0.40-0.91) and CRP (0.63; 95% CI: 0.39-0.88). Similar results were found for 30-day events. With respect to prognosis of 30-day adverse cardiovascular events, the area under the curve was 0.68 (95% CI: 0.45-0.91) for MPO and 0.67 (95% CI: 0.45-0.90) for CRP. Similar results were found for 30-day events. In the matched cohort of patients who were not cocaine users, performance of MPO (n = 66) and CRP (n = 86) was also poor. Conclusions: Myeloperoxidase and CRP are not useful for diagnosis or prognosis of patients with cocaineassociated chest pain. © 2013 Elsevier Inc. All rights reserved.
1. Introduction In 2009, an estimated 14.4 million Americans reported prior cocaine use, 1.6 million of whom had used cocaine within the past month [1]. Cocaine use leads to up to 40% of all emergency department (ED) drug-related encounters [2,3]. Chest pain is a common feature of cocaine-related ED visits, occurring in up to 40% of patient presentations [4]. In ED patients presenting with chest pain, acute coronary syndrome (ACS) and acute myocardial infarction (AMI) were found to occur 4 times more often in recent cocaine users, with a nearly 24-fold increase in incidence within the first hour after cocaine use [5,6]. Recently, higher levels of the inflammatory markers myeloperoxidase (MPO) and C-reactive protein (CRP) have been found to be associated with higher rates of adverse cardiovascular events, including AMI and ACS [7-10]. The pathophysiology of cocaine-associated ACS may be different than that of ACS unrelated to cocaine. Cocaine can cause coronary ⁎ Corresponding author. Tel.: +1 215 662 2767. E-mail address: [email protected] (J.E. Hollander). 0735-6757/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2012.11.026
artery vasospasm, but also causes inflammation and increased platelet aggregation [11-13]. Thus, it is theoretically possible that MPO and CRP may be more strongly associated with ACS in patients with cocaine-associated chest pain. Our primary aim was to evaluate the utility of MPO and CRP in the diagnosis and prognosis of ACS in patients presenting to the ED with cocaine-associated chest pain. For purposes of context, we also provide data on a matched cohort of non–cocaine-using patients. 2. Methods 2.1. Study design This was a secondary analysis of prospectively collected data from prior cohort studies of patients presenting to the ED with chest pain from 2004 to 2008. The primary aim of the overall study is to develop and evaluate novel biomarkers. Prior analysis from these studies, but not biomarker results, has been previously reported for subsets of patients [14-16]. For this analysis, our primary aim is to evaluate the utility of MPO and CRP in the diagnosis and prognosis of ACS in patients presenting to the ED with cocaine-associated chest pain. This
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
study included clinical data from the initial ED visit, blood samples collected at presentation, hospital course, and data collected via telephone 30 days after ED visit. The Institutional Review Board approved the study. Informed consent was obtained from each patient. 2.2. Setting The study was conducted in an urban tertiary referral center with an annual ED census between 55,000 and 59,000 during the enrollment period. 2.3. Selection of participants Patients presenting with chest pain consistent with potential ACS, who received an electrocardiogram (ECG) during their visit, were enrolled. This represents a nondifferentiated chest pain population that presents to an urban ED. Patients were excluded if they had chest trauma within the past week, temperature greater than 101°F, pregnancy, use of home oxygen, and/or metastatic cancer. From the overall subject population of patients presenting with chest pain, we selected patients with recent cocaine use. Cocaine use was determined by patient-reported cocaine use in the week before presentation or presence of cocaine metabolites in urine specimen. Patients older than 18 years were eligible to be included if they had self-reported cocaine use; patients not reporting cocaine use were eligible if older than 30 years. We then matched a comparison group of chest pain patients without recent cocaine use to those with cocaine use based on demographic attributes of age, sex, and race. The comparison group was not intended to be statistically compared with the cocaine group, but rather to place the results in context. 2.4. Data collection and processing Patients presenting with chest pain consistent with ACS were identified prospectively by trained research personnel who were present in the ED from 7:00 AM to midnight, 7 days a week. Structured data collection was performed using the Standardized Reporting Guidelines [17] in accordance with the Key Definitions [18] and was documented by the treating physician. This information includes patient-reported demographic characteristics, cardiac risk factors, chest pain characteristics, associated symptoms, medications, initial vital signs, ECG, treatment, and disposition. 2.5. Methods of measurement Blood samples were collected at the time of presentation after informed consent was obtained. Samples were obtained in lithiumheparin tubes for MPO measurement and serum separator tubes for CRP measurement. Serum/plasma was isolated via centrifugation at 3000g for 10 minutes and frozen within 2 hours of sample acquisition. Samples were stored at − 80°C. Samples were shipped to Diazyme Laboratories (Poway, CA) where they were maintained at − 80°C or colder and measured after first thaw. Myeloperoxidase levels were measured using the Diazyme MPO Enzymatic Assay Kit (Diazyme Laboratories). The assay range for MPO was 20 to 1300 ng/mL, with a detection limit of 13.1 ng/mL and total imprecision (coefficient of variation) of 4.1%, 2.8%, and 1.5% at MPO concentrations of 105, 300, and 720 ng/mL. C-reactive protein levels were measured using Diazyme High Sensitivity C-Reactive Protein Assay (Diazyme Laboratories), a quantitative immunoturbidimetric assay. The assay range for CRP was 0.2 to 20 mg/L, with a detection limit of 0.13 mg/L and total imprecision (coefficient of variation) of 7.3%, 2.4%, and 1.6% at CRP concentrations of 1.18, 3.62, and 15.56 mg/L.
665
2.6. Outcome measures The primary outcome was diagnosis of ACS during ED visit and a composite of death, AMI, and/or revascularization at 30 days. Diagnosis of AMI was consistent with the European Society of Cardiology/American College of Cardiology/American Heart Association guidelines [19-21]. Acute coronary syndrome was defined as an AMI or objective confirmation of unstable angina (reversible ischemia on provocative testing or coronary angiography demonstrating stenosis 70% or greater) per guideline recommendations [17]. Revascularization included percutaneous coronary intervention and coronary artery bypass grafting. Medical record reviews and telephone follow-up interviews (performed without knowledge of the MPO and CRP results) were used to determine clinical outcomes during the 30-day period following ED visit. Patients were questioned regarding diagnosis of ACS or AMI, cardiac procedures, and rehospitalizations within 30 days of index visit. Standardized diagnoses were determined and adjudicated by trained research personnel. For patients not reachable by telephone, follow-up was conducted via the Social Security Death Index and review of institutional medical records more than a year after the initial telephone call. 2.7. Primary data analysis Continuous variables are described using means and standard deviations or medians and interquartile ranges, depending on normalcy. 2.8. Analysis of biomarkers' diagnostic and prognostic performance Receiver operating characteristic (ROC) curves were calculated for each biomarker using aggregated data from the entire cohort (cocaine users and non–cocaine users). We chose to define a single cut point for each marker so that it would be clinically sensible: clinicians are not receptive to using different cut points in different patient populations. From these data, we derived the cut point value for each biomarker that would optimize sensitivity and specificity in diagnosis of ACS. We then applied this cut point to assess the number of ACS outcomes it would have diagnosed in each group compared with the standardized diagnoses determined by our protocol. For each biomarker within each group (cocaine use or no cocaine use), ROC curves were calculated using continuous variables for biomarkers and dichotomous outcome variables assigned as ACS or not ACS. Areas under the ROC curves (AUCs) were calculated with 95% confidence intervals (CIs). Receiver operating characteristic curves were also calculated for each biomarker using continuous variables for biomarkers and dichotomous outcome variables assigned to 30-day outcomes. Thirty-day outcomes were classified as adverse cardiovascular events, which included death, AMI, or revascularization, or no adverse cardiovascular event, which was defined as the absence of any reported death, AMI, or revascularization. Areas under the ROC curves were calculated with 95% CIs. We maintained the same cut point for prognosis as we used for diagnosis because we believe this is the way clinicians use laboratory values. 3. Results Baseline and biomarker data were available for 181 chest pain patients presenting to the ED during the enrollment period. Ninetyfive patients were classified as cocaine users; 66 of these patients reported cocaine use within the past week, and 29 patients who reported no cocaine use were found to have cocaine metabolites in their urine at ED presentation. Our matched nonusers group of 86 patients comprised the remaining patients who reported no cocaine
666
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
use within the past week and had a negative urine test result for cocaine metabolites. All 181 patients had CRP results available. Myeloperoxidase results were available for 82 of 95 cocaine users and 66 of 86 non–cocaine users. Some samples could not be used because of hemolysis of samples. Follow-up outcomes were available for 174 patients through direct follow-up (96%); a review of Social Security Death Index and medical records was performed for the remaining 7 patients. The baseline characteristics of the subject populations are presented in Table; mean age was 46.6 (SD 8.1) years, 90% were black, and 62% were male. Cocaine users were more likely than non– cocaine users to be current tobacco users. Cocaine users and non– cocaine users did not differ significantly in any other baseline characteristics, such as history of hypertension, hyperlipidemia, or Thrombolysis in Myocardial Infarction risk score [22]. A final diagnosis of ACS was just as likely in the patients with recent cocaine use compared with those without cocaine use (7% vs 9%; P = .6). No significant difference in biomarker levels was found between the 2 groups overall. Median MPO levels were similar for cocaine users and non–cocaine users (162 vs 136 ng/mL).The optimal cut point to maximize sensitivity and specificity was 242 ng/mL, with an overall sensitivity of 0.43 and a specificity of 0.75. Median CRP levels were similar for cocaine users and non–cocaine users (3 vs 5 mg/L). The optimal cut point was 11.91 mg/L, with a sensitivity of 0.67 and a specificity of 0.79.
3.1. ROC curves—diagnosis of ACS The AUC for MPO for the diagnosis of ACS was poor. In cocaine users, it was 0.65 (95% CI: 0.40-0.91), as compared with 0.46 (95% CI: 0.19-0.74) in non–cocaine users (Fig. 1). The 95% CIs for cocaine users and non–cocaine users overlap. Using the optimal cut point for a diagnosis of ACS, only 57% of the cocaine users and 29% of the non– cocaine users would have been correctly identified as having ACS. Of the patients who were not diagnosed with ACS, 75% of the cocaine users and 76% of non–cocaine users would have been correctly identified as not having ACS. Similarly, the AUC for CRP in cocaine users was poor. It was 0.63 (95% CI: 0.39-0.88), as compared with 0.73 (95% CI: 0.52-0.95) in non–cocaine users (Fig. 2). The 95% CIs for cocaine user and non– cocaine users overlap. Using the optimal cut point, only 43% of the cocaine users and 88% of the non–cocaine users would have been correctly identified as having ACS by an elevated CRP. Of the patients not diagnosed with ACS, 85% of the cocaine users and 72% of non–cocaine users would have been correctly identified as not having ACS. 3.2. ROC curves—30-day prognosis of adverse cardiovascular events The AUC for MPO for prognosis of 30-day adverse cardiovascular events in cocaine users was poor for both cocaine users (0.68; 95% CI:
Table Baseline characteristics of study participants Cocaine users (n = 95) Demographics Age (years) Male Black White Asian Risk factors Weight (lb) Tobacco use Hypertension Hyperlipidemia Diabetes mellitus Family history of CAD TIMI risk score Low (0-2) Moderate (3-5) High (N5) Cardiovascular history Prior MI Prior CABG CHF CAD Arrhythmia Presenting characteristics Heart rate Systolic BP (mm Hg) Diastolic BP (mm Hg) ST elevation ST depression T-wave abnormality Bundle branch block Medication taken/prescribed in past week ASA Antiplatelets β-Blockers Ca blockers Nitro ACE inhibitors Statins Biomarkers MPO (ng/mL) CRP (mg/L)
46.1 57 88 6 1 178 68 62 16 15 11
(39.7, 52.7) 62% 93% 6% 1% (155, 215) 72% 65% 17% 16% 12%
Nonusers (n = 86) 46.7 57 77 9 0 210 38 56 24 24 9
P value (42.7, 51.9) 66% 90% 11% 0.0%
Matched Matched Matched Matched Matched
(180, 251) 44 65% 28% 28% 11%
.2 b.0001 .98 .07 .05 .8 .9
84 11 0
88% 12% 0%
75 11 0
87% 13% 0%
26 1 15 25 5
27% 1 16% 26% 5%
11 3 18 16 7
13% 4% 21% 19% 8%
.02 .3 .8 .2 .4
(73, 98) (122, 155) (78, 98) 3% 6% 24% 1%
.3 .7 .9 .3 .4 .1 .3
27% 7% 24% 20% 4% 25% 22%
.6 .9 .9 .7 .002 .03 .06
(111, 235) (1, 15)
.8 .08
88 138 88 5 7 34 5 29 6 24 17 17 12 11 162 3
(75, 98) (126, 158) (79, 96) 5% 7% 36% 5% 31% 6% 25% 18% 18% 13% 12% (101, 253) (1, 9)
86 140 87 3 5 21 1 23 6 21 17 3 20 19 136 5
Data are shown as number (percentage) for dichotomous variables and median (25th, 75th percentile) for continuous variables. CAD, coronary artery disease; TIMI, thrombolysis in myocardial infarction; MI, myocardial infarction; CABG, coronary artery bypass grafting; CHF, congestive heart failure; BP, blood pressure.
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
667
Fig. 1. Receiver operating characteristic curve for MPO comparing cocaine users to non– cocaine users for diagnosis of ACS.
Fig. 3. Receiver operating characteristic curve for MPO comparing cocaine users to non– cocaine users for diagnosis of 30-day adverse cardiovascular events.
0.45-0.91) and non–cocaine users (0.45; 95% CI: 0.21-0.70) (Fig. 3). The 95% CIs for cocaine user and non–cocaine users overlap. Using the cut point of 242 ng/mL, 57% of the cocaine users and 38% of the non– cocaine users with 30-day adverse cardiovascular events would have been correctly identified. Similarly, the AUC for CRP for prognosis of 30-day adverse cardiovascular events was poor for both cocaine users (0.67; 95% CI: 0.45-0.90) and non–cocaine users (0.76; 95% CI: 0.56-0.96) (Fig. 4). The 95% CIs for cocaine user and non–cocaine users overlap. Using the cut point of 11.91 mg/L, 43% of the cocaine users and 63% of the non– cocaine users with 30-day adverse cardiovascular events would have been correctly identified with elevated CRP.
There has been no clear consensus on the role of MPO and CRP as markers of ACS in patients with ACS unrelated to cocaine. Preliminary studies indicated potential utility for MPO, but those results lacked broad applicability [7-9]. Prior to use of comtemporary troponin assays, Brennan et al [7] found diagnostic utility for MPO in a group with an unusually high ACS prevalence of 41%, not typical of an undifferentiated population presenting to the ED with chest pain [25]. Rudolph et al [26] found some utility in the subset of patients who presented within the first 120 minutes of symptom onset; however, most patients do not present this soon after symptom onset [27]. In contrast to these preliminary studies, our results align with
multicenter studies finding that MPO and CRP do not have independent utility for the diagnosis of ACS in a typical undifferentiated chest pain patient population presenting to the ED [28-31]. A significant body of research supports the value of these biomarkers for predicting long-term cardiovascular outcomes with definite ACS [32-37]. Evidence also points to the utility of MPO in predicting long-term incidence of coronary artery disease (CAD), regardless of ACS [38-40]. Apple et al [41] found MPO to have significant discriminatory utility for the prediction of 30-day adverse cardiovascular outcomes. Schaub et al [29] found MPO to be predictive of all-cause mortality rather than specifically adverse cardiovascular outcomes. CRP was predictive of all adverse outcomes [29]. Our results, which focus on cocaine users, found that neither biomarker is sufficiently informative. Our findings indicate that both MPO and CRP have limited utility in the acute diagnosis of ACS and 30-day prognosis in cocaine-associated chest pain. One potential explanation for this finding is that MPO is a marker of plaque instability and CRP is a marker of inflammation [10,23,24]. Although both plaque instability and inflammation may exist in patients with cocaine-associated ACS, cocaine-associated ACS is, in part, attributed to vasoconstriction in the setting of increased myocardial oxygen demand [11]. Thus, ischemia can occur without the inflammation and plaque rupture that typically precipitate ACS in non–cocaine users [10,23,24]. In addition, although MPO and CRP may predict risk of CAD, cocaine use is not associated with early development of CAD [42].
Fig. 2. Receiver operating characteristic curve for CRP comparing cocaine users to non– cocaine users for diagnosis of ACS.
Fig. 4. Receiver operating characteristic curve for CRP comparing cocaine users to non– cocaine users for diagnosis of 30-day adverse cardiovascular events.
4. Discussion
668
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
We believe that our findings suggest that MPO and CRP do not have sufficient utility to justify being incorporated into the acute care clinical environment [43-45], especially in patients with cocaineassociated chest pain. 5. Limitations Despite a relatively small sample size, our results showed sufficiently poor utility for these biomarkers, such that the upper limit of the CI still would not find these biomarkers useful. It is possible that ACS outcomes in our population were underreported given that not all patients received advanced diagnostic testing or high-sensitivity troponin testing, either of which could detect a greater number of ACS outcomes. However, our rates of ACS were similar to other prior ED-based chest pain studies of cocaine users and nondifferentiated populations [25,46]. In addition, we evaluated the biomarkers using one internally derived cut point from the maximal area under our ROC curve in a combined cohort of cocaine- and non–cocaine-using patients, rather than validating against empirically developed standards. As Morrow and Cook highlighted [47], the optimal process for evaluating a diagnostic biomarker must build on prior research and validate against previously tested cut points, rather than only using internally derived cut points. Our approach would bias our study toward finding more utility for the markers than they might truly have, and yet we found them not to be useful. In addition, our cut point for CRP (11.91 mg/L) lies between 2 common standards referenced in National Academy of Clinical Biochemistry laboratory medicine practice guidelines for ACS (10 and 15 mg/L) [48]. Our MPO cut point of 242 ng/mL is higher than the values in most previously published values, which have ranged from 76 to 210 ng/mL [9,28,34,38,41]. Preanalytical bias may have occurred because MPO samples were collected in lithium heparin tubes, rather than EDTA tubes [49], and without knowledge of heparin administration and timing. Data were not available for elapsed time between symptom onset and blood draw, so the influence of this factor could not be considered [26]. Finally, because cocaine use was determined by presence of cocaine metabolites in urine and/or self-report of any cocaine use within the past week, there may be variation in the acuity of cocaine intoxication affecting the cocaine-use patients at the time of clinical assessment. 6. Conclusions Myeloperoxidase and CRP do not provide sufficiently discriminatory diagnostic or prognostic utility in patients presenting to the ED with cocaine-associated chest pain. References [1] Substance Abuse and Mental Health Services Administration. Results from the 2009 National Survey on Drug Use and Health: volume I. Summary of national findings (Office of Applied Studies, NSDUH series H-38A, HHS publication no. SMA 10-4586). SAMHSA; 2010. [2] Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. Drug Abuse Warning Network, 2008: national estimates of drug-related emergency department visits. HHS publication no. SMA 11-4618. SAMHSA; 2011. [3] Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. The DAWN report: highlights of the 2009 Drug Abuse Warning Network (DAWN) findings on drug-related emergency department visits. SAMHSA; 2010. [4] Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med 1990;88:325-31. [5] Bosch X, Loma-Osorio P, Guasch E, et al. Prevalence, clinical characteristics and risk of myocardial infarction in patients with cocaine-related chest pain. Rev Esp Cardiol 2010;63:1028-34. [6] Mittleman M, Mintzer D, Maclure M, et al. Triggering of myocardial infarction by cocaine. Circulation 1999;99(21):2737-41. [7] Brennan ML, Penn MS, Van Lente F, et al. Prognostic value of myeloperoxidase in patients with chest pain. N Engl J Med 2003;349:1595-604.
[8] Esporcatte R, Rey HC, Rangel FO, et al. Predictive value of myeloperoxidase to identify high risk patients admitted to the hospital with acute chest pain. Arg Bras Cardiol 2007;89:377-84. [9] Sawicki M, Sypniewska G, Kozinski M, et al. Diagnostic efficacy of myeloperoxidase for the detection of acute coronary syndromes. Eur J Clin Invest 2011;41: 667-71. [10] Ramasamy I. Biochemical markers in acute coronary syndrome. Clin Chim Acta 2011;412:1279-96. [11] McCord J, Jneid H, Hollander JE, et al. Management of cocaine-associated chest pain and myocardial infarction: a scientific statement from the American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology. Circulation 2008;117:1897-907. [12] Wright N, Martin M, Goff T, et al. Cocaine and thrombosis: a narrative systematic review of clinical and in-vivo studies. Subst Abuse Treat Prev Policy 2007;2:27. [13] Kugelmass A, Oda A, Monahan K, et al. Platelets adhesion and aggregation: activation of human platelets by cocaine. Circulation 1993;88:876-83. [14] Chase M, Robey JL, Zogby KE, et al. Prospective validation of the Thrombolysis in Myocardial Infarction risk score in the emergency department chest pain population. Ann Emerg Med 2006;48:252-9. [15] Cruz CO, Meshberg EB, Shofer FS, et al. Interrater reliability and accuracy of clinicians and trained research assistants performing prospective data collection in emergency department patients with potential acute coronary syndrome. Ann Emerg Med 2009;54:1-7. [16] Pines JM, Isserman JA, Szyld D, et al. The effect of physician risk tolerance and the presence of an observation unit on decision making for ED patients with chest pain. Am J Emerg Med 2010;27:771-9. [17] Hollander JE, Blomkalns AL, Brogan GX, et al. Standardized reporting guidelines for studies evaluating risk stratification of ED patients with potential acute coronary syndromes. Acad Emerg Med 2004;11:1331-40. [18] Cannon CP, Battler A, Brindis RG, et al. American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients with acute coronary syndromes. A report of the American College of Cardiology Task Force on Clinical Data Standards (Acute Coronary Syndromes Writing Committee). J Am Coll Cardiol 2001;38:2114-30. [19] Hamm CW, Bassand JP, Agewall S, et al. ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2011 Sep 21;32:2999-3054. [20] Thygesen K, Alpert JS, White HD, on behalf of the Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction. Universal definition of myocardial infarction. Eur Heart J 2007;28:2525-38. [21] Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366-74. [22] Antman EM, Cohen M, Bernink PJLM, et al. The TIMI risk score for unstable angina and non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284:835-42. [23] Zaninotto M, Mion MM, Novello E, et al. New biochemical markers: from bench to bedside. Clin Chim Acta 2007;381:14-20. [24] Morrow DA. Appraisal of myeloperoxidase for evaluation of patients with suspected acute coronary syndromes. J Am Coll Cardiol 2007;49:2001-2. [25] Pope JH, Aufderheide TP, Ruthazer R, et al. Missed diagnosis of acute cardiac ischemia in the emergency department. N Engl J Med 2000;342:1163-70. [26] Rudolph V, Goldmann BU, Bos C, et al. Diagnostic value of MPO plasma levels in patients admitted for suspected myocardial infarction. Int J Cardiol 2011;153: 267-71. [27] Luepker RV, Raczynski JM, Osganian S, et al. Effect of a community intervention on patient delay and emergency medical service use in acute coronary heart disease: the Rapid Early Action for Coronary Treatment (REACT) Trial. JAMA 2000;284: 60-7. [28] Peacock WF, Nagurney J, Birkhahn R, et al. Myeloperoxidase in the diagnosis of acute coronary syndromes: the importance of spectrum. Am Heart J 2011;162: 893-9. [29] Schaub N, Reichlin T, Meune C, et al. Markers of plaque instability in the early diagnosis and risk stratification of acute myocardial infarction. Clin Chem 2011;58:246-56. [30] Cristell N, Cianflone D, Durante A, et al. High-sensitivity C-reactive protein is within normal levels at the very onset of first ST-segment elevation acute myocardial infarction in 41% of cases. J Am Coll Cardiol 2011;58:2654-61. [31] Diercks DB, Kirk JD, Naser S, et al. Value of high-sensitivity C-reactive protein in low risk chest pain observation unit patients. Int J Emerg Med 2011;4:1-5. [32] Mocatta TJ, Pilbrow AP, Cameron VA, et al. Plasma concentrations of myeloperoxidase predict mortality after myocardial infarction. J Am Coll Cardiol 2007;49: 1993-2000. [33] Scirica BM, Morrow DA, Cannon CP, et al. Clinical application of C-reactive protein across the spectrum of acute coronary syndromes. Clin Chem 2007;53: 1800-7. [34] Morrow DA, Sabatine MS, Brennan ML, et al. Concurrent evaluation of novel cardiac biomarkers in acute coronary syndrome: myeloperoxidase and soluble CD40 ligand and the risk of recurrent ischaemic events in TACTICS-TIMI 18. Eur Heart J 2008;29:1096-102.
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669 [35] Raposeiras-Roubin S, Barreiro Pardal C, Rodino Janeiro B, et al. High-sensitivity Creactive protein is a predictor of in-hospital cardiac events in acute myocardial infarction independently of GRACE risk score. Angiology 2012;63:30-4. [36] Heeschen C, Hamm CW, Bruemmer J, et al. Predictive value of C-reactive protein and troponin T in patients with unstable angina: a comparative analysis. J Am Coll Cardiol 2000;35:1535-42. [37] Morrow DA, Rifai N, Antman EM, et al. C-reactive protein is a potent predictor of mortality independently of and in combination with troponin T in acute coronary syndromes: a TIMI 11A substudy. J Am Coll Cardiol 1998;31:1460-5. [38] Meuwese MC, Stroes ES, Hazen SL, et al. Serum myeloperoxidase levels are associated with the future risk of coronary artery disease in apparently healthy individuals: the EPIC-Norfolk Prospective Population Study. J Am Coll Cardiol 2007;50:159-65. [39] Heslop CL, Frohlich JJ, Hill JS. Myeloperoxidase and C-reactive protein have combined utility for long-term prediction of cardiovascular mortality after coronary angiography. J Am Coll Cardiol 2010;55:1102-9. [40] Zhang R, Brennan ML, Fu X, et al. Association between myeloperoxidase levels and risk of coronary artery disease. JAMA 2001;286:2136-42. [41] Apple FS, Smith SW, Pearce LA, et al. Myeloperoxidase improves risk stratification in patients with ischemia and normal cardiac troponin I concentrations. Clin Chem 2011;57:603-8.
669
[42] Chang AM, Walsh KM, Shofer FS, et al. Relationship between cocaine use and coronary artery disease in patients with symptoms consistent with an acute coronary syndrome. Acad Emerg Med 2011;18:1-9. [43] Hlatky MA, Greenland P, Arnett DK, et al. Criteria for the evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart Association. Circulation 2009;119:2408-16. [44] Manolio T. Novel risk markers and clinical practice. N Engl J Med 2003;349: 1587-9. [45] Morrow DA, de Lemos JA. Benchmarks for the assessment of novel cardiovascular biomarkers. Circulation 2007;115:949-52. [46] Wang Y, Lindsell C, Pollack C. Self-reported cocaine use, emergency physician testing and outcomes in suspected acute coronary syndromes: a nested matched case-control study. BMJ 2012;2(3). [47] Morrow DA, Cook NR. Determining decision limits for new biomarkers: clinical and statistical considerations. Clin Chem 2011;57:1-3. [48] Morrow DA, Cannon CP, Jesse RL, et al. National Academy of Clinical Biochemistry laboratory medicine practice guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes. Clin Chem 2007;53:552-74. [49] Shih J, Datwyler SA, Hsu SC, et al. Effect of collection tube type and preanalytical handling on myeloperoxidase concentrations. Clin Chem 2008;54:1076-9.
Contents lists available at SciVerse ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Myeloperoxidase and C-reactive protein in patients with cocaine-associated chest pain Katie O'Conor a, Anna Marie Chang MD a, Alan H.B. Wu PhD b, Judd E. Hollander MD a,⁎ a b
Department of Emergency Medicine, University of Pennsylvania, Philadelphia, PA Department of Laboratory Medicine, University of California at San Francisco
a r t i c l e
i n f o
Article history: Received 7 September 2012 Received in revised form 31 October 2012 Accepted 23 November 2012
a b s t r a c t Background: Myeloperoxidase (MPO) and C-reactive protein (CRP) are markers of inflammation and elevated levels have been found in patients with acute coronary syndrome (ACS) unrelated to cocaine. We evaluated the utility of MPO and CRP for diagnosis of ACS and the prediction of 30-day adverse cardiovascular events in patients with cocaine-related chest pain. Methods: This is a secondary analysis from a prospective cohort study of ED patients who received evaluation for ACS. Structured data collection at presentation included demographics, chest pain history, laboratory results, and electrocardiographic data. Our primary outcome was diagnosis of ACS at index visit and 30-day adverse events. As a secondary analysis, we provide data on a matched cohort without cocaine use. Results: Baseline data and CRP were available for 95 cocaine users; 82 had MPO data also. Patients had a mean age of 46.6 (SD 8.1) years, 90% were black, and 62% were male. Acute coronary syndrome occurred in 7% of cocaine users. With respect to diagnosis of ACS, the area under the curve was poor for both MPO (0.65; 95% confidence interval [CI]: 0.40-0.91) and CRP (0.63; 95% CI: 0.39-0.88). Similar results were found for 30-day events. With respect to prognosis of 30-day adverse cardiovascular events, the area under the curve was 0.68 (95% CI: 0.45-0.91) for MPO and 0.67 (95% CI: 0.45-0.90) for CRP. Similar results were found for 30-day events. In the matched cohort of patients who were not cocaine users, performance of MPO (n = 66) and CRP (n = 86) was also poor. Conclusions: Myeloperoxidase and CRP are not useful for diagnosis or prognosis of patients with cocaineassociated chest pain. © 2013 Elsevier Inc. All rights reserved.
1. Introduction In 2009, an estimated 14.4 million Americans reported prior cocaine use, 1.6 million of whom had used cocaine within the past month [1]. Cocaine use leads to up to 40% of all emergency department (ED) drug-related encounters [2,3]. Chest pain is a common feature of cocaine-related ED visits, occurring in up to 40% of patient presentations [4]. In ED patients presenting with chest pain, acute coronary syndrome (ACS) and acute myocardial infarction (AMI) were found to occur 4 times more often in recent cocaine users, with a nearly 24-fold increase in incidence within the first hour after cocaine use [5,6]. Recently, higher levels of the inflammatory markers myeloperoxidase (MPO) and C-reactive protein (CRP) have been found to be associated with higher rates of adverse cardiovascular events, including AMI and ACS [7-10]. The pathophysiology of cocaine-associated ACS may be different than that of ACS unrelated to cocaine. Cocaine can cause coronary ⁎ Corresponding author. Tel.: +1 215 662 2767. E-mail address: [email protected] (J.E. Hollander). 0735-6757/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2012.11.026
artery vasospasm, but also causes inflammation and increased platelet aggregation [11-13]. Thus, it is theoretically possible that MPO and CRP may be more strongly associated with ACS in patients with cocaine-associated chest pain. Our primary aim was to evaluate the utility of MPO and CRP in the diagnosis and prognosis of ACS in patients presenting to the ED with cocaine-associated chest pain. For purposes of context, we also provide data on a matched cohort of non–cocaine-using patients. 2. Methods 2.1. Study design This was a secondary analysis of prospectively collected data from prior cohort studies of patients presenting to the ED with chest pain from 2004 to 2008. The primary aim of the overall study is to develop and evaluate novel biomarkers. Prior analysis from these studies, but not biomarker results, has been previously reported for subsets of patients [14-16]. For this analysis, our primary aim is to evaluate the utility of MPO and CRP in the diagnosis and prognosis of ACS in patients presenting to the ED with cocaine-associated chest pain. This
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
study included clinical data from the initial ED visit, blood samples collected at presentation, hospital course, and data collected via telephone 30 days after ED visit. The Institutional Review Board approved the study. Informed consent was obtained from each patient. 2.2. Setting The study was conducted in an urban tertiary referral center with an annual ED census between 55,000 and 59,000 during the enrollment period. 2.3. Selection of participants Patients presenting with chest pain consistent with potential ACS, who received an electrocardiogram (ECG) during their visit, were enrolled. This represents a nondifferentiated chest pain population that presents to an urban ED. Patients were excluded if they had chest trauma within the past week, temperature greater than 101°F, pregnancy, use of home oxygen, and/or metastatic cancer. From the overall subject population of patients presenting with chest pain, we selected patients with recent cocaine use. Cocaine use was determined by patient-reported cocaine use in the week before presentation or presence of cocaine metabolites in urine specimen. Patients older than 18 years were eligible to be included if they had self-reported cocaine use; patients not reporting cocaine use were eligible if older than 30 years. We then matched a comparison group of chest pain patients without recent cocaine use to those with cocaine use based on demographic attributes of age, sex, and race. The comparison group was not intended to be statistically compared with the cocaine group, but rather to place the results in context. 2.4. Data collection and processing Patients presenting with chest pain consistent with ACS were identified prospectively by trained research personnel who were present in the ED from 7:00 AM to midnight, 7 days a week. Structured data collection was performed using the Standardized Reporting Guidelines [17] in accordance with the Key Definitions [18] and was documented by the treating physician. This information includes patient-reported demographic characteristics, cardiac risk factors, chest pain characteristics, associated symptoms, medications, initial vital signs, ECG, treatment, and disposition. 2.5. Methods of measurement Blood samples were collected at the time of presentation after informed consent was obtained. Samples were obtained in lithiumheparin tubes for MPO measurement and serum separator tubes for CRP measurement. Serum/plasma was isolated via centrifugation at 3000g for 10 minutes and frozen within 2 hours of sample acquisition. Samples were stored at − 80°C. Samples were shipped to Diazyme Laboratories (Poway, CA) where they were maintained at − 80°C or colder and measured after first thaw. Myeloperoxidase levels were measured using the Diazyme MPO Enzymatic Assay Kit (Diazyme Laboratories). The assay range for MPO was 20 to 1300 ng/mL, with a detection limit of 13.1 ng/mL and total imprecision (coefficient of variation) of 4.1%, 2.8%, and 1.5% at MPO concentrations of 105, 300, and 720 ng/mL. C-reactive protein levels were measured using Diazyme High Sensitivity C-Reactive Protein Assay (Diazyme Laboratories), a quantitative immunoturbidimetric assay. The assay range for CRP was 0.2 to 20 mg/L, with a detection limit of 0.13 mg/L and total imprecision (coefficient of variation) of 7.3%, 2.4%, and 1.6% at CRP concentrations of 1.18, 3.62, and 15.56 mg/L.
665
2.6. Outcome measures The primary outcome was diagnosis of ACS during ED visit and a composite of death, AMI, and/or revascularization at 30 days. Diagnosis of AMI was consistent with the European Society of Cardiology/American College of Cardiology/American Heart Association guidelines [19-21]. Acute coronary syndrome was defined as an AMI or objective confirmation of unstable angina (reversible ischemia on provocative testing or coronary angiography demonstrating stenosis 70% or greater) per guideline recommendations [17]. Revascularization included percutaneous coronary intervention and coronary artery bypass grafting. Medical record reviews and telephone follow-up interviews (performed without knowledge of the MPO and CRP results) were used to determine clinical outcomes during the 30-day period following ED visit. Patients were questioned regarding diagnosis of ACS or AMI, cardiac procedures, and rehospitalizations within 30 days of index visit. Standardized diagnoses were determined and adjudicated by trained research personnel. For patients not reachable by telephone, follow-up was conducted via the Social Security Death Index and review of institutional medical records more than a year after the initial telephone call. 2.7. Primary data analysis Continuous variables are described using means and standard deviations or medians and interquartile ranges, depending on normalcy. 2.8. Analysis of biomarkers' diagnostic and prognostic performance Receiver operating characteristic (ROC) curves were calculated for each biomarker using aggregated data from the entire cohort (cocaine users and non–cocaine users). We chose to define a single cut point for each marker so that it would be clinically sensible: clinicians are not receptive to using different cut points in different patient populations. From these data, we derived the cut point value for each biomarker that would optimize sensitivity and specificity in diagnosis of ACS. We then applied this cut point to assess the number of ACS outcomes it would have diagnosed in each group compared with the standardized diagnoses determined by our protocol. For each biomarker within each group (cocaine use or no cocaine use), ROC curves were calculated using continuous variables for biomarkers and dichotomous outcome variables assigned as ACS or not ACS. Areas under the ROC curves (AUCs) were calculated with 95% confidence intervals (CIs). Receiver operating characteristic curves were also calculated for each biomarker using continuous variables for biomarkers and dichotomous outcome variables assigned to 30-day outcomes. Thirty-day outcomes were classified as adverse cardiovascular events, which included death, AMI, or revascularization, or no adverse cardiovascular event, which was defined as the absence of any reported death, AMI, or revascularization. Areas under the ROC curves were calculated with 95% CIs. We maintained the same cut point for prognosis as we used for diagnosis because we believe this is the way clinicians use laboratory values. 3. Results Baseline and biomarker data were available for 181 chest pain patients presenting to the ED during the enrollment period. Ninetyfive patients were classified as cocaine users; 66 of these patients reported cocaine use within the past week, and 29 patients who reported no cocaine use were found to have cocaine metabolites in their urine at ED presentation. Our matched nonusers group of 86 patients comprised the remaining patients who reported no cocaine
666
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
use within the past week and had a negative urine test result for cocaine metabolites. All 181 patients had CRP results available. Myeloperoxidase results were available for 82 of 95 cocaine users and 66 of 86 non–cocaine users. Some samples could not be used because of hemolysis of samples. Follow-up outcomes were available for 174 patients through direct follow-up (96%); a review of Social Security Death Index and medical records was performed for the remaining 7 patients. The baseline characteristics of the subject populations are presented in Table; mean age was 46.6 (SD 8.1) years, 90% were black, and 62% were male. Cocaine users were more likely than non– cocaine users to be current tobacco users. Cocaine users and non– cocaine users did not differ significantly in any other baseline characteristics, such as history of hypertension, hyperlipidemia, or Thrombolysis in Myocardial Infarction risk score [22]. A final diagnosis of ACS was just as likely in the patients with recent cocaine use compared with those without cocaine use (7% vs 9%; P = .6). No significant difference in biomarker levels was found between the 2 groups overall. Median MPO levels were similar for cocaine users and non–cocaine users (162 vs 136 ng/mL).The optimal cut point to maximize sensitivity and specificity was 242 ng/mL, with an overall sensitivity of 0.43 and a specificity of 0.75. Median CRP levels were similar for cocaine users and non–cocaine users (3 vs 5 mg/L). The optimal cut point was 11.91 mg/L, with a sensitivity of 0.67 and a specificity of 0.79.
3.1. ROC curves—diagnosis of ACS The AUC for MPO for the diagnosis of ACS was poor. In cocaine users, it was 0.65 (95% CI: 0.40-0.91), as compared with 0.46 (95% CI: 0.19-0.74) in non–cocaine users (Fig. 1). The 95% CIs for cocaine users and non–cocaine users overlap. Using the optimal cut point for a diagnosis of ACS, only 57% of the cocaine users and 29% of the non– cocaine users would have been correctly identified as having ACS. Of the patients who were not diagnosed with ACS, 75% of the cocaine users and 76% of non–cocaine users would have been correctly identified as not having ACS. Similarly, the AUC for CRP in cocaine users was poor. It was 0.63 (95% CI: 0.39-0.88), as compared with 0.73 (95% CI: 0.52-0.95) in non–cocaine users (Fig. 2). The 95% CIs for cocaine user and non– cocaine users overlap. Using the optimal cut point, only 43% of the cocaine users and 88% of the non–cocaine users would have been correctly identified as having ACS by an elevated CRP. Of the patients not diagnosed with ACS, 85% of the cocaine users and 72% of non–cocaine users would have been correctly identified as not having ACS. 3.2. ROC curves—30-day prognosis of adverse cardiovascular events The AUC for MPO for prognosis of 30-day adverse cardiovascular events in cocaine users was poor for both cocaine users (0.68; 95% CI:
Table Baseline characteristics of study participants Cocaine users (n = 95) Demographics Age (years) Male Black White Asian Risk factors Weight (lb) Tobacco use Hypertension Hyperlipidemia Diabetes mellitus Family history of CAD TIMI risk score Low (0-2) Moderate (3-5) High (N5) Cardiovascular history Prior MI Prior CABG CHF CAD Arrhythmia Presenting characteristics Heart rate Systolic BP (mm Hg) Diastolic BP (mm Hg) ST elevation ST depression T-wave abnormality Bundle branch block Medication taken/prescribed in past week ASA Antiplatelets β-Blockers Ca blockers Nitro ACE inhibitors Statins Biomarkers MPO (ng/mL) CRP (mg/L)
46.1 57 88 6 1 178 68 62 16 15 11
(39.7, 52.7) 62% 93% 6% 1% (155, 215) 72% 65% 17% 16% 12%
Nonusers (n = 86) 46.7 57 77 9 0 210 38 56 24 24 9
P value (42.7, 51.9) 66% 90% 11% 0.0%
Matched Matched Matched Matched Matched
(180, 251) 44 65% 28% 28% 11%
.2 b.0001 .98 .07 .05 .8 .9
84 11 0
88% 12% 0%
75 11 0
87% 13% 0%
26 1 15 25 5
27% 1 16% 26% 5%
11 3 18 16 7
13% 4% 21% 19% 8%
.02 .3 .8 .2 .4
(73, 98) (122, 155) (78, 98) 3% 6% 24% 1%
.3 .7 .9 .3 .4 .1 .3
27% 7% 24% 20% 4% 25% 22%
.6 .9 .9 .7 .002 .03 .06
(111, 235) (1, 15)
.8 .08
88 138 88 5 7 34 5 29 6 24 17 17 12 11 162 3
(75, 98) (126, 158) (79, 96) 5% 7% 36% 5% 31% 6% 25% 18% 18% 13% 12% (101, 253) (1, 9)
86 140 87 3 5 21 1 23 6 21 17 3 20 19 136 5
Data are shown as number (percentage) for dichotomous variables and median (25th, 75th percentile) for continuous variables. CAD, coronary artery disease; TIMI, thrombolysis in myocardial infarction; MI, myocardial infarction; CABG, coronary artery bypass grafting; CHF, congestive heart failure; BP, blood pressure.
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
667
Fig. 1. Receiver operating characteristic curve for MPO comparing cocaine users to non– cocaine users for diagnosis of ACS.
Fig. 3. Receiver operating characteristic curve for MPO comparing cocaine users to non– cocaine users for diagnosis of 30-day adverse cardiovascular events.
0.45-0.91) and non–cocaine users (0.45; 95% CI: 0.21-0.70) (Fig. 3). The 95% CIs for cocaine user and non–cocaine users overlap. Using the cut point of 242 ng/mL, 57% of the cocaine users and 38% of the non– cocaine users with 30-day adverse cardiovascular events would have been correctly identified. Similarly, the AUC for CRP for prognosis of 30-day adverse cardiovascular events was poor for both cocaine users (0.67; 95% CI: 0.45-0.90) and non–cocaine users (0.76; 95% CI: 0.56-0.96) (Fig. 4). The 95% CIs for cocaine user and non–cocaine users overlap. Using the cut point of 11.91 mg/L, 43% of the cocaine users and 63% of the non– cocaine users with 30-day adverse cardiovascular events would have been correctly identified with elevated CRP.
There has been no clear consensus on the role of MPO and CRP as markers of ACS in patients with ACS unrelated to cocaine. Preliminary studies indicated potential utility for MPO, but those results lacked broad applicability [7-9]. Prior to use of comtemporary troponin assays, Brennan et al [7] found diagnostic utility for MPO in a group with an unusually high ACS prevalence of 41%, not typical of an undifferentiated population presenting to the ED with chest pain [25]. Rudolph et al [26] found some utility in the subset of patients who presented within the first 120 minutes of symptom onset; however, most patients do not present this soon after symptom onset [27]. In contrast to these preliminary studies, our results align with
multicenter studies finding that MPO and CRP do not have independent utility for the diagnosis of ACS in a typical undifferentiated chest pain patient population presenting to the ED [28-31]. A significant body of research supports the value of these biomarkers for predicting long-term cardiovascular outcomes with definite ACS [32-37]. Evidence also points to the utility of MPO in predicting long-term incidence of coronary artery disease (CAD), regardless of ACS [38-40]. Apple et al [41] found MPO to have significant discriminatory utility for the prediction of 30-day adverse cardiovascular outcomes. Schaub et al [29] found MPO to be predictive of all-cause mortality rather than specifically adverse cardiovascular outcomes. CRP was predictive of all adverse outcomes [29]. Our results, which focus on cocaine users, found that neither biomarker is sufficiently informative. Our findings indicate that both MPO and CRP have limited utility in the acute diagnosis of ACS and 30-day prognosis in cocaine-associated chest pain. One potential explanation for this finding is that MPO is a marker of plaque instability and CRP is a marker of inflammation [10,23,24]. Although both plaque instability and inflammation may exist in patients with cocaine-associated ACS, cocaine-associated ACS is, in part, attributed to vasoconstriction in the setting of increased myocardial oxygen demand [11]. Thus, ischemia can occur without the inflammation and plaque rupture that typically precipitate ACS in non–cocaine users [10,23,24]. In addition, although MPO and CRP may predict risk of CAD, cocaine use is not associated with early development of CAD [42].
Fig. 2. Receiver operating characteristic curve for CRP comparing cocaine users to non– cocaine users for diagnosis of ACS.
Fig. 4. Receiver operating characteristic curve for CRP comparing cocaine users to non– cocaine users for diagnosis of 30-day adverse cardiovascular events.
4. Discussion
668
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669
We believe that our findings suggest that MPO and CRP do not have sufficient utility to justify being incorporated into the acute care clinical environment [43-45], especially in patients with cocaineassociated chest pain. 5. Limitations Despite a relatively small sample size, our results showed sufficiently poor utility for these biomarkers, such that the upper limit of the CI still would not find these biomarkers useful. It is possible that ACS outcomes in our population were underreported given that not all patients received advanced diagnostic testing or high-sensitivity troponin testing, either of which could detect a greater number of ACS outcomes. However, our rates of ACS were similar to other prior ED-based chest pain studies of cocaine users and nondifferentiated populations [25,46]. In addition, we evaluated the biomarkers using one internally derived cut point from the maximal area under our ROC curve in a combined cohort of cocaine- and non–cocaine-using patients, rather than validating against empirically developed standards. As Morrow and Cook highlighted [47], the optimal process for evaluating a diagnostic biomarker must build on prior research and validate against previously tested cut points, rather than only using internally derived cut points. Our approach would bias our study toward finding more utility for the markers than they might truly have, and yet we found them not to be useful. In addition, our cut point for CRP (11.91 mg/L) lies between 2 common standards referenced in National Academy of Clinical Biochemistry laboratory medicine practice guidelines for ACS (10 and 15 mg/L) [48]. Our MPO cut point of 242 ng/mL is higher than the values in most previously published values, which have ranged from 76 to 210 ng/mL [9,28,34,38,41]. Preanalytical bias may have occurred because MPO samples were collected in lithium heparin tubes, rather than EDTA tubes [49], and without knowledge of heparin administration and timing. Data were not available for elapsed time between symptom onset and blood draw, so the influence of this factor could not be considered [26]. Finally, because cocaine use was determined by presence of cocaine metabolites in urine and/or self-report of any cocaine use within the past week, there may be variation in the acuity of cocaine intoxication affecting the cocaine-use patients at the time of clinical assessment. 6. Conclusions Myeloperoxidase and CRP do not provide sufficiently discriminatory diagnostic or prognostic utility in patients presenting to the ED with cocaine-associated chest pain. References [1] Substance Abuse and Mental Health Services Administration. Results from the 2009 National Survey on Drug Use and Health: volume I. Summary of national findings (Office of Applied Studies, NSDUH series H-38A, HHS publication no. SMA 10-4586). SAMHSA; 2010. [2] Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. Drug Abuse Warning Network, 2008: national estimates of drug-related emergency department visits. HHS publication no. SMA 11-4618. SAMHSA; 2011. [3] Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality. The DAWN report: highlights of the 2009 Drug Abuse Warning Network (DAWN) findings on drug-related emergency department visits. SAMHSA; 2010. [4] Brody SL, Slovis CM, Wrenn KD. Cocaine-related medical problems: consecutive series of 233 patients. Am J Med 1990;88:325-31. [5] Bosch X, Loma-Osorio P, Guasch E, et al. Prevalence, clinical characteristics and risk of myocardial infarction in patients with cocaine-related chest pain. Rev Esp Cardiol 2010;63:1028-34. [6] Mittleman M, Mintzer D, Maclure M, et al. Triggering of myocardial infarction by cocaine. Circulation 1999;99(21):2737-41. [7] Brennan ML, Penn MS, Van Lente F, et al. Prognostic value of myeloperoxidase in patients with chest pain. N Engl J Med 2003;349:1595-604.
[8] Esporcatte R, Rey HC, Rangel FO, et al. Predictive value of myeloperoxidase to identify high risk patients admitted to the hospital with acute chest pain. Arg Bras Cardiol 2007;89:377-84. [9] Sawicki M, Sypniewska G, Kozinski M, et al. Diagnostic efficacy of myeloperoxidase for the detection of acute coronary syndromes. Eur J Clin Invest 2011;41: 667-71. [10] Ramasamy I. Biochemical markers in acute coronary syndrome. Clin Chim Acta 2011;412:1279-96. [11] McCord J, Jneid H, Hollander JE, et al. Management of cocaine-associated chest pain and myocardial infarction: a scientific statement from the American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology. Circulation 2008;117:1897-907. [12] Wright N, Martin M, Goff T, et al. Cocaine and thrombosis: a narrative systematic review of clinical and in-vivo studies. Subst Abuse Treat Prev Policy 2007;2:27. [13] Kugelmass A, Oda A, Monahan K, et al. Platelets adhesion and aggregation: activation of human platelets by cocaine. Circulation 1993;88:876-83. [14] Chase M, Robey JL, Zogby KE, et al. Prospective validation of the Thrombolysis in Myocardial Infarction risk score in the emergency department chest pain population. Ann Emerg Med 2006;48:252-9. [15] Cruz CO, Meshberg EB, Shofer FS, et al. Interrater reliability and accuracy of clinicians and trained research assistants performing prospective data collection in emergency department patients with potential acute coronary syndrome. Ann Emerg Med 2009;54:1-7. [16] Pines JM, Isserman JA, Szyld D, et al. The effect of physician risk tolerance and the presence of an observation unit on decision making for ED patients with chest pain. Am J Emerg Med 2010;27:771-9. [17] Hollander JE, Blomkalns AL, Brogan GX, et al. Standardized reporting guidelines for studies evaluating risk stratification of ED patients with potential acute coronary syndromes. Acad Emerg Med 2004;11:1331-40. [18] Cannon CP, Battler A, Brindis RG, et al. American College of Cardiology key data elements and definitions for measuring the clinical management and outcomes of patients with acute coronary syndromes. A report of the American College of Cardiology Task Force on Clinical Data Standards (Acute Coronary Syndromes Writing Committee). J Am Coll Cardiol 2001;38:2114-30. [19] Hamm CW, Bassand JP, Agewall S, et al. ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Eur Heart J 2011 Sep 21;32:2999-3054. [20] Thygesen K, Alpert JS, White HD, on behalf of the Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction. Universal definition of myocardial infarction. Eur Heart J 2007;28:2525-38. [21] Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction—summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366-74. [22] Antman EM, Cohen M, Bernink PJLM, et al. The TIMI risk score for unstable angina and non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA 2000;284:835-42. [23] Zaninotto M, Mion MM, Novello E, et al. New biochemical markers: from bench to bedside. Clin Chim Acta 2007;381:14-20. [24] Morrow DA. Appraisal of myeloperoxidase for evaluation of patients with suspected acute coronary syndromes. J Am Coll Cardiol 2007;49:2001-2. [25] Pope JH, Aufderheide TP, Ruthazer R, et al. Missed diagnosis of acute cardiac ischemia in the emergency department. N Engl J Med 2000;342:1163-70. [26] Rudolph V, Goldmann BU, Bos C, et al. Diagnostic value of MPO plasma levels in patients admitted for suspected myocardial infarction. Int J Cardiol 2011;153: 267-71. [27] Luepker RV, Raczynski JM, Osganian S, et al. Effect of a community intervention on patient delay and emergency medical service use in acute coronary heart disease: the Rapid Early Action for Coronary Treatment (REACT) Trial. JAMA 2000;284: 60-7. [28] Peacock WF, Nagurney J, Birkhahn R, et al. Myeloperoxidase in the diagnosis of acute coronary syndromes: the importance of spectrum. Am Heart J 2011;162: 893-9. [29] Schaub N, Reichlin T, Meune C, et al. Markers of plaque instability in the early diagnosis and risk stratification of acute myocardial infarction. Clin Chem 2011;58:246-56. [30] Cristell N, Cianflone D, Durante A, et al. High-sensitivity C-reactive protein is within normal levels at the very onset of first ST-segment elevation acute myocardial infarction in 41% of cases. J Am Coll Cardiol 2011;58:2654-61. [31] Diercks DB, Kirk JD, Naser S, et al. Value of high-sensitivity C-reactive protein in low risk chest pain observation unit patients. Int J Emerg Med 2011;4:1-5. [32] Mocatta TJ, Pilbrow AP, Cameron VA, et al. Plasma concentrations of myeloperoxidase predict mortality after myocardial infarction. J Am Coll Cardiol 2007;49: 1993-2000. [33] Scirica BM, Morrow DA, Cannon CP, et al. Clinical application of C-reactive protein across the spectrum of acute coronary syndromes. Clin Chem 2007;53: 1800-7. [34] Morrow DA, Sabatine MS, Brennan ML, et al. Concurrent evaluation of novel cardiac biomarkers in acute coronary syndrome: myeloperoxidase and soluble CD40 ligand and the risk of recurrent ischaemic events in TACTICS-TIMI 18. Eur Heart J 2008;29:1096-102.
K. O'Conor et al. / American Journal of Emergency Medicine 31 (2013) 664–669 [35] Raposeiras-Roubin S, Barreiro Pardal C, Rodino Janeiro B, et al. High-sensitivity Creactive protein is a predictor of in-hospital cardiac events in acute myocardial infarction independently of GRACE risk score. Angiology 2012;63:30-4. [36] Heeschen C, Hamm CW, Bruemmer J, et al. Predictive value of C-reactive protein and troponin T in patients with unstable angina: a comparative analysis. J Am Coll Cardiol 2000;35:1535-42. [37] Morrow DA, Rifai N, Antman EM, et al. C-reactive protein is a potent predictor of mortality independently of and in combination with troponin T in acute coronary syndromes: a TIMI 11A substudy. J Am Coll Cardiol 1998;31:1460-5. [38] Meuwese MC, Stroes ES, Hazen SL, et al. Serum myeloperoxidase levels are associated with the future risk of coronary artery disease in apparently healthy individuals: the EPIC-Norfolk Prospective Population Study. J Am Coll Cardiol 2007;50:159-65. [39] Heslop CL, Frohlich JJ, Hill JS. Myeloperoxidase and C-reactive protein have combined utility for long-term prediction of cardiovascular mortality after coronary angiography. J Am Coll Cardiol 2010;55:1102-9. [40] Zhang R, Brennan ML, Fu X, et al. Association between myeloperoxidase levels and risk of coronary artery disease. JAMA 2001;286:2136-42. [41] Apple FS, Smith SW, Pearce LA, et al. Myeloperoxidase improves risk stratification in patients with ischemia and normal cardiac troponin I concentrations. Clin Chem 2011;57:603-8.
669
[42] Chang AM, Walsh KM, Shofer FS, et al. Relationship between cocaine use and coronary artery disease in patients with symptoms consistent with an acute coronary syndrome. Acad Emerg Med 2011;18:1-9. [43] Hlatky MA, Greenland P, Arnett DK, et al. Criteria for the evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart Association. Circulation 2009;119:2408-16. [44] Manolio T. Novel risk markers and clinical practice. N Engl J Med 2003;349: 1587-9. [45] Morrow DA, de Lemos JA. Benchmarks for the assessment of novel cardiovascular biomarkers. Circulation 2007;115:949-52. [46] Wang Y, Lindsell C, Pollack C. Self-reported cocaine use, emergency physician testing and outcomes in suspected acute coronary syndromes: a nested matched case-control study. BMJ 2012;2(3). [47] Morrow DA, Cook NR. Determining decision limits for new biomarkers: clinical and statistical considerations. Clin Chem 2011;57:1-3. [48] Morrow DA, Cannon CP, Jesse RL, et al. National Academy of Clinical Biochemistry laboratory medicine practice guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes. Clin Chem 2007;53:552-74. [49] Shih J, Datwyler SA, Hsu SC, et al. Effect of collection tube type and preanalytical handling on myeloperoxidase concentrations. Clin Chem 2008;54:1076-9.