- Email: [email protected]
Pregnancy associated plasma protein-A and risk stratification of patients presenting with chest pain in the emergency department
International Journal of Cardiology 117 (2007) 365 – 369 www.elsevier.com/locate/ijcard
Pregnancy associated plasma protein-A and risk stratification of patients presenting with chest pain in the emergency department Ahmad A. Elesber a , Amir Lerman a , Ali E. Denktas a,1 , Zachary T. Resch b,2 , T. Jared Bunch a , Robert S. Schwartz c , Cheryl A. Conover b,⁎ a
c
Division of Cardiovascular Diseases, Department of Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, MN, USA b Minnesota Cardiovascular Research Institute, Minneapolis Heart Institute Foundation, Minneapolis, MN, USA Division of Endocrinology, Metabolism, and Nutrition, Department of Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, MN, USA Received 1 March 2006; received in revised form 23 April 2006; accepted 18 May 2006 Available online 20 July 2006
Abstract Background: The aim of this study was to evaluate the clinical utility of serum pregnancy associated plasma protein-A (PAPP-A) levels in assisting triage of an intermediate to high-risk patient presenting with chest pain in the Emergency Department and no definite evidence of an acute coronary syndrome. Methods: Serum levels of PAPP-A were measured in 59 patients presenting with chest pain to the Emergency Department. The patients were independently grouped according to the presence of acute coronary syndromes or the absence thereof. Results: In a multivariate model that corrected for age, sex, type of chest pain, number of risk factors, history of coronary artery disease, troponin levels, and non-specific ECG changes, PAPP-A levels were still predictive of a final diagnosis of acute coronary syndrome in patients presenting with chest pain to the Emergency Department (Odds Ratio, 2.093; 95th confidence intervals, 1.037–4.224; p = 0.039). Conclusions: Elevated serum PAPP-A levels were predictive of a diagnosis of acute coronary syndrome in intermediate- to high-risk patients presenting to the Emergency Department with chest pain and no definite evidence of an acute coronary syndrome. Thus, serum PAPP-A may be valuable as an adjunct, minimally invasive marker to improve risk stratification in chest pain patients. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Pregnancy-associated plasma protein A; Chest pain; Emergency Department
1. Introduction Chest pain is second only to abdominal pain as the most common reason for Emergency Department visits in the ⁎ Corresponding author. Mayo Clinic, 200 First Street SW, 5-194 Joseph, Rochester, MN 55905, USA. Tel.: +1 507 284 2511; fax: +1 507 255 4828. E-mail addresses: [email protected] (A.E. Denktas), [email protected] (Z.T. Resch), [email protected] (C.A. Conover). 1 Current address: Acibadem Bakirkoy Hospital, Halit Ziya Usakligil Cad 1, 34140 Bakirkoy Istanbul, Turkey. Tel.: +90 212 414 444; fax: +90 212 414 4400. 2 Current address: Department of Internal Medicine, Division of Endocrinology, D1110A Diabetes Center, One Hospital Drive, Columbia, MO 65211, USA. Tel.: +1 573 814 6000x3392; fax: +1 573 814 6551. 0167-5273/$ - see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2006.05.021
United States, making up 5.4% of all visits in 2000 [1]. While it has multiple causes, the most clinically important are the acute coronary syndromes, which must be differentiated from non-cardiac causes [2]. The ability to accurately triage chest pain patients in the Emergency Department remains elusive and is most challenging in patients with intermediate- to high-risk profiles and no definitive evidence of an acute coronary syndrome. Current blood-based markers such as the troponin and creatine kinase-MB reflect myocardial injury and are thus secondary phenomena and not necessarily elevated in unstable angina or early myocardial infarction [3–5]. A circulating marker that would reflect unstable plaque in coronary arteries could provide additional and powerful diagnostic information for patients with acute coronary
366
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
syndromes. Such a marker would also improve risk stratification in chest pain patients and might identify those at risk for future acute coronary syndromes. Serum levels of pregnancy associated plasma protein-A (PAPP-A), a recently identified metalloproteinase in the insulin-like growth factor system [6], were demonstrated to be elevated in men and women with proven unstable angina or acute myocardial infarction [7]. In another study, circulating PAPP-A was found to be a strong independent predictor of ischemic cardiac events and need for revascularization in patients with acute coronary syndromes whose troponin was negative [8]. Therefore, the current study was designed to assess whether there is an additive diagnostic value of measuring circulating PAPP-A levels to the known clinical information and laboratory tests available to Emergency Department physicians when evaluating a patient with chest pain of at least intermediate risk and no definitive diagnosis of an acute coronary syndrome. 2. Methods 2.1. Patients and design The study consisted of patients presenting with chest pain to the Emergency Department at St. Mary's Hospital, Rochester, MN. One hundred and six patients accepted to participate in the study and gave informed consent. Blood was drawn and sent for analysis. After completion of the medical evaluation, the paper records of these 106 patients were reviewed. Patient history, cardiac risk factors, electrocardiogram (ECG), stress test results, coronary angiogram results and laboratory values for serum troponin were abstracted. Patients of intermediate to high likelihood of having a significant coronary event at presentation to the emergency department were eligible for the study while patients with a definite acute coronary syndrome at presentation or very low likelihood of a significant coronary event were excluded. Accordingly, patients with one or more risk factors for coronary atherosclerosis regardless of the character of the chest pain and patients with atypical or typical chest pain were included. There were 59 patients that met these inclusion criteria and subsequently formed the study population. Excluded patients were (1) patients with noncardiac chest pain and no risk factor who were considered very unlikely to have an acute coronary syndrome (18 patients); (2) patients with ST elevation myocardial infarction or dynamic electrocardiographic ST segment changes ≥ 0.5 mm from baseline (4 patients); (3) patients with a positive initial troponin T (13 patients); (4) patients with significant renal dysfunction and serum creatinine>1.8 mg/ dL (5 patients); (5) patients with acute or chronic inflammatory diseases (7 patients); (6) and pregnant patients (none). This protocol was approved by the Mayo Clinic Institutional Review Board.
After reviewing the records of the complete evaluation that included a stress test and/or a coronary angiogram, a final diagnosis of the presence or absence of an acute coronary syndrome (ACS) was made. Patients were grouped independently by 2 reviewers (AAE and AED) who were blinded to the PAPP-A results. Acute coronary syndromes were defined as patients with a positive troponin T on subsequent blood draws, a positive treadmill stress test with Duke score ≤5, an imaging stress test with new areas of myocardial involvement, or “active” coronary artery disease (ulceration, thrombus) shown by subsequent coronary arteriography. All other patients who did not meet the above criteria for an ACS were considered to have noncardiac chest pain. 2.2. Definitions History of coronary artery disease was considered present in patients with a prior positive stress test, positive coronary angiogram, previous myocardial infarction or coronary revascularization procedure. Coronary artery disease (CAD) risk factors, as outlined by the National Cholesterol Program (NCEP), were age, gender, family history of disease, smoking, hypertension, diabetes mellitus, and hyperlipidemia [9]. The later five risk factors were collapsed into one variable (risk factor number, 0 to 5) reflecting the presence or absence of each one of them. Chest pain was considered typical if it met these 3 characteristics: (1) retrosternal; (2) precipitated by effort or stress; and (3) relieved by rest or nitroglycerin. Chest pain was considered atypical if it met any 2 characteristics and non-cardiac if it met only 1 characteristic. The glomerular filtration rate was estimated with the use of the Modified Diet in Renal Disease (MDRD) equation [10]. A troponin T > 0.03 ng/mL was considered positive. 2.3. Serum assays for PAPP-A Blood samples were taken by venipuncture on admission to the Emergency Department. Serum PAPP-A levels were determined by Ultra-sensitive PAPP-A ELISA kits kindly provided by Diagnostic Systems Laboratories, Inc. (Webster, TX). In view of the fact that circulating ACSrelated PAPP-A is different from circulating pregnancyrelated PAPP-A in that it is not complexed with proMBP [11], the test used in our study was developed and validated for non-pregnancy applications and uses different antibodies that specifically recognize epitopes on PAPP-A and not the pro-form of eosinophil major basic protein. Minimum sensitivity is 0.24 mIU/L, with intra- and interassay coefficients of variation of 4.7% and 4.2%, respectively. All patient samples were run together for the assay by a technician who was blinded to the ACS status of the patients.
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
2.4. Statistical analyses Continuous variables with symmetric distributions are presented as mean ± deviation. Those with moderate/severe skewness are summarized as median (first, third quartile). Discrete variables are presented as frequency and percentage. Groups comparisons are tested using Student's t-test (continuous, symmetric), Wilcoxon rank sum test (skewed continuous or ordinal discrete), and Pearson's chi-squared test (nominal). Covariates of interest for regression adjustment were age, sex, type of chest pain, number of risk factors, history of CAD, troponin, and non-specific ECG changes. Given the small sample size, only one parameter for covariate adjustment could be fit in the logistic regression model for final diagnosis. Thus, the covariate space was collapsed into one dimension by regressing the covariates on the log2transformed values of PAPP-A. The predicted values from this model were then used in the logistic regression model with log2-transformed values of PAPP-A to estimate partial associations with final diagnosis. The transformation of PAPP-A means that the resulting odds ratio should be interpreted as the expected relative increase in odds for a doubling of the PAPP-A value. 3. Results 3.1. Patients' characteristics Table 1 summarizes the clinical and laboratory characteristics of the patient population according to final diagnosis of acute coronary syndrome versus non-cardiac chest pain. After review of the complete medical evaluation including follow-up troponin levels, cardiac stress testing and coronary angiography 19 (32%) patients had a final diagnosis of acute coronary syndrome, while 40 (68%) patients had a final diagnosis of non-cardiac chest pain. As expected, patients with a final diagnosis of acute coronary syndrome were older, more commonly males, presented more often with typical angina, had a higher prevalence of known coronary artery disease, and a higher troponin and PAPP-A levels compared to patients with non-cardiac chest pain. Fig. 1 shows that the serum PAPP-A levels were significantly higher (p = 0.001) in patients with a final diagnosis of acute coronary syndrome (median, 2.0 mIU/L; 25th quartile, 1.2; 75th quartile, 4.9) compared to patients with a final diagnosis of non-cardiac chest pain (median, 1.2 mIU/L; 25th quartile, 0.7; 75th quartile, 1.6). The mean age of the total population was 63.1 ± 12.8 years with 64% males. All patients were Caucasian from southern Minnesota, western Wisconsin and northern Iowa. The mean body mass index was 30.5 ± 5.0 kg/m2 with 50% of the patient population being overweight (i.e., body mass index > 30 kg/m2). Thirty three patients (56%) had hypertension, 10 (17%) had diabetes mellitus, 33 (56%) had
367
Table 1 The clinical and laboratory characteristics of the patient population according to final diagnosis of acute coronary syndrome (ACS) versus non-cardiac chest pain (CP) Variable
Non-cardiac CP (N = 40)
ACS (N = 19)
Age (years) Male, n (%) Body mass index Number of risk factors ⁎ (%) No risk factors Any 1 risk factor Any 2 risk factors Any 3 risk factors Any 4 risk factors All 5 risk factors History of coronary artery disease, n (%) Angina, n (%) Non-cardiac Atypical Typical ECG results, n (%) Normal Non-specific Initial troponin, median (Q1, Q3) (ng/mL) Estimated GFR, median (Q1, Q3) (mL/min) PAPP-A, median (Q1, Q3) (mIU/L)
60.1 ± 12.6 22 (55%) 30.9 ± 5.2
69.4 ± 10.9 16 (84%) 29.4 ± 4.5
6 (15%) 16 (40%) 11 (28%) 5 (12%) 2 (5%) 0 (0%) 10 (25%)
0 (0%) 5 (28%) 6 (34%) 5 (28%) 1 (5%) 1 (5%) 13 (68%)
29 (72%) 6 (15%) 5 (13%)
3 (16%) 5 (26%) 11 (58%)
P-value 0.008 0.029 0.29 0.20
0.0014 <0.001
0.60 28 (70%) 12 (63%) 12 (30%) 7 (37%) 0.01 (0.01, 0.02) 0.02 (0.01, 0.03)
0.13
69.5 (60.8, 77.5) 63.0 (57.0, 76.0)
0.36
1.2 (0.7, 1.6)
0.002
2.0 (1.2, 4.9)
⁎ Risk factors included: male gender, smoking, hypertension, diabetes mellitus, and hyperlipidemia (as defined by the National Cholesterol Program).
hyperlipidemia, 9 (15%) were active smoker, and 23 (39%) had family history of coronary atherosclerosis. Twenty three (39%) patients had a history of CAD. Thirty two (54%) patients presented with non-cardiac chest pain, 11 (19%) with atypical chest pain, and 16 (27%) with typical chest pain. Nineteen (32%) patients had non-specific ST/T wave changes on ECG while 40 (68%) had normal ECG at presentation. Median glomerular filtration rate was 68.0 mL/ min (25th quartile, 57.0; 75th quartile, 76.0), median troponin was 0.010 ng/mL (25th quartile, 0.01; 75th quartile, 0.01), and median PAPP-A was 1.29 mIU/L (25th quartile, 0.79; 75th quartile, 1.90). 3.2. Incremental diagnostic value of PAPP-A levels In a univariate model, measurement of PAPP-A levels at initial evaluation was found to be predictive of a final diagnosis of acute coronary syndrome in patients presenting with chest pain to the Emergency Department (Odds Ratio, 2.47; 95th confidence intervals, 1.29–4.27; p = 0.0062). In a multivariate model that corrected for age, sex, type of chest pain, number of risk factors, history of coronary artery disease, troponin levels, and non-specific ECG changes, PAPP-A levels were still predictive of a final diagnosis of acute coronary syndrome in patients presenting with chest
368
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
Fig. 1. Serum PAPP-A levels box plot distribution according to final diagnosis. The box indicates the interquartile range and median. The circles outside the box represent outliers. CP: chest pain. ACS: acute coronary syndrome.
pain to the Emergency Department (Odds Ratio, 2.093; 95th confidence intervals, 1.037–4.224; p = 0.039). 4. Discussion The results of this study suggest that serum PAPP-A may be valuable as an adjunct, minimally invasive marker to risk stratify intermediate- to high-risk patients presenting to the Emergency Department with chest discomfort. With each doubling in serum PAPP-A levels in patients presenting with chest pain, the risk of having an acute coronary syndrome likewise doubles. Our study population included patients categorized as intermediate to high risk without definite evidence of an acute coronary syndrome. This study population is particularly challenging since a significant proportion of these patients can have either cardiac or non-cardiac chest pain. In our particular patient population 32% had a final diagnosis of acute coronary syndrome while 68% had a non-cardiac etiology despite an initial elevated risk categorization. In our study, serum PAPP-A levels were found to be elevated in acute coronary syndrome patients who were troponin-negative at presentation at the Emergency Department, similar to the findings of Lund et al. [8]. That study, however, examined a more select patient group admitted to the hospital with an acute coronary syndrome [8]. In a previous study of patients undergoing coronary arteriography, we found elevated serum PAPP-A to be a marker of unstable atherosclerotic lesions [7]. A report by Beaudeux et al. [12] also found mildly elevated PAPP-A levels in asymptomatic hyperlipidemic patients with advanced
hyper-/isoechoic carotid atherosclerotic lesions. Together, these studies support a relationship between unstable coronary artery disease and level of serum PAPP-A. This study expands those prior reports and includes a broader range of patients similar in characteristics to those who present to an Emergency Department. Detection of unstable or potentially unstable vascular lesions is an important clinical goal, especially in the earliest stages when therapeutic interventions, both invasive and noninvasive, are most likely to result in the greatest benefit. A key feature emerging in these detection strategies is the identification of processes that confer instability to coronary or cerebral vascular lesions, at this time understood to be principally inflammation. We showed previously that PAPP-A localizes in unstable human plaque using immunohistochemical methods [7]. PAPP-A is associated with locally infiltrating macrophages, a principal inflammatory component of culprit lesions. Whether circulating PAPP-A detected in peripheral blood arises from vulnerable plaques is uncertain, but this provides an intriguing hypothesis about enhanced PAPPA sensitivity and specificity for detecting acute coronary syndromes. Indeed, emerging data suggest that inflammation may not be localized to a specific plaque. More often, the entire coronary tree is inflamed resulting in “pancoronaritis” and diffuse arterial destabilization. This often results in distant plaque progression and many acute coronary syndromes as supported by multiple prior studies [13,14]. Spagnoli et al. [13] found immunohistochemical evidence that patients dying from acute myocardial infarction have diffuse inflammation of all three major coronary arteries, irrespective of the artery with the culprit lesion. Pathophysiologic processes destabilizing artery wall segments are likely in multiple locations throughout the major epicardial coronary arteries. Other studies have shown similar results [13,14]. Taken together, these studies challenge the notion that a single vulnerable plaque is responsible for coronary plaque instability and resulting acute coronary syndromes. Elevated systemic levels of inflammatory markers in subjects with acute coronary syndromes may thus be due to the diffuse nature of the arterial inflammation rather than from a single site, and their detection in peripheral blood may prove a viable strategy for risk determination. The limitation of our study lies in the relatively small number of patients. We also did not collect follow-up information for major adverse cardiovascular events. However, our main interest was the immediate risk stratification of patients presenting to the Emergency Department as required by the physician facing such an acute situation. Second, this study was a prospective but observational evaluation. Unobserved confounders may have affected the study results, and multivariate analyses may have failed to completely remove all confounders of those covariables that were measured. Finally, we acknowledge that waiting for a second troponin in the Emergency Department is a
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
reasonable strategy. Nonetheless, saving a 6-h wait for a second troponin in a busy Emergency Department can be of great help in the triaging of patient care towards the patients who need that the most. Add to that fact that only about 10% of our patients with a negative initial troponin had a positive troponin at follow-up (data not shown). To conclude, serum PAPP-A has the potential to be a valuable marker of acute coronary syndromes, in particular, as a predictor of clinical instability in acute coronary syndrome patients who are troponin-negative. The results of this study suggest that further investigation into the potential for PAPP-A as point-of-care assay for decisionmaking in the Emergency Department is indicated. Acknowledgement This work was supported by FNDT0151290Z-1 from the American Heart Association (CAC), 0225543Z from the American Heart Association (ZTR), and Endocrinology and Metabolism Training Grant DK-07352 (ZTR) from the National Institutes of Health. References [1] Ly N, McCaig LF. National Hospital Ambulatory Medical Care Survey: 2000 outpatient department summary. Adv Data 2002;327: 1– 27. [2] American Heart Association 2000 Heart and Stroke Statistical Update. In:http://www.americanheart.org/statistics/04cornry.html ed: American Heart Association, 2000. [3] Antman EM, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335(18):1342–9.
369
[4] Hamm CW, et al. Emergency room triage of patients with acute chest pain by means of rapid testing for cardiac troponin T or troponin I. N Engl J Med 1997;337(23):1648–53. [5] Puleo PR, et al. Use of a rapid assay of subforms of creatine kinase-MB to diagnose or rule out acute myocardial infarction. N Engl J Med 1994;331(9):561–6. [6] Lawrence JB, et al. The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc Natl Acad Sci U S A 1999;96(6):3149–53. [7] Bayes-Genis A, et al. Pregnancy-associated plasma protein A as a marker of acute coronary syndromes. N Engl J Med 2001;345(14): 1022–9. [8] Lund J, et al. Circulating pregnancy-associated plasma protein a predicts outcome in patients with acute coronary syndrome but no troponin I elevation. Circulation 2003;108(16):1924–6. [9] Expert Panel on Detection, E. and A. Treatment of High Blood Cholesterol in, Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 2001;285(19):2486–97. [10] Levey AS, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130(6):461–70. [11] Qin QP, et al. Molecular distinction of circulating pregnancyassociated plasma protein A in myocardial infarction and pregnancy. Clin Chem 2005;51(1):75–83. [12] Beaudeux JL, et al. Serum plasma pregnancy-associated protein A: a potential marker of echogenic carotid atherosclerotic plaques in asymptomatic hyperlipidemic subjects at high cardiovascular risk. Arterioscler Thromb Vasc Biol 2003;23(1):e7–e10. [13] Spagnoli LG, et al. Multicentric inflammation in epicardial coronary arteries of patients dying of acute myocardial infarction. J Am Coll Cardiol 2002;40(9):1579–88. [14] Buffon A, et al. Widespread coronary inflammation in unstable angina. N Engl J Med 2002;347(1):5–12.
Pregnancy associated plasma protein-A and risk stratification of patients presenting with chest pain in the emergency department Ahmad A. Elesber a , Amir Lerman a , Ali E. Denktas a,1 , Zachary T. Resch b,2 , T. Jared Bunch a , Robert S. Schwartz c , Cheryl A. Conover b,⁎ a
c
Division of Cardiovascular Diseases, Department of Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, MN, USA b Minnesota Cardiovascular Research Institute, Minneapolis Heart Institute Foundation, Minneapolis, MN, USA Division of Endocrinology, Metabolism, and Nutrition, Department of Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, MN, USA Received 1 March 2006; received in revised form 23 April 2006; accepted 18 May 2006 Available online 20 July 2006
Abstract Background: The aim of this study was to evaluate the clinical utility of serum pregnancy associated plasma protein-A (PAPP-A) levels in assisting triage of an intermediate to high-risk patient presenting with chest pain in the Emergency Department and no definite evidence of an acute coronary syndrome. Methods: Serum levels of PAPP-A were measured in 59 patients presenting with chest pain to the Emergency Department. The patients were independently grouped according to the presence of acute coronary syndromes or the absence thereof. Results: In a multivariate model that corrected for age, sex, type of chest pain, number of risk factors, history of coronary artery disease, troponin levels, and non-specific ECG changes, PAPP-A levels were still predictive of a final diagnosis of acute coronary syndrome in patients presenting with chest pain to the Emergency Department (Odds Ratio, 2.093; 95th confidence intervals, 1.037–4.224; p = 0.039). Conclusions: Elevated serum PAPP-A levels were predictive of a diagnosis of acute coronary syndrome in intermediate- to high-risk patients presenting to the Emergency Department with chest pain and no definite evidence of an acute coronary syndrome. Thus, serum PAPP-A may be valuable as an adjunct, minimally invasive marker to improve risk stratification in chest pain patients. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Pregnancy-associated plasma protein A; Chest pain; Emergency Department
1. Introduction Chest pain is second only to abdominal pain as the most common reason for Emergency Department visits in the ⁎ Corresponding author. Mayo Clinic, 200 First Street SW, 5-194 Joseph, Rochester, MN 55905, USA. Tel.: +1 507 284 2511; fax: +1 507 255 4828. E-mail addresses: [email protected] (A.E. Denktas), [email protected] (Z.T. Resch), [email protected] (C.A. Conover). 1 Current address: Acibadem Bakirkoy Hospital, Halit Ziya Usakligil Cad 1, 34140 Bakirkoy Istanbul, Turkey. Tel.: +90 212 414 444; fax: +90 212 414 4400. 2 Current address: Department of Internal Medicine, Division of Endocrinology, D1110A Diabetes Center, One Hospital Drive, Columbia, MO 65211, USA. Tel.: +1 573 814 6000x3392; fax: +1 573 814 6551. 0167-5273/$ - see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2006.05.021
United States, making up 5.4% of all visits in 2000 [1]. While it has multiple causes, the most clinically important are the acute coronary syndromes, which must be differentiated from non-cardiac causes [2]. The ability to accurately triage chest pain patients in the Emergency Department remains elusive and is most challenging in patients with intermediate- to high-risk profiles and no definitive evidence of an acute coronary syndrome. Current blood-based markers such as the troponin and creatine kinase-MB reflect myocardial injury and are thus secondary phenomena and not necessarily elevated in unstable angina or early myocardial infarction [3–5]. A circulating marker that would reflect unstable plaque in coronary arteries could provide additional and powerful diagnostic information for patients with acute coronary
366
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
syndromes. Such a marker would also improve risk stratification in chest pain patients and might identify those at risk for future acute coronary syndromes. Serum levels of pregnancy associated plasma protein-A (PAPP-A), a recently identified metalloproteinase in the insulin-like growth factor system [6], were demonstrated to be elevated in men and women with proven unstable angina or acute myocardial infarction [7]. In another study, circulating PAPP-A was found to be a strong independent predictor of ischemic cardiac events and need for revascularization in patients with acute coronary syndromes whose troponin was negative [8]. Therefore, the current study was designed to assess whether there is an additive diagnostic value of measuring circulating PAPP-A levels to the known clinical information and laboratory tests available to Emergency Department physicians when evaluating a patient with chest pain of at least intermediate risk and no definitive diagnosis of an acute coronary syndrome. 2. Methods 2.1. Patients and design The study consisted of patients presenting with chest pain to the Emergency Department at St. Mary's Hospital, Rochester, MN. One hundred and six patients accepted to participate in the study and gave informed consent. Blood was drawn and sent for analysis. After completion of the medical evaluation, the paper records of these 106 patients were reviewed. Patient history, cardiac risk factors, electrocardiogram (ECG), stress test results, coronary angiogram results and laboratory values for serum troponin were abstracted. Patients of intermediate to high likelihood of having a significant coronary event at presentation to the emergency department were eligible for the study while patients with a definite acute coronary syndrome at presentation or very low likelihood of a significant coronary event were excluded. Accordingly, patients with one or more risk factors for coronary atherosclerosis regardless of the character of the chest pain and patients with atypical or typical chest pain were included. There were 59 patients that met these inclusion criteria and subsequently formed the study population. Excluded patients were (1) patients with noncardiac chest pain and no risk factor who were considered very unlikely to have an acute coronary syndrome (18 patients); (2) patients with ST elevation myocardial infarction or dynamic electrocardiographic ST segment changes ≥ 0.5 mm from baseline (4 patients); (3) patients with a positive initial troponin T (13 patients); (4) patients with significant renal dysfunction and serum creatinine>1.8 mg/ dL (5 patients); (5) patients with acute or chronic inflammatory diseases (7 patients); (6) and pregnant patients (none). This protocol was approved by the Mayo Clinic Institutional Review Board.
After reviewing the records of the complete evaluation that included a stress test and/or a coronary angiogram, a final diagnosis of the presence or absence of an acute coronary syndrome (ACS) was made. Patients were grouped independently by 2 reviewers (AAE and AED) who were blinded to the PAPP-A results. Acute coronary syndromes were defined as patients with a positive troponin T on subsequent blood draws, a positive treadmill stress test with Duke score ≤5, an imaging stress test with new areas of myocardial involvement, or “active” coronary artery disease (ulceration, thrombus) shown by subsequent coronary arteriography. All other patients who did not meet the above criteria for an ACS were considered to have noncardiac chest pain. 2.2. Definitions History of coronary artery disease was considered present in patients with a prior positive stress test, positive coronary angiogram, previous myocardial infarction or coronary revascularization procedure. Coronary artery disease (CAD) risk factors, as outlined by the National Cholesterol Program (NCEP), were age, gender, family history of disease, smoking, hypertension, diabetes mellitus, and hyperlipidemia [9]. The later five risk factors were collapsed into one variable (risk factor number, 0 to 5) reflecting the presence or absence of each one of them. Chest pain was considered typical if it met these 3 characteristics: (1) retrosternal; (2) precipitated by effort or stress; and (3) relieved by rest or nitroglycerin. Chest pain was considered atypical if it met any 2 characteristics and non-cardiac if it met only 1 characteristic. The glomerular filtration rate was estimated with the use of the Modified Diet in Renal Disease (MDRD) equation [10]. A troponin T > 0.03 ng/mL was considered positive. 2.3. Serum assays for PAPP-A Blood samples were taken by venipuncture on admission to the Emergency Department. Serum PAPP-A levels were determined by Ultra-sensitive PAPP-A ELISA kits kindly provided by Diagnostic Systems Laboratories, Inc. (Webster, TX). In view of the fact that circulating ACSrelated PAPP-A is different from circulating pregnancyrelated PAPP-A in that it is not complexed with proMBP [11], the test used in our study was developed and validated for non-pregnancy applications and uses different antibodies that specifically recognize epitopes on PAPP-A and not the pro-form of eosinophil major basic protein. Minimum sensitivity is 0.24 mIU/L, with intra- and interassay coefficients of variation of 4.7% and 4.2%, respectively. All patient samples were run together for the assay by a technician who was blinded to the ACS status of the patients.
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
2.4. Statistical analyses Continuous variables with symmetric distributions are presented as mean ± deviation. Those with moderate/severe skewness are summarized as median (first, third quartile). Discrete variables are presented as frequency and percentage. Groups comparisons are tested using Student's t-test (continuous, symmetric), Wilcoxon rank sum test (skewed continuous or ordinal discrete), and Pearson's chi-squared test (nominal). Covariates of interest for regression adjustment were age, sex, type of chest pain, number of risk factors, history of CAD, troponin, and non-specific ECG changes. Given the small sample size, only one parameter for covariate adjustment could be fit in the logistic regression model for final diagnosis. Thus, the covariate space was collapsed into one dimension by regressing the covariates on the log2transformed values of PAPP-A. The predicted values from this model were then used in the logistic regression model with log2-transformed values of PAPP-A to estimate partial associations with final diagnosis. The transformation of PAPP-A means that the resulting odds ratio should be interpreted as the expected relative increase in odds for a doubling of the PAPP-A value. 3. Results 3.1. Patients' characteristics Table 1 summarizes the clinical and laboratory characteristics of the patient population according to final diagnosis of acute coronary syndrome versus non-cardiac chest pain. After review of the complete medical evaluation including follow-up troponin levels, cardiac stress testing and coronary angiography 19 (32%) patients had a final diagnosis of acute coronary syndrome, while 40 (68%) patients had a final diagnosis of non-cardiac chest pain. As expected, patients with a final diagnosis of acute coronary syndrome were older, more commonly males, presented more often with typical angina, had a higher prevalence of known coronary artery disease, and a higher troponin and PAPP-A levels compared to patients with non-cardiac chest pain. Fig. 1 shows that the serum PAPP-A levels were significantly higher (p = 0.001) in patients with a final diagnosis of acute coronary syndrome (median, 2.0 mIU/L; 25th quartile, 1.2; 75th quartile, 4.9) compared to patients with a final diagnosis of non-cardiac chest pain (median, 1.2 mIU/L; 25th quartile, 0.7; 75th quartile, 1.6). The mean age of the total population was 63.1 ± 12.8 years with 64% males. All patients were Caucasian from southern Minnesota, western Wisconsin and northern Iowa. The mean body mass index was 30.5 ± 5.0 kg/m2 with 50% of the patient population being overweight (i.e., body mass index > 30 kg/m2). Thirty three patients (56%) had hypertension, 10 (17%) had diabetes mellitus, 33 (56%) had
367
Table 1 The clinical and laboratory characteristics of the patient population according to final diagnosis of acute coronary syndrome (ACS) versus non-cardiac chest pain (CP) Variable
Non-cardiac CP (N = 40)
ACS (N = 19)
Age (years) Male, n (%) Body mass index Number of risk factors ⁎ (%) No risk factors Any 1 risk factor Any 2 risk factors Any 3 risk factors Any 4 risk factors All 5 risk factors History of coronary artery disease, n (%) Angina, n (%) Non-cardiac Atypical Typical ECG results, n (%) Normal Non-specific Initial troponin, median (Q1, Q3) (ng/mL) Estimated GFR, median (Q1, Q3) (mL/min) PAPP-A, median (Q1, Q3) (mIU/L)
60.1 ± 12.6 22 (55%) 30.9 ± 5.2
69.4 ± 10.9 16 (84%) 29.4 ± 4.5
6 (15%) 16 (40%) 11 (28%) 5 (12%) 2 (5%) 0 (0%) 10 (25%)
0 (0%) 5 (28%) 6 (34%) 5 (28%) 1 (5%) 1 (5%) 13 (68%)
29 (72%) 6 (15%) 5 (13%)
3 (16%) 5 (26%) 11 (58%)
P-value 0.008 0.029 0.29 0.20
0.0014 <0.001
0.60 28 (70%) 12 (63%) 12 (30%) 7 (37%) 0.01 (0.01, 0.02) 0.02 (0.01, 0.03)
0.13
69.5 (60.8, 77.5) 63.0 (57.0, 76.0)
0.36
1.2 (0.7, 1.6)
0.002
2.0 (1.2, 4.9)
⁎ Risk factors included: male gender, smoking, hypertension, diabetes mellitus, and hyperlipidemia (as defined by the National Cholesterol Program).
hyperlipidemia, 9 (15%) were active smoker, and 23 (39%) had family history of coronary atherosclerosis. Twenty three (39%) patients had a history of CAD. Thirty two (54%) patients presented with non-cardiac chest pain, 11 (19%) with atypical chest pain, and 16 (27%) with typical chest pain. Nineteen (32%) patients had non-specific ST/T wave changes on ECG while 40 (68%) had normal ECG at presentation. Median glomerular filtration rate was 68.0 mL/ min (25th quartile, 57.0; 75th quartile, 76.0), median troponin was 0.010 ng/mL (25th quartile, 0.01; 75th quartile, 0.01), and median PAPP-A was 1.29 mIU/L (25th quartile, 0.79; 75th quartile, 1.90). 3.2. Incremental diagnostic value of PAPP-A levels In a univariate model, measurement of PAPP-A levels at initial evaluation was found to be predictive of a final diagnosis of acute coronary syndrome in patients presenting with chest pain to the Emergency Department (Odds Ratio, 2.47; 95th confidence intervals, 1.29–4.27; p = 0.0062). In a multivariate model that corrected for age, sex, type of chest pain, number of risk factors, history of coronary artery disease, troponin levels, and non-specific ECG changes, PAPP-A levels were still predictive of a final diagnosis of acute coronary syndrome in patients presenting with chest
368
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
Fig. 1. Serum PAPP-A levels box plot distribution according to final diagnosis. The box indicates the interquartile range and median. The circles outside the box represent outliers. CP: chest pain. ACS: acute coronary syndrome.
pain to the Emergency Department (Odds Ratio, 2.093; 95th confidence intervals, 1.037–4.224; p = 0.039). 4. Discussion The results of this study suggest that serum PAPP-A may be valuable as an adjunct, minimally invasive marker to risk stratify intermediate- to high-risk patients presenting to the Emergency Department with chest discomfort. With each doubling in serum PAPP-A levels in patients presenting with chest pain, the risk of having an acute coronary syndrome likewise doubles. Our study population included patients categorized as intermediate to high risk without definite evidence of an acute coronary syndrome. This study population is particularly challenging since a significant proportion of these patients can have either cardiac or non-cardiac chest pain. In our particular patient population 32% had a final diagnosis of acute coronary syndrome while 68% had a non-cardiac etiology despite an initial elevated risk categorization. In our study, serum PAPP-A levels were found to be elevated in acute coronary syndrome patients who were troponin-negative at presentation at the Emergency Department, similar to the findings of Lund et al. [8]. That study, however, examined a more select patient group admitted to the hospital with an acute coronary syndrome [8]. In a previous study of patients undergoing coronary arteriography, we found elevated serum PAPP-A to be a marker of unstable atherosclerotic lesions [7]. A report by Beaudeux et al. [12] also found mildly elevated PAPP-A levels in asymptomatic hyperlipidemic patients with advanced
hyper-/isoechoic carotid atherosclerotic lesions. Together, these studies support a relationship between unstable coronary artery disease and level of serum PAPP-A. This study expands those prior reports and includes a broader range of patients similar in characteristics to those who present to an Emergency Department. Detection of unstable or potentially unstable vascular lesions is an important clinical goal, especially in the earliest stages when therapeutic interventions, both invasive and noninvasive, are most likely to result in the greatest benefit. A key feature emerging in these detection strategies is the identification of processes that confer instability to coronary or cerebral vascular lesions, at this time understood to be principally inflammation. We showed previously that PAPP-A localizes in unstable human plaque using immunohistochemical methods [7]. PAPP-A is associated with locally infiltrating macrophages, a principal inflammatory component of culprit lesions. Whether circulating PAPP-A detected in peripheral blood arises from vulnerable plaques is uncertain, but this provides an intriguing hypothesis about enhanced PAPPA sensitivity and specificity for detecting acute coronary syndromes. Indeed, emerging data suggest that inflammation may not be localized to a specific plaque. More often, the entire coronary tree is inflamed resulting in “pancoronaritis” and diffuse arterial destabilization. This often results in distant plaque progression and many acute coronary syndromes as supported by multiple prior studies [13,14]. Spagnoli et al. [13] found immunohistochemical evidence that patients dying from acute myocardial infarction have diffuse inflammation of all three major coronary arteries, irrespective of the artery with the culprit lesion. Pathophysiologic processes destabilizing artery wall segments are likely in multiple locations throughout the major epicardial coronary arteries. Other studies have shown similar results [13,14]. Taken together, these studies challenge the notion that a single vulnerable plaque is responsible for coronary plaque instability and resulting acute coronary syndromes. Elevated systemic levels of inflammatory markers in subjects with acute coronary syndromes may thus be due to the diffuse nature of the arterial inflammation rather than from a single site, and their detection in peripheral blood may prove a viable strategy for risk determination. The limitation of our study lies in the relatively small number of patients. We also did not collect follow-up information for major adverse cardiovascular events. However, our main interest was the immediate risk stratification of patients presenting to the Emergency Department as required by the physician facing such an acute situation. Second, this study was a prospective but observational evaluation. Unobserved confounders may have affected the study results, and multivariate analyses may have failed to completely remove all confounders of those covariables that were measured. Finally, we acknowledge that waiting for a second troponin in the Emergency Department is a
A.A. Elesber et al. / International Journal of Cardiology 117 (2007) 365–369
reasonable strategy. Nonetheless, saving a 6-h wait for a second troponin in a busy Emergency Department can be of great help in the triaging of patient care towards the patients who need that the most. Add to that fact that only about 10% of our patients with a negative initial troponin had a positive troponin at follow-up (data not shown). To conclude, serum PAPP-A has the potential to be a valuable marker of acute coronary syndromes, in particular, as a predictor of clinical instability in acute coronary syndrome patients who are troponin-negative. The results of this study suggest that further investigation into the potential for PAPP-A as point-of-care assay for decisionmaking in the Emergency Department is indicated. Acknowledgement This work was supported by FNDT0151290Z-1 from the American Heart Association (CAC), 0225543Z from the American Heart Association (ZTR), and Endocrinology and Metabolism Training Grant DK-07352 (ZTR) from the National Institutes of Health. References [1] Ly N, McCaig LF. National Hospital Ambulatory Medical Care Survey: 2000 outpatient department summary. Adv Data 2002;327: 1– 27. [2] American Heart Association 2000 Heart and Stroke Statistical Update. In:http://www.americanheart.org/statistics/04cornry.html ed: American Heart Association, 2000. [3] Antman EM, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335(18):1342–9.
369
[4] Hamm CW, et al. Emergency room triage of patients with acute chest pain by means of rapid testing for cardiac troponin T or troponin I. N Engl J Med 1997;337(23):1648–53. [5] Puleo PR, et al. Use of a rapid assay of subforms of creatine kinase-MB to diagnose or rule out acute myocardial infarction. N Engl J Med 1994;331(9):561–6. [6] Lawrence JB, et al. The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc Natl Acad Sci U S A 1999;96(6):3149–53. [7] Bayes-Genis A, et al. Pregnancy-associated plasma protein A as a marker of acute coronary syndromes. N Engl J Med 2001;345(14): 1022–9. [8] Lund J, et al. Circulating pregnancy-associated plasma protein a predicts outcome in patients with acute coronary syndrome but no troponin I elevation. Circulation 2003;108(16):1924–6. [9] Expert Panel on Detection, E. and A. Treatment of High Blood Cholesterol in, Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA 2001;285(19):2486–97. [10] Levey AS, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130(6):461–70. [11] Qin QP, et al. Molecular distinction of circulating pregnancyassociated plasma protein A in myocardial infarction and pregnancy. Clin Chem 2005;51(1):75–83. [12] Beaudeux JL, et al. Serum plasma pregnancy-associated protein A: a potential marker of echogenic carotid atherosclerotic plaques in asymptomatic hyperlipidemic subjects at high cardiovascular risk. Arterioscler Thromb Vasc Biol 2003;23(1):e7–e10. [13] Spagnoli LG, et al. Multicentric inflammation in epicardial coronary arteries of patients dying of acute myocardial infarction. J Am Coll Cardiol 2002;40(9):1579–88. [14] Buffon A, et al. Widespread coronary inflammation in unstable angina. N Engl J Med 2002;347(1):5–12.